Document j0Mr24pjeap4OaZ762bY0v5O

CURRICULUM VITAE DAVID L. EATON Professor of Environmental Health and Environmental Studies, and Associate Chairman, Department of Environmental Health, SC-34 School of Public Health and Community Medicine University of Washington Seattle, Washington 98195 (206)545-3785 . . .FAX(206)685-4696 PERSONAL Born August 15, 1952, Helena, Montana Married: Janet Children: Lucas (7), Kaley (4) EDUCATION 1978 University of Kansas Medical Center, Kansas City, Kansas, PhD, (Pharmacology) 1974 Montana State University, Bozeman, Montana, BSc, (Pre-medicine) PROFESSIONAL POSITIONS 1992-Present Professor, Department of Environmental Health, School of Public Health and Community Medicine, and Institute for Environmental Studies, College of Arts and Sciences, University of Washington 1991-Present A ssociate Chairman, Department of Environmental Health, School of Public Health and Community Medicine, University of Washington 1986-1991 D irector, Toxicology Program, Department of Environmental Health, School of Public Health and Community Medicine, University of Washington 1986-1992 A ssociate Professor, Department of Environmental Health, School of Public Health and Community Medicine, and Institute for Environmental Studies, College of Arts and Sciences, University of Washington 1979-1986 A ssistant Professor, Department of Environmental Health, School of Public Health and Community Medicine, and Institute for Environmental Studies, College of Arts and Sciences, University of Washington 19047 / xm E um n David L. Eaton Page 2 CURRICULUM VITAE 1978-1979 P ostd o cto ral Fellow in Toxicology, Department of Pharmacology, University of Kansas Medical Center, Kansas City, Kansas 1974-1978 P red o cto ral T rainee, US Public Health Service, Department of Pharmacology, University of Kansas Medical Center 1974 R esearch A ssistant, Department of Chemistry, Montana State University, Bozeman, Montana FELLOWSHIPS, HONORS, AWARDS Rohm & Haas Distinguished Professor of Public Health Sciences, 1992-97 NIH Pharmacology Research Associate Program Postdoctoral Fellowship Award, 1979-1981 (declined) Honorable Mention Overall, National Student Research Forum, Galveston, Texas, 1978 First Place Award, Pharmacology/Toxicology Division, National Student Research Forum, Galveston, Texas, 1978 First Place Award, University of Kansas Medical Center, Graduate Student Research Day, 1978 Who's Who in American Colleges and Universities, 1977 President, Graduate Student Council, University of Kansas Medical Center, 1977 First Place Award, Pharmacology/Toxicology Division, National Student Research Forum, Galveston, Texas, 1977 Second Place Overall, National Student Research Forum, Galveston, Texas, 1977 First Place Award, University of Kansas Medical Center, Graduate Student Research Day, 1977 Graduate with Distinction, Montana State University, 1974 CERTIFICATION Diplomate of the American Board of Toxicology, 1981, recertified 1985, 1990 PROFESSIONAL ORGANIZATIONS International Society for the Study of Xenobiotics (ISSX), 1992-present American Association for Cancer Research, 1988 - present Sigma Xi 1987-present Pacific Northwest Association of Toxicologists (Northwest Chapter, SOT); founding member, 1984-present, Vice President, 1984-1985; President, 1987-88, Councilor, 1988-89 Society of Toxicology, 1984-present Society for Environmental Toxicology and Chemistry, 1981-present, Regional Chapter Councilor, 1989-90 American Association for the Advancement of Science, 1977-present 19C48 I David L. Eaton Page 3 CURRICULUM VITAE PROFESSIONAL ACTIVITIES NATIONAL AND INTERNATIONAL Member, NEEHS Site Visit Team and Reviewer, Program Project Grant review, Oregon State University, 1992. Editorial Board, Environmental Carcinogenesis & Ecotoxicology Reviews, 1992-present Treasurer, American Board of Toxicology, Inc., 1991-94 M ember, Science Advisory Board, The Institute for Wildlife and Environmental Toxicology, Clemson University, Clemson, S.C. 1990-94 Councilor, Mechanisms Specialty Section, Society of Toxicology, 1990-93 Chairman, Membership Committee, Society of Toxicology, 1992-93 Membership Committee, (elected committee), Society of Toxicology, 1990-93 Board of Directors, American Board of Toxicology, Inc., 1990-94 Board of Publications, Society of Environmental Toxicology and Chemistry, 1989-91 Editorial Board, Toxicology and Applied Pharmacology, 1989-present Invited Participant, United States - Japan Cooperative Program on Development & Utilization of Natural Resources, Joint panel on toxic microorganisms symposium on "Cellular and Molecular Mode of Action of Selected Microbial Toxins in Foods and Feeds," National 4-H Center, Chevy Chase, MD, Oct. 31- Nov. 2, 1989 Member, Society of Toxicology ad hoc "Tox-90's Education Committee," 1988-present Invited Participant, Advisory workshop on significance and utilization of SARA Title III data, sponsored by USEPA, Chemical Manufacturer's Association and the National Academy of Sciences, NAS Headquarters, Washington, D.C., Oct. 18-19, 1988 Participant, 1987 Gordon Research Conference on Mechanisms of Toxicity, Kimball Union Academy, Meriden, New Hampshire, July 27-31, 1987 Member, NIH A d Hoc Scientific Review Panel for NIEHS contract on "Methods development for in vitro human metabolism of xenobiotics," February 19-21, 1985, Research Triangle Park, NC Committee on Public Communications, Society ofToxicology, 1984-1985 M anuscript R eferee, Aquatic Toxicology; Archives of Environmental Contamination and Toxicology; Biochemical Pharmacology; Biochemistry; Chemico-Biological Interactions; Environmental Science and Technology, Cancer Research; Fundamental and Applied Toxicology; Journal of Pharmacology and Experimental Therapeutics; Molecular Pharmacology; Pharmacology & Toxicology; Toxicology and Applied Pharmacology; Transactions of the American Fisheries Society COMMUNITY AND STATE Member, Northwest Consumer Food Safety Council, 1990-94 Chairman, Science Advisory Board (5 member board established by Initiative, pre empting above SAB), Washington State Department of Ecology, 1989-1991; Member 1989-92. Member, Science Advisory Board (14 member board established under legislative act), Washington State Department of Ecology, 1988-90 Member, Technical Work Group on Environmental/Regulatory Policy for the Puget Sound Water Quality Authority, 1987 G David L. Eaton Page 4 CURRICULUM VITAE Site V isitor, Western Washington University, Bellingham, WA; Requested to evaluate current occupational health and safety practices relating to toxic chemical storage and use at Western Washington University. Provided written report to Dean of the Graduate School, October 8, 1986 Member, Mayor's Advisory Task Force on PCB's, 1984-86 Member, Advisory Task Force to Assess Health Impacts of Sediment and Shellfish pollution at Eagle Harbor, WA. For Coalition of County and State Health Departments. October 1984-85 Member, Mayor's Health Advisory Panel on Gas Works Park, 1984 Member, Gypsy Moth Advisory Committee, Department of Agriculture, Washington State Medical Association, and Office of the Mayor, City of Seattle, 1983 Member, Emergency Insect Criteria Committee, State Department of Agriculture, 1982-86 Member, Washington State Pesticide Advisory Board (Gubernatorial Appointment), 1982-1985, reappointed 1986-1989 Member, State Department of Social and Health Services, Ad Hoc Committee to Review Health Implications of Totally Recycled Water Systems, 1981 UNIVERSITY Co-Chairman for Exhibits, University of Washington Health Sciences Open House, 1991 Member, Search Committee for new Chairman of Department of Environmental Health, 1990-91 Member, Civil Engineering Program Review Committee, University Programmatic Review by the Graduate School, 1990 Member, Environmental Studies Review Committee, campus-wide review of Environmental Studies programs and future directions, for the Office of the Provost, 1990 Chairman, University of Washington Technical Oversight and Advisory Committee for Ruston/Vashon Arsenic contamination remedial investigation feasibility study; consultant to Black and Veatch and Washington Department of Ecology, 1987 Member, Search Committee for Occupational Medicine faculty position (Chair, Dr. Linda Rosenstock), 1987 Member, Appointments, Promotions and Tenure Committee, Department of Environmental Health, 1986-present Member, Arsenic Pathways Review committee, to assist study group and local health departments interpret and present findings of CDC/UW. Arsenic Pathways Study, 1986 Member, Alcohol and Drug Abuse Institute Executive Council, 1986-present Member, Small Grants Review Committee, Alcoholism and Drug Abuse Institute, University of Washington, 1984-86 Chair, Curriculum and Teaching Policy Committee, Institute for Environmental Studies, 1983-present Member, Search Com m ittee for Dean of School of Public Health and Community Medicine, University of Washington (Chairman, Dean William Richardson), 1982 M ember, Curriculum and Teaching Policy committee, Department of Environmental Health, 1981-86; Chairman, 1985-86 Member, Search Committee for Director of Laboratories, Department of Environmental Health (Chairman, Dr. Ken Jackson), 1980 19050 David L. Eaton Page 5 CURRICULUM VITAE Core Faculty Member, Environmental Pathology Training Grant, Department of Pathology, University of Washington, 1980-present Member, Graduate Faculty, University of Washington, 1979-present PRESENTATIONS NATIONAL AND INTERNATIONAL Sem inar Speaker, Molecular basis for species differences in aflatoxin carcinogenicity, Chemical Industry Institute for Toxicology, Research Triangle Park, NC, June 4, 1992. Invited Speaker, "Risk Assessment in Managing Environmental Exposure Communications and Media Relations", Medical Group Management Association, Occupational Medicine Assembly Conference, Orlando, FL, February 18, 1992 Invited Speaker, Principles of Toxicology; Metals Toxicology; Carcinogenesis, mutagenesis and Teratogenesis lectures, as part of "-Risk Assessment, Management and the Communication of Drinking water Contamination, Continuing Education program sponsored by USEPA and NEHA, Portland, OR, June 27, 1991 Invited Alumni Speaker, The biochemical and molecular basis for species differences in susceptibility to aflatoxin Bi carcinogenesis, Department of Pharmacology, Toxicology and Therapeutics, University of Kansas Medical Center, Kansas City, KS April 9,1991 Invited Sem inar Speaker, The biochemical basis for species differences in aflatoxin carcinogenicity" Department of Environmental Health Sciences, Johns Hopkins University School Hygiene and Public Health, Baltimore, MD, February 12, 1990 Invited Sem inar Speaker, "The role of biotransformation in species susceptibility to aflatoxin Bi carcinogenesis", Medical Research Council, Toxicology Division, Carlshalton, England, Sept. 5, 1989 Participant and Presenter, Third International Conference on Glutathione STransferases, "Conjugation of aflatoxin Bi-8,9-epoxide by human liver glutathione Stransferases", Edinburgh, Scotland, August 31 - September 3, 1989 Invited Guest, National Public Radio Broadcast, Discussion on Alar in Apples, June 21, 1989 Invited Speaker, Pesticides in our food and Water. Putting the Risks in Perspective ',, public information seminar sponsored by the National Environmental Health Association Annual Meeting, Seattle, WA June 25, 1989 Invited Speaker, "Approaches to Risk Assessment and Risk Communication for Forest Use Herbicides", presented to Regional Forest Managers and other agency officials, Portland, Oregon, April 16, 1987 Invited Speaker, "Estimating Occupational and Public Health Risks of Pesticides", Northwest Aerial Applicators Association Convention, Pendleton, OR, November 12, 1986 Invited Speaker, "Toxicology of Wood Preservatives: Pentachlorophenol, Creosote and Arsenicals", given for Montana Department of Agriculture as part of wood preservative applicators' certification training program, Kalispell and Helena, MT, September 30 and October 1,1986 19051 David L. Eaton Page 6 CURRICULUM VITAE Session Co-Chairman, "Human Health Implications of Contaminated Seafood", International Symposium on Toxic Chemicals and Aquatic Life: Research and Management; provided closing remarks and session summary, Seattle, WA, September 16-18,1986 Sem inar Speaker, "Effects of enzyme induction on hepatobiliary disposition and DNA binding of Aflatoxin Bj in the rat"; Presented at Philadelphia College of Pharmacy and Science, Department of Pharmacology and Toxicology, October 12, 1984 COMMUNITY AND STATE Invited Speaker, "Fundamntalsof Toxicology," Part 1 & Part 2,1992 Northwest Federal Safety and Health Conference, Olympia, WA, April 29, 1992 Invited Speaker, "Basic Principles of Risk Assessment", American Public Works Association, Washington State Chapter, Spring Conference, March 26, 1992 Invited Speaker, "Toxic Chemicals, Human Health and the Environment", Science Enhancement for Teachers Program, University of Washington, April 27, 1991 Invited Speaker, "Are rodents good models for human responses to cancer-causing chemicals? Aflatoxin as an example. WEST `91 Conference, Seattle, Convention Center, April 11, 1991 Invited Speaker, "America's Epidemic of Chemicals and Cancer - Myth or Fact?" Inland Empire Agricultural Chemical Association 18th annual convention, Spokane, WA Dec. 11,1990 Invited Panel Participant, Land Based Marine Pollution in the Pacific Northwest, sponsored by the Ocean Studies Council, University of British Columbia, Vancouver, BC, Nov. 2,1991 Program Organizer and Speaker, "Human Health Risk Assessment: Process and Limitations", principal lecture for a symposium on uncertainties in risk assessment and risk management, given to upper level managers in the Washington Department of Ecology, March 13, 1990 Invited Speaker, "Risk Assessment in Foods", and Panel participant, "Chemical Residues in Foods", presented at Food Safety in Northwest Supermarkets, sponsored by Washington State University Cooperative Extension, Culinary Botany Northwest, and Department of Environmental Health, UW. October 12, 1989 Invited Guest, "Pesticides in Our Food", The Jim Altoff Show, KING 1090 Radio, October 9,1989 (1 hour) Invited Speaker, "Pesticide residues in our food and water- a hazard to your health, or a negligible risk? Saturday Seminar Series, sponsored by the Office of University Relations and UW Extension, Sept. 30, 1989 Sem inar Speaker, "Biochemical basis for Species Differences in Aflatoxin carcinogenesis", Washington State University Pharmacology/Toxicology Program Seminar, July 25,1989 Invited Participant, "Town Meeting", local television talk show focusing on health controversies over pesticide residues in food, KOMO TV, Seattle, WA, April 13, 1989. Invited Speaker, "Strengths and Weaknesses of Toxicological Evidence", presented at Continuing Education Program on "Legal Aspects of Occupational Health", University of Washington Education Resource Center, February 8, 1989, Seattle, WA. 19052 David L. Eaton Page 7 CURRICULUM VITAE Co-organizer and Speaker, "Pesticides in the Urban Environment: Health and Environmental Responsibility", program for 1988 annual convention of the International Pesticide Applicators Association, held in Bellevue, WA on Sept. 28-30, 1988 Invited Speaker, City Club ( a downtown business leaders civic group). Discussion of the procedures and difficulties encountered in utilizing science in decision making regarding public health risks. Seattle, WA, February 23, 1988 Invited Speaker, "Utilization of Scientific Information in Risk Management Decisions", presented to Division Mangers and Budget Coordinators for METRO, Battelle Conference Center, Seattle, WA, October 14, 1987 Invited Speaker, "Problems in Risk assessment and Risk Communication", King County Health Department annual retreat, Seattle Center, Seattle, WA, March 18, 1987 Invited Speaker, "Taking Care of Yourself - Pesticide Use and Human Health", presented at 77th Annual Meeting of the Western Washington Horticultural Association, Olympia, WA, January 6, 1987 Invited Speaker, "Toxic Chemicals and Public Health: Putting the Risks in Perspective", presented to the Seattle Chapter of the American Society of Safety Engineers, October 27,1986 Invited Speaker, "Principles of Toxicology", given at "Toxic Substances and Public Health Workshop", sponsored by the Indian Health Service, Seattle, WA, October 22, 1986 Invited Speaker, "Pesticides and Public Health: Risks and Uncertainties", presented at the International Pesticide Applicators Association Convention, Spokane, WA, October 2,1986 Invited Speaker, "Basic Principles of Toxicology", Risk Assessment Workshop for State Legislators Olympia, WA. Sponsored by Washington Department of Social and Health Services, September 25,1986 Invited Speaker, "Basic Principles of Toxicology and Risk Assessment", presented to Community Health nurses and doctors for the Seattle-King County Health Department, Seattle, WA, June 10, 1986 Invited Speaker, "Potential health effects of occupational Exposure to PCB's", presented to Snohomish County PUD utility workers, March 17, 1986 Invited Speaker, "Health Implications of Gaseous Emissions Associated with Midway Landfill", February 6, 1986, Kent, WA; sponsored by Washington State Department of Ecology Course Co-Organizer and Instructor, "Basic Principles of Toxicology for Environmental/Health Practitioners"; two day course given in Tacoma, WA (July 16 & 17,1985), and Spokane, WA (August 29 & 30,1985) Speaker, "Toxicology of PBC's, dioxins and dibenzofurans", given at a public meeting in Kitsap County at the request of Seattle City Light Environmental Affairs Division, Seattle, WA, June 17, 1985 Invited Speaker, "An evaluation of the health concerns over dioxins found in Eagle Harbor, WA", given a t a community meeting at the request of Dr. Willa Fisher, Director of the Kitsap County Health Department, July 11, 1985 Invited Speaker, "Health Impacts of Environmental Pollutants", given for the Continuing Medical Education Series, Riverton Hospital, July 19, 1985 David L. Eaton Page 8 CURRICULUM VITAE Invited Speaker, "Health Effects of PCB's", given to Seattle City Light employees on November 1, 1984, January 9, November 19, November 21, December 4, and December 11,1985 Course Organizer and Speaker, "Toxic Chemicals: Communicating Risks to the Public", forum for Scientist-media exchange, sponsored by the Society of Toxicology and the Department of Environmental Health, Battelle-Seattle Conference Center, Seattle, WA, April 27,1985 Invited Speaker, Public Perception of Health Impacts of Chronic Pesticide Exposure; International Pesticide Applicators Association Convention, Fife, WA, September 27, 1984 Course Organizer and Instructor, Toxicology Review Course for EPA Personnel; Seattle, WA, September 5 , 1984-January 15,1985 (attended by.45 EPA staff) Invited Speaker, Pesticides use for Gypsy Moth control; presented at Controversies in Poisoning Management and Occupational/Environmental Exposures, Children's Hospital and Medical Center, Seattle, WA, April 28-29, 1983 L ecturer, Recent Developments in biological Monitoring; Continuing Medical Education Course, Seattle, WA, March 18-19, 1983 Lecturer, Toxicology; short course presented to Hazards Assessment Program employees of NOAA, 3 lectures; NOAA Sand Point Facilities, Seattle, WA, March 10, 1983 Speaker, Health implications of Gypsy Moth control programs; Public Information Seminar, sponsored by Department of Environmental Health and Institute for Environmental Studies, University of Washington, March 10, 1983 Invited Speaker, Health effects of chronic pesticide exposure: a training program of health personnel: prevention, recognition and treatment of pesticide-related illness; sponsored by Yakima Valley Farm Workers Clinic and the Migrant Health Clinics in Washington, Oregon and Idaho, Pasco, WA, January 21-22, 1983 Invited Speaker, Clinical toxicology of pesticides; Northwest Poison Control Center Conference, Spokane, WA, April 22-23, 1982 Invited Speaker, A survey of toxic torts; Washington State Trial Lawyers Association Annual Meeting, Vancouver, BC, Canada, July 10, 1981 Invited Speaker, Carcinogens in the environment; Washington State Office of Environmental Education in Health Education, Everett, WA, March 10, 1981 Invited Speaker, Phenoxy acid herbicides-fact and fiction; Huxley College of Environmental Studies, Western Washington University. Bellingham, WA, January 27,1981 Invited Speaker, Central nervous system effects of industrial chemicals and physical agents; 1980 Occupational Health and Medical Conference, Spokane, WA, October 2224,1980 Invited Speaker, PCB's and human health. Washington State Office of Environmental Education and Health Education. Seattle, WA, February 27, 1980 UNIVERSITY Sem inar Speaker, "Delaney Amendment of Carcinogenic Substances". The 1992 Symposium on Science, Technology, arid Society, The Influence of Scientific Evidence on National Policy, April 28, 1992 Sem inar Speaker, "Glutathione S-transferases: Species and individual susceptibility to chemical carcinogens." Department of Environmental Health, UW, February 5, 1988 19054 David L. Eaton Page 9 CURRICULUM VITAE Course Co-Organizer and Speaker, "Working With Pesticides: Health and Safety Issues", Northwest Center for Occupational Safety and Health, University of Washington, Seattle, WA, January 16, 1987 Sem inar Speaker, "Role of Glutathione-S-Transferase in Detoxification of Aflatoxin Bi", Department of Environmental Health, University of Washington, Seattle, WA, December 4,1986 Invited Speaker, "Basic Principles of Toxicology", given at "Worker Health and Safety at Hazardous Waste Sites", continuing Education Program, Department of Environmental Health, University of Washington, Seattle, WA, December 4, 1986 Panel Member and D iscussant, "Scientist and the Media: M,aking Headlines on Your Own Terms", February 12, 1986; sponsored by University of Washington Medical Sciences Information Center Speaker and Program M oderator, Professional exchange on hazardous/toxic substances in and around the home - use and disposal; sponsored by Institute for Environmental Studies, University of Washington, Seattle, WA, June 18, 1984 Invited Speaker, Chemicals and cancer: Putting the risks in perspective; Continuing Education Course, Biologists Look at Life and Human Affairs, Biology Program, University of Washington, Seattle, WA, April 16, 1984 Course Organizer and Instructor, Legal aspects of toxicology; Continuing Education Course, Educational Resource Center, Department of Environmental Health, University of Washington, SeaTac Hotel, Seattle, WA, December 2-3,1983 Course Organizer and Instructor, Principles and practice of toxicology in environmental health; developed for Continuing Competency Education for Environmental Health Practitioners Program, University of Washington, Seattle, WA, January 25-March 15, 1982 Speaker, Central nervous system effects of chemicals and physical agents; Continuing Education Course, Educational Resource Center, University of Washington, Seattle, WA, April 20, 1981 Invited Speaker, Heavy metals in the environment; School of Public Health and Community Medicine's 10th Anniversary Celebration, University of Washington, Seattle, WA, September 25,1980 Invited Speaker, Occupational toxicology: principles of toxicology; Department of Environmental Health Continuing Education Program, University of Washington, Seattle, WA, September 19,1980 Sem inar Speaker, Carrier-mediated transport systems in isolated hepatocytes; Department of Environmental Pharmacology, University of Washington, Seattle, WA, November 5,1980 Sem inar Speaker, Effects of trace metals on metallothionein, glutathione, cytochrome P450 and heme oxygenase activity in rats. Department of Pathology, University of Washington. Seattle, WA, April 11, 1980 CURRENTLY FUNDED INDIVIDUAL AND COLLABORATIVE RESEARCH Program Director, NIH/NIEHS ES-04696 (Superfund program project), "Effects-Related Monitors of Toxic Exposure"; Annual Direct Costs, $1,682,598; 6/01/90 - 3/01/95; (Sheldon D. Murphy was Program Director from 9/29/87 until his death in May, 1990). 19055 David L. Eaton Page 10 CURRICULUM VITAE Co-Principal Investigator, NIH/NIEHS ES-04696 Superfund research project, "Glutathione perturbations with toxic exposures"; Annual Direct Costs, $96,263; 09/29/87 - 03/31/95 Principal Investigator, NIH/NIEHS ES-04696, Superfund Training Core; Annual Direct Costs, $94,303; 04/01/92-03/31/95. Principal Investigator, NIH/NIEHS, ES-05780, "Species differences in glutathione Stransferases", $129,527 Annual Direct Cost, 05/01/88 - 04/31/94. Principal Investigator, NIH/NIEHS/NCI, ES-03933, "Effect of enzyme inducers on liver preneoplastic lesions"; $127,580 Annual Direct Cost, 05/01/91 - 04/30/95 Principal Investigator, Washington Department of Health/UW Cooperative agreement for research projects in Toxicology and Public Health, 7/1/89-6/30/94, $243,000 TDC 19056 David L. Eaton Page 11 CURRICULUM VITAE BIBLIOGRAPHY vindicates students, post-doctoral fellows or research technologists under the direction of Dr. Eaton). RESEARCH PAPERS IN REFEREED JOURNALS 1. Eaton DL, Klaassen CD. Effects of acute administration of taurocholic and taurochenedeoxychlic acid on biliary lipid excretion. Proc Soc Exp Biol Med 151:198202,1975. 2. Eaton DL, Poisner AM. Plasma pseudorenin in rats after alteration in the reninangiotensin system. Proc Soc Exp Biol Med 154:6-8, 1977. 3. Eaton DL, Klaassen CD. Carrier-mediated transport of ouabain in isolated hpatocytes. J Pharmacol Exp Ther 285:480-488, 1978. 4. Iwamoto I, Eaton DL, Klaassen CD. Uptake of morphine and nalorphine by isolated rat hpatocytes. J Pharmacol Exp Ther 206:181-190,1978. 5. Eaton DL, Klaassen CD. Carrier-mediated transport of the organic cation procaineamide ethobromide by isolated rat liver parenchymal cells. J Pharmacol Exp Ther 206:595-606,1978. 6. Iga T, Eaton DL, Klaassen CD. Uptake of unconjugated bilirubin by isolated rat hepatocytes. Am J Physiol 236:C9-C14, 1979. 7. Eaton DL, Klaassen CD. Effects of microsomal enzyme inducers on carriermediated transport systems in isolated rat hepatocytes. J Pharmacol Exp Ther 208:381-385,1979. 8. Eaton DL, Klaassen CD. Effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin, kepone and polybrominated biphenyls on transport systems in isolated rat hepatocytes. Toxicol Appl Pharmacol 51:137-144,1979. 9. Eaton DL, Stacey NH, Wong KL, Klaassen CD. Dose-response effects of various metal ions on rat liver metallothionein, glutathione, heme oxygenase and cytochrome P-450. Toxicol Appl Pharmacol 55:393-402,1980. 10. Eaton DL. Biliary excretion of 2,4,5-trichlorophenoxyacetic acid in the rat. Toxicol Lett 14:175-181,1982. 11. Eaton DL, *Toal BF. Evaluation of the Cd/hemoglobin affinity assay for the rapid determination of metallothionein in biological tissues. Toxicol Appl Pharmacol 66:134-142,1982. 12. Eaton DL, *ToalBF. A simplified method of quantitating metallothionein in biological tissues. J Sci Total Environ 28:375-384,1983. 13. ^Carpenter LA, Eaton DL. The disposition of 2,4-dichlorophenoxyacetic acid (2,4-D) in the rainbow trout, Salmo gairdneri. Arch Environ Contam Tox 12:162-173, 1983. 14. *Stinson MD, E ato n DL. Concentrations of lead, cadmium, mercury and copper in the crayfish (Pacificasticus lenisculus) obtained from a lake receiving urban runoff. Arch Environ Contam Toxicol 12:693-700, 1983. 15. Woods JS, Fowler BA, Eaton DL. Studies on the mechanisms of thallium-mediated inhibition of hepatic mixed function oxidase activity: Correlation with inhibition of NADPH Cytochrome c (P-450) reductase. Biochem Pharmacol 33:571-576, 1984. 16. Woods JS, Eaton DL, *Lukens C. Studies on porphyrin metabolism in the kidney: Effects of trace metals and glutathione on renal uroporphyrinogen decarboxylase. Mol Pharmacol 26:336-341, 1984. 19057 David L. Eaton Page 12 CURRICULUM VITAE 17. Kalman D, E ato n DL, Schumacher RS, Covert D. Biological availability of lead in a paint aerosol. I. Physical and chemical characterization of a lead paint aerosol. Toxicol Lett 22:301-306,1984. 18. Eaton DL, Kalman DA, *Garvey D, Morgan M, Omenn GS. Biological availability of lead in a paint aerosol. 2. Absorption, distribution and excretion of intratracheally instilled lead paint particles in the rat. Toxicol Lett 22:307-313,1984. 19. Eaton DL. Short Communications: Effects of various trace metals on the binding of cadmium to rat hepatic metallothionein determined by the Cd/hemoglobin affinity assay. Toxicol Appl Pharmacol 78:158-162,1985. 20. *Monroe DH, *Holeski CJ, E ato n DL. Effects of single dose and repeated dose pretreatment with 2(3)-teri-butyl-4-hydroxyanisole (BHA) on the hepatobiliary disposition and covalent binding to DNA of Aflatoxin Bi in the rat. Food Chem Toxicol 24:1273-1281,1986. 21. Eaton DL, *Richards JA. Kinetic evaluation of carrier-mediated transport of ouabain and taurocholic acid in isolated rat hepatocytes. Evidence for independent transport systems. Biochem Pharmacol 35:2721-2725, 1986. 22. *Holeski CJ, Eaton DL, *Monroe DH, ^Bellamy GM. Effects of phnobarbital on the biliary excretion of Aflatoxin Pi-glucuronide and AflatoxinB i-S-glutathione in the rat. Xenobiotica 17:139-153,1987. 23. *Monroe DH, Eaton DL. Comparative effects of butylated hydroxyanisole (BHA) on the in vivo and in vitro biotransformation of Aflatoxin Bi (AFB) in rat and mouse. Toxicol Appl Pharmacol 90:401-409, 1987. 24. Geraci JP, Dunston SG, Jackson KL, Mariano MS, *Holeski C, Eaton DL. Bile loss in the acute intestinal radiation syndrome in rats. Rad Res 109:47-57, 1987. 25. Eaton DL, *Monroe DH, Kalman DA and *Bellamy G. Identification of a novel dihydroxy-metabolite of aflatoxin Bi formed both in vivo and in vitro in rats and mice. Chem Res Toxicol 1:108-114, 1988. 26. Franzblau A, Rosenstock L and E ato n DL. Use of inductively coupled plasmaatomic absorption spectroscopy (ICP-AES) in screening for trace metal exposures in an industrial population. Environ Res 46:15-24, 1988. 27. *Monroe DH and E ato n DL. Effects of modulation of hepatic glutathione on biotransformation and covalent binding of aflatoxin Bi to DNA in the mouse. Toxicol Appl Pharmacol 94:118-127, 1988. 28. *Ramsdell HS and Eaton DL. Modification of aflatoxin Bi biotransformation in vitro and DNA binding in vivo by dietary broccoli in rats. J Toxicol Environ Health 25:269275,1988. 29. Eaton DL, *Stapleton PL. Simultaneous determination of cytosolic glutathione Stransferase and microsomal epoxide hydrolase activity toward benzo[a]pyrene-4,5oxide by high performance liquid chromatography. Analytl Biochem 178:153-158, 1989. 30. *Ramsdell HS, Eaton DL. Species susceptibility to aflatoxin Bi carcinogenesis: Comparative kinetics of biotransformation. Cancer Res 50:615-620, 1990. 31. *Ramsdell HS, Eaton DL. Mouse liver glutathione S-transferase isoenzyme activity toward aflatoxin Bi-8,9-epoxide and benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide. Toxicol Appl Pharmacol 105:216-225, 1990. 32. Kavanagh TJ, Grossmann A, Jaecks EP, Jinneman JC, E ato n DL, Martin GM, Rabinovitch PS. Proliferative capacity of human peripheral blood lymphocytes sorted on the basis of glutathione content J Cell Physiol, 145:472-480,1990. 19058 David L. Eaton Page 13 CURRICULUM VITAE 33. *Ramsdell HS, Parkinson A, Eddy C, E aton DL. Bioactivation of aflatoxin Bi by human liver microsomes. Role of cytochrome P450IIIA isoenzymes. Toxicol Appl Pharmacol 108:436-447,1991. 34. *Trenga CA, Kunkel DD, E ato n DL, Costa LG. Effect of styrene oxide on rat brain glutathione. Neurotoxicology 12:165-178, 1991. 35. *Chen ZY, E ato n DL. Differential regulation of cytochrome P450 IIBi/2 by phnobarbital in hepatocellular hyperplastic nodules induced by aflatoxin Bi or diethynitrosamine plus 2-acetylaminofluorine in male F344 rats. Toxicol Appl Pharmacol 111:132-144,1991. 36. *Borroz Kl, *Ramsdell HS, E ato n D L Mouse strain differences in glutathione Stransferase activity and aflatoxin Bi biotransformation. Tox Lett 58:97-105, 1991. 37. Rosenstock L, Keefer M, Daniell WE, McConnell R, Claypoole K, Barnhart S, Costa LG, Demers P, E ato n DL, Morris S, Murphy SD. Chronic central nervous system effects of acute organophosphate pesticide intoxication. The Lancet 338:223-227, 1991. 38. *Anderson PN, E ato n DL, Murphy SD. Comparative metabolism of methyl parathion in intact and subcellular fractions of isolated rat hepatocytes. Fund. Appl Toxicol 18:221-226,1992. 39. Heckbert SH, Weiss NS, Homung SK, E aton DL, Motulsky AG. Glutathione Stransferase and epoxide hydrolase activity in human leukocytes in relation to the risk of lung and other smoking-related cancers. J Nat'l Cancer Inst 84:414-422, 1992. 40. *Buetler TM, E ato n DL. cDNA cloning, mRNA expression and induction of a-class glutathione S-transferases in mouse tissues. Cancer Res 52:314-318, 1992. 41. *Hamel DM, *White C, E ato n DL. Determination of y-glutamylcysteine synthetase and glutathione synthetase activity by HPLC. Toxicology Methods 1:273-288, 1992. 42. Kensler TW, Groopman JD, E ato n DL, Curphey TJ, Roebuck BD. Potent inhibition of aflatoxin-induced hepatic tumorigenesis by the monofunctional enzyme inducer l,2-dithiole-3-thione. Carcinogenesis 13:95-100,1992. 43. *Chen ZY, Farm F, Omiecinski CO, E ato n DL Association between growth stimulation by phnobarbital and expression of cytochromes P450 1A1, 1A2, 2Bi/2 and 3A1 in hepatic hyperplastic nodules in male F344 rats. Carcinogenesis 13:4, 675-682,1992. Manuscripts in Review: 1. *Chen Z-Y, E ato n DL. Association between responsiveness to phnobarbital induction of CYP2Bi/2 and 3A1 in rat hepatic hyperplastic nodules and their zonal origin. Environ Health Perspect. 2. Nazar-Stewart V, Motulsky AG, E ato n DL, *White, EJ, Homung, SK, Leng, Z-T, Stapleton, P and Weiss, NS. The Glutathione S-Transferase mu gene as a marker for susceptibility to lung carcinoma. New Eng J Med. 3. Kavanagh TJ, Grossman A, Jinneman JC, Kanner SB, *White CC, E ato n DL, Ledbetter JA, Rabinovitch PS. Glutathione depletion by l-chloro-2,4-dinitrobenzene inhibits T-cell receptor-stimulated transmembrane signal transduction in purified subsets of human peripheral blood lymphocytes. Toxicol Appl Pharmacol. David L. Eaton Page 14 CURRICULUM VITAE 4. *Chen ZY, E aton DL. Association between centrilobular origin and advanced differentiation phenotypic characteristics of hepatic hyperplastic nodules induced by aflatoxin Bi in male F344 rats. Carcinogenesis. 5. *Buetler, TM, *Slone, D and Eaton, DL. Comparison of the aflatoxin B1-8,9-epoxide conjugating activities of two bacterially expressed alpha class glutathione Stransferase isozymes from mouse and rat, Biochem Biophys Res Commun. 6. Buetler, TM and Eaton, DL. Amino acid sequence comparison, classification, and phylogenetic relationship between 65 glutathione S-transferases. Environ Carcin Ecotox Rev. 7. *Chen, ZY, *White, CC and Eaton, DL. Decreased expression of cytochrome P450 mRNA and related steroid hydroxylation activities in hepatic hyperplastic nodules in male F344 rats. Toxicol Appl Pharmacol. 8. Nazar-Stewart, V., Eaton, DL, White, E, *Stapleton, P and Weiss, NS. Markers for glutathione S-transferase and epoxide hydrolase enzymes in lung tissue and risk for lung carcinoma among smokers. BOOKS Eaton DL, Groopman JD, eds. The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural Significance. Academic Press, New York, in press, 1993. BOOK CHAPTERS. PUBLISHED REVIEW PAPERS. GENERAL EDUCATIONAL ARTICLE 1. Klaassen CD, Eaton DL, Cagen SZ. Hepatobiliary disposition of xenobiotics. In: Drug Metabolism Reviews. JW Bridges and LF Chasseaud, eds. London: John Wiley and Sons, 1981, p. 1-75. 2. Eaton DL. Toxicology of pesticides: A Northwest perspective. In: UW Medicine 1983; 9:11-17. 3. Eaton DL. Principles of Occupational Toxicology for Community Health Nurses. In: Occupational Health: A Self-Study Guide for Community Health Nurses, ed. S Holland, J Camp, J McNeil and AM Hughes, 1985. 4. Eaton DL, *Ramsdell HR, *Monroe DH. Biotransformation as a determinant of species susceptibility to aflatoxin Bi: in vitro studies in rat, mouse, monkey and human liver. In: Cellular and Molecular Mode of Action of Selected Microbial Toxins in Foods and Feeds, A Pohland, JL Richard, eds. Pergamon Press, New York, pp. 275-288,1990. 5. Eaton DL. Alar in Apples and Our Children's Health. In: Washington Public Health 8(Winter issue):8-9, 1990. 6. Eaton DL, Costa LG. Occupational exposure to pesticides in the state of Washington. In: Washington Public Health 8(Winter issue):26-28, 1990. 7. Eaton DL "America's Epidemic of Chemicals and Cancer" - Myth or Fact?" In: Environmental Science: A Framework for Decision Making, third edition, DD Chiras, ed., Benjamin Cummings Publishing Company, Menlo Park, CA, 429-431, 1990. David L. Eaton Page 15 CURRICULUM VITAE 8. E ato n DL, Gandolfi J, Dudley R, Mehendale HH, Medinski MM, Klaassen CD. In: Resource Guide to Careers in Toxicology, 1st edition, Society of Toxicology, Washington, D C., 1990; 2nd edition revised and published, 1991. 9. E ato n DL. "The Myths of the Cancer Epidemic" In: Human Biology: Health, Homeostasis and the Environment, DD Chiras, ed.,West Publishing Company, San Francisco, p. 425, 1991. 10. Klaassen CD, E ato n DL. Principles of Toxicology. In: Casarett and Doull's Toxicology: The Basic Science of Poisons, Fourth edition, MO Amdur, J Doull and CD Klaassen, eds., Pergammon Press, New York, pp. 12-49, 1991. 11. Omenn GS, Omiecinski CJ, E ato n DL. Ecogenetics of chemical carcinogens. In: Biotechnology and Human Genetic Predisposition to Disease, Cantor C, Caskey T, Hood L, Kamely D, Omenn GS, Editors, Wiley-Liss Inc, New York, pp. 81-93, 1990. 12. E ato n DL, Cherian MG. Determination of metallothionein in tissues by the Cadmium/Hemoglobin affinity assay. In: Methods in Enzymology, Metallobiochemistry, Riordan J, Valle B, Editors, Academic Press, New York, Chp. 13, pp. 83-88,1991. 13. E ato n DL, *Ramsdell HS. Species and diet related differences in aflatoxin biotransformation, In: Handbook of Applied Mycology, voi. 5, D. Bhatnagar, editor, Marcel Dekker, Inc., New York, Chap. 6, pp. 157-182, 1992. 14. Kalman DA, E ato n DL, Lorenzana R. Tracking hazardous emissions from municipal landfills. In: Washington Public Health 9: 50-51, 1991. 15. *Kimbell M, *Davis M, E ato n DL, Lorenzana RM. Washington Department of Health Guide to Physico-chemical, Toxicological and Regulatory Values for Priority Pollutants. Washington Department of Health, 1991. 16. E ato n DL, Robertson WO. Toxicology. In: Textbook of Occupational and Environmental Medicine, Rosenstock L, Cullen M, eds. WB Saunders Company, New York, in press, 1992. 17. E ato n DL, *Ramsdell HS, Neal G, Biotrasformations of Aflatoxins, in E aton DL, Groopman JD, eds. The Toxicology of Aflatoxins: Human Health, Veterinary and Agricultural Significance. Chp. 6, Academic Press, Orlando, in press, 1993. 18. *Gallagher E, Kavanagh TJ, E ato n DL. Methods in Toxicology, Vol. II: In Vitro Toxicity Indicators, CA Tyson, JM Frazier, eds., Academic Press, Orlando, in press, 1992. abstracts 1. E ato n DL, Bunag RD. Anesthetic depression of ouabain-induced arrhythmias in rats. Fed Proc 1976; 35:383. 2. E ato n DL, Poisner AM. A comparison of the effects of surgical and pharmacological intervention of plasma renin and pseudorenin levels. Pharmacologist 1976; 18:230. 3. E ato n DL, Klaassen CD. Kinetics of hepatic uptake of ouabain in isolated rat hepatocytes. Gastroenterology 1970; 72:A20. 4. E ato n DL, Klaassen CD. Carrier-mediated transport of ouabain in isolated hepatocytes. Fed Proc 1977; 36:960. 5. E ato n DL, Klaassen CD. Uptake of procaineamide ethobromide into isolated rat hepatocytes. Fed Proc 1978; 37:368A. 19081 David L. Eaton Page 16 CURRICULUM VITAE 6. E ato n DL, Klaassen CD. Effects of microsomal enzyme inducers on carriermediated transport systems in isolated rat hepatocytes. Gastroenterology 1978; 75:961. 7. E ato n DL, Klaassen CD. Alterations in hepatic transport systems in isolated rat hepatocytes after treatm ent with microsomal enzyme inducers. Toxicol Appi Pharmacol 1979; 48(Suppl):A187. 8. E aton DL, Stacey NH, Wong KL, Klaassen CD. Effects of twelve metal ions on metallothionein, glutathione, heme oxygenase and cytochrome P-450 content in rat liver and kidney. Toxicol Appi Pharmacol 1980; 50(Suppl):A290. 9. E ato n DL, *Toal B. A simplified method for quantitating metallothionein in biological tissues. Proc 21st Hanford Live Sciences Symposium, Biological Availability o f Trace Metals, J.981, p 48. 10. E ato n DL, *Toal B. A simple, sensitive and quantitative assay for metallothioneins. Toxicologist 1982; 2:317. 11. *Holeski CJ, E ato n DL. Effects of diethylmeleate on hepatobiliary disposition of Aflatoxin Bi in the rat. Fed Proc 1982; 41:7552. 12. Kalman DA, E ato n DL. Detection of glucuronide conjugates of xenobiotics by direct insertion mass spectrometry. Proc 30th Ann Conf Mass Spectrometry and Allied Topics, Amer Soc Mass Spectrom, June 6-11, 1982. 13. Woods JS, E ato n DL, *Lukens C. Studies on the renal etiology of mercury-induced uroporphyrinuria. Toxicologist 1984; 4:74. 14. Woods JS, E ato n DL, *Lukens CB. Effects of glutathione on the regulation of renal uroporphyrinogen decarboxylase. Toxicologist 1984; 4:32. 15. *Holeski CJ, *Bellamy GM, E ato n DL. Effects of phnobarbital (PB) pretreatment on hepatobiliary disposition of Aflatoxin Bi in the rat. Toxicologist 1984; 4:151. 16. *Monroe DH, *Holeski CJ, *Bellamy GM, E ato n DL. Effects of a purified versus standard rodent diet on hepatic biotransformation enzyme activities. Toxicologist 1984; 4:98. 17. *Holeski CJ, E ato n DL, *Bellamy GM. Analysis of Aflatoxin metabolites in bile following treatm ent with phnobarbital. PANWAT Proc 1984; 1:13. 18. *Monroe DH, *Holeski CJ, E ato n DL. Effects of butylated hydroxyanisole (BHA) on hepatobiliary disposition of Aflatoxin Bi (AFB) in the rat. PANWAT Proc 1984; 1:14; Toxicologist 1985; 5:149. 19. E ato n DL, ^Bellamy GM, Kalman DA. Identification of Aflatoxin Pi-glucuronide in bile of rats given Aflatoxin Bi. PANWAT Proc 1984; 1:22; Toxicologist 1985; 5:961. 20. *Holeski CJ, E ato n DL. Effects of hepatic glutathione depletion on the DNA binding and hepatobiliary disposition of Aflatoxin Bi in vivo. PANWAT Proc 1985; 2:16 & 2:32; Toxicologist 1985; 5:686. 21. *Anderson PN, E ato n DL, Murphy SD. Analysis of methyl parathion metabolites with reverse-phase ion-pair high-performance liquid chromatography. PANWAT Proc 1985; 2:28;Toxicologist 1986; 6:319. 22. Geissler F, E ato n DL, Faustman-Watts E. Biochemical studies on the role of biotransformation in mycotoxin-induced embryotoxicity in vitro. Toxicologist 1986; 6:379. 23. E ato n DL, *Richards JA. Kinetic evaluation of carrier-mediated transport of ouabain and taurocholic acid in isolated rat hepatocytes. Evidence ft- independent transport systems. Toxicologist 1986; 6:469. David L. Eaton Page 17 CURRICULUM VITAE 24. ""Holeski CJ, E ato n DL. Effects of glutathione depletion by buthionine sulfoximine (BSO) and diethyl maleate (DEM) on biliary excretion of Aflatoxin Bi (AFB) conjugates and covalent AFB-DNA binding. Toxicologist 1986; 6:605. 25. Eaton DL, ""Bellamy GM, Kalman DA, Monroe DH, Holeski CJ. Separation and partial characterization of 17 metabolites of Aflatoxin Bi (AFB) excreted in bile of rats. Toxicologist 1986; 6:994. 26. Monroe DH, E aton DL. Effects of butylated hydroxyanisole (BHA) on the in vivo and in vitro metabolism of Aflatoxin Bi (AFB) in the mouse liver. Toxicologist 1986; 6:995, PANWATProc 6:3, 1986. 27. *Monroe DH, E ato n DL. Comparative effects of butylated hydroxyanisole (BHA) on the in vivo and in vitro metabolism of Aflatoxin Bi (AFB) in rat liver. PANWAT Proc 3:6,1986, Toxicologist 7:146, 1987. 28. E ato n DL, ""Monroe DH, ""Bellamy GM. Development of an HPLC assay for four MFO hydroxylation enzymes towards Aflatoxin Bi and GSH-S-transferase towards Aflatoxin B-8,9-epoxide. PANWAT Proc 3:17, 1986, Toxicologist 7:872, 1987. 29. *Monroe DH, E ato n DL. Effects of modulation of hepatic glutathione on covalent binding of aflatoxin Bi to DNA in the mouse. PANWAT Proc 4:2, 1987. 30. ""Richards JA, E ato n DL, Homung SK, Motulsky AG, Hammock BD. Characterization of glutathione S-transferase activity towards various substrates in human mononuclear leukocytes. PANWAT Proc 4:7, 1987; Toxicologist 8:452, 1988. 31. ""Hagelstein H, E ato n DL. Characterization of liver, lung and leukocyte glutathione S-transferase activity in rat, mouse and monkey. PANWAT Proc 4:8, 1987. 32. ""Ramsdell HS, E ato n DL. Modification of aflatoxin Bi metabolism in vitro and DNA binding in vivo by dietary broccoli in rats. PANWAT Proc 4:21, 1987; Toxicologist 8:670,1988. 33. ""Rank JP, E ato n DL. Dmthylation of methyl organophosphates by rat hepatic glutathione S-transferases. PANWAT Proc 4:22, 1987; Toxicologist 8:691, 1988. 34. ""Stapleton PL, E ato n DL. Concurrent measurement of glutathione S-transferase and epoxide hydrolase activity by HPLC. Toxicologist 8:729,1988. 35. Faustman EM, ""Stapleton PL, E ato n DL. Characterization of rodent embryonic glutathione S-transferase activity toward various substrates. Toxicologist 8:742, 1988. 36. Eaton DL, ""Hamel DM. Induction of gamma-glutamylcysteine synthetase (GCS) activity by butylated hydroxyanisole in the mouse. PANWAT Proc 1988; 5:29\Toxicologist 9:759,1989. 37. ""Ramsdell HS, E ato n DL. Effects of butylated hydroxyanisole (BHA) on mouse liver glutathione S-transferase (GST) isoenzyme activity toward aflatoxin Bi-epoxide (AFBO). Toxicologist 9:832,1989. 38. ""Ramsdell H, E ato n DL. Human susceptibility to aflatoxin Bi carcinogenesis: liver microsomal biotransformation. PANWAT Proc 6:16, 1989. 39. Bailamme D, Grimsted B, E ato n D, ""Ramsdell H, Faustman EM. Developmental toxicity of hydralazine in vitro: role of metabolism. Toxicologist 9:114, 1989. 40. ""Chen Z-Y, E ato n DL. Comparison of glutathione S-transferase subunit composition and activity in liver, lung, and kidney of Wistar, Fishcher and Sprague-Dawley rats. Proceedings American Association Cancer Research 30:157 (#620), 1989. 19083 David L. Eaton Page 18 CURRICULUM VITAE 41. *Ramsdell HS, Eaton DL. Kinetics of aflatoxin Bi biotransformation in vitro by hepatic microsomes from the rat, mouse, monkey and human. Proceedings American Association Cancer Research 30:157 (#621), 1989. 42. *Trenga CA, Eaton DL, Costa CG. Effect of styrene oxide on rat brain glutathione. PANWAT Proceedings 6:5, 1989. 43. *Ramsdell HR, Eaton DL. Human susceptibility to aflatoxin Bi carcinogenesis: liver microsomal metabolism. PANWAT Proceedings 6:16, 1989. 44. Faustman E, Horike N, MacQueen H, E ato n DL. Is your health in jeopardy? Interactive educational computer program for junior and senior high school students. Toxicologist 10:354,1990. 45. Eaton DL, *Ramsdell HS. Human susceptibility to aflatoxin Bi carcinogenesis: Individual variations in biotransformation. Toxicologist 10:578, 1990. 46. *Michelson KD, Eaton DL. Substrate-specified inhibition of rat liver cytosolic glutathione S-transferase isoenzymes (GST) by chalcone. Toxicologist 10:796, 1990. 47. *Berlad LG, Eaton DL. Inhibition of human liver cytosolic glutathione S- transferase isoenzymes (GST) by tri-butyltin acetate (TBT), 2,4dichlorophenoxyacetic acid (2,4-D) and ethylene dichloride (EDC). Toxicologist 10:1137,1990. 48. *Trenga CA, Costa LG, E ato n DL. Regional distribution of glutathione in rat brain: Effect of styrene oxide. Toxicologist 10:1281,1990. 49. *Ramsdell HS, Eaton DL. Bioactivation of aflatoxin Bi by human liver microsomes: role of cytochrome P450IIIA isoenzymes. Toxicologist 10:1289, 1990. 50. *Chen ZY, Eaton DL. Phnobarbital (PB) induction of cytochrome P450IIBi/2 in hepatic hyperplastic nodules (HHN) produced by aflatoxin Bi (AFB) or Solt-Farber (SF) protocol in rats. Proceedings American Association Cancer Research 31:153 (#911), 1990. 51. TJ Kavanagh, A Grossman, JC Jinneman, Rabinovitch PS (Eaton DL consulted). Glutathione depletion inhibits transmembrane signal transduction in human CD4+ and CD8+ lymphocytes. Toxicologist 11:117, 1991. 52. ^Channel SR, Eaton DL, Kavanagh TJ. Regulation of intracellular glutathione in a rat diploid hepatic epithelial cell line. Toxicologist 11:127, 1991. 53. *Borroz KI, *Ramsdell HS, Eaton DL. Mouse strain differences in glutathione Stransferase activity and aflatoxin Bi biotransformation. Toxicologist 11:142, 1991. 54. *Ramsdell HS, *Slone DH, E ato n DL. Conjugation of aflatoxin Bi-8,9-epoxide by glutathione S-transferases from human and rodent liver. Toxicologist 11:143, 1991. 55. *Buetler MT, Eaton DL. Cloning and characterization of a murine alpha-class glutathione S-transferase cDNA homologous to rat GST Yc (GST 2-2). Proceedings American Association Cancer Research 32:821, 1991. 56. *Chen ZY, E ato n DL. Selective growth stimulation by phnobarbital (PB) and expression of different cytochromes P450 in hepatic hyperplastic nodules (HHN) induced by Solt-Farber protocol in male F344 rats. Proceedings American Association Cancer Research 32:904, 1991. 57. *Chen ZY, Eaton DL. Relationship between expression of cytochrome P450 enzymes and other phenotypic characteristics within hepatic hyperplastic nodules in F344 rats. Presented at "Cell Proliferationand Chemical Carcinogenesis" symposium; NIEHS, Research Triangle Park, NC, Jan.13-16, 1992. David L. Eaton Page 19 CURRICULUM VITAE 58. *Borroz Kl, E ato n DL. Development and utilization of a cDNA probe for gglutamylcysteine synthetase to study transcriptional regulation of glutathione biosynthesis. The Toxicologist 12:1613, 1992. 59. *Buetler MT, E a to n DL. cDNA cloning and mRNA expression of murine glutathione S-transferase and analysis of purified alpha class GST proteins expressed in E. coli. The Toxicologist 12:1633, 1992. 60. Ogbum CE, Orsbom MT, *White CC, E ato n DL, Martin GM, Kavanagh TJ. Mouse teratocarcinoma cells selected for oxygen resistance have increased glutathione content. The Toxicologist 12:191,1992. 61. *Chen ZY, E ato n , DL. Expression of cytochrome P450 genes involved in hydroxylation of steroid hormones in hepatocyte nodules from Male F344 rats. Proceedings American Association Cancer Research 33:753, 1992. 62. *Buetler TM, *Van Ness KP, E aton, DL. Alignment of multiple glutathione Stransferase amino acid sequences and prediction of structure-fimction relationship. Proceedings American Association Cancer Research 33:1175, 1992. 9/14/92 ' 19065 UW TEACHING RESPONSIBILITIES : Courses with principal responsibility: David L. Eaton Page 1 CV ADDENDUM ENVH/ENVS 515, Environmental and Occupational Toxicology, typical enrollment of 30-40 students, taught annually from 1980-1991 (as ENVH/ENVS 415 from 80-82). ENVH/ENVS 305/405, Toxic Chemicals in the Environment, typical enrollment of 5065 students, taught annually from 1982 - 1984, even numbered years from 86-present. Annually from 1992 on. ENVH 567 Environmental and Occupational Carcinogenesis, new course developed and taught on odd-numbered years from 1993 - on. Annual Guest Lectures: ENVS 101, Introduction to Environmental Studies, Fall and Spring quarters, 2-4 lectures HUBIO 555, Medicine, Health and Society, Winter quarter, 2 lectures. ENVH 411, Introduction to Environmental Health, Spring quarter, 1 lecture. ENVH 590c, Ecogenetics, Winter quarter, 1 lecture. HUBIO 567, Medical Pharmacology, Spring quarter, 2 lectures. C MED 520/530, Biology and Disease of Laboratory Animals, Fall quarter, 1 lecture. ENVH 567, Industrial Carcinogens, 1-2 lectures. Students/Fellows Precepted: Post-Doctoral Fellows: Dr. Howard Ramsdell, 1987-90. Dr. Zhi-Ying Chen, 1988-present Dr. Martin Buetler, 1990-present Dr. Evan Gallagher, 1991-present Doctoral Students: Carolyn Holeski (with Dr. N. Karle Mottet, Pathology), 1986 David H. Monroe (Ind. Ph.D.), 1987 Kirk Van Ness, in progress Master's Students (year graduated): Brian Toal, 1981 Leslie Carpenter, 1981 Stephan Dobratz, 1983 Jeffry Defenbach, 1984 David Monroe, 1984 Priscilla Anderson, 1986 Julia Richards, 1988 Karyn Michelson, 1989 Denise Bender, 1990 Carol Trenga (with Dr. Lucio Costa), 1990 Steve Channel (with Dr. Terry Kavanagh), 1990 Dana Stahl, in progress Margaret Stinson, 1981 Don Garvey, 1982 JefFBailet, 1984 Carl Lukens, 1984 Kim Lowry, 1985 Heidi Hagelstein, 1987 Denise Hamel, 1989 LynnBerlad, 1989 Ingrid Borroz, 1991 John Heinonen, 1992 19058 EXPIRED GRANTS P rin cip al Investigator, N1H/NIEHS, Carrier mediated transport of xenobiotics in isolated hepatocytes. 6/1/83-5/31/86, $245,000 DC P rin cip al Investigator, NIH/NIEHS/NCI, ES-03933, "Modification of aflatoxin disposition by dietary enzyme inducers"; $324,964 TDC , 05/01/88 - 04/30/91, (renewed as "Effects of enzyme inducers on liver preneoplastic lesions, NIEHS) C o-Investigator, Dana Foundation Grant in Genetics and Environmental Health, (PI, Dr. Amo Motulsky); 09/01/85-06/30/92; sub-project, "Genetic Differences in Glutathione S-transferase activity". Co-Investigator, US Forest Service (USDA); "Health Effects Implications of Pacific Northwest Region Vegetation Management", 10/01/86-9/30-88; C o-Investigator, Milbank Memorial Fund, "Illness and Injuries Among Washington State Agricultural Workers", PI-Dr. Linda Rosenstock, 11/01/86-04/30/88; C o-Investigator, NIH/NIOHS; "Peripheral markers of styrene toxicity", PI, Dr. Harvey Checkoway, 12/01/88-11/30/90; C o-Investigator, NIH/NHLBI; "Metabolism of propranolol in uremic rat liver", PI, Dr. Danny Shen, Dept. Pharmaceutics; 04/01/88-03/31/92 Numerous small institutional grants: SPHCM BRSG (2), Graduate School Young Investigator Award, American Cancer Society Institutional Grants (2), Departmental Industrial Hygiene Research Grant. 13067 David L. Eaton Page 1 CONSULTING ACTIVITIES PROFESSIONAL CONSULTING ACTIVITIES CONSULTING FIRMS B attelle Human Affairs Center, Seattle, WA. (Dr. James Woods) Advisory board member for Battelle contract to evaluate scientific review process utilized by EPA for pesticide evaluation for registration. Environmental Toxicology International, Seattle, WA. (Dr. Kathryn Kelly) Prepared evaluation of potential human health risks associated with DDT soil contamination at a utility sub-station. Harding-Lawson A ssociates, Novato, CA. (Dr. Michael Zander) Reviewed human health risk assessment portion of feasibility study of Milltown Reservoir CERCLA site containing arsenic residues. URS Engineers, Seattle, WA. (Mr. Bob Dexter) Expert consultant on health effects of industrial pollutants. CH2M H ill, Seattle, WA. (Mr. John Anthony) Expert consultant to evaluate potential human health hazards associated with commercial development of an abandon creosote plant site into condominium complex. (In association with Woodward-Clyde Associates, San Francisco, CA.) K.S. Crump & Company, Ruston, LA. (Dr. Kenny Crump) Assisted in the development and evaluation of "Worst-Case Analysis" for forest herbicide use on public lands in the State of Washington. Tetra Tech, Inc. (Dr. Robert Pastorak) Served as toxicology and risk assessment consultant on various projects associated with evaluating risks of consumption of contaminated seafood from Puget Sound. Tetra Tech, Inc. (Dr. Gary Pascoe) Assisted in the development of a survey of pesticide use in the greater Puget Sound area, including evaluation of potential environmental and human health significance. Param etrix, Inc. (Dr. Ramon Beluche) Reviewed human health components of various consultant documents on hazardous and solid waste site evaluation and clean-up plans. Woodward-Clyde Consultants, (Dr. William Zbitnoff) Peer-reviewed human health risk assessment on potential airborne exposures and ground water contamination from municipal landfill. PUBLIC AGENCIES Departm ent of Agriculture, State of Washington. (Mr. Art Losey) Provided expert evaluation of human health implications and risk assessment of chemical options, especially carbaryl, for control of Gypsy Moth in urban areas. Seattle/King County Health Department, Seattle, WA. (Mr. Chuck Kleeberg) Provided numerous expert consultations on local environmental health problems, including: Gas Works Park - evaluation of human health risks associated with a City Park located on a former coal gasification plant site which was heavily contaminated with polyaromatic hydrocarbons; assisted City/County in evaluating health risks from several solid waste landfill sites; assisted in evaluating human health risks from consumption of contaminated seafood in Puget Sound area. Bureau o f Land M anagement, Washington, D.C. Evaluated worst-case analysis of herbicides for federal forest lands. 19C.S8 David L. Eaton Page 2 CONSULTING ACTIVITIES Tacoma/Pierce County Health Department, Washington. (Mr. Doug Pierce) Assisted in evaluating human health risks associated with arsenic residues in the community from ASARCO smelter emissions. Kitsap County Health Department, Washington. (Dr. Willa Fisher) Assisted in evaluating human health risks associated with consumption of shellfish contaminated with polyaromatic hydrocarbons and other industrial waste products in Eagle Harbor, WA. Seattle City Light, Environmental Affairs Division (Ms. Clair Dyckman) Reviewed toxicology and environmental effects of pesticides used by SCL, and prepared written evaluation of pesticide inventory and use practices. N orthw est Poison Control Center, Children's Hospital and Medical Center. (Dr. William Robertson) Provide occasional telephone consultations in pesticide and unusual toxic chemical poisoning. U.S. F orest Service, Pacific Northwest Region. (Mr. Gary Larsen, Group Leader, Vegetation Management) Reviewed toxicologic data on forest use of herbicides, and assisted U.S. Forest Service in preparation, evaluation and interpretation of risk assessments (worst case analyses) as part of EIS. City o f Seattle, Solid Waste Division (Ms. Lin Robinson). Provided written evaluation of a draft Endangerment Assessment for the Midway Landfill. LITIGATION EXPERIENCE Aaby, Knies and Robinson, Bellingham, WA. (Mr. Robert Knies, plaintiffs attorney) Expert witness on occupational accident case involving high level 2,4-D exposure. Reed, McClure, Moceri & Thonn, P.S., Seattle, WA. (Mr. Dennis Smith, defendant's attorney) Provided expert consultation and review on occupational exposure to ethylene oxide. King County Prosecutor's Office, Seattle, WA. (Mr. Michael Linnabury, representing county as defendant.) Provided extensive evaluation, review and expert testimony regarding scientific information relevant to purported claims of adverse human health impacts resulting from living near a municipal landfill site. County was a defendant. R ichter, Wimberley & Ericson, P.S., Spokane, WA. (Mr. Gary Bloom, Plaintiffs attorney, class-action suit) Provided expert consultation and opinions regarding potential human health effects of chronic exposure to cyanide in drinking water (Spokane River aquifer). M alagone & Velure, Eugene, OR. (Mr. Peter McSwain, plaintiffs'attorney) Reviewed documents and provide opinion (unfavorable) on occupational injury claim form solvent (toluene) exposure and possible relatedness to cardiovascular accident. Ferguson & Burdell, Seattle, WA. (Mr. Thomas Wolfendale, defendant's attorney) Expert evaluation of personal injury claim related to exposure to carbonless copy paper. Winer, Bennet, Harman & Tauman, P.C., Portland, OR. (Mr. Charles Tauman, plaintiffs' attorney) Provided expert evaluation and opinion regarding relationship between neurological deficits and chronic, very high level exposure to organic solvents in a work place. * David L. Eaton Page 3 CONSULTING ACTIVITIES U.S. Department of Justice D istrict of New Mexico. (Mr. Raymond Hamilton, Assistant U.S. Attorney, on behalf of U.S. Forest Service as defendant) Provided expert opinion (in testimony) on extent or significance of public health effects that complainants alleged would occur if application of carbaryl to national forest lands was allowed. G. Lee Raaen, Attomey-at-Law, Seattle, WA. (Lee Raaen, plaintiff s attorney) Provided consultation and expert testimony on relationship between arsenic exposure and diagnosed peripheral neuropathy. Agency Legal Services Bureau Attorney General's Office State of Montana. (Opal Winebrennar, Assistant Attorney General) Provided consultation and expert opinion on behalf of defendant (State of Montana) in workers compensation case claiming that leukemia resulted from herbicide (2,4-D) exposure. Bogle and Gates Seattle, WA. (Mr. David Bateman, defendant's attorney, for Olympic Pipeline Company.) Assisted Department of Engineering Program health risk assessment and public communications of risk surrounding pipeline leak of gasoline products in residential neighborhood. Seattle City Attorney's Office, Seattle, WA. (Mr. Dennis McClairen and others) Provided consultation and expert opinion on likelihood of health risks occurring to residents near a municipal landfill. Reed, McClure, Moceri, Thonn & Moriarty, Seattle, WA (Mr. Dennis Smith, Defendant's attorney) Provide consultation as to state of knowledge in 1960's of toxicologic hazards of various chemicals identified at a hazardous waste site. W illiams, Kastner & Gibbs, Seattle, WA (Mr. Roy Umlauf, defendant's attorney) Provide consultation as to effects of specific chemicals identified at a hazardous waste site. U.S. Departm ent o f Ju stice, Washington, D. C. (Mr. Kevin Murphy, Trial Attorney, Lands Division) Expert Testimony in criminal prosecution case for illegal disposal of pesticides, Spokane, WA. Betts, Patterson & M ines, Seattle, WA (Mr. John Allison, defendant's attorney) Provide research, consultation and expert testimony on potential health impacts of consumption of groundwater contaminated with small amounts of EDB. Shidler, McBroom & Gates, Seattle, WA (Mr. Joel Summer, plaintiff s attorney) Provide, research, consultation and expert opinion on herbicide (paraquat)-related death. Preston, Thorgrimson, E llis & Holman, Seattle, WA (Mr. Ross Macfarlane, attorney for pulp and paper processor) Provide consultation and guidance on proper disposal of "hogged" fuel ash to ensure protection of human health and environment. Underwood, Campbell, Brock & Cerutti, P.S., Spokane, WA (Mr. Jay Leipham, defendants attorney). Provide research, consultation and expert opinion regarding health impacts of consumption of groundwater contaminated with nitrates. Moulton, Bellingham , Longo & Mather, P.S., Billings, MT (Ms. Tamara A. Scully, defendants attorney); Provide research, consultation and expert opinion regarding health impacts of atrazine herbicide. Lukins & Annis, P.S., Spokane, WA. (Mr Ted Stiles, defendants attorney). Provide research, consultation and expert opinion regarding health impacts of malathion exposure. P erkins Coie, Seattle, WA (Mr. Mark Schneider, defendants attorney). Provide research, consultation and expert opinion regarding health and environmental impacts of trichlorethane and trichloroethylene in groundwater. lirio David L. Eaton Page 4 CONSULTING ACTIVITIES Heller, Ehrman, White & McAuliffe, Seattle, WA (Mr. Steve Anderson, defendants attorney). Provide research, consultation and expert opinion regarding health impacts of variety of pollutants associated with Superfund hazardous waste site. H elsell, Fetterman, Martin, Todd & Hokanson, Seattle, WA (Mr. Mark Rising, defendants attorney); Provide research, consultation and expert opinion regarding health and environmental impacts of waste products from steel production process. Carney, Stephenson, Badley, Smith, Mueller & Spellman, P.S. (Mr. Dave Bratz, defendant's attorney) Provide research, consultation and expert opinion regarding toxicologic effects of methyl ethyl ketone peroxide. Rivkin, Radler, D unne & Bayh, Uniondale, NY (Mr. Edward T. Finneran, III, defendant's attorney); Provide research, consultation and expert opinion regarding health and environmental impacts of dioxane in groundwater. W illiams, Kastner & Gibbs, Seattle, WA (Mr. Mark Myers, defendants attorney). Provide research, consultation and expert opinion on health impacts of occupational exposure to wood products containing copper arsenate preservative. Huey, Guilday, K uersteiner & Tucker, P.S., Tallahassee, FL (Mr. Ralph DeMeo, defendant's attorney). Provide research, consultation, and expert opinion on health risks of EDB in groundwater Snohomish County, Solid Waste D ivision (Ms. Karen Nahkajuri) Provide testimony regarding extent, or lack, of potential public health risks from chemicals at planned municipal landfill. 19071 i R fi lm Arch Occup Environ Health (1990) 62:513-520 -ryK .? OccupationalEnvironmental Health Springer-Verlag 1990 A meta-analysis of exposure to phenoxy acid herbicides and chlorophenols in relation to risk of soft tissue sarcoma Christine Cole Johnson, Marcia Feingold, and Barbara Tilley i ! Division of Biostatistics, Research Epidemiology, and Computing, Henry Ford Health System, 23725 Northwestern Highway, Southfield, Michigan 48075, USA Received January 10 / Accepted September 27, 1990 : Summary. This analysis combines the results of the epi demiologic studies which appeared in the literature from 1979-1987 concerning the associations between phenoxy acid herbicides (PH) and chlorophenols (CP) and the oc currence of soft tissue sarcoma. Cohort studies and casecontrol studies are considered separately. After specify ing criteria for inclusion, the total number of deaths due to cancer occurring in selected cohorts of people exposed to PH or CP was determined. The proportion of soft tis sue sarcoma cases observed versus expected in the co hort studies, calculated using United States SEER regis try data, resulted in a proportional mortality ratio of 3.5 (95% confidence interval of 0,7-10.3). Excluding the in- ^ [dal reports, the selected case-control studies had homo geneous risk estimates with a summary odds ratio of 1.1 (95% confidence interval of 0.9-1.4). Other attempts to assess the combined data did not yield results indicating an association. These meta-analyses do not provide strong support for an association between the specified herbi cides and soft tissue sarcoma. assessing summary measures and the consistency of re sults. Methods To our knowledge, we obtained and reviewed all of the papers published from 1979-1987 dealing specifically with epidemiologic studies of PH and CP and STS risk. A collection of papers and re ports were provided by Dynamac Corporation to the authors. In addition, the authors conducted a comprehensive literature search using MEDLINE for all years through I9S6. and by review of pa pers cited in the collected material. The 1986 and 1987 issues of en vironmental and occupational health journals were reviewed to ob tain recent articles that might not have otherwise been included. All articles with mention of PH, CP. or dioxin and cancer were evaluated. The analyses presented in this paper include only co hort and case-control studies; case series reports were excluded be cause of the lack of comparison groups, and proportional mortality studies were excluded because the data provided in these studies could not be accommodated in our analytical approach. Key words: Epidemiology - Soft-tissue neoplasms - Di oxins - Pesticides Introduction In 1979, case-series and case-control epidemiologic stud ies from Sweden indicted phenoxy acid pesticides (PH) and chlorophenols (CP) as human carcinogens. Of par ticular concern was exposure to the dioxin contaminant (2.3,7,8-tetrachlorodibenzo-p-dioxin or TCDD) of the phenoxy acid herbicide 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and chlorophenols. These reports were followed by a series of epidemiologic studies from all over the world of the risk for cancer in general (1,-2, 18, 24, 26, 28, 31, 38, 39], and for soft tissue sarcomas and lympho mas in particular, associated with exposures to PH and CP. This report presents meta-analyses of data pertaining to soft tissue sarcoma (STS), and draws conclusions by Offprint requests to: C. Cole Johnson Zt.. a ,Date_I 0 ^ I'O'' `"'ttMS*. MNELEAT10TA Cohort studies. Cohort studies of the association between PH, CP and human cancer came from two different sources; 1) industrial exposures involved in the manufacture of these products and 2) ex posures involving the application of these substances. The studies from manufacturing industries included cohorts who routinely came into contact with these substances as a part of their work as well as several groups of workers who experienced high intensity exposures of short duration due to accidents. Table 1 presents a listing of the cohort studies of exposures associated with PH and CP and STS mortality or incidence (2, 16, 24, 26, 28, 31, 36, 38, 39], A number of commentaries presented additional data on over lapping groups of subjects from these cohorts (7, 12. 19. 22, 25]. However, the subject groups selected for this analysis were mutu ally exclusive. The studies varied in exposure criteria, although all persons were considered exposed in some degree to PH or CP. Those stud ies that were more general with regard to exposure (e.g., all pes ticides) or with only a short follow-up period were excluded. Some studies had disease incidence endpoints; others measured mortal ity. As this analysis used cancer-specific and total cancer mortality, studies with only incidence data were excluded. All the studies were limited to white males with the exception that 3.2% of the Michigan study subjects were American Indian, Black or Hispanic males [26]. No age-specific person-years of follow-up were available from the reports, prohibiting a direct re-analysis. Few of the --studies presented the total number of person-years of follow-up. 2 J fi 5 14 Table 1. Summary of cohort studies and risk estimates for soft tissue sarcomajU Location [Ref.] Exposure No. of sub O /E jects of risk Estimate Sweden [2] PH Axdsstrn *~ta. ( 207 0/0.03b 95% Confidence interval Railroad workers exposed > 45 days from 1957-72: followed 1957-1978: 10 years latency incorporated Finland [28] R iiW ^ U r PH $2 1926 0/0.1 Herbicide applicators exposed 2 14 days from 1955-71; followed 1972-1980: 10 years latency incorporated Denmark [24]c Lyvl^ I I S S - 2,4-D; cresol 3390 940d 4/1.09 1/0.16 3.7 1.0-9.4 6.3 0.2-34.8 2 herbicide manufacturing plants; all employees from 1947-1982; followed 1947-1982; 10 years latency incorporated: 4 STS (2 leiomyosarcomas. one dermatofibrosarcoma of the back, one sarcoma of the retroperitoneum) " `r. v . Michigan [26] 2.4,5-T 2187 1/0.4 2.5 0.1-13.9 Chemical workers emploved 1940-1980; followed 1940^iys;; no latency incorporated: includes 61 subjects involved in a 1964 accident Sweden [18] J IWO West Virginia [38] PH 2,4,5-T 161 0/0.1 Forest workers employed > 5 days from 1954-1967; followed 1954-1978 121 l/0.05b 22.2 0.6-123.8 Chemical workers accidentally exposed and diagnosed with chloracne in 1949; followed 1949-1978; no latency incorporated; number expected for STS grouped with "other cancer sites" Federal Republic of Germany [31] T h e is t 1. 2,4,5-T Z 74 0/0.02 - - Chemical workers accidentally exposed in 1953; followed to 1979; no latency incorporated West Virginia [39] 1< 7 * 3 2,4,5-T. 884 1/0.18b 5.7 0.1-31.8 Plant chemical workers with potential exposure: > 1 year employment from 1955-1977: followed 1955-1977; no latency incorporated; number expected for STS was grouped with "other cancer sites" Sweden [36]* PH 354620 nc 7215d nc 0.9 0.8-1.0 0.8 0.3-1.9 Swedish agricultural and forestry workers in 1960 compared to other Swedish employed men; followed 1961-1979; high exposure category included those in 1 silviculture -- -- i-- i -................ . -- r~ - -- * Cl = confidence interval; O /E = observed/expected; nc = not calculated as necessary data not presented; indicates that zero (0) cases - r were observed so a risk estimate was not calculated: PH = phenoxy acid herbicides; 2,4-D = 2,4-dichlorophenoxyacetic acid; 2,4,5- T = 2.4,5-trichlorophenoxyacetic acid; STS = soft tissue sarcoma. Reference 2 superseded an initial study [1]. Reference 26 superseded three earlier reports [4, 8, 27] b Estimated expected number based on the information that 0.5% of the total cancer cases ascertained in the study could be expected to have STS, as calculated from SEER registry data [37] * These risk estimates related to cancer incidence rather than mortality d Higher exposed subgroup 19073 !t 515 Table 2. Summary of case control studies and risk estimates for soft tissue sarcoma Location [Ref.] Exposure' No. of cases No. of Risk 95% Confidence Comments1' controls estimate interval Sw eden [16] 1 l*T 7 l CP& PH CP Sweden [10] H /tlcis** e.i~ * New Zealand [30] S vm'tH - CP& PH PH CP PH CP 52 206 46 201 40 193 5.7` 2.7-11.8 5.3e 2.4-11.5 6.6e 2.1-20.9 110 219 4.7C 2.3-9.5 92 207 17.0C 2.1-140.0 96 214 3.3C 1.3-8.6 82 92 1.2 0.6-2.6 82 92 1.5 0.5-4.7 * New York State [13] <fZ4 Vietnam service Agent orange England [3] * l yta i a u a M cm u rt Agricultural and forestry occupations Farming11 281 252 1961 1961 281 252 1961 1961 (lalv (33) V tuV is + * i . 3 Agricultural occupations in high herbicide use area: males & females Surviving women with high herbicide exposure 68 158 20 56 Washington [35]u Occupations with 128 694 f t * . 1 . 1 1 * 7 exPsure to PH Occupations with exposure to CP 128 694 Kansas[17] H r* ? < * tei(. Fanners Hx chioracne Fanners Herbicide use (mostly 2,4-D) 128 694 128 694 133 948 133 948 0.5 0.2-1.2 0.7 0.3-1.8 1.1 0.8-1.5 1.7` 1.0-2.9 1.8 1.0-3.2 2.7 0.6-11.1 0.8 0.5-1.2 1.0 0.7-1.5 1.3 0.8-1.9 3.3 0.8-14.0 1.1 0.7-1.6 0.8 0.5-1.3 Hospital cases admitted from 1970-1977, general popula tion controls, extensive matching. Positive exposure if > 1 day and 5 years preceding diagnosis Incident cases in 5 counties from 1974-1978; positive exposure if > 1 day at least 5 years preceding diagnosis Incident cases 1967-1980, registry cancer controls matched on year of diagnosis and age. Unmatched in analysis. Positive exposure if > 1 day and 5 years preceding diagnosis Incident cases from 1962-1980, 18-29 years old; controls from motor vehicle license listing and mortality files, partial pathology review Incident cases > 15 years old from 1968-1976. Cancer con trols. Occupation at diag nosis from registry. Matching factors not stated but matched analysis Incident cases S 20 years old from 1981-1983. Controls selected from voting list and mortality files Incident registry cases 1981-1984, 20-79 years; popu lation control group matched on age and vital status; interview data Incident cases from 1976-1982 > 21 years old. Controls from general population; random digit dialing. Medicare lists and mortality files: telephone interviews; pathology review * CP = chlorophenols, PH = phenoxy acid herbicides Unless otherwise noted, studies are limited to white males c P < 0.05 Could not recalculate odds ratios and confidence limits, reported as stated in manuscript For each study, relative risk estimates and confidence intervals (based on the Poisson distribution) were calculated directly from the data where available (Table 1). Many of the risk estimates were elevated, but the confidence intervals were wide and no risk estimate was statistically significant at the 0.05 level. In all these studies, exposure was relatively well documented through person nel records. However, the number of exposed subjects in each study was small, and STS's are very uncommon, with an overall age-adjusted incidence rate of 2.4 per 100,000 [37], indicating that the studies had a low power to detect elevated risk estimates. Thus, conclusive results cannot be determined from these indi vidual reports. The combined analysis for the cohort studies involved the cal culation of a crude proportional mortality ratio (PMR) and 95% 516 confidence interval across all studies assessing the percentage of all cancer deaths due to STS. The percentage of cancer deaths in the combined cohorts due to STS was compared to the same percent age calculated for white males > 20 years as derived from the com bined 1973-1981 SEER registries [37], Younger males were ex cluded from the comparison population as all of the studies were of employed or retired males. Many studies reported data or results with and without a latency period incorporated in the analysis: if so, the data taking latency into account were selected. Case-control studies. There were eight case-control studies (Table 2) which varied with respect to selection of living or deceased cases, the source of cases, and the selection of controls [3, 10, 13, 16, 17, 30, 33, 35], For the case-control studies, odds ratios and confidence limits were recalculated (with some exceptions as noted) using the normal approximation to the logarithm of the odds ratio as suggested by Woolf [34]. Most of the studies had moderately increased risk estimates (Table 2). The classification of a subject as exposed varied both by the source of data (i.e., per sonal interview or available documents) and the criteria for expo sure (i.e., job titles versus specific contact with a substance). Studies in which the appropriate data could be abstracted were used to calculate a summary odds ratio and 95% confidence inter val. The data from one 2 x 2 contingency table per study were in cluded. Subjects with unknown exposures were excluded from analysis. One study in which the raw data could not be abstracted was excluded [35]. For each study where different pesticides were analyzed together and separately, the data representing combined pesticide exposure were selected. If data using both occupational titles and specific chemical exposure were available, the specific chemical exposure data were used. If only occupational titles were used as surrogate measures for exposure, but the data were strati fied by exposure potential (e.g., residence in a high PH use county), the category with criteria defining the highest potential exposure was selected for the summary analysis. In some instances, cases and controls were matched in the re ported analyses. Sometimes the authors noted that dissolution of the matching did not affect the results and presented unmatched analyses: for others this procedure could be accomplished from data included in the manuscripts. Dissolving the matching never resulted in a substantial change in the results. Therefore, where applicable, the matching was dissolved for the studies utilized in the summary odds ratio analyses. The odds ratios from the case-control studies were plotted against the number of cases in each study. This plot was done to determine if the odds ratios tended to cluster around a central point, with larger samples more closely clustered around a single true population value (if such exists). Dose-response evaluation. For an assessment of dose-response in both the cohort and case-control studies, risk estimates are pre sented where results were derived for presumably low or general exposure categories as well as high exposure groups within a popu lation. Categories of higher exposure were defined by methods in dicating heavy exposure (spraying or mixing the material, or expo sure during an accident), clinical evidence of high exposure (a history of chloracne), a longer duration of exposure, or other mea sures as constructed and presented by a paper's authors. A second analysis o f dose-response compared risk estimates within the time period covered by a study, presuming that exposures have declined as technology has improved. Results The PMR for the combined cohort studies was 3.5 (95% confidence interval 0.7-10.3) (Table 3). The PMR using only cases confirmed by pathological review was 1.2 (95% confidence interval 0.03-6.55). -4 orTable 3. Crude proportional mortality ratio combined cohorts of white males exposed toforphSeTnSoxbvasaecdidtosn 7 chlorophenols compared to all white male deaths. 10 SEER re tries combined, 1973-1977* *,v ".Y-: [Ref.] Number Causes of death Number Cancer STS deaths deaths deaths STS deaths post review 9 12] 207 [281 1926 [26] 2187 [18] 161 [38] 121 PM 74 [39] 884 20 144 370 29 32 21 163 Observed Expected' PMR Lower limit o f 195% Cl Upper limit of 95% Cl 6 26 81 5 9 7 _35 169 0 0 1 0 1 0 1 3 0.85 3.53 0.73 10.31 0 0 0 0 lb 0 0_____ 1 0.85 1.18 0.03 6.55 s * 5 i x t 1 i V * * STS = soft tissue sarcoma; Cl = confidence interval, calculated based on Poisson distribution b The specific diagnosis was malignant fibrous histiocytoma, con firmed by expert pathology review. The duration between first ex posure and death was 30 years c Expected numbers calculated using percentage of all cancer deaths, NCI's SEER prouram, 1973-1977, white male 2 20 years, which is 0.5% for STS [37] Table 4. Data used for summary odds ratio analyses* Exposure' [Ref.] Ex posed STS cases Ex posed con trols Non- Non ex ex pose d pose d STS con cases trols OR & 95% Cl* PH & CP [16] 19 19 33 187 5.7 (2.7-11.8) PH & CP [10] 25 13 85 206 4.7 (2.3- 9.5) PH [30] 18 17 64 75 1.2 (0.6- 2.6) PH (Agent orange) [13] 7 10 245 242 0.7 (0.3- l.S) PH (mostly 2,4-D) [17] 22 192 111 756 0.8 (0.5- 1.3) Agricultural & forestry occu pations [3] 93 82 1868 1879 1.1 (0.8- 1.5) Agricultural occupations in high PH use area [33] 29 47 39 111 1.8 (1.0- 3.2) * Studies could only be included if numbers of cases and controls by exposure status could be determined from report r b Individual confidence intervals are calculated using the normal -= approximation to the logarithm of the odds ratio as suggested b y Woolf [34]; OR = odds ratio. C l = confidence interval rjj|E e STS = soft tissue sarcoma, PH = phenoxy acid herbicides, CP " j f chlorophenols -njjg The Mantel-Haenszel odds ratio combining data across all case-control studies (Table 4) was 1.4 (95% confidence limits 1.1, 1.7, P < 0.001). However, the Breslow-Day test for homogeneity revealed that fhe individual esti- 19075 CASES Fig. 1. Distribution of odds ratios from the case-control studies by the number of cases in each study mates were disparate ( P < 0.001). The Swedish studies that initially reported these associations had the highest estimates and have also received the most criticism [5-7, 14, 15]. Without the Swedish studies, the Breslow-Day test was insignificant (P < 0 .2 5 ), and the summary odds ratio was 1.1 (95% confidence interval 0.9-1.4). No as sociation was evident (%2mh = 0 .7 , P < 0 .4 2 ). 517 Figure 1 shows odds ratios from the case-control stud ies (both statistically significant and not) compared to the number of cases in each study. There is no indication that the odds ratios from studies with more cases cluster more tightly around a single population value compared to odds ratios from smaller studies. Thus there is no im plication that the studies are all samples from a single ex posure population with a single true odds ratio. Three of the seven study locations in Table 5 [24, 33, 38, 39] displayed an increased risk with increased expo sure levels. In addition, the category in the Washington State study [35] implying definite and intense exposure, as indicated by a history of chloracne, had a relatively higher estimate of 3.3. In the Michigan cohort [26], the one case which occurred was in the high exposure cate gory. The exposure categories were relatively less reli able in the remainder of the studies [13, 36]. A second approach to evaluate dose-response was based on the assumption that manufacturing processes and environments have improved over time and have been accompanied by a decrease in contaminants and lower exposure levels for workers. Figure 2 displays risk estimates against the midyear of eligibility of the cases, specifically, the midyear of the time period when a case Table 5. Summary of study results by degree of exposure1 Location [Ref.] Sweden [36] Denmark [24] Michigan [26] West Virginia [38, 39]c New York State [13] Italy [33] Washington [35] Method Cohort Cohort Cohort Cohort Case-control Case-control Case-control Degree of exposure*1 Low High General High Low Moderate High General High (accident) General High General High None (PH) Low (PH) Moderate (PH) High (PH) None (CP) Low (CP) Moderate (CP) High (CP) High (chloracne) STS risk estimate 0.9 0.8 3.7 6.3 (1 obs/0 exp) 5.7 22.2 0.5 0.7 1.8 2.7 1.0 0.6 1.0 0.9 1.0 0.9 0.9 0.9 3.3 95% Confidence interval 0.8-1.0 0.3-1.9 1.0-9.4 0.2-34.8 - 0.1-31.8 0.6-123.8 0.2-1.2 0.3-1.8 1.0-3.2 0.6-11.1 0.3-1.1 0.6-1.7 0.4-1.9 0.5-1.6 0.6-1.5 0.5-1.8 0.8-14.0 * indicates that zero (0) cases were observed, so a risk estimate was not calculated. The number of subjects in the exposure groups can be determined from Tables 1 and 3 and Reference 35 STS = soft tissue sarcoma, PH = phenoxyacid herbicides, CP = chlorophenols ' Crude risk estimates were calculated. Since expected numbers were not available in the paper, the SEER registry data [37] showing that 0.5% of the total cancer cases ascertained in the study could be expected to be soft tissue sarcomas were used to generate an expected num ber. In this paper, the soft tissue sarcoma case was misdassified in the "skin cancer" category 13078 518 RISK ESTIMATE 2 5 1------------------------------------------------ 20 - 15 10 1960 1970 1960 MIDYEAR 1990 Fig. 2. Distribution of risk estimates by midyear of eligibility for case ascertainment could be ascertained either through a diagnosis or death due to STS. All studies, including both manufacturing and application exposures, are included. Since survival for the tumors under study is relatively poor, the graph does not distinguish between cases identified at death versus diagnosis. There is no definite trend of decreasing risk estimates associated with a later case ascertainment. Discussion A major concern when combining the results of studies is the possibility that negative or inconclusive results re main unpublished and, therefore, unknown. The selec tion bias that can result may invalidate the conclusions of a collective evaluation of all published studies on a given topic. Because of the widespread use of CP and PH in agricultural and lumbering occupations and in military endeavors, such as Vietnam, studies of these agents have been conducted in a relatively politicized atmosphere. This has resulted in the publication of a number of nega tive or inconclusive reports. Thus, publication bias may not be a major problem here. The biological plausibility of dioxin contaminants be having as tumorigens has been demonstrated in several animals studies identifying this substance as a general chemical carcinogen [9, 23, 29, 32], Recent reviews con clude, however, that the evidence for carcinogenicity of PH irrespective of their dioxin contaminant is inadequate [ 11, 20, 21]. In some studies in this series, it was evident that ex posure periods overlapped with the time periods where cases could be ascertained. Most studies addressed this problem by imposing a latency period - disease had to occur after a designated time from first exposure. Even so, most studies used 5 or 10 years, which may be too short. The durations between first exposure at work and death for two pathologically confirmed STS cases were 32 and 30 years [7, 38]. It is impossible to recalculate es timates using different latency periods without the raw data. It is likely that there was misclassification for expo sure, particularly when occupational titles or interview data from subjects or their next of kin were the only indi cators of exposure. If misclassification occurred unsolcctively among the exposed versus the non-exposed. rink estimates are biased toward one. Studies in which pathological material was not re viewed probably resulted in the misclassification of some cases as STS is difficult to categorize. The pathology material from four cases occurring in three chemical worker cohorts [26, 38, 39] along with three other cases was reviewed at the Armed Forces Institute of Patholouv and by another independent pathologist [12]. Two of the first four cases were found to be incorrectly classified as STS. The two expert reviewers agreed only 71% of the time as to histological subtype and in only two of seven cases did these reviewers and the original pathologist agree. In the case-control studies reporting results from pathological reviews, a substantial number of subjects originally classified with STS were reclassified to nonSTS diagnoses, e.g. 6.3% [30], 9.3% [13], 19% [17] and 23.2% [33]. Thus, any expected numbers for STS gener ated from general population cancer registries can be as sumed to include a large number of misclassified cases. It would be misleading, therefore, to compare study co horts in which diagnoses have been corrected by pathol ogy review to general population statistics. Therefore, the unreviewed PMR in Table 3 is the more valid PMR. The cohort studies give inconclusive results even when combined (Table 3). The proportion of deaths is higher than expected; however, the confidence intervals include 1.0. One other case of STS (a confirmed malig nant fibrous histiocytoma) occurred among one of the chemical worker cohorts [26] but was not included in the reported study or on Table 3. This subject died in 1983. and the last day of follow-up for this SMR study was De cember 31, 1982. Including this case yields 4 observed (prior to pathologic review) versus 0.85 expected for a PMR of 4.71 (95% confidence limits of 1.28 and 12.05). Other cancer deaths occurring in 1983 in the study cohort are not included, but based on the total number of can cer deaths between 1940-1982 (n = 8 1 ) adding one year's worth of deaths would have little effect on the per cent of STS cases expected, and consequently the PMR. The case-control study summary results are more in formative. When the initial studies were excluded - as initial cases or clusters that stimulate a hypothesis are not counted in an individual study -- the summary odds ratio for STS was 1.1 with a tight confidence interval (0.9-1.4). No conclusions indicating a high risk or no risk can be drawn from Fig. 1, which was based on the case-control studies. Individual study results by degree of exposure pre sented in Table 5 and Fig. 2, in which year is a proxy for dose, are not strongly suggestive of a dose-response effeet. Some risk estimates for the high exposure groups are elevated, but the confidence limits are wide and in-^? elude 1.0, indicating low power and small sample sizes in ,.r; these subcategories. One difficulty with the approach in Figure 2 is that the case ascertainment period for many of the studies covered a wide range of years. The cohort results are equivocal. The case-control studies, with the exception of the earliest reports, do not support the proposed associations. The two most recent studies in Washington and Kansas appear to be well-con ducted studies that were designed to avoid the problems encountered in earlier studies (17, 35], The Kansas study did not find associations with herbicide use and STS. The results of the Washington study were conclusively negative, although there was an elevated but nonstatistically significant association of STS and a history of chioracne, indicating exposure to TCDD. In summary, this assessment does not provide strong evidence that expo sure to PH and CP is associated with the occurrence of STS. Acknowledgements. The authors gratefully acknowledge Mr. Gre gory Piazza, Ms. Cheryl Morris, and Ms. Marie A.Cuelio for ex pert manuscript preparation. This research was performed under contract to Dynamac Corporation, Rockville, Maryland, for the Environmental Protection Agency. References 1. Axelson O, Sundell L (1974) Herbicide exposure, mortality and tumor incidence. An epidemiological investigation on Swedish railroad workers. Work Environ Health 11:21-28 2. Axelson O, Sundell L, Anderrson K, Edling C, Hogstedt C, Kling H (1980) Herbicide exposure and tumor mortality. An updated epidemiologic investigation on Swedish railroad work ers. Scand J Work Environ Health 6:73-79 3. Balarajan R, Acheson ED (1984) Soft tissue sarcomas in agri culture and forestry workers. J Epid Comm Health 38:113-116 4. Bond GG, Ott MG, Brenner FE, Cook RR (1983) Medical and morbidity surveillance findings among employees poten tially exposed to TCDD. Br J Ind Med 40:318-324 5. Coggon D, Acheson ED (1982) Do phenoxy herbicides cause cancer in man? Lancet, May 8 6. Cole P (1980) Direct testimony before the Administrator of the United States of America in regard to the Dow Chemical Com pany et al. FIFRA Document No. 415, et al, Washington, DC 7. Cook R (1981) Dioxin, chloracne, and soft tissue sarcoma. Letter, Lancet 8. Cook RR, Townsend JC, Ott G, Silverstein LG (1980) Mortal. ity experience of employees exposed to 2,3.7,8-tetrachlorodibenzo-p-dioxin (TCDD). J Occup Med 22:531-532 9. Current Intelligence Bulletin 40 (1984) 2,3,7.8-tetrachloro-dibenzo-p-dioxin. DHHS (NIOSH) Publication No. 84-104 10. Eriksson M, Hardell L, Berg NO, MollerT, Axelson O (1981) Soft-tissue sarcomas and exposure to chemical substances: a case-referent study. Br J Ind Med 38:27-33 11. Expert panel report on carcinogenicity of 2.4-D (1987) Cana dian Centre for Toxicology, Guelph. Ont, Canada 12. Fingerhut MA, Halperin WE, Honchur PA, Smith A B, Groth DH, Russell WO (1984) An evaluation of reports of dioxin ex posure and soft tissue sarcoma pathology among chemical workers in the United States. Scand J Work Environ Health 10:299-303 13. Greenwald P, Kovasznay B, Collins DN, Them'ault G (1984) Sar comas of soft tissues after Vietnam service. JN C I73:1107-1109 14. Hardell L (1981) Relation of soft-tissue sarcoma, malignant lymphoma and colon cancer to phenoxy adds, chlorophenols and other agents. Scand J Work Environ Health 7:119-130 15. Hardell L, Eriksson M (1981) Phenoxy acids, chlorophenols and cancer. Review, 1981. Lakartidningen 78:2861-2862. Translated by Literature Research Co, Annandale. VA 16. Hardell L, Sandstrom A (1979) Case-control study: soft-tissue sarcomas and exposure to phenoxyacetic adds or chlorophenols. BrJ Cancer 39:711-717 17. Hoar SK, Blair A , Holmes FF, Boysen CD, Robel RJ, Hoover R, Fraumeni JF Jr (1986) Agricultural herbicide use and risk of lymphoma and soft-tissue sarcoma. JAMA 256:1141-1147 519 18. Hogstedt C, Westerlund B (1980) Cohort study of the fatality causes in forest workers exposed and unexposed to phenoxy acid preparations. Lakartidningen 77:1828-1831. Translated by Literature Research Corp., Annandale, VA 19. Honchar PA, Halperin WE (1981) 2,4,5-T, Trichlorophenol, and soft tissue sarcoma. Letter, Lancet 20. Huff JE, Moore JA, Saracci R, Tomatis L (1980) Long-term hazards of polychlorinated dibenzodioxins and polychlorinated dibenzofurans. Environ Health Perspectives 36:221-240 21. IARC Monographs on the evaluation of the carcinogenic risk of chemicals to humans. Some halogenated hydrocarbons and pesticide exposures (1986) World Health Organ. Internat'I Agency for Research on Cancer, volume 41 22. Johnson FE, Kugler MA, Brown SM (1981) Soft tissue sar comas and chlorinated phenols. Letter, Lancet 23. Kociba RJ, Keyes DB, Beyer H, et al (1978) Results of a two year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol Appl Pharmacol 46: 279-303 24. Lynge E (1985) A follow-up study of cancer incidence among workers in manufacture of phenoxy herbicides in Denmark. Br JCancer 52:259-270 25. Moses M, Selikoff IJ (1981) Soft tissue sarcomas, phenoxy her bicides and chlorinated phenols. Letter, Lancet 26. Ott MG, Olson RA, Cook RR, Bond GG (1987) Cohort mor tality study of chemical workers with potential exposure to the higher chlorinated dioxins. J Occup Med 29:422-429 27. Ott MG, Holder BB, Olson RD (1980) A mortality analysis of employees engaged in the manufacture of 2,4,5-trichlorophenoxyacetic acid. J Occup Med 22:47-50 28. Riihimaki V, Asp S, Hernberg S (1982) Mortality of 2,4-dichlorophenoxyacetic acid and 2,4,5-triochlorophenoxyacetic acid herbicide applicators in Finland. First report of an ongoing pro spective cohort study. Scand J Work Environ Health 8:37-42 29. Scientific Review Committee of the American Academy of Clinical Toxicology (1985) Vet Human Toxicol 25:434-438 30. Smith AH, Pearce NE, Fisher DO, Giles HJ, Teague CA, Howard JK (1984) Soft tissue sarcoma and exposure to phenoxyherbicides and chlorophenols in New Zealand. JNCI 73: 1111-1117 31. Thiess AM. Frentzel-Beyme R, Link R (1982) Mortality study of persons exposed to dioxin in a trichlorophenol-process acci dent that'occurred in the BASF AG on November 17, 1953. Am J Indust Med 3:179-189 32. Van Miller JP, Lalich JJ. Allen JR (1977) Increased incidence of neoplasms in rats exposed to low levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Chemosphere 9:537-544 33. Vincis P, Terracini B, Ciccone G, Cignetti A, Colombo E, Donna A , Maffi L, Pisa R. Ricci P, Zanini E. Comba P (1986) Phenoxy herbicides and soft-tissue sarcomas in female rice weeders. A population-based case-referent study. Scand J Work Environ Health 13:9-17 34. Woolf B (1955) On estimating the relationship between blood group and disease. Ann Human Genet 19:251-253 35. Woods JS, Polissar L, Severson RK, Heuser LS, Kulander BG (1987) Soft tissue sarcoma and non-Hodgkin's lymphoma in re lation to phenoxy herbicide and chlorinated phenol exposure in western Washington. JNCI 78:899-910 36. Wiklund K, Holm L (1986) Soft tissue sarcoma risk in Swedish agricultural and forestry workers. JNCI 76:229-234 37. Young JL, Percy CL, Asire AJ (1981) Cancer.incidence and mortality in the United States, 1973-77. National Cancer Insti tute Monograph No. 57 38. Zack J, Suskind R (1980) The mortality experience of workers exposed to tetrachlorodibenzodioxin in a,trichlorophenol pro cess accident. J Occup Med 22:11-14 39. Zack JA, Gaffey WR (1983) A mortality study of workers employed at the Monsanto company plant in Nitro. West Vir ginia. Proceedings of symposium on Chlorinated Dioxins and related compounds, October 25-29, 1981, Arlington, VA, pages 575-590 520 ,i Appendix A p p e n d i x Table l (data for Fig. 1) [Ref.] Odds ratio 95% confidence interval (35] 1.3 [35] 3.3 [30] 1.5 (301 1.2 [13] 0.7 (13] 0.5 (17] 0.8 [17] 1.1 [33] 1.8 [33] 2.7 [16] 5.7* [10] 4.7* [3] 1.1 [3] 1.7* 0 .8 - 1.9 0.8-14.0 0.5- 4.7 0.6 - 2.6 0 .3 - 1.8 0 .2 - 1.2 0 .5 - 1.3 0 .7 - 1.6 1.0- 3.2 0.6-11.1 2.7-11.8 2 .3 - 9.5 0 .8 - 1.5 1.0- 2.9 P C 0.05 No. cases 128 128 82 82 252 281 133 133 68 20 52 110 1961 1961 A p p e n d i x Table 2 (data for Fig. 2) Eligibility years [Ref.] Risk estimate 95% confidence interval 1972-1980 [28] 1961-1979 [36] 1947-1982 (24] 1940-1982(26] 1954-1978(18] 1949-1978 [38]** 1955-1977 [39]** 1970-1977(16] 1974-1978(10] 1967-1980(30] 1962-1980(13] 1968-1976 [3] 1981-1983 [33] 1981-1984 [35] 1976-1982 [17] _ 0.9 3.7 2.5 - 22.2 5.7 5.7* 4.7* 1.2 0.5 1.1 1.8 0.8 1.1 - 0 .8 - 1.0 1.0- 9.4 0 .1 - 13.9 - 0.6-123.8 0 .1 - 31.8 2 .7 - 11.8 2 .3 - 9.5 0.6- 2.6 0 .2 - 1.2 0 .8 - 1.5 1.0- 3.2 0 .5 - 1.2 0 .7 - 1.6 * P < 0.05 * Crude estimates calculated based on the estimate that 0.5% of the total cases could be expected to have STS as calculated from SEER registry [10] 19079 Toxicology Human Exposure to 2,3,7,8-TCDD and Risk of Cancer Eric S. Johnson, M.B.; B.S., Ph.D. ABSTRACT of 29 pesticides evaluated by the International Agency for Research on Cancer (IARC) were judged to be animal carcin ogens.33 This review does not consider risk of cancer among ground troops in the Vietnam war since it has not been shown that this group had levels of TCDD exposure above back ground. However, persons in the military who sprayed her bicides in Vietnam are considered since their exposure to TCDD has been well documented. Most of the evidence for the carcinogenicity of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) in humans has centered around whether it causes malignant lymphomas-(ML) and softtissue sarcomas (STS). A critical review of the literature in dicates that the evidence does not support a causal role for TCDD in the etiology of ML. For STS, the evidence does not specifically incriminate TCDD either, although there is room for doubt. Cancers of other sites, particularly of the respiratory system and thyroid which were found to be statistically sig nificantly in excess in either of the two largest studies of com bined cohorts of occupationally exposed workers, were iden tified as candidate tumors for which a possible etiological role of TCDD might need investigation in future studies. Key W ords: 2,3,7,8-tetrachlorodibenzo-p-dioxin, TCDD, phenoxyacetic acids, chlorophenols, soft-tissue sarcoma, ma lignant lymphoma, cancer. I. INTRODUCTION There have been suggestions from case reports that exposure to phenoxyacetic acid herbicides, chlorophenols and their con taminant dioxins, especially 2,3,7,8-TCDD, may be associated with the development of malignant lymphomas (ML) and softtissue sarcomas (STS).1-10 A large number of analytic epide miological studies have investigated this putative association. Many studies examined the risk of ML among agricultural workers or among persons exposed to different groups of pesticides in general, but did not examine specific pesti cides.11-34 A majority of these studies have shown an associ ation between non-Hodgkin's lymphoma (NHL) or Hodgkin's disease (HD) and agricultural practice, or groups of pesticides such as insecticides and herbicides.11-23 Since these studies did not investigate specifically the role of phenoxy herbicides/chlorophenols/dioxins, they are not considered in this review. How ever, they leave open the question of which of the numerous exposures prevalent in agricultural occupational populations (including specific insecticides, herbicides, fungicides, fumi gants, rodenticides, fertilizers, solvents, other chemicals, and oncogenic viruses of cattle and chickens) is responsible indi vidually or jointly for this association. It is to be noted that phenoxy herbicides and chlorophenols are just 2 of the 600 to 1100 pesticides or active pesticide ingredients agricultural workers could be exposed to occupationally, and that 18 out II. CASE-CONTROL STUDIES OF MALIGNANT LYMPHOMAS Ten distinct case-control studies have examined the risk of ML from exposure to phenoxys/chlorophenols,36-30 Table 1. Four of the ten studies observed a statistically significant as sociation in the study population as a whole,37-39'46 and three more studies reported a statistically significant association only in a subset of the study population.42-43'4Thus, 70% of casecontrol studies which have investigated phenoxy acids and chlorophenols have reported some form of a statistical asso ciation between these compounds and ML. Two of the ten studies reported a statistically significant association with NHL in general,46or skin NHL in particular,43 but did not indicate whether the phenoxy acids that subjects were exposed to included 2,4,5-T. Eight studies provided in formation on possible TCDD exposure either specifically or by implication.36'38'39'42'44'45'4-30 Only one out of these eight studies reported a statistically significant association with phenoxys containing TCDD.38 Thus, at least 70% of the total of ten case-control studies did not show a statistically significant association between ML and exposure to 2,3,7,8-TCDD. III. CASE-CONTROL STUDIES OF SOFTTISSUE SARCOMAS There have been 15 distinct case-control studies of STS,39-j'48'31-68'j'able 2. One study did not find an association with Agent Orange (2,4,5-T and 2,4-D in a 50:50 ratio) ex posure in the military in Vietnam among Vietnam veterans, or with exposure to herbicides outside of the military.60 Another reported a statistically significant 1,7-fold risk associated with farming, but no increased risk was observed among other ag ricultural workers with possible exposure to phenoxy acids/ chlorophenols.31 These two studies are not considered further since they provide no information specifically on phenoxy/ chlorophenol exposure. Of the remaining 13 studies which did, E ric S. Jo h n so n , M .B .; B .S ., P h .D ., Environm ental and M olecular Epidem iology S ection, Epidem iology B ranch, Division o f Biometry and Risk A ssessm ent, National Institute of Environmental Health Sciences, P.O . Box 12233, Research Triangle Park. NC 27709. 1992 Ex. 1 .Date 451 Witness. A-taTT JOANNE LEAT10TA 1980 Critical Review s In -- >16 1 ase-Control Studies of ML and Exposure to }henoxy Acids or Chlorophenois D isease E x p o su re O dds ratio R e f. fH L 4L 4HL iD 4HL 4H L (skin) NHL (all) NHL NHL NHL NHL NHL HD 2 ,4 ,5-T PCP P henoxys ex D & M only* Phenoxys C h lo ro p h e n o is 2,4-D & MCPA 2 ,4 ,5 -T 2 ,4 -D Phenoxys Phenoxys Phenoxy acid Phenoxy acid C h lorophenois Phenoxys C h lorophenois 2,4-D 2,4,5-T Phenoxys (E & L > 15 years)' Farm er Sprayed forests with herbicides 2 ,4 ,5 -T 2,4-D 2,4-D* 2,4-D-4 Phenoxys C h lo ro p h en o is 2,4,5-T (gorse) Phenoxys C h lorophenois 2,4,5-T (gorse) Phenoxys C h lo ro p h en o is A gent Orange Phenoxys C hlorophenois Agent Orange 1.8 1.86* 3.5* 4.8* 4.3* 33.7* 1.36 2.6* 3.8 4.9* 10.0* 1.3 1.2 0 .9 1.0 0.7 1.0 1.7* 1.3* 4.8* 1.6 1.5 1.1 1.8* 1.4 1.3 1.4 1.0 1.4 0.9 0 .9 '' l.lb 1.1 0.9 l.l6 0.8 3 6 , 37 38 39, 40 46 43 48 41, 42, 44 45 49, 50 Statistically significant. * Calculated from publication. * A djusted for organophosphate exposure. * Adjusted for fungicide exposure. * Phenoxys excluding 2,4-D and M CPA only. ' E xposure and latency > IS years. nly 4 showed a statistically significant association with exosure to these compounds in the study group as a whole.52-55-57-68 Tiree more studies showed a statistically significant association i a subset of the study population.48-53,66 Thus, a little over alf of the case-control studies investigating exposure to phenxy/chlorophenols reported a statistically significant associaon in one form or another between this exposure and STS. ven studies provided some specific information on exObure to 2,3,7,8-TCDD or TCDD-containing phenoxys/chlo- Table 2 Case-Control Studies of STS and Exposure to Phenoxy Acids or Chlorophenois E x p o su re O dds ratio R ef. Phenoxy acids C h lo ro p h en o is N on-2,4,5-T Phenoxys 2,4,5-T and other phenoxys Phenoxys N on-2,4,5-T Phenoxys 2 ,4 ,5 -T Phenoxys during 1950s 2,4,5-T during 1950s PCP (high grade) Non-TCP chlorophenois (high grade) Phenoxys C h lo ro p h en o is 2,4-D or M CPA Phenoxys Phenoxys 2,3,7,8-TC D D 2,3,7,8-TC D D Chlorophenol (low grade) Chlorophenol (low grade) Chlorophenol (high grade) Chlorophenol (high grade) C h lo ro p h en o is Phenoxys Agent Orange Phenoxys (live men) Phenoxys (live w om en) Phenoxys(dead men) Phenoxys (dead women) Phenoxys (live w om en, age < 7 5 .years, exp. 1950-1955) Phenoxys C h lo ro p h e n o is H igh level phenoxys (w ith Scandinavian names) High and low chlorophenol (with Scandinavian names) Lumber grader Agent O range, dioxin, or 2 ,4 ,5 -T Phenoxys 2 .4 ,5-T Phenoxys C h lo ro p h en o is Phenoxys Chlorinated dioxins (tetra- and higher) V ietnam veterans Agent Orange H erb icid es Pesticides Insecticides Chem ical w ork or phenoxys/ chlorophenois 6.8* 3.3* 4 .2 17.0* 1.3 0 .6 1.8 2.4* 2.9* 3.9* 5.3* 5.3* 6.6* 5.7 3.3* (Pop.) 2.2 (Cancer) 3.5* (Pop.) 3.1* (Cancer) 1.0 (P o p .) 1.0" (Cancer) 0.3 (Pop.) 0.5 (Cancer) 1.4* 0 .7 0 .7 0 .9 2.7 0.2 1.1 15.5* 0.8 1.0 2.8 7.2* 2.7* 0.7 No association No association 1.4 1.3 0.7 No association 0.8 No association No association No association No association 1.8 (Cancer) 52 53 55, 56 57 68 66 48 54 39, 40, 58, 59 61-63 64 65 60 67 190152 Volume 21, Issue 6 Toxicology Table 2 (continued) Case-Control Studies of STS and Exposure to Phenoxy Acids or Chlorophenols E xposure Chem ical work or phenoxys/ chlorophenols G ardeners G ardeners P h e n o x y s/c h lo ro p h e n o ls G ardeners, groundsmen Foresters, woodmen Acricultural workers Agricultural & forestry Farm ers, farm m anagers, mar ket gardeners ' O dds ratio 1.6 (Pop.) Inf. (Cancer) 4.1* (Pop.) 4 .2 (Pop.) 0.7 1.0 1.0 1.2 1.7* N ote: Pop. = population controls; Cancer = cancer controls. * Statistically significant. b Calculated from publication. rophenols, or stated that exposure to phenoxys may include 2.4.5- T.39,48,52-55,57,61-66 Of these 11 studies, only 3 observed a statistically significant association with these compounds in the study group as a whole.52,53-57A further two studies showed such an association in a subset of the study population.53,66 Thus, less than half of the case-control studies investigating TCDD exposure have reported a statistically significant asso ciation of any sort between STS and TCDD exposure. IV. COHORT STUDIES OF WORKERS IN MANUFACTURING/FORMULATING PLANTS Eight independent cohorts of persons with possible exposure to TCDD during the manufacture/formulation of 2,4,5-T or 2.4.5- TCP have been published to date,69-37Table 3. Five were small and obviously of inadequate size to evaluate risk of STS or ML.69-73,83,84,36 Three of these did not record a single case of STS or ML;69-73 one of them recorded one case each of STS and HD;83 and one observed a nonstatistically significant two fold risk of NHL, but recorded no case of STS or HD.86 Of the remaining three larger studies,79,80,83,87 none reported a sta tistically significant excess of STS or ML in the study group as a whole, although one study79,80 reported a statistically sig nificant excess of STS in the subset of workers with chloracne based on two cases, one of which on histological review was found not to be a case of STS. Another study87 reported a statistically significant excess of STS (based on four cases) in workers in the shipping department of the plant where exposure to several other chemicals occurred, as well as phenoxy acids, but not among workers involved in phenoxy production. Fur thermore, MCPA was the major phenoxy acid produced in this plant, although limited amount of exposure to 2,4,5-T and 2.4.5- trichlorophenol also occurred. In this same cohort, a statistically significant risk of ML was observed not among workers engaged in the production of phenoxy acids, but rather in workers manufacturing other chemicals. Isolated reports of a statistically significant risk of cancer of the bladder,84 hemopoietic/lymphatic systems,78 prostate,80 rectum, uterine cervix, lung87have been made on one occasion only, and these reports have not been confirmed by the other studies. Two studies reported a significant excess of stomach cancer;71,80 however, in one of them,80 none of the stomach cancers occurred among the subset of workers with chloracne, although 0.4 were expected. V. COHORT STUDIES OF WORKERS SPRAYING/APPLYING PHENOXY HERBICIDES - Cancer mortality (Table 4) and incidence (Table 5) have been studied in 15 distinct cohorts of sprayers/applica tors.85,88-107 None of these studies reported a statistically sig nificant increased risk of STS, and only one reported a statis tically significant risk of 1.4 for ML.105 Of the 15 cohorts, only 10 specifically reported possible exposure to 2,4,5-T with potential exposure to TCDD.88-102Eight of these were too small to evaluate risk of STS or ML adequately, and none reported a case of either tumor.88-94,96,97,102 Of the remaining two larger cohorts,95,101 none reported a statistically significant risk of STS or ML. There were also isolated reports of statistically significant risks of cancer of the testicles,101,102 skin,93,102 and lung94 among workers in cohorts with the potential for exposure to TCDD. In cohorts not specifically identified as having a potential for TCDD exposure, isolated cancers reported to be significantly in excess include cancer of the nose,85 bladder,103skin,105 and lung.104,106,107 VI. POOLED COHORT STUDIES IN REGISTRIES The results of two registries from the International Agency for Research on Cancer (IARC) and the U.S. National Institute of Environmental Health Sciences (NIEHS) and the National Institute of Occupational Safety and Health (NIOSH) U.S.A., which have pooled together a large number of occupational cohorts including many of the published ones, have recently been reported,108,109Table 3. The NIOSH registry109,110studied workers with possible exposure to TCDD during the manu facture/formulation of 2,4,5-T, 2,4,5-TCP, and hexachlorophene, whereas the IARC/NIEHS registry108 includes workers exposed to all types of phenoxy acids and chlorophenols both during their manufacture and use, and thus consisted of both TCDD-exposed and -unexposed subjects. A statistically significant excess of NHL or HD was not 1992 453 Critical Review s In Table 3 Cohort Studies of Workers in Manufacturing/Formulating Plants -- Observed Numbers of Deaths/Cases, and Relative Risks (RR) S tu d y Com pound T otal cancer 2,4,5-T 2,4,5-T 2,4,5-TC P Zober Update of Cohort 1 R eferen ces Cohort 2 71 and 72 Cohort 3 Chloracne, etc. 2,4,5-T 2,4-D MCPA TCDD 2 .4 .5 - T 2 .4 .5 - T C P Update o f Reference 76 2,4-D 2 ,4 ,5 -T MCPA Update o f References 76 and 77 2 ,4 ,5 -T 2,4-D C h lo ro p h en o ls (chloracne) Update o f Reference 81 2,4,5-TC P 2.4.5-T 2 .4 .5 - T C P MCPA, 2,4,5-T 2,4,5-T 2 ,4 -D /B /P M CPA/B/P, 2,4,6- TCP Lynge M ales MCPA, 2,4-D , 2 ,4 ,5 -T F em ales N IO SH Com bined Study Fingerhut 2,4,5-T 2,4,5-TCP, PCP, hexachlorophene 1ARC/NIEHS Com bined Study K ogevinas 2 ,4 ,5 -T /T P , 2,4,-D /B/P, M CPA/B/P, PCP, 2,4,5-TC P. 2 .4 .6 - T C P , 2 .3 .4 .6 - T C P 8 2 7 17 9 11 3 16 1 3 61 81 26 95 6 5 9 9 297 37 159 49 265 505 ' Statistically significant. 'o n -H o d g k in s ly m p h o m a and H o d g k in 's d ise a se co m b in ed . RR 1.2 -- 1.7 1.7 1.3 1.7 0.5 1.4 1.9 1.0 1.0 1.2 1.0 0.7 0.6 1.0 0.8 1.0 1.0 1.0 0.9 1.2 >1 STS cases 0 0 0 0 0 0 0 0 0 1 1 1 0 2 1 2* I 1 1 0 5 4 0 4 4 RR 5.0 20.0 1.1 2.7 3.7" 3.4 2.1 NHL cases 0 0 0 0 0 0 0 0 0 RR 1 HD cases 0 0 0 0 0 0 0 0 0 RR Ref. 69 70 71 72 73 74 0 5 2.4 0 1 5 1.9 1 2 1.8 1 75 76 77 78 6 2.1 0 I 0 80 81 0 0 0 2 0.4 2 2.3 0 I 0 1 0 79 83 84 85 86 T 1.3 b 87 l6 b 10 1.4 3 1.2 109,110 11 7 < 1 108 454 Volume 21, Issue 6 Toxicology Table 4 Cohort Mortality Studies of Users (Sprayers and Applicators) -- Observed Numbers of Deaths and Relative Risks Com pound 2,4-D 2 ,4 ,5 -T 2,4-D , 2,4,5-T 2,4-D /P, 2 ,4 ,5 -T /P 2,4,-D , 2,4,5-T U pdate 2,4-D , 2 ,4 ,5 -T 2,4,-D , 2,4,5-T -- -- M CPA, 2,4-D 2,4-D , 2 ,4 ,5 -T 2,4-D , 2 ,4 ,5 -T 2,4-D , 2 ,4 ,5 -T T otal cancer cases 26 5 18 2 6 13 84 47 232 297 169 12 6 RR STS 0.7 0 0.8 0 1.1 0 0.7 0 1.1 0 0.9 0 1.1 0.8 -- 1.1 -- 1.0 1 --1 0.8 1 0.9 0 RR NHL RR HD 00 00 00 00 01 00 ---- ---- ---- 1.4 3* 0 .9 a 00 00 * N o n -H o d g k in 's ly m phom a and H o d g k in 's disease com bined. R ef. 88 89 90 91 92 93 94 103 104 85 95 96 97 observed in either of these two studies. In fact, in the NIOSH study, the risk of NHL in workers exposed to TCDD for 1 or more years was 0.93, while in the IARC/NEEHS registry, the risk for exposure to phenoxys/chlorophenols was 0.97. Both registries reported a nonstatistically significant two- to three fold risk of STS among persons with the potential for exposure to TCDD, based on fou r cases each. In the NIOSH registry, STS risk was statistically significant in workers exposed to TCDD for 1or more years when a latency period of 20 or more years was allowed for. (These workers had an estimated mean serum TCDD level at the time of last exposure of 3600 ppt.) However, the workers had spent much longer periods of employment exposed to chemicals other than 2,4,5-T, 2,4,5-trichlorophenol, or TCDD. Moreover, only 2 of 12 plants reported the total of 4 cases of STS (2 in each plant). Two of the four cases (one from each plant) were sub sequently found not to be STS on histology review. Moreover, in one of the plants, there was evidence that the risk given for STS may have been overestimated for the following reasons: (1) a proportion of the person-years of observation for persons known not to have died from STS was excluded from the analysis; (2) 139 workers with chloracne, and thus possible high TCDD exposure representing at least 24% of the cohort, were not included in the analysis, and it was known that no death from STS as the underlying cause of death had occurred among them; (3) 45% of the cohort who were alive were ex cluded from the dose-response analysis. Unfortunately, it is not known by how much the STS risk given in the paper was therefore overestimated because of these problems. It is esti mated though that substantial risk would still remain. No cases of STS occurred among workers exposed to TCDD for less than 1 year, even though for those with a 20-year latency it was estimated that their mean serum level of TCDD at the time of last exposure was possibly around 640 ppt. However, inter pretation is limited because of insufficient power in this subgroup. Therefore, although the NIOSH study was a welldesigned and carefully executed study, and of significant im portance since it involved the largest number of workers with the highest TCDD exposure ever studied, the results with re gard to STS are tentative because of the paucity of histologi cally confirmed STS (total = 2), insufficient statistical power, and problems mentioned. In the IARC/NIEHS registry, there was no difference in risk of STS among phenoxy/chlorophenol-exposed persons with a potential for TCDD exposure and those without (twofold risk), and the increased risk was seen in the sprayer cohorts but not 1992 455 Critical Review s In Table 5 Cohort Cancer Incidence Studies of Users (Sprayers, Applicators) -- Observed Numbers of Cases and Relative Risks T o ta l cancer Com pound cases RR STS RR NHL RR HD R ef. 2,4-D , 2 ,4 ,5 -T 2,4-D , MCPA 2,4-D , 2,4,5-T 2,4-D , 2,4,5-T 2,4-D , 2 ,4 ,5 -T 2,4-D , MCPA 2,4-D , 2,4,5-T -- 2,4-D , 2 ,4 ,5 -T 2,4-D , 2 ,4 ,5 -T 26 0.7 0 0 0 98 30 0 0 0 106 8 0.8. 0 0 0 89 8 1.6 -- -- -- 91 35 0 1 0 93 169 1 2 2 107 558 0 .9 6 0 .9 21 1.0 11 9 9 -1 0 1 631 0.7 -- 21 1.2 1 45` 1.4" 105 1 0 102 10 0 0 2 97 1 Non-Hodgkin's lymphoma and Hodgkin's disease combined. b Statistically significant. a the production cohorts in which TCDD exposure would be xpected to be highest. The observation of similar increased isk of STS in both TCDD-exposed and TCDD-unexposed ersons raises the possibility that the increased risk of STS een in the NIOSH study may be due not to TCDD but to other oncomitant exposure(s). Thus, caution should be exercised in aterpreting these findings at this time. Thyroid cancer mortality was statistically significantly inreased for workers exposed to phenoxys/chlorophenols in the ARC/NIEHS registry with four observed deaths, and a fourold increased mortality, both for sprayers and production /orkers. Risk was highest in workers with probable TCDD xposure. Unfortunately, the NIOSH study did not provide lformation on this site. A statistically significant increased risk of all cancers com ined was reported in the NIOSH registry, and the risk inreased with duration of exposure and latency. Also, a statiscally significant excess of respiratory cancer was observed at igher exposures. Slightly elevated risks of all cancers, and ancer of the lung were also reported in the IARC/NIEHS igistry, however, they were not statistically significant, and o dose-response relationship was evident. A mortality study of the Seveso population in Italy exposed y environmental contamination as a result of an industrial cf'-'J".nt has been published.1" Unfortunately, follow-up was for the first 10 years since the incident, hence a sufficient latency period has not been accrued. Cancer of the corpus uteri was statistically significantly increased in women residing in the area where dioxin contamination was lowest. SMRs for certain cancers were highly elevated; these were soft-tissue sarcoma, HD, gallbladder and biliary duct, peritoneum, mel anoma, pleura, other endocrine and thyroid, but they were based on one to three deaths only, and there were no consistent patterns that might help relate their occurrence specifically to TCDD at this time. It is pertinent to note that this cohort is not plagued as much as occupational cohorts by the issue of other concomitant exposures and thus might be of importance in the future in helping toward determining whether any ob served risk of cancer is due to TCDD or other exposures. VII. ASSESSMENT OF EVIDENCE A. Prelude Any evaluation of the evidence for a causal association be tween 2,3,7,8-TCDD and cancer in humans must take into account the following: 1. Occupational exposure to TCDD occurs either during the manufacture of phenoxy acids, chlorophenols, and hexachlorophene, or during the use of these compounds 56 Volume 21, Issue 6 . 9 0 ? '*sj u* Toxicology (spraying or application). Occupational exposure to TCDD is on average considerably much higher in manufacturing workers than in sprayers and applicators, e.g., in the NIOSH study, the mean serum TCDD level in a sample of manufacturing workers at the time exposure ceased was estimated to be 2000 ppt with a range of up to 32,000 ppt;110the corresponding highest serum TCDD level ever recorded in a sprayer was <1600 ppt."2 2. In virtually all the case-control studies, for practical pur poses, risks given were for exposure during the use of phenoxy acids and chlorophenols, rather than during the manufacture of these substances. Hence, risks reported in case-control studies and cohort studies of sprayers/ applicators should be expected to be lower than those given for exposure in manufacturing plants. 3. Workers exposed to phenoxy acids/chlorophenols are concomitantly exposed to a wide variety of different pes ticides. For example, in the NIOSH study, it was stated that TCDD-exposed workers were also potentially ex posed to some of the thousands of chemicals handled in the factories concerned, and in fact had spent five times as long exposed to these other chemicals than to TCDD. Similarly, every single individual who has sprayed phen oxy acids has also sprayed other chemicals. Moreover, some of these other chemicals are well-known causes of STS or ML in humans or animals. For example, the insecticide ethylene dibromide, which is used as a fu migant for various grains (com, wheat, rice, etc.), causes hemangiosarcomas in animals;113 similarly, arsenicals widely used as insecticides, herbicides, and wood pre servatives cause angiosarcoma of the liver in humans and possibly lymphomas in humans and animals;113 vinyl chloride, which causes angiosarcoma of the liver and other sites and lymphomas in humans and animals, has been used as a pesticide aerosol propellant.113-114 Man ufacturing plants producing phenoxys and chlorophenols not only produce arsenicals and ethylene dibromide, but also make p-aminobiphenyl, a known cause of angio sarcomas in animals.84113 This means that an excess occurrence of STS or ML in sprayers or workers in pesticide manufacturing plants will not be unexpected. The issue would be to identify which of these chemicals is (are) responsible. In case-control studies in par ticular, the demonstration of an excess risk in sprayers or ag ricultural workers who had used phenoxys/chlorophenols is by itself not enough to conclude that these compounds are the cause of the excess. To determine this, further analyses must be done which attempt to obtain a risk estimate for exposure to phenoxys/chlorophenols/dioxins while controlling for other candidate chemicals shown to be, or likely to be, associated with the excess. It is a fact that every single case-control study o f M L in civilian populations included in this review has re ported a statistically significant association with one or more concomitant exposures; these include DDT (RR = 1.5), chloramben (RR = 2.2);36 methyl bromide (RR = 2.8), insecticides (RR = 1.9);37 asbestos (RR = 2.0), organic solvents (RR = 2.4);38 fungicides (RR = 2 .1 ), triazines (RR = 2.5), insec ticides (RR = 2 .8 ), amides (RR = 2.9), trifluralin (RR = 12.5);39 organophosphates (RR = 1.9), diazinon (RR = 2.0), chlordane (RR = 2.1), malathion (RR = 2.2 ), dyfonate (RR = 2.4);42 solvents (RR = 1.9), creosote (RR = 9.4 and 10.7);46 solvents (RR = 1 . 4 and 3.4);43 solvents (RR = 1.4), lead arsenate (RR = 1.6), DDT (RR = 1.8);4* animal on cogenic viruses -- meat works (RR = 1.8).4445 It is noted that this list is made up predominantly of pesticides. In this situation, it is reasonable to say that an observed excess risk cannot be attributed to exposure to phenoxys/chlorophenols/ dioxins unless reasonable attempts have been made to control for these other exposures. It is pertinent to note that investi gators have not explored the role of these other exposures to any significant degree in case-control studies of STS as was done for ML, and in the few instances this was attempted, there was a suggestion also that the pattern was the same, although mostly, statistically significance was not achieved because of the smaller sample size of STS studies.40-48-52--53-35-57-63-63-47 B. Malignant Lymphoma At least 70% of all case-control studies of ML and phen- oxyacetic acid/chlorophenol exposure do not show a statisti cally significant association for exposure to 2,3,7,8-TCDD. Only one case-control study of ML, conducted in Sweden, reported a statistically significant association with exposure to TCDD-containing phenoxy acid.38Paradoxically, in that study, although based on small numbers, the crude odds ratio for exposure to 2,4-D and MCPA only was calculated to be 33.7 (p <0.0001) as compared with 4.8 for exposure to all phenoxy acids combined including the TCDD-containing 2,4,5-T. In fact when subjects exposed to 2,4-D and MCPA only, and chlorophenols were excluded, the odds ratio for mixed expo sure to phenoxy acids (TCDD and non-TCDD phenoxys) was only 3.5 (p <0.0001). This paradox was not limited to this study only -- in the study by Hoar et al. ,39 the odds ratio for 2,4-D exposure was 2.6 (p <0.05) and only a nonsignificant 1.3 for 2,4,5-T; in another study by Hoar Zahm et al.,42 the odds ratio for 2,4-D, controlling for fungicide exposure, was 1.8 (p <0.5) and for 2,4,5-T, a nonsignificant 1.6; in the studies by Cantor et al.36 and Everett et al.,37 the odds ratio for pentachlorophenol (PCP) exposure was 1.9 (p <0.05) and for 2,4,5-T, a nonsignificant 1.8. In the remaining four studies which provided information for 2,4,5-T and non-2,4,5-T phen oxys, neither was associated with an increased risk of ML.44-43-48-30 These data therefore do not support a causal role for 2,3,7,8-TCDD in the etiology of ML- For these data to be consistent with a causal role for TCDD, 1992 457 13038 Critical Review s In -e would have to entertain the possibility that dioxins/furans ier than 2,3,7,8-TCDD, which occur in both 2,4,5-T and non-2,4,5-T phenoxys do so at toxic equivalent concentrations greatly in excess of the concentration at which TCDD exposure occurs in 2,4,5-T. However, a study of dioxin and furan con centrations in older preparations of 2,4-D and 2,4,5-T in Swe den at a detection limit of 0.01 to 0.05 ppm does not support this possibility since the highest concentration of any isomer was that of TCDD (1.1 ppm) which was found only in 2,4,5T preparations. No other dioxin isomers were detected, and only tetra- and penta-dibenzofurans were detected at a con centration of 0.1 ppm or less.115Similar results were obtained from analyses of European samples of 2,4,5-T.116 It is important to note that the single investigator reporting a statistically significant excess of ML due to exposure to TCDD-containing phenoxys also tested fat samples from phenoxy-exposed patients with ML for the presence of TCDD. Rather unexpectedly, no elevated levels of TCDD were found in the patients.117 This again suggests that the excess risk of ML in that study was not due to TCDD exposure. None of the 18 cohort studies of workers potentially exposed to TCDD have reported a statistically significant excess of ML. This does not seem to be due to inadequate sample size of many of these studies since the two largest studies of workers in manufacturing plants where TCDD exposure is highest (NIOSH and IARC/NEEHS), which have combined many such orts together, did not find an increased risk of ML. Morev. _r, in the only cohort study of workers in manufacturing/ formulating plants in which it was possible to separate out to a large extent workers engaged in phenoxy acid production from workers engaged in the production of other chemicals (pigments), a statistically significant excess of ML observed in the factory as a whole was found to be confined to workers engaged in making pigments, and not among workers handling phenoxys.87 It is concluded therefore that the overwhelming evidence does not support a causal association between ML and exposure to 2,3,7,8-TCDD. This conclusion is consistent with the animal data. No association between TCDD exposure and ML in male animals has been reported, except for the development of thymic lymphomas in mice given high toxic doses of TCDD during infancy.118 C. Soft-Tissue Sarcoma Only five case-control studies have reported a statistically significant association between STS and phenoxys/chlorophenols containing TCDD.52-53-55,57-66 Four of these studies were conducted by the same investigators in Sweden,52-53-55-57 and the fifth was done in Italy.66 The Italian study did not find an association between STS and phenoxy acid exposure in the study as a whole, neither in male STS cases whether living or dead, nor in female dead cases. A nonstatistically significant twofold risk for phenoxy exposure was observed in living fer ' STS cases, based on only five exposed cases, and a statistically significant relative risk of 15.5 was reported for living women less than 75 years of age exposed to phenoxys between 1950 and 1955. However, there were no official rec ords documenting whether or not 2,4,5-T was used during 1950 to 1955. Also, two of the five cases were chondrosarcomas of the lower limb and thorax, hence not true STS, and in the third case, the histology was uncertain. The Swedish studies were unique in that they were not only conducted by the same investigators, but also were done in the same country using the same questionnaire instrument and study design. Thus, a high degree of replicability in findings is ex pected. Unfortunately, this has not been the case (Table 6). The inconsistency evidendy makes interpretation of the Swed ish studies difficult, especially in reference to TCDD-contain ing phenoxys and chlorophenols. Furthermore, there is evi dence of bias in these studies. Although the main tool in these studies was a mailed questionnaire, all respondents who had reported work in agriculture and forestry or whose response to questions on exposure to phenoxys/chlorophenols was unclear were contacted again and reinterviewed by telephone on ex posure to these compounds. In both the first two studies,52-55 Table 6 Lack of Replicability of Swedish SoftTissue Sarcoma Studies Com pound F irst study Second study E riksson 2,4,5-T & other phenoxys Phenoxys with no TCDD C h lorophenols Chlorophenols with no TCDD P henoxy acids e x p o su re in 1950s 2 ,4 ,5 -T exposure in 1950s P henoxy acids e x p o su re in 1960s Phenoxy acids (highest ex- posed group) 17.0* 4.2* 3.3* -- -- -- -- -- 1.8 0.6 -- 5.3* 2.4* 2.9* 0.7 2.1 H ardell C hlorophenols Chlorophenols (high grade) Phenoxys 6.6* 0 .7 (Pop.) 0.8 (Cancer) -- 0.3 (Pop.) 0.5 (Cancer) 5.3* 3.3* (all phenoxys) 3.1*. 3.5* (TCDD Ph) Note: Pop. = population controls; Cancer = cancer controls; Ph = phenoxys. ` Statistically significant. 458 Volume 21, Issue 6 Toxicology there was evidence of interviewer bias,56 in spite of the fact that interviewers were blind to the status of cases and controls (Table 7). In the first study by Eriksson et al.,52 the relative risk of STS for exposure to phenoxy acids using information from both questionnaire and telephone interview was 6.8 as compared with 4.1 using the questionnaire information only; similarly, in the first study by Hardell and Sandstrom,55 the relative risks were 5.3 and 2.6, respectively (a twofold dif ference). The telephone interviews resulted in exposed controls being disproportionately reclassified as unexposed. Similarly, there was evidence of possible recall bias in these studies as shown in the second study by Hardell and Eriksson57 -- a statistically significant risk of 4.1 was obtained for exposure to phenoxy acids using population controls, and a borderline statistically significant risk of 2.4 was obtained using cancer controls (Table 7). Table 7 Swedish Studies Interviewer Bias (Phenoxy Acids) O dds ratios Q uestionnaire Interview & questionnaire 2.6* 5.3* 4.1* 6.8* * Statistically significant. Recall Bias R ef. 55 52 O dds ratios U sing population controls 4.1* Using can cer controls 2.4 ' Statistically significant. Apart from these difficulties, these studies do not provide evidence to incriminate TCDD specifically because the initial finding of a 17-fold risk associated with TCDD exposure52was not subsequently confirmed. As seen in Table 6 for both repeat studies, risks for exposure to phenoxys/chlorophenols which do not contain TCDD were either greater than or equal to those for exposure to TCDD-containing compounds. A similar par adox is that the risk for STS obtained in the Swedish studies for exposure to phenoxys/chlorophenols during occupational use is much higher than those reported in the NIOSH study which included workers in manufacturing plants with much higher TCDD exposure. Furthermore, whereas in the Swedish studies the potential exposure to TCDD was brief, e.g., in one of these studies,55 of the 11 cases exposed to 2,4,5-T, only 2 were exposed for more than a year (all 10 controls were exposed for less than a year), the NIOSH study has not shown as yet any increased risk of STS in workers who were exposed to TCDD for less than a year and who had mean TCDD levels much higher than those reported among Swedish STS cases. As with ML, it is pertinent to note that adipose tissue collected by Hardell from STS patients who had sprayed phenoxy her bicides did not show increased levels of TCDD."7 Also, the absence of dioxin isomers other than 2,3,7,8-TCDD in the different phenoxy herbicide preparations at concentrations sig nificant enough to explain the occurrence of excess STS115-116 does not support a causal role for dioxins in general in the etiology of STS, as claimed by Swedish investigators.53 This is consistent with the fact that a majority of case-control studies of STS have not showh an association with TCDD. It is to be noted also that TCDD is not known to cause STS in male animals. However, since the two largest cohort studies (NIOSH and IARC/NIEHS) reported an excess of STS, in spite of the paucity of confirmed cases of STS (only two) in the NIOSH study, and in spite of the absence of risk in manufacturing workers in the IARC/NIEHS study, a remotely possible role of TCDD in the etiology of STS cannot be completely ruled out in the future. The following is now necessary: (1) further follow-up of the NIOSH, IARC/NIEHS, and Seveso cohorts, and accrual of more STS cases to determine if there is a real excess risk in TCDD-exposed cohorts; and (2) nested casecontrol studies within the NIOSH and LARC/NIEHS cohorts should be conducted to investigate specifically whether any excess risk observed is due to TCDD exposure or other con comitant chemical exposure(s). This requirement is obligatory considering that (1) no case-control study has adequately con trolled for concomitant exposures; (2) in the NIOSH study, workers spent five times as long exposed to other chemicals than to TCDD; and (3) in the IARC/NIEHS study, the nonstatistically significant twofold risk of STS was the same in both TCDD-exposed and unexposed subjects. The identifica tion of plants within the two registeries in which workers were engaged in the production of phenoxys/chlorophenols almost exclusively will also be a great asset to the study of possible carcinogenic effects of TCDD, phenoxys, and chlorophenols. D. Other Cancer Sites As is evident from the review, there has been no consistent demonstration of excess risk of cancers of other sites either in the cohort studies. A significant excess of cancer of the lung, stomach, skin, bladder, and testicles only has been reported in more than one individual cohort study. However, such re ports have been fragmentary and not corroborated in the ma jority of studies. Further investigations of these cancer sites are needed. The demonstration of a statistically significant excess risk of respiratory cancer in the NIOSH study and of thyroid cancer in the IARC/NIEHS study in TCDD-exposed workers makes these sites of particular interest for a possible causal role of TCDD in their etiology since TCDD exposure at very high doses has been shown to cause lung cancer in male and female animals,"9120 while the thyroid is the site 1992 459 Critical Review s In n to be most sensitive to the carcinogenic action of TCDD . .imals.119 These sites therefore warrant further detailed tvestigations in the future. IN. CONCLUSION It is concluded that whereas the evidence from this review ses not support a causal role for TCDD specifically in the :currence of ML and possibly also STS, TCDD-exposed ocipational male populations have shown an excess of certain meets, particularly lung and thyroid cancers, which are known ibe caused by TCDD in male animals. Therefore, it is possible >at TCDD may be carcinogenic in humans. However, the /idence indicates that previous epidemiologic studies may ive focused unduly on sites which may not be the targets, or e most sensitive targets, for its carcinogenic action (ML, and jssibly STS). There is a need therefore for serious considation to be given in the future to evaluating the role of TCDD the etiology of cancers other than ML and STS. EFERENCES 1. B is h o p , C . M . a n d J o n e s , A . H ., N o n -H o d g k in 's ly m p h o m a o f the scalp in w orkers exposed to dioxins. Lancet, ii, 369, 1981. O ls s o n , H . a n d B r a n d t, L ., N o n -H o d g k in 's lym phom a o f the skin and occupational exposure to herbicides. Lancet, ii, 579, 1981. 3 . C o o k , R . R ., D io x in , ch lo racn e, and so ft-tissu e sa rco m a. La n cet, i, 618, 1981. 4 . F in g erh u t, M -, H alp erin , W . E ., H o n ch ar, P . A ., S m ith, A . B., G ro th , D . H ., a n d R ussell, W . O ., A n evaluation of reports o f dioxin exposure and soft-tissue sarcom a pathology am ong chem ical workers in the U nited States, Scand. J. Work E nviron. H ealth. 10, 299, 1984. 5 . H ard ell, L ., Phenoxyacetic acids and cancer -- a clinical observation, Lakartidningen, 74, 2753, 1977. 6 . H o n c h a r, P . A . an d H alp erin , W . E ., 2,4,5-T , trichlorophenol and soft-tissue sarcom a. Lancet, i, 268, 1981. 7 . H ope, W ., R ussell, W ., an d W eiss, S ., Porphyria cutanea tarda and sarcom a in w orkers exposed to 2,3,7,8-tetrachlorodibenzodioxin -- M issouri, JA M A , 251, 1534, 1984. 8. Jo h n so n , F . E ., K u g ler, M . A ., a n d B row n, S. M ., Soft-tissue sarcom as and chlorinated phenols. Lancet, ii, 40, 1981. 9. M oses, M . a n d Selikoff, I. J ., Soft-tissue sarcom as, phenoxy her bicides and chlorinated phenols. Lancet, i, 1370, 1981. 10. S a rn ia , P . R . a n d Ja co b s, J . , Thoracic soft-tissue sarcom a in Vietnam veterans exposed to A gent O range, N . E ngl. J. M ed ., 306, 1109, 1982. 11. B inaschi, S ., S an to ro , G ., an d In tro in i, U ., Hypothesis o f occu p atio n al e tio lo g y o f H o d g k in 's and n o n -H o d g k in 's lym phom a: a casecontrol study, G. Ital. M ed. Lav., 7, 181, 1985. 12. B row nson, R . C . an d R eif, J ., A cancer registry-based study of occupational risk for lym phom a, m ultiple m yelom a and leukem ia, /nr. J. Epidem iol.. 17, 27, 1988. 13. B uesching, D . P . an d W o llstad t, L ., C ancer mortality among fanners, J. N atl. Cancer Inst., 72, 503, 1984. 14. B u rm e iste r, L . F ., C ancer mortality in Iow a farm ers, 1971-78, J. N atl. Cancer Inst., 66, 461, 1981. 15. B u rm e iste r, L . F ., E v e re tt, G . D ., V an L ie r, S. F ., a n d Isacson, P ., Selected can cer m ortality and farm practices in Iowa, A m . J. E p idem iol., 118, 72. 1983. 16. C a n to r, K . P ., Farm ing and m ortality from non-H odgkin's lym phom a: a case-control study, Int. J. Cancer. 29, 239, 1982. 17. G iles, G . G ., L ic k iss, J . N ., B a ik ie, M . J . , L ow enthal, R . M ., a n d P a n to n , J . , M yeloproliferative and lym phoproliferative disorders in T asm ania, 1972-80: occupational and fam ilial aspects, J. Natl. Cancer Inst., 72, 1233, 1984. 18. P e a rc e , N . E ., S m ith , A . H ., a n d F ish e r, D . O ., M alignant lym phom a and m ultiple m yelom a linked with agricultural occupation in a N ew Zealand cancer registry-based study, Am . J. Epidem iol., 121, 225, 1985. 19. R eif, J . , P e a rc e , N ., K aw ac h i, I ., a n d F ra s e r, J . , Soft-tissue sar com a, and non-H odgkin's lym phom a and other cancers in N ew Zealand forestry w orkers, Int. J. Cancer. 43, 49, 1989. 20. S aftlas, A . F ., B lair, A ., C an to r, K . P ., H anrahan, L ., and A n d erso n , H . A ., Cancer and other causes o f death among W isconsin farm ers. Am . J. Ind. M ed., 11, 119, 1987. 21. S c h u m a c h e r, M . C -, F anning occupations and mortality from nonH o d g k in 's lym p h o m a in U tah. A case-control study, J. O ccup. M e d ., 27, 580, 1985. 22. W eisen b u rg er, D . D ., A rm itag e, J . O ., an d P u rtilo , D . T ., Envi ro nm ental ep id em io lo g y o f n o n -H o d g k in 's lym phom a (N H L) in E ast ern N ebraska (A bstr.), Fed. P roc., 45 , 640, 1986. 23. W ik lu n d , K ., L in d efo rs, B ., an d H olm , L ., Risk of m alignant lym phom a in Sw edish agricultural and forestry w orkers, Br. J. Ind. M ed., 45, 19, 1988. 24. D elzell, E . a n d G ru flien n an , S ., M ortality among white and non w hite farmers in N orth C arolina, 1976-1978, Am . J. Epidem iol., 121, 391, 1985. 25. F asal, E ., Ja c k so n , E . W ., an d K lau b er, M . R ., Leukemia and lym phom a m ortality and farm residence. A m . J. Epidem iol., 87, 267, 1968. 26. F ranceschi, S ., S errain o , D ., Bidoli, E ., T alam ini, R ., T irelli, U ., C a rb o n e, A ., a n d L a V ecchia, C ., The epidem iology of non-Hodg kin 's lym phom a in the N orth-East of Italy: a hospital-based case-control study, Leukem ia R es.. 13, 465, 1989. 27. G allagher, R . P ., T hrelfall, W . J ., Jeffries, E ., B and, P . R ., Spinelli, J . , a n d C o ld m an ,. A . J . , Cancer and aplastic anem ia in British Colum bia farm ers, J. Natl. Cancer Inst., 72, 1311, 1984. 28. M ilh am , S ., Leukem ia and multiple m yelom a in farm ers. Am . J. Epidem iol., 94, 307, 1971. 29. P rie ste r, W . A . a n d M aso n , T . J ., Hum an cancer mortality in relation to poultry population, by county, in 10 southeastern states, J. N atl. Cancer Inst., 53, 45, 1974. 30. S ch u m ach er, M . C . an d D elzell, E ., A death certifcate case-control stu d y o f n o n -H o d g k in 's lym phom a and occupation in m en in N orth Carolina, Am . J. Ind. M ed., 13, 317, 1988. 31. D u b ro w , R ., P au lso n , J . O ., an d W in d ian , R. W ., Farm ing and m alignant lym phom a in H ancock C ounty, O hio, Br. J. Ind. M ed ., 45, 25, 1988. 32. S ta rk , A ., H off, M ., a n d S to n e, R ., Cancer incidence and death from all causes in a cohort o f farm ers (A bstr.), M eeting o f the Society for Epidemiologic Research, Am. J. Epidem iol., 118, 431, 1983. 33. B race, K . D ., A gricultural chem ical use, type o f farm production, and farm county cancer m ortality in the rural U nited States: an ecological m odel, D iss. A b str. Int. [B J.. 46, 2879, 1986. 34. W illiam s, R . R ., Stegens, N . L ., an d G oldsm ith, J . R ., A ssociations o f cancer site and type with occupation and industry from the Third National C ancer Survey Interview, J. Natl. Cancer Inst., 59, 1147, 1977. 30 Volume 21, Issue 6 Toxicology 3 5 . Council R eport, C a n cer risk o f pesticides in agricultural workers, JA M A , 260, 959, 1988. 36. C antor, K ,, E verett, G ,, B lair, A ., G ibson, R ., Schum an, L ., and Is a c s o n , P . , F a n n in g a n d n o n -H o d g k in 's ly m phom a (A b str.), M eeting of the Society for Epidem iologic Research, Am. J. Epidem iol., 122. 535, 1985. 37. E verett, G ., B lair, A ., C an to r, K ., G ibson, R ., and Van L ier, S ,, Environm ental chem ical exposures as risk factors for leukem ia and non-H odgkin's lym phom a (A bstr.), M eeting of the Society for Epi dem iologic R esearch, A m . J . Epidem iol.. 122, 535, 1985. 38. H ard ell, L ., E rik sso n , M ., L en n er, P ., an d L undgren, E ., M alig nant lym phom a and exposure to chem icals, especially organic solvents, chlorophenols and phenoxy acids: a case-control study, Br. J. Cancer. 43, 169, 1981. 39. H o ar, S. K ., B la ir, A ., H olm es, F . F ., B oysen, C . D ., R obel, R . J ., H oover, R ., an d F ra u m e n i, J . J ., J r ., Agricultural herbicide use and risk o f lym phom a and soft-tissue sarcom a, JAM A. 256, 1141, 1986. 40. H o ar Z ahm , S ., B lair, A ., H olm es, F . F ., Boysen, C. D ., and R o b el, R . J . , A case-referent study o f soft-tissue sarcom a and H odg k in 's disease. Fanning and insecticide use, Scand. J. Work Environ. H ealth, 14, 224, 1988. 41. H o ar Z ahm , S ., W eisenburger, D. D ., B abbitt, P. A ., Saal, R . C ., C a n to r, K . P ., a n d B la ir, A ., A case-control study o f non-H odgkin's lym phom a and agricultural factors in Eastern N ebraska (A bstr.), So ciety for Epidem iologic Research, Am . J. E pidem iol., 128,901, 1988. 42. H o ar Z ahm , S ., W eisenburger, D. D ., B abbitt, P . A ., Saal, R . C ., V au g h t, J . B ., C a n to r, K . P ., an d B lair, A ., A case-control study o f non-H odgkin's lym phom a and the herbicide 2,4-dichlorophenoxyacetic acid (2,4-D ) in Eastern N ebraska, Epidemiology, 1 ,3 4 9 , 1990. 43. O lsso n , H . a n d B r a n d t, L ., R isk o f non-H odgkin's lym phom a am ong m en occupationally exposed to organic solvents, Scand. J. W ork E n viron. H ealth, 14, 246, 1988. 44. Pearce, N . E ., S m ith, A . H ., H ow ard, J . K ., Sheppard, R . A ., G ile s , H . A ., a n d T e a g u e , C . A ., N o n -H o d g k in 's lym phom a and exposure to phenoxy herbicides, chlorophenols, fencing w ork, and m eat works em ploym ent: a case-control study, Br. J. Ind. M ed., 43, 75, 1986. 45. P earce, N . E ., S h ep p ard , R . A ., Sm ith, A . H ., and Teague, C . A ., N on-H odgkin's lym phom a an d farm ing: an expanded case-control study, Int. J. Cancer, 39, 155, 1987. 46. P ersson, B ., D ah lan d er, A ., F redriksson, M ., B rage, H . N ., O hlso n , C .-G ., a n d A xelson, O ., M alignant lymphomas and occupational exposures, Br. J. Ind. M ed., 46, 516, 1989. 47. W e is e n b u rg e r, D . D ., E n v iro n m en tal ep id em io lo g y o f n o n-H odgkin's lym phom a in Eastern N ebraska, Am . J. Ind. M ed., 18, 303, 1990. 48. W oods, J . S ., Polissar, L ., Severson, R . K ., H euser, L . S ., and K u la n d e r , B . G . , S o ft-tissu e sa rc o m a an d no n -H o d gkin's lym phom a in relation to ph en o x y herbicide and chlorinated phenol exposure in w estern W ashington, J . N atl. C ancer Inst., 78, 899, 1987. 49. The Selected Cancers Cooperative Study Group, The association o f selected cancers w ith service in the US M ilitary in Vietnam . I. NonH o d g k in 's ly m p h o m a . A r c h . In te rn . M e d ., 150, 2 4 7 3 , 1990. 50. The Selected Cancers Cooperative Study Group, The association of selected cancers w ith service in the US M ilitary in Vietnam. HI. H odg k in 's disease, nasal cancer, nasopharyngeal cancer, and primary liver cancer, Arch. Intern. M ed., 150, 2495, 1990. 51. B a la ra ja n , R . a n d A cheson, E . D ., Soft-tissue sarcomas in agriculture and forestry w orkers, J. Epidem iol. Commun. Health, 38, 113, 1984. 52. E rik sso n , M ., H ard ell, L ., B erg, N. O ., M oller, T ., and Axelson, O ., S oft tissue sarcom as and exposure to chem ical substances: a casereferent study, B r. J. Ind. M ed., 38, 27, 1981. 53. E rik sso n , M ., H a rd e ll, L ., a n d A d am i, H ., Exposure to dioxins as a risk factor for soft tissue sarcom a: a population-based case-control study, J. Natl. C ancer Inst., 82, 486, 1990. 54. G reenw ald, P ,, K ovasznay, B ., Collins, D . N ., and T h erriau lt, G ., Sarcom as of soft-tissues after Vietnam service, J. Natl. Cancer Inst., 73, 1107, 1984. 55. H ard ell, L . an d S an d stro m , A ., Case-control study: soft-tissue sar com a and exposure to phenoxyacetic acids o r chlorophenols, Br. J. Cancer. 39, 711, 1979. 56. H a rd e ll, L ., Relation o f soft-tissue sarcom a, malignant lymphoma and colon cancer to phenoxy acids, chlorophenols and other agents, Scand. J. W ork Environ. H ealth, 7, 119, 1981. 57. H a rd e ll, L . a n d E rik sso n , M ., The association between soft tissue sarcom as and exposure to phenoxyacetic acids. A new case-referent study. Cancer, 62, 652, 1988. 58. H o ar Z ahm , S ., B lair, A ., Holm es, F . F ., Boysen, C. D ., and R o b el, R . J . , A case-control study of soft-tissue sarcoma: farming and pesticide use (A b str.), 12th Annual M eeting o f the Am erican Society of Preventive O ncology, Bethesda, M aryland, M arch 14-15, 1988, Prev. M ed.. 17, 241, 1988. 59. H o ar Z ahm , S ., B lair, A ., H olm es, F. F ., Boysen, C . O ., Robel, R . J . , a n d F ra u m e n i, J . F ., J r ., A case-control study o f soft-tissue sarcom a. Am . J. Epidem iol., 130, 665, 1989. 60. K ang, H ., E nziger, F ., B reslin, P ., Feil, M ., L ee, Y ., and S hep ard , B . , Soft tissue sarcom a and military service in Vietnam: a case-control study, J. Natl. Cancer Inst., 79, 693, 1987. 61. Sm ith, A . H ., F isher, D. O ., Pearce, N ., and Teague, C . A ., Do agricultural chem icals cause soft-tissue sarcom a? Initial findings o f a case-control study in N ew Zealand, Comm un. H ealth Studies. 6, 114, 1982. 62. S m ith, A . H ., F isher, D . O ., G iles, H . J ., and P earce, N ., The N ew Zealand soft-tissues sarcom a case-control study: interview find ings concerning phenoxyacetic acid exposure, Chemosphere, 12, 565, 1983. 63. Sm ith, A . H ., P earce, N . E ., F ish er, D. O ., Giles, H . J ., T eague, C . A ., a n d H o w ard , J . K ., Soft-tissue sarcom a and exposure to phenoxyherbicides and chlorophenols in N ew Zealand, J. Natl. Cancer Inst., 72, 111, 1984. 64. S m ith , A . H . an d P earce, N . E ., Update on soft-tissue sarcoma and phenoxyherbicides in N ew Z ealand, Chemosphere, 15, 1795, 1986. 65. Sobel, W ., B ond, C . G ., Skow ronski, B . J ., B row nson, P . J ., and C ook, R . R ., Soft-tissue sarcoma: case-control study, J. Occup. M ed., 28, 804, 1986. 66. Vineis, P ., T e rra d n i, B ., Ciccone, G ., C Ignetti, A ., C olom bo, E ., D onna, A ., M ail!, L ., P isa, R ., R icci, P ,, Z an in i, E ., and C om ba, P ., Phenoxy herbicides and soft-tissue sarcom as in female rice-w eeders: a population-based case-control study, Scand. J. Work Environ. H ealth, 13, 9, 1987. 67. W ingren, G ., F redrikson, M ., N oorlind B rage, H ., N ordenskjold, B ., a n d A xelson, O ., Soft tissue sarcom a and occupational exposures. Cancer. 66, 806, 1990. 68. The Selected Cancers Cooperative Study G roup, The association of selected cancers with service in the U .S. M ilitary in Vietnam. II. Softtissue and other sarcom as. A rch. Intern. M ed., 150, 2485, 1990. 69. D ald eru p , L . M . an d Z eile n rath , D ., Dioxin exposure: 20 year fol low-up, Lancet, ii, 1134, 1983. 70. Pazderova-V ejlupkova, J ., N em cova, M ., Pickova, J ., and Jirasek, L ,, T he developm ent and prognosis of chronic intoxication by tetrachlorodibenzo-p-dioxin in m en. A rch. Environ. H ealth, 36, 5 , 1981. 71. T hiess, A . M ., F rentzel-B eym e, R ., an d L in k , R ., M ortality study o f persons exposed to dioxin in a trichlorophenol-process accident that occurred in the BA SF AG on N ovem ber 17, 1953, Am. J. Ind. M ed., 3, 179, 1982. 72. Von L eh n ert, G ., an d Szadkow ski, D ., Z ur Humankanzerogenitat von 2,3,7,8-TC D D (-U nfallversicherungsrechtliche Beuneilung-), Arbeitsm ed. Sozialmed. P raeventivem ed., 10, 225, 1985. 73. Z o b e r, A ., M e sserer, P ., an d H u b e r, P ., Thirty-four-year mortality 1992 461 13090 Critical Review s In follow -u p o f B A S F em ployees e x p o sed to 2 ,3 ,7 ,8-T C D D after the 1953 accident, /nr. A rch. O ccup. Environ. H ealth, 62, 139, 1990. 74. O tt, G ., H o ld er, B . B ., a n d O lson, R . D ., A m ortality analysis of em ployees engaged in the m anufacture o f 2,4,5-trichlorophenoxyacetic acid, J. O ccup. M ed., 22, 47, 1980. 75. C ook, R . R ., T ow nsend, J . C ., O tt, G ., and Silverstein, L. G ., M ortality experience o f em ployees exposed to 2 ,3 ,7 ,8-tetrachlorodibenzo-p-dioxin (TCD D ), J. Occup. M ed., 22, 530, 1980. 76. C ook, R . R ., B ond, G . G ., O lson, R . A ., O tt, M . G ., and G ondek, M . R ., E valuation o f the m ortality experience o f w orkers exposed to the chlorinated dioxins, Chemosphere, 15, 1969, 1986. 77. O tt, M . G -, O lson, R . A ., C ook, R . R ., and B ond, G . G ., Cohort m ortality study o f chem ical w orkers w ith potential exposure to the higher chlorinated dioxins, J. Occup. M ed., 29, 422. 1987. 78. B ond, G . G ., W etterstroem , N . H ., R oush, G . J ., M cL aren, E . A ., L ip p s, T . E ., a n d C ook, R . R ., Cause specific mortality among em ployees engaged in the m anufacture, form ulation, or packaging of 2,4-dichloFophenoxyacedc acid and related salts, B r. J. Ind. M ed.. 45, 98, 1988. 79. B ond, G . G ., M cL aren, E . A ., L ipps, T . E ., an d C ook, R. R ., Effect o f reclassification o f chloracne cases. J. Occup. M ed., 32, 423, 1990. 80. Bond, G . G ., M claren, E . A ., L ipps, T . E ., and Cook, R. R ., U pdate o f m ortality am ong chem ical workers with potential exposure to the higher chlorinated dioxins, J. Occup. M ed., 31, 121, 1989. 81. B ond, G . G ., C ook, R . R ., B renner, F . E ., and M cL aren, E. A ., Evaluation o f m ortality patterns am ong chemical workers with chlor acne, Chemosphere, 16, 2117, 1987. 82. B ond, G . G ., O tt, M . G ., B re n n e r, F . E ., an d C ook, R . R ., M edical and m orbidity surveillance findings am ong em ployees potentially ex posed to T C D D , Br. J. Ind. M ed., 40, 318, 1983. Z ack , J . A . an d S usk in d , R . R ., The mortality experience o f workers exposed to tetrachlorodibenzodioxin in a trichlorophenol process ac cident, J. O ccup. M ed., 22, 11, 1980. 84. Z ack , J . A . a n d G affey, W . R ., A mortality study o f workers em ployed at the M onsanto com pany plant in Nitro, W est Virginia, E n viron. Sci. R es., 26, 575, 1983. 85. Coggon, D ., P an n ett, B ., W in ter, P. D ., A cheson, E . D ., and B onsall, J ., M ortality o f workers exposed to 2 m ethyl-4 chlorophenoxyacetic acid, Scand. J. W ork Environ. Health, 12, 448, 1986. 86. C oggon, D ., P a n n e tt, B ., a n d W in ter, P . D ., M ortality and incidence o f cancer at four factories making phenoxy herbicides, Br. J. Ind. M ed., 48, 173, 1991. 87. L y n g e, E ., A follow -up o f cancer incidence am ong w orkers in m an ufacture o f phenoxy herbicides in D enm ark, Br. J . Cancer, 52, 259, 1985. 88. R iihim aki, V ., A sp, S -, a n d H ern b erg , S ., M ortality o f 2,4-dichlorophenoxyacetic acid and 2,4,5-trichlorophenoxyacetic acid herbi cide applicators in Finland, Scand. J. Work Environ. H ealth. 8, 37, 1982. 89. H ogstedt, C . an d W esteriu n d , B ., Mortality o f forest workers ex posed and exposed to phenoxy acid preparations, Lakartidningen. 11, 1828, 1980. 90. G reen , L . M -, A cohort m ortality study of forestry workers exposed to phenoxy acid herbicides, Br. J. Ind. M ed., 48, 234, 1991. 91. A xelson, O . a n d S u n d ell, L ., Herbicide exposure, mortality and tum or incidence. A n epidem iological investigation o f Sw edish railroad w ork ers, W ork Environ. H ealth, 11, 21, 1974. 92. Axelson, O ., S undell, L ., A ndersson, K ., E dling, C ., H ogstedt, C ., an d K ling, H ., H erbicide exposure and tum or mortality. An up dated epidem iologic investigation on Swedish railroad workers, Scand. J. W ork Environ. H ealth, 6, 73, 1980. "'S. M o r g a n , D . P ., L in , L . I . , a n d S a ik a ly , H . H ., M orbidity and m ortality in workers occupationally exposed to pesticides. Arch. En viron. Contam. Toxicol., 9, 349, 1980. 94. B lair, A ., G raum an, D. J ., L ubin, J . H ., and Fraum eni, J . F ., Lung cancer and other causes of death am ong licensed pesticide ap plicators, J. Natl. Cancer Inst.. 71, 31, 1983. 95. R oyal Com m ission on the use and effects o f chemical agents on Aus tralian personnel in V ietnam , Final Report, Vol. 4, C ancer, Parlia m entary Paper N o. 291/1985, 1985. 96. M ich alek , J . E ., W olfe, W . H ., an d M in e r, J . C ., Health stares of A ir Force veterans occupationally exposed to herbicides in Vietnam , JA M A , 264. 1832. 1990. 97. T h o m a s, T . L . a n d K ang, H . K ., M ortality and morbidity among A rm y Chem ical Corps Vietnam Veterans: a preliminary report. Am. J. Ind. M ed., 18, 665, 1990. 98. R iih im ak i, V ., A sp , S ., P u k k a la , E ., a n d H ern b erg , S -, Mortality and can cer m orbidity am ong chlorinated phenoxyacid applicators in Finland, Chem osphere, 12, 779, 1983. 99. W ik lu n d , K ., D ich , J ,, an d H olm , L ., Risk o f m alignant lymphoma in Swedish pesticide appliets, Br. J. Cancer, 56, 505, 1987. 100. W ik lu n d , K ., D ic h , J . , a n d H o lm , L -, S o ft-tissue sarco m a risk in Sw edish licensed pesticide applicators, J. O ccup. M ed.. 30. 801, 1988. 101. W ik lu n d , K ., D ich, J ., H olm , L .-E ,, a n d E k lu n d , G ., Risk of cancer in pesticide applicators in Sw edish agriculture, Br. J. Ind. M ed .. 46, 809, 1989. 102. W olfe, W . H -, M ichalek, J . E ., M iner, J . C ., R ahe, A ., Silva, J ., T hom as, W . F ., G rubbs, W . D ., L ustik, M . B-, K arrison, T . G ., R o e g n er, R . H -, an d W illiam s, D. E ., Health stares of Air Force V eterans occupationally exposed to herbicides in V ietnam . I. Physical health, JAM A. 264, 1824, 1990. 103. W an g , H . H . a n d M acM ah o n , B ., M ortality o f pesticide applicators, J. O ccup. M ed., 21. 741, 1979. 104. M acM ahon, B ., M onson, R . R ., W ang, H . H ., and Z heng, T ., A second follow -up o f m ortality in a cohort o f pesticide applicators. J. O ccup. M ed., 30, 429, 1988. 105. C o rra o , G ., C a lleri, M ., C a rle, F ., R usso, R ., B esia, S ., and Piccioni, P ., C ancer risk in a cohort o f licensed pesticide users. Scand. J. W ork E nviron. H ealth, 15, 203, 1989. 106. B a rth e l, E ., H igh incidence o f lung cancer in persons w ith chronic professional exposure to pesticides in agriculture, Z. Erkrand. Atm . O rg., 146, 266, 1976. 107. B a rth e l, E ., Increased risk o f lung cancer in pesticide-exposed male agricultural w orkers, J. Toxicol. Environ. H ealth, 8, 1027, 1981. 108. K ogevinas, M ., S a ra c ri, R ., B ertazzi, P . A ., Bueno de M esquita, H . B ., Coggon, D ., G reen, L . M ., Johnson, E . S ., K auppinen, T ., L 'A b b e, K . A ., L itto rin , M ., L ynge, E ., M a th ew s, J . D ., Neub erg er, M ., O sm an , J ., P earce, N ., a n d W inklem an, R ., The IARC International R egister o f Persons Exposed to Phenoxy Herbicides and Contam inants, presented at the 23rd Int. Cong. O ccup. Health, Mon treal, Septem ber, 1990. 109. F in g erh u t, M ., H aiperin, W . E ., M arlow , D. A ., Piacitelli, L . A ., H o n ch ar, P . A ., Sw eeney, M . H ., G riefe, A . L ., D ill, P . A ., Kyle S te e n la n d , A . B ., an d S u ru d a , A . J ., C ancer m ortality in workers exposed to 2 ,3 ,7 ,8-tetrachlorodibenzo-p-dioxin, N . Engl. J. M ed.. 324, 212, 1991. 110. F in g e rh u t, M ., H a ip e rin , W . E ., M arlo w , D . A ., P iacitelli, L . A ., H o n ch ar, P. A ., Sw eeney, M . H ., G riefe, A . L ., D ill, P . A ., Kyle S teen lan d , A . B ., a n d S u ru d a , A . J ., M ortality among U .S. workers em ployed in the production o f chem icals contam inated with 2 ,3 ,7 ,8 tetrachlorodibenzo-p-dioxin (TCDD), Final report, January 1991. N T IS# PB 91-125971, U .S. D epartment o f H ealth and H um an Services, Public Health Service, Centers for Disease Control, National Institute of O c cupational Safety and Health, Industrywide Studies Branch, Cincinnati, OH 45226. 462 Volume 21, Issue 6 Toxicology 111. B ertazzi, P . A ., Z occhetti, C -, P esatori, A. C ., G uercilena, S., S an a ric o , M ., an d R ad ice, L ., Ten-year study o f the population involved in the Seveso incident in 1976, A m . J. Epidem iol., 129, 1187, 1989. 112. P atterso n , O. G ., J r ., F ingerhut, M . A ., R oberts, D. W ., Need ham , L . L ., H aring Sweeney, M ., M arlow , D. A ., A ndrew s, J . S ., a n d H a lp erin , W . E ., Level o f polychlorinated dibenzo-p-dioxins and dibenzofurans in workers exposed to 2,3,7,8-tetrachlorodibenzo-pdioxin. A m . J. lnd. M ed.. 16. 135, 1989. 113. International Agency for Research on Cancer, IARC m onographs on the evaluation o f carcinogenic risks to hum ans. O verall evaluations of carcinogenicity: an updating o f IA R C m onographs, Vols. 1 to 42, Supplem ent 7, Lyon, France, 1987. 114. International Agency for Research on Cancer, IARC m onographs on the evaluation o f carcinogenic risks to hum ans. Som e m onom ers, plas tics and synthetic elastom ers, and acrolein, Vol. 19, Lyon, France, 1979. 115. N o rstro m , A ., R ap p e, C ., L in d ah l, R ., an d B u ser, H .-R ., Analysis o f som e older Scandinavian form ulations o f 2,4-dichlorophenoxy acetic acid and 2,4,5-trichiorophenoxyacetic acid for contents of chlorinated dibenzo-p-dioxins and dibenzofurans, Scand. J. Work Environ. H ealth, 5, 375, 1979. 116. R a p p e, C . an d B u ser, H . R ., Identification and quantification of polychlorinated dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs) in 2,4,5-T -ester form ulations and herbicide O range, Chem osphere, 5, 431, 1978. 117. N y g ren ,M ., H an sso n ,M .,R a p p e , C ., D om m ellof, L ., and H ardell, L ., A nalysis of polychlorinated dibenzo-p-dioxins and dibenzofurans in adipose tissue from soft-tissue sarcom a patients and controls, pre sented before the Division of Environm ental C hem istry, Am erican Chem ical Society, M iam i, FL, April 1985. 118. D ella P o rta , G ., D rag an i, T . A ., an d S ozzi, G ., Carcinogenic effects of infantile and long-term 2 ,3 ,7 ,8-tetrachlorodibenzo-p-dioxin treat ment in the m ouse, Tumori. 73, 99, 1987. 119. National Toxicology Program , Technical Report Series N o. 209, Car cinogenesis bioassay of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (CAS No. 1746-01-6) in O sbom e-M endel rats and B6C 3F, m ice (G avage Study), N T P -80-31, NTH Publication N o . 82-1765, U .S . D epartm ent o f Health and H um an Services, Public Health Service, National Institutes of H ealth, 1982. 120. V an M iller, J . P ., L alich , J . J . , an d A llen, J . R ., Increased incidence o f neoplasm s in rats exposed to low levels o f 2 ,3 ,7 ,8-tetrachlorodi benzo-p-dioxin, Chemosphere, 9, 537, 1977. 121. R ao, M . S ., S u b b arao , V ., P rasad , J . D ., and Scarpelli, D . G ., Carcinogenicity o f 2 ,3 ,7 ,8-tetrachlorodibenzo-p-dioxin in the Syrian golden ham ster, Carcinogenesis, 9, 1677, 1988. 1992 463 REGULATORY TOXICOLOGY ANDPHARMACOLOGY 13, 150-169 (1991) Use of Epidemiology Data to A ss e s s the Cancer Risk of 2,3,7,8-Tetrachlorodibenzo-p-dioxin L in d a T o l l e f s o n O ffice o f N utrition and Food Sciences. C enterfo r F ood S a fety a n d A pplied N utrition, Food and D rug A dm inistration. W ashington, D .C. 20204 R e c e iv e d A u g u s t 1. 1 9 9 0 M ost countries have assessed the h u m an cancer risk o f 2,3 ,7 ,8-tetrachlorodibenzo-p-dioxin (T C D D ) by extrapolating from anim al d ata. T h e 1600-fold variation for acceptable intakes es tim ated by the A m ericans, C anadians, an d E uropeans, how ever, indicates a large scientific un certainty about T C D D toxicity. R egulators are attem p tin g to use h u m an epidem iology data to elim inate som e o f the lim itations inherent in using anim al d ata to quantitatively assess cancer risk in h u m an s, particularly now th at tech n iq u es for m easuring T C D D levels in adipose tissue and serum are available. T his paper presents an overview o f the m ajor epidem iologic studies that associate cancer with T C D D exposure. T he actual exposures experienced by the study populations are em phasized. T h e T C D D serum levels o f th e Seveso, Italy, residents and o f the N IO SH D ioxin R egistry participants indicate th at these populations hold th e best prom ise for future assessm ent o f the hum an carcinogenic risk o f dioxin. Based o n the an im al data, how ever, the calculated risk fro m these ex p o su res is still m o d erate co m p a re d w ith high b a ck g ro u n d can cer incidence p resent in m o s t in d u s tria liz e d c o u n trie s . 199l Academic Press. Inc. INTRODUCTION 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a by-product of the manufacture of polychlorinated phenols, particularly the herbicide 2,4,5-trichlorophenol (2,4,5TCP) and the germicide hexachlorophene (Ayres et al., 1985). TCDD seems to be the most toxic of the dioxin isomers that were evaluated in animal toxicology studies (IARC, 1978). Human evidence of adverse health effects associated with TCDD is derived from occupational, environmental, and military exposure. Routine occupa tional use of compounds contaminated with TCDD and industrial accidents caused the heaviest exposure. Most countries have extrapolated the human risk of TCDD from animal data. These values are shown in Table 1. U.S. regulatory agencies have estimated riskspecific doses based on linear-at-Iow-dose extrapolation from the Kociba bioassay data (Kociba et al., 1978, 1979). These doses represent the level in pg/kg body wt/day that may result in at most a lifetime risk of 10~6 for cancer. European and Canadian regulators generally do not believe that estimation of dioxin risk by the traditional 150 0273-2300/91 53.00 Copyright <9 1991 by Academic Press, Inc. All rights o f reproduction in any form reserved. er Risk l N utrition. ibenzo-p-dioxin table intakes esge scientific unm iology d ata to ly assess cancer i adipose tissue )gic studies that jd y populations N IO SH D ioxin ture assessment : calculated risk :e present he manufacture ophenol (2,4,5seems to be the cicology studies with TCDD is .outine occupaccidents caused n animal data, estimated riskia bioassay data >dy wt/day that and Canadian the traditional CANCER RISK OF 2,3,7,8-TCDD 151 TABLE 1 Risk-Specific (1 0 -6) doses and acceptable Doses of T C D D Published by Various Countries and U .S. agencies C o u n try D ose (pg/kg/day) End point M odel for RA" U .S. EPA U .S. C D C U .S. FD A Federal Republic o f G erm any T he N etherlands C anada (H ealth and W elfare) 0.006 ' 0.03 0.06 1.0 4.0 10.0 Cancer Cancer Cancer C ancer + reproductive C ancer + reproductive C ancer + reproductive LMS4 LMS LMS S P (1000) SF (250) SF (100) * RA, Risk assessm ent. 4 LM S, Linearized m ultistage m odel. c SF, Safety factor. linear-at-low-dose extrapolation method is appropriate because of the evidence that TCDD is not mutagenic. Canada, Germany, the Netherlands, and Switzerland base a level of 1-10 pg/kg/day as a maximum daily intake on the application of a 100-- 1000 safety factor to the "no observed effect level" identified in the animal bioassay (Kociba et al., 1978, 1979). No specific risk is calculated when this safety factor ap proach is used. As shown in Table 1, these estimates of acceptable intakes, which vary over 1600-fold, indicate the large scientific uncertainty surrounding TCDD potency estimates. Using animal bioassay data to quantitatively assess cancer risk in humans has been criticized because of the inherent uncertainties in extrapolating from laboratory animals to humans. Predictions are frequently made even though the mechanism of carci nogenesis is unknown, pharmacokinetic data are unavailable, the metabolic processes vary widely among species and between animals and humans, or extremely large doses are required to produce cancer in chronic bioassays. To eliminate some ofthese uncertainties, investigators are attempting to use human epidemiology data to quantitatively assess cancer risk in humans. Use of epidemiology data has many advantages because humans are evaluated and relevant exposure levels and routes of exposure can be analyzed. The majority of the available epidemiologic studies on the association of cancer to TCDD exposure provide little evidence that dioxin is a potent carcinogen in humans. However, the converse is also true--currently available evidence is inadequate to rule out TCDD as a human carcinogen. The ep idemiologic studies are limited by small numbers of exposed persons and hence low statistical power for identifying an effect, insufficient follow-up considering the latency required for a chronic disease such as cancer, and imprecise exposure assessments. The conclusions that can be drawn from the available human studies of TCDD exposure are particularly limited because of the uncertain accuracy of the exposure assessment. Techniques for measuring TCDD levels in either adipose tissue or serum are now available. Unfortunately, fat biopsies are invasive, and approximately 10 g of adipose tissue is needed to determine 1-10 parts per trillion (ppt) of TCDD (Stanley et al., 1985). A significant breakthrough occurred when Patterson et al. (1988) found a high correlation between adipose tissue and serum levels of TCDD when expressed 152 LINDA TOLLEFSON on a lipid-weight basis. The mean of the partitioning ratio of adipose tissue to serum TCDD was 1.09 with a 95% confidence interval (Cl) of 0.97-1.21. The ability to measure TCDD in serum years after exposure provides an opportunity to link observed health effects in human populations exposed to TCDD with actual measurements of exposure. The following is an overview of the major epidemiologic studies associating cancer with TCDD exposure. The actual exposures experienced by the study populations and the availability of data on TCDD levels in adipose tissue or serum are emphasized. The information that the epidemiologic studies provide about the human carcinogenic risk from dioxin is assessed from available exposure data. MAJOR EPIDEMIOLOGIC STUDIES OF THE ASSOCIATION OF CANCER WITH TCDD EXPOSURE B A S F , Ludw igshafen, Federal Republic o f Germ any-- 1953 On November 17, 1953, an accidental reaction occurred during the hydrolysis of 1,2,4,5-tetrachlorobenzol to 2,4,5-trichlorphenoxyacetic acid (2,4,5-T) in the BASF Ludwigshafen plant in the Federal Republic of Germany (FRG), and TCDD was formed in this reaction. In 1982, Thiess et al. published a mortality study on the 70 persons who were exposed in 1953 and on the 4 additional persons who were exposed for short periods during cleaning and testing procedures in 1954 and 1955. Of the 74 persons, 66 suffered chloracne. All 74 persons were traced and the vital status of each was assessed through June 30, 1980, and their mortality experience was compared with both internal and external control groups. External comparison groups consisted of the total population of the local town, the local district, and the FRG. Internal comparison groups consisted of two control groups matched by age and date of first employment and gathered from cohorts in other studies for which follow-up had been completed. Among those exposed to dioxin, 21 persons had died compared with 18 or 19 predicted by each external control group. Seven cancer deaths occurred among the exposed group compared with 4.1 expected from the external controls. The only sig nificant site-specific finding was 3 stomach cancer deaths compared with 0.024-0.034 expected. When compared with the 2 matched internal control groups, excesses of stomach and lung cancer were found in the dioxin-exposed group (3 stomach vs 0 or 1 and 3 lung vs 0 or 1). Recently, Zober et al. (1990) published the results of a mortality study of BASF employees who had been exposed to TCDD during or after the 1953 accident and followed through 1987. Approximately 250 employees were divided into three cohorts based on exposure status, and their mortality experience was compared with FRG rates. The composition of cohort C1in this study was comparable to that of the cohort studied by Thiess et al. (1982); i.e., 66 individuals were identical. Three deaths were from stomach cancer compared with one expected in the Cl cohort (not significant). For those individuals with chloracne (N = 114), a statistically significant increase in risk for all malignant neoplasms existed for those individuals > 20 years after first exposure. The standard mortality ratio (SMR) was 139. The study of Zober et al. (1990) was sufficiently powerful to identify a 2.5- to 3-fold increase in risk for all 19035 se tissue to serum 21. The ability to ty to link observed measurements of associating cancer y populations and 1 are emphasized, .man carcinogenic : ia t io n the hydrolysis of >-T) in the BASF and TCDD was y study on the 70 vho were exposed d 1955. Of the 74 ' status of each ~~ was compared i groups consisted le FRG. Internal : and date of first llow-up had been :d with 18 or 19 urred among the ols. The only sigwith 0.024-0.034 oups, excesses of 1stomach vs 0 or y study of BASF 153 accident and nto three cohorts pared with FRG that of the cohort hree deaths were (not significant). Scant increase in ) years after first / of Zober et al. se in risk for all CANCER RISK OF 2,3,7,8-TCDD 153 malignant neoplasms combined. However, the study has essentially no power to identify increased risks of even threefold for specific cancer sites. In 1985, 32 years after the industrial accident, adipose tissue was taken from six workers who were exposed to TCDD at BASF (Schecter and Ryan, 1988). Each of the six workers had experienced chloracne as well as other acute effects from the exposure. The average concentration of TCDD on a lipid-weight basis was 49 ppt with a range of 11-141 ppt. Assuming that the human body is a one-compartment model for dioxin distribution and excretion and assuming a dioxin half-life of approximately 7 years (Pirkle et al., 1989), the average 49 ppt level 32 years after exposure corresponds to an adipose tissue-level at the time of the accide'nt of approximately 1000 ppt on a lipid-weight basis. For a 70-kg man, assuming 15% lipid content, the mean body burden at the time of the accident would have been approximately 10.5 jig, which corresponds to a dose of 150 ng/kg body wt. Furthermore, if we assume that these workers had no further TCDD exposure and that their average age at the time of exposure was 35 years, their lifetime (70 years) average daily dose (LADD) propor tionality could be estimated from (150 ng/kg)(l/35 years)(l year/365 days) = 12 pg/kg/day. The lowest concentration of TCDD in adipose tissue measured 32 years after the accident was 11 ppt on a lipid-weight basis, which corresponds to approximately 280 ppt at the time of the accident. Based on the same assumptions and calculations, the LADD for 280 ppt is approximately equivalent to 3 pg/kg/day. This calculation, which is only an approximation, was based on limited adipose tissue data. If variable dose rate is an important factor in risk assessment or if doses received at certain life stages result in a disproportionate risk, then the use of LADD is ques tionable. Such information does not exist for dioxin. However, dioxin does have a long half-life in humans. Thus, the use of LADD to assess carcinogenic risk is probably an acceptable assumption. The LADD calculated by the method shown here is actually the average daily dose received over the worker's remaining lifetime, which is approximately 35 years. An alternative approach, which would average the dose received over the entire 70-year life, results in LADDs of 6 pg/kg body wt/day for the average TCDD exposure and 1.5 pg/kg body wt/day for the lowest exposure. Arguments can be made to support either approach; for our purposes, the discrepancy is probably not large, considering the other approximations used in these calculations. The TCDD levels measured in adipose tissue or serum are dependent upon the worker's nutritional status, which can be affected by disease or life-style changes. For example, the malnutrition associated with many chronic diseases could cause a decrease in adipose tissue, resulting in an increase in lipid and serum TCDD levels. Conversely, a substantial gain of adipose tissue would result in a dilution or lowering of lipid and serum TCDD levels. Although these potential fluctuations of individual lipid and serum TCDD levels over time introduce another area of uncertainty in the LADD calculations, the average or median values of the worker population are protected from this instability to some extent. Zober et al. (1990) reported serum TCDD levels for 28 BASF workers, who were tested in 1986. Cohort Cl had the highest median value of 24.5 ppt on a lipid-weight basis; the other two cohorts had serum TCDD levels that were essentially the same as background (9.5 and 8.4 ppt). These serum TCDD levels are similar to what would 19038 154 LINDA TOLLEFSON have been measured in adipose tissue, given the high correlation between adipose tissue and serum TCDD when expressed on a lipid-weight basis (Patterson et al., 1988). In 1989, two workers from cohort Cl, who had been exposed in the 1953 accident, had serum TCDD levels of 255 and 553 ppt (Table 8 in Zober et al., 1990). Currently, one (255 ppt) has acute myeloid leucosis and the other (553 ppt) has liver carcinoma without cirrhosis (Zober et al., 1990). By using a 7-year half-life, these serum levels can be extrapolated to approximately 10,000 and 20,000 ppt at the time of exposure. However, these are isolated examples of high exposure; the foregoing calculations reveal that as a result of the accident, the majority of the BASF workers have experienced a lifetime TCDD exposure that is not excessively higher than known background TCDD exposure levels. The available data collected from studies measuring human adipose tissue, blood, and milk levels of TCDD support the existence of a background level of TCDD in the general population, at least in industrialized coun tries. Ryan et al. (1985) measured TCDD in human fat obtained from 1979 to 1981 in Ontario. The tissues were taken from the abdominal region of 21 deceased hospital patients of both sexes, primarily 60 years of age and older. The average level of TCDD was 10.7 5.4 ppt; the highest level was 21.8 ppt. Graham et al. (1986) reported a mean level of 7 ppt TCDD in adipose tissue from 35 autopsy specimens from St. Louis, and Patterson et al. (1986) reported a mean level of 7 ppt TCDD in 35 adipose tissues from autopsy specimens from Georgia and Utah. Both these studies found that levels of TCDD increased with age, as expected. Ono et al. (1986) reported a mean level of 9 ppt TCDD from 13 adipose specimens in Japan. The U.S. Environmental Protection Agency found a mean level of 5 ppt TCDD is an analysis of 46 composite human adipose tissues prepared from over 900 specimens representing three age groups and all regions of the United States (Stanley et al., 1986). Commoner et al. (1986), who used a half-life estimate of 5 years and simplified models of pharmacokinetics to calculate body burdens of dioxin, showed that a con tinuous daily dose of approximately 1 pg TCDD/kg body wt must occur in order to maintain the background adipose tissue levels at 10 ppt in industrialized countries. M onsanto Co., Nitro, West Virginia-- 1949 One other significant accident during the trichlorophenol (TCP) process resulted in TCDD exposure for which the workers have been followed approximately 30 years, a sufficient period to allow for chronic diseases such as cancer to be clinically diagnosed. The accident occurred on March 8, 1949, at the Monsanto Co. plant in Nitro, West Virginia. The factory produced 2,4,5-T from 2,4,5-TCP. In 1980, Zack and Suskind reported on the results of a standardized mortality analysis conducted on all male workers who developed chloracne attributable to the 1949 accident. The vital status of the 121 members of this cohort was ascertained as of December 31, 1978, almost 30 years after the accident. All 121 members of the cohort were traced. The mortality of the cohort members was compared with that of the U.S. population. The SMR for deaths from all causes was 69; 32 deaths were observed and 46.4 were expected. Excess cancer deaths were found in only three sites: 5 deaths from lung cancer (2.85 expected), 3 deaths from neoplasms of lymphatic and hematopoietic i between adipose 5 (Patterson et a i, posed in the 1953 Zober et a i. 1990). (553 ppt) has liver 'ear half-life, these )00 ppt at the time ;ure; the foregoing the BASF workers higher than known l studies measuring the existence of a idustrialized counfrom 1979 to 1981 1 deceased hospital age level of TCDD idipose tissue from ) reported a mean ;from Georgia and h age, as expected, adipose specimens lean level of 5 ppt >' *from over 900 States (Stanley :ars and simplified showed that a cont occur in order to ialized countries. P) process resulted iximately 30 years, linically diagnosed, ant in Nitro, West dardized mortality attributable to the was ascertained as !1 members of the he U.S. population, ved and 46.4 were deaths from lung and hematopoietic CANCER RISK OF 2,3,7,8-TCDD 155 tissue (0.88 expected), and 1 fibrous histiocytoma (0.15 expected). None of these was statistically significant. The Zack and Suskind study lacks the power necessary to determine a statistically significant increase in cancers of most sites. The cohort was very small, and no internal control group was used for comparison. All members of the cohort had chloracne, which signifies TCDD exposure. However, exposure to TCDD need not result in chloracne in order for the potential chronic effects of TCDD to occur. Therefore, finding an appropriate internal control group that would not be misclassified by an inaccurate exposure assessment would be difficult at this plant. Zack and Gaffey (1983) reported on the mortality experience of Nitro plant male hourly employees who were working at the plant a few years after the accident. The study cohort consisted of employees working for at least 1 year between January 1, 1955, and December 31, 1977. Workers were considered exposed if they were assigned to an area of 2,4,5-T production. The data might be confounded because other chem icals used at this plant included p-aminobiphenyl (PAB), a known human bladder carcinogen. 2,4,5-T exposure was assessed for decedents only and was determined by the worker's assignment to a 2,4,5-T operation. Employees with plantwide responsi bilities were considered not to have been exposed to 2,4,5-T. Vital status of each cohort member was ascertained as of December 31, 1977. Mortality rates of the total plant study cohort were compared with those of the ageand race-standardized U.S. population. All 884 men in the study cohort were traced; 163 deaths were identified. The all-cause SMR was 103 and the SMR for all malignant neoplasms was 113. Significantly elevated SMRs occurred for bladder cancer and for arteriosclerotic heart disease: The SMR for bladder cancer was 989 (9 deaths observed and 0.91 expected), and the SMR for arteriosclerotic heart disease was 133. Zack and Gaffey (1983) attributed the excess bladder cancer to PAB exposure. The cohort of this mortality analysis was larger than the cohort of employees with chloracne studied by Zack and Suskind (1980). However, whether all members of this cohort were actually exposed to TCDD or what the level of exposure was is not clear. The criteria used to determine exposure to dioxin are questionable and could reduce the strength of any association due to random misclassification bias, as with the chlor acne study. Although there is no quantitative exposure information on the Monsanto workers, their TCDD levels are assumed to be similar to those of the BASF workers. Therefore, a significant excess of cancers is unlikely to occur among these workers. N IO S H D io x in Registry The National Institute for Occupational Safety and Health (NIOSH) maintains the NIOSH Dioxin Registry, which consists of approximately 6000 workers in 14 U.S. facilities who were identified as being employed in the production of dioxin-contam inated chemicals such as 2,4,5-T and 2,4,5-TCP (Fingerhut et al., 1984). These sites were assessed for adequacy of personnel and medical records and exposure data before being included in the registry (Halperin et ai, 1982). Recently, NIOSH investigators reported results from a cross-sectional epidemiologic study of 400 living workers who were employed at 2 of the 14 plants (Fingerhut et al.. 1989; Sweeney et al., 1989). Serum from 135 workers and from 54 unexposed referents was analyzed for TCDD on a lipid-weight basis. The mean TCDD level among the exposed workers was 266 156 LINDA TOLLEFSON ppt (range, 2-3389 ppt) compared with 8 ppt (range, 1.7-19.7 ppt) among the unex posed referents (Sweeney et al., 1989). Using a half-life of 7 years, the investigators calculated blood levels on a lipid-weight basis of up to 30,000 ppt TCDD at the approximate time of exposure. The mean extrapolated blood TCDD level present in workers at the time of termination of employment was 2658 ppt (Sweeney et al., 1989). We would like to compare these workers' lifetime TCDD exposure to background levels, as we did for the BASF workers. However, the NIOSH workers did not experience an accidental one-time exposure but instead were exposed continuously over their working lifetimes. Estimating LADD under these conditions is much more difficult because we need to know not only cumulative exposure but also how long the workers were exposed. The mean body burden at termination of employment can be calculated by using the mean extrapolated blood TCDD level of 2658 ppt in a 70-kg man, as suming 15% lipid content, from (2658 ng/kg)(70 kg)(0.15)(l Mg/1000 ng) = 28 A body burden of 28 n% corresponds to a dose of approximately 400 ng/kg body wt. Assuming that the workers' average age at termination of employment was 55 years and their average working life was 25 years and that the workers had no further TCDD exposure after termination of employment, the mean dose of 400 ng/kg body wt at termination of employment can be averaged over their working life plus their remaining life (70 --55 years) to calculate LADD proportionality: (400 ng/kg)( 1/(25 + 15) years)(l year/365 days) = 27 pg/kg/day. These calculations only estimate an approximate LADD to compare with background levels in industrialized populations and are not meant to indicate absolute levels. A LADD for TCDD of approximately 27 pg/kg/day, which is over 25 times the estimated background dose, is a substantial exposure. Results from a NIOSH mortality study of workers from the 14 sites are expected to be available in late 1990. Currently, no direct measurements of exposure exist for these workers; instead, duration of exposure is used as a surrogate of dose. An exposure matrix is being constructed from process and task descriptions, safety records, and analytical data reporting TCDD levels in substances produced at the companies (Fingerhut et al., 1989). It is probably reasonable to assume that the workers enrolled in the mortality study have approximately the same exposure as that measured for the workers in the morbidity study at 2 of the 14 plants. To calculate an approximate lifetime cancer risk that this dose represents, the calculated LADD of 27 pg/kg/day can be compared with the Food and Drug Administration's (FDA) risk-specific dose based on the rat studies of Kociba et al. (1978, 1979). FDA's dose of 0.06 pg TCDD/ kg body wt/day would result in a lifetime risk of at most 10-6 for cancer, therefore, 27 pg/kg/day would correspond to a risk of (27/0.06) X 10-6 = 4.5 X 10-4. Thus, using the worst-case estimate, at most 4 to 5 of 10,000 workers with this level ofTCDD exposure would be expected to die of cancer. The NIOSH Mortality Study may not be powerful enough to determine this excess risk, given the large background cancer rate in the United States. Herbicide Exposure in Sweden-- 1970s In 1979, Hardell and Sandstrom reported on a matched case-control study of softtissue sarcoma cases (STS) performed in northern Sweden where forestry is a major _ io n g the unexthe investigators ipt TCDD at the DD level present ppt (Sweeney et ire to background lid not experience uously over their ich more difficult / long the workers . can be calculated a 70-kg man, as- '0 ng/kg body wt. nent was 55 years no further TCDD ng/kg body wt at us their remaining kg/day. with background absolute levels. A the estimated sites are expected exposure exist for lose. An exposure ifety records, and ; companies (Finorkers enrolled in measured for the i an approximate ) of 27 pg/kg/day risk-specific dose if 0.06 pg TCDD/ cancer, therefore, 1.5 X 1CT4. Thus, bis level of TCDD ty Study may not ackground cancer itrol study of softorestry is a major CANCER RISK OF 2,3,7,8-TCDD 157 occupation. The cases were identified from admissions records of an oncology unit from 1970 to 1977; tumor histology was reviewed by a pathologist, who identified 52 cases of STS and 208 controls. Four controls matched by age, sex, year of death, and residence were selected for each case. Deaths from malignant tumors or suicide were excluded from the control selection. Exposure to chemicals, especially phenoxyacetic acids or chlorophenols, was assessed via a questionnaire mailed to the living cases and controls or to next-of-kin surrogates for the dead cases and controls. Supplementary information was gathered through a telephone survey by an interviewer blinded to the case/control status. In addition, a questionnaire was also sent to the employers of persons who stated work in the forest, pulp industries, or sawmills as an occupation, to verily use of phenoxy herbicides. Exposure to phenoxyacetic acids or chlorophenols was reported in 36.5% of the cases and in 9.2% of the controls. This finding results in an odds ratio for STS of 5.7 (95% Cl = 2.9-11.3). To validate the information gathered from surrogates of the dead cases and controls, the living and dead groups were analyzed separately. The odds ratio for STS was 9.9 for the 21 living patients and their controls and 3.8 for the 31 dead patients and their controls. Thus, the surrogates may have underestimated the risk. Analyses were also performed separately for those exposed only to phenoxy acetic acids (mixed with 2,4,5-T) and for those exposed only to chlorophenols; the odds ratios remained high and statistically significant for each (5.3 for phenoxyacetic acids and 6.6 for chlorophenols). A larger case-control study by these same investigators was published in 1981 (Er iksson et at, 1981). The study included all patients with STS, living or dead, in southern Sweden, where herbicides are used for agriculture, who had been reported to the national cancer registry from 1974 to 1978. Their histology slides were examined by two pathologists. For every living case, two controls were selected who were matched by age and residence; for every dead case, the controls were matched by age, residence, and year of death. Cancer cases and suicides were excluded. After the pathology review, 11 cases were omitted because of a previous wrong or uncertain diagnosis; 110 cases and 220 controls remained in the study. Exposure was assessed by postal questionnaire, as in the previous study, and in addition, an attempt was made to determine exposure to phenoxy acids free from polychlorinated dibenzodioxins and dibenzofurans. Cases with exposure to phenoxy acids and chlorophenols within 5 years of diagnosis of STS were excluded. The results of the study in southern Sweden were similar to those for northern Sweden: The odds ratio for STS associated with exposure to phenoxy acids or chlo rophenols was 5.1 (95% Cl = 2.5-10.4) and to phenoxy acids alone was 6.8 (95% Cl = 2.6-17.3). The investigators also subdivided the cases and controls into two categories based on the expected presence or absence ofTCDD in the phenoxy acids. For example, 2,4,5-T is known to contain TCDD, whereas 4-chioro-2-methylphenoxyacetic acid, 2,4-dichlorophenoxyacetic acid (2,4-D), monochlorophenoxyacetic acid, and others are believed not to contain TCDD. The odds ratio associated with 2,4,5-T was 17; with non-TCDD herbicides it was 4.2. Thus the authors concluded that the risk of STS was also associated with uncontaminated phenoxy acids. However, 2,4-D may have been contaminated with TCDD because the facilities for 2,4,5-T production were also used to produce 2,4-D (Lilienfeid and Gallo, 1989). These Swedish case-control studies on STS (Eriksson et al., 1981; Hardell and Sandstrom, 1979) have been interpreted as showing strong evidence for an association 158 LINDA TOLLEFSON of STS with TCDD. The major limitations of the studies concern the amount of exposure to TCDD. Exposure history was assessed retrospectively by questionnaire at a time when herbicide use was receiving media attention (Coggon and Acheson, 1982). This publicity could have influenced the STS cases to overestimate their exposure (information bias) or have contributed to recall bias (2,4,5-T was banned in Sweden in 1977). Furthermore, medical surveillance of herbicide workers could have been increased, possibly leading to bias. On the other hand, the odds ratios identified in the Swedish studies are quite high--too large to be explained by information, recall, or detection biases. The latency periods in the studies were sufficient (mean latency = 20 years) for carcinogenic effects to be initiated, the investigators matched on known confounders of risk for STS, and all tumor diagnoses were reviewed by pathologists-- a crucial control technique when dealing with a tumor as difficult to diagnose as STS. The publicized deficiencies of the Swedish studies generally represent limitations inherent in case-control studies. Unfortunately, the only realistic approach to study rare tumors such as STS is by case-control; cohort analyses of STS are inappropriate because many person-years and 15-20 years of follow-up are required to obtain tumors of this type. Recently, Hardell and Eriksson (1988) conducted a case-control study of STS in Sweden to test the reproducibility of their previous findings. Only patients with the diagnosis of STS based upon histopathologic examination were included. Two control groups were used to evaluate the effect of recall bias: a population-based referent group and an additional group of patients with another type of malignant disease. Exposure was assessed by questionnaire, and follow-up telephone interviews were performed on those subjects who had worked in forestry, agriculture, horticulture, carpentry, or in sawmills for some period since 1948, when phenoxyacetic acids and chlorophenols were first used in Sweden. All questionnaires and interviews were analyzed in a blinded manner. Cases exposed to phenoxyacetic acids produced an odds ratio of 3.3 for STS when compared with the population referents and an odds ratio of 2.2 when compared with the cancer referents. Because the risk ratio decreased when the control group with malignant disease other than STS was used, some recall bias cannot be excluded; yet we cannot rule out that phenoxyacetic acids may be associated with a general carci nogenic effect. No differences in risk were obtained for exposure to chlorophenols, in contrast to previous studies; however, the number of cases (4) exposed to chlorophenols in this study was low. Hardell and Eriksson (1988), who also assessed the risk of STS associated with exposure to TCDD as a contaminant in phenoxyacetic acids, found an odds ratio of 3.5 compared with the population-based control group and an odds ratio of 3.1 compared with the cancer control group. Another Swedish case-control study ofSTS failed to find an association with phenoxy herbicides. In 1986, Wiklund and Holm reported on the risk of STS associated with exposure to phenoxy herbicides in 354,620 Swedish men employed for at least 1week in I960 in agriculture or forestry. The control group consisted of 1,725.845 Swedish men employed in other industries during the same time period. The cohort was sub divided into six groups based on occupation. Cases of malignant tumors were identified via a nationwide cancer registry for the years 1960-1979. The relative risk for STS was 0.9, and subanalyses did not reveal any group with an elevated risk of STS. Limitations of this study include use of occupational classification from a national census as a surrogate for TCDD exposure and possible underascertainment of STS 131Gl n the amount of y questionnaire at dAcheson, 1982). ite their exposure ianned in Sweden could have been atios identified in formation, recall, :nt (mean latency latched on known by pathologists-- i diagnose as STS. resent limitations ipproach to study are inappropriate 1to obtain tumors 1 study of STS in patients with the ided. Two control ;ed referent group disease. Exposure 'ere performed on :, carpentry, or in chlorophenols d in a blinded 3.3 for STS when :n compared with mtrol group with t be excluded; yet h a general carcichlorophenols, in I to chlorophenols id the risk of STS :etic acids, found roup and an odds tion with phenoxy S associated with for at least 1week ,725,845 Swedish e cohort was subirs were identified itive risk for STS risk of STS. n from a national tainment of STS CANCER RISK OF 2,3.7,8-TCDD 159 cases through use of a national cancer registry. Agricultural and forestry workers may not have access to high-quality medical care, and therefore cases of STS may be un derdiagnosed. On the other hand, the study is large and the follow-up period long enough to reveal any excess STS cases. H erbicide Exposure in N ew Zealand and the United States Phenoxy herbicides, particularly 2,4,5-T, were widely used in New Zealand since the late 1940s. Researchers there undertook a case-control study of STS after the Swedish results were published (Smith et al., 1984). All male cases of STS reported to New Zealand's national cancer registry between 1976 and 1980 were included in the study. All histology reports were reviewed by pathologists. Controls were selected for each case from patients in the registry with other types of cancer. Cases and controls were contacted by a single telephone interviewer who was blinded to the case/control status. Exposure was ascertained by questions regarding employment and use of specific chemicals. Interviews were completed for 82 cases and 92 controls. The results, how ever, were not consistent with the Swedish studies. For cases having phenoxyherbicide exposure that lasted more than 1day and occurred more than 5 years preceding cancer registration, the estimate of relative risk (the odds ratio) was 1.3 (90% Cl = 0.5-4.5). Using only cancer patients as controls may have introduced misclassification bias in that dioxiirexposure could have possibly contributed to the cancers. The investigators undertook a further study of 51 more recently diagnosed STS cases (Smith and Pearce, 1986). Cases from 1976 through 1982 were selected from the same cancer registry as before. Histology reports were reviewed for each case, as in the earlier study. The 315 controls were selected from controls being used concur rently for another study of lymphoma and multiple myeloma. Only six cases were exposed, resulting in an odds ratio estimate of 0.7 (90% Cl = 0.3-1.5). The National Cancer Institute conducted a case-control study of STS, Hodgkin's disease, and non-Hodgkin's lymphoma in Kansas and found farm herbicide use to be associated with non-Hodgkin's lymphoma (Hoar et al., 1986). All cases of STS, Hodg kin's, and non-Hodgkin's lymphoma among white male Kansas residents from 1976 to 1982 were identified through a University of Kansas population-based cancer reg istry. Pathology specimens for 87% of the cases were reviewed by a panel of three pathologists. Controls were white males from the Kansas general population, matched to each case by age and vital status. Three controls were used for each case. The patients and controls, or their next of kin, were interviewed by telephone. Questions on farming practices included which herbicides and insecticides were used, how many days per year the men were exposed, and what protective equipment was used. For the 22 cases of STS that reported having used herbicides, the odds ratio was 0.9 when compared with 192 controls. For the 40 cases of non-Hodgkin's lymphoma involving herbicides, the odds ratio was 1.6. Men exposed to herbicides more than 20 days per year had an odds ratio of 6.0 for increased risk of non-Hodgkin's lymphoma (95% Cl = 1.9-19.5). Also, a significantly increased odds ratio was associated with the non-Hodgkin's lymphoma cases who had used phenoxyacetic acids. The study results, which are based on questionnaire data, may be subject to recall bias. Only three patients had ever used 2,4,5-T in this study; the herbicide most frequently used was J. ) 2 160 LINDA TOLLEFSON 2,4-D. Therefore, the study results suggest that the herbicide 2,4-D is not related to the development of STS. However, the results do not provide adequate evidence to refute the Swedish study that found an association between 2,4,5-T and STS. No quantitative data on TCDD exposure exist for any of the studies on the asso ciation of herbicide exposure to STS or other cancers. Although useful from a qualitative perspective, the data cannot be used to calculate the human carcinogenic risk of dioxin. South Vietnam-- 1962-1971 In January 1962, a program to use herbicides for defoliation and crop destruction began in Vietnam under the code name "Ranch Hand." Six herbicides were used (Agents Orange, Purple, Pink, Green, White, and Blue). During the 9 years the program was operational, approximately 19 million gallons of these herbicides were sprayed over South Vietnam, of which approximately 12 million gallons were Agent Orange (Wolfe et a i, 1985). Other than Agents White and Blue, most of the herbicides used in Vietnam were mixtures of 2,4-D and 2,4,5-T (Halperin et a i, 1982). Agents Pink, Green, and Purple each contained up to 66,000 ppb of TCDD; Agent Orange was used exclusively after July 1, 1965, and contained 2000 ppb of TCDD (Ayres et ai, 1985). In early 1978, the Veterans Administration established the Agent Orange Registry to identify all Vietnam veterans exposed to Agent Orange, to provide a mechanism for follow-up, and to obtain information on their current health status (Young et ai, 1985). The usefulness of the data in this registry is limited because TCDD exposure and health problems are self-reported and the participants are self-selected. In 1979, the U.S. Air Force began a prospective epidemiologic study of chronic health effects among Air Force personnel who conducted the aerial spraying of Agent Orange ("Ranch Handers"). Wolfe et a i (1985) reported on the analysis of health information data for 1174 Ranch Handers and controls who were sent questionnaires and for 1045 Ranch Handers and controls who underwent physical examination. Five controls, who were matched by race, age, and job category, were selected for each Ranch Hander. The controls were assigned to Air Force units that flew cargo aircraft in Southeast Asia. Each living Ranch Hander and the first of his living controls were selected for the questionnaire and physical examination phases. An exposure index for each job category was developed from mission records and tour data that included the amount and type of herbicide sprayed each month as well as the number of Ranch Handers in each job category who were involved in the spraying. The in vestigators were also able to determine the months each individual was assigned to the Ranch Hand operation. Exposure was determined as the effective number of gallons of herbicides to which each individual had been potentially exposed. Both the inter viewer and the examining physician were blinded as to the exposure status-of the subject. The few deaths which had occurred among the cohort at the time of this study did not indicate increased mortality compared with the controls. This finding is not un expected because of the young age of the cohort. Results were essentially negative. After adjustment for education, however, significantly more Ranch Handers than controls demonstrated hypochondria, depression, hysteria, and schizophrenia. Kang et a i (1986) reported on a study of Vietnam veterans in which 234 veterans with STS who had served in the U.S. military between 1964 and 1975 were compared 191. >is not related to luate evidence to and STS. idies on the assofrom a qualitative nic risk of dioxin. crop destruction ricides were used /ears the program des were sprayed ire Agent Orange in Vietnam were Ireen, and Purple 1exclusively after 5). In early 1978, ry to identify all ;m for follow-up, t al., 1985). The osure and health r v of chronic paying of Agent nalysis of health nt questionnaires examination, were selected for s that flew cargo lis living controls ;es. An exposure id tour data that ell as the number graying. The in. was assigned to lumber of gallons 1. Both the interure status of the of this study did inding is not unmtially negative. :h Handers than rophrenia. ich 234 veterans 5 were compared CANCER RISK OF 2,3,7,8-TCDD 161 with 13,496 patients without STS selected from a Vietnam-era patient population. No association was found between service in Vietnam and STS. However, in this study Vietnam service was used as a measure of exposure to TCDD and not enough time had elapsed for the diagnosis of STS. Kang et al. (1987) did another case-control study in which 217 STS patients who were selected from the Armed Forces Institute of Pathology were matched to 599 controls for service in Vietnam, exposure to chemicals, medical history, and life-style. The investigators found that the risk of STS was nonsignificantly increased for veterans with combat experience when compared with the risk for those without (odds ratio = 2.6), and also nonsignificantly increased for the combat veterans assigned to the region that was sprayed the most (odds ratio = 8.6). Again, the classification of exposure is questionable, and the study has little power to determine an effect. Thomas and Kang (1989) have recently reported the results of a study of Army Chemical Corps Vietnam veterans, i.e., men who were responsible for the mixing and application of herbicides. The study cohort consists of approximately 1000 veterans who served between 1965 and 1971 and whose vital status was followed through December 31, 1987. The only potentially important finding was 3 diagnoses of Hodg kin's disease compared with 0.002, which was expected based on prevalence rates among all male veterans in the Army's Agent Orange Registry. In an early study of Vietnam veterans in Australia (Fett et al., 1984), the only excess neoplasm found was 2 cases of STS vs 0.64 expected, whereas in the control group no STS was found and 0.84 was expected. The study used service in Vietnam as a surrogate for exposure. Fett et al. (1987a,b) published two reports on a retrospective cohort study that examined the number and causes of deaths among Australian Army veterans who served during the Vietnam era. Ofthose studied, 19,205 served in Vietnam and 25,677 served only in Australia. All men were traced from the end of their service, between 1966 and 1973, until January 1, 1982. Thus, the maximum follow-up time was 16 years. The first paper (Fett et al., 1987a) reported that the mortality of Vietnam veterans was well below that expected based on the Australian male death rates (SMR = 73) but statistically significantly higher than that of non-Vietnam veterans (relative risk = 1.3; 95% Cl = 1.1-1.5). The low mortality rate of the cohort compared with the Australian general male population could be explained by missed deaths, the enlistment requirement for good health, or the continual health screening practices of the Army. When stratified by assigned corps, the engineers' death rate for Vietnam veterans was 2.5 times that of non-Vietnam veterans (P = 0.001). Battle injuries were not important in contributing to the relative mortality of Vietnam and non-Vietnam veterans. The second paper (Fett et al., 1987b) reported the analysis of specific causes of death. Over 98% of the deaths among veterans were included in five classes of causes of death. Approximately three-fourths of the deaths were due to accidents, poisonings, and violence (external causes). The other four classes were neoplasms (14%), diseases of the circulatory system (7%), diseases of the digestive system (2%), and mental dis orders (2%). In the last two classes, alcohol abuse was a factor in most deaths among Vietnam veterans. The mortality rates for the Vietnam veterans were significantly increased only for external causes (relative risk = 1.3) and diseases of the digestive system (relative risk = 9.0). However, the numbers of deaths from specific cancers are quite small and the follow-up period is too brief to ascertain carcinogenic effects. As 19104 162 LINDA TOLLHFSON in other studies of Vietnam veterans, service in Vietnam was the surrogate for exposure to any potentially putative agent. Levels of TCDD in adipose tissue taken from Vietnam-era veterans have no cor relation with known exposure to TCDD and service in Vietnam (Gross et al., 1984; Lee and Hobson, 1985). Investigators from the Veterans Administration reported on the results of adipose tissue analyses from 20 veterans who believed they were exposed to Agent Orange in Vietnam, 10 veterans who had not served in Vietnam and who had no known exposure to TCDD, and 3 Air Force officers who had handled Agent Orange in field tests or in disposal operations (Lee and Hobson, 1985). Of the 10 unexposed veterans, 9 had 3-14 ppt TCDD in their adipose tissue. All 3 officers had repeated contact with TCDD and had been exposed within the previous 2 years; yet their adipose tissue contained 4, 5, and 6 ppt TCDD. The veterans who believed they were exposed to Agent Orange had tissue levels ranging from 0 to 99 ppt; the two high levels were 35 and 99 ppt and none of the others were above 13 ppt. The veterans were divided into three groups as to their likely past exposure to Agent Orange, based on military records. The most heavily exposed group consisted of the 2 veterans with levels of 35 and 99 ppt TCDD in their adipose tissue and 1veteran with no measurable TCDD. The other two groups had similar levels of adipose tissue TCDD; none were above 13 ppt TCDD. The adipose tissue TCDD levels also did not correlate with perceived or real health status. Kang et al. (1989) recently reported similar results. Mean TCDD levels in adipose tissue from 40 Vietnam veterans (13.4 ppt) did not differ appreciably from mean levels from either 80 non-Vietnam veterans (12.5 ppt) or 80 civilians (15.8 ppt). Also, the results showed no association between adipose tissue TCDD levels and any estimate of Agent Orange exposure opportunity based on military records. With the exception of military personnel who actually handled or sprayed Agent Orange in Vietnam, most Vietnam veterans have current TCDD body burdens no different from the average background level. From the Veterans Administration data, the highest adipose tissue TCDD level was 99 ppt on a lipid-weight basis approximately 20 years, or 3 half-lives, after exposure. This finding corresponds to 800 ppt at the time of exposure, a body burden of 8.4 ig, or a dose of 120 ng/kg. The LADD is calculated by (120 ng/kg)(l/50 years)(l year/365 days) = 6.5 pg/kg/day. This calculation assumes that the veteran was 20 years old at the time of exposure. The adipose tissue TCDD level of 13 ppt corresponds to 104 ppt at the time of exposure, a body burden of 1 tig, a dose of 16 ng/kg, and a LADD of 0.9 pg/kg/day. IC M E S A , Seveso, Ita ly-- July 10, 1976 On July 10, 1976, an accident occurred at the ICMESA factory during production of 2,4,5-TCP, which is used as an intermediate for hexachlorophene. The factory was located near Seveso, about 20 miles north of Milan. A plugged exhaust pipe caused an exothermic reaction that released high concentrations of TCDD into the atmosphere of an urban area southeast of the factory, which had a population of several thousand. Two to three days after the accident, small animals such as birds, rabbits, and chickens were dying, and a few children and adults were complaining of skin lesions and nausea 19105 ogate for exposure rans have no corjross et al., 1984; ation reported on :hey were exposed Vietnam and who id handled Agent 1985). Of the 10 All 3 officers had vious 2 years; yet vho believed they 99 ppt; the two ppt. The veterans nt Orange, based e 2 veterans with th no measurable CDD; none were ot correlate with levels in adipose 'rom mean levels 8 ppt). Also, the ind any estimate .'ayed Agent ody burdens no linistration data, is approximately ) 800 ppt at the The LADD is ay. me of exposure, ime of exposure, :/day. ring production The factory was ust pipe caused the atmosphere veral thousand, s, and chickens ons and nausea CANCER RISK. OF 2,3,7,8-TCDD 163 (Fanelli et al., 1980). The company analyzed the soil, dust, and vegetation in and outside the factory in an attempt to assess the amount and location of TCDD. Because concentrations as high as 15,000 ppb were found in grass, but much less was found in soil, all vegetation was cleared (Reggiani, 1978). Approximately 2 weeks after the accident, the contaminated area was divided into three zones based on the soil analyses. Zone A with > 10 ng TCDD/kg soil was evac uated. Zone B with 0.1-10 jugTCDD/kg soil was not evacuated, but recommendations were issued to reduce the possibility of exposure. Zone R had <0.1 /ig TCDD/kg soil, and safety measures were restricted to prohibition of food production and consumption (Reggiani, 1978). The Seveso Cancer Registry was set up in 1981 -to record cases of cancer occurring among the population in the entire Lombardia region (Puntoni et al., 1986). Cancer cases were identified through admission and discharge hospital forms. In 1986, the 1975-1981 site-specific cancer incidence rates were available. At that time, the incidence of STS was higher in the exposed group (517 per 100,000) than in the unexposed group (3.3 per 100,000) (Merlo and Puntoni, 1986; Puntoni et al., 1986). The numbers were very small, however, and comparable STS incidence rates did not exist for the area before 1975. Worldwide STS incidence rates are usually 2 per 100,000 (Waterhouse et al., 1982). In 1989, Bertazzi et al. published the results of a 10-year mortality study of the population involved in the Seveso incident (Bertazzi et al., 1989). Exposure was cat egorized by residence: At the date of the accident or their first entry into the area, subjects were classified as living in Zone A, Zone B, or Zone R. Persons living outside the contamination boundaries were considered to be unexposed. Both national and local mortality rates were used as reference standards. The cohort of people who had been living in the uncontaminated study area was determined to be the most suitable control population because they shared dietary habits and social and educational background with the exposed population. This population was followed according to the same criteria and methods and for the same amount of time. Mortality of all persons residing in the area between the date of the accident and December 31, 1984, was examined for the period July 10, 1976, to December 31, 1986. More than 99% of the subjects were traced. The all-cause mortality rates were similar among the zones and to the reference rate. In the analysis, cause-specific mor tality results for Zone A indicated an elevated mortality from circulatory diseases in both sexes. One death from cancer of the gallbladder and biliary tract among females resulted in a significantly high relative risk of 21.9. The results for Zone B were non significant among males. Among fmales, however, the two deaths from cancer of the gallbladder and biliary tract gave a nonsignificant relative risk of 3.5, and one death from STS was statistically significant when compared with the control population. In Zone R, two deaths from STS among males resulted in a nonsignificant but high relative risk. Females had a significantly elevated relative risk for uterine cancer. The vital status of all the chloracne cases was ascertained (N = 193); no deaths have occurred. The authors suggest that the excess deaths from cardiovascular problems may have been due to stress in the aftermath of the accident. The increased relative risks of specific cancer sites, even when statistically significant, were based on very few cases, and therefore the study results cannot be considered a reliable indication of a TCDD effect. Other limitations of the study include an insufficient latency period (10 years) for cancer and possible misclassification of exposure status. The residents of the un- 164 LINDA TOLLEFSON contaminated area could have been exposed to dioxin, because the only exposure indicator used was soil levels of TCDD. Residents of Zone A were evacuated after a few weeks, whereas residents of Zone B remained in an area considered to be very contaminated. Their compliance with issued regulations regarding consumption of products from the area has not been estimated. Also, data suggest that some parts of Zone R were more contaminated than initially thought (Caramaschi et al,, 1981). Nevertheless, > 10 years of follow-up is needed before the chronic effects of the Seveso accident can be fully evaluated. Centers for Disease Control (CDC) has approximately 30,000 frozen blood specimens from people exposed to TCDD at Seveso. A few of the specimens have been analyzed and the serum TCDD levels have been estimated on a lipid-weight basis at the time of the accident. Analyses of serum from 19 individuals who lived in Zone A indicate a range of TCDD levels from 828 to 56,000 ppt; the latter measurement was made in a child who was 4 years old at the time of the accident (Needham et al., 1989). The most recently reported half-life for TCDD of 7.1 years was used for this analysis (Pirkle et al., 1989). The presence of chioracne was not correlated with a threshold level of dioxin; significant overlap of TCDD levels existed between those with and without chioracne (Patterson et al., 1989). A serum level of 56,000 ppt on a lipid-weight basis in a 4-year-old child with a 30% body lipid level corresponds to a body burden of approximately 340 Mg, (56,000 ng/kg)(20 kg)(0.30)(l Mg/1000 ng) = 340 Mg, which is equivalent to a dose of approximately 17,000 ng/kg. If we assume that the child had no further exposure, his LADD (74 years) proportionality could be esti mated by (17,000 ng/kg)(l/70 years)(l year/365 days) = 660 pg/kg/day. The lowest serum level of TCDD among the persons with chioracne, 828 ppt, was measured in a 16-year-old girl and corresponds to the following body burden, dose, and LADD: (828 ng/kg)(50 kg)(0.20)(l Mg/1000 ng) = 8.3 Mg body burden (8.3 Mg/50 kg)(1000 ng/1 Mg) = 166 ng/kg dose (166 ng/kg)(l/55 years)(l year/365 days) = 8.3 pg/kg/day LADD. These approximate calculations reveal that the Seveso residents clearly have some of the highest TCDD exposures known to date; thus, they have significant potential to document whether dioxin exposure results in any chronic effects. The 4-year-old child with the highest exposure has a calculated LADD of 660 pg TCDD/kg body wt/ day, which is quite high compared with the other documented human exposures to TCDD. However, when this approximate LADD is related to FDA's risk-specific dose based on the rat study (Kociba et al., 1978, 1979), we find that the child's risk of death from cancer would result in a 10-2 risk at most. Because FDA's dose of 0.06 pg TCDD/ kg body wt/day will result in at most a lifetime risk of 10-6 for cancer, 660 pg/kg/day would correspond to a risk of 660/0.06 X 10-6 = 1.1 X 10-2. DISCUSSION The evidence for an association of TCDD with human cancer is equivocal. The workers known to be exposed to TCDD at the Monsanto plant in Nitro, West Virginia, the only exposure e evacuated after a isidered to be very lg consumption of that some parts of tschi et aL, 1981). fleets of the Seveso ;n blood specimens ave been analyzed t basis at the time n Zone A indicate ment was made in et al.. 1989). The his analysis (Pirkle threshold level of with and without i child with a 30% 0 Mg, g, s assume that the ity could be esti- %t -a y . me, 828 ppt, was >dy burden, dose, arden LADD. learly have some lificant potential ;. The 4-year-old DD/kg body wt/ tan exposures to risk-specific dose Id's risk of death f 0.06 pgTCDD/ r, 660 pg/kg/day equivocal. The a, West Virginia, CANCER RISK OF 2,3,7,8-TCDD 165 in 1949 have shown a nonsignificant excess of lymphatic and hematopoietic cancers, and the workers exposed at the BASF plant in Germany in 1953 have shown an excess of stomach cancer. The risk ratio for each type of cancer was small. Moreover, the positive findings of the BASF and Monsanto studies have not been duplicated in other epidemiologic studies because the cohorts have frequently been too small to have sufficient power to identify a small increase in cancer risk; were too recently exposed to develop cancer, which has a long latency period; or were exposed to other chemicals concurrently. The small number of workers involved in these accidents and the modest levels of TCDD exposure experienced by the workers make it unlikely that an excess of cancer will be determined in epidemiologic studies of these cohorts compared with the high background incidence rate of cancer in the United States. An assessment of the available data for an association between TCDD exposure and STS reveals inconsistent findings. Case-control studies are the only practical method to study risk factors for very rare diseases such as STS, but then assessing exposure retrospectively introduces a number of inherent problems. Also, selection of appropriate controls who must have low or zero probability of exposure to TCDD is very difficult. Very high relative risks of STS were found in the Swedish studies, but the risks were associated with exposure to "TCDD-contaminated herbicides" and therefore included exposure to a myriad of chemicals. Although the most recent study of Hardell and Eriksson (1988) did associate STS with TCDD, the results are not useful in risk assessment because quantitative exposure information was not available. In addition, corroborative evidence has not been found from studies in other countries. Serum specimens from the Ranch Handers are being analyzed by CDC, but the Vietnam experience will not likely reveal a human carcinogenic risk from dioxin. Although the Ranch Handers have shown no increase in the incidence of cancer, the cohort is composed of people who are still young; furthermore, the TCDD exposure from Agent Orange experienced by the Ranch Handers was, for the most part, not much different from the TCDD exposure experienced by a person living in an indus trialized country. Such exposures imply excess lifetime cancer risks too minute to isolate in industrialized populations with high annual rates of cancer incidence. More over, although the results of additional epidemiologic studies on the Ranch Handers, which will relate serum levels of dioxin to an assessment of mortality, are anticipated, the follow-up periods covered by the studies will still be inadequate to result in many cancers. Serum from the Seveso residents indicates that this population was significantly exposed to TCDD; CDC has estimated that several hundred of the participants probably had exposures resulting in serum concentrations of TCDD from 800 to over 50,000 ppt on a lipid-weight basis at the time of the accident (Needham et al.. 1989). Although the incidences of some cancers among those in the Seveso cohort are slightly increased over background, the numbers are too small to place much confidence in the findings. Also, the Seveso accident, which is relatively recent, has been studied for only 10-12 years to date. An additional important limitation of the Seveso data concerns the young age of the highly exposed cohort members, the majority of whom were children or young adults at the time of the accident in 1976. Therefore, the problem of a short latency period is compounded unless dioxin has the potential to increase the incidence of childhood cancers, and no evidence in human or animal studies supports such speculation. 13108 166 LINDA TOLLEFSON The results from the NIOSH Mortality and Morbidity Studies, which use data from NIOSH's Dioxin Registry, are expected in late 1990. This cohort also has documented high exposures to TCDD, although the percentage of the cohort that actually expe rienced the higher levels is not clear. The period of follow-up is long (up to 30 years), the serum levels of 135 workers at termination of employment average 2658 ppt, and because the exposure is occupational in nature, age is not a limiting factor (Sweeney et al., 1989). Yet, the calculated m axim um lifetime cancer risk predicted from the TCDD exposures experienced by the NIOSH cohort is only 4.5 X 10-4, and it is unlikely that this level of excess cancer risk can be determined by an epidemiol ogy study. , The findings of the epidemiology studies, which have quantitative exposure data, are summarized in Table 2. The studies are severely limited by low statistical power, except for the NIOSH cohort, whose power will depend upon the percentage of cohort members having high exposure histories and long follow-up periods. The maximum carcinogenic risks from dioxin implied by the studies (Table 2) were calculated by relating the approximate LADD of each of the exposed populations to FDA's risk- specific (10-6) dose based on the Kociba rat study (Kociba et al., 1978, 1979). The true risk may range from zero up to the maximum shown in Table 2. The risk to the Monsanto workers is unknown, but is probably <2 X 10-4, the calculated risk to the BASF workers. With the documented high exposure levels, the Seveso and NIOSH data hold the best promise for future use in the assessment of the human carcinogenic risk of dioxin, with the qualification that the maximum calculated risk of the Seveso and NIOSH cohorts based on the animal data is still moderate compared with high background cancer incidence rates for most industrialized countries. Therefore, we cannot be certain that the Seveso and NIOSH epidemiology studies will be sensitive enough to identify a significant excess of cancer. TABLE 2 Maximum Carcinogenic Risk to Populations and Workers Exposed to T C D D P o p u la tio n exposed BASF-- FRG M onsanto-- W est V irginia N IOSH D ioxin R e g istry V ietnam veterans Excess cancer Stom ach ? Lung ? Lym phatic ? H em atopoietic R esu lts expected late 1990 None Seveso ? None Pow er considerations Positive N egative Long latency Long latency Sm all cohort Low exposure Sm all cohort Low exposure High exposure Long latency Highly exposed subcohort m ay be sm all None High exposure Short latency Young cohort Low exposure Short latency H igh exposures in young M axim um " risk 2 X 1(T4 ? <2 x 10~4 4.5 X 10-4 1 X 1 0 '4 i x 1er2 ' F ro m F D A 's risk -specific ( 10- *) d o se b a se d o n th e K o c ib a ra t s tu d y (K o c ib a et al., 1978, 1979). isio e ucn use data from o has documented hat actually expeg (up to 30 years), age 2658 ppt, and g factor (Sweeney redicted from the X 10-4, and it is by an epidemiol- ve exposure data, statistical power, rcentage of cohort s. The maximum 'ere calculated by ns to FDA's risk1978, 1979). The 2. The risk to the :ulated risk to the veso and NIOSH man carcinogenic risk of the Seveso npared with high es. Therefore, we : 1be sensitive d to TCDD -- M axim um " risk 2 X 10"4 <2 x icr4 'h o rt 4.5 X 10-4 1 X 10-4 1 X 10-J ing . 1978, 1979). CANCER RISK OF 2.3.7.8-TCDD 167 REFERENCES Ayres, S. M ,, Webb, K. B., Evans, R . G ., and Mikes, J. (1985). Is 2 .3 .7 ,8 -T C D D (d io x in ) a c a rc in o g e n for hum ans? Environ. H ealth Perspect. 62, 329-335. Bertazzi, P. a ., Zocchetti, C,, Pesatori. a . C,, Guercilena, S., Sanarico. M.. and Radice, L. (1989). Ten-year mortality study of the population involved in the Seveso incident in 1976. A m . J. E pidem iol. 129(6), 1187-1200. Caramaschi, R ., Del CORNO, G ., and Favaretti, C . (1981). C h lo ra c n e fo llo w in g e n v iro n m e n ta l c o n tam ination by T C D D in Seveso, Italy. Ini. J. E pidem iol. 10, 135-143. COGGON, D ., and ACHESON, E. D. (1982). D o p h e n o x y h erb icid es ca u se c a n c e r in m an ? L a n cet 1, 1057-- 1059. Commoner. B ,, Webster, T ., and Shapiro. K. (1 9 8 6 ). E n v ir o n m e n ta l L e v e ls a n d H ea lth E ffec ts o f C h lo rinated D ioxins a n d Furans. Presented at the A m erican A ssociation for the A dvancem ent o f Science A nnual M eeting, Philadelphia. PA. Eriksson, M ., Hardell, L ., Berg, N . O .. Moller, T ., and Axelson, O . (1 9 8 1 ). S o ft-tissue sa rc o m a s an d exposure to chem ical substances: A case-referent study. Br. J. Ind. M ed. 38, 27-3 3 . Fanelli. R ., Bertoni, M . P., Castelu, M . G ., Chiabrando, G , Martelu. G . P., Noseda, A .. Garattini, S ., Binaghi, C , Marazza, V.. and Pezza, F. (1 9 8 0 ). 2 ,3 ,7 ,8 -T e tra c h lo ro d ib e n z o -p -d io x in to x ic effects an d tissue levels in an im als from the co n tam in ated area o f Seveso, Italy. Arch. Environ. C ontam . Toxicol. 9(5), 569-577. Fett, M . J., Dunn, M ., and Adena, M . A. (1984). A u stra lia n V eterans H e a lth S tu d ies: T h e M o rta lity R ep o rt. P a rt I: .1 R e tro sp e c tiv e C o h o rt S tu d y o f M o r ta lity a m o n g A u stra lia n N a tio n a l S e rv ic e m e n o f th e Vietnam Conflict Era a nd an E xecutive S u m m a ry o f the M ortality Report. A ustralian G overnm ent Printing Service, C anberra, A ustralia. Fett, M . J., Adena, M . a ., Cobbin, D . M ,, and Dunn, M . ( 1987a). M o rta lity a m o n g A u stralian c o n sc rip ts of th e V ie tn a m co n flict era. I. D e a th fro m all cau ses. A m . J. E p id e m io l. 125(5), 8 6 9 -8 7 7 . Fett, M . J., Nairn, J. R., Cobbin, D. M ., and Adena, M . A. (1987b). M o rtality a m o n g A u stralian c o n scrip ts o f the V ietnam conflict era. II. Causes o f death. A m . J. E pidem iol. 125(5), 878-884. Fingerhut, M . A ., Marlow, D . A ., Honchar, P. A ., and Halperin, W . E. (1984). T h e N IO S H O c cu p atio n al D ioxin Registry. In P ublic H ealth R isk s o f the D io xin s (W . W . Low rance, Ed.), pp. 3 6 1-366. W m . K au fm an n , Los Angeles. Fingerhut, M . a ., Sweeney, M . H ,, Patterson, D . G .. Piacitelli, L. a ., Morris, J. A ., Marlow, D. A ., Hornung, R. w ,, Cameron, L. w ,, Connally, L. B., Needham, L. L.. and Halperin, w . e . (1989). Levels o f 2,3.7.8-tetrachlorodibenzo-p-dioxin in the serum o f U .S. chem ical w orkers exposed to dioxin contam inated products: Interim results. Chem osphere 19(1-6), 835-840. Graham, M., Hileman, F. D., Orth, R. G., Wendling, J. M ., and Wilson, J. D. (1986). C h lo ro carb o n s in ad ip o se tissue fro m a M isso u ri p o p u la tio n . C h e m o sp h ere 15, 1595-1600. Gross, M . L ., Lay, J. O ., Lyon, P. a ., Lippstreu, D ., Kangas, N ,, Harless, R . L ,, Taylor, S. E., and Dupuy, A. E. (1984). 2 ,3 ,7 ,8 -T e tra c h lo ro d ib e n z o -p -d io x in levels in a d ip o s e tissu e o f V ie tn a m v ete ra n s. Environ. Res. 33, 261-268. Halperin, w . e ., Honchar, P. a ., and Fingerhut, M . a . (1982). D io x in : A n o verview . A m . S ta t. 3 6(3), 285-289. Hardell, L., and Sandstrom, A . (1979). C a se -c o n tro l stu d y : S o ft-tissu e s a rc o m a s a n d e x p o s u re to p h e n - oxyacetic acids o r chlorophenols. Br. J. Cancer 39, 711-717. Hardell, L,, and Eriksson, M . (1 9 8 8 ). T h e a s so c ia tio n b e tw e e n so ft-tissu e sa rc o m a s a n d e x p o s u re to phenoxyacetic acids. Cancer 62, 652-656. Hoar. S. K., Blair, A., Holmes, F. F., Boysen, D. G , Robel, J. R., Hoover, R.. and Flammen, J. F. (1986). Agricultural herbicide use and risk o f lymphoma and soft-tissue sarcoma. J. A m . M ed. A ssoc. 256(9), 1141-1147. In tern atio n al Agency for R esearch o n C ancer (IA R C ) (1978). L ong-T erm H a za rd s o f P olychlorinated Di- benzodioxins and Polychlorinated D ibenzofurans. Joint N IEH S/IA R C W orking G roup R eport. IARC Internal Technical R eport 78-001. IARC, Lyon, France. Kang, H. K., Weatherbee, L., Bresun, P. P,, Lee, Y., and Shepard, B. M. (1986). Soft-tissue sarcomas and military service in Vietnam: A case comparison group analysis of hospital patients. J. Occup. M ed. 28(12), 1215-1218. Kang, H. K, Enziger, F., Breslin, P., Feil, M., Lee, Y,, and Shepard, B. (1987). Soft-tissue sarcoma and military service in Vietnam: A case-control study. J. N atl. C ancer Inst. 79(4), 693-699. 168 LINDA TOLLEFSON Kang, H.. Watanbe, J., and Remmers, J. (1 9 8 9 ). D io x in s a n d D ib e n zo fu ra n s in A d ip o s e T is s u e o f U.S. Vietnam Veterans a n d Their Controls. Presented at the N inth International Sym posium on C hlorinated D ioxins and Related C om pounds, T o ro n to , O ntario, Septem ber 1989. Kociba, R . J ., Keyes, D . G., Beyer, J. E., Carreon, R . M ,, Wade, C. E., Dittenber, D . a ., Kalnins, R . P., Frauson, L., Park, C. N., Barnard, S. D ., Hummell, R . A., and Humiston, G. C. (1978). R esults o f a tw o-year ch ro n ic toxicity an d oncogenicity study o f 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol. Appl. Pharm acol. 46, 279-303. Kociba, R . J., Keyes, D. G,, Beyer, J. E., Carreon, R . M., and Gehring, P. J. (1979). Long-term toxicologic studies o f 2,3,7,8-tetrachlorodibenzo-p-dioxin (T C D D ) in laboratory animals. A n n . N . Y. A cad. Sci. 320, 379-406. Lee, L. ., and Hobson, L. B. (1 9 8 5 ). 2 ,3 ,7 ,8 -T e tra c h lo ro d ib e n z o -p -d io x in (T C D D ) in b o d y fat o f V ie tn a m v eteran s a n d o th e r m en . In C h lo rin a ted D io xin s a n d D ibenzofurans in the T otal E nvironm ent I I (L. H. K eith, C. R appe, an d G . C houdbary, Eds.), pp. 197-204. B utterw orth, Boston. LlLlENFELD, D . E., and Gallo, M . a . (1 9 8 9 ). 2 ,4 -D , 2 ,4 ,5 -T , a n d 2 ,3 ,7 ,8 -T C D D : A n overv iew . E p id e m io l. Rev. I I , 28-58. Merlo, F., and Puntoni, R. (1 9 8 6 ). S o ft-tissu e sa rc o m a s , m a lig n a n t ly m p h o m a s, a n d 2 ,3 ,7 ,8 -T C D D exposure in Seveso. L ancet 2, 1455. Needham, L. L., Mocarelli, P., Patterson, D . G ., and Andrews, J. S. (1 9 8 9 ). P ilo t S tu d y (P h a s e I) fo r D eterm ining 2.3,7.8-Tetrachlorodibenzo-p-dioxin Levels in Serum o f Seveso Residents. Presented at the N inth International Sym posium on C hlorinated D ioxins and Related C om pounds, T oronto, O ntario, Septem ber 1989. Ono, M ., Wakjmoto, T ., and Tatsukawa, R . (1986). P o ly ch lo rin ated dib en zo -p -d io x in s a n d d ib e n z o fu ra n s in h u m an adipose tissues o f Ja p an . C hem osphere 15, 1629-1634. Patterson, D . G ., Holler, J . S., Smith, S. J ., Liddle, J . A ., Sampson, E. J., and N eedham, L. L. (1986). H u m an adipose d ata for 2,3,7,8-tetrachlorodibenzo-p-dioxin in certain U.S. sam ples. C hem osphere 15, 2055-2060. Patterson, D. G., Needham, L. L., Pirkle, J. L., Roberts, D. w ., Bagby, J., Garrett, w . a .. Andrews, J. S., Falk, H., Bernert, J. T., Sampson, F. J., and Hour, V. N. (1988). C o rre la tio n b etw een s e ru m an d adipose tissue levels o f 2,3,7,8-tetrachlorodibenzo-p-dioxin in 50 persons from M issouri. Arch. Environ. Contam . Toxicol. 17, 139-143. Patterson, D . G ., Mocarelli, P ., Needham, L . L., and Andrews, J . S. (1989). P ilo t S tu d y (P h a s e II) fo r D eterm ining PCD D a n d PC D F Levels in S erum o f Seveso R esidents C ollected at the T im e o f Exposure: F uture Plans. Presented at the N in th International Sym posium on C hlorinated D ioxins an d R elated C om pounds. T oronto, O ntario, Septem ber 1989. Pirkle, J . L ,, Wolfe, W . h ., Patterson, D . G ., Needham, L. L., Michalek, J. E., Miner, J. C ., Peterson, M. R ., and Phillips, D . L. (1 9 8 9 ). E s tim a te s o f th e half-life o f 2 ,3 ,7 ,8 -te tra c h io ro d ib e n z o -p -d io x in in V ietnam veterans o f O peration R anch H and. J. Toxicol. Environ. H ealth 27, 165-171. Puntoni, R ., Merlo, A ., Fini, a ., Meazza, L ,, and Santi, L. (1986). S o ft-tissu e s a rc o m a s in S eveso. Lancet 2, 525. Reggiani, G . (1978). M e d ic a l p ro b le m s ra ise d b y th e T C D D c o n ta m in a tio n in S eveso, Italy. A rch . T o x ic o l. 40, 161-188. Ryan, J . J ., Williams, D . T ., and Lau, B. P .-Y . (1 9 8 5 ). A n aly sis o f h u m a n fat tissu e fo r 2 ,3 .7 ,8 -te tra - chlorodibenzo-p-dioxin an d chlorinated dibenzofuran residues. In Chlorinated D ioxins a n d D ibenzofurans in th e T otal E nvironm ent I I (L. H . K eith, C. R appe. an d G . C houdbary, Eds.), pp. 205-214. B utterw orth, B o sto n . Schecter, A ., and Ryan, J. J . (1 9 8 8 ). P o ly c h lo rin a te d d ib e n z o -p -d io x in a n d d ib e n z o fu ra n levels in h u m a n adipose tissue from w orkers 32 years after occupational exposure to 2,3,7,8-TC D D . C hem osphere 17(5), 915-920. Smith, a . H ., Pearce, N . E., Fisher, D . O .. G iles, H . J., Teague, C . a ., and Howard, J. A. (1984). Soft-tissue sarcom a and exposure to phenoxyherbicides an d chlorophenols in New Zealand. J. Natl. Cancer Inst. 73(5), 1111-1117. Smith, A . H ., and Pearce, N . E. (1 9 8 6 ). U p d a te o n so ft-tissu e sa rc o m a a n d p h e n o x y h e rb ic id e s in N ew Zealand. Chem osphere 15, 1795-1798. Stanley, J . S., Going, J . E ,, Redford, D . P., Kutz, F. w ,, and Young, a . l . (1 9 8 5 ). A su rv e y o f analytical m ethods for m easurem ent o f polychlorinated dibenzo-p-dioxins (PC D D ) and polychlorinated d ibenzofurans (PC D F ) in h u m an adipose tissues. In C hlorinated D ioxins an d D ibenzofurans in the T otal E nvironm ent I I (L. H . K eith, C. R appe, a n d G . C houdbary, Eds.), pp. 181-195. B utterw orth, Boston. in A dipose T issue o f U.S. /m posium on C hlorinated T E N B E R , D. A.. KaLNINS, Humiston, G. C. (1978). :hlorodibenzo-p-dioxin in , P. J. (1979). Long-term ' anim als. Ann. N. Y. Acad. D) in b o d y fat o f V ie tn a m ta l E nvironm ent I I (L. H. A n overview. E pidem iol. >mas, an d 2,3,7,8-TC D D 39). P ilo t S tu d y (P h a se I) 0 Residents. Presented at ounds, Toronto, O ntario, iio x in s and dibenzofurans ., and Needham, L. L. .S. sam ples. C h em o sp h ere rrett, W. a ., Andrews, >rreIation between serum 1 ' ` ;ssouri. Arch. Environ. h - Pilot S tu d y (Phase II) 'a t the Tim e o fExposure: ed D ioxins and Related Miner, J. G , Peterson, ilorodibenzo-p-dioxin in 55-171. su e sarcom as in Seveso. 'eso, Italy. Arch. Toxicol. t tissue for 2,3,7,8-tetraoxins and D ibenzofurans >. 2 0 5 - 2 1 4 . B u tte rw o rth , zo fu ran levels in hum an )D . C hem osphere 17(5), Howard, J. A. (1984). Z ealand. J. Natl. Cancer m oxyherbicides in New L. (1985). A survey o f D) a n d poly ch lo rin ated lenzofurans in the Total B utterw orth, Boston. CANCER RISK OF 2,3,7,8-TCDD 169 Stanley, J . S., Boggess, K . E., Onstot, J., Sack, T . M ., and Remmers, J . C . (1 9 8 6 ). P C D D s a n d P C D F s in h u m a n a d ip o s e tissu e fro m th e E PA F Y 8 2 N H A T S repo sito ry . C h e m o sp h e r e 15, 1 6 0 5 -1 6 1 2 . Sweeney, M . H ., Fingerhut, M . a ., Patterson, D . G ., Piacitelli, L. a ., Morris, J . A ., Marlow, D. A ,, Hornung, R . w ,, Cameron, L. w ,, Connally, L. B.. Needham, L. L ., and Halperin, w . e . (1989). S eru m 2.3.7,3-Tetrachlorodibenzo-p-< iiox\n (2,3.7,8-TC D D ) Levels o f U.S. C hem ical W orkers E xposed to D ioxin-C ontam inated M aterials: P relim inary Report. Presented at the N inth International Sym posium on C hlorinated D ioxins and R elated C om pounds, T oronto, O ntario, Septem ber 1989. Thiess, A . M ,, Frentzel-Beyme, R ., and Link, R . (1982). M o rta lity stu d y o f p e rso n s ex p o se d to d io x in in a triclorophenol-process accid en t th a t o ccurred in the BA SF A G o n N o v em b er 17, 1953. A m . J. Ind. M ed. 3, 179-189. Thomas, T . L., and Kang, H . K . (1 9 8 9 ). M o r ta lity S tu d y o f A r m y C h e m ic a l C o rp s V ie tn a m V eterans. Presented at the N inth International Sym posium o n C hlorinated Dioxins an d R elated C om pounds, T oronto, O ntario, Septem ber 1989. Waterhouse, J., Muir, C , Shanmugaratnam, K , and Powell, J . (E ds.) (1 9 8 2 ). C a n c e r In c id e n c e in Five Continents. IA R C Sei. Publ. N o. 42. IA R C , L yon, France. Wiklund, K, and Holm, L. E. (1986). Soft-tissue sarcoma risk in Swedish agricultural and forestry workers. J. N atl. C ancer Inst. 76(2), 229-234. Wolfe, w . H ., Lathrop, G . R ., Albanese, R . A ., and Moynahan, P. M . (1985). A n ep id e m io lo g ic investigation o f health effects in A ir Force personnel follow ing exposure to herbicides an d associated dioxins. Chem osphere 14, 707-716. Young, A. L., Flicker, M . R ., Kang, H . K ., and Shepard, B. M . (1 9 8 5 ). H e a lth su rv e illa n c e o f V ie tn a m veterans claim ing A gent O range exposure. In C hlorinated D ioxins a n d D ibenzofurans in the Total E n vironm ent I I (L . H . K eith , C. R appe, and G . C houdbary, Eds.), pp. 167-179. B utterw orth, Boston. Zack, J. A ., and Suskind, R . R . (1 9 8 0 ). T h e m o rta lity e x p erien ce o f w o rk e rs ex p o se d to te tra c h lo ro d i- benzodioxin in a trichlorophenol process accident. J. Occup. M ed. 22(1), 11-14. Zack, J. A., and Gaffey, W . R. (1 9 8 3 ). A m o rta lity s tu d y o f w o rk ers e m p lo y e d a t th e M o n s a n to C o m p a n y plant in N itro, W est V irginia. In H um a n a n d Environm ental R isks o f C hlorinated D ioxins a n d R elated C o m p o u n d s (R. E. T u c k e r, A. L. Y o u n g , a n d A. P. G ra y , Eds.), p p . 5 7 5 - 5 9 1 . P le n u m , L o n d o n . Zober, A ., Messerer, P., and Huber, P. (1 9 9 0 ). T h irty -fo u r-y e a r m o rta lity fo llo w -u p o f B A S F e m p lo y e e s exposed to 2,3,7,8-T C D D after the 1953 accident. Ini. Arch. Occup. Environ. H ealth 62, 139-157. .3112 ( t C hem osphere, V o l.2 5 , N o s.1 - 2 ,pp 2 1 9 -2 2 2 , 1992 P rin te d in G reat B rita in 0 0 4 5 -6 5 3 5 /9 2 $ 5 .0 0 + 0 .0 0 Pergam on P re s s L td . B IO L O G IC A L L Y M E A S U R E D H U M A N E X P O S U R E T O 2,3,7,8- , TETRA CH LO R O D IB EN ZO -P-D IO X IN a n d h u m a n c a n c e r A.H. Sm ith, M .L.W arner School of Public H ealth U n iv ersity of C alifo rn ia, B erkeley, C A , U.S.A. ABSTRACT, I T he developm ent o f m eth o d s to biologically m easu re h u m a n ex p o su re to 2,3,7,8I tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) has becom e an im p o rtan t break th ro u g h for j epidem iological research. Because of the long h alf-life of 2,3,7,8-TCDD in hum ans, ' exceeding 5 years, biological m easurem ents give an indication o f hum an exposure i occurring over m any years. In this paper, we review cancer epidem iology studies of 2,3,7,8-TCDD in the light of biological m easurem ents o f exposure an d conclude th a t no ! - hum an cancer has so far been established to be caused by exposure to 2,3,7,8-TCDD. H ow ever increased cancer risks in tw o cohort studies in v o lv in g sig n ifican t exposure and su ffic ie n t tim e o f follow -up raise the possibility th a t 2,3,7,8-TC D D is a hu m an carcinogen. 1MT.RQP.UCTJQN T he developm ent o f m ethods to m easure 2 ,3 ,7 ,8-tetrachlorodibenzo-p-dioxin (2,3,7,8I TCD D ) in h u m an adipose tissue, and subsequently in the blood, have becom e | im portant breakthroughs for epidem iological research. Because of the long half-life of ! 2,3,7,8-TCDD in hum ans, exceeding 5 years, biological m easurem ents give an indication of hum an exposure occurring over m any years and reflect cum ulative dose ' received from o ccu p atio n al exposure. Such in fo rm a tio n is v alu ab le for causal ; inference since it allows identification of those studies involving sig n ifican t exposure | to 2,3,7,8-TCDD, and com parison of the results from d iffe re n t studies taking into I account d ifferen ces in exposure. Below, we review can cer epidem iology studies of 2,3,7,8-TCDD in the light of biological m easurem ents of exposure. METHODS Cancer epidem iology studies were identified w hich could be linked to biological m easurem ents o f exposure to 2,3,7,8-TCDD. M ost o f these studies involved biological m easurem ents on a subset o f the cohorts being studied. On the basis of these m easurem ents, inference could be m ade about the likely extent of exposure o f the w hole cohort. In som e cases, such as herbicide spraying, there have been no cohort studies involving w orkers with biologically m easured exposure to 2,3,7,8-TCDD. H ow ever the ex ten t of exposure associated w ith spraying 2,4,5-T co n tam in ated w ith 2,3,7,8-TCDD can be assessed on the basis o f a study o f serum 2,3,7,8-TCDD levels o f a g ro u p o f p ro fe s sio n a l s p ra y e rs (1). C a u sa l in f e re n c e w as b a s e d on th e fin d in g s o f th e collected studies tak in g into account the extent o f exposure to 2,3,7,8-TCDD and the adequacy of follow -up in term s of latency. 219 19113 BESUL.TS Studies of 2,3,7,8-TCDD-exposed populations that have used biological m easurem ents to assess exposure, and w hich could be lin k ed to cancer studies, are presented in Table 1. T h e l e v e l s p r e s e n t e d m a y b e c o m p a r e d t o l e v e l s in t h e g e n e r a l p o p u l a t i o n w h i c h a v e ra g e a ro u n d 5-10 p g /g lip id (2). T h e c o h o rt w ith th e h ig h e st av e ra g e levels o f 2,3,7,8-TCDD am ong those subjects m easured involves Seveso residents w ho were ex p o sed fo llo w in g an e x p lo sion in a c h e m ic a l m a n u f a c tu r in g p la n t (3). H o w e v e r the m easurem ents involved only 19 subjects. T h e cohort w ith the second highest average level involves U.S. chem ical p la n t w o rk ers w ith an av erage of 233 p g /g based on 253 m e a s u re m e n ts (4). T h e a v e ra g e level fo r 16 em p lo y e e s in the B A SF c h e m ic a l p la n t in G erm an y w as estim ated to be 74 p g /g (5). T h is is an ap p ro x im a te estim ate d eriv ed from d a ta presen ted in a b a r g rap h . R a n c h H a n d V eteran s w ere exposed to 2,3,7,8TCD D contam inated A gent O range in V ietn am w ith the highest exposure occurring am ong the en listed n o n -fly in g p erso n n el (6). A m ed ian o f 24 p g /g w as rep o rted for this group but the range w ent up to 618 pg /g indicating that some of the men had high exposures. T a b l e 1: E x p o s e d p o p u l a t i o n s t h a t h a v e b i o l o g i c a l m e a s u r e m e n t s o f 2 , 3 , 7 , 8 - T C D D a n d can be linked w ith cancer studies Population Number of M easurem ents Serum TCD D Level1 A verage (Range) Y ears Since Last E x posure R eferen ce U.S. C h e m ic a l W orkers BASF E m ployees Ranch Hand E n liste d Personnel Seveso R e sid en ts3 New Zealand 2,4,5-T S prayers 253 16 407 19 9 233 (2 - 3,400) 74 (0 - 553) 243 (0 - 618) 12,558 (828 56,000) 53 (3 - 131) 1 pg per gram of lipid 2 M edian value reported 3 Residents of Zone A 4 Y e a rs sin c e la s t s i g n i f i c a n t e x p o s u re ( R e f e r e n c e 7) 15 - 37 18 - 34 15 - 25 <1 164 (-) (5) (6) (3) (1) Findings reported for the four cancer epidem iology studies th at have used biom arkers of exposure and indicate sig n ifican t exposure to 2,3,7,8-TCDD are presented in Table 2. T h e hig h est ex p o sed g ro u p id e n tifie d in T a b le 1 in v o lv es Seveso residents. H o w e v e r d u ra tio n of follow -up o f this cohort is in su ffic ie n t for causal in feren c e. A fte r 10 years o f follow -up, no sig n ifican t fin d in g s have em erged in the highest exposure group (Z o n e A ) (8). 221 T a b le 2: C a n c e r f i n d in g s o f s t u d i e s w ith b io lo g ic a l m a r k e r s o f e x p o s u re Population N um ber of Y ears Since Subjects First Exposure Study Findings S M R ( 9 5 % C.l.) R eferen ce U.S. C h e m ic a l W orkers 1,520* BASF E m ployees Ranch Hand V eteran s Seveso Residents of Zone A H42 . 1,261 556 >20 >20 15-25 A ll C a n c e rs 1.5 (1.2-1.8) R e s p ira to ry C a n c e rs 1.4 (1.0-1.9) S o ft T issue S arcom a 9.2 (1.9-2.7) A ll C a n c e rs 2.0 (1.1-3.41 L u n g C a n c e r 2.5 (0.8-5.9) A ll C a n c e rs 0.7 (0.3-1.1) 10 A ll C a n c e rs3 0.9 (0.3-2.9) L u n g C a n c e r3 2.0 (0.5-8.1) (4) (5). (9) (8) 1 Subcohort o f w orkers w ith > 1 yr o f exposure and > 20 yrs latency. 2 Subcohort of w orkers that developed chloracne and w ith > 20 yrs since first exposure. 3 M ales only. T h e m ost im p o rta n t fin d in g s a re those re p o rte d fo r th e U.S. ch e m ic a l p la n t w o rk e rs since the cohort w as large, involved sig n ifican t exposure to 2,3,7,8-TCDD and included a su b -g ro u p w ith s u f f ic ie n t d u r a tio n o f fo llo w -u p (4). T h e m a in fin d in g s fo r those w ith at least 1 y e a r o f ex p o s u re a n d a t least 20 y ears o f fo llo w -u p a re p re s e n te d in T a b le 2. N o te w o rth y fin d in g s also a p p e a r in the stu d y o f B A S F e m p lo y ees (S). R e su lts are presented in T able 2 for a subcohort o f w orkers that developed chloracne and had at least 20 years o f follow -up. D u ratio n o f follow -up o f R a n ch H a n d V eteran s is not yet su ffic ie n t to m ake the stu d y u se fu l fo r c au sal in fe re n c e (T a b le 2). In a d d itio n , the m e d ia n e x p o s u re o f 24 p g /g ( T a b le 1) e v e n f o r th e m o st h e a v ily e x p o s e d g r o u p is n o t as g r e a t as th o se r e p o r te d fo r the above cohorts. D IS C U S S IO N Studies th at use biological m easures o f h um an exposure to 2,3,7,8-TCDD help to place results reported fo r o th er studies in perspective in term s o f relev an t levels o f exposure. Results from the study o f professional pesticide applicators in New Z ealand show th at significant exposure beyond background levels can be found in ground-level ap p licato rs w h o h a v e sp ray ed 2,4,5-T fo r m an y years. E v id e n c e th a t ex p o su re to 2 ^ ,7 ,8 -T C D D causes h u m an can ce r is largely based on stu d ies o f subjects w h o recall periods o f exposure less th an 90 days. O n the basis o f the fin d in g s reported fo r the New Z ealand study, it seems unlikely that exposures o f such short duration w ould sig nificantly increase serum levels o f 2,3,7,8-TCDD beyond b ack g ro u n d levels. T hus, if exposure to p h en o x y h erb icid es co n tam in ated w ith 2,3,7,8-TCD D causes can cer in hum ans, the epidem iologic evidence would likely come from studies of individuals with exposure occurring over m any years. There are no such studies and therefore pesticide user studies are not relevent to causal inference concerning 2,3,7,8-TCDD effects. Based on th e stu d y fin d in g s rep o rted in T a b le 2, the m ost in fo rm a tiv e h u m a n stu d y con cern in g 2,3,7,8-T C D D pub lish ed to d ate is the co h o rt o f U.S. ch em ical p lan t w o rk e rs (4). A lth o u g h th is stu d y d id n o t re v e a l an in c re a se d risk o f can c e rs o f the ly m p h atic or h em ato p o ietic tissue w hich have been postulated to be related to 2,3,7,8T C D D exposure, it supports new hypotheses concerning some other cancer sites. It provides the only h u m an evidence th at so ft tissue sarcom a m ig h t be related to 2,3,7,8T C D D exposure. In ad d itio n , it raises the possibility th a t th ere m ight be a generalized increased risk o f can ce r resulting from high levels o f exposure to 2,3,7,8-TC D D since there was an increase in m ortality from all cancers com bined, largely attrib u tab le to cancers of the respiratory and gastrointestinal tracts. H ow ever on the basis of current pu b licatio n s o f fin d in g s in this cohort, it is not possible to ex clu d e effe cts from o th er occupational exposures such as asbestos w hich m ust be the presum ed cause of the tw o m esotheliom as reported. It is n o tew o rth y th a t the BASF cohort fo u n d increased risks o f all cancers com bined, also largely a ttrib u tab le to respiratory and gastrointestinal tract cancers. H ow ever a case of pleural m esotheliom a was reported for this sm all cohort. Thus, although two cohorts provide evidence that 2,3,7,8-T C D D m ay increase risks of a variety o f cancers and there are no cohorts involving significant exposure w hich provide evidence against this hypothesis; asbestos, w hich has been found to increase the risks o f both respiratory and gastrointestinal tract cancers, probably accounts for at least part of the observed excess cancers in b oth th e BA SF co h o rt an d th e U.S. ch em ical p lan t w orkers cohort. In conclusion, w hen the results of cancer epidem iology studies are review ed in light of biological m easurem ents o f exposure, no h u m an cancer site has yet been established to be cau sed b y e x p o s u re to 2,3,7,8-T C D D . H o w e v e r th e re is e v id e n c e fo r an in c re a se in th e o v e ra ll risk o f c a n c e r b ased on tw o stu d ie s , w ith no e v id e n c e to th e c o n tra ry . It is expected that follow -up over the next ten years of these and o ther cohorts experiencing significant exposure will provide valuable inform ation for causal inference con cern in g 2,3,7,8-TC D D and h u m an cancer. REFERENCES 1. S m ith , A .H , P a tte rso n , J r , D .G , W a rn e r, M.L. et al. (1991) J N a t C a n c e r In st (In Press). 2 . P a t t e r s o n , J r , D . G , F i n g e r h u t , M . A , R o b e r t s , D . W . e t al. ( 1 9 8 9 ) A m J I n d M e d l f i , 1 3 5 - 1 4 6 . 9876543 3. M ocarelli, P ,, N eedham , L.L., M aro cch i, A , et al. (1991) J T oxicol E n v iro n H e a l t h 22, 3 5 7 - 3 6 6 . 4. F in g e rh u t, M .A , H alp erin , W .E , M arlow , D .A , et al. (1991) N E JM 224, 212-218. 5 . Z o b e r , A , M e s s e r e r , P , H u b e r , P . ( 1 9 9 0 ) I n t A r c h O c c u p E n v i r o n H e a l t h 62, 139-157. 6. W olfe, W .H , M ichalek, J .E , M iner, J.C. et al. (1990) JA M A 24. 1824-1831. 7. S m ith , A .H . a n d P earce, NJE. (1986) C h e m o sp h ere 15, 1795-1798. 8. B e r t a z z i , P . A , 2 U > cch ett i, C , P e s a t o r i , A.C. e t al. ( 1 9 8 9 ) A J E 122. 1 1 8 7 - 1 2 0 0 . 9. M ich alek , J .E , W olfe, W.H. a n d M in er, J.C. (1990) JA M A 2 M . 1832-1836. Chcmosphere ,V o i.1 2 ,N o .2 ,p p 2 0 9 -2 1 5 ,1 9 e3 printed in Great B rita in 004 5-6535/83/020209--0 7 5 0 3 .0 0 /0 31983 Pera croon P r e ss Jutd. XiQJlCE This n-.2t.eri3; I " : . - , fytpyright Jaw -^2- Cci-^J. EXPOSURE or FLORIDA AJR9QAT AQUATIC WEED APPLICATORS TO 2 ,4-0ICHLQR0PtCKKYACETIC ACID (2,4-0) H. N. Nigg and J . H. Stsaper U niversity o f F lo rid , I n stitu te o f Food and Agricultural Scien ces, A gricultural Research and Eoucatlon Center, 700 Experiment S ta tio n Road, Labs Alfred, Fl 33650 ABSTRACT florid alrboat handgun applicators ere monitored far exposure to 2 ,4 : Qlctvlaraphenoxyacetlc a d d (2 ,4 -0 ). Four ap p licators ere monitored 1th a ir sam plers, -c e llu lo se patches, and 24 hr urine samples on 10 separate days. Estimated to ta l body exposure averaged 15 2 mgAu-, o f hlch 7*X as t o the leg s and feet 1th an od ltlon al 18S to the hands and arms. Estimated resp iratory exposure as about 0.Q3X o f the to ta l oody exposure. Tenty-four hr urinary 2 ,4 -0 ranged from 0.190-0.645 mg. The use o f disposatile c o v e ra lls and effectiv e hand protection ould markedly reduce th is exposure. INTRODUCTION The e xposure o f agricultural workers t o p e stic id e s has received Increasing a tten tio n from federal regulatory agencies. Markers a sso cia ted w ith the application of chemicals appear to be the aost highly exposed group w ithin the *farmworker' designation. Exposure stu d ies o f 2 ,4 -0 and 2,4,5-Trlchlorophenoxyacetlc a d d any a lso be u sefu l In estimating Vietnam veteran exposure to dioxi n In Agent Orange (1 ). O tiorophenoxyacetlc a d d herbicide exposure and rela ted stu dies have been reviewed ( 2) . The purpose o f th is study as t o determ ine the 2 ,4 -0 exposure o f Florida a irtx n t handgun herbicide applicators. EXPER IM EN TA L M aterials and tethods To, two in a lrb o a t crews fro Polk County Environmental S ervices were chosen for th is stu oy. Folk County Environmental Services i s the county agency resp onsible for osqulto and aquatic weed co n tro l. A lrboat crews are routinely used for water h yadoth control Id lek es, r iv e r s, stream s, and ca n a ls. V ita l s t a t is t ic s far the participating su b jects appear in Table 1. Table 1 . V ital s t a t is t ic s o f F lorida alrboat spray crews applying 2 ,4 -0 far aq u atic weed c o n tro l. Folk County, Florida. Subject Age H eight Ndgrrt lb (kg) Length of employment JD 27 6' 0- 155 (70.4) 1 yr 6 mo CM 54 5* 8- 165 (74.9) 6 yr KP 26 5 ' 9* 150 (68.1) . 1 yr SS 43 5* 9 -1 /2 * 165 (7 4 .9 ) . 2 yr 19117 210 These crews apply the dimethylaudne s a lt of 2,4-0 at a rate o f 2 qt of 47. 4%2, 4-0 amine/jQ water. The formulation o f 2 ,4 -0 M ine was 3 .a lbs 2, 4-0 acid equivalent per gallon tSxithcm M ill Creek Prod., In c ., Tampa, f t 33601) or a concentration in the finished spray o f 0.03a lb A l/gal (4 .6 gra A l/1 ). E itf* t u r n s o f M aico-TroitDqrift a o a itiv e (tttlco Chemical C o., Chicago, H 60636) was added to each 30 gal to control d r if t , while one crewman drove the airb oat, the other applied 3-4 tanks over a 1 hr period. The men then changed p la c es, so that each subject applied 2 ,4 -0 and operated the a lrto a t for approximately 1 hr w hile exposure was monitored. 2 ,4 -0 was applied with a Bear adjustable handgun with trigger g rip , tarin g the exposure period each man wore q -c e llu lc sc pads and a personal a ir sampler from which exposure estim ates were la ter drawn. These methods have been aescrlbed in d e ta il (3 ). The o -c e llu io ^ pads were preextracted by timelin g 10 pads in 1 1 o f methylene chloride for 10 min. Foam plugj were pxeextxacted with 40 ml o f methylene chloride fallowed by 40 ml of methanol as described below. a ir sampler exposure periods were t i med with a sampler c lock, and the foam pU^ retain ers west then covered with aluminum f o il . A ll sm iles were returned to the laboratory on ic e ana stored a t -20*C far periods ranging from 0-48 hr prior to ex tra ctio n . The a -c e llu lo se pads were sized to 2.3 In. x 2 .3 in . with a paper cu tter ana then scctiarea in to quarters with s c isso r s. Cutting surfaces were rinsed with methanol between sam ples. The quarters were placed in 8 or jars and shaken 2X in 40 ml o f methanol at 330 rpm for 10 min. V c nmalned methanol extraction s were taken almost to dryness a t 40C on a rotary evaporator ano transferred in to 10 ml o f 1:1 aceto n itrile:w a ter, ta e ml o f th is extraction was reduced to dryness in a 10 ml cen trifu ge tube on a M? evaporator a t 40C. The remaining residue was transferred in 3 rin ses o f methanol to a 230 ml b o ilin g fla sk for methyla tio n . Foam a ir sampler plugs were removed from th eir g la ss retain ers with forceps ana shaken 2X In 80 ml o f methanol for 10 min a t 330 spa. The plug was squeezed with a p e stle a fte r each extraction and the methanol was decanted in to a 230 ml b o ilin g fla sk . The combined methanol extracts were taken almost to dryness a t 40*C on a rotary evaporator. Ten ml o f methanol was added far m ethylatian. An adaptation o f the S e ll and M altlen (4) methyla tio n method was used for GUC an alyses. A g la ss b ailin g bead and 1 ml o f 8F3 in methanol (P ierce Chemical C o., Rockford, XL 61103) was added to each fla sk . This so lu tio n was bailed gently an a hot p late u n til almost dry (solution darkens and white fumes appear). Ten ml o f deionized water was added to the fla sk and i t as allaed to cool Co room temperature. Ten ml o f lsooctane was added, the fla sk was shaken for 30 s e c , and the lsooctane brouput Into the neck o f the fla sk with deionized w ater. The lsooctane transferred to a brown g la ss storage b o ttle over anhydrous sodium s u lfa te . A ll samples exe stored a t -20*C. Recove r ie s (mean S.E.M .) o f standard 2, 4-0 were: foam plug 90 2* (10 u9 le v e l); g-ce llu lose pads 9S , 2X (l"yg le v e l) . K> In terferin g m aterials were d etected in preextracted pads or foam plugs through the above procedure. Twenty-four hr urine samples were taken from noon o f each monitoring day u n til noon o f the fallow ing day. Che week a fte r spraying stopped three consecutive 24-hr urine samples were c o lle cted . Urine sample s were returned to the laboratory ano stored a t - 2CPC p rior to extraction and d erlv a tlra tio n . Urine samples era thawed, to ta l volumes measured, and 13 *1 a llxyjots plac ed in to a 230 ml b o ilin g fla sk . A g la ss bead, 13 ml o f water, and 2 ml o f 37* potassium hydroxide were added. This mixture was b o iled an a hot p late for 30 min, allow ed to co?! and transferred q u an titatively to a 230 ml separatory funnel with 100 ml o f 1 :1 methanolrwater. This mixture as extracted once w ith 23 mi o f methylene ch lorid e ana the eth ylen e chloride discarded. The so lu tio n was a c id ifie d with 2 ml o f concentrated phosphoric acid and extracted three tim es with 23 ml o f methylene ch lo rid e. The combired methylene chloride extracts ere taken to dryness on a rotary evaporator a t 40C and the residue derlvatlzed for g is chromatography as aescrlbed above. 21,1 Blank and fo rtified urines re extracted and derivatizeo with each se t o f f ie ld u rin es. Also, 100 g o f standard 2,4,5-T (2 ,4 ,5 - Trichloropnerxwyacetic acid , ethyl e s te r ) was added to each urine except for control "blanks." Internal 2,4,5-T was used to denote in e ffic ie n t extractions and/or reaction s. F o r tifie d urines contained 1.5 w9 o f standard 2 ,4 -0 . Blank urines were blank throughout. Recovery o f standard 2 ,4 -0 fro fo r tifie d urine samples averaged iQ3 *_ 3* (ean S.E.H.) . C alculation o f urinary 2 ,4 -0 included the correction for the ra tio o f M. a. 2,4-0 m ethylestex/2,4-0 a c id , i . e . , 1 .1 7 . Standard 2,4-0 acid was obtained fro Oow Chemical Co., Midland, Ml 46640, with a lis t e d purity o f 99.5*. Standard 2 ,4 -0 e th y le ste r and 2 ,4 ,5 -T acid were prepared by nogen, In c ., - W atsonville, CA 95077. Tracer 550 gas chromatographic con d ition s for the methylated product were: 1 .8 a , 2 mm 1 .0 . silan ized g la ss coluan, 5X SP210Q/1X SP24Q1 on 100-120 esh Spelcoport , 210aC in je c tio n , 165*C column, electron capture 300C, *2 40 ml/min. Methods for estim ating s p e c ific body areas fro body height and weight have been previously described (3 ). RESULTS AC DISCUSSION Table 2 contains the mean flu x (yg/c2 hr) onto the exposure pads them selves, by pad location. Comparing these means through an F -te st y ield ed no s ta t is t ic a lly s ig n ific a n t differences among workers far any o f the seven ped lo ca tio n s. Comparing the o v e r a ll means by pad location yielded F20.6, d f-5 /2 66, s ig n ific a n t a t the 99% le v e l. Five exposure groupings, differing, a t the 99X le v e l, emerge:sh in ( h ip e s t ) ; th igfi; w rist; forearm and hand; c h e st ano shoulder; and back (low est). For these groupings, hand accumulation ra te, presented in Table 2 in ag/hr, was converted in to flu x by d iv id in g by 900 cm2, a representative value for the surface area o f two hands, one o f the three accum ilation rates presented in Table 2 showed any sign ifican t difference among workers, Mean o v e r a ll estim ated to ta l body exposure was 15 ^ 2 ag/hr. The respiratory exposure estim ate was 0.002 0.001 mg/hr. ~ Table 2 . Flux o f 2,4-0 on body pads and accum ulation ra tes on body regions o f F lorida aquatic weed spray ap p licators. Suotect an jD Chi P 5s - SEh n . Flux (ug/cm? hr) * an various cads: flack Chest Foreaim Shin Moulder Thigh V rlst .07 - .02 .20 - .06 .62 .19 2.42 - .81 .i9 - .07 1.07 7 .21 .79 7 .26 .05 - .03 .1 3 - .03 .16 - .04 1.53 7 .29 .17 7 .05 .56 7 .12 .2 8 7 .06 .01 - .01 .19 - .04 .71 7 .27 2 .7 8 7 1.09 .2 0 7 .11 1.12 7 .32 .06 - .04 .41 - .11 .5 2 .20 2 .8 3 7 1.06 .23 7 .09 .78 7 .U 1.06 7 .42 cb .05 .01 .23 7 .04 .50 - .10 2.38 * .42 .20 - .04 .88 - .11 .1 0 d 39 39 39 39 39 39 39 Itl H 5 ll <'Oh " e la t i o n rate '8'hr) on body T^dlor-________ Hands fep lratoiy to ta l body .24 .04 001 7 .001 1 6 -4 .33 * .07 .00 1 .0 0 0 10-2 .61 - .11 .7 0 - .30 .46 .08 .0 0 2 * .001 .003 T .002 .002 - . .001 17 7 4 1 8 -5 1 5 -2 39 38 39 over saopling period, per su b je ct, error o f the mean. 2Jtln over sampling period, a ll su b jects coccined. o f rep lication s, a ll su b jects cocoin ed . ' 18119 212 F -test comparisons o f * exposure to body areas showed no sig n ific a n t differen ce aoorq orders, and these data were coocined in Table 3. Approximately 7AX o f the e s timated tota l (wv exposure was to the le g s and f e e t . Arms ( -1QX) and hands ( -6*) ere next, followed by ctest* head and neck, and back. Respiratory exposure, as a percentage of the estimated to ta l body . ' exposure, was low. Table 3- Percent o f estimated to ta l body exposure of Florida 2 ,4 -0 aquatic weed spray applicators to their various body areas. Body area Head neck Chest Back Aims Hands Legs fe e t Resp. Hsan X _ SEX 4b 1.4 0.1 5.8 0.6 1-2 * 0.3 9.8 - 1.0 8.4 1.2 73.6 . 1.9 .028 .009 n 40 40 40 40 40 40 38 Hear percent o f estim ated to ta l body exposure, a ll su b jects cw o in ed . ^Standard error o f the mean. cKmi>er o f r e p lic a tio n s, a ll subjects com ined. C orrelation o f urlr-iry ? ,* - ft~^<^-"*Tttmatrn f o n i tEE) i i m * i i i m piowort i r n i r r i tifi|l (Table 4 ). This d iffic u lty I s I llu s tr a te d by subject CH. CMhad the h igh est aean urinary ex cretion ' concentration and r a te , but th e low est mean estim ated to ta l body exposure. The d iffic u lty o f ' correlating estim ated 2 ,4 -0 exposure with urinary m etabolites has been noted by Al-Jabery (5 ). Lavy e t a l (4 ) a lso had d iff ic u lt y in correlatin g estim ated exposure to urinary m etabolites in 2 ,4 ,5 r f applicators. 2 ,4 -0 i s rapidly excreted by man. Seuexhoff e t al_. (7 ) recovered 87.4-106.3X o f the Ingested dose o f 2 ,4 -0 in human urine over 14* h r. The elim ination h a lf - lif e ranged from 10-2 0.4 -2 8 .4 _ 2 .4 hr far a in g le 5 mg/Vg oral dose. Kohll e t a l. (8 ) recovered 73* o f a 3 mgy*4 oral dose**9f 2 ,4 -0 in huaan urine over 94 hr a fter adm inistration, w ith an elim in a tio n half-U *e o f 33.0 3 .1 h r . Feldman and Haibach (9) recovered from human urine 100X o f an i . v . dose o f 2 ,4 -0 over a 120 hr p eriod , th e elim ination h a lf - lif e being 13 hr; however, only 3 .8 , 2-4X o f * top ical dose to the forearm was recovered in 120 h r. Ac cording to th ese au th ors, the**mriatlch observed in dermal penetratio n re s u lted froa p h ysiological variation among the su b je cts. This variation, combined w ith low dermal penetration, overlapping d oses, and v a ria b le exposure estim ates probably precludes ary correla tio n o f 2,4-4) dermal dose and urinary 2 ,4 -0 in our ****** In the Lavy e t a l . (6 ) stu d y , 2,4,5-T backpack sprayers were dermally exposed to 0-20-2-5* ag/kQ and excreted 0.0 2 3 -0 .0 7 4 mg/Vcg 2 ,4 ,3 -T . Oexmal exposure for Florida 2 ,4 -0 handgun applicators ranged from 0-134-0.250 mg/kg and they excreted 0.0027-0.0084 mg/kg. ftjte n tia l exposure for F lorida a p p lica to rs was about fiv e tim es le s s , and excretion about ten tim es l 53* 1C a Table 4 . Mean u rin a ry e x c re tio n o f 2 ,4 - 0 by s u b je c t, and c o r r e la tio n between d a lly u rin a ry e x c re tio n and d a lly estimated total body exposure (ETBE). JO CH KP SS Mean ET B E (m g/hr) from Tab le 2 Mean u rln e ry 2 ,4 - 0 (ppb) Mean u r ln e r y 2 ,4 - 0 (g /2 4 h r ) 1 6 * 1 *b o f 116 * 1? (10 ) 190 * 27 (10 ) 10 * 2 (10) 671 * 198 (101 645 * 160 (10 ) 17 * 4 (10) 404 1 67 (8 ) 315 * 53 (8 ) 18 1 5 19) 464 * 161 (9) 402 * 120 iy ) O e l l y ppb 2 ,4 - 0 v s . same day ET B E D a l l y vg/24 h r 2 ,4 - 0 v s . same day ET B E D a lly ppb 2 ,4 -0 v s . previous day ETBE O a ily vg/24 hr 2 ,4 -0 v s . previous day ETBE , 4 4 d -.0 3 -.0 8 *.2 2 -.0 0 -.1 6 . 7 6 *.6 9 .3 4 .3 4 -.9 6 .23 -.3 4 -.3 2 -.3 6 -.3 0 Mae a n o v e r s a m p l i n g p e r i o d . ^Standard e rro r o r the mean. Climber o f re p lic a tio n s . ^Correlation c o e ffic ie n t. 213 19121 I 214 than for forest orders. The spray concentration o f 2 ,4 ,S-T as 0.16 10 A I/gal compared to 0.036 10 A l/gal 2 ,4 -0 in th is study. Under sim ilar application conditions, the p esticid e concentration apparently determ ines exposure le v e l, a point also made oy other authors ( 1qj ^ OXXUSIO^ froa the pattern o f exposure observed here, a practical reouction cxild oe afforaeo by the ^ 1 use o f disposable c o v e r a lls and hand p rotection . Exposures reported here are low and would n j appear to represent e ith e r acute or chronic hazard to Florida handgun 2,4-0 a p p lica to rs. /OUOHUEDCEhEKT Me acknowledge the tetfm lca l assista n ce o f R. Queen, T. Bailey, S. P o lin sk i, R. M lllia*s, and R. Hartman. Although the research described in th is a r tic le has been funded wholly or in part by the US Environmental P rotection Agency throu^i Grant tinder R60647A to the University of Florida, i t has not been subjected to Agency review and therefore does not n e cessa rily reflect the views o f the Agency and no o f f ic ia l endorsement should be Inferred. REFERSfC5 1. K.M. Stevens. Agent orange to x ic ity : a quantitative perspective. Hman T oxicol. JL, 31 usai). 2. H-L. Lerg, J.C . Ramsey, V.H. Braun, and T.L. Lavy. Review o f studies with 2,4,5-T rlchlorophenoxyacetic acid in humans including applicators under f ie ld conditions, pp. 133-136, in P esticid e residues and exposure (ed. J. R. Plimmer]. ACS Syap- S er. #162, Amer. Che*. S o c ., NT (1982). i 3. C.A. Mojeck, H.N. td gg, J.H. Stamper, and O.E. Bradway. worker exposuire to eth io n in Florida c itr u s . Arch. Environ. Gontam. Toxicol. 10, 723 (1961). 4. C.R. Sell' and J.C . H altlen . Rapid procedure for the determination of 2 .4- O ichlcrophenoxyacetic acid in dermal exposure pads, urine, and p ersp iration taken fioa agricultural workers exposed to the h erb icid e. J. Agr. Food Che*, (in p r e s s). i 3 . X.A.R. Al-Jabery. P e stic id e exposure stu d ies: D irect and indirect d etectio n o f adsorption o f 2 ,4 -0 and pzonamide herb icid es In the guinea pig and occupationally exposed workers. Q lss. Abstr. In t. 8 . 41, 2947 (1981). 6 . T.L. Lavy, J .S . Shepard, and J.O . M attlce. Exposure aeasureaents o f a p p lica to rs spraying (2,4,3-Trichlorophenoxy) a c e tic a d d In the fo re st. J . Agrlc. Food Che*. 28, 6 2 6 (1960)- 7 . M.. Sauexhoff, V.H. 8raun, G.E. 81aln, and P .J. Gehrlng. The fa te o f 2.4- d e h lorophenoxy a c g tic acid (2 ,4 -0 ) follow ing oral administration to man. To x ico l ogy * , 3 (1977). B. J.O. K ohll, R-M. Khana, 8.H . Gupta, H.M. Ohar, J .S . Tandon, and K.P. S ir c a r . Absorption and excretion o f 2 ,4-Qichlorophenoxym cetlc acid in man. Xenobiotlca 4,, 97 (1 9 7 4 ). 9 . R.J. Feldaan and H .I. Halbach. Percutaneous penetration o f some p e stic id e s Mid herbicides Inman. T oxicol. Appl. Riazmacol. 28, 126 (1974): 10. H.R. M olfe, J-F . Armstrong, and w.F. Ouxham. P esticid e exposure fro* con cen trate spraying. Arch. Environ, tilth . 13, 340 (1966). 19122 21S 11. C.A. ojeck, H.H. Klgg, R*s* firaaan, J.H. Stamper, and R.L. R ouseff. Worker exposure to arsenic In Florida grapefruit spray o p era tio n s. Aron. Environ. Gontaa. T oxicol. In press. (1982). Deceived in The Ffetherlands iSeries 3. *099. . F lorid a A gricultural Experiment S ta tio n s Jcxirnal (fteoelved in OSA 9 December 1982) 131 Cancer Causes and Control, 3, 25 -3 0 Occupation and soft-tissue sarcoma in northeastern Italy Diego Serraino, Silvia Franceschi, Carlo La Vecchia, and Antonino Carbone (Received. 12 A u g u st 1991; accepted in revised form 30 September 1991) The influence of occupation and exposure to different agents on the risk of developing soft-tissue sarcoma (STS) was assessed in a case-control study based on 93 cases of STS (53 men and 40 women) and 721 controls (371 men and 350 women), conducted in northeastern Italy. N o risk elevation was found in subjects employed in agriculture (odds ratio [OR] for > 10 years = 0.8,95 percent confidence interval [Cl] = 0.4-1.5), nor in those who reported exposure to pesticides or herbicides (OR = 0.4, Cl = 0.1-1.2). Similarly, neither occupation in the furniture, upholstery, and mechanics industries, nor exposure to livestock or meat processing, wood dust, metal dust, and dyes or paints were associated with STS risk. Workers who reported exposure to chemical agents or to benzene or other solvents for more than 10 years had, respectively, a 1.8-fold (Cl = 0.7-4.4) and a 2.2-fold (Cl = 0.9-5.5) higher risk of developing STS. Although the small number of STS cases limits the interpretation of the study results, these findings weigh against the hyphothesis that pesticides, herbicides, or other exposures related to agriculture, play an important role in the etiology of STS. The direct associations with exposure to chemical agents and benzene or other solvents, albeit not statistically significant, may pro vide a useful hint for future investigations. Key words: Agriculture, benzene, case-control study, chemicals, Italy, occupation, soft-tissue sarcoma. Introduction Soft-tissue sarcomas (STS) are a group of rare and heterogeneous malignant neoplasms1'3whose etiology is still largely unknown.4 Most of the epidemiologic studies have focused on STS risk associated with occu pation in agriculture and related exposures, but incon sistent results have been reported. Two Swedish case-control studies, conducted in the late 1970s and early 1980s, showed a nearly sixfold excess risk in individuals exposed to phenoxyacetic acid herbicides and chlorophenols.5-6An excess of STS also was noted among industrial workers exposed to herbicides7 and among farmers in Washington State (United States).3Other studies, including arecord link age between occupational histories and the Swedish Cancer Environmental Registry9 and a study on a cohort of licensed pesticide applicators,10did not con firm such association. Little information is available on the influence of occupations other than agriculture and related exposures on the risk of developing STS.41'-12 In the present study, the role of agricultural and nonagricultural occupations and exposures in STS risk was assessed, using data from acase-control study conduc ted in northeastern Italy. D rs S e r ra in o a n d F ra n cesch i a re w i t h th e E p id e m io lo g y U n it, a n d D r C a r b o n e is w i t h th e D e p a r tm e n t o f P a th o lo g y a t th e A v ia n o C a n c e r C e n te r, A v ia n o , Ita ly . D r L a V ecch ia is w i t h th e "M a rio N e g r i " I n s titu te f o r P h a rm a c o lo g ic a l R esea rch , M ila n , I ta ly a n d th e I n s titu te o f S o c ia l a n d P re v e n tiv e M edicine, L ausanne, S w itze rla n d . A ddress correspondence to D r Serraino a t th e E pidem iology U nit, A v ia n o C ancer C enter, Via P edem ontana Occ. 33081 A v ia n o (P N ), Ita ly. T he w o rk w as supported b y th e co n tribution o f the Italian Association fo r C ancer Research, M ilan, a n d th e Ita lia n N a tio n a l R esearch C o u n c il ( C N R A p p lied P roject "C lin ica l A p p lica tio n o f O ncological Research, " C ontract 87.01344.44). 9 1992 Rapid Communications of Oxford Ltd Cancer Causes and Control. Voi 3. 1992 25 D . Serraino ec al Materials and methods Since June, 1985, we have been conducting a hospitalbased case-control study on risk factors for STS in the Friuli-Venezia Giulia region, northeastern Italy. Methods and results concerning the association be tween STS and nonoccupational risk factors (based on 88 cases and 610 controls) have been described elsewhere. Cases The cases were histologically confirmed STS, diag nosed within two years before interview, who had been admitted as inpatients or referred for follow-up to the outpatient clinics of the Aviano Cancer Center or to the general hospitals in the study area. The present analysis is based on 93 cases (53 men and 40 women), aged 16 to 79 years (median age: 52 years), interviewed before June 1991. According to the classification of Enzinger and Weiss,13the most frequent cell type was sarcoma of the fibrous tissue (43 percent), followed by myomatus sarcomas (17 percent), liposarcomas (13 percent), and various other cell types (27 percent). Kaposi's sarcoma linked to human immunodeficiency virus infection and visceral sarcomas were excluded. All the histologic specimens were reviewed by one pathologist (A.C.). Controls The control group consisted of 721 patients (371 men and 350 women), aged 17 to 79 years (median age: 54 years), admitted to the study hospitals for a wide spec trum of diseases. Specifically excluded from the com parison group were all patients whose diagnoses at hospital admission included malignant disorders, con ditions related to alcohol and tobacco consumption, or any disease which might have resulted in diet modifi cation, e.g. disorders of respiratory and digestive tracts, cardiovascular diseases, diabetes, etc. Hematologic, allergic, or autoimmune diseases also were excluded. Twenty-eight percent of the controls were admitted for traumatic conditions (mainly fractures and sprains), 32 percent for musculoskeletal diseases (mainly low-back pain and disk disorders), 17 percent for minor conditions requiring surgical operations, and 23 percent for other illnesses such as disorders of the ear, nose, throat, skin, or teeth. Questionnaire Information was elicited on socioeconomic and anthropometric indicators including: education and social class (based on the head of the household's occu pation);1'*smoking and consumption of alcohol, and of 26 Cancer Causes and Control. Vol 3. 1992 14 selected food items; and medical history. The ques tionnaire also included information on the age at start ing and stopping employment in 17 industries or occupations, on the subject's role in the industry or occupation in terms of direct involvement in pro duction aspects, and on exposure to 15 occupational agents or groups of agents. Exposure histories were truncated, for both cases and controls, at the time of diagnosis. D a ta analysis For each occupation or exposure with at least five `exposed' cases, odds ratios (OR), adjusted for age and sex by the Mantel-Haenszel procedure,15and their 95 percent confidence intervals (Cl),1'1 were computed. Sex- and age-adjusted tests for trend in risk were based on the Mantel chi-square test.17 Results STS cases and controls were comparable regarding sex, age, education and social class (Table 1). The overall percentage of industrial manual workers among STS cases (45 percent) and controls (38 percent) were also similar (not shown in Tables). Table 1. Distribution of cases of soft-tissue sarcomaand con trols bv sex, age, education, and social class. Pordenone, Italy, 1985-91 tos. ft Sex M ale Fem ale XT Age (years) 39 40-49 50-59 60-69 x; E ducation (years) 5 6-8 3=9 Xf f o r tr e n d Social class I-II (highest) III IV -V Xf f o r tre n d O ther C ases (* = 93) No. % C ontrols (" = 721) No. % 53 57 371 52 40 43 350 49 0.80, F = 0.37 21 23 179 25 17 18 118 16 23 25 165 23 19 20 168 23 13 14 91 13 0.89, F = 0.93 54 58 425 59 14 15 143 20 25 27 153 21 0.53, F = 0.47 78 34 5 27 29 211 29 55 59 446 62 0.73, F = 0.39 44 30 4 19125 Risk factors for soft-tissue sarcoma Table 2. Distribution of cases of soft-tissue sarcoma and controls by duration of various occupations or exposures, Pordenone, Italy, 1985-91 O ccupation' A griculture C hem ical Rubber D vescuff P a in tin g P rin tin g P e tro le u m Pharm aceutical F urniture/upholstery A sbestos C oal/gas Food processing M echanics L eather tanning N uclear Exposure L ivestock/m eat processing C hem ical agents Gas M etal dust Plastic resins/glues O il R adiation D ves/paints W ood dust B en zen e/so lv en ts A sbestos E lectricitv/radar C oal tar H erbicides/pesticides A lum inum 1-10 Years No. % Cases ( = 93) > 10 Y ears No. % 2 2.2 1 1.1 1 1.1 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 1.1 0 0.0 2 2.2 1 1.1 2 2.2 0 0.0 0 0.0 17 18.3 1 1.1 0 0.0 1 1.1 0 0.0 0 0.0 0 0.0 0 0.0 7 7.5 1 1.1 0 0.0 2 2.2 5 5.4 0 0.0 0 0.0 2 2.2 2 2.2 1 li 6 6.5 1 1.1 0 0.0 0 0.0 4 4.3 5 5.4 2 2.2 0 0.0 0 0.0 0 0.0 3 3.2 2 2.2 10 10.3 8 8.6 3 3.2 5 5.4 3 3.2 4 4.3 0 0.0 4 4.3 6 6.5 3 8.6 1 1.1 1 1.1 1 1.1 3 3.2 0 0.0 C ontrols __________________ ( = 721)_______________ 1-10 Y ears > 10 Y ears No. % No. % 34 4.7 2 0.3 1 0.1 4 0.6 6 0.8 2 0.3 2 0.3 0 0.0 36 5.0 0 0.0 5 0.7 3 1.1 21 2.9 2 0.3 0 0.0 149 20.7 1 0.2 1 0.1 1 0.1 6 0.8 1 0.1 0 0.0 2 0.3 45 6.2 0 0.0 2 0.3 3 1.1 40 5.5 0 0.0 0 0.0 12 1.7 16 2.2 8 1.1 26 3.6 18 2.5 12 1.7 1 0.1 32 4.4 47 6.5 29 4.0 4 0.6 0 0.0 6 0.8 29 4.0 13 1.8 58 8.0 35 4.9 17 2.4 38 5.3 21 2.9 14 1.9 2 0.3 32 4.4 46 6.4 27 3.7 4 0.6 3 0.4 5 0.7 56 7.8 14 1.4 ` O n ly individuals em ployed in p ro d u ctio n aspects are included. The distribution of STS cases and controls according to the duration of employment in major industrial occupations and reported exposure to various agents, or groups of agents, is provided in Table 2. Only occu pations in agriculture, and the furniture or upholstery and mechanics industries, as well as exposures to live stock or meat processing, chemical agents, metal dust, dyes or paints, wood dust, solvents or benzene, and herbicides or pesticides were reported by at least five STS cases. The related ORs are given in Table 3. No significant association of STS risk with occu pation in agriculture (OR for >10 years = 0.8, Cl = 0.4-1.5), or exposure to herbicides or pesticides (OR = 0.4, CI = 0.1-1.2) emerged (Table 3). Con versely, an approximately twofold increase in STS risk was noted among subjects who reported more than a 10-year exposure to chemical agents (different from pesticides or herbicides) (OR =1.8, Cl = 0.7-4.4) or to benzene or other solvents (OR = 2.2, CI = 0.9-5.5) (Table 3). Analyses also were performed for men and women separately. Although no significant differences be tween sexes emerged, it should be noted that STS risk was somewhat higher in women exposed to benzene or other solvents for more than 10 years (OR = 3.7, Cl = 0.7-17.0) than in men (OR = 1.8, Cl = 0.6-5.6) (not shown). Discussion The present results provide further evidence against the possibility that employment in agriculture and, in Cancer Causes and Control. Vol 3. 1992 27 19126 D . Serraino et al Table 3. Odds ratios (OR)4and 95% confidence intervals (Cl) for soft-tissue sarcoma for selected occupations and exposures,14according to duration. Pordenone, Italy 1985-91 O ccupation o r exposure D uration o f occupation or exposure < = 10 y ea rs > 10 years OR Cl OR Cl A griculture 0.4 0 .1 -1 .7 0.8 0 .4 -1 .5 F urniture/ upholstery M echanics 0.2 0 .0 -1 .5 1.1 0 .4 - 2 .7 0.7 0.1-3.3 0.9 0 .3 -2 .4 L ivestock/m eac processing 1.4 0 .2 -7 .0 1.4 0 .6 - 3 .0 C h e m ic a l a g e n ts 1.1 0 .2 -5 .6 1.8 0 .7 - 4 .4 M etal dust 1.8 0 .6 -5 .0 0.9 0 .3 - 2 .6 D yes/paints 0.9 0 .3 -2 .9 0.9 0 .2 -2 .7 W ood dust 0.8 0.3-2.3 0.9 0 .3 -2 .4 Benzene/ solvents H erbicides/ pesticides 0.5 0 .1 -2 .2 2.2 0 .9 -5 .5 0.7 0.2-2.5 0.4 0 .1 -1 .2 X8, (tre n d ) 0.79, F = 0.37 0.11, F = 0.74 0.19, P = 0.66 0.96, F = 0.33 2 .1 3 ,P = 0.15 0.13, F = 0.72 0.07, F = 0.79 0.09, P = 0.77 1.90, P = Q .U 3.14, F = 0.08 ' N o n ex p o sed su b jects as reference category; adjusted fo r age in quinquennia, and sex, using the M antel-H aenszei procedure. b O n ly o ccu p atio n s and exposures w ith at least five exposed cases. particular, exposure to pesticides or herbicides play an important role in the etiology of STS. Because the study was based on only 93 cases of STS, there was insufficient power to evaluate all the subgroups-- exposed to specific agricultural factors--that might be at risk. However, the study was able to exclude, at the level of statistical significance, an increased STS risk of greater than 20 percent from more than 10-years' exposure to herbicides or pesticides. This may be a use ful contribution, considering the rarity of the disease and the recent debate on this topic. In fact, the results of some epidemiologic studies seem to indicate that farmers have elevated risks for a few nonepithelial cancers, including STS.1819However, the evidence remains inconclusive. A risk of 0.9 was found in a cohort of Swedish agricultural and forestry workers potentially exposed to phenoxy acid herbi cides,9and acase-control study conducted in New Zea land was also negative." STS risks lower than unity were found among agricultural and forestry workers and gardeners in astudy based on 15 cancer registries in England and Wales.81 Moreover, no evidence of increased STS risk emerged among professional pesti 28 Cancer Causes and Control. Voi 3. 1992 cide applicators in the US" or in Germany.- A popu lation-based case-control study of STS and malignant lymphoma from Kansas (US) showed a nearly sixfold higher risk of developing non-Hodgkin's lymphoma among herbicide users, but no increased risk for STS.84 In Italy, the association between STS risk and agri cultural exposure to phenoxy herbicides has been investigated in one population-based case-referent study from an area where rice growing is a predomi nant agricultural activity.85The study, which included 68 cases (37 men and 31 women) and 158 controls, found a relative risk of 0.9 in men and a nearly threefold, nonsignificant, excess STS risk among women who potentially were exposed to phenoxy herbicides.85 Among nonagricultural occupations, an increased, albeit inconsistent, risk of STS among woodworkers has been reported.11-86One possible explanation for the reported excess STS risk among woodworkers is the frequency of exposure to phenoxy acid herbicides or chlorophenols in that occupation.4:rThe present inves tigation, in agreement with other studies on wood working and STS risk which showed negative results,8S-89found no association either with occupation in the furniture or upholstery industry or with reported exposure to wood dust. The slightly increased risk of STS associated with livestock handling or meat processing, although not statistically significant, seems to be consistent with an excess STS risk among abattoir workers reported by Pearce et Eight STS cases (seven men and one woman) in this study reported an exposure to chemical agents for more than 10 years, with a nearly twofold, albeit non significant, increased risk of STS. Three men reported exposure to sodium bicarbonate and two to ferrocyanide. Exposure to other chemicals (e.g. formaldehyde) found to be important in some studies on STS'1-'8was not documented in the present investigation. Similarly, STS cases exposed for more than 10 years to benzene or other solvents had a nearly twofold, although nonsig nificant, increased risk of STS. Although benzene exposure was associated consistently with an elevated risk of acute leukemia,33-35 the few studies which have investigated the association of benzene exposure with STS showed very modest, nonsignificant risk elevations.36-57 Both positive and negative results should be viewed with caution, considering the limits of the present study. Interview data were not validated with infor mation from other sources, and questions were re stricted to a selected series of general items. Importantly, not all potentially relevant occupations (e.g. leather industry38 were well-represented in the study area and the overall number of exposed cases in most industrial occupations precluded detailed analv- -s 19127  Risk factors for soft-tissue sarcoma ses of time-risk relationships and more precise esti mations of ORs. It is unlikely, however, that these findings are affected seriously by bias, since there is little scope for supposing a differential recall between cases and controls for these broad categories of occu pations and exposures. As yet, no hypothesis on STS etiology has gained sufficient attention in Italy to rep resent a major distorting factor. Moreover, a com panion study of lymphoid neoplasms, based on the same design and data collection method, actually did find elevated risks among individuals employed in agriculture.;g+= In conclusion, the present study, which took place in an area where approximately one-quarter of the study population had a history of employment in agriculture, weighs against the possibility that herbicides and pesti cides, or other exposures related to agriculture, play an important role in the etiology of STS. The ORs above unity for exposure to benzene or other solvents and chemicals agents, albeit statistically nonsignificant, may provide useful leads for future investigations. Acknowledgements--We thank Ms Tiziana Angelin, Maria Grazia Valentini, and Derna Gerdol for inter viewing cases and controls, and Ms Ilaria Calderan for editorial assistance. References 1. Pickle LW, Mason TJ, Howard N , Hoover R, Fraumeni JF. Atlas of US Cancer Mortality Among Whites: 19501980. Washington, DC: US Dept of Health and Human Services, Public Health Service, National Institutes of Health, 1987; D H H S Pub. N o. (NIH ) 87-2900, 180. 2. Levi F, Maisonneuve P, Filiberti R, La Vecchia C, Boyle P. Cancer incidence and mortality in Europe. Soz Praeventivm ed 1989; 34 (Suppl. 2): S32-S33. 3. La Vecchia C, Negri E, Decarli A, Fasoli M, and Cislaghi C. Cancer mortality in Italy: an overview of age-specific and age-standardized trends from 1955 to 1984. Tumori 1990;76:87-166. 4. Tucker MA, Fraumeni JF. Soft-tissue. In: Schottenfeld D, Fraumeni JF Jr, eds. Cancer Epidemiology and Pre vention. Philadelphia, PA: WB Saunders Company, 1982: 827-36. 5. Hardell L, Sandstrom A. Case-control study: Soft-tissue sarcomas and exposure to phenoxy acetic acids or chlorophenols. Br J Cancer 1979; 39: 711-17. 6. Eriksson M, Hardell L, Berg N O , Moller T, Axelson O. Soft-tissue sarcomas and exposure to chemical sub stances: a case-referent study. Br J Ind M ed 1981; 38: 27-33. 7. Ott MG, Holder BB, Olson RD. A mortality analysis of employees engaged in the manufacture of 2,4,5,-trichlorophenoxyacetic a c id ./Occup Med 1980; 22:47-50. 8. Milham S Jr. Occupational mortality in Washington State 1950-1979. Cincinnati, OH: National Institute for Occupational Safety and Health, 1983; DHHS Pub. No. (NIOSH ) 83-116. 9. Wiklund K, Holm LE. Soft-tissue sarcoma risk in Swed ish agricultural and forestry w o rk ers./N C / 1986; 76: 229-34. 10. Wiklund K, Dick J, Holm LE. Soft-tissue sarcoma risk in Swedish licensed pesticide applicators. J Occup Med 1988; 30: 801-4. 11. Zahm SH, Blair A, Holmes FF, Bovsen CD, Robel RJ, Fraumeni JF Jr. A case-control study of soft-tissue sar coma. Am J Epidemiol 1989; 130: 665-74. 12. Serraino D, Franceschi S, Talamini R, Frustaci S, La Vec chia C. Non-occupational risk factors for adult soft-tis sue sarcoma in northern Italy. Cancer Causes Control 1991;2: 157-64. 13. Enzinger FM, Weiss SW. Soft-tissue Tumors. St Louis, MO: CV Mosby, 1983. 14. Office of Population Census and Surveys. Classification o f Occupation 1980. London: HMSO, 1980. 15. Mantel N , Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease./iVC/ 1959; 22: 719-48. 16. Cornfield J. A statistical problem arising from retrospec tive studies. In: Neyman J, ed. Proceedings o f the Third Berkely Symposium, IV. Berkelv, CA: University of California Press, 1956. 17. Mantel N . Chi-square test with one degree of freedom: extension of the Mantel-Haenszel procedure. J Am Stat Assoc 1963; 58: 690-700. 18. Blair A, Malker H, Cantor KP, Burmeister L, Wiklund K. Cancer among farmers. A review. Scand J Work Environ Health 1985; 11: 397-407. 19. Blair A, Zahm SH. Herbicides and cancer: a review and discussion of methodologic issues. In: Band EP, ed. Recent Results in Cancer Research. Occupational Cancer Epidemiology. Berlin-Heidelberg, Germany: SpringerVerlag, 1990; 120: 132-45. 20. Smith A H , Pearce NE, Fisher D O , Giles HJ, Teague CA, Howard JK. Soft-tissue sarcoma and exposure to phenoxyherbicides' and chlorophenols in N ew Zealand. J N C I 1984; 73: 1111-17. 21. Balarajan R, Acheson ED. Soft-tissue sarcomas in agri culture and forestry workers. J Epidemiol Community Health 1984; 38: 113-6. 22. Wang H H , MacMahon B. Mortality of pesticides appli cators./ Occup M ed 1979; 21: 741-4. 23. Barthel E. Cancer risk in pesticide-exposed agricultural workers. Arch Geschwulstforsch 1981; 51: 579-85. 24. Hoar SK, Blair A, Holmes F. Agriculture herbicide use and risk of lymphoma and soft-tissue sarcoma. JAMA 1986; 256: 1141-7. 25. Vineis P, Terracini B, Ciccone G, et al. Phenoxy herbi cides and soft-tissue sarcomas in female rice weeders. A population based case-referent study. Scand J Work Environ Health 1987; 13: 9-17. 26. Smith A H , Fisher DO, Pearce NE, Teague CA. D o agri cultural chemicals cause soft-tissue sarcoma? Initial find ings of a case-control study in N ew Zealand. Community Health Stud 1982; 6: 114-19. 27. Eriksson M, Hardell L, Berg N O , Moller T, Axelson O. Cancer Causes and Control. Vol 3. 1992 2 9 13128 4 D . Serraino et al t Soft-tissue sarcomas and exposure to chemical sub update of mortality among chemical workers exposed to stances: a case-referent study. Br J Ind Med 1981; 38: benzene. BrJ Ind Med 1986; 43: 685-91. 27-33. 34. Yin SN, Li GL, Tain FD, et al. Leukemia in benzene 28. International Agency for Research on Cancer. Wood, workers: a retrospective cohort studv. Br J Ind Med Leather and Some Associated Industries. Lyon, France: 1987;44:124-8. IARC, 1981; I A R C Monogr Eval Carcinog Risk Chem 35. McCraw DS, Joyner RE, Cole P. Excess leukemia in i Humans, VoL 25. refinery population./ Occup M ed 1985; 27:220-2. 29. Sterling TD, Stoffman LD, Sterling DA, Mate G. Health 36. Woods JS, Polissar L, Severson RK, Heuser LS, Kulan- effects of chlorophenol wood preservatives on sawmill der BG. Soft-tissue sarcoma and non-Hodgkin's lym workers. In tJ Health Serv 1982; 12:559-71. phoma in relation to phenoxyherbicide and chlorinated 30. Pearce N, Smith AH, ReifJS. Increased risk of soft tissue phenol exposure in Western W ashington./IVCf1987; 78: sarcoma, malignant lymphoma, and acute myeloid leu 899-910. kemia in abattoir workers. A m J Ind Med 1988; 14(1): 37. Eriksson M, Hardell L, Adami H O . Exposure to dioxins 63-72. as a risk factor for soft-tissue sarcoma: a population- 31. Stayner LT, Elliot L, Blade L, Keenlyside R, Halperin W. based case-control study. J N C I 1990; 82:486-90. A retrospective cohort mortality study of workers 38. Constantini AS, Paci E, Miligi L, Buiatti E, Martelli C, exposed to formaldehyde in the garment industry. Am J Lenzi S. Cancer mortality among workers in the Tuscan Ind M ed 1988; 13(6): 667-81. tanning industry. BrJ Ind M ed 1989; 46(6): 384-8. 32. Hardell L, Moss A, Osmond D, Volberding P. Exposure 39. La Vecchia C, Negri E, D 'Avanzo B, Franceschi S. to hair dyes and polychlorinated dibenzo-p-dioxins in Occupation and lymphoid neoplasms. BrJ Cancer 1989; AIDS patients with Kaposi's sarcoma: an epidemiolog 60: 385-8. ical investigation. Cancer Detection Prev 1987; Suppl. 1: 40. Franceschi S, Serraino D, La Vecchia C, Bidoli E, Tirelli 567-70. U. Occupation and risk of Hodgkin's disease in North- 33. Bond GG, McLaren EA, Baldwin CL, Cook RR. An East Italy. In t J Cancer 1991; 48: 831-5. 3 0 Cancer Causes and Control. Vol 3. 1992 13129 M ed. O ncol. & T u m o r P harm acoiher. V ol. 4, N o. 3/4, pp. 207-2L 7, 1987 P rinted in G re a t B ritain 1 0736-0118/87 $3.00 + .00 Pergam on Journals Ltd. PESTICIDES AND CANCER RISKS IN AGRICULTURE OLAV AXELSON Department of Occupational Medicine, University Hospital, 581 85 Linkping. Sweden (Received 22 May 1987; accepted 1June 1987) The work environment in agriculture is complex, with many potentially hazardous exposures, but the overall mortality from cancer and other causes is rather low among farmers. However, several studies have consistently indicated an excess of certain cancer forms. Lymphomas, leukemias, multiple myeloma and also malignancies of connective tissue attract especial interest, as being possibly associated with the use of pesticides. Phenoxy acid herbicides may play an etiological role, especially for non-Hodgkin's lymphoma, whereas the findings are more ambiguous for Hodgkin's disease and soft-tissue sarcoma, perhaps indicating an interaction with co-factors. The issue has been controversial for many years, however, and one of its aspects involves the use of phenoxy acids in the Vietnam war. Furthermore, DOT has been associated with lung cancer in mixed exposure situa tions, and with chronic lymphatic leukemia. Arsenical pesticides may have caused skin cancer in vine-growers. Further studies, especially of specific user groups and producers, may avoid the complex exposure situation in agriculture. Key words: DDT, Phenoxy acids, Leukemia, Lymphoma, Sarcoma. INTRODUCTION Agriculture means a complex work environment with exposures to dusts of organic and inorganic materials, micro-organisms including zoonoic viruses, pesticides, fertilizers and fumigants. Other pertinent exposures may relate to repair and main tenance of machinery and buildings, often implying heavy but short time exposures to agents like gaso line, oils, lubricants, paints and wood preservatives, mineral fibres, etc. Engine exhausts are inherent in modem agriculture, and other exposure to combus tion products may derive from the burning of waste material, e.g. plastic sacks used for transportation of fertilizers or animal food. Usually, inhalation would be the most important route of exposure to these various agents, but skin absorption can also be con siderable for some chemicals. In spite of the many potentially adverse occupa tional exposures, that might be identified in the agri cultural environment, the overall mortality of farmers tends to be low in comparison to that found for other occupational groups.1-2'3,4 Tobacco and alcohol-related disorders especially, including can cer, seem to occur less frequently among fanners, as reported by several authors.5'6,7 Many of the poten tially adverse exposures that nowadays can be found in agriculture appeared only recently, and may therefore not yet have caused any clear effects on health and survival. The exposure to pesticides in agriculture nevertheless necessitates particular con cern, since these agents exert strong biological acti vities and therefore have a potential also for adverse health effects. These involve cancer risks as subject to consideration in this review, especially from the epidemiological point of view and with emphasis on the controversial phenoxy acid herbicides. PESTICIDES AND PESTICIDE EXPOSURE In contrast to other chemicals (except drugs), pes ticides are subject to special registration in many countries, after some sort of evaluation of the toxic properties. This procedure, together with recom mendations for protection, might be thought of as hindering untoward effects on humans. At the time of registration, there has often been incomplete toxicologic information, however, especially on long-term effects. This point can be well illustrated by the fact that a working group at the International Agency for Research on Cancer (IARC), as late as in March 1987, had to conclude that the information on carcinogenicity from animal testing was inad equate for such widely used pesticides as the 208 O la v Axelson phenoxy acid herbicides 2,4-D and 2,4,5-T; for MCPA there were no data.3 Many years after their introduction on the market, several pesticides have become increasingly sus pected of carcinogenic activity. Still, relatively few of these compounds have been labelled as carci nogens or as probably or possibly carcinogenic to humans, e.g. referred to as IARC groups 2A and 2B, respectively.3 Some examples in this respect are shown in Table 1; none has been classified as carci nogenic without reservations (IARC group 1). Also, those compounds considered as possible or (espe cially) probable carcinogens may be thought of as being most likely to actually cause cancer in humans. The uptake of pesticides by exposed workers has been studied in some instances and seems to be best assessed for phenoxy acids. Hence, urinary concen trations of workers exposed to phenoxy acids have been reported from many countries as summarized by IARC.9 For small groups of spraymen, mean con centrations of up to 8 mg l-1 have been found as in forestry workers (for 2,4-D in tractor spraying;10). Spraying in farming was found to result in consider ably lower levels, e.g. 0.2-0.3 mg l-1. Professional spraymen may sometimes reach three to six times higher concentrations, however.11 Not surprisingly, aerial spraying is more favour able than tractor and other ground-spraying, with an approximately ten-fold reduction in exposure as re flected in the urinary herbicide concentrations. Still, Table 1. Some pesticides and pesticide exposures* considered by IARC, as grouped with regard to carcino genicity to humans Degree of evidence Compound or exposure Humans Animals Groupt Amitrol|| Chlordecone (Kepone) Chlorophenols't Chlorophenoxy herbicides*|| Creosotes^ DDT Dichlorobromopropane Hexachlorobenzenet Mirex Toxapherre I ND L L L I ND ND ND I S 2B S 2B -- 2B -- 2B S 2A S 2B s 2B s 2B s 2B s 2B tCarcinogenic to humans (group 1), probably (group 2 A) or possibly (group 2 B) carcinogenic to humans. S = suffi cient, L = limited and I = inadequate evidence of carcino genicity. ND = no data. Information obtained from IARC monographs up to vol. 41, 1987 and supplement 7.8 Fungicide; wood preservative. Insecticide. ||Herbicide. aerial spraying tends to be less well accepted by the public, mainly for ecological reasons and especially with regard to spraying of forest land. The contamination of clothes by pesticides has' been common in the past, especially over the front of the thighs and legs. This is noteworthy, since the skin absorption seems to be particularly good from the thighs. A habit of combining suntanning with the spraying of pesticides of low acute toxicity seems to have been common in some countries, e.g. in Sweden, resulting in a wetting of the naked skin by phenoxy acids or other pesticides and probably leading to efficient absorption through the skin. In clinical contacts with patients with alleged symptoms from pesticide exposure, one may also be told that clogged nozzles of the spraying equipment were sometimes cleaned by blowing with the mouth. This procedure apparently means another important route of exposure in some instances. Absorption of pesticides through the skin, and even by oral con tact, is therefore sometimes likely to play a greater role than exposure by inhalation. As a consequence, the dose obtained by an individual may to a great ex tent depend on behavioural factors and vary consid erably, even among individuals performing the same type of spraying. Only limited protection should therefore be expected through the standards set for air concentrations of pesticides in many countries. OBSERVATIONS ON CANCER AND PESTICIDE EXPOSURE The first reports more clearly associating a cancer risk with exposure to pesticides appeared in the 1970s. For example, Swedish railroad workers with a mixed exposure to amitrol, diurone, phenoxy acids, potassium chlorate, monurone and other agents were found to have suffered from significantly more cancers than expected, especially those who had a combined exposure to amitrol and phenoxy acids.12-13 An excess of lung cancer was also seen in pesticide applicators in the DDR, again exposed to a mixture of preparations, both insecticides and herbicides.14-1S A case series of soft-tissue sarcoma patients was reported by Hardell in 1977, suggesting exposure to phenoxy acids as playing an etiological role.16 This clinical observation prompted epidemi ologic studies in Sweden of soft-tissue sarcomas, carrying over also to a study on lymphomas.17-18-19 and followed by investigations in other countries, as discussed below. There has also been a number of cancer studies among farmers and other agricultural groups since the late 1970s, but with less specific information with regard to pesticide exposure. A review of most of these studies by Blair et al.1K) suggested that in spite of a generally low cancer morbidity among farmers, some cancer forms tend to quite consistently appear in excess, namely of the lip. stomach, skin (nonmelanotic forms), prostate, brain and connective tis sue, as well as Hodgkin's disease, multiple myeloma and leukemia. A decreased risk seems to be present for colorectal, bladder and lung cancer. Rather re cently, additional studies have essentially confirmed this cancer pattern,21" '23 and some interest has also been directed to testicular cancer in some other stud ies, three out of four being positive in this respect regarding agricultural workers.24-25-26-27 Etiologicalfy. solar radiation seems to explain the excess of lip and skin cancer, whereas physical activ ity may account for the low morbidity from colorec tal cancer.28'29 and the well-known lower smoking rate in rural populations may determine the de creased rate of bladder and lung cancer. For the lat ter, one may speculate that exposure to endotoxins in dust from hay and grain''" may also play a preven tive role.31 Stomach cancer may be thought of as relating to nitrates in food and water as a precursor of N nitrosoamines. Interestingly, a large number of pesticides may react with nitrite to form /V-nitroso compounds, e.g. atrazin, carbaryl, ferbam, glyphosate, MCPA (as an amine salt), simazine, thiram, ziram and several others.32 However, in view of the latency required for cancer to develop, factors likely to have caused stomach cancer in rural areas must have appeared at an earlier point in time than the various pesticides used in agriculture. It is also of some interest in this context, that a rather recent re port somewhat surprisingly shows an inverse associ ation between nitrates and nitrites in saliva and stomach cancer rates.33 Some of the other increased cancer rates among farmers like those of the prostate and brain have no good explanations, although Blair et a/.20 in their review noted some relation of prostate cancer to poultry breeding and horticulture. Some studies seem to suggest pesticide uSe in agriculture as possibly playing a role in the etiology of brain tumours.34'35,36 A number of studies have considered leukemias and agricultural activities. As summarized by Blair et a i.20 there is no consistent etiological explanation although poultry production seemed associated in some studies but not in others. Furthermore, in studies from the U.S. various leukemia forms have shown associations with farming, as involving insec ticide use, but also dairy production, as well as the growing of crops. A tendency to a higher risk was Pesticides and cancer risks 209 seen for those born after 1900, a phenomenon suggesting factors in modern agriculture to be re sponsible. A simiar age-pattern has been found in some of the studies of multiple myeloma, and again. * an increased risk appeared as possibly associated with pesticide use.20 A recent series of Swedish studies on acute myeloid leukemia, chronic lymphatic leukemia and multiple myeloma may help shed some further light on etiological factors for these disorders, however, especially as the exposure information is relatively detailed.37'38'39 Hence, with regard to acute myeloid leukemia, there appeared a risk ratio of 2.3 for ex posure to poultry but other determinants were also identified (see Table 2). For chronic lymphatic leukemia, a risk ratio of 6.0 was obtained for DDT exposure. Horse contact also appeared as a risk factor, and so did exposure to engine exhausts, all of these exposures being present in agriculture. DDT has also appeared as a risk factor, although rather weak, in the aforementioned case-control studies on soft-tissue sarcoma and lymphoma in Sweden, as well as among a small group of forestry foreman,40 and also in the U.S.41 Some indications of an increased risk of lung cancer in relation to DDT exposure have also been obtained in a few studies with mixed pesticide exposures, which leads to a confounding of the situation, however.15'42'43 Alveolar cell carcinoma has been reported in indi viduals with granulomatous disease of the lungs after inhalation of DDT powder.44 For multiple myeloma,38 engine exhausts appeared as a determinant, in addition to creosote and contact with fresh wood, the latter exposure also being related to chronic lymphatic leukemia but only with borderline significance.39 It is noteworthy that farming as an occupation did not remain as a risk factor, when these other determinants as occurring in farming were accounted for, i.e. engine exhausts, fresh wood and creosote. A puzzling observation that multiple myeloma seemed to correlate signifi cantly negatively with background radiation, i.e. mainly tantamount to living in wooden houses, whether causally or spuriously, remains unsolved (whereas the chronic lymphatic and acute myeloid leukemias had no association and a positive relation, respectively). It also remains to be seen, however, whether other studies with as detailed exposure information will confirm or refute these various associations. Some reports suggest arsenical pesticides as causing skin cancer in vineyard workers,45'46 and an excess of lung cancer has been seen in pesticide pro ducers with arsenic exposure47 and users, although with mixed exposure.15 Also, cases of liver angi- 13132 210 O la v Axelson Table 2. Some significantly increased risk ratios, RR, by Mantel-Haenszel analyses* or logistic regressiont regarding farming activities, acute myeloid leukemia (AML), chronic lymphatic leukemia (CLL) and multiple myeloma Disease Risk factor RR 95% Cl RR, other factorst AML CLL MM Poultry 2.3t 1.0-5.1 DDT Horse 6.0* 1.5-23 2.5* 1.2-5.1 Creosote 4.7' 1.2-18 Fresh wood 2.6* 1.1-5.7 Electrical work: 3.8; psychotherapeutic drugs: 3.6; styrene: 18.9. three cases Engine exhausts: 2.2 (Fresh wood: 2.4; solvents 1.7) Engine exhausts: 2.1 tR R s in parentheses with 95% Cl including unity; for AML a dose-response pattern was also seen with regard to background radiation from building materials. MM with reversed association in this respect. sarcoma have been reported as associated with arsenical pesticides.48 49 The overall importance of arsenical pesticides in cancer causation in agriculture seems quite limited, however, since the use of such pesticides is restricted. SOFT-TISSUE SARCOMAS, LYMPHOMAS AND PHENOXY HERBICIDES As mentioned, studies with non-specific exposure information have indicated an excess risk of malig nancies of the connective tissue, Hodgkin's disease and non-Hodgkin's lymphoma among farmers.20 In this regard, the studies from Sweden1718 may be taken as a confirmation of the risk and as an elucida tion the specific factors involved as showing a risk ratio of about 5-6. for soft-tissue sarcoma and phenoxy acid exposure alone, and a risk of 3-6 for chlorophenol exposure alone; taking phenoxy acids and chlorophenols either alone or together, the risk ratios were about 5. In one of the studies,18 an attempt was made to distinguish the effect from phenoxy acids without tetrachloro-dibenzo-dioxin (TCDD) contamination, but effects were seen re gardless of the possibility of this impurity's presence. However, the risk appeared to be higher when TCDD might have been involved. Similarly, as for soft-tissue sarcomas, malignant lymphomas of the Hodgkin and non-Hodgkin type taken together were also found to be associated with exposure to phenoxy acids and chlorophenols, with risk ratios of about 5 and 4, respectively. In addition, there appeared to be an about three-fold risk as re lated to solvent exposure, especially involving tetraand trichloroethylene, benzene and styrene.10 These Swedish studies have prompted a number of cohort and case-control studies to confirm or re fute the possible associations of these malignancies and exposure to phenoxy acids and chlorophenols. Some results of the more important of these various studies are given in Tables 3-5, but a further com ment may be justified in certain respects for some of the studies, especially regarding validity and power. Existing reviews may also be consulted for addi tional information and other aspects as also in volving TCDD exposure.66'6768 First, regarding the cohort studies, only the largest of these 52-3;>would have some power to detect a risk of soft-tissue sarcoma as also appearing in one of them (Lynge, 1985).r'2 The situation with regard to power is somewhat better for lymphoma although not found in any clear excess. The observation of nasal cancer in one of them36 is of some interest, since nasal cancer has been associated especially with chlorophenol exposure, e.g. in sawmill work, wood dust being controlled.69 An increased risk was also seen for phenoxy acid exposure, although not reaching significance. It may also be remarked that the latency periods elapsed so far since exposure in many of the studies, e.g. with regard to exposure in Vietnam, is still quite short. In the Swedish stud ies referred to, the average latency periods were 15-20 yr for the two disorders studied.70 The case-control studies listed in Table 4 may all be characterized as overall non-positive (except for the study on ovarial mesothelioma), but showing Pesticides a n d cancer risks 211 Table 3. Results in some cohort studies on phenoxy acid exposure, soft-tissue sarcoma (STS) and other cancer Author and year Character of population Agents involved Type of cancer(s) Obs/expected or risk ratio; remarks Hogstedt and Westerlund (1980)40 Barthel (1981)15 244 forestry workers 1658 males in agriculture Riihimki et al. 1926 herbicide (1983)50 sprayers Lynge (1984, 1985)51,52 4563 chemical workers Ott et al. (1980)53 204 chemical workers Lawrence et al. 1496 Vietnam (1985)54 veterans 2,4-D; 2,4,5-T 2,4-D; MCPA and other pesticides 2,4-D; 2,4,5-T (2,4,5-T) 2,4-D; MCPA and other 2,4,5-T 2,4,5-T 2,4-D Coggon et al. (1986)55 5784 users MCPA and producers All cancer Lung cancer 5/1.4 (foremen) 3/8.4 (workers) 50/27.5 All cancer All ca 26/36.5 (no type in excess) 156/160.6 STS Lung cancer All cancer 5/1.8 11/5.3 1/1.3 STS All ca Nasal ca RR = 1.09 (95% Cl, 0.18-6.7 vs non-Vietnam veterans) 297/314.0 3/0.61 Table 4. A ranking of risk ratios, RR. obtained in case-control studies of soft-tissue sarcoma and ovarial mesothelioma in relation to specific information of phenoxy herbicide exposure Author RR Remarks Donna et al. (1984)36 Woods et al. (1987)37 Vineis et al. (1986)38 Olsson (1987)39 Smith et al. (1984)"' 4.4 Ovarial mesothelioma 2.8 Scandinavians onlv 2.7 Women with early exposure 2.4 Finland and Sweden 1.6 Probably or definitely exposed Hoar et al. (1986)61 0.9 Use of 2,4-D on crops Kang et al. (1986)62 0.8 Vietnam veterans Greenvald et al. (1984)63 0.7 Vietnam veterans Smith and Pearce (1986)64 0.7 RR = 1.1 by including earlier data from New Zealand increased risk ratios for certain subgroups, e.g. for women with early and relevant phenoxy acid exposure.38 A quite remarkable observation is re ported by Woods e ta l.,51 who in a study in Washing ton State found an increased risk among a subset of cases and controls with Scandinavian names. This finding together with the observations from Sweden17,18 and Denmark51,52 could indicate the operation of some genetic or cultural co-factor. It may be mentioned, however, that a record-linkage study using census data on occupation and cancer registry information in Sweden71 failed to confirm the findings from the case-control studies. On the other hand, a British study based on data from can cer registration has shown a slightly but significantly increased risk of soft-tissue sarcomas among 212 O la v Axelson Table 5. A ranking of risk ratios, RR. obtained in case-control studies of Hodgkin's (H) and non-Hodgkin's (n-H) lymphoma in relation to specific information on phenoxy herbicide exposure Author H/n-H RR Remarks Woods et al. (1987)57 n-H 4.8 Forestry herbicide applicators: 1.3 for farmers. No excess in Scandinavians Hoar et al. (1986)m H 0.9 n-H 1.6 Exposure for more than 20 davs/yr with RR = 6.0 for n-H Pearce et al. n-H 1.3 Cancer controls; RR = 1.0 with ( 1986)65 . - population controls farmers, farm managers and market gardeners but not in other subgroups in forestry and farming.72 Based on occupational titles, the elevated risks seen in the Swedish case-control studies of soft-tissue sarcoma and lymphoma were reduced to 1.4 or less as indicated in Table 6.70 Regarding lymphomas, the aforementioned study from Washington State57 indicate an increased risk for individuals in farming and forestry with exposure to phenoxy herbicides. A study from Kansas61 strongly supports the findings from Sweden regard ing non-Hodgkin's lymphoma with a clear and strong dose-response relationship, but, quite re markably, no risk was seen either for Hodgkin's dis ease or soft-tissue sarcoma. The overall exposure situation may have been different to that in Sweden, however, with a resulting differential effect on the cancer pattern. Again, the observations with regard to ethnicity attract interest in view of the report by Woods et al.3l The studies from New Zealand have provided no clear support for any effect of phenoxy acid exposure with regard to soft-tissue sarcoma or non-Hodgkin's lymphoma, even if some slightly in creased risk ratios appeared in the analyses of cer tain subgroups. Instead, an indication was obtained that meat work could be a risk factor both for softtissue sarcoma and non-Hodgkin's lymphoma.60-63 Fencing appeared as another risk factor for nonHodgkin's lymphoma. CONTROVERSIES AND DISCUSSIONS For regulatory purposes, evaluations have been made from time to time regarding the possible can- Table 6. Exposure to phenoxy acids (Ph) and chlorophenols (Ch) of soft-tissue sarcoma (Eriksson et a l.[>>) and lymphoma (Hardell et a l.19) cases in comparison to referents Type of cases; referents Fanning and forestry work Ph Ch None Other occupations Ph Ch None Soft-tissue sarcomas Referents Risk ratios 13 1 53 6.4 -- 17 1 10 68 39 0 5 167 1.1 -- 4.9 (1.0) Lymphomas Referents Risk ratios 35 5 37 23 1 114 4.1 -- 0.9 6 15 71 1 7 189 -- 5.7 (1.0) The risk ratios with regard to occupation (farming and forestry work) can be derived from the table and are 1.4 for soft-tissue sarcomas and 1.2 for lymphomas. cer effects of pesticides, and the possible carci nogenic effects of the phenoxy acids have become especially controversial. Most of the evaluations of this kind have not been published in the open litera ture and can hardly be considered as scientific in character, but rather as complying to economic in terests. The resulting, quite problematic, situation has hardly been improved by the fact that the use of phenoxy acids in the Vietnam war has also led to alleged disorders in the Vietnamese population67 and to compensation claims from veterans suffering from cancer and other illnesses. The issue became even more intricate as a result of falsifications of figures and facts from the published studies by a Royal Commission on the Use and Effects of Chemical Agents on Australian Personnel in Vietnam.73-74'76 Hence, instead of making a serious and independent evaluation, this Australian com mission simply copied several hundred pages of a submission, conclusions included, as compiled by lawyers representing a chemical company, namely Monsanto Australia.'6 However, in contrast to the Australian approach, there are also some remarks on a scientific basis to the Swedish studies on soft-tissue sarcoma and lymphoma associated with phenoxy acids and chlorophenols. A couple of these comments might be considered here,77-78 particularly with regard to the question of a possible misclassification of expo sure as subject to concern, i.e. with a suspicion of either memory bias among the subjects regarding their exposure or a tendency by the investigators to ascribe exposure to the cases rather than to the con trols. Now, in this critique, little attention seems to have been paid to the fact that measures were under taken to avoid such a bias, i.e. the application of a special epidemiologic technique in two of the stud ies, that might have revealed a bias in this respect.79-80 The principle can be seen in Table 6, where exposed and non-exposed subjects within the domain of farming and forestry work are compared to a reference outside this domain. Had there been an observation bias with regard to exposure, this would have meant an overstreaming of unexposed cases to the exposed category and/or of exposed con trols or referents to the unexposed subjects, result ing in a decreased risk for non-exposed individuals within farming and forestry in comparison to the outside reference group. No such phenomenon appears, however, but the possibility remains that there is for some reason an increased risk among farmers and forestry workers, and that this in creased risk in combination with an observation bias happens to be split up just the way it is, i.e. so that the non-exposed farmers and forestry work Pesticides and cancer risks 213 ers get just the normal risk. It would be remarkable, however, with such a precise splitting up of a gen erally operating risk, not only in one but in two consecutive studies on this particular issue and for two different disorders. It is also difficult to explain the basis for any operation of a general risk as detached from herbicide exposure in these occupa tions. Furthermore, for chlorophenol exposure, there was a good agreement between the answers obtained in the questionnaires and the information given by the employers. Unfortunately, however, no employer data on exposure were available for phenoxy acids. In view of the points just referred to, it is remark able to see how the debate78 on the possible cancer risks from phenoxy acids tends to accumulate vari ous perceptions of bias, stemming from unidenti fied sources, and as witnessing that the original re ports can hardly have been read with sufficient care. It should not be forgotten either, that if a significant recall bias had existed, studies on colon cancer and liver cancer undertaken in close connection with the soft-tissue sarcoma and lymphoma studies would also have reported an effect due to phenoxy acid and chlorophenol exposure.81-82The existence of specific information on the validity has also been acknow ledged,9 and further views on the scientific debate around the phenoxy acids can be found elsewhere.66 In view of the additional studies that have recently appeared on the scene, and as briefly reviewed above, the overall picture as to the effects of phen oxy acids remains somewhat confused. The results of the study by Hoar et al.bi have now focused the interest on non-Hodgkin's lymphoma as a possible effect of phenoxy acids, but on the other hand the well-conducted studies from New Zealand do not provide much support for any effects at all. However, non-positive studies do usually not attract much criticism, and there is perhaps a reason in this context to make some remarks relating also to the validity of the studies from New Zealand. First, it should be recalled that random misclassi fication (as for that matter systematical misclassifica tion) of exposure might cause a bias towards a risk ratio of unity, and especially so if another risk factor is also operating. Therefore, it would have been de sirable for exposure to both phenoxy acids and chlorophenols to have been more clearly displayed in the data. There is also an unexplained and re markably increased risk for railway workers of 3.2 with regard to soft-tissue sarcoma,60 at the same time as the percentage of railway workers among the controls is as high as 7.6% (7/92). A similarly high figure for railway workers is found in the study of non-Hodgkin's lymphoma,65 i.e. 12 and 11% for ' 19136 214 O la v Axelson cancer and population controls, respectively. Since the composition of a study population is supposed to be reflected by the controls in a case-control study,83 the surprisingly high proportion of railroad workers raises the question whether there could have been some sort of selection bias with regard to the con trols in these studies. However, there is no obvious reason why a bias in this respect should occur. EPILOGUE In an attempt to summarize the possible cancer effects from pesticides, it might be suggested that arsenic compounds are rather likely to be carcino genic, as in good agreement with other observations on arsenic and cancer, although the studies on arsenic pesticides alone would be less convincing. DDT has also become more clearly, if not convinc ingly, associated with human cancer and there is sufficient evidence for carcinogenicity in animals. Therefore, even if the direct evidences for a cancer risk are rather limited in humans, the overall infor mation seems rather convincing. The quantitative role of DDT for those cancer forms appearing in ex cess among farmers is not at all clear, however. A definite evaluation of the carcinogenic effect of phenoxy acids seems to remain rather remote, since the epidemiologic studies appear as inconsistent but still without any clearly identifiable flaws. The con clusion at present might therefore be that these com pounds are unlikely to be complete carcinogens and might even require particular circumstances to exert activity. Animal data do not provide much help for judgements on carcinogenicity as they are missing for MCPA and inadequate for 2,4,5-T and 2,4-D, even if some animal studies do exist that are indica tive of a cancer risk.9 A study from New Zealand, showing a significant relation between smallintestinal adenocarcinoma in slaughtered sheep and exposure to phenoxy acid or picolinic acid or both, through sprayed feedstuff, is quite remarkable.8 No particular guidance regarding the carcinogenic effects of phenoxy acids is obtained from studies of genotoxic effects such as chromosome aberrations, sister chromatid exchanges and micronuclei, since again, these studies are equivocal both in humans and in experimental systems.8-85 It could be of some interest in this context, however, that both 2,4,5-T and 2,4-D are able to inhibit intercellular com munication.8 Furthermore, 2,4-D and MCPA can in duce peroxisome proliferation in the liver of Chinese hamsters and decrease serum lipid levels,86 i.e. sim ilar to the drug clofibrate. Hepatic peroxisome proliferators are known to be carcinogenic, presumably through producing hydrogen peroxide and reactive oxygen radicals.87,88 Judging from the existing know ledge based on experimental research as well as epi demiologic observations, and thereby taking tl^e in consistencies noted into consideration, it seems as if further investigations into phenoxy acids might very well help also the general understanding of carcino genic mechanisms, namely if the somewhat puzzling pattern seen at present would turn out as a model for various interfering phenomena and co-factors in the process of carcinogenesis. With regard to epidemiology, it seems rather clear, however, that cancer risks are associated at least with the use of mixtures of pesticides in agricul ture, whereas the effects of isolated compounds are more difficult to assess. It may not necessarily be further studies on farmers, but rather on special groups of other users and producers, that will help assess more definitely the health hazards from pesti cides, since the farmers may have a too complex ex posure situation. It should also be a concern for the future, that even more potent herbicides seem to be under introduction at present, i.e. compounds of which only a few grams are needed per hectare of land, obviously making the hygienic control and the epidemiologic risk evaluations even more difficult than today. REFERENCES 1. Guralnick L: Mortality by occupation and cause of death. Vital Statistics Special Report 53(3). Washing ton DC, Department of Health, Education and Wel fare (1963). 2. Registrar General: The Registrar General's decennial supplement. England and Wales. Occupational mor tality tables. London, HMSO (1971). 3. Fox A J, Goldblatt P O: Longitudinal study of sociodemographic mortalitv differentials, 1971-1975. London, HMSO (1980). 4. McMichael A J, Hartshorne J M: Mortality risks in Australian men by occupational groups, 1968-1978. Med J Aust 1, 253 (1982). 5. Burmeister L F, Morgan D P: Mortality in Iowa farmers and farm laborers, 1971-1978. J occup Med 24. 898 (1982). 6. Wiklund K: Swedish agricultural workers: a group with a decreased risk of cancer. Cancer 51, 566 (1983). 7. Gallagher R P, Threlfall W J, Jeffries E: Cancer and aplastic anemia in British Columbia farmers. J natn Cancer Inst 72, 1311 (1984). 8. International Agency for Research on Cancer: Mono graphs on the evaluation of the carcinogenic risk of chemicals to humans, Supplement 7. Lyon, IARC (in press). 9. International Agency for Research on Cancer: Mono- >inif ry ong or he er at ulire be ial ;lp >ti- iX - :he be of of the ;ult e of ingVel- mial nor- / of 975. <s in .978. [owa Med roup 566 r and natn lonosj- ->f iono- graphs on the Evaluation o f the Carcinogenic Risk o f Chemicals to Humans, Vol. 4L. Lyon. IARC (1986). LO. Kolmodin-Hedman B, Erne K: Estimation o f occupa tional exposure to phenoxy acids (2,4-D and 2,4,5-T). Archs Toxic. Suppl. 4, 318 (1980). 11. Kolmodin-Hedman B, Hglund S. kerblom M: Studies on phenoxy acid herbicides. I. Field study. Occupational exposure to phenoxy acid herbicides (MCPA, dichlorprop, mecoprop and 2,4-D) in agri culture. Archs Toxic 54, 257 (1983). 12. Axelson O. Sundell L: Herbicide exposure, mortality and tumor incidence: an epidemiological investigation on Swedish railroad workers. Work envir Hlth 11,21 (1974). 13. Axelson O, Sundell L, Andersson K, Edling C, Hogstedt C, Kling H: Herbicide exposure and tumor mor tality. An updated epidemiologic investigation on Swedish railroad workers. ScandJ Work envir Hlth 6. 73 (1980). 14. Barthel E: Gehuftes Vorkommen von Bronchial krebs bei beruflicher Pestizidexpostition in der Lantwirtschaft. Z Erkrank Atmungsorg 146. 266 (1976). 15. Barthel E: Increased risk of lung cancer in pesticideexposed male agricultural workers. J. Toxic envir Hlth 8, 1027 (1981). 16. Hardell L: Malignant mesenchymal tumours and ex posure to phenoxy acids. A clinical observation (in Swedish). Lkartidningen 74, 2753 (1977). 17. Hardell L, Sandstrm A: Case-control study: softtissue sarcomas and exposure to phenoxyacetic acids or chlorophenols. Br J Cancer 39, 711 (1979). 18. Eriksson M. Hardell L, Berg N O, Mller T, Axelson O: Soft tissue sarcomas and exposure to chemical subtances: a case-referent study. Br J ind Med 38, 27 (1981). 19. Hardell L, Eriksson M, Lenner P, Lundgren E: Malignant lymphoma and exposure to chemicals, especially organic solvents, chlorophenols and phe noxy acids: a case-control study. B rJ Cancer 43, 169 (1981). 20. Blair A , Malker H, Cantor K P, Burmeister L, Wiklund K: Cancer among farmers. A review. Scand J Work envir Hlth 1 1 , 397 (1985). 21. Schumacher M C: Farming occupations and mortality from non-Hodgkin's lymphoma in Utah. A case-con trol study. J occup Med 27, 580 (1985). 22. Pearce N E, Sheppard R A , Howard J K, Fraser J, Lilley B M: Leukemia among New Zealand agricul tural workers. A cancer registry-based study. Am J Epidemiol 12 4 , 402 (1986). 23. Saftlas A F, Blair A , Cantor K P, Hanrahan L, Anderson H A: Cancer and other causes of death among Wisconsin farmers. A m J ind Med 11, 119 (1987). 24. Mills P K, Newell G R, Johnson D E: Testicular can cer associated with employment in agriculture and oil and natural gas extraction. Lancet 1, 207 (1984). 25. McDowell M, Balarajan R: Testicular cancer and employment in agriculture. Lancet 1, 510 (1984). 26. Sewell C M, Castle S P, Hull H F: Testicular cancer and employment in agriculture and oil and natural gas extraction. Lancet 1, 553 (1986). Pesticides and cancer risks 215 27. Wiklund K, Dich J, Holm L-E: Testicular cancer among agricultural workers and licensed pesticide applicators in Sweden. Scand J Work envir Hlth 12, 630 (1986). 28. Garabrant D H, Peters J M. Mack T M, Bernstein L: ' Job activity and colon cancer risks. Am J Epidemiol 119, 1005 (1984). 29. Gerhardsson M, Norell S E, Kiviranta H, Pedersen N L, Ahlbom A: Sedentary jobs and colon cancer. Am J Epidemiol 123, 775 (1986). 30. Rylander R: Organic dusts and lung reactions. Ex posure characteristics and mechanisms for disease. Scand J Work envir Hlth 11, 199 (1985). 31. Enterline P E, Sykftra J L. Keleti G, Lange J H: Endotoxins, cotton dust and cancer. Lancet 2. 934 (1985). 32. International Agency for Research on Cancer: Mono graphs on the evaluation of the carcinogenic risk of chemicals to humans, Vol 30. Lyon, IRC (1983). 33. Forman D, Al-Dabbagh S, Doll R: Nitrates, nitrites and gastric cancer in Great Britain. Nature 313. 620 (1985). 34. Gold E, Gordis L, Tonascia J, Szklo M: Risk factors for brain tumors in children. Am J Epidemiol 109. 309 (1979). 35. Unger M, Olsen J: Organochlorine compounds in the adipose tissue of deceased patients with and without cancer. Envir Res 23, 257 (1980). 36. Musicco M, Filippini G, Bordo B M. Melotto A, Morello G, Benino F: Gliomas and agricultural ex posure to carcinogens: a case-control study. Am J Epidemiol 116, 782 (1982). 37. Flodin U , Fredriksson M, Axelson O, Persson B, Hardell L: Background radiation, electrical work and some other exposures associated with acute myeloid leukemia in a case-referent study. Archs envir Hlth 41, 77 (1986). 38. Flodin U: Etiological studies on certain blood malignancies by means of case-control technique (in Swedish). Stockholm, Arbetsmiljofondens sammanfattningar No. 980 (1986). 39. Flodin U, Fredriksson M, Persson B, Axelson O: Chronic lymphatic leukemia and engine exhausts, freshwood and DDT. A case-control study. Br J ind Med (in press). 40. Hogstedt C, Westerlund B: Cohort studies on causes of death of forestry workers with and without ex posure to phenoxy acid preparations (in Swedish). Lakartidningen 77, 1828 (1980). 41. Cantor K, Everett G, Blair A, Gibson R. Schuman L, Isacson P: Fanning and non-Hodgkin's lymphoma. A m J Epidemiol 122, 535 (1984). 42. Wang H H, McMahon B: Mortality among pesticide applicators. J occup Med 21, 741 (1979). 43. Blair A, Grauman A J, Lubin J H, Fraumeni J F Jr.Lung cancer and other causes of death among li cenced pesticide applicators. J natn Cancer Inst 71, 31 (1983). 44. Pimentel J C, Menezes A P: Granulomatous disease with alveolar cell carcinoma due to inhalation of powdered DDT. Pathology 165, 140 (1979). 45. Lchtrath H: The consequences of chronic arsenic 1 ij i >j 8 216 O la v Axelson poisoning among Moselle wine growers. J Cancer Res clin Oncol 105. 173 (1983). 46. Thiers H. Colomb G, Moulin G, Colin L: Le cancer cutan arsenical des viticulteurs du Beaujolais. Ann Derm 94. 133 (1967). 47. Mabuchi K. Lilienfeid A M, Snell L M: Cancer and occupational exposure to arsenic: a study of pesticide workers. Prev Med 9, 51 (1980). 48. Brady J, Liberatore F, Harper P, Greenwald P, Bur nett W, Davies J N P, Bishop M, Polan A, Vianna N: Angiosarcoma of the liver: an epidemiologic survey. J natn Cancer Inst 59, 1383 (1977). 49. El Zayadi A. Khalil A, El Samny N, Hamza M R, Selim O: Hepatic angiosarcoma among Egyptian farmers exposed to pesticides. Hepatogastroent 33, 148 (1986). 50. Riihimaki V, Asp S, Pukkala E, Hemberg S: Mortal ity of 2.4-dichlorophenoxyacetic acid and 2,4,5trichlorophenoxyacetic acid herbicide applicators in Finland. Chemosphere 12, 779 (1983). 51. Lynge E: Weed killers and cancer. Cohort Study o f Cancer Outcome among Workers in Production o f Phenoxy Acids in a Chemical Factory (in Danish). Copenhagen. Arbejdsmiljofondet (1984). 52. Lynge E: A follow-up study of cancer incidence among workers in manufacture of phenoxy herbicides in Denmark. Br J Cancer 52, 259 (1985). 53. Ott M G. Holder B B, Olsen R D: A mortality analy sis of employees engaged in the manufacture of 2,4,5trichlorophenoxvacetic acid. J occup Med 22, 47 (1980). 54. Lawrence C E. Reilly A A. Quickenton P, Green wald P, Page W F, Kuntz A J: Mortality patterns of New York State Vietnam veterans. Am J publ Hlth 75, 277 (1985). 55. Coggon D , Pannett B, Winter P D , Acheson E D, Bonsall J: Mortality of workers exposed to 2-methyl4-chlorophenoxyacetic acid. ScandJ Work envir Hlth 12. 448 (1986). 56. Donna A , Betta P-G, Robutti F, Crosignani P, Berrino F, Bellingeri D: Ovarian msothlial tumors and herbicides: a case-control study. Carcinogenesis 5, 941 (1984). 57. Woods J S, Polissar L, Severson R K, Heuser L S, Kulander B G: Soft-tissue sarcoma and nonHodgkin's lymphoma in relation to phenoxy herbicide and chlorinated phenol exposure in western Washing ton. J natn Cancer Inst (in press). 58. Vineis P, Terracini B, Ciccone G, Cignetti A , Col ombo E, Donna A , Maffi L, Pisa R, Ricci P, Zanini E, Comba P: Phenoxy herbicides and soft-tissue sar comas in female rice weeders. A population-based case-referent study. Scand J Work envir Hlth 13, 9 (1986). 59. Olsson H: Summary of epidemiologic findings in the study by the Scandinavian sarcoma group (in Swed ish). Abstract to the 13th meeting of the Scandinavian Sarcoma Group, Copenhagen, 5-6 March, 1987. 60. Smith A H, Pearce N E, Fisher D O, Giles H J, Teague C A, Howard J K: Soft tissue sarcoma and ex posure to phenoxy herbicides and chlorophenols in New Zealand. J natn Cancer Inst 73, 1111 (1984). 61. Hoar S K, Blair A , Holmes F F. Boysen C D. Robel R J, Hoover R, Fraumeni Jr J F: Agriculture herbi cide use and risk of lymphoma and soft-tissue sar coma. J Am med Assoc 256, 1141 (1986). 62. Kang H K, Weatherbee L, Breslin P P, Lee Y. She pard B M: Soft tissue sarcomas and military service in Vietnam: a case comparison group analysis of hospital patients. J occup Med 28, 1215 (1986). 63. Greenwald P, Kovasznay B. Collins D N, Therriault G: Sarcomas of soft tissues after Vietnam service. J natn Cancer Inst 73, 1107 (1984). 64. Smith A H, Pearce N E: Update on soft tissue sarcoma and phenoxy herbicides in New Zealand. Chemosphere (in press). 65. Pearce N E, Smith A H, Howard J K. Sheppard R A, Giles H J, Teague C A: Non-Hodgkin's lymphoma and exposure to phenoxy herbicides, chlorophenols, fencing work and meat works employment. A casecontrol study. Br J ind Med 43, 75 (1986). 66. Axelson O: The health effects of phenoxy acid herbi cides, in Harrington J M (ed): Recent Advances in Occupational Health, p. 253. Edinburgh. Churchill Livingstone (1984). 67. Sterling T D, Arundel A: Review of recent Vietna mese studies on the carcinogenic and teratogenic effects of phenoxv herbicide exposure. IntJ Hlth Serv 16, 265 (1986). ' 68. Sterling T D, Arundel A: Health effects of phenoxy herbicides. A review. ScandJ Work envir Hlth 12.161 (1986). 69. Hardell L, Johansson B, Axelson O: Epidemiological study of nasal and nasopharyngeal cancer and their relation to phenoxy acid or chlorophenol exposure. Am J ind Med 3, 247 (1982). 70. Hardell L. Axelson O: Phenoxyherbicides and other pesticides in the etiology of cancer: some comments on Swedish experiences, in Becker C E, Coye M J (eds): Cancer Prevention. Strategies in the Workplace. Washington, DC, Hemisphere (1986). 71. Wiklund K, Holm L-E: Soft tissue sarcoma risk in Swedish agricultural and forestry workers. J natn Cancer Inst 76. 229 (1986). 72. Balarajan R, Acheson F D: Soft tissue sarcomas in agriculture and forestry workers. J Epidemiol com mun Hlth 38, 113 (1984). 73. Royal Commission on the Use and Effects of Chemi cal Agents on Australian Personnel in Vietnam: Final report, voi. 4: Cancer. Canberra. Australian Govern ment Publishing Service (1985). 74. Axelson O (ed): Rebuttals of the Final report on can cer by the Royal Commission on the Use and Effects of Chemical Agents on Australian Personnel in Viet nam. Report no. LiU-YMED-R-6. Linkping, Uni versity of Linkping (1986). 75. Armstrong B K: Storm in a cup of 2.4,5-T. Med J Aust 144, 284 (1986). 76. Thomas A S, Saracci R, Humphrey G F, Poliak J K, Axelson O, Hardell L (letters): Storm in a cup of 2,4,5-T Med J Aust 144, 611 (1986). 77. Coggon D , Acheson E D: Do phenoxy herbicides cause cancer in man? Lancet 1, 1057 (1982). 78. Colton T: Herbicide exposure and cancer. J Am Med Assoc 256, 1176 (1986). 79. Axelson O: A note on observational bias in casereferent studies in occupational health epidemiology. Scand J Work envir Hlth 6, 80 (1980). . ii. Tockman M S: Epidemiology in the workplace: The problem of misclassification. J occup Med 24, 21 (1982). 81. Hardell L: Relation of soft-tissue sarcoma, malignant lymphoma and colon cancer to phenoxy acids, chlorophenols and other agents. Scand J Work envir Hlth 7, 119 (1981). 82. Hardell L, Bengtsson N O, Johnsson U. Eriksson S, Larsson L G: Aetiological aspects on primary liver cancer with special regard to alcohol, organic solvents and acute intermittent porphyria -- an epidemiologi cal investigation. Br J Cancer 50, 389 (1984). Pesticides and cancer risks 217 83. Axelson O: Elucidation of some epidemiologic prin ciples. Scand J Work envir Hlth 9, 231 (1983). 84. Newell K W, Ross A D, Renner R M: Phenoxy and picolinic acid herbicides and small-intestinal carci noma in sheep. Lancet 2, 1301 (1984). 85. Linnainmaa K: Genotoxicity of phenoxy acid herbi cides 2,4-D and MCPA. Dissertation. Helsinki, University of Helsinki and Institute of Occupational Health (1983). 86. Vainio H. Nickels J, Linnainmaa K: Phenoxy acid herbicides cause peroxisome proliferation in Chinese hamsters: Scand J Work envir Hlth 8, 70 (1982). 87. Reddy J K, Azarnoff D L: Hypolipidaemic hepatic peroxisome proliferators form a novel class of chemi cal carcinogens. `Nature 283, 397 (1980). 88. Knuutila S: Role of free radicals in genetic damage (mutation). Med Biol 62, 110 (1984). 1 ydenhuoltohenkilostbn oikeudelli-j u u s o f health care personnel]. Lha-J' a a 2 (1983): 18, 8 - 9 . \ kinen kuorm itus lahikasvattajien ja? hoitohenkilO kunnan tyossa [Sues-] ions o f professional personnel at in-] :are and institutions for the m ental-; stitute o f O ccupational H ealth, H el-j portti 2). I p iirteiden m ittaam inen kyselymene-j ten t o f w o rk characteristics by ques-* . T ydterveyslaitoksen tutkim uksia 2j leal stress an d disease, a co nceptual; ion EK, R ahe R H , ed. Life stress an d ' T h o m a s , S p rin g fie ld , 1L 1974, ppy im ongst nurses. In: Cooper C L, M ar-; collar an d professional stress. John' [Chester 1980, pp 1 9 - 6 1 . $ m h o ita ja n tyotyytyvaisyys ja tyossa 3 s a tis f a c tio n a n d w o rk stre s s exper-'. U n iv e rsity o f H e lsin k i. H elsin k i 1976.^ e G . V A ldshandlingar pa en psykiat-] ing. E n retrospektiv studie [Handling^ ute psychiatric health care: A retro-j a k a r tid n in g e n 74 (1977) 4 6 0 6 --4 6 0 7 .! nhoidossa tapahtuvien vahinkojen kore n satio n for accidents occurring dur-' . S u o m Idakaril 34(1979) 1155 -- 1158.| mi H . A p u laisla ak arin tyOhonsi-j ,s (A ssistant physicians' w ork re ining], S uom laakaril 36 (1981): 20; so n E , K afry DB. From tedium to p er| he F ree P ress, New Y ork, NY 1981. | y k ia trise n sa iraalan tyO olojen ja ty0n-; n se k d n iid en v u o ro v a ik u tu k se n yhtey-] psyy k k iseen terveyteen: kuvaileva osa' o f the w ork conditions and personnel h o sp ita l a n d th eir relatio n sh ip w ith the h e a lth : A d escrip tio n ]. In stitu te o f Oc^ 1th, H elsin k i 1981. I h e n k ilo k u n n a n k okem uksista tyOstaatr iden parissa [Experiences o f health carq p atien ts w ith ch ro n ic illnesses.]. Helsin^ u rsin g , H elsinki 1981. 1 tress o f life. M cG raw -H ill B ook, New >. ^ srisk er i arbetsm iljO n (Risks o f violence iro n m en t]. A rbetarskyddsstyrelscn, Solna 1, e d . T e rv e y d e n h u o lto a la n tyOsuojeluon-. :ms o f lab o r p ro tectio n in h ealth c arer pksen tu tk im u k sia 2 (1984): suppl. :a tio n : 20 S eptem b er 1986 ORIGINAL ARTICLES .V iW J W o r k E n v ir o n H e a lth 13 (1986) 9 -- 17 Phenoxy herbicides and soft-tissue sarcom as in female rice weeders A population-based case-referent study cy Paolo V ineis, MD,1 B enedetto Terracini, MD,1 Giovannino C icco n e, MD,1 A lessandro C ignetti,2 Eva C olom bo,1 Adalberto Donna, MD,3 Lidio Maffi,1 Roberto Pisa, MD,* Paolo Ricci, MD,5 Ermanno Zanini,3 Pietro Comba, DSc,6 VINEIS P , T E R R A C IN I B, CICC O N E G, C IG N ETTI A, CO LO M BO E, D O N N A A. M A FFI L, PISA R, R IC C I P, Z A N IN I E, C O M B A P. Phenoxy herbicides an d soft-tissue sarcom as in fem ale rice w eed ers: A p o p u la tio n -b a s e d c a s e -re fe re n t stu d y . S c a n d J W o rk E n v iro n H e a lth 13 (1987) 9 -- 17. A p o p u la tio n based case-referent study w as cond u cted in an area o f n o rth ern Italy w here rice grow ing is the p red o m i nant agricultural activity an d phenoxy herbicides have been used since 1950. M anual rice w eeding was form erly perform ed by a seasonal fem ale w orking p opulation; in the early 1950s these w om en were co n currently exposed to chem ical herbicides. Sixty-eight persons representing incident an d histologically re vised cases (31 w o m en ) a n d 158 p o p u la tio n refe ren ts (73 w om en) w ere in terv ie w ed . T h e cases w ere h isto logically confirm ed independently by two blinded pathologists, and exposure to phenoxy herbicides was assessed by tw o b linded pesticide researchers. A n age-ad ju sted odds ratio o f 0.91 w as fo u n d for the living m en (w ith suspect exposures; no m an diagnosed as a case had been exposed with certainty to phenoxy h e rb ic id e s). A m o n g th e living w o m e n th e rela tiv e risk w as 2.7 (90 'to c o n fid e n c e in te rv a l 0 .5 9 -- 12.37), an d it fu rth e r increased w hen a tte n tio n w as restricted to w om en exposed in the w hole 1950-- 1955 p erio d and to younger age groups. K ey term s: case-control study, occupational exposure, rice grow ing. Chlorinated phenoxy acids, namely, 2,4,5-trichloro,'henoxy acetic acid (2,4,5-T), 2,4-dichlorophenoxy .cctic acid (2,4-D) and 2-methylchlorophenoxyacetic acid (MCPA), have been manufactured since the early 1940s and extensively used as herbicides in agriculture and forestry since the 1950s. A mixture of 2,4,5-T and 2,4-D (Agent Orange) was used as a defoliant during the Vietnam war. An association between occupational exposure to phenoxy herbicides, both in their manufacture and use, and the subsequent occurrence of soft-tissue sarcomas has been suspected in some reports (2, 8, 9, 14, 18), 'ui not confirmed by others (12, 27). A few studies 1 Unit o f C ancer Epidem iology, U niversity and M ain H os pital, T o rin o , Italy. Institute o f A gronom ical C hem istry, University o f T orino, T orino, Italy. D epartm ent o f Pathology, M ain H ospital, A lessandria, Ita ly. D epartm ent o f Pathology, Ospedale Cervello, Palerm o, Ita- niM itute o f P ath o lo g y , U niversity o f V erona, V erona, Ita ly. Istituto Superiore de S anit, R om a, Italy. Reprint requests to: D r P V ineis, Servizio di E pidem iologia dei T u m o ri, Istitu to di A n a to m ia P ato lo g ica, Via S an ten a ' 1-10126 T o rin o , Italy. have not been sufficiently informative (10, 21,24,26). (See also the reviews in references 1 and 6). The Inter national Agency for Research on Cancer, chiefly on the basis of reports by Eriksson et al (8) and Hardell & SandstrOm (14), evaluated the evidence on the car cinogenicity of phenoxy acid herbicides to humans as limited (15). Most of the aforementioned authors emphasized the need for further knowledge. Therefore we decided to undertake an investigation in the provinces of Vercelli, Novara, and Alessandria (northern Italy), where rice growing is the predominant agricultural activity. The total population in this area is approximately 1.5 mil lion. Phenoxy herbicides were introduced in large amounts in the early 1950s and have been extensively used up to now. According to official statistics (16, 17), more than 4.8 t of 2,4-D and 21.71 of MCPA were used in the three provinces in 1959 (data not available for 2,4,5-T). The corresponding figures for 1969 were 24.5 t for 2,4-D, 29.4 t for MCPA and 125.1 t for 2,4,5-T. 2,4,5-T was officially banned in 1970. Before the introduction o f chemical herbicides, a female work ing population was seasonally engaged in rice weeding. For a period around 1950-- 1955 rice weeding was still done manually although chemical weeding was being experimented with. It is therefore reasonable to assume that the rice weeders working during this period were exposed to phenoxy herbicides, mainly through skin contact. 19141. 9 Subjects and methods Case identification and ascertainment New cases of soft-tissue sarcoma with a proved or sus pected histological diagnosis in 1981 -1983 (age of per son > 20 years) were identified through all of the pa thology departments o f the three provinces (hospitals of Vercelli, Novara, Alessandria, Borgomanero, Biella, Casale Monferrato). The search was extended to six pathology departments in the city of Torino and to the National Cancer Institute in Milan, where residents in the three provinces could be referred for treatment. Overall, 135 potential cases were identified (116 in hos pitals located in the three provinces and 19 in Torino or Milan). Visceral sarcomas were excluded. Histological specimens (with the exception o f the cases diagnosed in Torino and Milan and four other cases) were independently reviewed by two patholo gists (AD, RP), blindly with respect to the exposure status and to previous diagnosis. The classification of the World Health Organization (WHO) was adopted for the soft-tissue sarcomas (7). Of the 135 identified cases, 18 were excluded because the person's town of residence was unknown or the person was not traced after five attempts (table 1). A f ter the histological review, 26 cases were discarded. For five others the two pathologists agreed on the diagno sis of soft-tissue sarcoma but did not agree on the his tological type. After a check o f the clinical records, four cases turned out to be visceral sarcomas and were excluded. For nine histologically confirmed cases, we met with refusal to participate, and for two in which the subject was deceased no relative could be traced. Of the remaining 75 cases, 68 were incident (ie, newly diagnosed in 1981 --1983), and seven turned out to be prevalent (ie, diagnosed prior to 1981) during a check of the clinical records. The seven prevalent cases were T a b le 1. D is trib u tio n o f c a s e s an d re fe re n ts by c a te g o rie s o f in c lu s io n /e x c lu s io n . Cases R e feren ts O rig in a lly en ro lle d E x clu d ed o r lost R e s id e n c e u n k n o w n U n tra c e d in d iv id u a l R e fu s a l* R e lative n o t tra c e d 0 C h an g e of resid en ce0 H is to lo g ic d ia g n o s is not c o n firm e d V isceral sarco m a0 P rev alen t c a s e 0 H is to lo g ic a lly co n firm e d and in te rv ie w e d n o n v is c e ra l cas e s In te rv ie w e d re fe re n ts 135 10 8 9 2 1 26 4 7 68 * 208 20 26 - 4 * * * 158 In c lu d in g in d iv id u a ls no t a v a ila b le fo r in te rv ie w d u e to ill ness, d e a fn e s s o r m en tal d iso rd ers. R e la tiv e s o f d e c e a s e d in d iv id u a ls w e re no t ava ilab le. M o v e d to o th e r p ro v in c e s ; in s tric t s e n s e n o t e lig ib le . T h e fo u rv is c e ra l an d seven p re v a le n t s a rc o m a s w e re id e n ti fie d as su c h a fte r a c h e c k o f th e c lin ic a l re c o rd s a fte r th e c a s e s h ad b ee n e n ro lle d . (S e e th e tex t.) 10 not included in the analyses. Of the 68 subjects remain ing, 44 were living and 24 were deceased. The age-standardized annual incidence rate (includ ing interviewed and noninterviewed subjects and ex cluding prevalent and visceral cases) was 2.8/100 000 [95 % confidence limits (CL) 1.3--4.2] for the men and 2.0 (95 % CL 0 .8 -3 .1 ) for the women (age > 20 years). These rates do not take into account the his tological revision in order to be comparable to the rates reported in 1976-1977 by the Cancer Registry of the Province o f Varese (an area close to the study area). The latter rates (per 100 000, age a 20 years) were 2.3 (95 % CL 0 .4 -4 .1 ) for the men and 1.5 (95 *7o CL 0.06--2.9) for the women (23). (All the rates have been age-standardized on the basis of the world population.) Living referents A random sample o f eligible referents (30 living men and 30 living women aged > 20 years from each prov ince) with a distribution, by municipality, representa tive of the population of each province was drawn from electoral rosters. The overall number of referents was approximately twice the expected number of eli gible cases. In fact, a total of 168 individuals from 65 municipalities were identified versus 180 expected. (Four electoral offices did not respond at all, and a few offices provided a lower than required number of individuals, for a total of 12 losses.) Deceased referents Thirty-seven of the 135 persons originally enrolled as cases were deceased at the time of the interview. There fore, for each o f these 37 persons, two deceased indi viduals, to be matched to the deceased subject by age ( 4 years), sex, and year of death, were identified as referents from the same municipality as the deceased subject through the demographic offices o f the muni cipalities. There was no selection of the cause of death, with the exception of the exclusion of one suicide (for ethical reasons). Seventeen deceased subjects with soft-tissue sarco ma had two matched referents. Only one referent met the matching criteria for six cases. No individual with the corresponding requirements was found for five cases. The municipalities did not respond for eight referents. And, finally, no referent was sought for five persons with soft-tissue sarcoma because they were en rolled in the National Cancer Institute in Milan after the scheduled end o f the study. A total of 40 deceased referents were therefore known for interview purposes, and in fact the relatives of 36 were interviewed. Interviews Each enrolled subject was given a code number unre lated to the nature of their case-referent status. All sub sequent procedures until the final stage of the analy sis were blind. Each living subject was mailed an invi- 13142 t t;...;on to be inter rise purpose of tl vitation was folio to schedule an in letters inviting th< for the interview side. For each ini by telephone or b made before the i F?r deceased pe: ' the same pr \ simplified p jects who, for ai but agreed to pr information on i in this way. The an exposure asst the direct interv All the intervi terviewer (LM). graphic infor ., (iii) a lifejob titles, addre of products or cr apeutic treatme changes, includi in rice-growing; eluded a list of sure to phenoxy jobs in sawmil highway and rt. trket gardenii .: for those w: rice growing an information wa entailing the h; tion of herbicidt specific jobs in' weeding and wa edge of exposu to phenoxy acid lart of the que? ith pesticide : Demographi pation andedu collected from lities of resider Assessment oj Two experts wi riculture (EZ, phenoxy ac ;ews. This ass two and in a bl used for readi: category 1 = could not be n posed. Attem; f + nalyses. Of the 68 subjects remain-' and 24 were deceased. * ized annual incidence rate (includ-: i noninterviewed subjects and ex-' id visceral cases) was 2.8/100 000 imits (CL) 1.3--4.2] for the men] 0.8 --3.1) for the women (age > ! ;s do not take into account the his- ! order to be comparable to the rates! .977 by the Cancer Registry o f the' (an area close to the study area).;. 100 000, age > 20 years) were 2.3 ' .) for the men and 1.5 (95 7o CL` omen (23). (All the rates have been j . the basis of the world population.) }f eligible referents (30 living men n aged > 20 years from each prov-j rtion, by municipality, representa- i tion o f each province was drawn! ts. The overall number of referents^ twice the expected number of eli-i , a total of 168 individuals from 65 ; 'entified versus 180 expected.) ..-s did not respond at all, and a; :d a lower than required number of I total o f 12 losses.) { i : 135 persons originally enrolled as 1at the time of the interview. Therelese 37 persons, two deceased indi t e d to the deceased subject by age and year of death, were identified re same municipality as the deceased e demographic offices of the muni- is no selection of the cause o f death, ; o f the exclusion of one suicide (for ised subjects with soft-tissue sarco-5 led referents. Only one referent met j ria for six cases. No individual with | ; requirements was found for fivej ipalities did not respond for eighttally, no referent was sought for five j :issue sarcoma because they were en- ] jnal Cancer Institute in Milan after I o f the study. A total of 40 deceased : efore known for interview purposes, datives of 36 were interviewed. ijcct was given a code number unreo f their case-referent status. All subes until the final stage o f the analych living subject was mailed an invi- ation to be interviewed, with no mention of the pre cise purpose of the study. Whenever possible, the in vitation was followed by telephone calls in an attempt [o schedule an interview date. Otherwise, additional letters inviting the individual to suggest a possible date for the interview were sent with a prepaid envelope in side. For each individual, at least five attempts (either by telephone or by a home visit by the interviewer) were made before the individual was considered untraceable. For deceased persons the next-of-kin was identified, od the same procedure was followed. A simplified postal questionnaire was sent to sub jects who, for any reason, refused a direct interview but agreed to provide information by mail. Overall, information on 16 cases and 37 referents was obtained in this way. The postal questionnaire was suitable for an exposure assessment identical to that provided by the direct interviews. Ail the interviews were carried out by a trained in terviewer (LM). The questionnaires collected (i) de mographic information, (ii) a life-long smoking his,;ry, (iji) a life-long occupational history (including job titles, addresses of the plants or farms, and types of products or crops), (iv) a history of radiological ther apeutic treatment, and (v) a history o f residence changes, including a specific question on periods spent in rice-growing areas. In addition the questionnaire in cluded a list of job titles potentially involving expo sure to phenoxy herbicides or chlorophenols (including jobs in sawmills, the paper and leather industries, highway and railway maintenance, and flower and market gardening). Two additional sheets were filled ut for those who had worked in agriculture, one for rice growing and one for other crops. In these sheets information was collected on the Involvement in jobs entailing the handling, transportation, and distribu tion of herbicides; herbicide spraying procedures; other specific jobs involving (or not) herbicide exposure (ie, weeding and water regulation in rice growing); knowl edge of exposure to herbicides (and more specifically to phenoxy acids or chlorophenols) or insecticides. This part of the questionnaire was prepared in cooperation vith pesticide researchers. Demographic information (including current occu pation and educational level) for nonrespondents was collected from demographic offices of the municipa lities of residence. Assessment o f exposure Two experts with experience in chemical aspects of ag riculture (EZ, AC) were asked to assess the exposure > phenoxy acids for each job recorded in the inter'Cws. This assessment was done independently by the two and in a blind manner. The following criteria were used for reading and interpreting the questionnaires: category 1 = not exposed, category 2 = exposure could not be ruled out, and category 3 = certainly ex posed. Attempts to distinguish between skin contact and other paths of exposure and between low and high levels of exposure proved not to be feasible. Rice weeders were considered to be exposed to phenoxy herbi cides when they worked after 1950 and did not work exclusively in a small rice allotment o f their own. Cat egory 2 was used particularly for people engaged af ter 1950 in com, wheat and pasture growing. As a first step, a separate exposure assessment was performed by each o f the two experts. Subsequently, a joint as sessment was made, in which they reached a common evaluation for each recorded job. The latter assessment was used in the analyses. Herbicide spraying in the area has been chiefly done by tractor. Coding o f occupational histories In addition to the aforementioned exposure assess ment, all occupational histories were coded according to the classification for job titles of the International Labour Organization (4) and the United Nations clas sification for economic activities of plants or farms (22). All coding work was done by a single person (GC) in a blind manner. Analyses The analyses were performed with an HP-61 calcula tor using programs for epidemiology (25). Ageadjusted relative risks were computed according to Mantel-Haenszel's estimator (19); 90 % confidence in tervals were obtained with Miettinen's approximate procedure (20) (the lower limit of each interval cor responding to a 95 % one -tailed confidence limit). In addition to age (in four groups), therapeutic X rays and smoking were included as potential confounders. The X-ray exposure represents a known risk factor for soft-tissue sarcoma (3), but there is no evidence of association with herbicide exposure. Smoking is not known to be a risk factor for this kind of tumor, but it was conservatively included as a potential confounder. Results Response and characteristics o f the subjects Table 2 reports the educational levels represented by the persons with soft-tissue sarcoma broken down into respondents and nonrespondents. The presented edu cational levels are those o f the persons scheduled for interview (ie, the next-of-kin of deceased subjects). Responding men showed an unbalanced distribution for the cases and referents in the different educational categories ( < 6 years: 26 cases and 37 referents, > 6 years: 10 cases and 43 referents; unknown: 1 case and 5 referents); the distribution of the women was more balanced ( < 6 years: 17 cases and 42 referents; > 6 years: 12 cases and 30 referents; unknown: 2 cases and 1 referent). 11 f T a b le 2. E d u c a tio n a l le v e ls (years o f s c h o o lin g ) fo r th e re s p o n d in g an d n o n re s p o n d in g livin g s u b je c ts o r th e n e x t-o f-k in of d e c e a s e d s u b je c ts . N o n re s p o n d in g in d iv id u a ls In c lu d e re fu s a ls , u n tra c e d in d iv id u a ls , in d iv id u a ls w h o s e re la tiv e s w e re n o t Iden tifie d , and in d ivid u als w ho had ch a n g ed re s id e n c e . Y ears of sch o o lin g <6 >6 U nknow n Livin g s u b je c ts N on respo ndents R espond ents C ases R eferen ts C ases R eferen ts 10 18 1 11 5 13 28 56 14 65 21 N e x t-o f-k in o f d e c e a s e d s u b je c ts N o nrespon dents R espond ents C ases R eferen ts C ases R eferen ts 4 3 1 4 15 23 88 15 T a b le 3. D is tr ib u tio n o f th e liv in g s u b je c ts b y a g e g r o u p a n d e x p o s u r e le v e l (1 = n o t e x p o s e d , 2 = e x p o s u r e c o u ld n o t be ru led o u t. 3 = e x p o s e d ). E x p o s u re re fe rs to a n y c a le n d a r p e rio d . T h e n u m b e r o f s u b je c ts fo r w h o m In fo rm a tio n c a m e from p o s ta l q u e s tio n n a ire s is p re s e n te d in p a re n th e s e s . E x p o su re level C ateg o ry 1 Cases R eferen ts C ateg o ry 2 Cases R eferen ts C ateg o ry 3 Cases R eferen ts M en <55 W om en 7 (3) 36 (8) 1 (D 4 2 5 (2) 3 6 (10) 1 _ 2 A g e gro u p (years) 55 - 6 4 6 5 -7 4 M en W om en M en W om en 63 (1) 12 (2) 1 2 _ - 1 6 (2) 1 1 2 1 5 _ __ 6 (3) 9 (5) _ _ 3 (D 1 M en >75 W om en 7 (1) 1 (1) _ _ - 4 (2) 2 (1) 1 (1) _ - T a b le 4. D is trib u tio n o f th e livin g fe m a le s u b je c ts by ag e g r o u p a n d e x p o s u r e le v e l (1 = n o t e x p o s e d , 2 = e x p o s u r e c o u ld no t be ruled o u t. 3 = exp o sed ). O n ly w o m e n e x p o s e d c o n tin u o u s ly d u rin g 1 9 5 0 -1 9 5 5 have b een in c lu d e d . E x p o s u re level C ateg o ry 1 Cases R eferen ts C ateg o ry 2 Cases R eferen ts C ateg o ry 3 Cases R eferen ts <55 5 36 -- - -- - A ge group (years) 55 --64 6 5 -7 4 16 69 ---- -- 12 1- >75 4 2 -- - -- Among the men 26 (30.5 Vo) of 85 referents and 15 (40.5 Vo) of 37 subjects with soft-tissue sarcoma had worked in agriculture during at least some period of their worklife. The women with such occupations were 21 (28.5 Vo) of 73 referents and 14 (45.5 Vo) of 31 women with soft-tissue sarcoma. Two male referents and no men with soft-tissue sarcoma were assigned a job title code o f rice grower. The respective numbers for the women were three and three (crude odds ratio 2.5). This code was the only job title code mentioning rice growing, and it was assigned when rice growing was the only activity performed in a given work period. Living subjects Table 3 reports the absolute numbers o f living subjects with soft-tissue sarcoma and their referents by expo sure levels and age groups for the two sexes. No men with soft-tissue sarcoma were definitely exposed, and exposure could not be ruled out for only two, the ageadjusted odds ratio being 0.91 (90 Vo CL 0.21 --3.91). Among the alive women, four subjects with softtissue sarcoma and five of their referents were certainly exposed, and one and two, respectively, belonged to exposure category 2. This distribution corresponds to an odds ratio o f 2.42 (90 Vo CL 0.56 - 10.34) when ex posure categories 2 and 3 are combined and an odds ratio of 2.70 (90 Vo CL 0.59--12.37) for category 3 alone. The odds ratios were age-adjusted in the four age groups. Table 4 shows the distribution o f the living female subjects who were engaged in relevant jobs in the whole 1950--1955 period. Three with soft-tissue sarcoma and one referent met these criteria, all from exposure cat egory 3. Unexposed were 16 women with soft-tissue sarcoma and 53 female referents, as shown in table 3. Deceased subjects Table 5 reports the age and exposure distribution for the deceased subjects. There were 24 next-of-kin of per sons with soft-tissue sarcoma interviewed and 36 nextof-kin o f referents. Three women and no men with soft-tissue sarcoma belonged to exposure category 3, and one woman and one man with soft-tissue sarco ma were classified into category 2. The odds ratio (ageadjusted in two age groups) for the women in expo sure categories 2 and 3 was 1.05 (90 Vo CL 0.21 --5.1). When matched pairs or triplets were considered, there T .e 5. Di: c.. e from p Exposure st. Unexposed Cases Referents Exposed3 Cases sferents Categorie were only t with soft-ti: definitely e and the setreferents (b eaory 3. T1 :le subje 'hen th cuts were c important ^ male refere versus 3 ot of 13 dece: same char; This comp, of deceaset the inclusic unfound. Therapeut as potentic one womai no female radiation l therefore. As for s: of 28 worn f 83 male .:us were s ents inforn findings in women wi; be interpretion of agr soft-tissue Histology "idle 6 re; gical typ<_ ' gives the the expose soft-tissue there were mas, two i { iv in g s u b je c ts o r th e n ex t-o f-kin of iu a ls w h o se relatives w ere no t den-' ofo f -k in d e c e a s e d s u b je c ts id e n ts R espond ents fe re n ts C ases R eferen ts 3 15 23 ; 1 8 8; 4 1 5: :p o s e d . 2 = e x p o s u re c o u ld n o t be s fo r w h o m in fo rm a tio n c a m e from ale 5. D is trib u tio n o f d e c e a s e d s u b je c ts by age gro u p and e x p o s u re s ta tu s . T h e n u m b e r o f s u b je c ts fo r w h o m in fo rm a tio n :j u r e fr o m p o s t a l q u e s t i o n n a i r e s is p r e s e n t e d in p a r e n t h e s e s . Exposure statu s M en <55 W om en A g e group (years) 5 5 -6 4 6 5 -7 4 M en W om en M en W om en M en 2 75 W om en Unexposed Cases R eferen ts Exposed3 Cases -efe re n ts 5 5 (4) -- 1 1 -- -- 51 62 ---- 2 2 (1) 5 (2) -- 1 1 1 1 1 1 3 (1) 3 4 3 6 (1) C a te g o rie s 2 (e x p o s u re c o u ld n o t be ru le d o u t) and 3 (e x p o s ed ) h ave b e e n c o m b in e d u n d e r th is h ea d in g . -7 4 W om en M en 275 W om en 6 (3) 9 (5) 7 (1) 1 (1) 4 (2) ' 2 (1); - - 1 (1) 3 (1) _ -- il*<- oups for the two sexes. No men ma were definitely exposed, and: : ruled out for only two, the age-j eing0.91 (90% CL 0 .2 1 -3 .9 1 ).; vomen, four subjects with soft--: 'e o f their referents were certainlyd two, respectively, belonged to: This distribution corresponds to (90 % CL 0 .56-10.34) when ex-| md 3 are combined and an odds| CL 0.59--12.37) for category 3: os were age-adjusted in the four| distribution of the living female] raged in relevant jobs in the whole] 'hree with soft-tissue sarcoma and], ;e criteria, all from exposure cat-! were 16 women with soft-tissue} tie referents, as shown in table 3. ] age and exposure distribution fori ;. There were 24 next-of-kin of per-j sarcoma interviewed and 36 next-] Three women and no men with; y ' 'nged to exposure category 3,! : man with soft-tissue sarco-j to category 2. The odds ratio (age-j : groups) for the women in expo-j i 3 was 1.05 (90 % CL 0.21 - 5.1).j s or triplets were considered, there? were only two: One was a triplet in which the subject with soft-tissue sarcoma and the two referents had been definitely exposed to phenoxy herbicides (category 3), and the second was a triplet in which one of the two referents (but not the sarcoma subject) belonged to cat egory 3. Therefore a matched analysis of deceased fe: :ie subjects was uninformative. '-Vhen the towns of residence of the deceased refer ents were compared with those of the living referents, important differences appeared. Six of the 23 deceased male referents lived in four towns of rice-growing areas versus 3 of 62 living referents. Among the women, 6 of 13 deceased referents lived in two towns with the same characteristics, versus 2 of 60 living referents. This comparison strongly suggests that overmatching of deceased subjects was introduced into the study by -he inclusion of residence as a matching factor. Confounding Therapeutic X rays and smoking were considered as potential confounders. Overall, only two men and one woman with soft-tissue sarcoma, and one male and no female referents had been exposed to therapeutic radiation before 1981. Confounding by this variable, therefore, can be ruled out. As for smoking, 24 (65 %) of 37 men and 3 (11 %) ot 28 women with soft-tissue sarcoma and 54 (65 %) 'i 83 male referents and 21 (40 %) o f 52 female refer ents were smokers or exsmokers. (For two male refer ents information on smoking was not available.) These findings indicate a clear deficit of smokers among the women with soft-tissue sarcoma. Such a pattern can be interpreted as a consequence of the greater propor tion of agricultural workers among the persons with ^oft-tissue sarcoma. rlktology and site o f the sarcomas 'le 6 reports the distribution of the cases by histo- 'cicai type according to the WHO classification. Table ='ves the histological type and site of the tumors for the exposed cases. Among the exposed women with 'toft-tissue sarcoma (categories 2 and 3 of exposure), there were two chondrosarcomas, two Kaposi sarco mas, two malignant fibrous histiocytomas, one chor- T a b le 6. D is trib u tio n o f th e in te rv ie w e d in c id e n t h is to lo g ic a l ly re v is e d c a s e s by h is to lo g ic a l ty p e , a c c o rd in g to th e c la s s i fic a tio n o f th e W o rld H e a lth O rg a n iza tio n . H is to lo g ic a l type D e rm a to fib ro s a rc o m a p ro tu b eran s L ip o sarco m a M y x o id lip o s a rc o m a L eio m yo sa rc o m a R habdom yosarcom a A n g io s a rc o m a M a lig n a n t h e m a n g io p e ric y to m a O th e r v a s c u la r m a lig n a n t tu m o rs S y n o vial s a rc o m a M a lig n a n t s c h w a n n o m a O th e r m a lig n a n t tu m o rs o f th e perip h eral nerves M a lig n a n t tu m o rs of n o n c h ro m a ffin tis s u e C hordom a C h o n d ro s a rc o m a of soft parts M a lig n a n t g ian t cell tu m o r of soft p a rts M a lig n a n t fib ro x a n th o m a (m a lig n a n t h is tio c y to m a ) Kaposi sarcom a U n c e rta in T o ta l3 M en 1 2 3 4 1 1 t 1 3 2 _ 2 1 3 7 1 33 W om en l 3 2 1 t 1 1 _ 1 1 3 - 5 6 4 30 3 T h e o rig in a l d ia g n o s is o f th e five in te rv ie w e d in c id e n t c a s e s w h ic h w e re not review ed w e re o n e d e rm a to flb ro s a rc o m a pro tu b e ra n s . o n e a n g io s a rc o m a , on e m a lig n a n t h is tio c y to m a , and o n e K aposi s a rc o m a a m o n g th e m en and o n e m a lig n a n t s c h w a n n o m a am ong th e w o m e n . Th e se cas es w ere in clu d ed In th e a n a ly s e s b e c a u s e o f th e c o n fid e n c e of th e h ig h level o f th e h is to lo g ic a l d ia g n o s is . O n e o f the m en b e lo n g e d to ex p o s u re c a te g o ry 2. w h ile th e o th ers w ere u n e x p o s e d . doma, and one leiomyosarcoma. (One case was un certain with respect to histological type.) No gross dif ferences were noticed between the exposed and unex posed women with respect to the sites of the tumors. Their locations were as follows: head and neck (0 exposed/3 unexposed), retroperitoneal (1 exposed/2 unexposed), upper limbs (1 exposed/4 unexposed), lower limbs (3 exposed/7 unexposed), thorax (1 exposed/3 unexposed), abdomen (1 exposed/1 unex posed), pelvis (2 exposed/1 unexposed), site not spe cified (0 exposed/1 unexposed). j -d A V " ' 13 T a b le 7. S e x , a g e , e x p o s u re c a te g o ry , jo b title , y e a rs o f e x p o s u re , an d h is to lo g y and s ite o f tu m o r (If a c a s e ) fo r th e e x p o s e d s u b je c ts w ith s o ft-tis s u e s a rc o m a (cases) and th e e x p o s e d re fe re n ts . * C a se/re feren t s ta tu s L ivin g c a s e L ivin g c a s e Deceased re fe re n t Livin g cas e Deceased re fe re n t Deceased re fe re n t Deceased re fe re n t Deceased re fe re n t Deceased re fe re n t L ivin g c a s e Deceased re fe re n t Deceased re fe re n t Deceased re fe re n t Deceased re fe re n t Deceased re fe re n t Livin g referen t L ivin g referen t Deceased case Livin g referen t L ivin g re fe re n t Livin g referen t L ivin g re fe re n t Deceased case L ivin g re fe re n t Deceased case L ivin g c a s e Deceased case L ivin g c a s e Livin g re fe re n t Deceased case Livin g c a s e L ivin g re fe re n t Livin g re fe re n t Livin g re fe re n t Livin g re fe re n t Sex F em ale F em ale F em ale M a le ' M a le F em ale F em ale M a le M ale F e m a le M a le F e m a le F em ale F em ale M a le M a le F e m a le F em ale M a le F em ale Fem ale M a le M a le F em ale Fem ale A ge Exposure (years) categ o ry * J o b title 84 2 R ice w e e d e r 71 3 R ic e w e e d e r 77 3 R ice w e e d e r 55 2 . R ice w e e d e r 62 2 Farm lab o rer in ric e g ro w in g 82 3 R ice w e e d e r 84 3 R ice w e e d e r 86 2 W h eat, corn and pastu re g ro w er 43 3 W a te r reg u lato r in ric e fie ld s 74 3 R ice w e e d e r 79 2 W h eat and corn grow er 77 3 R ice w e e d e r 79 3 R ice w e e d e r 81 2 C o rn a n d w h e a t grow er 62 2 W h e a t, pastu re and corn grow er 54 2 F lo w er g ard en er 56 2 C o rn and pastu re grow er 79 2 W h eat and corn grow er 47 3 R ice g ro w e r 52 3 R ice w e e d e r 64 3 R ic e w e e d e r 44 3 R ice w e e d e r 88 2 W h eat and corn grow er 69 3 R ice w e e d e r 77 3 R ice w e e d e r F em ale 73 3 R ice w e e d e r F e m a le 84 3 R ice w e e d e r M a le 64 2 W h eat and corn grow er F em ale 62 3 R ice w e e d e r F em ale 75 3 R ice w e e d e r F em ale 60 3 R ice w e e d e r F e m a le 47 2 M arket g ard e n er F e m a le 50 3 W h e a t, corn and pastu re g ro w er M a le 65 2 W h eat, corn and p as tu re g ro w er M a le 47 2 W h eat, corn and p astu re grow er Years of e x p o s u re Unknow n 1 9 5 0 -1 9 6 3 1 9 5 0 -1 9 5 8 1 9 5 0 -1 9 8 4 1 9 5 0 -1 9 5 2 1 9 5 0 -1 9 6 0 195 8 -1 9 7 1 1 9 5 0 -1 9 7 4 1 9 5 4 -1 9 6 6 1 9 5 0 -1 9 6 7 195 0 -1 9 7 1 1 9 5 0 -1 9 6 0 1 9 5 0 -1 9 7 0 1 9 5 0 -1 9 6 7 1 9 5 0 -1 9 7 0 1 9 5 0 -1 9 8 4 1 9 5 2 -1 9 8 4 1 9 5 0 -1 9 8 0 1 9 6 0 -1 9 8 4 1 9 6 3 -1 9 6 8 196 0 -1 9 7 1 1 9 5 5 -1 9 6 8 1 9 5 0 -1 9 8 0 1 9 5 5 -1 9 7 0 1 9 5 0 -1 9 6 6 1 9 6 1 -1 9 6 3 1 9 5 0 -1 9 6 0 1 9 6 0 -1 9 7 0 1 9 5 0 -1 9 7 8 1 9 5 0 -1 9 7 0 1 9 5 0 -1 9 6 3 1 9 7 4 -1 9 8 4 1 9 7 0 -1 9 8 4 1 9 7 5 -1 9 8 3 1 9 7 8 -1 9 8 3 H is to lo g y and site o f tu m o r U n c e rta in h is to lo g y lo w e r lim b K aposi, u p p e r lim b D e rm a to fib ro sarco m a p rotube rans, s ite unknow n C hondrosarcom a, lo w e r lim b C h o rd o m a , pelvis L eio m yo sa rc o m a, u p p e r lim b M a lig n a n t fib ro u s h is tio c y to m a , p e lv is C hondrosarcom a. th o ra x K a p o si, lo w e r lim b L ip o sarco m a , head-n eck M a lig n a n t h is tio c y to m a , re tro p e rito n e a l L eio m yo sa rc o m a, abdom en (c o n tin u e d ) 14 19146 te of tumor (if a case) for the exposed ' ars of p o s a re kn o w n 5 0 -1 9 6 3 1 5 0 -1 9 5 8 1 5 0 -1 9 8 4 1 5 0 -1 9 5 2 H isto lo g y and site o f tum or U n certa in h is to lo g y , lo w e r lim b K aposi, u p p er lim b D e rm a to fib ro sarcom a protube rans, site unknow n i tole 7. Continued. C ase/re feren t s ta tu s Livin g re fe re n t Living re fe re n t L iv in g re fe re n t D eceased referent Sex M a le M a le M a le M a le A ge Exposure (years) categ o ry* Job title 30 2 W h eat, corn and pasture g row er 47 2 C o rn and p asture grow er 63 2 W h eat and corn grow er* 77 3 Farm laborer in ric e g ro w in g : = e x p o s u re c o u ld no t be ruled o ut; 3 = e xp o sed . Aiso e n g a g e d in ra ilw a y m a in te n a n c e (1 9 5 3 - 1 9 7 8 ). Y ears of e x p o s u re 1 9 6 7 -1 9 8 4 1 9 5 3 -1 9 5 7 1 9 7 0 -1 9 8 2 1 9 5 9 -1 9 8 4 1 9 5 0 -1 9 6 8 H isto lo g y and site o f tum or 1 5 0 -1 9 6 0 3 5 8 -1 9 7 1 3 5 0 -1 9 7 4 3 5 4 -1 9 6 6 9 5 0 -1 9 6 7 9 5 0 -1 9 7 1 9 5 0 -1 9 6 0 9 5 0 -1 9 7 0 -1 9 6 7 ;9 5 0 - 1 9 7 0 1 9 5 0 -1 9 8 4 1 9 5 2 -1 9 8 4 1 9 5 0 -1 9 8 0 1 9 6 0 -1 9 8 4 1 9 6 3 -1 9 6 8 1 9 6 0 -1 9 7 1 1 9 5 5 -1 9 6 8 1 9 5 0 -1 9 8 0 1 9 5 5 -1 9 7 0 1 9 5 0 -1 9 6 6 1 9 6 1 -1 9 6 3 1 9 5 0 -1 9 6 0 1 9 6 0 -1 9 7 0 1 9 5 0 -1 9 7 8 1 9 5 0 -1 9 7 0 1 9 5 0 -1 9 6 3 1 9 7 4 -1 9 8 4 1 9 7 0 -1 9 8 4 '" 7 5 -1 9 8 3 ..,'8 -1 9 8 3 C hondrosarcom a, lo w e r lim b C h o rd o m a , pelvis Leiom yosarcom a, upp er lim b M a lig n a n t fib ro u s h is tio c y to m a . p e lv is C hondrosarcom a, th o ra x K aposi, lo w e r lim b Lip o sarco m a , head-n eck M a lig n a n t h is tio c y to m a , retro p e rito n e a l L eio m yo sa rc o m a, abdom en (c o n tin u e d ) l Table 7 reports the ages at the time o f the interview, vital status, job titles, years o f start and cessation of exposure, and the level of exposure (as assessed by ex perts) for the exposed subjects, in addition to the his tological types and tumor sites of those with soft-tissue sarcoma. exposure o f persons with prevalent cases o f soft-tissue sarcoma and those with visceral sarcomas Of the five women interviewed who were considered to have prevalent cases of soft-tissue sarcoma, four were unexposed and one was definitely exposed, and they belonged to the 55- to 64-year age group. Two persons with visceral sarcoma were interviewed; they were both unexposed. Discussion Testing the hypothesis of an association between ex posure to phenoxy acids in agricultural practice and soft-tissue sarcomas required a study to be carried out (i) on a population basis and (ii) in an area where phe noxy acids have been widely used. The present study meets both these requirements. The most important exposures took place in the early ''-Os, due to the concurrent presence of manual weedng and chemical spraying. Indeed, an increased risk ot soft-tissue sarcoma was shown among the female rice weeders who had been employed in the early 1950s. Exposure among the men was negligible. Although the present study is limited by a low power, on qualitative grounds our findings are con sistent with both those obtained by Hardeil & Sandstrom (14) and Eriksson et al (8) in similar populationbased studies and with those of reports on small se- s of cases among workers exposed to chlorophenols u/or their derivatives in the chemical industry (9, iS). Studies failing to suggest an association either (i) se lected referents from cancer patient files (2) or adopt ed an approach involving proportional mortality ra tios, which implied some loss in sensitivity (10, 21), or (ii) used a crude categorization of exposures, in terms of occupation at the time of diagnosis (2) or at the time of death (21). A misclassification of exposure may lead to an underestimation of the relative risk, if not a sys tematic bias (28). Intensity and duration of exposure should also be kept in mind when different studies are compared, al though no clear picture of the levels of exposure in dif ferent jobs and different countries is available. The statistical power of the present study is low since we had planned a sample size suitable for increases in risk of the same order of magnitude of those reported by Swedish authors (8, 14). This outcome is reflected by the large confidence intervals around the relative risks. The one-tailed lower confidence limit was, in most instances, lower than one. Indeed, the target pop ulation (female rice weeders affected by soft-tissue sar comas) was inherently limited in size. Major biases were prevented. The similarity o f the incidence rates of soft-tissue sarcoma in the study area and in the near province of Varese suggests that case ascertainment was likely to be complete. The analysis was restricted to histolog ically revised incident cases. Population referents were selected from electoral rosters. Respondence was rea sonable. (Nonrespondents were 9 o f 80 for the con firmed incident cases and 26 of 208 referents.) Selec tion bias is therefore a remote explanation o f the find ings. As table 8 suggests, almost all of the exposed liv ing females were in fact in the lowest educational lev el (relative risk in the category < 6 = 3.86, 90 ?o CL 0.87--17.15). The interviewer was thoroughly trained, all questionnaires were coded biindly with respect to caseness, and exposures were blindly assessed by ex pert pesticide researchers. However, the relative risks may have been under estimated because of the following three types of prob lems; 1. A median latency time of 15 years has been es timated for soft-tissue sarcomas after exposure to chemical agents (13); Fingerhut et al (9) reported fig ures in the range 9 - 2 9 years. Thus a proportion of the cases induced in rice weeders by exposures in the 15 1914*3 .} .... ofT a b le 8 . D i s t r ib u t i o n living fe m a le s u b je c ts by a g e gro u p , e x p o s u re c a te g o ry , an d e d u c a tio n a l level.1 Len gth of sch o o lin g <55 Unex posed Exposed0 A g e g ro u p (years) 5 5 -6 4 6 5 -7 4 Unex posed Exposed0 Unex posed Exposed0 < 6 years0 Cases R eferen ts > 6 years Cases R eferen ts 1 3 T h e re are tw o v a lu e s m is sin g . 0 C a te g o rie s 2 (e x p o s u re c o u ld n o t b e ru le d o u t) and 3 (e x p o s e d ) are c o m b in e d u n d e r th is h ea d in g . 0 A g e -a d ju s te d o d d s ratio: 3 .86 (90 % c o n fid e n c e lim its 0 .8 7 - 1 7 .1 5 ) . >75 U n e x Exposel p o s ed _____ | It'. early 1950s may have been diagnosed before 1981, when our study started. 2. Elderly women ( 7 5 years) were less likely to be engaged in highly-exposing jobs in the 1950s, since they were already 45 years of age or older in that period. Their inclusion in the study may have intro duced an underestimation o f the risks. 3. The procedure of matching deceased referents to the deceased persons with soft-tissue sarcoma with respect to town o f residence may have artifactually raised the proportion of exposed referents. The pro portion of definitely exposed women was similar among the deceased (3 of 10) and living (4 of 21) wo men with soft-tissue sarcoma but quite different among the deceased (5 of 13) and living (5 o f 60) referents. The choice o f a group of living referents, extracted from the general population, for comparison with the deceased subjects inducted as cases is considered to be the best option in case-referent studies; the only flaw is the quality of the interview, obtained from the nextof-kin of deceased subjects (5, 11). If we compare all of the female subjects with soft-tissue sarcoma (living + deceased) with the living referents only, the ageadjusted odds ratio is 2.71 (90 % CL 0.51 --14.17). We intend, in any case, to select another reference group of deceased individuals, unmatched to the subjects with soft-tissue sarcoma in respect to town o f residence. In conclusion, our study showed an odds ratio of 2.7 for living women who were exposed to phenoxy herbicides in any period of their life. When we re stricted our attention to living women less than 75 years of age and exposed in the 1950--1955 period, an ageadjusted odds ratio o f 15.5 (90 Vo CL 1.3 --180.3) was determined (table 4). When exposure to phenoxy her bicides among living females who had a regular job in agriculture (ie, excluding unexposed nonagricultural workers) was analyzed, the age-adjusted odds ratio (Mantel-Haenszel) for exposure category 3 was 3.0. The number of definitely exposed subjects with softtissue sarcoma and referents was negligible among the men. There is no specialized job such as pesticide ap- 16 plicator in rice growing in Piedmont, and annual hoj 13. bicide spraying is performed, among numerous otl jobs, by unspecialized rice growers or farm Iaboren No dose-response relationship could be considen 14. because we were not able to distinguish between hn| and low levels of exposure in the exposure assessmeL (with the exception of the a priori hypothesis concqbj ing the 1950-1955 period). Most of the exposed : jects (both those with soft-tissue sarcoma and theJJ ferents) had been employed in rice weeding for mg than 10 years (table 7). 16. (17. Acknowledgments This research was supported by Regione Piemonte ( cerca biomedica finalizzata) and by CSI-PiemonteJ addition G Ciccone received a fellowship from th e| sociazione Italiana per la Ricerca sul Cancro. We thank also Professor B Giau (Istituto di nomia Agraria, Universit di Torino) for his skill advice and the heads of the pathology department: the hospitals of Biella, Novara, Vercelli, Borgom ro, Casale Monferrato and of the National Cancen stitute in Milan for access to their files and the : mission of histological slides. References 1. A x e lso n O . T h e h e a lth e ffe c ts o f p h e n o x y a c id hi cides. In: H arrin g to n JM , ed. Recent advances in p a tio n a i h e a lth . C h u rch ill L iv in g sto n e, E d in b u rg h 1! pp 253-266. 2. B a la ra ja n R . A c h e s o n E D . S o ft tissu e s a rc o m a s ii riculture and forestry w orkers. J E pidem iol Comm ty H ealth 38 (1984) 1 1 3 -1 1 6 . 3. Boice JD . C ancer follow ing m edical irradiation, cer 47 (1981) 1 0 8 1 -1 0 9 0 . 4. B u re a u In te rn a tio n a l d u T ra v a il. C la s sific a tio n inti tionale type des professions. A tar SA , G enve 1 5. C a lle E E . C rite ria fo r selection o f d e c e d e n t v ersus li| c o n tro ls in a m o rta lity case-co n tro l stu d y . A m J Ep m iol 120 (1984) 6 3 5 -6 4 2 . 6. C oggon D. A cheson ED. Do phenoxy herbicides can ce r in m an? L ancet 1 (1982) 1057-- 1059. 48r 'i 3 l <posed 3 1 * 12 7 5 Unexposed 1 2 ftE x p o s e d "' l i 11 1 :1 h is h e a d in g . ing in Piedmont, and annual herrformed, among numerous other :d rice growers or farm laborers, relationship could be considered : able to distinguish between high Dosure in the exposure assessmeni df the a priori hypothesis concernperiod). Most of the exposed suBth soft-tissue sarcoma and the remployed in rice weeding for mort e 7). *. E n zin g er F R , L a tte s R , T o rlo n i R,< e d . H isto lo g ic a k ivping o f soft-tissue tum ours: International classification f tum ours 3. W orld H ealth O rganization, Geneva i% 9. ,v E rik sso n M , H a rd e ll L , B erg N O , M lle r T , A x elso n O. Soft-tissue sarcom as and exposure to chem ical sub stances: A case-referent study. Br J Ind M ed 38 (1981) 27-33. 9 . Fingerhut M A , H alperin W E, H onchar PA , Sm ith AB, G roth D M , Russell W O . A n evaluation o f reports of dioxin exposure an d so ft tissue sarcom a pathology am ong chem ical w orkers in the U nited States. S cand J W ork E n v iro n H e a lth 10 (1984) 299-- 303. It). G allag h er R P . T h re lfa il W J . C a n c e r a n d o c c u p a tio n a l exposure to chlorophenols. Lancet 2 (1982) 48. t iordis L. S h o u ld dead cases be m atched to dead con-:ols? A m J E p id e m io l 115 (1982) 1 - 5 . .2. G ieenw ald P , K ovasznay B, C ollins D N , T h erriau lt G . Sarcom as o f so ft tissues after V ietnam service. J N atl C ancer Inst 73 (1984) 1 1 0 7 -1 1 0 9 . 13. H ard ell L . E p id e m io lo g ic a l stu d ie s o n so ft-tis su e sa r coma and m alignant lym phom a and their relation to phe noxy acid o r chlorophenol exposure. U m e U niversity, Ume 1981. (D octoral dissertation). 14. Hardell L , S a n d str m , A . C a se-co n tro l study: Soft-tissue sarcom as and exposure to phenoxyacetic acids o r chlo rophenols. Br J C ancer 39 (1979) 711-- 717. !v International A gency for Research on C ancer. Chemi.als and in dustrial processes associated w ith cancer in .iiiuans. L y o n 1979. (IA R C m o n o g ra p h s o n th e e v al uation o f the carcinogenic risk o f chem icals to hum ans, supplem ent 4). 16. Istitu to C e n tra le di S ta tistic a . A n n u a rio di sta tis tic a ag raria. R om a 1959. 17. Istitu to C e n tra le di S ta tistic a . A n n u a rio di sta tis tic a a g raria. R om a 1969. 18. Lynge E. U k ru td tsm id ler o g K rdft. A rb ejd sm iljafo n d et, C openhagen 1984. 19. M antel N , H aenszel W . Statistical aspects o f the analy sis o f d a ta fro m retro sp ectiv e studies o f disease. J N atl C ancer Inst 22 (1959) 7 1 9 -7 4 8 . 20. M iettinen O S. E stim ability and estim ation in casereferen t studies. A m J E pidem iol 103 (1976) 226 --235. 21. M ilham S. H erbicides, occupation and cancer. Lancet 1 (1982) 1 4 6 4 -1 4 6 5 . 22. N ations U nies. Index de la classification internationale type, p ar in d u strie, d e toutes les b ranches d 'activit co nom ique. New Y o rk , NY 1975. 23. Regione L o m b ard ia. E pidem iologia dei tum ori m aligni: Incidenza e m o rta lit in provincia di V arese: 1976 - 77. M ilano 1981. 24. R iihim ki V , A sp S, H ernberg S. M ortality o f 2,4-dichlorophenoxyacetic acid and 2,4,5-trichlorophenoxyacetic acid herbicide applicators in F in lan d . S cand J W o rk E n v iro n H ealth 8 (1982) 37 --42. 25. R othm an K, Boice U D . E pidem iologie analysis w ith a program m able calculator. US D epartm ent o f H ealth, E ducation an d W elfare, W ashington, DC 1979. 26. S arm a P R , Ja c o b s J . T h o racic so ft-tissu e sa rco m a in V ietnam veterans exposed to Agent O range. N Engl J M ed 306 (1982) 1109. 27. Sm ith A H , P earce N E , Fisher D O , Giles H J, T eague C A , H ow ard JK . Soft-tissue sarcom as an d exposure to phenoxy-herbicides an d chlorophenols in New Z ealand. J N a tl C a n c e r In st 73 (1984) 1111 -- 1117. 28. T ockm an M S. E pidem iology in the w orkplace: T he problem , o f m isclassification. J O ccup M ed 24 (1982) 2 1 -2 4 . R eceived fo r p u b lic a tio n : 21 J a n u a r y 1986 jpported by Regione Piemonte (Ri- alizzata) and by CSl-Piemonte. In : received a fellowship from the As- i per la Ricerca sul Cancro. | Professor B Giau (Istituto di lc- niversit di Torino) for his skilled ds o f the pathology departments f ella, Novara, Vercelli, Borgoman- rato and of the National Cancer In- )r access to their files and the sub- )gical slides. f ? ne h e a lth effects o f phenoxy acid herb! in g to n J M , e d . R e cen t a d v a n c e s in occtt- t. C h u rc h ill L iv in g sto n e , E d in b u rg h 19W- S A cheson E D . S oft tissue sarcom as in af fo restry w o rk ers. J E pidem iol Communi- (1984) 113 -- 116. ) n c e r fo llo w in g m e d ic a l ir ra d ia tio n . Cay- 1081-1090. tatio n al d u T ravail. C lassification intern ie s p r o fe s s io n s . A ta r S A , G e n ev e 1969. te r ia fo r se le c tio n o f d e c e d e n t v ersu s livini m o rta lity case-co n tro l stu d y . A m J Epidc 84' '1 5 --642. | u i E D . Do phenoxy herbicides cau? in: _ a n c e t 1 (1982) 1057-- 1059. 19149 17 i 1-L 11/- t U American Journal of Industrial Medicine 18:133-148 (1990) Epidemiologic Studies of Cancer in Agricultural Workers Neil Pearce, PhD, and John S. Reif, dvm, msc The incidence of cancer in agricultural workers is generally low, in part due to the low prevalence of cigarette smoking in this group. However, agricultural workers have elevated risks for several specific cancer types including leukemia, Hodgkin's disease. non-Hodgkin's lymphoma, multiple myeloma, and cancers of the lip, stomach, prostate, brain, and connective tissue. Two major groups of risk factors have been proposed as causes of hematologic malignancies in agricultural workers. The first group includes various agricultural chemicals. In particular, several studies have found increased risks of malignant lymphoma and soft tissue sarcoma in persons exposed to phenoxy herbi cides. However, the evidence is inconsistent and there is a wide variation in relative risk estimates. The second group of risk factors includes various animal viruses. There is currently little evidence concerning the zoonotic nature or human carcinogenicity of these viruses. However, leads have been suggested by recent evidence of increased risks of hematologic malignancies in abattoir workers, veterinarians, and meat inspectors. A third hypothesis, for which little evidence is currently available, is that agricultural work may involve prolonged antigenic stimulus leading to lymphoprolifration. The factors responsible for the increased risks for cancers other than hematologic malignancies are not well understood but may also involve exposure to chemicals or viruses. Key words: agricultural chemicals, animal viruses, antigenic stimulation, occupational cancer INTRODUCTION Agricultural work involves a wide variety of tasks, each involving potentially carcinogenic exposures. Farmers and farm laborers may come into contact with animal viruses, bacteria and fungi, pesticides, solvents, fuels and oils, and dusts [Blair et al., 1985; Lawhorne, 1976]. Other agricultural workers experience some of the same exposures. These include fencing workers, abattoir workers, meat inspec tors, and veterinarians. In this paper, available data on cancer risks in agricultural workers are reviewed. We begin by reviewing studies of overall cancer mortality and site-specific cancers in farmers. The major hypotheses concerning cancer risks in agricultural workers are then discussed, and the epidemiologic evidence for each is W ellington School o f M edicine, W ellington, New Z ealand (N .P.). C olorado State U niversity, Fort C oilins (J.S .R .). A ddress reprint requests to D r. N eil Pearce, U niversity o f O tag o , D epartm ent o f M ed icin e, W ellington School o f M edicine, P.O . Box 7343, W ellington. N ew Z ealand. A ccepted for publication M arch 30, 1990. 1990 Wiley-Liss, Inc. U-- ; '-jvJI I 134 Pearce and Reif TABLE I.Studies ofOverall Cancer Mortality in Agricultural Workers R eferen ce Howe and L in d s a y (1983) R e g istra rG en eral's (1978) Pearce and H ow ard (1978)" W iklund (1986a) W alrath et al. ( 1985)b C ountry and tim e period Canada 1965-1969 E ngland and W ales 1970-1972 New Zealand 1974-1978 Sweden 1961-1979 U nited States 1970-1984 S tu d y type M ortality M ortality M ortality incidence M ortality Study population 10% sam ple o f all m ales M ales aged 15-64 M ales aged 15-64 All agricultural w o rk ers V eterans with life insurance aged 31-84 T o ta l cancers in agricultural w orkers 75 __ * 659 24.763 824 "P re v io u sly u n p u b lish e d d a ta fro m th is stu d y . bA d a p te d fro m B la ir e t al. (u n p u b lis h e d ). R elative risk 0 .8 7 0 .9 2 0 .9 9 0 .8 2 0 .9 0 95% confidence interval 0.6 8 -1 .0 9 0.92 -1 .0 7 0 .81-0.83 0 .8 4 -0 .9 6 reviewed. Finally, we discuss some of the difficulties of investigating cancer risks in agricultural workers and suggest some priorities for further research. STUDIES OF CANCER MORTALITY IN AGRICULTURAL WORKERS A recent publication reviewed the epidemiologic studies of the cancer experi ence of agricultural workers in a number of countries and regions [Blair et al., 1985]. The present article is not intended to review all the available studies. Instead, one major study has been selected from each country. In each instance, a cancer mortality or incidence study was selected (proportional mortality and case-control studies were not considered) on the basis of the study size, availability of site specific data, and availability of appropriate data for the calculation of confidence intervals. The char acteristics of the selected studies [Howe and Lindsay, 1983; Office of Population Census and Surveys, 1978; Pearce and Howard, 1986; Wiklund and Holm, 1986a; Walrath et al., 1985] are presented in Table I. Cancers for Which Agricultural Workers Have Decreased Risks Despite the presence of potentially carcinogenic exposures, overall cancer in cidence and mortality in farmers is generally lower than for the general population (Table I). In particular, farmers have lower than average risks for cancers of the lung, bladder, nose, colon, rectum, and liver (Table II). These patterns are clearest in the Swedish study since this involved the largest numbers of cases. However, similar patterns occur in Canada, England and Wales, New Zealand, and the United States (Tables I and II). The decreased risks for cancer of the lung and bladder are presum ably due to the low prevalence of cigarette smoking in agricultural workers [Blair et rs 95% -al R elative confidence ; risk interval 0 .8 7 0.6 8 -1 .0 9 0 .9 2 -- 0 .9 9 0 .9 2 -1 .0 7 0 .8 2 0 .81-0.83 0 .9 0 0 .8 4 -0 .9 6 ligating cancer risks in search. WORKERS ; o f the cancer experins [Blair et al.. 1985]. : studies. Instead, one ace, a cancer mortality ;e-control studies were site specific data, and ;e intervals. The char- Office of Population nd and Holm, 1986a; d Risks -es, overall cancer inhe general population or cancers of the lung, rns are clearest in the es. However, similar and the United States d bladder are presumural workers [Blair et Cancer in Agricultural Workers 135 TABLE II. Studies of Cancer Types That Show a Decreased Risk for Agricultural Workers C o u n try Canada E ngland and W ales New Zealand Sweden U nited States C ountry Canada England and W ales New Zealand Sweden U nited States Lung R elative risk 0 .6 9 0 .8 4 0 .8 6 0 .3 6 0 .8 3 95% confidence interval 0 .4 2 -1 .0 8 ____ 0 .74-1.00 0 .3 4 -0 .3 8 0 .7 0 -0 .9 8 C olon R elative risk 95% confidence interval 0 .5 3 1.20 1.17 0 .7 2 0.71 0 .14-1.36 -- 0 .9 2 -1 .4 8 0 .6 9 -0 .7 6 0 .57-0.88 B lad d e r R e la tiv e risk 95% confidence interval ____ 0 .7 0 0.91 0.61 0 .5 0 ____ -- 0 .4 7 -1 .5 9 0 .57-0.65 0 .31-0.77 R e ctu m R elative risk 95% confidence interval 1.16 1.00 0 .8 0 0 .8 9 0 .7 9 0 .3 1 -2 .9 7 -- 0.5 4 -1 .1 5 0 .8 4 -0 .9 4 0.5 4 -1 .1 2 Nasal R elative risk 95% confidence interval ____ -- -- 0 .9 2 -- ____ ____ -- 0 .73-1.15 -- L iv e r R e la tiv e risk 95% confidence interval ___ -- 0 .5 9 0 .4 4 0 .8 3 ____ -- 0 .2 4 -1 .2 2 0 .37-0.52 0 .50-1.30 al., 1985], a pattern that has been observed in New Zealand [Department of Statistics, 1983], Sweden [Central Bureau of Statistics, 1965], and the United States [Sterling and Weinkam, 1976], The decreased risks of colon cancer in agricultural workers may be due to the protective effect of high physical activity [Blair et al., 1985; Garabrant et al., 1984]. The reasons for the low risks of cancer of the rectum, liver, and nose are currently unknown [Blair et al., 1985]. Cancers For Which Agricultural Workers Have increased Risks Agricultural workers have also been found to have elevated risks for several types of cancer. Increased risks have been most firmly established for hematologic malignancies, although there is little evidence for an increased risk of Hodgkin's disease in the countries considered here (Table III). Farmers also appear to have increased risks for cancers of the lip, stomach, prostate, and brain (Table IV). Some proportionate mortality studies have also found excess risks for connective tissue [Blair et al., 1985], but this was not observed in the Swedish study considered here (Table IV). The excess risks presented in Tables III and IV are small, but this is to be expected given the crude occupational information used in these studies. Further more, it is likely that the elevated risks for specific cancers are due to exposures confined to particular subgroups of agricultural workers. The excess risks in these subgroups would thus be greater than those presented in Tables III and IV. These findings are also of interest when considered in conjunction with time trends in mortality for these cancer sites. Mortality from multiple myeloma. non-Hodgkin's lymphoma, leukemia, and cancers of the brain and connective tissue has increased in the United States during the period 1950-1980, particularly in farming states [Pickle et al., 1987], Similar patterns have been observed for mortality and incidence in New Zealand [Pearce et al., 1986; Smith et al., 1984; Pearce et al., 1985], but there has been little increase in incidence in Sweden [Wiklund and Holm, 1986a]. 19152 136 Pearce and Reif TABLE III. Studies of Cancer Types That Show an Increased Risk for Agricultural Workers: Hematologic Malignancies H odg k in 's disease N on-H odgkin's lym phom a M ultiple m yelom a C o u n try Canada E ngland and W ales New Zealand Sweden U nited States R e la tiv e risk __ 1.03 0 .7 9 1.06 -- 95% confidence interval _ -- 0 .1 6 -2 .2 9 0 .9 3 -1 .2 0 -- R elative risk 0 .6 2 1.12 . 1.41 0 .9 8 -- 95% confidence Interval 0 .1 3 -1 .8 1 -- 0 .9 1 -2 .0 8 0 .9 1 -1 .0 6 -- R e la tiv e risk -- -- 1.55 1.20 -- 95% confidence interval -- -- 0.82 -2 .6 5 1 .0 9 -1 .3 3 -- Leukem ia T otal hem atologic m alignancies C o u n try R e la tiv e risk 95% confidence Interval R elative risk 95% confidence Interval Canada England and W ales New Zealand Sweden U nited States 0 .3 3 -- 1.18 0 .9 5 1.13 0 .0 0 -1 .8 4 -- 0 .7 8 -1 .7 2 0 .8 9 -1 .0 2 0 .8 5 -1 .4 7 0 .5 5 -- 1.29 1.04 -- 0.2 2 -1 .1 4 -- 1 .0 0 -1 .6 3 1 .0 0 -1 .0 8 -- TABLE IV. Studies of Cancer Types That Show an Increased Risk for Agricultural Workers: Cancers Other Than Hematologic Malignancies Stom ach L ip B rain C o u n try Canada E ngland and W ales New Zealand Sweden U nited States R e la tiv e risk _ 0 .9 7 0 .8 5 1.05 1.07 95% confidence interval _ -- 0 .5 6 -1 .2 4 1 .0 1 -1 .0 9 0.8 2 -1 .3 8 R elative risk -- -- -- 1.92 -- 95% confidence interval -- -- -- 1 .7 4 -2 .1 1 -- R elative risk 1.77 0 .9 9 1.10 1.01 1.04 95% confidence interval 0 .4 8 -4 .5 3 -- 0 .7 5 -1 .5 5 0 .9 4 -1 .0 8 0 .60-1.51 Prostate C onnective tissue C o u n try Canada E ngland and W ales New Zealand Sweden U nited States R e la tiv e risk 1.16 1.22 1.61 0 .9 6 1.06 95% confidence interval 0 .42-2.53 -- 1 .1 2 -2 .2 5 0 .9 4 -0 .9 9 0 .8 8 -1 .2 7 R e la tiv e risk _ -- -- 0 .9 8 -- 95% confidence interval _ -- -- 0 .85-1.13 -- GENERAL HYPOTHESES FOR THE INCREASED CANCER RISKS IN FARMERS The factors responsible for the increased risks of certain cancers in agricultural workers are not well understood. Hematologic malignancies have been most exten sively considered, and most studies have focused on two general groups of risk >r Agricultural W orkers: Multiple myeloma 95% e Relative confidence risk interval I 3 1.55 0.82-2.65 5 1.20 1.09-1.33 ---- 'e 3 >r A g r i c u l t u r a l W o r k e r s : Brain 95% e R elative confidence risk interval 1.77 0 .4 8 -4 .5 3 0 .9 9 -- 1.10 0 .75-1.55 1 1.01 0 .9 4 - 1 .0 8 1.04 0 .60-1.51 e 3 :er n cancers in agricultural ; have been most extengeneral groups of risk Cancer in Agricultural Workers 137 factors: agricultural chemicals and farm animal viruses. A third hypothesis, which has * not been studied in depth, is a possible role of chronic antigenic stimulation. These general hypotheses will be discussed with particular reference to hematologic malig nancies. The factors responsible for the increased risks for cancer of sites other than the lymphatic and hematopoietic system are even less well understood but may also involve exposure to chemicals or farm animal viruses [Reif et al., in press], CHEMICALS Exposure to agricultural chemicals has been the primary focus of most epide miologic studies of agricultural workers. Initial studies attempted to relate risks for leukemia [Blair and Thomas. 1979; Blair and White, 1981] and other hematologic malignancies [Burmeister et al.. 1983; Cantor, 1982; Cantor and Blair, 1984; Saftlas et al., 1987] with specific agricultural practices based on county of residence. In general, these studies have not provided a clear pattern of exposure and risk, although associations between herbicide use, fertilizers, insecticides, and various crops such as corn and soybeans have been found by this ecologic approach. In the last decade, more studies have collected information on individual chemical exposures, particu larly to insecticides and herbicides. Insecticides Arsenic is the insecticide that has been most clearly established as a human carcinogen. It is an established lung carcinogen and may also cause hematologic malignancies [Axelson, 1980], Arsenic compounds have been used as pesticides for more than a century, particularly in sheep-dips and vineyards, although their use has declined in recent decades [Pershagen. 1983]. They are still used for the preservation of farm fencing materials and a recent New Zealand study found a small excess risk of non-Hodgkin's lymphoma in fencing workers [Pearce et al., 1986, 1987]. How ever, arsenic compounds have not been widely used in other branches of agriculture and would appear to be of limited importance for cancer causation in agricultural workers [Axelson, 1987]. Studies of other insecticides have included manufacturers of organochlorine compounds, and termite and agricultural applicators. Individual exposures have not been documented but most workers appear to have been exposed to various organochlorines including chlordane, heptachlor, DDT, dieldrin, and aldrin. These studies have recently been reviewed by Blair in an unpublished manuscript who concluded that they show small but consistent excesses of lung cancer. Although the specific insecticides that caused the lung cancer excess could not be identified, the organo chlorine pesticides were the most likely candidates since they were the predominant insecticides during the relevant time periods. These findings are of considerable interest, but they do not explain the cancer mortality patterns in other agricultural workers who generally have low lung cancer risks (Table II). Of more relevance are the case-control studies of hematologic ma lignancies and soft tissue sarcoma that have collected information on insecticide exposure. In an unpublished manuscript by Blair et al., it was reported that weak but consistent excesses exist for some commonly used organochlorines including DDT and lindane. One unpublished study by Blair and two case-control studies [Hoar et al., 1986; Hoar et al., 1988] have also reported associations between hematologic * * v < :-i f v . 138 Pearce and Reif T A B L E V . R elative R isk E stim ates F rom C ase-C o n tro l S tudies o f Soft T issue S arco m a. N o n -H o d g k in 's L y m p h o m a a n d H o d g k in 's D isease a n d E x p o su re to P h en o x y H erb icid es* Study Exposed cases Relative risk 95<7c confidence interval Soft tissue sarcom a Hardell and Sandstrom (1979) Eriksson et al. (1981) H ardell and Eriksson (1988) Sm ith et al. (1984) Sm ith and Pearce (1986) H oar et al. ( 1986)a W oods et al. (1987) V ineis et al. (1987) N on-H odgkin's lym phom a Hardell et al. (1981) H oar et al. (1986) Pearce et al. (1987) W oods et al. (1987) B lair et al. (unpublished) H oar Zahm et al. (1988) Persson et al. (1989) H o dgkin's disease Hardell et al. (1981) H oar et al. (1986)a 13 14 9 21 6 15 b 5 27 24 44 b 8 b 12 14 13 5.3 2 .4 -1 1 .5 6.8 2 .6 -1 7 .3 3.3 1 .4 -8 .1 1.3 0 .6 - 2 .9 0.7 0 .3 -1 .9 1.4 0 .8 - 2 .3 0.9 0 .4 -1 .9 2.4 0 .5 -1 2 .4 4.7 2 .6 -8 .3 2.2 1 .2-4.1 1.0 0 .6 - 1 .7 1.2 0 .8 - 1 .9 2.0 0 .7 -5 .2 1.5 0 .9 - 2 .4 1.6 0 .7 - 3 .5 5.0 2 .4 -1 0 .2 1.0 0 .5 - 1 .7 in c lu d e s 32 cases o f H o dgkin's disease and 84 cases o f non-H odgkin's lym phom a. aA dapted from B lair et al. [unpublished). hN o t re p o rte d . malignancies and several organophosphate insecticides, which have largely replaced the organochlorines. Herbicides The carcinogenic effects of herbicides have been more intensively studied than those of insecticides. Attention has been focused on the phenoxy herbicides, a group which includes 2,4,5-trichlorophenoxyacetic acid (2,4.5-T) and 2.4-dichlorophenoxyacetic acid (2,4-D); 2,4,5-T has been of particular interest because early formulations were contaminated with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). 2,4-D and other commonly used phenoxy herbicides are not contaminated by TCDD but are contaminated with other dioxins and dibenzofurans. The current interest in phenoxy herbicides was initiated by a series of casecontrol studies in Sweden that found increased risks of soft tissue sarcoma [Hardell and Sandstrom, 1979; Eriksson et ai., 1981] and malignant lymphoma [Hardell et al., 1981] in agricultural workers who had been exposed to phenoxy herbicides (Table V). These findings have been further investigated in case-control studies in Sweden [Hardell and Eriksson, 1988; Persson et al., 1989], New Zealand [Smith et al.. 1984; Pearce et al., 1987; Smith and Pearce, 1986], the United States [Hoar et al.. 1986. 1988; Woods et al., 1987], and Italy [Vineis et al., 1987], The findings of these studies are summarized in Table V. The principal feature of the table is the considerable variation in relative risk estimates. For soft tissue sarcoma, the original two Swedish studies [Hardell and Sandstrom. 1979; Eriksson et 18135 thosi class for S' apprc contr cohoi Hodg fount. 0.6- ; relati' study explai bias oi of hea been f. of new Howe' was ca worker bias w; have us risk est Kansas little e su e Sarcom a, xv H erbicides* 95'3- confidence interval 2 .4 - 11.5 2 .6 - 17.3 1.4- 8.1 0. 6- 2.9 0.3-1.9 0 .8- 2.3 0.4-1.9 0.5-12.4 2. 61.2-4.1 0. 6- 1.7 0. 8- 1.9 0.7-5.2 0.9-2.4 0.7-3.5 8.3 2 .4 - 10.2 0.5-1.7 jm a. ve largely replaced isively studied than herbicides, a group and 2,4-dichloro:rest because early -p-dioxin (TCDD). iminated by TCDD y a series of casee sarcoma [Hardell )ma [Hardell et al., rbicides (Table V). studies in Sweden Smith et ai., 1984; Tioar et al.. 1986, le principal feature es. For soft tissue , 1979; Eriksson et al.. 1981] found relative risks of 5 or 6, and a more recent study [Hardell and Eriksson, 1988] found a relative risk of 3.3. The estimates in the two New Zealand [Smith et al., 1984; Smith and Pearce, 1986] and two United States [Hoar et al.. 1986; Woods et al., 1987] studies were much closer to 1.0. Vineis [Vineis et al.. 1987] found a two-fold risk of soft tissue sarcoma in female rice weeders in Italy who had contact with 2,4,5-T. Several case-control studies have found elevated risks for non-Hodgkin's lymphoma In persons exposed to phenoxy herbicides. Hardens study [Hardell et al.. 1981] found a six-fold risk, whereas a more recent Swedish study by Persson [Persson et al., 1989] found a relative risk of 1.6 for malignant lymphoma (primarily nonHodgkin's lymphoma). Hoar [1986] found a two-fold risk in Kansas, but the risk was more than seven-fold in persons who reported using herbicides for more than 20 days per year. In the initial findings of a similar study in Nebraska [Hoar et al., 1988], a relative risk of 1.5 has been reported both for 2,4,5-T and 2,4-D. In Iowa and Minnesota, Blair reported in an unpublished manuscript that a two-fold risk occurring 20 years after exposure to 2,4,5-T was found, but little evidence was found for an increased risk for 2,4-D. On the other hand, Pearce [Pearce et al., 1987] in New Zealand, and Woods [Woods et al., 1987] in Washington State, found little evidence of elevated risks for non-Hodgkin's lymphoma in persons exposed to phenoxy herbicides. The two case-control studies of Hodgkin's disease also show a marked disparity in their findings (Table V). A series of cohort studies in Sweden has also yielded findings at variance with those of the Swedish case-control studies. A study of 354,620 Swedish men who were classified as farmers or forestry workers in the I960 census found a relative risk of 0.9 for soft tissue sarcoma (95% confidence interval 0.8-1.0), whereas a relative risk of approximately 1.5 would have been expected in this group on the basis of the casecontrol study findings [Wiklund and Holm. 1986b]. A further analysis of the same cohort found no evidence of an increased risk of non-Hodgkin's lymphoma or Hodgkin's disease [Wiklund et al., 1988]. A similar study of pesticide applicators found relative risks of 1.0 for non-Hodgkin's lymphoma (95% confidence interval 0.6-1.5) and 1.2 for Hodgkin's disease (95% confidence interval 0.6-2.2), whereas relative risks of approximately 4.5 were expected on the basis of the case-control study findings [Wiklund et al., 1987]. The wide range of relative risk estimates is puzzling, and a number of possible explanations have been offered. The Swedish studies have been criticized with regard to the possibility of recall bias or interviewer bias when comparing the exposure history of cancer cases with those of healthy controls drawn from the general population. Such tendencies could have been fostered " by suggestion during previous interviews in hospital and by the effects of newspaper publicity on patients and their relatives" [Coggon and Acheson, 1982]. However, a simple check, based on a principle suggested by Axelson [Axelson, 1980], was carried out by examining the findings for occupations in which some, but not all, workers had been exposed to phenoxy herbicides. It was concluded that serious recall bias was unlikely to have occurred. Swedish case-control studies of other cancer sites have used the same study design and have not found six-told risks, although the relative risk estimates were slightly elevated [Hardell, 1981; Hardell etal., 1982, 1984], Hoar's Kansas study [Hoar et al., 1986] used a general population control group and found little evidence for an increased risk for soft tissue sarcoma or Hodgkin's disease LVl_ll>-- 0 - 1 1-1- I U H U ;; j ! ;|; ij ; ; ;j ;' ii jj Ij ii 19156 ! 140 Pearce and Reif (Table V). The New Zealand non-Hodgkin's lymphoma study used both general population controls and other cancer controls at a time when there was considerable media publicity concerning possible health effects of 2,4,5-T and found similar reported exposure patterns in the two groups [Pearce et al., 1986], The recent Swedish soft tissue sarcoma study [Hardell and Eriksson, 1988] also used both general pop ulation controls and cancer controls and also found similar reported exposure patterns in the two control groups. Thus, although the information bias hypothesis cannot be excluded, there is currently little evidence to support it [Axelson, 1985]. The validity of the control group should also be considered for those studies that have not found an increased risk. The New Zealand studies used other cancer patients as controls in order to minimize the possibility of information bias. Selection bias could occur using this study design if other cancers were also caused by phenoxy herbicide exposure. However, any such bias is likely to be small [Smith et al.. 1988] since the overall cancer mortality in New Zealand farmers is virtually identical to national rates (Table I). Furthermore, as noted above, one New Zealand study [Pearce et al.. 1986] and one Swedish study [Hardell and Eriksson, 1988] used both general population and other cancer controls, and in both studies, the two control groups gave similar findings. Differences in case ascertainment could also be responsible for the differing study findings. The New Zealand case-control studies of soft tissue sarcoma were confined to cases registered under International Classification of Diseases (ICD) [World Health Organization, 1977] code 171, whereas only one half of the Swedish cases were registered under ICD code 171. However, the effect estimates in this subgroup were similar to those in the cases registered under other ICD codes (L Hardell. personal communication). Thus, although all the published case-control studies have strengths and weak nesses. it is difficult to identify methodological problems that could cause five-fold differences in the relative risk estimates. Accordingly, the characteristics of the populations and exposures studied should also be considered. If the TCDD contaminant of 2,4,5-T is involved, then it is possible that the levels of contaminants have been different in Sweden, New Zealand, the United States, and Italy. However, available evidence does not- suggest major differences between Sweden and New Zealand [Smith et al., 1984], Furthermore, the Kansas study predominantly involved exposure to 2,4-D, and the Swedish studies also do not necessarily implicate TCDD since, in at least one study [Eriksson et al., 1981], an elevated risk for non-Hodgkin's lymphoma was also observed in persons exposed to 2,4-D and other phenoxy herbicides that do not contain TCDD [Axelson, 1987], Differences in spraying practices could also be important. Owing to the climate, spraying in Sweden is usually carried out intensively during a 2-3-month period, whereas spraying in New Zealand and Washington State occurs intermittently over a longer period [Woods et al., 1987], These differences could result in Swedish her bicide sprayers receiving a relatively high absorbed dose. This conjecture is consis tent with the finding of Hoar [Hoar et al., 1986] of markedly increased risks of non-Hodgkin's lymphoma (but not soft tissue sarcoma or Hodgkin's disease) in persons spraying herbicides for more than 20 days per year. However, a re-analysis of the New Zealand non-Hodgkin's lymphoma data found a relative risk of 1.1 (95% confidence interval 0.3-4.1) in persons spraying 20 days or more per year (Pearce, unpublished data). used both general re was considerable and found similar The recent Swedish 1 both general popd exposure patterns /pothesis cannot be i, 1985]. or those studies that ther cancer patients )ias. Selection bias :aused by phenoxy Sm ith et al., 1988] irtually identical to aland study [Pearce ] used both general :ontrol groups gave le for the differing ssue sarcoma were of Diseases (ICD) lalf of the Swedish :t estimates in this *' ICD codes (L trengths and weakuld cause five-fold aracteristics of the s possible that the ;aland, the United : major differences rmore, the Kansas studies also do not n et al., 1981], an persons exposed to Axelson, 1987]. ving to the climate, 2-3-month period, itermittently over a lit in Swedish heranjecture is consis- increased risks of gkin's disease) in ever, a re-analysis 'e risk of 1.1 (95% j per year (Pearce, Cancer in Agricultural Workers 141 A further possibility is that phenoxy herbicide exposure could be carcinogenic only when occurring jointly with other exposures present in Sweden and Kansas, but not New Zealand [Pearce et al., 1986, 1987]. Some support for this conjecture has been provided by the study of Woods [Woods et al., 1987]. Overall there was little evidence of elevated risks, but when the analysis was restricted to subjects with Scandinavian surnames, the odds ratio estimates for soft tissue sarcoma (but not non-Hodgkin's lymphoma) were elevated for phenoxy herbicide exposure (odds ratio = 2.8) and chlorophenol exposure (odds ratio = 7.2). Furthermore, the distribution of histologic types was similar to that in the Swedish studies. Woods et al. [1987] have suggested that faetors specific to Scandinavian descent may contribute to in creased risks of soft tissue sarcoma when exposed to certain chemicals. The relevant factors could be genetic, a hypothesis that is consistent with at least some studies of the binding affinity of the TCDD contaminant in tissues [Woods et al., 1987], Al ternatively, the relevant factors could be environmental, including the use of other agricultural chemicals in combination with phenoxy herbicides and chlorophenols. For example, in the recent Swedish malignant lymphoma study [Persson et al., 1989] the relative risk associated with exposure to phenoxy herbicides in combination with exposure to solvents and fresh wood was 2.7 (90% confidence interval 0.7-9.4), whereas exposure to these factors alone was associated with relatively modest in creases in risk. Consideration of such joint effects can be expected to play an in creasingly important role in occupational epidemiology [Axelson, 1988], The hypothesis that phenoxy herbicides may only be carcinogenic in combina tion with other factors is of particular interest in light of recent developments in the understanding of mechanisms of carcinogenesis. The phenoxy herbicides are not classic mutagens, and there is currently little evidence that they are direct animal carcinogens. However, it is plausible that they could be tumor promoters. Promotion may involve a variety of mechanisms including interference with DNA repair mech anisms, cytotoxity. inhibition of cellular differentiation, immunosuppression, mitogenesis, and inhibition of intercellular communication [Williams, 1984]. These ef fects do not appear to involve genetic change and would require concomitant exposures to other carcinogens. Similar mechanisms have been suggested for arsenic, which is an established human carcinogen [Pershagen, 1983] despite the lack of evidence of animal carcinogenicity [Landrigan, 1981]. Viruses Although epidemiologic evidence for the role of oncogenic animal viruses in human cancer is currently weak [Blair et al., 1985], a number of potentially zoonotic viruses exist in the agricultural environment. These include the herpes virus that causes Marek's disease in poultry [Witter, 1972], the avian leukosis virus [Burmeister and Purchase, 1970], papilloma viruses in cattle [Jarrett et al., 1978; Jarrett, 1980; Smith and Campo, 1985], and perhaps other currently undetected viruses. The rapid proliferation of information concerning retroviruses suggests that other members of this family will be found in domestic animals as well as humans. Most interest has centered on the bovine leukemia virus (BLV), an exogenous C-type retrovirus that has been established as the etiologic agent of the adult form of bovine lymphosarcoma (BLS) [Miller et al., 1969], The virus is related closely to HTLV-1, the cause of adult T-cell leukemia in humans [Jakobs, 1983]. BLV has been found in herds in most countries, including the United States and New Zealand  ry* 142 Pearce and Reif [Parrish et al., 1982], The virus has also been found in meat and unpasteurized milk [Ferrer et al., 1981]. Although BLV is killed by pasteurization, consumption of unpasteurized milk is extremely common among families and rural residents [Donham et al., 1977]. BLV can also infect sheep and goats, and can cause cancer in sheep [Jakobs, 1983], It is capable of infecting human cells and inducing syncytia in tissue culture [Diglio and Ferrer, 1976]. Preliminary human studies have not shown evidence of infection [Donham et al., 1977], although this may be because current laboratory techniques or knowledge of the mechanism of infection are not sufficiently devel oped. In vivo, BLV infected cells or tumor cells do not express BLV [Kettman et al., 1978], Therefore the failure to find antibodies to BLV in human seroepidemiologic studies should not be taken as definite evidence against infection. Numerous reports of human leukemia clusters have been published that have attempted to show a relationship between cattle or the dairy industry and the occur rence of various hematologic malignancies [Blair and Thomas., 1979; Bartsch et al., 1975; Heath. 1970; Henricson and Ringertz, 1968]. Most studies have shown little evidence of an association, although a geographic correlation has been demonstrated in Sweden and the Soviet Union [Khoklova and Rakhamin, 1970] between areas with a high frequency of BLS and high human leukemia rates. In Iowa. Donham [Donham et al., 1980] found a high positive correlation between the incidence of acute lym phatic leukemia (ALL) in males and cattle density, which was most striking for dairy cattle. A Swedish case-control study found that leukemia patients had BLS in their cattle herds more frequently than did neighborhood controls [Kvarnfors et al., 1975]. However, a recent United States case-control study [Donham et al., 1987] found that the prevalence of antibodies to BLV in dairy herds to which cases of acute lymphatic leukemia (ALL) had been exposed was actually lower than that found in herds with which the controls had had contact. Abattoir Workers Thus, most studies of farm animal viruses have not found an association with human cancer, although this may be due to the difficulties of assessing exposure to these viruses in farmers. Possible leads have been suggested by recent studies of abattoir workers, a group that may have high exposure to animal viruses [Pearce et al., 1988], Table VI presents the findings of five New Zealand case-control studies that have found increased risks of soft tissue sarcoma (STS), non-Hodgkin's lymphoma (NHL), and acute myeloid leukemia (AML) in abattoir workers [Pearce et al., 1986; Smith et al., 1984; Pearce et al., 1986,1987; Smith and Pearce, 1986]. The table also shows the cohort study by Johnson [Johnson et al., 1986a] of mortality in 13.844 white male members of a meat-cutters union in Baltimore, MD, of whom 3,025 had worked in abattoirs. The findings are difficult to interpret due to the small numbers of deaths, but Johnson's data show some consistency with the New Zealand findings. In particular, there were elevated risks for Hodgkin's disease (ICD 201) and for the category of " cancer of other lymphatic tissue," which included non-Hodgkin's lym phoma cases registered under ICD 202 and multiple myeloma (ICD 203). No in creased risk was observed for non-Hodgkin's lymphoma cases registered under ICD 200. There was also no increased risk for all leukemias combined, but specific data for acute myeloid leukemia were not presented. isteurized milk onsumption of esidents [Don- sheep [Jakobs, l tissue culture /n evidence of -ent laboratory iciently devel<ettman et al., iepidemiologic shed that have and the occurBartsch et al., /e shown little t demonstrated een areas with ham [Donham of acute lymiking for dairy 1 BLS in their ; et al., 1975]. 871 found that ymphatic in herds with sociation with g exposure to ent studies of ses [Pearce et )1 studies that l's lymphoma e et al., 1986; The table also Tty in 13,844 om 3,025 had mall numbers land findings. 1) and for the )dgkin's lym203). No ined under ICD specific data Cancer in Agricultural Workers 143 TABLE VI. Relative Risk Estimates for Studies Showing Excess Risks of Soft Tissue Sarcoma, Non-Hodgkin's Lymphoma or Acute Myeloid Leukemia in Abattoir Workers* Study ICD 95% Site Code Exposed cases Relative risk confidence interval Smith et al. (1984) STS Smith and Pearce (1986) STS 171 171 19 2.8 1.1-7.3 11 1.6 0.8-3.5 Pearce et al. (1987) NHL 200 24 1.8 0.9-3.5 Pearce et al. (1987) NHL 202 19 1.7 0.9-3.4 Johnson et al. (1986a) NHL 200 1 0.4 0.0-2.2 NHL. 202.203 5 2.2 0.7-5.1 MM 208 Johnson et al. ( 1986b) HD 201 4 2.4 0.8-7.5 Pearce et al. (1986) AML 205.0 9 2.5 1.1-5.7 *STS. soft tissue sarcoma: NHL. non-Hodgkin's lymphoma: MM. multiple myeloma; HD. Hodgkin's disease: AML. acute myeloid leukemia. Abattoir work may involve significant chemical exposures; 2.4,6-TCP, which contains TCDD as a contaminant, has been used in the treatment of pelts [Smith et al., 1984; Glover et al., 1975], However, the New Zealand studies found that pelt department workers have similar risks to those of other abattoir workers [Pearce et al., 1988], Various plastics have been used to wrap meat, and the thermal decom position products may include benzene, phthalic anhydride, and phthalates. Johnson et al. [1986b] found a three-fold risk of mortality from myeloid leukemia and nonHodgkin's lymphoma among female workers in the meat department of retail food stores. Elevated lung cancer rates in butchers and meatcutters have also been ob served in several countries [Reif et al., 1989] and have also been attributed to exposures in the wrapping process [Johnson et al., 1986a.b]. However, a recent Swedish study [Gustavsson et al., 1987] did not find that lung cancer risks were associated with wrapping, and it appears that in most abattoirs only a few workers are directly exposed to fumes from the wrapping process. Johnson et al. [1986a] have also noted that the greatest risk for Hodgkin's disease in their study of male abattoir workers was associated with slaughtering, particularly of cattle, pigs, and sheep. This finding is consistent with the suggestion of a viral etiology [Pearce et al., 1986; Johnson et al., 1986a], and this is currently the leading hypothesis to explain the findings for abattoir workers [Pearce et al., 1988]. Other Relevant Populations Veterinarians are another occupational group with exposure to farm animals. There is inconsistent evidence that veterinarians may be at increased risk for cancer of hematologic malignancies. Matanowski and Lilienfeld [1976] reported an SMR of 1.48 for Hodgkin's disease in a mortality study of 20,000 veterinarians. Blair and Hayes [1982] studied 5,016 deaths in white male veterinarians and found elevated risks for Hodgkin's disease (PMR= 1.87, 95% confidence interval 1.11-2.96) and cancer of other lymphatic tissue (PMR = 1.92, 95% confidence interval 1.26-2.82). Interestingly, the PMR for all hematologic malignancies was elevated in meat in spectors (PMR = 3.36, 95% confidence interval 1.44-6.57), a finding consistent with the increased risks in abattoir workers. :: iii ; I :\ i !:!i !! \ 144 Pearce and Reif Chronic Antigenic Stimulation A third group of possible etiologic factors is suggested by the hypothesis that multiple myeloma and other hematologic malignancies may occur after prolonged antigenic stimulation [Greene, 1982; Blattner, 1982]. Gallagher et al. [1983] found elevated risks of multiple myeloma in farmers and in persons reporting a history of allergy. However, the role of allergies did not explain the excess risk for fanners. Cunningham [1976] has suggested that ingestion of anirpal proteins, particularly bovine protein, results in chronic stimulation of lymphoid tissue which may act in combination with other factors such as oncogenic viruses to induce malignant change. There are other sources of chronic antigenic stimulation that may also be more prevalent in agricultural communities. Chronic exposure to mitogenic substances from a grain elevator has been postulated to be a factor in a cluster of Hodgkin's disease cases [Schwartz, 1978], and a proportionate mortality study of workers in the grain industry [Alvanja et al., 1987] found elevated risks for hematologic malignan cies. It has also been suggested that multiple myeloma may develop following chronic antigenic stimulation from repeated courses of allergen vaccines for diseases such as hay fever [Woodroffe, 1972]. Thus, the chronic antigenic stimulation hypothesis warrants further investigation, particularly with regard to hematologic malignancies in abattoir workers who have continual exposure to animal protein. DISCUSSION Given the wide variety of exposures involved, it is perhaps not surprising that the reasons for the increased risks for certain cancers in agricultural workers remain unclear. The problem of multiple exposures is a limitation that applies to all epide miologic studies and has not prevented the discovery of other environmental carcin ogens such as alcohol and cigarette smoking. In principle, the carcinogenic effects of specific herbicides and insecticides can be validly measured, providing appropriate information is collected on all other potentially confounding agricultural exposures. Nevertheless, an identification problem occurs when agricultural workers are ex Ai-- posed, perhaps simultaneously, to several chemicals. This problem is of particular concern when assessing the contributions of the so-called inert ingredients. For ex ample, if a particular herbicide is always used in combination with particular types of solvents and emulsifiers in a particular country, then it may be very difficult to identify the agents responsible for an increased cancer risk associated with use of the herbicide. These considerations imply that the most efficient means of ascertaining the health effects of pesticide exposure in agricultural workers may be to study workers who produce these pesticides, rather than the agricultural workers who use them. However, exposures in pesticide producers may be much lower than exposures in users. Furthermore, if particular pesticides are only carcinogenic in combination with other exposures in the agricultural environment, then these effects may not be identified in studies of production workers. Thus, it is also important to study pro fessional pesticide applicators who use a variety of agricultural chemicals, provided that it is possible for the effects of specific chemicals or combinations of chemicals to be isolated. Studies of farmers are unlikely to clarify the situation because of the difficulty of identifying the effects of specific risk factors [Axelson, 1987], In this regard, the international collaborative cohort studies being conducted by the Inter- 13181 r ;ted by the hypothesis that nay occur after prolonged lagher et al. [1983] found sons reporting a history of le excess risk for farmers, mai proteins, particularly d tissue which may act in >induce malignant change, n that may also be more ; to mitogenic substances in a cluster of Hodgkin's lity study of workers in the for hematologic malignandevelop following chronic ccines for diseases such as lie stimulation hypothesis hematologic malignancies al protein. r ^aps not surprising that , .ultural workers remain n that applies to all epidether environmental carcinthe carcinogenic effects of red, providing appropriate ;ng agricultural exposures, ricultural workers are exis problem is of particular . inert ingredients. For ex ion with particular types of may be very difficult to . associated with use of the ient means of ascertaining workers may be to study cultural workers who use nuch lower than exposures rcinogenic in combination n these effects may not be so important to study proItural chemicals, provided :ombinations of chemicals le situation because of the s [Axelson, 1987], In this ig conducted by the Inter- Cancer in Agricultural Workers 145 % national Agency for Research on Cancer [1990] and the United States National Institute for Occupational Safety and Health [Fingerhut et al., 1984] will be of particular value in establishing whether phenoxy herbicides cause cancer in agricul tural workers. These cohort studies of production workers and professional pesticide applicators will avoid many of the methodological problems, real or imaginary, of the previously conducted case-control studies. Studying the potential human carcinoge nicity of animal viruses presents even more problems, since there are difficulties with documenting exposures to animal viruses in agricultural workers, or even in deter mining whether exposure occurs at all. In this respect, there may be considerable advantages to studying abattoir workers. They form a large enumerable group suitable for cohort studies [Johnson et al., 1986a,b]. They do not experience the wide variety of exposures experienced by farmers but have been demonstrated to have similarly increased risks for hematologic malignancies. Furthermore, exposure to animal vi ruses and other biological agents is likely to be high in slaughtering but relatively low in other abattoir workers. Thus, further studies of abattoir workers, rather than studies of farmers, may be the most productive avenue for investigating the causes of the increased risks for those cancer sites associated with contact with farm animals. REFERENCES A lavanja M C R , R ush G A , S tew art P , B lair A ( 1987): P roportionate m ortality study o f w orkers in the grain industry. JN C I 78:247-252. A xelson O ( 1980): A note on o b serv atio n al bias in case-referen t stu d ies in o ccu p atio n al health ep id em i ology. Scand J W ork E nviron Hlth 6 :8 0 -8 2 . A xelson 0 (1987): Pesticides and can cer risks in agriculture. M ed O ncol T u m o r P harm acother 4 :2 0 7 217. A xelson O (1988): O ccu p atio n al c a n c e r Interaction o f factors. In E ylenbosch W J, V an L arabeke N, D epoorter AM (eds): " Prim ary Prevention o f C a n c e r." N ew York: Raven Press. Axelson O (1980): A rsenic com pounds and cancer. J Toxicol E nviron H lth 6 :1 2 2 9 -1 2 3 5 . A xelson O (1 9 8 5 ): T h e c a se -re fe re n t stu d y : S o m e co m m e n ts o n Its stru c tu re , m erits an d lim ita tio n s. Scand J W ork Environ H lth 11:207-213. Bartsch D C , Springher F, Falk H (1975): A cute nonlym phocytic leukem ia: A n adult cluster. JA M A 2 3 2 :1 3 3 3 -1 3 3 6 . B lair A . M alker H . C antor K P, B urm eister L, W iklund K (1985): C ancer am ong farm ers: A review . Scand J W ork E nviron H lth 11:397-407. B lair A. T hom as T L (1979): L eukem ia am ong N ebraska farm ers: A death certificate study. Am J Epidem iol 11 0 :2 6 4 -2 7 3 . Blair A . W hite D W (1981): D eath certificate study o f leukem ia am one farm ers from W isconsin. JN CI 6 6 :1 0 2 7 -1 0 3 0 . Blair A. H ayes HM Jr (1982): M ortality patterns am ong US veterinarians: 1947-1977: An expanded study. Int J E pidem iol 11:391-397. Blattner W A (1982): M ultiple m yelom a and m acroglobulem ia. In: Schottenfeld D , Fraum eni, Jr JF (eds): "C an cer E pidem iology and P revention." Philadelphia: W B Saunders. B u rm eister L F , E verett G D , V an L ie r S F , Isacson P (1983): S elected m o rta lity and farm p ractices in Iow a. A m J E pidem iol 118:72-77. B urm eister B R , Purchase H G (1970): O ccurrence, transm ission and oncogenic spectrum o f the avian leukosis viruses. Bibl H aem atol 3 6 :8 3 -9 5 . C antor KP (1982): Farm ing and m ortality from N o n -H odgkin's lym phom a: A case-control study. Int J Cancer 22:239-247. C antor K P, B lair A (1984): F arm ing and m ortality from m ultiple m yelom a: A case-control study w ith the use o f death certificates. JN CI 72:251-255. Central B ureau o f Statistics (1965): " S m oking habits in Sw eden: A m ail survey-- S pring 1 9 63." S tock holm : C entral B ureau o f S tatistics. 19162 146 Pearce and Reif C o g g o n D , A c h e s o n E d (1 9 8 2 ): D o p h e n o x y h e rb ic id e s c a u s e c a n c e r in m a n ? L a n c e t i: 1 0 5 7 -1 0 5 9 . C un n in g h am A S (19 7 6 ): L y m p h o m as and an im al-p ro tein c o n su m p tio n . L ancet ii: 1 1 8 4 -1 1 8 6 . D epartm ent o f Statistics (1983): " N ew Zealand C ensus of Population and D w ellings 1981: Bulletin on C igarette S m oking." W ellington: D epartm ent o f Statistics. D iglio C A , Ferrer JF (1976): Induction o f syncytia by the bovine C type leukem ia virus. C ancer Res 3 6 :1 0 5 6 -1 0 6 7 . D onham K J, B erg JW , Savin RS (1980): E pidem iologic relationships o f the bovine population and hum an leukem ia in Iow a. Am J E pidem iol 1 1 2 :8 0 -9 2 . D onham K J, B urm eister LF. V an L ier S F . G reiner T C (1987): R elationships o f bovine leukem ia virus prevalence in dairy herds and density o f dairy cattle to hum an lym phocytic leukem ia. A m J Vet Res 4 8 :2 3 5 -2 3 8 . D onham K J. V anD er M aaten M J. M iller JM . K ruse BC . R ubino M J (1977): Sereopidem iologic studies on the possible relationships o f hum an and bovine leukem ia: B rief com m unication. J N atl C ancer Inst 59:851-853. E riksson M , H ard ell L. B erg N O . M o ller T , A xelson O (1981): S o ft tissue sarco m as and ex p o su re to chem ical substances: A case-referent study. Br J Ind M ed 38:2 7 -3 3 . Ferrer JF , K enyon SJ, G upta P (1981): M ilk of dairy cow s frequently contains a leukem ogenic virus. Science 213:1014-1016. Fingerhut M A , M arlow D A . H onchar PA. H alperin W E (1984): T he N IO SH occupational dioxin registry. In: L o w ran ce W W (ed): " P u b lic H ealth R isks o f the D io x in s ." L os A lto s. C A : W illiam K aufm an. Inc., pp 361-366. G allag h er R P , S p in elli JJ. E lw ood JM . S k ip p en D H (1983): A llerg ies and agricultural ex p o su re as risk factors for m ultiple m yelom a. B r J C ancer 4 8 :8 5 3 -8 5 7 . G arabrant D H , Peters JM . M ack T M . B ernstein L (1984): Job activity and colon cancer risks. Am J E pidem iol 119:1005-1014. G lover JC . H ughes IR. V ivian G W (1975): M ould grow th on pickled pelt. J Soc Leath Tech Chem 5 9 :7 4 -7 8 . G reene M H (1982): N on-H odgkin's lym phom a and m ycosis fungoides. In: S chottenfeld D . F raum eni JF Jr (eds): " C a n c e r Epidem iology and P re v en tio n ." P hiladelphia: W B Saunders. G ustavsson P. Fellenius E. H ogstedt C (1987): Possible causes o f increased lung can ce r incidence am ong butchers and slaughterhouse w orkers. Scand J W ork E nviron H ealth 1 3 :518-523. Hardell L . S andstrom A (1979): C ase-control study: Soft-tissue sarcom as and exposure to phenoxyacetic acids or chlorophenols. Br J C ancer 39:711-717. Hardell L, B engtsson N O . Eriksson S , Larsson LG (1984): A etiological aspects on prim ary liver cancer w ith special regard to alcohol, organic solvents and acute interm ittent porphyriaan epidem iological investigation. Br J C ancer 5 0 :3 8 9 -3 9 7 . Hardell L , Johansson B . A xelson O (1982): Epidem iological study o f nasal and nasopharyngeal cancer and th eir relation to phenoxy acid o r ch lorophenol ex p o su re. A m J Ind M ed 3 :2 4 7 -2 5 7 . Hardell L (1981): R elation o f soft tissue sarcom a, m alignant lym phom a and colon cancer to phenoxy acids, chlorophenols and other agents. Scand J W ork Environ H lth 7 :1 1 9 -1 3 0 . H ardell L . E riksson M , L enner P, L undgren E (1981): M alignant ly m p h o m a and exposure to chem icals, especially organic solvents, chlorophenols and phenoxy acids: A case-control study. B r J C ancer 4 3 :1 6 9 -1 7 6 . H ardell L, E riksson M (1988): T he association betw een soft tissue sarcom as and exposure to p h enoxy acetic acids: A new case-referent study. C ancer 6 2 :6 5 2 -6 5 6 . H eath C W J r (1970): H um an leukem ia: G enetic and environm ental clusters. Bibl H aem atol 3 6 :6 4 9 -6 5 3 . H enricson B . R ingertz N (1968): C om parative statistical studies on the geographical distribution o f hum an and bovine leukem ia in Sw eden. B ibl H aem atol 3 1 :3 3 0 -3 3 2 . H oar SK . B lair A . H olm es FF, Boysen C D , Robel R J. H oover R, Fraum eni JF (1986): A gricultural herbicide use and risk o f lym phom a and soft tissue sarcom a. JA M A 256:1141-1147. H o arZ ah m S , W eisenbergerD D , B abbitt P A . Saal R C , C antor K P, B lair A (1988): A case-control study o f n o n -H o d g k in 's lym phom a and agricultural factors in eastern N ebraska (ab stract). A m J E pi dem iol 128:901. H ow e G R . L indsay JP (1983): A follow -up study o f a ten-percent sam ple o f the C anadian labor force: I. C a n cer m o rtality in m ales. 1 9 6 5 -7 3 . JN C I 7 0 :3 7 -4 4 . IA RC W orking G roup (1990): Phenoxy A cid H erbicides and C ontam inants: D escription of the IA RC International R egister o f W orkers. Am J Ind M ed 1 8 :3 9 -4 6 . ;r " in m a n ? L a n c e t i: 1057--1059. t. Lancet ii:l 184-1186. an d D w ellings 1981: Bulletin on ty p e leukem ia virus. C ancer Res ' the bovine population and human o n sh ip s o f bovine leukem ia virus lphocytic leukem ia. Am J Vet Res 1977): Sereopidem iologic studies e f com m unication. J Natl C ancer tissu e sarco m as and ex posure to 7-33. / co n tain s a leukem ogenic virus. )S H occupational dioxin registry, o s A ltos. C A : W illiam K aufm an, a n d ag ricultural exposure as risk y and colon cancer risks. Am J i pelt. J Soc Leath Tech Chem " chottenfeld D . Fraum eni JF , Saunders. sed lung cancer incidence among :alth 13:518-523. s an d exposure to phenoxyacetic aspects on prim ary liver cancer en t porphyriaan epidem iological iasal and nasopharyngeal cancer i J Ind M ed 3:247-257. ia a n d c o lo n c a n c e r to ph en o x y !th 7 :1 1 9 -1 3 0 . a m a and exposure to chem icals, case-control study. Br J Cancer )m as an d exposure to phenoxy- rs. Bibl H aem atol 3 6 :6 4 9 -6 5 3 . ^graphical distribution o f hum an lum eni JF (1986): A gricultural 4A 256:1141-1147. A (1988): A case-control study 'leb rask a (abstract). A m J Epi- o f th e C a n a d ia n la b o r force: I. ants: D escription o f the 1ARC Cancer in Agricultural Workers 147 akobs R M (1983): B ovine L ym phom a. In: O lson R G , K rakow ska S , B lakeslee JR Jr (eds): " C o m p a r ative Pathobiology o f V iral D iseases." B oca R aton FL: C R C Press. Jurrett W FH (1980): B racken fern and p apillom a virus in bovine alim entary cancer. B R M ed Bull 3 6 :7 9 -8 1 . Jarrett W FH . M cN eil PE, G rim shaw W TR . Selm an IE, M cIntyre W1 (1978): High incidence area o f cattle cancer w ith a possible interaction betw een an environm ental carcinogen and a papillom a virus. N ature 2 7 4 :215-217. Johnson E S, Fischm an H R . M atanoski G M . D iam ond E (1986a): C ancer m ortality am ong white m ales in the m eat in d u stry . J O ccu p M ed 28:23 - 3 2 . Johnson E S . F isch m an H R , M atanoski G M , D iam ond E (1986b): O ccu rren ce o f can ce r in w o m en in the m eat industry. B r J Ind M ed 4 3 :597-604. K ettm an R , Burny A , K leuter Y (1978): D istribution o f bovine leukem ia virus proviral D N A sequences in tissues o f anim als w ith enzootic bovine leukosis. L eukem ia Res 2 :2 3 -3 2 . K hoklova M P. R akham in PP (1970): C om parative study on geographical distribution o f hum an and cattle leukosis. Bibl H aem atol 3 6 :6 5 4 -6 5 8 . K vam fors E, H enricson B , H ugoson G (1975): A statistical study on farm and village level on the possible relations betw een hum an leukem ia and bovine leucosis. A cta Vet Scand 16:163-169. Landrigan PJ (1981): A rsenic: State o f the an . Am J Ind M ed 2 :5 -1 4 . Law hom e L (1976): The health o f farm ers. J Iowa M ed Soc 6 6 :4 0 9 -4 1 8 . M atanow ski G M . L ilienfeld A (1976): " H o w V eterinarians D ie ." Presented at the 104th M eeting o f the A m erican Public H ealth A ssociation. M iam i Beach, FL. M iller JM . M iller L D , O lson C . G illette KG (1969): V irus-like panicles in phytohem aglutinin sim ulated cultures w ith reference to bovine lym phosarcom a. J N atl C ancer Inst 4 3 :1297-1305. O tfice o f Population C ensus and Surveys 11978): "O ccupational M o rality , 1970-1972. D ecennial Sup plem ent." Londom H M SO . Parrish C R . O liv er R E , W eddell W . and C orrin K C (1982): B ovine leukem ia virus infection in N ew Zealand cattle. N Z Vet J 3 0:56-58. Pearce N E . Sheppard R A , H ow ard JK . Fraser J, Lilley BM (1986): Leukem ia am ong New Zealand agricultural w orkers: A cancer registry based study. Am J E pidem iol 124:402-409. Pearce N E , Sheppard R A , Sm ith A H . T eague CA (1987): N on-H odgkin's lym phom a and farm ing: An expanded case-control study. Int J C ancer 3 9 :155-161. Pearce N E , Sm ith A H , H ow ard JK , Sheppard R A , G iles H J, T eague C A (1986): N on-H odgkin's lym phom a and exposure to phenoxyherbicides. chlorophenols. fencing w ork, and m eat w orks em ploym ent: A case-control study. Br J Ind M ed 4 3 :7 5 -8 3 . Pearce N E . H ow ard JK (1986): O ccupation, social class and m ale cancer m ortality in N ew Z ealand. 1 9 7 4 -7 8 . Int J Epidem iol 1 5 :456-462. Pearce N E , S m ith A H . R eif JS (1988): Increased risks o f soft tissue sarcom a, m alignant lym phom a, and acute m yeloid leukem ia in abattoir w orkers. A m J Ind M ed 1 4 :6 3 -7 2 . Pearce N E . Sm ith A H , Fisher DO (1985): M alignant lym phom a and m ultiple m yelom a linked w ith agricultural occupations in a N ew Z ealand can cer registry-based study. A m J Epidem iol 121: 225-237. Pershagen G (1983): T he epidem iology o f hum an arsenic exposure. In Fow ler BA (ed): "B io lo g ical and Environm ental Effects o f A rsen ic." A m sterdam :lsevier. Persson B, D ahlander A M , Fredriksson M , N oorlind B rage H, O hlson C G , A xelson O (1989): M alignant lynm phom a and occupational exposure: Br J Ind M ed 4 6:516-520. Pickle LW , M ason T J, H ow ard N , H oover R , Fraum eni JF (1987): "A tlas o f U S. C ancer M ortality am ong W hites: 1 9 5 0 -1 9 8 0 ." W ashington DC: US D epartm ent o f H ealth and Hum an Services. [D H H S publication N o. (N1H) 87-2900], R e if J, Pearce N , F raser J (1989): C ancer risks in N ew Z ealand m eat w orkers. S cand J W ork E nviron HIth 15:24-29. R e if J, Pearce N , F raser J (1989): C ancer risks in N ew Z ealan d farm ers. Int J E p id em io l, 18:768-- 774. Saftlas A F, B lair A , C antor K P. H anrahan L. A nderson H A (1987): C ancer and other causes o f death am ong W isconsin farm ers. Am J Ind M ed 11:119-129. S chw artz R S, C allen JP , S ilva J J r (1978): A cluster o f H o d g k in 's disease in a sm all com m unity: Evidence o f environm ental factors. Am J Epidem iol 108:19-26. Sm ith A H , Pearce N E, F isher D O , G iles H J. Teague C A , H ow ard JK (1984): Soft tissue sarcom a and & i. : x 2:2 148 Pearce and Reif exposure to phenoxyherbicides and chlorophenols in New Zealand. J Natl Cancer Inst 73:111I1117. Smith AH. Pearce NE, Callas PW (1988): Cancer case-control studies with other cancers as controls. Int J Epidemiol 17:298-306. Smith AH. Pearce NE (1986): Update on soft tissue sarcoma and phenoxyherbicides in New Zealand. Chemosphere 15:1795-1798. Smith KT. Campo MS (1985): The biology of papillomaviruses and their role in oncogenesis. Anticancer Res 5:31-48. Sterling TD. Weinkam JJ (1976): Smoking characteristics by type of employment. J Occup Med 18: 743-754. Vineis P, Terracini B. Ciccone G. Cignetti A. Colombo E. Donna A. Maffi L, Pisa R, Ricci P, Zanini E, Comba P (1987): Phenoxy herbicides and soft-tissue sarcomas in female rice weeders: A population-based case-referent study. Scand J Work Environ HIth 13:9-17. Walrath J. Rogot E. Murray J. Blair A (1985): " Mortality Patterns among US Veterans by Occupation and Smoking Status.'' (NIH) Publication No. 85-2756. Washington. DC: US Government Printing Office. Wiklund K. Holm L-E (1986a): Trends in cancer risk among Swedish agricultural workers. J Natl Cancer Inst 77:657-664. Wiklund K. Holm L-E ( 1986b): Soft tissue sarcoma risk in Swedish agricultural and forestry workers. JNCI 76:229-234. Wiklund K. Dich J. Holm L-E (1987): Risk of malignant lymphoma in Swedish pesticide appliers. Br J Cancer 56:505-508. Wiklund K. Lindefors B-M, Holm L-E (1988): Risk of malignant lymphomna in Swedish agricultural and forestry workers. Br J Ind Med 45:19-24. Williams GM (1984): Modulation of chemical carcinogenesis by xenobiotics. Fund Appi Toxicol 34: 310-320. Witter RL (1972): Epidemiology of Marek's disease: A review. In: Biggs PM. de The G. Payne LN (eds): "Oncogenesis and Herpesviruses." LyomlARC. Woodroffe AJ (1972): Multiple myeloma associated with long history of hyposensitisation with allergen vaccines. Lancet 1:99 (letter). Woods JS. Polissar L. Severson RK. Heuser LS. Kulander BG (1987): Soft tissue sarcoma and nonHodgkin's lymphoma in relation to phenoxyherbicide and chlorinated phenol exposure in western Washington. J Natl Cancer Inst 78:899-910. World Health Organization (1977): " Manual of the International Statistical Classification of Diseases, Injuries and Causes of Death." 9th revision. Geneva:WHO. 18185 fi JE n, y -3 1 3 . 800 J rinati, in th e unnica Soft Tissue Sarcom a Risk in Swedish u atln g among Licensed Pesticide Applicators M eet- le M em i prac- Kerstin Wiklund, Dr M ed Sc; Jan Dich, MSc; and Lars-Erik Holm, MD er et al ; Two Swedish studies have suggested th a t exposure to phenoxy acid herbicides m ay cause so it tissue sarcom a. T his asso ciation has been subsequently in vestig a ted in a num ber o f epidemiologic studies. R esu lts are inconsistent. In this stu d y the risk o f so ft tissue sarcom a in a cohort o f 20,245 Swedish pesticide applicators, o f whom 72% were es tim ated to have been exposed to phenoxy acid herbicides, was analyzed. The cohort was observed by m eans o f the Swedish Cancer R eg ister from date o f license u n til Dec 31, 1984, or until death i f it occurred p rio r to th a t date. The m ean follow up tim e was 13.9years. Seven p a tien ts w ith so ft tissue sarcom a were observed compared w ith 7.7 expected on th e basis o f the incidence in Sweden. The rela tive ris k was found to be 0.9 (95% confidence interval, 0.4 to 1.9). This finding is consistent w ith th e hypothesis th a t exposure o fphenoxy acid herbicides does n o t increase risk o f so ft tissue sarcom a. Two Swedish studies have reported a sixfold increase in the risk of soft tissue sarcom a following' exposure to phenoxy acid herbicides.14 This relationship has been investigated in a great number of subsequent studies.3-34 Reviews of these are somewhat contradictory.38-38 One review of delayed health hazards of pesticide exposure concluded that neither phenoxy herbicides not dioxins can be said un equivocally to cause cancer in hum ans.38 A review of health effects of phenoxy herbicides suggested th a t ex posure may be associated with an increased incidence of cancer.36 Phenoxy acid herbicides were introduced in Sweden in the late 1940s. Use of these herbicides increased greatly until the most recent years, when a slight decrease has been observed.37 From the Departments of Cancer Epidemiology and General Oncol ogy, Badlumhemmet, Karolinska University Hospital, Stockholm, Swe den. 009e-L738/8a/301(K>80lS0a.00 Copyright by American Occupational Medical Association The compound most widely used in Swedish agricul ture is MCPA (4-chloro-2-methylphenoxyacetic acid; CAS: 94-76-6). In the mid-1960s two other phenoxy acid herbicides were introduced: mecoprop (()-2-(4-chloroo-tolyloxy) propionic acid; CAS: 7085-19-0) and dichlorprop (()-2-(2,4-dichlorophenoxy) propionic acid; CAS: 120-36-5). These two compounds constituted approxi mately 25% of the total use of phenoxy acid herbicides in 1970 and about 40% in 1985. The compound 2,4-D (2,4-dichlorophenoxyacetic acid; CAS: 94-75-7) and 2,4,5-T (2,4,5-trichlorophenoxyacetic acid; CAS: 93-765) have also been used, although to a lesser extent. Phenoxy acid herbicides contain toxic contaminants such as dioxins and dibenzofurans. MCPA, 2,4-D, me coprop, and dichlorprop are not supposed to contain 2,3,7,8-TCDD (2,3,7,8-tetrachlorodibenzo-para-dioxin; CAS: 1746-01-6), which is the most toxic dioxin. How ever, 2,4,5-T was prohibited in Sweden in 1977 because of contamination with 2,3,7,8-TCDD. Most of the phenoxy acid herbicides sold in Sweden are used in agriculture or forestry. Since 1965 a license has been mandatory for using the most acutely toxic pesticides in agriculture.38-*0 Licensed pesticide appli cators handle phenoxy acid herbicides to a greater extent than do agricultural and forestry workers in general. In two surveys we have found th at 72% of licensed pesticide applicators compared with 16% of agricultural and forestry workers used phenoxy acid herbicides one day or more during the 1950s, 1960s, or 1970s. In the remainder of the Swedish population, the proportion of persons exposed to phenoxy acid herbi cides can be estim ated to be about 1%.41 Of those, very few have been exposed occupationally. Exposure from private use of such herbicides is very low. Swedish licensed pesticide applicators are thus an appropriate group of subjects for a study of the effects of phenoxy acid herbicides exposure. The object of the present study was to analyze the risk of soft tissue sarcoma. Journal of Occupational Medicine/Volume 30 No. 10/October 1988 19186 801 Methods Cohort Since 1965, completion of a four-day course on pesti cide application has been a prerequisite for obtaining1a license for using the most acutely toxic pesticides in agriculture. As of 1976, a total of 20,296 licenses had been issued. Thirty-three duplicates and 18 persons with incomplete identification were excluded, resulting in a cohort of 20,245 persons. Distribution of the cohort by year of birth is shown in Table 1. Almost 50% of the cohort were born in 1935 or later. Approximately 99% were men and 1% women. One third lived in the south ernmost part of Sweden, where the use of herbicides is highest in the country. Half of the licenses were issued in 1965 or 1966. Exposure TABLE 2 E s tim ate d E x p o s u re to P h enoxy A d d H erbicides A m on g th e C o h o rt of L ic e n s e d P e s tid d e A p p lic a to rs in A g riculture, by Y e a r o f B irth Y ear of Birth Percent Exposed 1905-1914 1915-1924 1925-1934 1935-1944 1945- 32 b 70 65 81 79 TABLE 3 Use of Protective Clothing Among Licensed Pesticide Applicators who Applied Phenoxy Add Herbicides in the 1950s, 1960s and 1970s Use of P rotective Clothing 1950s (n = 53) 1960s (n = 120) 1970s (n = 157) Never or seldom used any 42 23 9 protective clothing Used protective mask 19 44 58 Used protective glasses 26 31 51 Used protective gloves (as 49 75 89 only protection) 19 19 16 Used protective dress 23 33 41 A questionnaire was mailed to a random sample of 273 persons in the cohort to study the use of pesticides and protective clothing, tobacco use habits, and occu pations during the 1950s, 1960s, and 1970s. The re sponse ra te was 83%. The sample size was chosen so that the range of the 95% confidence interval for the percentage of phenoxy acid herbicide users would be 5%, if the sample showed th a t 75% of respondents had been users. The proportion of persons exposed to phen oxy acid herbicides one day or more was estimated to be 72% (66% to 78%). The proportions in the five yearof-birth groups are shown in Table 2. Herbicide use during the 1950s, 1960s, and 1970s was 19%, 49% and 67%, respectively. Forty-two percent of those who used phenoxy acid herbicides during the 1950s reported that they never or seldom wore any protective clothing, compared with 23% in the 1960s and 9% in the 1970s (Table 3). Gloves were the most commonly used protective cloth ing. The survey also showed th at most cohort members were employed in agriculture and/or forestry (70%) and horticulture (10%). Only a few worked full-time as pesticide applicators. The remaining 20% held a variety of occupations closely related to agriculture, but some worked in occupations such as building and construc tion. TABLE 1 D istribu tion b y Y e a r o f B irth o f th e C o h o rt o f L ic e n s e d P e sticid e A p p lic a to rs in A g ric u ltu re Year of Birth Percent (N = 20,245) -1904 1905-1914 1915-1924 1925-1934 1935-1944 1945-1954 1955Total 2.3 10.0 19.6 19.9 23.8 22.7 1.8 100 Follow-up The cohort was observed by means of the Swedish Cancer Register from date of license until Dec 31, 1984, or until death if it occurred prior to that date. The Cancer R egister was established in 1958. Reporting of all m alignant and some benign tumors has been com pulsory for almost all physicians and pathologists from the b e g i n n in g . A modified version of the 7th revision of the Classification of Diseases (ICD 7), has been used throughout to code anatomic sites.48 This study included all registered cases of malignant tum ors in connective tissue or muscle (ICD 7 = 197). The histologic diagnoses included were: angiosarcoma, leiomyosarcoma, rhabdomyosarcoma, myogenic sar coma, fibrosarcoma, malignant fibrous histiocytoma, myxosarcoma, liposarcoma, extraskeletal chondrosar coma, malignant synovialoma, undifferentiated and sar comatous tumor, spool-shaped sarcoma, Enzinger tu mors, giant cell sarcoma, and embryonal rhabdo myosarcoma. The cases were reviewed by a pathologist and all were confirmed. Statistical Analysis The expected number of cases (F ) was calculated from the age-specific incidence in 5-year classes for respective years in the entire Swedish population. The ratio of cases observed (O) to the number expected forms the standardized incidence ratio (SIR). Observed and expected numbers of cases were also calculated 5 and 10 years, respectively, after date of license to allow for " latency." The 95% confidence intervals were cal culated by means of a Poisson distribution table. Results The number of person-years of follow-up in the cohort was 281,550. When no "latency" time was allowed, the 802 Soft Tissue Sarcoma in Pesticide Applicators/Wiklund et ai { rtr* vho )s 57) sdish L984, The ig of comf Di- used nani .97). orna, sarDma, >sarsartubdo>gist ited for The cted rved k 5 Uow cal- ort the it al Li TABLE 4 Cases of Soft Tissue Sarcoma, SIR, and 95% Confidence Interval in the Cohort of Licensed Pesticide Applicators for Latency Periods of 0,5, and 10 Years Latency Time, yr Cases Observed Expected SIR 95% Confidence interval 0 7 7.7 0.9 0.4-1.9 5 6 6.0 1.0 0.4-2.2 10 4 3.9 1.0 0.3-2.7 TABLE 5 The 7 Cases of Soft Tissue Sarcoma Age at Year of Year of ICD Diagnosis Diagnosis License (7th) Histologic Diagnosis 63 1978 1966 197.0 Fibrosarcoma 61 1975 1965 197.3 Rhabdomyosarcoma 60 1983 1965 197.3 Malignant fibrous histiocytoma 52 1973 1965 197.3 Undifferentiated 41 1967 1965 197.1 Leiomyosarcoma 38 1979 1965 197.0 Rhabdomyosarcoma 28 1978 1969 197.3 Fibrosarcoma observed number of soft tissue sarcomas was 7, com pared with 7.7 expected (Table 4). The corresponding: SIR was 0.9 (0.4 to 1.9). The SIR for 5 years' latency was 1.0 (0.4 to 2.2) and for 10 years', 1.0 (0.3 to 2.7). The observed numbers of cases were 6 and 4, respec tively. Among those to whom a license was issued in 1965 or 1966, 6 cases were observed and 5.0 expected; the SIR was calculated to be 1.2 (0.4 to 2.6). The mean obser vation tim e was 16.1 person-years. Some data for the seven patients with soft tissue sarcoma are shbwn in Table 5. For all patients, diagnosis was based on histologic examination of biopsy material, and the diagnoses were confirmed in a review by a pathologist. Age a t diagnosis ranged from 28 to 63 years. Discussion In this cohort of licensed pesticide applicators in Sweden no excess risk of soft tissue sarcoma was found, although 72% (confidence interval, 66% to 78%) of the persons in the cohort were estimated to have been exposed to phenoxy acid herbicides one day or more. If the relative risk for exposed versus unexposed persons is tru ly 6,1,9 the SIR in this study should have been 4.5. Latency periods have been estim ated to be 15 and 20 years, respectively, in the two Swedish case-control studies.41 The mean follow-up time in this study was 13.9 years. However, time since first exposure is longer,because 19% were exposed one day or more during the 1950s. Regardless, the cohort may have been observed for too short a time to reveal any excess risk and it is therefore im portant to continue the follow-up. Estimation of the use of protective clothing is of importance since dermal absorption of pesticides is higher th an uptake from the oral cavity or the lungs.43 The absorption of phenoxy acid herbicides can be 30 to 40 times higher without protective clothing.44 However, in some c u e s wearing contam inated clothing may in crease the exposure.43 The use of protective clothing among pesticide Appli cators was not frequent, despite special education. One explanation could be th at it is more uncomfortable than the acute reaction to exposure. Pesticide applicators are recruited primarily from farmers, who have a decreased risk of cancer in gen eral.43 Farm ers are healthier and thus utilize health services less than other occupational groups do.47 One can therefore suspect th at some tumors have not been detected. However, in a study of a sample of cases not reported in the Cancer Register, the registration deficit was found to be 4%.48 In a follow-up study, agricultural workers were found not to have any increased registra tion deficit for all cancers. We have no data regarding a possible deficit for soft tissue sarcoma. Since the relative risk for soft tissue sarcoma in Swedish farmers was found to be 0.98 (0.84-1.15), one can assume no deficit or a very small one for this type of tumor. The initial Swedish case-control studies have been criticized for basing exposure data on recall of the patient or of relatives if the patient was deceased.49 The studies were made during the 1970s, when phenoxy acid herbicides and their possible health effects attracted much attention in Sweden, which could have influenced the data on exposure.37,49 One advantage of the present study is th at the expo sure status ("having a license for pesticide application") was established before the disease was detected, and therefore has not been influenced by media attention. Another advantage is the high proportion of exposed subjects in the cohort in comparison with the general population. A major disadvantage of the present cohort study is the lack of individual exposure data. We have informa tion for only a sample of the cohort and thus the esti mates of pesticide use and use of protective clothing refer to the whole cohort and not to individuals. In summary, the results of this study are consistent with the hypothesis that exposure to phenoxy acid her bicides does not increase risk of soft tissue sarcoma. Acknowledgments This study was supported by grants 83:91 and 85:66 from the Stockholm Cancer Society. References 1. Hardell L, Sandstrom A: Case-control study: Soft tissue sarco mas and exposure to phenoxyacetic acids or chiorophenois. B r J Cancer 1979;39:711-717. 2. Eriksson M, Hardell L, Berg NO, et al: Soft-tissue sarcomas and exposure to chemical substances. A case-referent study. B r J Ind Med 1981;38:27-33. 3. Cook RR: Dioxin, chloracne, and soft tissue sarcoma. Lancet Journal of Occupational Medicine/Volume 30 No. 10/October 1988 13108803 1981;1:618-819. 26. Barthel E: Cancer risk in pesticide exposed agricultural work 4. Honchar PA, Halperin WE: 2,4,5-T, triehlorophenol and soft- ers. Arch Oeschwulstforsch 1981;5l/7:579--685 (In German). tissue sarcoma. Lancet 1981;i:268--269. 27. Balarajan R, Acheson ED: Soft tissue sarcomas in agriculture 6. Moses M. Selikoff IJ: Soft tissue sarcomas, phenoxy herbicides, and forestry workers. J Epidemiol Community Health 1984:38:118. and chlorinated phenols. Lancet 1981:1:1370. 116. 6. Johnson FE, S u tle r MA, Brown SM: Soft tissue sarcomas and 28. Kilpatrick B: Mortality from malignant disease of connective chlorinated phenols. Lancet 1981:11:40. and other soft tissue in agricultural workers in England and Wales. 7. Ott MO, Holder BB, Olson RD: A mortality analysis of employ Symposium abstract. Dioxin 86, Fukuoka, Japan, Sept 16-19, 1986. ees engaged in the manufacture of 2,4,5-trlchlorophenoxyacetio acid. 29. Greenwald P, Kovasznay B, Collins DN, et ah Sarcomas of soft J Ocoup lie d 1980;22:47-60. tissues after Vietnam service. JNCI 1984;73:1107-1109. 3. Fingerhut MA. Halperin WE, Honchar PA, et ah An evaluation 30. Coggon D, Pannett B, Winter PD. et al: Mortality of workers of reports of dioxin exposure and soft tissue sarcoma pathology among exposed to 2-methyl-4-chlorophenoxyacetic add. Scand J Work Envi chemical workers in the United States. Scand J Work Environ H ealth ron H ealth 1986;12:448-454. 1984:10:299-303. 31. Lawrence CE. Reilly AA. Quickenton P, et al: Mortality pattern 9. Lynge E: A follow-up study of cancer incidence among workers ofNew York state Vietnam veterans. Am J Public Health 1985;75:277- in manufacture of phenoxy herbicides in Denmark. B r J Cancer 279. 1986:62:269-270. 32. Wlklund K, Holnr L-E: Soft tissue sarcoma risk in Swedish 10. Sarma PB, Jacobs J: Thoracic soft tissue sarcoma in Vietnam agricultural and forestry workers. JNCI 1986;76:229-234. veterans exposed to Agent Orange. I f Engl J Med 1982:306:1109. 33. Hogstedt C, Westerlund B: Kohortstudie av dodsorsaker for 11. Puntoni B, Merlo F, Fini A, et al: Soft tissue sarcoma In Seveso. skogsarbetare med och utan exposition for fenoxisyrapreparat. Lak- Lancet 1986;ii:525. artldningen 1980;77:1828-1831. 12. Kogan MD, Clapp BW: Mortality among Vietnam veterans in 34. Axelson 0, Sundell L, Andersson K, et al: Herbicide exposure Massachusetts, 1972-1983. Boston, Massachusetts Department of and tumor mortality. An updated epidemiologic investigation on Swed Public Health, Division of Health Statistics and Research, 1986. ish railroad workers. Scand J Work Environ H ealth 1980;6:73-79. 13. Anderson HA, Hanrahan LP, Jensen M, at ah Wisconsin Viet 35. Sharp DS, Eskenazi B, Harrison R, et ah Delayed health nam veterans mortality study: Proportionate mortality ratio results. hazards of pesticide exposure. Annu Rev Publia Health 1986:7:441- Madison, WI, State of Wisconsin, Department of Health and Social 471. Services, 1986. 36. Sterling TD, Arundel PV: Health effects of phenoxy herbicides. 14. More AA, Heydinger DK: Vietnam-era veterans mortality Scand J Work Environ Health 1986;12:161-173. study: West Virginia residents 1968-1983. Charleston, WV, West 37. Backstrom J: The phenoxy acid problem in Sweden. Ecol Bull Virginia Department of Health, Vietnam-era Veterans mortality study 1978:27:108--121. committee, 1986. 38. SFS 1985: 426 Law on Chemical Products. Liber forlag, Stock 15. Smith AH. Pearce NE, Fisher DO, et al: Soft tissue sarcoma holm, 1986. and exposure to phenoxy herbicides and chlorophenols in New Zealand. 39. SFS 1985: 835 Ordinance on Chemical Products. Liber fSrlag, JNCI 1984;73:1111-1117. Stockholm, 1986. 16. Kang HK. Weatherbee L, Breslin PP, et al: Soft tissue sarcomas 40. SFS 1985:836 Forordningen om bekampningsmedeL Ordinance and military service in Vietnam: A case comparison group analysis of an Pesticides). Liber forlag, Stockholm, 1986 (in Swedish). hospital patients. J Occup Med 1986:28:1215--1318. 41. Hardell L: Epidemiological studies on soft tissue sarcoma and 17. Woods J, Pollssar L, Severson B, et al: Soft tissue sarcoma and malignant lymphoma and their relation to phenoxy acid or chloro- non-Hodgkin's lymphoma in relation to phenoxy herbicides and chlo- phenol exposure, ffmea University medical dissertations. New series, rophenol exposure in Western Washington State, USA. Symposium no. 65,1981. abstract. Dioxin 86, Fukuoka, Japan, Sept 16-19, 1986. 42. World Health Organization: International classification of dis 18. Hoar SK, Blair A, Holmes FF, et ah Agricultural herbicide use eases, injuries, and causes of death, 1955 revision. Geneva, 1957. and risk of lymphoma and soft-tissue sarcoma. JAMA 1986:256:1141- 43. Kolmodin-Hedman B, Hoglund S, Swensson A, et ah Studies on 1147. phenoxy acid herbicides. H. Oral and dermal uptake and elimination 19. Riihimaki V, Asp S, Pukkala E, et ah Mortality and cancer in urine of MCPA in humans. Arch Toxicol 1983;54:267-273. morbidity among chlorinated phenoxyacid applicators in Finland. 44. Nordiska Expertgruppen for gransvardesdokumentatlon; Kol Chemoaphere 1983;12:779-784. modin-Hedman B: No. 23 MCPA (4-klor-2-methylfenoxisyra). Arbeta 20. Milham S: Herbicides, occupation, and cancer. Lancet ash Halsa 1981:14 (in Swedish). 1982;hl464-1465. 45. Wester RC, Maibach HI: In vivo percutaneous absorption and 21. Blondell JM: Cancer mortality and pesticide use among rural decontamination of pesticides in humans. J Toxicol Environ Health farm counties in the United States, 1950-1969. Am J Epidemiol 1986;16:25-37. 1979:110:356. 46. Wiklund K, Holm L-E: Trends in cancer risks among Swedish 22. Suskind B, Hertzberg VS: Human health effects of 2,4,5-T and agricultural workers. JNCI 1986;77:657-684. its toxic contaminants. JAM A 1984^51:2372-2380. 47. Den jamlika sjukvarden. Hals och sjukvard infor 90-talet. 23. Gallagher BP, ThreUall WJ: Cancer and occupational exposure Stockholm: Allmanna forlaget, 1984. (SOU 1984:41). to chlorophenols. Lancet 1984;ii:48. 48. Mattsson N, Wallgren A: Completeness of the Swedish Cancer 24. Wang HH. MacMahon B: Mortality of pesticide applicators. J Register. Non-notified cases recorded on death certificates in 1978. Occup Med 1979:21:741--744. Acta Radiol Oncol 1984;23:305-313. 25. Blair A, Grauman DJ, Lubin JH, et al: Lung cancer and other 49. Colton T: Herbicide exposure and cancer. JAMA causes of death among licensed applicators. JNCI 1983;71:31-37. 1986;258:1178-1178. 804 Soft Tissue Sarcoma in Pesticide Applicators/Wiklund et ai 19169 The associations between soft tissue sarcoma (STS) and occupational exposures were studied in a case-referent study in the southeast of Sweden. Exposure information was obtained through mailed questionnaires to 96 cases, 450 randomly selected population referents, and 200 cancer referents. Odds ratios (OR), were calculated for various occupational groups, and particularly, for occupations with ( potential exposure to chlorinated phenoxy herbicides and chlorophenols. In the analyses based on population referents, increased risks for soft tissue sarcoma were seen for especially gardeners (OR = 4.1), but also railroad workers (OR = 3.1); construction workers with exposure to impregnating agents (OR = 2.3), asbestos (OR = 1.8), or pressure impregnating agents (OR = 1.7); and unspecified chemical workers with potential exposure to phenoxy herbicides and/or chlorophenols (OR = 1.6). A sim ilar pattern appeared when cancer referents were used although the numerical values of the odds ratios became different. A grouping of jobs resulted in Mantel-Haensel OR from 1.5 to 1.9 for farmers and forestry workers, dependent on referents used and even more increased OR for railroad workers and unspecified chemical workers with potential exposure to phenoxy herbicides and chlorophenols. The results of the study confirm rather than refute that phenoxy herbicides and chlorophenols could be of tiologie importance for STS; the high risk for gardeners, although based on a small number of individuals, was unexpected and remains unclear. Also, since other cancers were used as referents, no definite problems of recall bias should obtain in this material. None of the exposed groups had a higher proportion of smokers than the unexposed group. C a n c e r 66:806-811,1990. ara t u o n m tr r p i m nam i S OMEWHAT CONFLICTING RESULTS about a possible association of cancer and occupational exposure to pesticides have been reported in a number of studies from different countries. An early observation of increased cancer risks in connection with herbicide exposure ap peared in the early 1970s in a study of Swedish railroad workers engaged in brush and weed killing.1,2Two Swed ish case-referent studies on soft tissue sarcoma (STS) in the late 1970s and early 1980s3,4showed odds ratios (OR) of 5.3 and 6.8 for exposure to chlorinated phenoxy her- F ro m th e D e p a rtm e n ts o f ' O c c u p a tio n a l M e d icin e a n d fOncoiogy, U niversity H ospital, Linkping, Sweden. Supported by grants from the Local C ancer Fund in the county o f O stergotland. Address for reprints: G . W ingren, BSc, D epartm ent o f O ccupational M edicine, U niversity H ospital, S-58I 85 Linkping, Sw eden. A ccepted for publication February 2, 1990. bicides and OR of 6.6 and 3.3 for exposure to chloro phenols, respectively. A recent Swedish case-referent study has again sug gested an association of STS with exposure to phenoxy herbicides with a roughly three-fold increase in risk but the earlier seen association with exposure to chlorophenols was not reproduced.5 It is worth noting, however, that record linkage in the so-called Cancer-Environmental Registry has failed to find any association between STS and agricultural or forestry work.6Another Swedish study on licensed pesticide applicators has also failed to find any association.7 In a study from Kansas, no association between agri cultural herbicides and STS was found8 whereas an as sociation appeared in the same material for STS and in secticides used in farming.9 The current case-referent study was undertaken to obtain further elucidation of the 19170 806 e to chloro- : again sugto phenoxy in risk but lorophenols wever, that ironmental tween STS edish study Lied to find tween agnreas an as- .nd inSfc-referent ition of the No. 4 So ft T issue Sarc om a a n d Occupation W ingren et al. 0UT possible associations between STS and different occupa and smoking. Ninety percent confidence intervals (Cl) tional exposures, especially those likely to involve use of were calculated according to Cornfield's method as de phenoxy herbicides or chlorophenols. scribed by Breslow and Day.11A calculated OR was con sidered statistically significant if the lower limit of its as Materials and Methods sociated Cl exceeded unity. The statistical analyses were performed with the Epilog statistical package for personal From the Regional Cancer Register at the University computers (Epicenter Software, Epilog, version 3, Pasa Hospital in Linkping, Sweden, 96 cases of STS with ICD dena, CA, 1986). 7 codes 174, 196 (except osteosarcomas), and 197 were identified during the period 1975 through 1982. All were men, 25 to 80 years old at time of diagnosis, from the Results region delivering cases to the hospital, i.e., the counties of stergtland, Jnkping, and Kalmar. Sixty-four are alive and 32 are dead. Two different sets of referents were drawn, one consisting of 200 alive and 250 dead individ uals, randomly selected in 1982 from the population in these counties and in the same ages as the cases. The sec ond referent group consisted of 200 individuals from the same three counties who had a cancer diagnosis other than sarcoma during the study period. Of these, 111 were ! alive and 89 were dead, all with ages 25 to 80 years at time of diagnosis. Information regarding exposure was collected through a contracted organization working with epidemiologic materials in various respects. Each person when alive, or the next-of-kin of dead individuals, received an introduc tory letter and a questionnaire with 16 main questions and subspecifications about occupations in general and some specific jobs in particular. Questions were asked about occupational exposure to pesticides, wood preser vatives, solvents, and other chemicals, as well as various leisure time exposures and smoking habits. If the questionnaire had not been properly completed or when a person indicated work as a farmer or forestry worker, additional telephone interviews were made ) (blindly with regard to case-referent status) by the con tracted organization. After receiving the returned ques tionnaires, the first page with information about casereferent status, name, address, and personal identification number was removed and not attached again until the final processing of the material for the analyses. Detailed questions about pesticide exposure were asked, but very few subjects were able to provide specific infor After losses due to refusals, failure to identify a nextof-kin, questionnaires which were not returned, and eth ical restrictions regarding recently dead individuals, the final study material consisted of 550 individuals, Table 1. Odds ratio analyses were performed for all occupational groups that could be defined from the questionnaires. Sig nificantly increased risks were seen for gardeners and construction workers exposed to asbestos (Tables 2 and 3). Somewhat elevated risks also appeared for other oc cupational categories, although none of them significant (Tables 2 and 3). Further analyses were conducted on specificjobs (called "risk jobs" in the tables) where exposure to phenoxy her bicides and/or chlorophenols were likely to have occurred, namely farmers, forestry workers, gardeners, railroad workers, road workers, unspecified chemical workers (with or without stated exposure to phenoxy herbicides and/or chlorophenols (PHC)/non-PHC), sawmill workers, con struction workers, and carpenters. Farmers and forestry workers were grouped together since almost all forestry workers in southern Sweden also have been working with farming. Another grouping of all potentially chlorophenol ex posed workers, Le., sawmill workers, construction workers and carpenters, was also made. In these further analyses, eight cases of lymphomas (ICD 7 codes 200-202) were removed from the cancer referent group because of the many reported findings of an association between lym phomas and exposure to phenoxy herbicides and chlo rophenols.12In total, 331 individuals were exposed to one or more of these particularjobs. The unexposed category mation about the names of the pesticides to which they had been exposed. The analyses were, therefore, limited to evaluatingjob associated risks only; analogous to using job-exposure matrices, certain work tasks were considered Table 1. Distribution of Alive and Dead Individuals Among Cases and Referents to have involved exposure in the various respects ac Alive Deceased Total counted for in the "Results" section. In addition to calculations of crude OR, the statistical analyses of the data were based on the Mantel-Haensel Cases Cancer referents Population referents 32 58 159 39 106 156 71 164 315 procedure for the calculation of adjusted OR (MH-OR).10 In some of the analyses, adjustments were made for age 550 19171 808 Cancer A u g u st 15 1990 v0|. ^ T a b l e 2. E levated R isks for S om e Specific Jo b s a n d S m o k in g W ith O d d s R atio > 1.2 an d T w o o r M ore C ases E xposed* G ardeners R ailroad work C onstruction work, asbestos exposure Unspecified chem ical w ork, potential exposure to phenoxy herbicides or chlorophenols F arm er w ork, exposure to fertilizers C onstruction work, other insulation exposure C onstruction work, exposure to im pregnating agents C o n stru ctio n work, exposure to pressure im pregnating agents Saw m ill w ork, sawing Sm okers OR Infinite 1.9 5.9 1.8 1.3 3.3 4.8 3.5 1.5 1.3 90% CI 2.8 0 .3 -9 .1 1.3-36 0 .8 -4 .0 0 .6 -2 .7 0 .9 -1 2 0 .7 -5 3 0 .4 -4 3 0 .4 -5 .6 0 .3 -4 .5 No. o f cases exposed 6 4 7 } 14 16 8 2 3 5 5 No. o f referents exposed 0 5 3 20 30 6 4 2 8 9 CI: confidence interval; O R: odds ratio. * O th er cancer cases were taken as referents. was taken as individuals in jobs, where exposures to phenoxy herbicides or chlorophenols were unlikely. In Table 4, the comparisons involved cancer referents and in Table 5 population referents. In these two tables, the exposed individuals could have had a combination of jobs with potential exposure to phenoxy herbicides and/ or chlorophenols. In Tables 6 and 7 the same analyses are presented but here the individuals were exposed only to one such job, i.e., no combinations of jobs were ac cepted, which reduced the number of individuals eligible in this context. Age and smoking were controlled for in the analyses. A latency time requirement of 5 years was applied in these analyses, i.e.. an occupational exposure less than 5 years before diagnosis was not taken into account. Separations of alive and dead individuals did not change the results in any specific direction, and these analyses are therefore omitted in this report. Discussion The results from this study, although based on small numbers, corroborate some of the findings from other epidemiologic studies regarding an increased risk of STS among farmers and forestry workers. However, other oc cupations such as gardeners, railroad workers, unspecified chemical workers, and workers in contact with wood, had an increased risk of STS. These occupations may imply exposure to either phenoxy herbicides or chlorophenols, or both, but obviously other agents also. It is somewhat surprising, however, that the occupa tional group with the highest risk for STS was gardeners. Table 3. E le v ated R isk s fo r S o m e Specific J o b s a n d S m o k in g W ith O d d s R a tio > 1.2 a n d T w o or M o r e C a ses E xposed* G ardeners R ailroad work C onstruction work, asbestos exposure U nspecified chem ical w ork, potential exposure to phenoxy herbicides or chlorophenols F arm er w ork, exposure to fertilizers C onstruction work, other insulation exposure C onstruction work, exposure to im pregnating agents C onstruction work, exposure to pressure im pregnating agents Saw m ill w ork, sawing Sm okers O R 90% CI 4.1 1 .0 -1 4 3.1 0 .6 - 1 4 1.8 0.6--4.8 1.6 0 .8 -3 .3 01.3 .6 -2 .6 1.3 0 .5 - 3 .2 2.3 0 .5 -8 .9 1.7 0 .3 -7 .3 1.5 0 .4 - 4 .5 1.5 0 .4 - 4 .6 N o. o f cases exposed 6 4 7 14 16 8 4 3 5 5 N o. o f referents exposed 7 6 18 41 57 28 8 8 15 15 O R; odds ratio; CI: confidence interval. A population sam ple was taken as referents. Table 4. E x p o su re to S o m e R isk J o b s A sso ciated W ith th e U se o f P h e n o x y a c e tic A cids No- of re f * * exposed 0 5 3 20 30 6 4 2 i 9 are therefore A ie (>r) a54 55-69 7('-80 ; Nonsm okers Sm okers N onsm okers Sm okers Nonsm okers Sm okers OR M -H OR 90% Cl Cases Cancer Ref Cases Cancer Ref Cases Cancer Ref Cases Cancer Ref Cases Cancer R ef Cases Cancer Ref Cases Cancer Ref Unexposec 5 4 7 16 1 5 6 22 2 4 1 15 22 66 1.0 1.0 -- Farm ers, forestry w o rk ers G ardeners 3 3 1 4 4 7 11 13 2 11 7 14 28 52 1.6 1.9 (0 .9 --4.3) 0 0 0 0 2 2 0 2 0 0 0 6 0 -- -- -- R ailroad w o rk ers 0 0 1 2 0 0 3 1 0' I 0 0 4 4 3.0 2.6 (0 .5 -1 6 ) Road w orkers 0 0 1 1 0 0 0 0 0 1 0 3 1 5 0.6 0.9 (0 .0 -1 2 ) PHC U nspecified chem ical w orkers (non-PH C) 12 0 02 04 01 01 l2 06 o 0 01 03 06 2 10 0 18 -- 1.7 -- 2.1 -- (0.7-6.1) Sawm ill w orkers, constriction workers, and carpenters i 0 1 3 3 3 8 8 2 8 6 16 21 38 1.7 2.6 ( 1 .0 -6 .6 ) O R: odds ratio: Cl: confidence interval: M -H: M antel-H aensel; PHC: unspecified chem ical w orkers exposed to phenoxy herbicides a n d /o r chlorophenols: Ref: referent. * U nexposed are individuals never exposed to any o f the risk jobs. C o m b in ed jo b s. i.e.. ex p o su re to m o re th an o ne o f th e risk jo b s can occur. C ancer cases other than sarcom a and lym phom a are taken as referents. ;ed on small ^ ">m oth er xofST S er, other oc. unspecified h w ood, had m ay im ply o r o p h e n o ls, the occupas gardeners* i* .`fe re n ts sed 1 ) * This association has, to our know ledge, not been reported in an y other stud y. T h e ch em ica l exp osu re situ ation a m o n g gardeners is u n clear. O n ly tw o o f a to ta l o f 13 gardeners stated that they had w orked w ith pesticide spraying, and on ly on e o f th ese (a case) cou ld tell w hat kind o f pesticide he had used (T oxid ol, a b rom op h os preparation). H igh er occu p ation al risks appeared, especially for those in dividu als w ith exp osu re to m ore than on e o f the job s studied. T h e n um ber o f cases and referents w ere very m uch reduced w hen on ly on e specific and nonoverlapping job w as studied, indicating that seasonal w ork or m ove m en t betw een the studied job s was com m on . T h e use o f a cancer referent group should have m ini- Table 5. E x p o su re to S o m e R isk J o b s A sso ciated W ith th e U se of P h e n o x y a c e tic A cid s a n d C h lo ro p h e n o ls* Unspecified Sawm ill w orkers, Farm ers. chem ical construction Age forestry R ailroad Road w orkers workers, and (yr) U nexposed w orkers G ardeners w orkers workers PHC (non-PH C) carpenters s54 55-69 70-80 Nonsm okers Sm okers Nonsm okers Sm okers N onsm okers Sm okers OR M -H OR 90% Cl Cases Population R ef Cases Population R ef Cases Population R ef Cases Population R ef Cases Population R ef Cases Population R ef Cases Population R ef 5 20 7 41 1 5 6 27 2 10 1 24 22 127 1.0 1.0 -- 3 8 1 16 4 17 11 25 2 18 7 29 28 113 1.4 1.5 (0 .7 -3 .0 ) 0 0 0 2 2 0 2 2 2 2 0 1 6 7 4.9 4.4 (1 .0 -1 8 ) 0 0 1 2 0 0 3 1 0 1 0 1 4 5 4.6 4.8 (0 .8 -2 4 ) 0 0 1 2 0 2 0 2 0 1 0 4 1 11 0.5 0.7 (0.0-6.2) 1 0 0 0 0 0 1 1 0 1 0 1 2 3 3.8 4.2 (0.4-41) 2 6 2 13 1 2 2 5 0 2 3 7 10 35 1.6 1.7 (0 .7 --4 .3 ) 1 7 1 18 3 13 8 20 2 7 6 17 21 82 1.5 1.4 (0 .7 -3 .0 ) O R : odds ratio; Cl: confidence interval; M -H : M antel-H aensel; PHC: unspecified chem ical workers exposed to phenoxy herbicides an d /o r chlorophenols; Ref: referent. * U nexposed are individuals never exposed to any o f the risk jobs. C o m b in ed jo b s, i.e., exposure to m o re th an o n e o f th e risk jo b s can occur. A p o p u latio n based sam ple is taken as referents. 19173 810 C a n c e r A ugust 15 1990 Vol.66 Table 6. E x p o su re to S o m e R isk J o b s A ssociated W ith th e U se of P h en o x y acetic A cids a n d C h lo ro p h e n o ls' Age (yr) 54 55-69 70-80 N onsm okers Sm okers N onsm okers Sm okers N onsm okers Sm okers OR M -H OR 90% Cl U nexposed Cases Cancer Ref Cases Cancer Ref Cases Cancer Ref Cases Cancer Ref Cases Cancer Ref Cases Cancer Ref Cases Cancer Ref 5 4 7 16 1 5 6 22 2 4 1 15 22 66 1 .0 1.0 -- Farm ers, forestry w orkers 1 3 1 2 2 5 5 10 1 5 4 3 14 28 1.5 1.6 (0 .6 --4.4) G ardeners 0 0 0 0 1 0 I 0 1 0 0 0 3 0 -- -- -- R a ilro a d w orkers 0 0 1 0 0 0 0 0 0 0 0 0 1 0 -- -- -- Road w orkers 0 0 0 0 0 0 0 0 0 1 0 1 0 2 -- ___ -- PHC 0 0 0 0 0 0 1 0 0 0 0 0 1 0 -- -- -- U nspecified chem ical w o rk ers (non-PH C ) 2 0 2 3 0 1 1 6 0 l 2 4 7 15 1.4 1.6 (0 .4 -5 .4 ) Saw m ill'w orkere, construction workers, and carpenters 0 0 0 1 0 1 2 6 0 2 2 6 4 16 0.7 1.1 (0 .2 --4.7) O R : odds ratio; C l; confidence interval; M -H: M antel-H aensel; PHC: unspecified chem ical w orkers exposed to phenoxy herbicides an d /o r chlorophenols; Ref: referent. * U nexposed are individuals never exposed to an y o f the risk jobs. N o co m b in e d jo b s are accepted, as reducing th e m aterial. C an cer diag noses oth er th an sarcom a and lym phom a are taken as referents. m ized the recall bias problem that often has been discussed w hen case-referen t stud ies focu s on m alignant diseases. H en ce, a recall bias, if existin g, is likely to be sim ilar for the cases and for the can cer referents. Furtherm ore, an association betw een the exposures and som e other cancers than sarcom a (an d ly m p h o m a ) w ou ld result in a decreased risk. Instead, the O R in this stud y in creased w h en using cancer referents. T h is m ay indicate that th e population w orking in the particular jo b s w ith p oten tial exposure, i.e., m o s tly o u td o o r a n d p h y s ic a lly a c t iv e w o r k , is h ea lth y w ith regard to oth er cancers. T h e S T S risk for railroad w ork ers w as h igh er w h en p o p u la tio n referen ts w ere in- Table 7. E x p o su re to S o m e R isk J o b s A sso ciated W ith th e U se o f P h e n o x y a c e tic A cids a n d C h lo ro p h e n o ls* Saw m ill workers, Farm ers, U n sp ecified construction Age forestry R ailroad Road chem ical workers, and (yr) Unexposed workers G ardeners workers workers PH C w orkers carpenters 54 55-69 70-80 N onsm okers Sm okers N onsm okers Sm okers N onsm okers Sm okers OR M -H O R 90% Cl Cases Population R ef Cases Population R ef Cases Population Ref Cases Population R ef Cases Population Ref Cases Population R ef Cases Population R ef 5 20 7 41 1 5 6 27 2 10 1 24 22 127 1 .0 1.0 -- 1 5 I 7 2 11 5 12 1 12 4 16 14 63 1.3 1.4 (0.6-3.4) 0 0 0 0 1 0 1 0 l 0 0 0 3 0 -- -- -- 0 00 2 0 00 6 1 00 2 2 00 7 0 00 0 0 00 2 0 01 1 1 01 2 0 00 0 0 01 0 0 00 2 0 01 3 1 01 7 3 03 20 1.9 -- 1.9 2.0 1.8 -- 2.1 2.0 (0 .1 -2 3 ) -- (0 .1 -3 3 ) (0 .7 --6 .7 ) 0 4 0 10 0 5 2 10 0 2 2 4 4 35 0.6 0.7 (0 .1 -2 .3 ) OR: odds ratio; C l: confidence interval: M -H: M antel-H aensel: PH C: unspecified chem ical w orkers exposed to phenoxy herbicides an d /o r chlorophenols; Ref: refe ren t * U nexposed are individuals never exposed to an y o f the risk jobs, N o com b in ed jobs are accepted, as reducing th e m aterial. A population sam ple is taken as referents. f v<JL66 >ls* < -rifled lical <ers ?HC) Sawmill worker^ construction workers, and carpenteis i I ; t 5 .4) 0 0 0 1 ! 2 6. 0 2 2 6 4 16 0.7 1.1 (0.2-4.7) 1 to any of the risk jobs. ie material. Cancer diagtaken as referents. crea sed w hen using :hat the population ntial exposure, :ive w ork, is healthy TS risk for railroad . referen ts w ere in- nols* Sawmill workers, ied . construction al workers, and s carpenters 0 4 0 10 0 5 2 10 0 2 2 4 4 35 0.6 0.7 ) (0.1-2.3) i a**" of the riskjobs, n. !. A population No. 4 So f t T is s u e S a r c o m a a n d O c c u p a t i o n W ingren et al. 8 volved rather than can cer referents, in dicating the pos sibility o f a h id d en association betw een railroad w ork and st m e c a n c e r fo r m s r e p r e s e n te d a m o n g th e c a n c e r refer ents. A lso, the fact that n o system atic differences w ere found between answ ers from alive and the next-of-kin o f dead individuals in dicates that there sh ould be n o im portant recall b ias in th is m aterial. It m ay be n oted in th is co n tex t that in a recen t valid ation study from U S , n o difference was fou n d b etw een cases and referents w ith regard to recall o f p reviou s w ork areas, th at is, th e risk o f recall bias is unlikely to be o f any im portance in case-referen t studies fc se d o n o c c u p a t io n a l t it le s .13 T h e p r o p o r tio n o f s m o k e r s did n ot differ m u ch b etw een the u n exp osed and the ex posed groups excep t for farm ers, forestry workers, and gardeners, w ho had a som ew hat low er proportion o f sm okers. H ence, w ith regard to these occupations, there m ight have b een so m e u ncontrolled con fou n d in g from s m o k in g a ffe c tin g th e re fe r e n t s id e 14,15 w h e n c a n c e r ref erents w ere used. E ven if the size o f this stud y is rather sm all as a result c f the relative rareness o f ST S, and d esp ite the lack o f specific exp osu re in form ation , the overall results seem to confirm than refute earlier observations suggesting a re lation o f ST S and, especially, exposure to p h en oxy her bicides. H ow ever, this m aterial sh ow s a low er degree o f risk in th e co m b in e d grou p o f o ccu p a tio n s w ith p oten tial exposure for chlorophenols, whereas construction workers w ith sp ecified exp osu re for im pregn ating and insulating agents had relatively higher risks for ST S. T h e strong effect teen for gardeners is difficult to exp lain a n d has to be considered further in n ew studies. REFERENCES 1. A xelso n O . S u n d e ll L. H e rb ic id e e x p o s u re , m o rta lity a n d turn incidence: A n epidem iological investigation on Swedish railroad work Work Environ Health 1974; 11:2 1 -2 8 . 2. Axelson O . Sundell L. A ndersson K , Edling C, H ogstedt C, Ki H. Herbicide exposure an d tum o u r m ortality: A n updated epidem iolo investigation o n Sw edish railroad w orkers. ScandJ Work Environ Hei. 1980; 6:73-79. 3. H ardell L S andstrom A. C ase-co n tro l study: Soft-tissue sarcor. a n d ex p o su re to p h e n o x y a cetic a c id s o r c h lo ro p h en o ls. Br J Car. 1979;39:711-717. 4. Eriksson M , H ardell L. Berg N O , M oller T , A xelson O. Soft-tis sarcom as an d exposure to chem ical substances: A case-referent stu B rJlnd Med 1981; 3 8 :2 7 -3 3 . 5. H ardell L, Eriksson M . T he association betw een soft tissue sarco r a n d ex p o su re to p h e n o x y a c e tic acids: A new case re fe re n t stu d y . Can 1988; 62:652-656. 6. W ik lu n d K , H o lm L E . S o ft tissu e s a rc o m a risk in S w edish a:: c u ltu ra l a n d fo restry w o rk ers. J Natl Cancer Inst 1986; 7 6 :2 2 9 -2 3 4 . 7. W iklund K, D ick J, H olm LE. Soft tissue sarcom a risk in Swed licen sed p esticid e a p p lic a to rs. J Occnp Med 1988; 3 0 :8 0 1 -8 0 4 . 8. H o a r S K , B lair A, H o lm e s F F et al. A g ric u ltu ra l h erb icid e use a risk o f ly m p h o m a a n d so ft-tissu e sa rc o m a . JAMA 1 9 8 6 :2 5 6 :1 1 4 1 -1 1 - 9. H o a r Z a h m S, B la ir A , H o lm e s F F , B oysen C D , R o b e l R J. A ca> c o n tro l s tu d y o f so ft-tissu e sa rc o m a s a n d H o d g k in 's d isease, fa rm in g a in secticid e use. Scand J Work Environ Health 1988; 14:224-230. 10. M antel N , H aen szel W . S tatistical asp ect o f analysis o f d a ta frc retro sp ectiv e stu d ies o f disease. J Natl Cancer Inst 1959; 2 3 :7 1 9 -7 4 8 11. Breslow N E . D ay N E . T h e an aly sis o f case-co n tro l studies. S ta tistic a l M e th o d s in C a n c e r R e se a rc h , vol. 1, n o . 32. L yon: 1A1 Scientific Publications, 1980; 133-137. 12. A xelson O , Flodin U , H ardell L A c o m m e n t on the referen series w ith regard to m ultiple exposure evaluations in a case-refere stu d y . Scand J Work Environ Health 1982; (Suppl) 8 :1 5 -1 9 . 13. B ond G G , B odner K M , Sobel W , S hellenberger R J, Flores G V a lid a tio n o f w ork h isto rie s o b ta in e d fro m in terv iew s. Am J Epidem 1988;128:343-351. 14. A xelson O . E lu c id a tio n o f s o m e ep id e m io lo g ic principles. Sea. J Work Environ Health 1983; 9 :2 3 1 -2 4 0 . 15. P earce N , C heckow ay H . C a se -c o n tro l studies using o th er diseas a s c o n tro ls: P ro b le m s o f e x c lu d in g e x p o s u re re la te d d iseases. Am J E idemiol 1988; 1 2 7 :8 5 1 -8 5 6 . . ; :{ ? ><: | {*) n :* `P i 'i : *13 i f*5 ; 7 :y ir i The Evaluation of Trends in Soft Tissue Sarcoma According to Diagnostic Criteria and Consumption of Phenoxy Herbicides ELSEBETH LYNGE, HANS HENRIK STORM, AND OLE MOLLER JENSEN The possible association between exposure to phenoxy herbicides and development of soft tissue sar comas has been studied in both case-control and cohort studies. In these studies soft tissue sarcoma cases have been identified from either deaths with malignant neoplasms of the connective tissue (ICD-8 171) as the underlying cause of death, from incident cancer cases with the same topography code, or from incident cases of extraskeletal sarcomas, excluding lympho and reticulosarcomas. The current study shows that the choice of source material has a considerable influence on the registered occurrence of soft tissue sarcomas in a well defined population. In 1978 to 1982 an annual number of 33 deaths from ICD-8 171 was registered in Denmark, whereas the annual number of diagnosed extraskeletal sarcomas was 231. The distribution over histologic types differed between tumours coded to the connective tissue and extraskeletal sarcomas coded to specific organs. Furthermore, during the period 1943 to 1982 the mortality from ICD-8 171 increased slightly, whereas the incidence from all extraskeletal sarcomas decreased slightly, except for males since 1970. Time trends based on any of the three source materials should, however, be interpreted with caution. Cancer 60:1896-1901,1987. S OFT TISSUE SARCOMAS (S T S ) d e fin e d a s e x tr a s k e le tal sarcom as oth er th a n lym p h osarcom as a n d reti c u lo s a r c o m a s c o n s t i t u t e o n l y 1% o f i n c i d e n t c a n c e r c a se s in d e v e lo p e d c o u n t r i e s .1,2 N e v e r t h e le s s , t h e y h a v e attracted substantial a tten tion sin ce 1977, w hen clin ical observations in S w ed en in d icated an association w ith exposure to p henoxy herb icides.3 T w o su bsequ en t ca secontrol studies, on e in n orthern4 an d o n e in southern Sw eden3 sh ow ed asso cia tio n s w ith h erb icide spraying w ith relative risks o f 5 .3 a n d 6 .8 , resp ectively. T h e three m ain types o f p h en oxy herb icides, 2 ,4 ,5 -T , 2 ,4 -D and M C P A , h ave b een w id ely u sed as w ee d k illers in agri culture an d forestry, an d 2 ,4 ,5 -T an d 2 ,4 -D w ere used extensively for d efoliation in th e V ietn am W ar. In 1981, four cases o f ST S w ere reported from sm all cohorts o f U S w orkers w h o had been exp osed acciden tally to the toxic d io x in 2 ,3 ,7 ,8 -T C D D in m an u factu re o f 2 ,4 ,5 - T .6-7 T h e s e o b s e r v a t io n s p r o m p t e d s e v e r a l fu r ther studies, the results o f w h ich have b een in con sisten t. N o cases o f STS w ere thu s fou n d in a coh ort o f F in nish From the Danish Cancer Registry, Institute of Cancer Epidemiol ogy. Danish Cancer Society, Copenhagen, Denmark. The authors thank Knud Juei, Danish Institute of Clinical Epidemi ology, for providing mortality figures. Address for reprints: Elsebeth Lynge, Danish Cancer Registry, Insti tute of Cancer Epidemiology, Danish Cancer Society, Landskronagade 66. DK-2100 Kabenhavn 0 , Denmark. A c c e p te d fo r p u b lic a tio n A p ril 21, 1987. m e n , w h o h a d s p r a y e d 2 ,4 ,5 - T a n d 2 , 4 - D 8-9; a w ea k , statistically n onsign ificant association betw een ST S and 2 ,4 ,5 -T an d 2 ,4 -D exp osu re w as foun d in a N ew Zea la n d c a s e - c o n t r o l s t u d y 10; a n e x c e s s risk a t t h e b o r d e r o f statistical significance w as fou n d in a coh ort o f workers in m a n u fa c tu r e o f M C P A i n D e n m a r k 11; a n d n o c o n s is ten t pattern o f excess risk o f S T S w as seen w ith either d uration or frequency o f h erbicide use in a recent casec o n t r o l s t u d y fr o m K a n s a s .12 In evalu atin g th ese in co n sisten t results, atten tion has b een given to m eth ologic w eaknesses inherent in ca seco n tro l stud ies, su ch as th e p ossib le recall bias in self-re ported exposure data, and to the accuracy o f th e expo sure data based o n em p lo y m en t records. L ittle attention has b een paid, h ow ever, to differences in sou rce m ate rials u sed for id en tification o f ST S cases. T ab le 1 sh ow s that ST S cases in the reported stud ies have b een identi fied from eith er d eath certificate d ata, from can cer regis try data w ith in form ation on topography on ly, or from ca n cer registry d ata w ith in fo rm a tio n o n b oth topogra p hy an d m orp hology. T h ese studies therefore cover ei ther d eaths from co n n ectiv e tissu e tum ours (IC D -8 171 or eq u iv a len t cod es in oth er IC D versions), in cid en t can cer cases w ith th e sam e site-specific IC D cod e, or in cid en t cancer cases w ith m orp h ology cod es for sar co m a s in com b in ation w ith topography cod es for soft tissu es. T h ese three sou rces for case id en tification are all availab le in D en m ark . T h e current study illustrates h ow j 1896 19176 ngto des^ i i weak, TS and :w Zea)rder of vorkeis co n sist either t case- o n has 1 caseself-re- expoe n tio n m atesh ow s d en tiregis from , jograrer ei8 171 :ident le, or % i -jft x e all ihow So. 3 Soft Tissue Sarcomas and Phenoxy Herbicides L y n g e et al. 1897 T able l. Epidemiologic Studies of the Relationship Between Soft Tissue Sarcoma and Exposure to Phenoxy Herbicides Classified by Study Design and Disease Indicator Studies reference Inclusion of No. of 95% Disease STS cases in STS confidence indicator organs cases RR interval Case-control studies Sweden, north/ph exposure (Harden and Sandstrm4) Sweden, south/ph exposure (Eriksson5) New Zeaiand/ph exposure (Smith et at."3) England & Wales/farmers, etc. (Balarajan and Acheson21) >-.".vYork State. US/Vietnam service (Greenwald a t~ ) patients, US/Vietnam service (Kang et at.13) Veterans. Mass, US/Vietnam service (Kogan and Clapp24) Kansas, US/ph exposure (Hoar et a/.l2) liaiy/ph exposure (Vineis et a/.25)|| Cohort studies Railway workers, Sweden (Axelson et al.3*) Forest workers. Sweden (Hogsted and Westerlund27) Sprayers, Finland (Riihimaeki et at.') Sprayers, Victoria, Australia (Royal Commission2*) Census-based cohorts. Sweden (Wiklund and Holm29) Unch Handers, US (US Air Force20) Vietnam veterans, Australia (Australian Veterans Health Studies21) 2, 4. 5-T manufacture, Dow Chemical. US (Ott et at.32) 2. 4, 5-T manufacture. Monsanto, US (Zack and Suskind22) TCDD-exposed, Dow Chemical. US (Cook et al.3*) TCDD-exposed Monsanto, US (Zack and Gaffey25) TCDD-exposed, BASF, GFR (Thiess et al.36) MCPA-manufacture. Denmark (Lynge") MCPA-manufacture, UK (Coggon et al.32) Incidence Incidence Incidence Incidence Incidence Incidence Mortality Incidence Incidence Mortality Mortality IncidencelT Mortality Incidence Mortality Mortality Mortality Mortality Mortality Mortality Mortality Incidence Mortality + + - - _ - - + - _ _ _ - + - 46 99 82 1961 281 234 ND 133 17 0 0 0 1 258# 0 2 0 1** 1* 1* 0 5 1 5.3 6.8 1.6 1.15t 0.70 0.83 5.16* 1.0 2.70 -- -- 1.43 0.9 -- 3.1 _ ND ND ND -- 2.72 1.06 2.4-11.5 2.6-17.3 0.7-3.3* 0.83-1.59 0.17-2.92 0.63-1.09 2.39-11.14 0.7-1.6 0.6-12.4* -- -- 0.04-7.96 0.8-1.1 -- 0.4-11.3 ND ND ND -- 0.88-6.34 0.03-5.93 ` 90%confidence interval. t Subcohort of farmers, farm managers and market gardeners, N = 42. RR = 1.7, CI95 = 1.00-2.88. $ Compared to non-Vietnam veterans from same era. Worked or lived on farmland. II Only living women included. 7 Also analyzed for mortality.* # Only subcohorts "land and/or animal husbandry" and "silvicul ture" included. ** In total three observed cases equivalent to 2.9% of deaths in the four cohorts, where approximately 0.07% was expected,6 subsequent pathology review showed only 2 of these cases to be STS.28 VA: Veterans Administration; Mass: Massachusetts; ND: no data; STS: soft tissue sarcoma; RR: relative risk. the choice of source material influences the registered .'ccurrence of, an d trends in , ST S in a w ell-d efin ed p op u la tio n . M aterial and M ethods Mortality Data A ll d eaths w ith m alignant n eop lasm o f th e co n n ectiv e tissue as th e underlying cau se o f death during th e period 1951 to 1982 w ere selected for analysis from the an nu al death certificate registrations. F or the years 1951 to 1957 these w ere coded as IC D -6 197, for 1958 to 1968 as IC D -7 197, and for 1969 to 1982 as IC D -8 171; for abbreviation th e n otation " deaths IC D -8 171" is used. M ortality d ata for this rubric are n o t availab le before 1951. Incidence D ata P op u lation -b ased cancer registration w as co m m en ced in D en m ark in 1942. From 1943 to 1977, notified cases w e r e c o d e d a c c o r d in g to a m o d if ie d v e r s io n o f I C D - 7 , 13 and from 1978 according to IC D for O n cology (IC D O ) . 14 A c o m p u t e r p r o g r a m h a s b e e n d e v e lo p e d w h e r e b y IC D -0 co d es are tran slated to th e m od ified IC D -7 codes, thu s allow in g con sisten t tabu lations over tim e. M align an t n eop lasm s o f th e co n n ectiv e tissu es have been cod ed as IC D -7 197; for abbreviation the n otation " in cid en t cases IC D -7 197" is u sed . S o ft tissu e sa rcom a cases located to specified organs have b een classified by m e a n s o f a s p e c ia l D a n is h c o d e , 13 e x c lu d in g ly m p h o s a r co m a s an d reticulosarcom as an d exclu d in g cases w ith b on e tu m ors (IC D -7 196) an d co n n ectiv e tissue tum ors 19177 ... .3.... li: (in a ..... tetai --i *r* I>1 ht'.r~. . rJ i' h - .i * r > *!*> i Table 2. Average, Annual Number of Soft Tissue Sarcoma Cases and Age-Standardized Rate per 100,000 in [Denmark 1978-1982 Based on Different Indicators for Occurrence of the Disease Men Women Disease indicator No. of cases Agestandardized rate* No. of cases Agestandardized rate* Deaths ICD-8 171 18 0.55 15 0.37 Incident cases ICD-7 197 39 1.21 28 0.67 Incident cases other STS 75 2.23 89 2.25 All incident STS cases 114 3.44 117 2.93 * World Standard Population. STS: soft tissue sarcoma. (IC D -7 197); for ab b reviation the n o ta tio n " in cid en t cases oth er ST S" is used. Tabulations B oth m ortality an d in cid en ce data have been tabu lated by sex, 5-year age groups, and 5-year calendar pe- T a ble 3. Incident Cases of ICD-7 197 and Incident Cases of Other STS Among Males and Females in Denmark 1978-1982 by Histologic Subtype Men (%) Women (%) ICD-O morphology code Behavior 3-9 ICD-7 197 (N - 195) Other STS (N - 376) ICD-7 197 (N - 141) Other STS (N =443) 8800-8804 Sarcoma NOS 13 20 18 19 8810-8814 Fibrosarcoma 11 8 17 6 8830 Malignant fibrous histiocytoma 23 8 20 4 8850-8860 Liposarcoma 23 9 21 7 8890 Leiomyosarcoma 6 18 9 31 8900-8920 Rhabdomyosarcoma 10 7 5 1 9540-9560 Neurogenic sarcoma 5 5 Other specified sarcomas* 14 26 11 27 Total 100 100 100 100 O ther specified sarcomas: 8710, 8831-8840, 8891-8895, 8930, 8980-8991, 9020-9044, 9120-9140, 9170, 9220-9240, 9270-9290, 9330, 9471, 9480-9481, 9530-9539, 9581. SIS: soft tissue sarcomas. Fio. I. Age-standardized (World Standard Population) mortality from malignant neoplasms of the connective tissue (ICD-8 171) and age-standardized incidence of all soft tissue sarcomas in Denmark, 1943 through 1982. Notice: Incidence figures for 1943 through 1977 include neuroblastomas. riods. A ge-standardized rates based o n th e W orld Stan dard Population are u sed for com parison over tim e. R esu lts T ab le 2 show s the average an n u al n um b er in 1978 to 1982 o f deaths from IC D -8 171, o f in cid en t cases cod ed to IC D -7 197, and o f in cid en t cases o f other ST S. T he m ortality figures give a n a n n u a l n u m b er o f ST S cases in D en m ark o f 33 and age-standardized rates per 100,000 o f 0.55 and 0.37 for m en an d w om en , respectively. A total annual num ber o f 231 in cid en t ST S cases is ob served w hen the tw o groups o f " in cid en t cases IC D -7 197" an d "incid en t cases oth er ST S " are co m b in ed . T h e age-standardized rates per 100,000 for all ST S cases are 3.44 an d 2.93 for m ales an d fem ales, respectively. T able 3 show s the d istrib ution by histologic types o f in cid en t cases o f IC D -7 197 and o f in cid en t cases o f o th e r S T S a m o n g m e n in 19 7 8 to 1 9 8 2 . M a lig n a n t fi brous h istiocytom a an d lip osarcom a are the predom i nant cell types am on g in cid en t IC D -7 197 cases, w hereas sarcom a n ot otherw ise sp ecified (N O S ) and 19178 T VoLio :iDENCE, L STS LES 3 vIALES T No. 8 \ Soft Tissue Sarcomas and Phenoxy Herbicides L y n g e et al. B. -------- MALES -------- FEMALES 1899 \ RTAUTY, - 8 171 -ES "> IALES T 1980 .EAR >n) mortality > 8 171) and in Denmark, hrough 1977 'orld Stan:r t im e . in 1978 to tses coded STS. The S cases in r 100,000 ctively. A se s is obxs IC D -7 in ed . The cases are 'eiy. : types o f : cases of a n t fi^ - d om i>7 c a s e s , iO S) and a u N Q 2 - < a z < ( O 1950 1960 1970 1980 1950 1960 1970 1980 YEAR F igs. 2A a n d 2B. Age-standardized (World Standard Population) incidence of (A) malignant neoplasms of the connective tissue (ICD-7 197) and of (B) other soft tissue sarcomas in Denmark, 1943 through 1982. Notice: Incidence figures for other soft tissue sarcomas for 1943 through '.977 include neuroblastomas. leiom yosarcom a are m ore frequent am ong the incident cases o f other sarcom as than am ong the in cid en t cases o f IC D -7 197. Figure 1 sh ow s th e tim e trends o f ST S in D en m ark based o n age-stan d ard ized m ortality figures an d agestand ard ized in cid en ce figures for all S T S, resp ectively. A slight increase is seen in th e m ortality from deaths IC D -8 171. T h e trend in in cid en ce o f all STS has been decreasing for both m a le an d fem ale patients u n til 1970, an d a slight increase has b een registered in the in cid en ce for m en after this p oin t in tim e. A ge-standardized rates for in cid en t cases o f IC D -7 197 and for in cid en t cases o f oth er S T S are p resen ted in Figures 2 A an d 2B . T h e in cid en ce o f IC D -7 197 has b een decreasing for both sexes, w hereas a slight increase is seen in the in cid en ce o f other ST S in m ale patients an d the in cid en ce o f other ST S has been virtually stable over tim e for w o m en . Discussion S oft tissue sarcom as form a m orphologically heteroge n eou s group o f tu m ors an d risk factors are k n ow n for a fe w o f t h e s e , e.g., th o r o tr a s t 15 a n d v in y l c h lo r id e 16 fo r an giosarcom a o f th e liver, an d v o n R ecklingh ausen n eu r o f ib r o m a t o s is fo r n e u r o s a r c o m a .17 T h e r e a r e in d i c a tion s o f io n izin g radiation b ein g a risk factor for other types o f STS as w ell, and o f ch ild h ood STS ten din g to aggregate in siblings and to be associated w ith other cancers in fam ily m em bers.2 P h en oxy herbicide m ay b e an ad d ition al risk factor c o m m o n for various typ es o f ST S, as suggested by the ob servation in Sw eden an d su b seq u en t studies. In the Sw edish case-co n tro l studies, o f STS an d exp o sure to p h en oxy herbicides, selection o f cases w as based o n p ath ology registration system s, w here in cid en t cases o f b oth sarcom as in the con n ective tissue and sarcom as o f other anatom ical sites co u ld b e identified. In subse q u en t studies, STS cases h ave been id en tified from death certificates, from can cer registers w ith topography cod es on ly or from cancer registers sim ilar to th ose used in S w ed en . T h e current stu d y sh o w s th at th e m ortality fig ures reflect on ly a fraction o f the d isease, and that b oth the occurrence o f STS and the tim e trends in ST S vary 19179 1900 C a n c e r October 15 1987 Vol.6Q IN 1000 KG ------------- E S TIM A TED FIG U R E S FO R 1 9 4 7 - 5 6 5 5 5 5 ? PERIOD FOR W HICH CANCER INCIDE NC E FIGURES ARE AVAILABLE, GIVEN A 2 0 TEAR LATENCY TIME FIG. 3. A n n u al c o n su m p tio n o f p h en o x y h erbicides in D e n m a rk , 1947 through 1985. con sid erab ly in a w ell-d efin ed p op u lation , d ep en d in g on w hich o f the three data sources is used. T h ese differences do n ot necessarily im p ly, that the estim a ted relative risks, for d e v elo p m en t o f S T S after exp osu re to p h en oxy herbicide, w ill vary b etw een stud ies w h ere ST S cases are id en tified in different w ays. T h e data sou rce used for id en tification o f ST S cases has a m ajor influence, how ever, o n the statistical p ow er o f coh ort studies. A n illustration o f this is given in the D an ish cohort study o f w orkers em p loyed in m an u fac ture o f M C P A w here five ST S cases w ere fo u n d in the D a n ish C ancer R egistry w h en in form ation o n b o th to pography and m orphology w as used for id en tification o f cases. O n ly on e o f these cases had the topography cod e IC D -7 197 on the cancer register record, an d o n ly this p atien t h ad the IC D -8 171 co d e o n th e d eath certifi c a t e .11 It has previously been p oin ted ou t that th e frequency distribution o f histologic types o f STS ascertained in the N e w Z ealan d ca se-co n tr o l stu d y differed from th o se in the S w edish studies. V ariation in p ath ologic classifica tio n or differences in etiologic factors h ave b een put fo r w a r d a s p o s s ib le e x p la n a t io n s .18 In th is s tu d y differ, e n c e s w ere fo u n d b etw een th e d istrib u tion o f histoIpgjc subtypes o f in cid en t cases o f IC D -7 and in cid en t cases of o th er sarcom as. O u r data th u s in d icate, that a possible exp lan ation for the difference found betw een N ew 7 ^ _ la n d a n d S w ed en is th at th e N ew Z ealan d stu d y in. elud ed on ly incident cases o f IC D -7, w hereas the Swed ish stu d ies in clu d ed in cid en t cases o f all ST S. N ew Z ealan d data sh ow b oth the in cid en ce (ICD-9 171) an d th e m ortality o f ST S to be higher for m en than for w o m e n , an d sh ow trend s over the p eriod 1955 to 1 9 7 9 to b e in c r e a s in g fo r b o th s e x e s .10 T h e a g e-sta n d a rd iz ed in c id e n c e (IC D -7 1 97) a n d m o rta lity figures for D en m a rk for 1978 to 1982 are low er, th an the equiva le n t fig u res for N e w Z e a la n d fo r 1975 to 1 9 7 9 . T h e r e is little d ifferen ce b etw e e n th e figures for th e tw o sex es in th e in c id e n c e o f a ll S T S in D en m ark ; a n d th e trend s in in c id e n c e o f all S T S sh o w a sligh t d ecrease o v er tim e in D en m ark ex cep t for m en sin ce 1970. T h ese differences b etw een th e figures for N ew Z ealand a n d for D enm ark are n o tica b le, b ecau se N ew Z ealan d is k n ow n to have a la rg e c o n s u m p t i o n o f 2 , 4 ,5 - T , 18 w h e r e a s v e r y lim ite d q u a n tit ie s o f 2 ,4 ,5 - T h a v e b e e n u se d in D e n m a r k .19 H ow ever, as for other rare cancers tim e trends in STS sh ou ld b e interpreted w ith ca u tio n . T h e proportion o f h istologically verified cancer cases notified to the D an ish C an cer R egistry has in creased from 68% in 1953 to 1 9 5 7 to 90% in 1 9 7 8 to 1 9 8 2 ,20 th u s in c r e a s in g t h e pro p ortion o f cases p oten tially co d ed b y m o rp h ology as sarcom as. In the K an sas ca se -c o n tr o l stu d y o n ly 81% o f th e S T S cases id en tified from can cer registry data could b e c o n f ir m e d in a p a th o lo g y r e v ie w .12 T h e r e fo r e , a p o ssi ble effect o f an increase in th e p rop ortion o f histologi cally verified cases in a can cer registry m igh t also b e that som e "STS cases" are elim inated. T im e trend s m a y also b e in flu en ced b y ch an ges in classification. T h us, neuroblastom as cou ld not be sepa rated from n eu rofibrosarcom as for the period 1943 through 1977 w here the m odified version o f the IC D -7 c o d e 13 w a s u se d ; a n d a lth o u g h th e m o d if ie d I C D -7 c o d e allow s site-sp ecific cod in g o f all extraskeletal sarcom as, fibrosarcom as and ch on d rosarcom as m ay have been co d ed to the b on e. T h is is u n lik ely to h app en w ith the I C D - 0 c o d e , 14 w h e r e to p o g r a p h ic a n d m o r p h o lo g ic d a ta are co d ed separately. A n ad dition al reason, for reservation in interpretation o f tim e trends in ST S in relation to tim e trends in phen o xy h erb icid es co n su m p tio n , is the laten cy tim e. In D en m ark , 1982 w as the last year for w h ich cancer inci d en ce figures are available. If w e assu m e that there is a 20-year latency period betw een exposure to p henoxy h erb icid e an d a p ossib le ex c ess risk for d ev elo p m en t o f ST S, th e exp osu re for w hich w e are n ow able to m easure a p ossib le effect, is the p h en oxy herbicide con su m p tion 19180 No. 8 Soft Tissue Sarcomas and Phenoxy Herbicides L y n g e et al. 1901 before 1962. A n illustration o f th is is given in Figure 3. As th e c o n s u m p t io n o f p h e n o x y h e r b ic id e s b e fo r e 1 9 6 2 w i lim ited , n egative resu lts based on tim e trends should a t p resen t b e in te rp re te d w ith c a u tio n , b u t sur veillance o f the trend in the near future is w arranted. REFERENCES 1. Young JL, Percy CL, Asire AJ. Surveillance, epidemiology and end results: Incidence and Mortality data 1973-77. Natl Cancer Inst Monogr no. 57. Bethesda Maryland: US Department of Health and Human Services, 1981. 2. Tucker M, Fraumeni JF. Soft tissue. Jn: Schottenfeld D, Fraurreni JF, eds. Cancer Epidemiology and Prevention. Philadelphia: W B F iders Company, 1982; 827-836. ... Hardell L. Malignant mesenchymal tumors and exposure to phenoxy herbicides: A clinical observation. Ldkartidningen 1977; 74:2753-2754 (In Swedish). 4. Hardell L, Sandstrom A. A case-control study: Soft-tissue sar comas and exposure to phenoxyacetic acids or chlorophenols. Br J Cancer 1979; 39:711-717. 5. Eriksson M, Hardell L, BergNO, MllerT, Axelson O. Soft tissue sarcomas and exposure to chemical substances: A case-referent study. BrJIndMed 1981; 38:27-33. 6. Honchar PA, Halperin WE. 2,4,5-T, trichlorophenol and softtissue sarcoma. Lancet 1981; i:268. 7. Cook RR. Dioxin, chloracne, and soft tissue sarcoma. Lancet 31; 1:618-619. 3. Riihimaeki V, Sisko A, Hemberg S. MortaUty of 2.4-dichlorophenoxy-acetic acid and 2,4,5-trichlorophenoxyacetic acid herbicide applicators in Finland. Scand J Work Environ Health 1982; 8:37-42. 9. Riihimaeki V, Asp S, Pukkala E, Hemberg S. Mortality and cancer morbidity among chlorinated phenoxyacid applicators in Fin land. Chemophere 1983; 12:779-784. 10. Smith AH, Pearce NE, Fisher DO, Giles HJ, Teague CA. Soft tissue sarcoma and exposure to phenoxyherbicides and chlorophenols in New Zealand. J Natl Cancer Inst 1984; 73:1111-1117. 11. Lynge E. A follow-up study of cancer incidence among workers nmanufacture of phenoxy herbicides in Denmark. BrJ Cancer 1985; ' '.:259-270. 12. Hoar SK. Blair A. Holmes FF et al. Agricultural herbicide use and risk of lymphoma and soft-tissue sarcoma. JAMA 1986; 256:1141-1147. 13. Clemmesen J. Statistical studies in the aetiology of malignant neoplasms: IV. Denmark 1943-67. Acta PatholMicrobiol Scand 1974; (Suppl) 247:19. 14. World Health Organization. ICD-O. International Classifica tion of Diseases for Oncology. Geneva: World Health Organization, 1976. 15. Committee on the Biological Effects of Ionizing Radiations. The Effects on Populations of Exposure to Low Levels of Ionizing Radiation: 1980. Washington. DC: National Academy Press, 1980; 372-376. 16. Creech JL, Johnson MN. Angiosarcoma of the liver in the manufacture of polyvinyl chloride. J Occup Med 1974; 16:150-151. 17. Sorensen SA, Mulvihill JJ, Nielsen A. Long-term follow-up of von Recklinghausen neurofibromatosis. N Engl J Med 1986; 314:1010-1015. 18. Smith AH, Fisher DO, Pearce N, Teague CA. Do agricultural chemicals cause soft tissue sarcoma? Initial findings of a case-control study in New Zealand. Community Health Stud 1982; 6:114-119. 19. ICemikaliekontrollen. Statistical figures for consumption and sale of pesticides. Lyngby (In Danish). 20. Jensen OM, Storm HH, Jensen HJ. Registration in Denmark and the study of multiple primary cancers, 1943-80. Natl Cancer Inst Monogr 1985; 68:245-251. 21. Balarajan R, Acheson ED. Soft tissue sarcomas in agriculture and forestry workers. J Epidemiol Commun Health 1984; 38:113-116. 22. Greenwald P, Kovasznay B, Collins DN, Therriault G. Sar comas of soft tissues after Vietnam Service. J Natl Cancer Inst 198473:1107-1109. 23. Kang HK, Weatherbee L, Breslin P, Lee Y, Shepard BM. Soft tissue sarcoma and military service in Vietnam: A rav; comparison group analysis of hospital patients. J Occup Med 1986; 28:1215-1218. 24. Kogan M, Clapp R. Mortality among Vietnam veterans in Massachusetts 1972-1983. Am J-Epidemiol 1985; 122:523. 25. Vineis P, Terracini B, Ciccone G et al. Phenoxy herbicides and soft-tissue sarcomas in female rice weeders: A population-based casereferent study. Scand J Work Environ Health 1987; 13:9-17. 26. Axelson O, Sundell L, Andersson K, Edling C, Kling H. Herbi cide exposure and tumor mortality. Scand J Work Environ Health 1980; 6:73-79. 27. Hogsted C, Westeriund B. Cohort study of mortality among forestry workers with and without exposure for phenoxy herbicides Ldkartidningen 1980; 77:1828-1831 (in Swedish). 28. Royal Commission on the use and effects of chemical agents on Australian personnel in Vietnam: Final report, vol. 4. Canberra: Aus tralian Government Publishing Service, 1985: 71-74. 29. Wiklund K, Holm L-E. Soft tissue sarcoma risk in Swedish agricultural and forestry workers. J Natl Cancer Inst 1986; 76:229234. 30. United States Air Force, Project Ranch Hand II. An Epidemio logic Investigation of Health Effects in Air Force Personnel Following Exposure to Herbicides. Texas: USAF, 1983. 31. Australian Veterans Health Studies. The Mortality Report: I. A Retrospective Cohort Study of Mortality among Australian National Servicemen of the Vietnam Conflict Era, and an Executive Summary ofthe Mortality Report. Canberra: Australian Government Publishing Service, 1984. 32. Ott MG, Holder BB, Olson RD. A mortality analysis of em ployees engaged in the manufacture of 2,4,5-trichlorophenoxyacetic acid. J Occup Med 1980; 22:47-50. 33. Zack JA, Suskind R. The mortality experience of workers ex posed to tetrachlorodibenzo-dioxin in a trichlorophenol process acci dent J Occup Med 1980; 22:11-14. 34. Cook RR, Townsend JO, Ott MG, Silverstein LG. Mortality experience of employees exposed to 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD). J Occup Med 1980; 22:530-532. 35. Zack JA, Gaffey WR. A mortality study of workers employed at the Monsanto company plant in Nitro, West Virginia. Environ Sci Res 1983;26:575-591. 36. Thiess AM, Frentzel-Beyme R, Link R. Mortality study of per sons exposed to dioxin in a trichlorophenol-process accident that oc curred in the BASF AG on November 17, 1983. Am J Ind Med 1982; 3:179-189. 37. Coggon D, Pannett B, Winter PD, Acheson ED, Bonsall J. Mor tality of workers exposed to 2 methyl-4 chlorophenoxyacetic acid ScandJ Work Environ Health 1986; 12:448-454. 38. Fingerhut MA, Halperin WE, Honcher PA, Smith AB, Groth DH, Russel WO. An evaluation of reports of dioxin exposure and soft tissue sarcoma pathology among chemical workers in United States. ScandJ Work Environ Health 1984; 10:299-303. 19181 234 British Journal of Industrial Mediente 1991;48:234-238 A cohort mortality study of forestry workers exposed to phenoxy acid herbicides L ois M G reen - Abstract A cohort mortality study was undertaken of forestry workers at a public electrical utility who had worked for six months or more during 1950-82 and who were routinely exposed to herbicides including phenoxy acids. A total of 1222 men with 25 274 years at risk experienced 80 deaths. Ascertainment of vital state at the end of follow up was 95*5%. The male popula tion of the province (Ontario) was used as the reference group. Overall, no excess mortality was found in this cohort relative to the reference population. A statistically sig nificant increase in deaths occurred, however, due to suicide (SMR = 210, 95% confidence interval, 95% Cl 105--376) for the cohort as a whole. No deaths were seen due to cancers such as soft tissue sarcoma and non-Hodgkin's lym phoma that have been cited by other authors as being associated with exposure to phenoxy acid herbicides. Although the cohort is not large, the absence of deaths due to these cancers is consistent with findings from other studies with sufficient numbers to allow examination of specific risks. The cohort is still young, however, and at the end of follow up most had not reached an age when increased incidence of cancer would normally be expected. Phenoxy acid herbicides comprise a large part of vegetation management programmes in Canada. Several epidemiological studies have linked exposure to these herbicides with specific cancers such as soft tissue sarcomas and non-Hodgkin's lymphomas.M A cohort mortality study was undertaken at a public utility (Ontario Hydro) that is responsible for the generation, transmission, and distribution of electricity to the province. The group under study belonged to the forestry trade. T he workers use herbicides including phenoxy acids for brush control Health Services Department, Ontario Hydro, 700 University Avenue, Toronto, Ontario, fjnarfa MSG 1X6 L M Green and line clearing along rights of way of power transmission and distribution lines. Concern regard ing the possible associations with cancer and exposure to phenoxy add herbiddes prompted the investigation. Control of vegetation is an integral part of forestry activities at this utility with herbidde application accounting for most of the control programme since 1950. Brush control and line clearing are performed throughout the year and although frequency varies seasonally, most forestry workers, are exposed to herbiddes on a weekly basis throughout the year. In the summer foliar application is province wide whereas in the colder months, basal sprays and stump treatments are routine. Exposure to herbidde occurs by breathing the aerosols or by direct contact either from touching recently sprayed vegetation or spray equipment; key factors affecting exposure include weather conditions, the height of brush and terrain, and the type o f spraying operation. Although measurements o f personal exposure were not made until 1980, it can be assumed that failure to wear protective equipment and poor work practices in the 1950s and 1960s resulted in exposures to herbiddes that were high and that were certainly greater than those likely to be experienced by most members of the general population. The work activities of this occupational group also indicate that exposure to herbiddes may be regarded as proportional to duration o f employment in the forestry trade at the company. Table 1 gives a list o f herbiddes that were used during the period under study. N ot all herbiddes were used over the entire study period; 2,4-dichlorophenoxyacetic add (2,4-D) and until March 1979, 2,4,5-trichlorophenoxyacetic add (2,4,5-T) were the most commonly used. Work with phenoxy add herbiddes increased steadily after 1950 because of the need to control greater land area under new lines, and because this means of vegetation management was the most labour and cost efiident. In a 10 year period, the amount of active ingredient of phenoxy add herbidde used increased by almost 50%. The exact 2,3,7,8-tetrachlorodibenzo-para-dioxin (TC D D ) content of the 2,4,5-T used within the company was not known, but in 1974 the permissible concentration of TC D D in 2,4,5-T registered in Canada was less than 0-5 ppm. This was reduced to Jl;48:234-238 posed y of power cem regardcancer and rotnpted the rt o f forestry : application ranime since e performed uency varies exposed to : the year. In '-nnce wide n d stump jiride occurs ontact either io n or spray sure include and terrain, ai exposure ssumed that d poor work resulted in nd that were experienced ilation. The group also ' be regarded m ent in the it were used 11 herbicides !,4-dichloroVlarch 1979, -T ) were the henoxy acid 3 because of er new lines, management In a 10 year c of phenoxy 50%. -para-dioxin vithin the permissible registered in is reduced to A cohort mortality study o fforestry workers exposed to phenoxy acid herbicides 235 Table 1 Herbicide use within Ontario Hydro Chemicalfamilies of herbicides (generic!trade name) Phenoxy acids: 2,4-D (weed spray) 2,4-D + 2,4-D P (Desormone; Silvaprop; Diphenoprop) 2.4.5- T 2,4-D -1- 2,4,5-T (Brashkill) 2.4.5- TP (Silvex, K una) Atrazine + 2,4-D Pidoram/2,4-D (Tordon 101) S-Triarinea: A rm ine Simazine (Simmaprim) Substituted urea: Dioxin (Kartnex) Monuron (Ureabor; Telvar UW*) Tebuthiuron (Spike) Other: Aminocrizole (Amitrol) Ammonium sulphamate (Amate) Arsenic trioxide Bromadl (Hyvar) Dicamba (Banvel; Dycleer) Fosamine ammonium (Krenite) Glyphosate (Roundup) Maleic hydrazide (M H 30) Paraquat (Gramoxone) Picloram (Tordon 10K) Sodium TGA Stump spray "Relative to study period. A pproxim ate period of use* 1950-82 1979-82 1950-79 1950-79 1952-79 1973-82 1965-82 1970-82 1970-82 1955-82 1950-73 1978-82 1968-82 1950-71 1952-4 1958-82 1973-82 1981-2 1975-82 1970-5 1963-82 1965-82 1950-82 1950-82 accomplished using internal company records, drivers' licence records (Ministry of Transportation and Communications) and the Canadian Mortality Data Base (Statistics Canada). Vital state was verified only from official sources. For all deaths but one, the underlying cause o f death was obtained from the official death certificate. Cause of death and the International Classification of Diseases (ICD) code were recorded as stated on the death certificate to maintain comparability with the statistics used for the reference population. For one person, only the fact of death could be confirmed from official records and the cause o f death was obtained from other contacts. Person-years at risk were calculated with a program based on the standard person-years approa ch.67Expected numbers of deaths were derived from Ontario provincial death rates for males using 15 five year age groups and seven calendar periods centred on the official five year censuses. Standardised mortality ratios (SMRs) were calculated for main ICD cause o f death categories and for suicides, lung cancer, and stomach cancer. Tests of statistical significance were applied assuming that the observed deaths had a Poisson distribution, the mean and variance of which were equal to those for the expected number. The 95% Confidence intervals (95% C l) were calculated according to Byar's formula.3 0-1 ppm in 1975 and was required to be undetectable with analytical methodology that was sensitive to concentrations o f OT ppm T C D D . Before 1974, TC D D content of 2,4,5-T may have been 60120 ppm. In Canada currently registered 2,4-D must have a total dioxin content of no more than 30 parts per billion (ppb) and no more than 10 ppb of any dioxin isomer.3 Because of a different formulation process, T C D D is not expected to occur in 2,4-D, but the legislative requirement is that it must be undetectable at a detection limit o f 1 ppb. By the nature of the routine duties o f the forestry workers at this utility, employment during the period of study may be considered synonymous with exposure to phenoxy add herbicides. The object was to determine whether exposure to phenoxy adds in this group o f workers was associated with a mortality experience different from that of the total male population o f Ontario. Subjects and methods A cohort o f 1222 workers was identified who had worked six months or more in the forestry trade ata public utility (Ontario Hydro) at some time during the period 1 January 1950 to 31 December 1982. The cohort was identified from records within the com pany. The follow up period coindded with the period of ascertainment--namely, 1950-82 and was Results Ascertainment of vital state was complete for 95-5% of the cohort. Fifty five members were lost to follow up after a mean duration in the forestry trade of 4-2 years. O f the 1222 cohort members, age specific personyears could not be calculated for two because o f the unavailability of birth dates. Therefore, the personyears calculation for the cohort includes 1220 members who contributed 25 274 person-years dur ing the 33 year follow up. T he employment-state of the Cohort as of the end of follow up is indicative of the trade's stability, most being still employed as foresters (table 2). Sixty per cent of the cohort has been followed up for 20 years or more with the mean length o f follow Table 2 Employment state by duration inforestry trade Employment state as of 31 December 1982 Active employment in forestry trade Active employment elsewhere in corporation Retired Left service Deceased Vital state unknown Total Years in forestry 'trade < 5 5-14 ? IS 99 182 34 100 13 156 117 21 18 39 Id 350 436 260 69 39 27 41 436 19133 236 Green up equal to 21 years (table 3); the main contributors are those cohort members with a long duration in the forestry trade. For the 361 cohort members who worked 20 years or more, the average duration in the trade was 28-6 years with a range of 20-46 years. About 30% of the person-years at risk are allocated where time since first entry to the forestry trade is 15 years or longer and years employed in the trade exceeds 10 (table 4). The 80 deaths identified for the period 1950-82 contributed to an SMR of 94 for mortality from ail causes indicating that this cohort experienced an overall mortality lower than that for men of cor responding age in the general population o f Ontario, but not significantly so. This was also the case for mortality from circulatory diseases, diseases o f the nervous system, respiratory system, genitourinary system, and digestive system (table 5). Seventeen deaths occurred due to malignant neo plasms (table 6). One neoplasm, acoustic neuroma, was benign and death was due to a complication of therapy. Inspection of the death certificate for one other cohort member showed that the cause of death, bladder cancer, was incorrectly coded. The ICD code applied by the vital statistics official to the underlying cause reflected a complication of the treatment of the disease (abdominal wall fistula) as probably the underlying cause of death. Due to the small number of deaths for each type of cancer, expected numbers were calculated only for lung and stomach cancer. No significant excesses in cancer Table 3 Duration offollow up and employment state Employment stau as of 31 December 982 Years offollow up < 1 0 10-19 > 2 0 Totai Table 5 Observed (Obs) and expected ( Exp) deaths, SMRs, and 95% Cls for major causes of death* Cause of death Obs Exp SM R !95% C l) All causes 80 Diseases o f circulatory system 27 External causes (accidents) 25 Neoplasms (malignant and benign) 18 Diseases of nervous system and sense organs Diseases o f respiratory system 3 Diseases o f digestive system Diseases o f genitourinary system 4 l 85-02 29-39 23-04 16-48 2*06 94(75-117) 92(61-134) 109 (70-160) 109(65-173) 97(11-350) 3-3 91 (19-265) 4-54 88(24-225) 100 100(1-556) Male Ontario population rates as a standard. Table 6 Numbers o f deaths due to m alignant neoplasms Type of cancer No ( 1950-82) Lung Stomach Pancreas Colon Bladder* Thyroid Leukaemia (acute myelogenous) Prostate Tongue Kidney Total 5 2 2 1 1 1 1 1 1 17 One other death was coded on the official death certificate to "diseases of the digestive system" but bladder cancer should be regarded as the underlying cause o f death. mortality were seen: there were five observed deaths from lung cancer v 4-57 expected and two observed deaths from stomach cancer v 1-05 expected. The SMR of 210 -for suicide with 11 observed deaths v 5-25 expected was the only statistically significant finding (p = 0-04). Active employment in forestry 187 116 238 541 trade Active employment elsewhere in 8 23 172 203 corporation Retired -- -- - 43 43 Left service 16 53 231 300 Deceased 18 14 48 80 Vital state unknown 44 8 3 55 Total 273 214 735 1222 Table 4 Distribution ofperson-years by duration of employment in theforestry trade and time sincefirst exposure Duration in forestry trade (y) <5 5-9 10-14 S15 Total Time sincefirst exposure 1 - ............ <10 10-14 15-19 *20 Total 62890 3926-9 -- -- 10215-9 860-6 636-9 2809-4 -- 4306-9 719-5 528-3 579-2 1957-8 3784-8 1055-1 676-0 839-8 4395-8 6966-7 8924-2 5768-1 4228-4 6353-6 25274-3 Discussion As expected, this cohort shows evidence of the healthy worker effect. T he low SMRs for mortality due to all causes and diseases of the circulatory system reflect the selection and survival of healthy and fit people into the physically demanding forestry occupation.4"12 A slight but non-significant excess of deaths occurred for all causes o f mortality for those employed less than five years, which is a finding characteristic of most occupational cohorts.13 No clear explanation exists for the significant twofold excess in suicides or the fact that suicide risk decreases with years spent in the forestry trade. With one exception, there was no personal medical infor mation which might help to explain this excess. One suicide death might have been accounted for by the cohort member's medical history that included a progressively disabling neurological disease. 19184 Green deaths, * !M R 195% C l I 94(75-1175 92 (61-134) 09 (70-160) 09 (65-173) 97(11-350) 91 (19-265) 88 (24-225) 00(1-556) neoplasms S o { 1950-82) 25 2 2 1 1 1 1 1 certificate to ncer should be :rved deaths so observed peered. The 'ed deaths v f significant nice of the or mortality circulatory 1 of healthy ling forestry ; o f deaths for those is a finding rts.13 significant suicide risk trade. With :dical inforexcess. One 4 for by the iciuded a use. A cohort mortality study o fforestry workers exposed to phenoxy acid herbicides 237 Hogstedt and Westerlund,14 in their cohort study of forestry workers in Sweden, reported an increase in violent deaths. Six of the 12 "violent deaths" were suicides. The standardised proportionate mortality ratios and standardised mortality odds ratios cal culated by Kogan and Clapp15 in their study of Massachusetts Vietnam veterans showed statistically significant excesses for suicides.. Other, studies relat ing either to exposure to phenoxy acid herbicide or to forestry work have not shown such excesses but few were cohort investigations in which a complete picture of the range of outcomes was obtained. In my study, the suicides appeared to be clustered in the period 1957-67 and in view o f this, consideration might be given to the theory o f imitative behaviour.14 Alternatively, a different explanation is worthy of speculation. Although it may seem that increased risk of suicide would be unrelated to forestry work or more specifically to exposure to phenoxy acid herbi cide, some published reports describe neurological toxicity associated with exposure to phenoxy add herbiddes. A fanning student committed suidde by ingestion of the dimethylamine salt of 2,4-D; the necropsy examination showed degenerative ganglion cell changes in the brain, although the substance did not appear to concentrate in the brain. The authors,17 with the provision that the cerebral changes were not due to anoxia, suggested that the nervous tissue must have been extremely sensitive to this substance. Oliver1* described personality changes and neurological disturbances in two laboratory workers who were exposed to 2,3,7,8-tetrachlorodibenzo1,4-dioxin. In a health survey o f workers in a factory which made 2,4,5-T, a correlation was found be tween the severity o f chloracne, known to be assodated with dioxin exposure, and a high score on the manic scale o f the Minnesota multiphasic person ality inventory.19 It is therefore reasonable to con sider whether neurological toxidty assodated with exposure to phenoxy add herbidde might have psychiatric manifestations such as suidde.20Whether the increased risk o f suidde is a chance finding resulting from testing for multiple hypotheses or whether it is real merits further study. The knowledge that this study was being conduc ted prompted an inquiry from one cohort member who had recently experienced a basilar artery throm bosis with infarction of the oedpitai lobe at age 53 years. Both he and his physidan raised the question of a causal relation with exposure to herbiddes. T he documented exposure for this cohort member while employed with the company was 2,4,5-T, arsenic trioxide, and monuron. The death o f one member of the cohort, at 44 years o f age was also due to basilar artery thrombosis. Such a diagnosis is uncommon at those ages and although the study does not permit a full determination o f inddence o f specific health outcomes, the question of basilar artery thrombosis as a late complication o f exposure to phenoxy add herbiddes is worthy o f consideration. At least one report exists dting changes in the central nervous system due to an unusual type o f severe athero sclerosis of the cerebral arteries in a 57 year old subject working in production of sodium 2,4,5-trichlorophenoxyacetate and butyl ester of 2,4,5-trichlorophenoxyacetate add, and with documented chloracne linked to chronic T C D D intoxication.21 Although this investigation was prompted by concern regarding findings in Sweden of soft tissue sarcomas and exposure to phenoxy add herbiddes,1 no deaths from cancer of this type were found. Nor did the study show any deaths due to non-Hodgkin's lymphoma, which other authors have shown to be assodated with exposure to this type o f herbidde.25 Because o f the small size o f the cohort, the statistical power to detect real increases in risk o f nonHodgkin's lymphoma or soft-tissue sarcoma, however, is low. This study only had sufficient statistical power (84%) to detect a sevenfold increase in non-Hodgkin's lymphoma. For soft tissue sar coma, the power was 80% to detect a 27-fold increase, with two observed deaths. Despite this, the absence o f deaths due to soft tissue sarcoma or nonHodgkin's lymphoma is consistent with findings in other studies2*"24that, as described by Johnson,25had suffident numbers to examine the risks for these cancers. The cohort is characterised by a group of young men who entered the forestry trade in the 1950s and who have mostly remained in this trade group throughout their career. Many o f the cohort members are only now approaching the age when mortality would be expected. At the end of follow up, 96% o f the person-years experience was contributed by subjects less than 55 years of age with an average age at the end of follow up of 43 years. It would therefore still be possible for an excess o f cancer deaths to be observed in further follow up. Although the follow up period of 33 years might appear adequate, the results can only be interpreted with caution. Although not directly relevant to the purpose of this investigation, it is o f interest that only one death was due to acute myelogenous leukaemia. An increased risk of this leukaemia subtype has been found in occupational groups thought to be exposed to electromagnetic fields.24"2* The primary function of the forestry trade is to perform line clearing adjacent to Ontario Hydro lines and associated apparatus and linemen at this utility work close to both transmission and distribution lines. Movement between the forestry and line trade is not uncommon. Leukaemia is a cancer with a relatively short latent period, and if a leukaemogenic risk associated with electromagnetic fields surrounding power lines exists, some evidence might be expected even in a small cohort. In summary, the results agree with findings from 238 Green other studies in which no association was found between phenoxy acid herbicide exposure and certain cancers. The cohort is small, however, and extended follow up is required. The excess of suicides in the cohort was an unexpected finding and prompts speculation regarding the biological plausibility o f the result in relation to exposure to herbicide. Evidence from the epidemiological studies completed thus far cannot discount the carcinogenic potential of phenoxy acid herbicides. It would seem from this study, however, that if such a risk exists, it is low and requires a long latent period. Consultation on the analysis was provided by Mr Jim Julian, a statistician with McMaster University, Hamilton, Ontario. 1 Eriksson M , Hardell L , Berg N O , Moller T , Axelson O. Softtissue sarcomas and exposure to chemical substances: a case referent study. Br J Ind M id 1981;38:27-33. 2 Hardeil L, Eriksson M , Lenner P, Lundgren E. Malignant lymphoma and exposure to chemicals, especially organic solvents, chlorophenols, and phenoxy adds: a case control study. B r J Cancer 1981;43:16^-76. 3 Hoar S K , Blair A, Holmes FF, t t al. Agricultural herbidde use and risk of lymphoma and soft-tissue sarcoma. J Am M ed Assoc 1986056:1141-7. 4 Woods JS, Polissar L, Sevenson RK, Henser L S, Kulander BG. Soft tissue sarcoma and non-Hodgkin's lymphoma in relation to phenoxy herbidde and chlorinated phenol exposure in western Washington. J N a tl Cancer Inst 1987;78:899-910. 5 Agriculture Canada memorandum to registrants R -l-2 1 6 , April 15, 1983. 6 Case RAM , Lea AJ. Mustard gas poisoning, chronic bronchitis and lung cancer. British Journal of Preventive and Social Medicine 1955;9:62-72. 7 Doll R . Mortality from lung cancer in asbestos workers. B r J Ind Med 1955;12:81-6. 8 Breslow N E , Day N E . Statistical methods in cancer research. Vol II. The design and analysis of cohort studies. Lyon: Inter national Agency for Research on Cancer, 1987. 9 Fox AJ, Collier PF. Low mortality rates in industrial cohort studies due to selection for work and survival in the industry. British Journal o f Preventive and Social Medicine 197600: 225-30. 10 Koskela R -S, Jarvinen E, Kolari PJ. Effect o f cohort definition and follow-up length on occupational mortality rates. Scand J Work Environ Health 1984;10:311-6. 11 Pearce Checkoway H, Shy C. Time-related factors as potential confounders and effect modifiers in studies based on an occupational cohort. Scand J Work Environ Health 1986;12:97-107. 12 M onson RR. Observations on the healthy worker effect. J Occup Med 198608:425-33. 13 D oll R. Occupational Cancer: a hazard for epidemiologists. Int J Epidemiol 1985;14:22-31. 14 Hogstedt C, Westerlund B. Cohort study o f the fatality causes in forest workers exposed and unexposed to phenoxy acid preparations. Lakartidmngen 1980;77:1828-31. (Translated for EPA by Literature Research Corporation.) 15 Kogan M D , Clapp RW. M ortality among Vietnam veterans in Massachusetts, 1972-1983. Boston: Massachusetts Office of Commissioner o f Veterans Services, Agent Orange Program, Massachusetts Department of Public Health, Division of Health Statistics and Research, 1985. 16 Phillips DP. T he influence o f suggestion on suicide: substantive and theoretical implications of the Werther effect. American Sociological Review 1974;39:340-54. 17 Nielsen K, Kaempe B, Jensen-Holm J. Fatal poisoning in man by 2,4-dichlorophenoxyacetic add (2,4-D ); determination of the agent in forensic materials. Acta pharmacologtca et toxicologica 1965;22:224-34. 18 Oliver RM. T oxic effects o f 2,3,70-tetrachlorodibenzo- 1,4dioxin in laboratory workers. B r J Ind M ed 1975;32:49-53. 19 Poland AP, Smith D , Metier G , Possick P. A health survey of workers in a 2,4-D and 2 ,4 0 -T plant. Arch Environ Health 1971;22:316-27. 20 Green LM . Suicide and exposure to phenoxy add herbiddes. Scand J Work Environ Health 1987;13:460. 21 Pazderova-Vejlupkova J, Lukas E, Nemcova M , Pickova J, Jirasek L. The development and prognosis o f chronic intoxi cation by tetrachIorodiben2o*p-dioxin in men. Arch Environ Health 1981;36:5-11. 22 Coggoa D , Panned B, Winter PD, Acheson ED, Bonsall J. Mortality o f workers exposed to 2-methyl-4-ch!orophenoxyacetic add. Scand J Work Environ Health 1986;12:448-54. 23 Royal Commission on the use and effects o f chemical agents on Australian personnd in Vietnam. Final report, cancer Canberra: 1985. (Parliamentary paper 291.) 24 Wiklund K , D ich J, Holm L. Soft-tissue sarcoma risk in Swedish licensed pestidde applications. J Occup M ed 1988;30:801-4. 25 Johnson ES. Association between soft-tissue sarcomas, malig nant lymphomas and phenoxy herbiddes/chlorophenois: evidence from occupational cohort studies. Fund A ppl Toxicol 1990;14:219-34. 26 Calle E, Savitz D . Leukaemia in occupational groups with presumed exposure to electrical and magnetic fields. New England Journal of Medicine 1985013:1476-7. 27 Milham S. Mortality from leukaemia in workers exposed to electrical and magnetic fields. New England Journal of Medicine 1982007:249. 28 Wright WE, Peters JM, Mack M . Leukaemia in workers exposed to eiectritai and magnetic fields. Lancet 19S2;ii:l 160--1. Accepted 10 September 1990 fI Brii L ir E: At UT fr< (n: Pg co. tw m fr< lo' le: tr: W( fo T1 er w: di co tic blc bit tic Pr ch ce: ac gl' ex' sb in ro ce: ac pe po me th. De Mi ES Sc: Sai Yi 19106 '.lie, de Wolff tidase and N h Chun Acta unopcpudasc [ and renal . 5. ) d glutathione ds Enzymol rendadon of v)l 1981;77: moglobin and 1986. oxic effects o f 1 1):217-44. t* monitoring occupadonal 249-57. : nephrotoxic -c and future tological and rce. Arm Clin S. Changes in iuman eryth7 Pharmacol iuman erythr marker of thione Sj p Environ udies on the iuman eryth- iolecuiar and from H u m an s. Biochemist- letabolism of Arch Toxicol British Journal of Industrial Medicine 1991;48:173-178 173 Mortality and incidence of cancer at four factories making phenoxy herbicides D avid C oggon, B rian P ann ett, P aul W in ter Abstract users, and those eating contaminated foods. Judge To assess the possible carcinogenicity of phen ments about the toxicity of new compounds in man oxy herbicides and related chlorophenols and are usually based on extrapolation from laboratory dioxins, the International Agency for Research experiments in vitro and in animals. Many pesticides on Cancer is coordinating an international have now been in use long enough, however, that it is collaborative study ofworkers exposed to these possible to assess even their chronic effects more compounds in their production or use. Four directly. Such assessment is important because British cohorts of chemical manufacturers laboratory investigation may not always reliably which have been recruited to the survey are predict toxicity in man. described. They comprise a total of 2239 men One group of compounds that have attracted employed during 1963-85. These subjects were particular attention are the phenoxy herbicides (for <I traced to 31 December 1987 through the example, 2,4,5-trichlorophenoxyaceric acid (2,4,5- National Health Service Central Register and T), 2,4-dichlorophenoxyacetic acid (2,4-D), the National Insurance Index, and their mor 2 methyl-4 chlorophenoxyacetic acid (MCPA)). D is tality compared with that in the national covered in the 1940s, these chemicals have found I population. Two deaths were from non- extensive worldwide use in agriculture and forestry, Hodgkin's lymphoma with 0*87 expected. Both as well as being sold on a smaller scale for domestic deaths occurred more than 10 years after first application in gardens. Initial case-control studies in exposure to phenoxy compounds. One further Sweden suggested that exposure to phenoxy com non-Hodgkin's lymphoma was registered in a pounds and structurally related chlorophenols might living subject with probable exposure to phen carry an increased risk of soft tissue sarcoma12 and oxy compounds. No cases ofsoft tissue sarcoma lymphoma.5This idea was supported by the findings or Hodgkin's disease were recorded. A non of four cohort studies of chemical manufacturers in significant excess of lung cancer (19 deaths the United States.^ Subsequent investigations, observed, 14*2 expected) is probably attribut however, have been less conclusive."22 Several have able to chance or a confounding effect of shown associations between the manufacture or use smoking. In one cohort only there was of phenoxy compounds and soft tissue sarcoma or increased mortality from circulatory disease lymphoma, but risk estimates have been smaller than (34 deaths observed, 20*4 expected). A nested in the earlier studies and usually not statistically case-control study did not point to any significant. occupational cause for this excess, but further The statistical uncertainty surrounding small evaluation will be needed during continued increases in risk for rare diseases presents a problem follow up. I in evaluation. T o help resolve this issue, the Inter national Agency for Research on Cancer (IARC) has The regulation of pesticides requires that the utility established an international collaborative survey of of products be balanced against any risks to the workers exposed to phenoxy acids and related com environment and to the health of manufacturers, pounds.23The study includes cohorts from six Brit ish companies. Data on two have already been published.124Here we describe the other four British workforces in the IARC survey, and report their mortality and cancer incidence. The findings are I MRC Environmental Epidemiology Unit, Univer relevant not only to the assessment of phenoxy sity of Southampton, Southampton General Hospi tal, Southampton S 09 4XY D Coggon, B Pannett, P Winter compounds but also to other chemicals that have been manufactured and formulated at the same factories. 174 Coggort, Pannen, Winter Table l Definition of cohort Factory *Phenoxy herbicides and chlorophenols produced and formulated (unless otherwise stated, dates are for both production andformulation) Other main products f Definition of cohort Number of men analysed Number of men traced A 2,4,5-T 1968-78 '- Picric acid 2.4 - D and 2,4-D P 1 9 5 4 - Dinitro-o-butyl phenol 2.4- DB 1965- Dtnitro-o-cresol MCPA and MCPP 1954 - Sim azine MCPB 1970- Aminotriazole PCPA and PAA (produced only) 1969 - Oxynils All manual employees * during April 1975October 1985 1104 B 2,4,5 -T (formulated only) 1970-74 Oichiocarbamaces All weekly paid employees 271 2,4 * 0 , 2,4-D P and 2,4-DB 1969 - Organophosphorus compounds during March 196- MCPA and MCPP 1 9 6 9 - Carbamates November 1985 MCPB formulated 1 9 6 9 - Dinocap produced 1975 - Toiuidine compounds Urea herbicides Imidazole compounds Phthalimide compounds C 2,4,5 -T (formulated only) 1959-81 Aminophylline 2,4,-D (formulated only) 1959-61 M etron id azole MCPB 1956- Sulphonamides PBA (produced only) 1956-75 Oxynils Dinocap Asulam Sodamide Diflufenican (1) All process workers on the phenoxy plant during January. 1963--December 1984 (2) All formulators and packers during January 1982-- December 1984 (3) All maintenance workers during July 1967December 1984 345 D 2.4.5- T (formulated only) 1960-79 Metallic soaps 2.4- D 1949- Plasticiser esters 2.4 - D P 1955 -- Naphthenic add MCPA 1951 - Alkyl phenols MCPP produced 1955- Phthalate esters formulated 1965 - 2.4.6- TCP (produced only) 1983 - Alt weekly paid employees during A pnl 1969December 1985 519 1082 261 343 503 *2,4,5-T=2,4,5-Trichlorophenoxyaceric add, 2,4-D = 2,4-dichlorophenoxyaceric add, 2,4-DP=*2,4-dichlorophenoxypropionic add, 2,4-DB=2,4dichlorophenoxybutyric add, MCPA= 2 methyl-4 chlorophenoxyaceric add, MCPP= 2 methyl-4 chlorophenoxpropionic add, MCPB =2 methyl-4 chiorophenoxybutyric add, PCPA parachlorophenoxyacenc add, PAA^phenoxyacedc add, PBA--phenoxybutyric add, 2,4,6-TCP=2,4,6trichlorophenol. tAs only one woman met these criteria, the study was restricted to men. Method Subjects worked at four factories each of which manufactured and formulated a range of chemicals including phenoxy herbicides (table 1). T he com pound 2,4,5-T was produced at factory A during 1968-78, but at the other factories it was only formulated. Manufacturing processes were similar at all four factories. The relevant substituted phenol was reacted with monochloroacedc acid to produce phenoxyaceric compounds, with a-chloropropionic acid to make phenoxypropionic compounds, and with y-butyrolactone to produce phenoxybutyric compounds. The 2,4,6-trichlorophenol was synthe sised at factory O by chlorination of dichlorophenol. Table 1 sets out the criteria for entry to the cohort. Scope for recruitment was limited by the availability of complete records from which subjects could be identified. At factories A, B, and D information was obtained from personnel or wages files. At factory C process and maintenance workers were identified from bound registers kept by shift foremen, and formulators and packers from personnel records. For each member of the cohort we abstracted (as far as they were available) the name, date of birth, address, national insurance number, and job history. Job histories were used to classify subjects accord ing to their potential exposure to phenoxy com pounds and chlorophenols. At factories A and D records were sufficiently detailed to distinguish jobs that entailed work with these chemicals. The few subjects who worked only in ,non-phenoxy plants were considered to have only "background" exposure. At factory C process operators were iden tified from the shift register on the phenoxy plant, and thus are known to have worked with phenoxy compounds. Maintenance workers and formulators and packers were also considered to be exposed, although a few may not have come into contact with the compounds if they only worked in the job for a short period. The employment records at Factory B 19108 Pannett, Winter mber of t lysed A Number of men traced 1082 1 261 5 343 503 >nic add, 2 ,4 -D B -2 ,4 :id, MCPB = 2 methyl-4 id d , 2,4,6-TCP = 2,4,6- oremen, and l records. For ted (as far as irth, address, cory. ijects accordlenoxy com es A and D cinguish jobs ils. The few moxy plants background" :s were idenenoxy plant, 'ith phenoxy formulators be exposed, contact with 'ue job for a 7actory B M ortality and incidence of cancer at four factories making phenoxy herbicides 175 Table 2 Observed and expected mortality by cause: all subjects Came of dtath Deaths observed Analysis based on national rates Deaths expected SMR 95% C l Expected numbers locally adjusted Deaths expected SMR 95% C l All causes Grcuiatory disease Respiratory disease Injury and poisoning All cancer Cancer o f lung Soft tissue sarcoma Hodgkin's disease Non-Hodgkin's lymphoma 152 74 8 19 37 19 0 0 2 136-22 63-76 11-78 12-27 36-84 14-15 0-18 0-43 0-87 112 116 68 155 100 134 0 0 229 94-131 91-146 29-134 93-242 71-138 81-210 0-2087 0-855 28-827 134-73 63-94 11-74 11-06 36-66 14-21 0-18 0-38 0-73 113 116 68 172 101 134 0 0 272 96-132 91-145 29-134 103-268 71-139 80-209 0-2058 0-976 33-983 Deaths were also observed from cancer of mouth (1), oesophagus (1), 3tomach (2), small intestine (1), large intestine (2), rectum (2), bowel not further specified (1), pancreas (2), larynx (1), bone (1), testis (1), and bladder (1). did not distinguish between work on different processes, and individual exposures therefore could not be assigned with confidence. Management estimated that approximately 50% of employees would have worked with phenoxy compounds during campaign periods (January-May), but ail staff were likely to walk through the production areas on occasions. No environmental or personal monitoring for phenoxy compounds, chiorophenols, or con taminant dioxins had been carried out at any o f the factories. Subjects were traced through the National Health Service Central Register and National Insurance Index up to 31 December 1987. For those who had died we obtained a copy of the death certificate with the underlying cause of death coded to the ninth revision of the International Classification o f D is eases. We were also notified of any cancers registered among living members o f the cohort. Mortality ratios standardised for age (in five year bands), sex, and five year calendar period o f death were calculated by the person-years method with rates for England and Wales as standard. In some analyses a second set of mortality ratios was derived with expected numbers o f deaths for each factory adjusted in proportion to standardised mortality ratios (SMRs) during 1974-85 for the local authority area in which the factory was situated. Confidence intervals for mortality ratios were based on the Poisson distribution. Results A total of 2256 men satisfied the criteria for entry to the cohort, but 17 had to be excluded from the analysis because their date of birth was unavailable. Of the remaining 2239 men, 2189 (97-8%) were traced, including 152 who had died and 19 who had emigrated. These 19 were followed up to the date that they left the country, while untraced subjects were considered to be at risk up to their last known date of employment. Table 1 shows the numbers of men traced from each factory. Table 2 shows mortality by cause for the full cohort. The total of deaths observed during the follow up period was a little higher than expected from national rates, and this was due largely to excesses o f circulatory disease (SM R = 116,95% Cl 91-146) and of deaths from injury and poisoning (SM R = 155, 95% C l 93-242). Overall cancer mortality was close to expectation, but there was a small excess of deaths from lung tumours (19 observed v 14-15 expected). Two deaths from nonHodgkin's lymphoma were recorded compared with Table 3 Mortality byfactoryfor selected causes Came of death Factory A Deaths Deaths observed expected All causes Grcuiatory disease Respiratory disease Injury and poisoning All cancer Cancer o f lung Non-Hodgkin's lymphoma 62 34 1 7 12 6 0 44*70 20-42 3-41 4-77 12-26 4-63 0-31 Expected numbers are based on national races. Factory Q Deaths Deaths observed expected 17 5' 1 3 5 3 0 11-90 5-29 1-25 1-43 2-83 0-99 0-07 Factory C Deaths Deaths observed expected 22 30-13 12 14-31 0 2-33 3 2-59 6 8-34 4 3-22 1 0-20 Factory D Deaths Deaths observed expected 51 49*50 23 23-75 6 4*79 6 3-48 14 13-42 6 5-30 1 0-29 176 Coggon, Pannen, Winter Table 4 Mortality from selected causes in men at Factories A, C, and D with greater than background exposure to phenoxy compounds andjor chlorophenols Cause of death Grculatory disease All cancer Cancer of lung Non-Hodgkin's lymphoma Subjects followed up from first exposure above background Deaths observed SMR 59 117 30 102 14 123 2 282 Subjects followed upfrom 10 years after first exposure above background Deaths observed SMR 31 no 13 80 6 92 2 588 SM Rs are baaed on national ratea. 0-87 expected, but there were no deaths from Hodgkin's disease or soft tissue sarcoma. Adjustment for local differences in mortality had little effect on expected numbers. Table 3 summarises mortality patterns by factory. The excesses of lung cancer and of injury and poisoning were spread across all factories, whereas the high rate of circulatory disease was confined to factory A (34 deaths observed v 20-42 expected). With adjustment for local mortality, the expected number of circulatory deaths at factory A rose to 24-49, and the excess fell just short of statistical significance at a 5% level. The two deaths from nonHodgkin's lymphoma occurred at factories C and D. Table 4 shows mortality ratios for subjects at factories A, C, and D with more than background exposure to phenoxy compounds or chlorophenols. Standardised mortality ratios for circulatory disease and all cancers were similar to those for the full cohort, whereas that for lung cancer was rather lower, especially after allowance for a latency o f 10 years from first exposure. T he two cases of nonHodgkin's lymphoma both occurred more than 10 years after first exposure to levels above back grou n d-one man had been a packer for 21 years and the other a process operator in phenoxy acid synth esis for one month. Only one further case of non-Hodgkin's lym phoma had been registered among living members of the cohort--a process worker at factory B. There were no registrations of Hodgkin's disease or soft tissue sarcoma. Eight of the 19 deaths from injury and poisoning occurred while the men concerned were still working at the factories under study. Of these, two were due to accidents at work, two to accidents outside work, three to suicide or possible suicide, and one to homicide. To investigate further the excess of circulatory disease at factory A, we carried out a nested casecontrol study. Each o f the 34 cases at the factory was matched with four controls who were under follow up at the time that the case died, and whose date of birth was as dose as possible to that of the case. Year of birth was matched to within one year for 82% of controls, and the mean year of birth of controls was one year later than that of cases. Assodations with having ever worked in various departments were examined by conditional logistic regression for matched sets.25Table 5 summarises the findings. No individual department was clearly assodated with a risk of circulatory disease. Risk was significantly increased among the aggregate of smaller depart ments (the `other' category in table 5), but detailed examination did not suggest a common underlying factor. Discussion Although the workforces that we studied were exposed to a multiplidty of chemicals, this only presents a problem of interpretation in relation to positive findings. Any reassurance deriving from Table 5 Risk of circulatory disease at Factory A according to department Department Oxynils Nitration products Triazines Engineering Effluent Plane cleaning Other Subjects evef employed in department Cases (n-=*34) Controls ( n ~ 36) 0 16 11 04 4 19 28 25 15 19 Relative risk 0 40 0 0-8 10 1-6 6-2 95% C l 0*3-64-0 0-3-2-6 0*2--4*7 0-3-8-2 2-3-16-4 i s i ro Parmett, Winter tre to phenoxy 10 y e a n afterfirst ind poisoning i still working vo were due to sutside work, , and one to )f circulatory nested casele factory was under follow :>se date of case. Year ir for 82% of ' controls was ciations with tments were :gression for findings. No ciated with a significantly aller departbut detailed a underlying rudied were s, this only t relation to riving from 9S% c i 0-3-64-0 0-3-2-6 A-7 3-2 - v-16-4 M ortality and incidence o f cancer at four factories making phenoxy herbicides 177 risks which were not increased extends to all exposures experienced by the cohort. Unfortunately, levels of exposure to specific chemicals could not be precisely estimated. In particular, no hygiene data were available for phenoxy compounds, for the chlorophenols from which they were made, or for the dioxins with which they may have been contamin ated. From descriptions of processes and methods of operation provided by longstanding employees, however, we estimate that exposures during the early pan of the study period would generally have been higher than more recently. Thus, any risks from current working practices at the factories are likely to be lower than those in the workers studied. No soft tissue sarcomas or Hodgkin's lymphomas were recorded in the study cohort during the period of follow up. There were, however, two deaths from non-Hodgkin's lymphoma compared with 0-87 expected, and one further case was registered in a subject who was still alive. Expected numbers of incident cancers could not be reliably estimated because the completeness of cancer registration in Britain has varied between regions and over time, but our findings accord with those of most previous studies in suggesting that any risk o f soft tissue sarcoma or lymphoma from occupational contact with phenoxy compounds is small.*"" A more precise estimate of the risk of these rare tumours will be possible from analysis of the full IARC survey with its much larger study population. The largest excess of cancer found was for lung tumours (19 deaths observed v 14-2 expected), and this derived from all four factories. Most o f the deaths occurred, however, less than 10 years after first exposure to phenoxy compounds or in subjects with only background exposure. After allowance for a latency o f 10 years from first exposure, the increased risk disappeared (table 4). This and the absence of a comparable association with lung cancer in most other cohort studies of phenoxy workers57*1419" suggests that phenoxy compounds are unlikely to be responsible for the excess. A more likely explanation is chance or a confounding effect of smoking. Data on the smoking habits of cohort members were not available. The only diagnostic category for which we found a clear increase in risk was circulatory disease, and this was confined to factory A where 34 deaths were observed v 24-5 expected (after adjustment for local patterns of mortality). The finding is particularly .remarkable since occupational cohorts usually exhibit a deficit o f cardiovascular deaths because of the healthy worker effect. The limitation o f the excess to one factory and the absence of similar associations in other cohorts o f phenoxy workers57,41920 argue against a hazard from the phenoxy process, but it was possible that some other aspect of work at factory A might be responsible. We therefore carried out a nested case-control study to examine other possible occupational risk factors, but could find none that would explain the observation. The excess could perhaps be due to a combination of smoking and other non-occupational causes, but it will require further evaluation during continued follow up. Despite the limitations imposed by small numbers and the potential difficulties in unravelling the effects of multiple exposures, it is important that studies such as this are carried out. They provide an essential audit of the extrapolation from laboratory data on which most decisions for the regulation o f pesticides are based. We will continue to follow up these workforces and review their mortality and cancer incidence at intervals. We thank the staff o f the N H S Central Register and the National Insurance Index for their assistance in tracing subjects and Rosemarie Kirby for her help with the computing. The study was supported by a grant from the International Agency for Research on Cancer. Requests for reprints to: Dr D Coggon. 1 Hardell L, Sandscrom A. Case-control study: soft tissue sar comas and exposure to pheaoxyaceric adds or chlorophenols. Br J Cancer 1979;39:711-7. 2 Eriksson M , Hardell L , Berg N O , Mller T , Axelson O. Soft tissue sarcomas and exposure to chemical substances: a casereferent study. Br J Ind M ed 1981;38:27-33. 3 Hardell L, Eriksson M, Lenner P, Lundgren E. Malignant lymphoma and exposure to chemicals, especially organic solvents, chlorophenols and phenoxy adds: a case-control study. Br J Cancer !981;43:16^-76. 4 Honchar PA, Halperin WE. 2,4,5-T , trichlorophenol and soft tissue sarcomas. Lancet 19814:268-9. 5 Zack JA, Suskind RR. T he mortality experience of workers exposed to tetrachlorodibenzodioxin in a trichlorophenol process accident. J Occup M ed 1980;22:11-4. 6 Cook RR, Townsend JC, Ott G, Siiverstdn LG. Mortality experience of employees exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (T C D D ). J Occup M ed 1980;22:530-2. 7 O n M G, Holder BB, Olson RD. A mortality analysis of employees engaged in the manufacture of 2,4,5-crichlorophenoxyacetic acid. J Occup M ed 1980;22:47-50. 8 Riihimaki V, Asp S, Hemberg S. Mortality of 2,4-dichlorophenoxyacenc add and 2,4,5-crichlorophenoxyacedc add herbicide applicators in Finland. Scand J Work Environ Health 1982;8:37-42. 9 Lathrop G D , Moynahan PM , Albanese RA, Wolfe WH. An epidemiologic investigation o f health effects in airforce personnel following exposure to herbicides: baseline mortality study results. Brooks Air Force Base, Texas: United States Air Force School o f Aerospace M edidne, 1983. 10 Greenwald P, Kovasznay 8 , Collins O N , Therriault G. Sar comas o f soft tissues after Vietnam service. J N atl Cancer Inst 1984;73:1107-9. 11 Smith AH, Pearce N E, Fisher D O , Giles HJ, Teague CA, Howard JK. Soft tissue sarcoma and exposure to phenoxy herbicides and chlorophenols in New Zealand. J N atl Cancer Inst 1984;73:1111-7. 12 Royal Commission on the Use and Effects of Chemical Agents on Australian Personnel in Vietnam. Final report July / 985. Vo! 4. Canberra: Australian Government Publishing Service, 1985. 13 Lynge E. A follow-up study o f cancer incidence among workers in manufacture of phenoxy herbiddes in Denmark. Br J Cancer 1985;52:25^-70. 178 Coggon, Pannett, Winter 14 Coggoa D , Pannett B, Winter PD , Acheson E D , Bonsall J. Mortality o f workers exposed to 2-methyl-4-chIorophenoxyacedc add. Scand J Work Environ Health 1986; 12:448-54. 15 Hoar SK, Blair A , Holmes FF, et al. Agricultural herbidde use and risk o f lymphoma and soft tissue sarcoma. JA M A 1986;256:1141-7. 16 Pearce NE, Smith A H , Howard JK, Sheppard RA, G iles HJ, Teague CA. N on-H odgkin's lymphoma and exposure to phenoxy herbiddes, chlorophenols, fencing work, and meat works employment: a case-control study. B r J Ind M ed 1986;43:75-83. 17 Woods JS, Polissar L , Severson R K , Heuser L S, Kulander BG. Soft tissue sarcoma and non-Hodgkin's lymphoma in relation to phenoxy herbidde and chlorinated phenol, exposure in western Washington. J H ad Cancer ins1 1987;78:899-910. 18 Veneis P, Terradni B, Ciccone G , et al. Phenoxy herbiddes and soft tissue sarcomas in female rice weeders: A populationbased case-referent study. Scand J Work Environ Health 1987;13:9-17. 19 Ott M G, Olson RA, Cook RR, Bond GG. Cohort mortality study o f chemical workers w ith potential exposure to the higher chlorinated dioxins. J Occup Med 1987;29:422-9. 20 Bond G G , Wetterstroem N H , Roush GJ, McLaren EA, Lipps T E , Cook RR. Cause specific mortality among employees engaged in the manufacture, formulation, or packaging of 2,4dichlorophenoxyaccdc add and related salts. Br J Ind Med 1988;45:98-105. 1 Persson B, Dahlander A, Ftedriksson M , Brage H N , Ohlson CG, Axelson O. Malignant lymphomas and occupational exposures. Br J Ind Med 1989;45:516-20. 22 Eriksson M , Hardeil L , Adami HO. Exposure to dioxins as a risk factor for soft tissue sarcoma; a population-based case-control study. J N atl Cancer Inst 1990;82:486-90. 23 Johnson ES, Winkelmann R, L 'Abbe KA, et al. Phenoxy add herbicides and contaminants: Description o f the IARC inter national register o f workers. Am J Ind M ed 1990;18:39-45. 24 Bishop C M , Jones A H . Non-Hodgkins lymphoma of the scalp in workers exposed to dioxins. Lancet 1981^1:369. 25 8 reslow N E , Day NE. Statistical methods in cancer research. Vol /. The analysis of case-control studies. Lyon: International Agency for Research on Cancer, 1980:248-79. Accepted 1 October 1990 Correspondence and editorials The British Journal o f Industrial Medicine wel comes correspondence relating to any of the material appearing in the journal. Results from preliminary or small scale studies may also be published in the correspondence column if this seems appropriate. Letters should be not more than 500 words in length and contain a minimum of references. Tables and figures should be kept to an absolute minimum. Letters are accepted on the understanding that they may be subject to editorial revision and shortening. The journal now also publishes editorials which are normally specially commissioned. The Editor welcomes suggestions regarding suitable topics; those wishing to submit an editorial, how ever. should do so only after discussion with the Editor. 1 Phenoxy herbicides and chiorophenols: a case control study on soft tissue sarcoma and malignant lymphoma J.G. Smith1& A.J. Christophers2 1Statistical Centre. Peter MacCallum Cancer Institute. 481 Little Lonsdale Street, Melbourne. Victoria: 2Department of Pharmacology. University o f Melbourne. Victoria. Australia. Sum m ary A case co n tro l study o n patients w ith soft tissue sarcom a an d m alignant lym phom a was under taken to test w hether there was any association between these diseases and past exposure to chlorinated phenoxy acid herbicides o r chiorophenols. It w as carried o ut over the period 1982-1988 in V ictoria. A ustralia. T h irty m ales w ith soft tissue sarco m a a n d 52 m ales with m alignant lym phom a w ere m atched by age, place of residence and sex w ith one population control and one cancer control each. E xposure was assessed by personal interview s conducted by an occupational hygienist. E xposures w ithin 5 years prior to diagnosis o f each m a tc h e d case w ere ig n o re d , b o th fo r th e cases a n d 'th e ir m a tc h e d c o n tro ls . T he estim ated relative risks for definite o r probable exposure to chlorinated phenoxy com pounds or c h io ro p h en o ls fo r a t least 1 d ay w ere 1.0 (9 5 % confidence in terv al (C l): 0 .3 - 3 .1 ) fo r so ft tissue sa rco m a and 1.5 (9 5 % C l: 0 .6 - 3 .7 ) fo r m a lig n a n t ly m p h o m a . W h e n th e c r ite rio n fo r e x p o s u re w as ra ise d to m o re th a n 30 days, the estim ated relative risks w ere 2.0 (95% C l: 0 .5 -8 .0 ) fo r soft tissue sarcom a a n d 2.7 (95% C l: 0 .7 -9 .6 ) for m alignant lym phom a. A dditional analyses were carried o u t for exposure o f at least 1 day to phenoxy herbicides alone o r chiorophenols alone. N one o f the estim ated relative risks was significantly greater than unity. A report of a case control study published in 1979 by Hardell and Sandstrdm in Sweden was the first to claim an associa tion between exposure to phenoxy herbicides and soft tissue sarcoma in humans (Hardell & Sandstrom, 1979). The authors also claimed to have found a link between exposure to chiorophenols and the same cancer and it was suggested that the carcinogenic agent in both types of chemicals could be a polychlorinated dibenzodioxin. This paper was soon followed by one reporting a similar study in a different area of Sweden (Eriksson et a i, 1981) and another one suggesting a link between exposure to these chemicals and malignant lymphoma (Hardell et al,, 1981). Prior to 1979, the claims of serious health effects from phenoxy herbicides had been confined to birth defects and other adverse pregnancy out comes. The three Swedish studies were the first linking exposure of phenoxy herbicides with human cancer and they aroused considerable concern among public health authorities around the world. In 1981 the Health Department o f the State o f Victoria, Australia, decided to support an investigation o f the possible association between exposure to phenoxy herbicides or chior ophenols and the development of soft tissue sarcoma or malignant lymphoma, using a matched case-control study design. The design specified each case of soft tissue sarcoma or malignant lymphoma to be matched by age, sex and place of residence to one control without cancer drawn from the Victorian population and one control with cancer drawn from the same group o f hospitals as the cases. Only living cases and controls were to be used and exposure was to be assessed by means of face-to-face interviews with the subjects themselves; no interviews with relatives were to be conducted. The results of this study are now reported. The main chlorinated phenoxy herbicides used in Victoria have been 2,4-dichlorophenoxy acetic acid (2,4-D ),'2,4,5trichlorophenoxy acetic acid (2,4,5-T) and 4-chioro-2-methylphenoxy acetic acid (MCPA) and their esters and amines. Other chlorinated phenoxy agricultural chemicals used to a much lesser extent are 4-chlorophenoxy acetic acid (CPA), S3 ------------------ C orrespondence: J.G . Sm ith, Statistical C entre, Peter M acC allum C an cer Institute, 481 Little L onsdale Street, M elbourne V ictoria 3000, A ustralia. R eceived 6 D ecem ber 1990; an d in revised form 5 A ugust 1991. 2-[4-chloro-2-methylphenoxy] propionic acid (mecoprop), ( )-2-[2,4,5-trichlorophenoxy] propionic acid (fenoprop), so dium 2-[2.4-dichlorophenoxy] ethyl sulphate (2,4-DES-sodium), 4-(4-chloro-2-methylphenoxy] butyric acid (MCPB), 4-(2,4-dichlorophenoxy] butyric acid (2,4-DB) and methyl ()-2-[4-(2,4-dichiorophenoxy) phenoxy] propionate (diclofop-methyl). At least two factories in Victoria have been involved in the manufacture of phenoxy herbicides and chiorophenols. Phenoxy herbicides have been widely used in Victoria to control broad-leaved weeds in cereal crops (wheat, oats, barley) and to control blackberry and other noxious weeds. Small amounts of chlorinated phenoxy com pounds are used to prevent pre-harvest dropping of fruit. Clofibrate, a chlorinated phenoxy drug, has been used to treat coronary heart disease, diabetes and high cholesterol. The main chlorophenol used in Victoria is pentachlorophenol and its sodium salt. Other chiorophenols or salts used have been 2,4,6-trichlorophenate, sodium 2,3,4,6-tetrachlorophenate, 2-benzyl-4-chlorophenol, 3-methyl-4-chlorophenol, 3,5-dimethyl-4-chlorophenoi, 3,5-dimethyl-2.4-dichlorophenol and bis(5-chloro-2-hydroxyphenyl) methane. Chiorophenols have been used mainly as wood preservatives, and also in leather tanning, as paint preservatives and anti mould treat ments for walls, as disinfectants, as preservatives in adhe sives, for slime control in paper pulp, for preserving size used on textiles, for mothproofing wool, as herbicides, and for various other fungicidal and bactericidal purposes. Subjects and methods The population from which the cases and cancer controls were selected consisted of male cancer patients registered by the Victorian Cancer Registry after l January 1982 who were aged 30 years or more at registration, who were patients at any of six major Melbourne hospitals and who were still alive at the time o f selection for the study. Since 1 January 1982 all cases of cancer diagnosed in Victoria, with the exception of non-melanoma skin cancer, have been required to be registered with the Cancer Registry. The study was restricted to six hospitals because the Cancer Registry was not permitted to contact patients before written informed consent of the patient had been obtained by the hospital concerned. The lower age limit of 30 years was specified in order to exclude young men whose cancers were unlikely to iue under)rinated astraila, place of ersonal o f each m ds or m a and th a n 30 3-7-9.6) henoxy :er than oprop), op), soES-sodviCPB), methyl 'diclobeen ,es and used in 1 crops d other ty comf fruit, used to lesterol. tchlorolts used tchloroiphenol. ophenol phenols also in d treatn adheize used and for controls :ered by ho were dents at ere still January i the squired idy was ;try was aformed hospital cified in likely to be caused by occupational exposures, taking into account a ' reasonable latency period. Only living patients were included as it was considered that occupational data from relatives was likely to be incomplete and possibly inaccurate. Cases were men with soft tissue sarcoma either coded as [CD 171 (World Health Organisation. 1977) or coded to other sites, and men with malignant lymphoma coded as 1CD 200. ICD201 or ICD 202. The study continued until interviews had been obtained from 30 patients with soft tissue sarcoma and 52 with malignant lymphoma. The his tological diagnoses for the 82 cases are shown in Table I. together with their ICD codes. Histological types and ICD codes recorded on the Cancer Registry were confirmed from individual hospital records. No review of pathology speci mens was undertaken. The cases were first diagnosed between 1976 and 1987 with 89% first diagnosed after 1 January 1982. The median age at diagnosis was 59 years (range 37 to 87) for soft tissue sarcoma cases, 39.5 years (range 28 to 57) for Hodgkin's disease cases and 59.5 years (range 30 to 87) for non-Hodgkin's lymphoma cases. For each case, one control with another type of cancer was randomly selected by Cancer Registry staff, matching for sex, age within 3 years and Statistical Division of current resi dence. There are 12 Statistical Divisions in Victoria, with the city of Melbourne comprising one. Patients with leukaemia, multiple myeloma or sarcoma of bone were not eligible as cancer controls because the aetiology of these diseases may be similar to that of lymphoma or soft tissue sarcoma. The cancer controls had 23 different ICD codes from ICD 140 to ICD 194 with the largest groups being ICD 162 -- bronchus and lung (17 controls), ICD 185 -- prostate (nine controls), ICD 153 -- colon (six controls) and ICD 154 -- rectum (six controls). Population controls were selected at random from the Electoral Register by Cancer Registry staff. It is compulsory for all Australian citizens over the age of 18 years to be on the Electoral Register. One population control was matched to each case by sex, age within 3 years and Statistical Division of current residence. Population controls who had had cancer (apart from non-melanoma skin cancer) were not eligible. All selection and matching of cases and controls was done by Cancer Registry staff, independently of the principal investigators (the authors). Five controls differed by age from their matched cases by more than 3 years (3.1 to 5.4 years). One case was resident in New South Wales, although he attended a Melbourne hospital. He was matched with two controls resident in the nearest Statistical Division in Vic toria. All other controls satisfied the matching criteria. Written consent for a personal interview was sought from each case and cancer control. In the letters the purpose of the study was described as an investigation of possible associa tions between occupations and cancer. Herbicides or other chemicals were not mentioned. Initially, written consent was also required from population controls but the response rate was so low that the interviews with the eight population controls who consented were considered to be possibly biased and were discarded. From then on population controls were contacted in writing and given the opportunity to refuse to participate. If there was no written refusal, the interviewer then contacted the potential control by telephone or in per son and, if verbal consent was obtained, the man was inter viewed. This procedure meant that the interviewer could not be blind with regard to the population controls. However in a face-to-face interview involving a person's life history, blindness with respect to cancer status is virtually impossible. Response rates in cases and controls were calculated after excluding 101 patients who were sent letters but who did not satisfy the eligibility criteria, 55 who were found to be dead by the time the letter would have been received, 14 who were reported to be no longer at the address on the Cancer Registry and 18 who were sent an incorrect letter which stated that either the patient or his relative could be inter viewed (administrative error by one hospital). Two of those sent the incorrect letter were interviewed and included in the study as cases but were not included in the calculation of the response rates. Of the 301 remaining cancer patients who were sent letters, 187 agreed to be interviewed, 25 refused and 89 did not reply. Assuming that those who did not reply were in fact refusals, the response rates were 70% for cases and 56% for cancer controls. This was the most pessimistic view as some of those who did not reply might have never T able I H istological types an d IC D -9 codes o f cases H istology o f so ft tissue sarcom a cases (30) M alignant fibrous histiocytom a Leiom yosarcom a Sarcom a, n o t otherw ise specified (N O S) L ip o sarco m a, well d ifferentiated L iposarcom a, N O S M yxoid liposarcom a Spindle cell sarcom a Fibrosarcom a, NO S D erm atofibrosarcom a, N O S E pitheloid leiom yosarcom a Synovial sarcom a, N O S C lear cell sarco m a o f tendons an d aponeuroses K a p o s i's s a rc o m a M alignant haem angiopericytom a C hondrosarcom a N O S (extraosseous) M alignant neurilem m om a No. patients 9 5 2 2 1 1 1 1 1 1 1 1 1 1 1 1 IC D code all 171 152, 158, th re e 171 b o th 171 b o th 171 171 158 171 158 173 171 171 171 173 171 171 171 H istology o f m alignant lym phom a cases (52) H o d g k in 's d ise a se - n o d u la r sclero sis N O S H o d g k in 's d ise a se - m ix ed c e llu la rity H o d g k in 's d ise a se - ly m p h o c y tic p re d o m in a n c e R eticulosarcom a L ym phocytic, poorly differentiated NOS Im m unobiastic type M ixed lym phocytic-histiocytic NOS L ym phocytic, poorly differentiated, nodular M ixed lym phocytic-histiocytic, nodular Lymphoma, NOS U ndifferentiated cell type N O S N odular NOS C o n v o lu ted cell type N O S H airy cell leukaem ia 6 3 1 15 3 2 2 8 4 2 2 2 1 .1 all 201 all 201 201 all 200 ail 200 both 200 both 200 all 202 all 202 both 202 both 202 both 202 202 202 444 J.G. SMITH et al. received their letters due to death or change o f address. It is likely that some of the cases and cancer controls refused, or did not reply to, the request for interview because they were too ill. Eleven were later found to be within 3 months of death and another six gave other health-related reasons for refusing. Sixteen of those who agreed to interview could not be interviewed because of illness or administrative problems. Of 160 population controls selected, 30 were not able to be contacted and six were ineligible because they had had cancer prior to the introduction o f compulsory cancer registration. Of the remaining 124 contacted, 37 refused and 87 were interviewed, giving a response rate of 70%. Nine eligible cancer controls and five eligible population controls who were interviewed were later excluded by Cancer Registry staff (without knowledge of the interview results). They were found not to match any cases or were superfluous to the requirement of one population control and one cancer control for each case (the control with the closest match was retained if two were available). Of those who consented, 30 men with soft tissue sarcoma and 52 men with malignant lymphoma, as well as 164 match ed controls (82 cancer controls and 82 population controls), were interviewed and included in the study. Unfortunately the distribution of interviews throughout the study was not uniform because of early administrative problems and the change to the method of obtaining population controls when it was realised that written consent was not feasible for these controls. Cases were interviewed between 1982 and 1988 with 80% interviewed prior to 1986, cancer controls between 1982 and 1988 with 18% interviewed prior to 1986 and population controls were all interviewed between 1986 and 1988. How ever all interviews were conducted by the same person (J.G.S.) who was blind as to the case/control status of the cases and the cancer controls and who was unaware of the distribution of cases and controls at the time. The extra controls and ineligible subjects who were interviewed, but later excluded helped to ensure the interviewer's blindness. The percentage of subjects reporting exposure did not vary significantly with the year of interview over the 6 years of the study (P = 0.63, chi square test, 5 degrees of freedom) and there was no apparent trend over the years (P = 0.69, chi square test for trend). The lengthy duration of the study was due to the rarity of soft tissue sarcoma cases and some administrative problems. The interviewer was an occupational hygienist with experi ence in pesticide exposures. In the interview, which usually lasted about 45 min, a comprehensive occupational history was obtained, plus details on educadon. leisure activities, and alcohol and tobacco consumption. Details on the nature and duration of exposure were sought if the subject reported any occupation or activity likely to involve the chemicals of interest. The fact that clofibrate could be a source of exposure to chlorinated phenoxy compounds was not realised until after the first ten subjects had been interviewed. From then on all subjects (except for one accidental omission) were asked about medications for diabetes, high blood pressure or high cholesterol to determine exposure to clofibrate. In addi tion to the occupational history, four specific questions were asked concerning work in the country or living on a farm, work with asbestos, use of pesticides, herbicides or wood preservatives, and work with lead. Positive answers were probed for further details. The questions on asbestos and lead were for camouflage purposes only. The subject was not told of the main purpose of the study, namely exposure to phenoxy herbicides or chlorophenols. Exposures to phenoxy herbicides or chlorophenols or clofibrate were coded as none, possible or definite/probable (Table II). This was done by the interviewer while still blind as to each subject's status and matched triad. In some instan ces this required extensive consultation with industry experts because subjects quoted superseded trade names or had worked in factories which were no longer in operation. For the analyses, exposures within 5 years prior to the year of diagnosis of a case were ignored, both for the case and his matched controls. A subject whose total lifetime exposure was less than 8 h (i.e. I day) was counted as not exposed. The main analyses were carried out by comparing matched triads for exposures, i.e. case vs both matched controls. Soft tissue sarcomas and lymphomas were analysed separately. The method of conditional logistic regression described by Bresiow and Day (1980, Chapter 7) was used to estimate relative risks and their approximate confidence intervals and test the null hypotheses. In the regression model for exposure to chlorinated phenoxy compounds or chlorophenols there were two indicator variables, one for possible exposure and one for definite exposure. The estimated relative risks re ported are for definite exposure, adjusted for the estimated risk of possible exposure, i.e. with the possible exposure Table II Exposures to chlorinated phenoxy com pounds or chlorophenols (N um bers exposed for m ore th an 30 days show n in parentheses) S o ft tissue sarcoma M alignant lym phom a E xposure Pop. Cancer Pop. Cancer Case control control Case control control D efinite/probable Spraying phenoxy herbicides O ther phenoxy acids PO P w ood preservatives' Sodium PCP as house painter P C P in carpet glues as carpet layer C hlorophenot disinfectant as cleaner C lofibrate m edication Total definite/probable 5'(4) 4 b( 1) 2(0) 1(1) 5(4) 7(2) 2(1) 7d(3) 9 bd( l ) 4'(2) KD 1(0) 1(0) 1(0) 2(2) HD 1(1) KD 3(2) 11(6) 11(2) 6(3) Possible U nknow n disinfectants U nknow n w ood preservatives o r possibly handling treated w ood Possibly exposed to PCP laurate a t w oollen mill U nknow n herbicides U nknow n chem icals in tanneries Total possible 2(2) KD KD 1(0) 2(2) l(D 3(2) KD 1(0) 3(0) 2(0) 3(3) 3(2) 4(1) 5(3) HD 'O n e subject also m ade phenoxy acid derivatives in a laboratory ( $ 30 days); "O ne subject also used pentachlorophenol w ood preservatives 30 days); 'O n e subject also exposed to sodium pentachlorophenate in m anufacture o f anim al glue ( > 30 days, already co u n ted as ex p o se d > 30 d a y s to p h e n o x y h e rb icid es): dO n e su b je ct a lso possibly ex p o se d to so d iu m pentachlorophenate in latex glues as shoem aker (possible exposure > 3 0 days, definite exposure to phenoxy herbicides ^ 30 days): 'P C P = pentachlorophenol or m m achloroohenate. . with experivhich usually tonal history activities, and tature and -ported any chemicals of a source of s not realised iewed. From nission) were d pressure or rate. In addiuestions were g on a farm, des or wood mswers were asbestos and bject was not ' exposure to rophenols or nite/probable tile still blind some instanlustry experts imes or had peration. For > the year of case and his ime exposure not exposed, iring matched controls. Soft :d separately, scribed by to estimate intervals and l for exposure phenols there exposure and tive risks rethe estimated ible exposure .ariable in the model. The score statistic was used to test each null hypothesis. Because there were three levels of exposure to the chemicals o f interest -- none, possible, definite--the test statistic was compared with the chi square distribution with two degrees of freedom. The a priori hypotheses of the study were whether ex posure of at least 1 day to chlorinated phenoxy herbicides or chlorophenols was associated with the development of soft tissue sarcoma or malignant lymphoma. Subsequent analyses were also carried out to investigate whether there was any risk associated with exposure of at least 1 day to chlorinated phenoxy herbicides alone or chlorophenols alone and wheth er there was any risk associated with exposure to either group of chemicals for more than 30 days. In addition, analyses o f the matched pairs, cases vs popula tion controls and cases vs cancer controls, were carried out. For the matched pairs analyses, the classical methods des cribed by Breslow and Day (1980. p. 182) were used to estimate relative risks and test the null hypotheses. The score test statistic was calculated and compared to the chi square distribution with two degrees of ffeedom". Approximate confidence intervals were calculated using conditional logistic regression (program STRAT, Breslow & Day, 1980 or soft ware package GLIM, Adena & Wilson, 1982). The confidence intervals reported for the matched pairs are of limited validity, because no adjustment has been made for the fact that the two comparisons made to test each hypothesis are not independent, i.e. the same cases are used in each matched pair comparison. Matched pair analyses of cases vs population controls only were carried out to test for possible effects of smoking tobacco (non-smoker, past smoker or current smoker of at least one cigarette per day for as long as 6 months) and drinking alcohol (non-drinker, past drinker or current drink er of more than 100 grams of alcohol per year). A 5 year latency period prior to the year of diagnosis of the matched cases was applied as in previous analyses. Comparisons of cases with cancer controls were not made for smoking and drinking because several of the cancers among the cancer controls are known to be strongly associated with tobacco and alcohol, namely lung, larynx, kidney and bladder. In each of the two regression models for smoking and drinking there were two indicator variables, one for past smoking/ drinking and one for current smoking/drinking. For these analyses the estimated relative risk for current smokers after adjustment for the risk for past smokers is reported and the estimated relative risk for past smokers after adjustment for the risk for current smokers is reported. Similarly the est imated relative risks are reported for current and past drinkers. As a hypothesis generating exercise, all occupations of at least 5 years' duration prior to the date of interview were coded according to the Australian Standard Classification of Occupations (Castles, 1986). Cases and controls were com pared to see if there were any significant clusters of soft tissue sarcoma or malignant lymphoma cases in each occupa tion. A latency period of 5 years prior to the year of diag nosis of the matched cases was applied. Any occupation which had a cluster of five or more cases was then analysed using the classical method for matched triads and dichoto mous exposures described by Breslow & Day (1980, p. 169). R esults Forty-three men (16 cases, 18 population controls, nine cancer controls) were definitely or probably exposed and 17 men (seven cases, six population controls, four cancer con trols) were possibly exposed to chlorinated phenoxy com pounds or chlorophenols for at least 1 day prior to 5 years before the year of diagnosis of the case in each matched triad. The exposures are listed in Table II. The total amount o f exposure for those involved in spraying phenoxy her bicides ranged from 8 h to 122 weeks, the latter being tht length of exposure o f a population control. The total amoum of exposure for those using pentachlorophenol as a wood presei^ative ranged from 14 h to 190 days, the latter being the length of exposure of a cancer control. There were nc significant differences between population controls anc cancer controls with respect to definite exposure (soft tissue sarcoma controls 7/30 vs 3/30, P = 0.30; malignant lym phoma controls 11/52 vs 6/52, P = 0.29, Fisher exact test two tailed). Hence the cancer controls and population cont rols were combined and matched triad analyses were done The data were tabulated according to the exposure of eacr subject in the triad. The matched triad data for testing th. two main hypotheses of the study are shown in Table III. th-. estimated relative risks with approximate 95% confidence intervals are shown in the first line of Table IV. Neither o the relative risks is significantly different from unity ( ? > 0 . for both). Analyses were also carried out for phenoxy herbicides am chlorophenols separately, using the same exposure criteria. Ir the phenoxy herbicide analysis, exposure to chlorophenols c clofibrate was counted as no exposure, and in the chlorc phenol analysis, exposure to phenoxy herbicides or clofibrat was counted as no exposure. The estimated relative risks ar shown in the second and third lines o f Table IV. None i statistically significant ( P > 0 .4 for all comparisons). An analysis was also carried out for exposure o f more thar 30 days to either chlorinated phenoxy compounds or chloro phenols prior to 5 years before the year of diagnosis of th matched cases (a subject exposed for 30 days or less wa counted as not exposed). The estimated relative risks ar shown in the fourth line of Table IV. They are higher tha. the risks for one day's minimum exposure but they are sti. not statistically significant (f' > 0 .2 for both comparisons Because of the small numbers exposed for more than 3: days, the confidence intervals are relatively wide. The results of the matched pair analyses are shown i Table V. The confidence intervals are wider than thos obtained for the matched triad analyses because only half th number of controls are used in each comparison and becaus pairs in which both the case and his matched control ar Table III Exposure o f m atched triads to chlorinated phenoxy com pounds o r chlorophenols for a t least one day, 5 years' latency period E xposure o f controls Exposure o f case B o th none 1 none, l possible A . S o ft tissue sarcom a N one 10 Possible 2 D efinite 4 T otal 16 4 0 0 4 3 . M alignant lym phom a N one 25 Possible 2 D efin ite 6 T otal 33 3 0 2 5 B o th possible 0 0 0 0 0 0 0 0 I none. / possible. 1 definite 1 definite 80 10 10 10 0 70 10 21 10 1 B o th definite 0 0 0 0 2 1 0 3 Total 22 3 5 30 37 4 11 52 446 J.G. SMITH et at. equally exposed are ignored in the analysis. (There were no triads in which the case and both his controls were equally exposed so the matched triad analysis was able to make full use of the data.) None of the estimated relative risks for the matched pairs was significantly different from unity (/ > 0.1 for all comparisons). The estimated relative risks for smoking tobacco or drin king alcohol are shown in Table VI. No statistically signifi cant associations were found (/> 0.1 for all comparisons). When all occupations were examined for significant clus ters of cases it was found that the 30 soft tissue sarcoma cases had 67 occupations o f at least 5 years' duration prior to the date of interview, among which there were 52 different occupation codes. The 52 malignant lymphoma cases had 111 occupations of at least 5 years' duration with 80 different occupation codes. After applying a 5 years' latency period and analysing the data as matched triads, no statistically significant clusters were found for any occupation when coded to a six digit level of specialization (/* > 0.1). When similar occupations were combined to the extent that the first four digits o f the occupation codes were the same, there was still no significant clusters (P > 0 .1 ). However the study was designed to test two specific hypotheses rather than generate new ones, and the number of cases was too small to be likely to find statistically significant clusters unless a very strong association were present. Thirteen cases (16%) and 25 cancer controls (30%) volun teered opinions as to the causes of their cancers although this was not asked by the interviewer. No cases, who gave an opinion, suspected phenoxy herbicides or chlorophenols and only one cancer control did so. Thus recall bias on the part of the subjects was not apparent. Discussion The results of this study do not support the hypotheses that exposure to chlorinated phenoxy herbicides or chlorophenols causes soft tissue sarcoma or malignant lymphoma. In the main analysis the relative risks of developing soft tissue sarcoma or malignant lymphoma following exposure of at least 1 day to these chemicals were estimated to be 1.0 (95% Cl: 0.3-3.1) and 1.5 (95% Cl: 0.6-3.7) respectively. When the relative risks were estimated from the matched pairs, the risks calculated from cases vs cancer controls were generally higher than the risks estimated from cases vs population controls, although none of the relative risks was statistically significant at the 0.1 level. If a higher relative risk had been found when comparing cases with population con trols than when comparing cases with cancer controls, j[ could have been explained in terms o f recall bias. As it s however, there is no obvious explanation for the variation between estimates of relative risk apart from chance. The number of subjects exposed was too small to be able to estimate with reasonable power the relative risks f0r exposure of more than 30 days to phenoxy herbicides or chlorophenols or for exposure of at least 1 day to chloro phenols alone. In the three early Swedish studies the estimated (matched) relative risks for exposure o f at least 1 day to chlorinated phenoxy herbicides or chlorophenols were 6.2 and 5.1 for soft tissue sarcoma (Hardell & Sandstrom, 1979; Eriksson et al., 1981) and 6.0 for malignant lymphoma (Hardell et al., 1981). In recent years the Swedish researchers have carried out two more case-control studies on soft tissue sarcoma in which the estimated relative risks were much lower. In a study reported in 1988 the estimated relative risk (unmat ched, stratified) for phenoxy acetic acids alone was 3.3 and for chlorophenols alone was less than unity (actual estimate not reported; Hardell & Eriksson, 1988). Conversely, in a study published in 1990, no significant risk was found for phenoxy herbicides alone (RR 1.3, 95% Cl: 0.7-2.6) but the estimated relative risk for chlorophenols alone was significant (RR 5.2, 95% Cl: 1.7-16.3; Eriksson et al., 1990). The estimated relative risk (matched) for exposure to either phenoxy herbicides or chlorophenols was 1.8 (95% Cl: 1.1-3.0) in the 1990 study. Many other cohort and case-control studies have been carried out to investigate the possible association between phenoxy herbicides or chlorophenols and soft tissue sarcoma or lymphoma. Of at least 18 studies of cohorts or workers known to be exposed to phenoxy herbicides or chlorophen ols, only one has found a statistically significant association with either o f these cancers. This is a study by Lynge (1985) who found a significant excess of soft tissue sarcomas in workers employed in manufacturing phenoxy herbicides in Denmark. A recent nationwide cohort study of 12 USA plants manufacturing chemicals contaminated with 2,3.7,8tetrachlorodibenzo-p-dioxin found an increased incidence of soft tissue sarcoma but the statistical significance of the results are in doubt because of misciassification of soft tissue sarcoma on death certificates (Fingerhut et al., 1991). Case control studies investigating actual exposures to the chemicals of interest, rather than presumptive exposures associated with occupations or military service in Vietnam, have been carried out in New Zealand (Smith er al., 1983, T able IV E stim ated relative risks for exposure to chlorinated phenoxy com pounds o r chlorophenols, 5 years' latency period, m atched triads, (95% confidence intervals in parentheses) Exposure S o ft tissue sarcom a M alignant lym phom a Phenoxy herbicides or chlorophenols 1 day Phenoxy herbicides 1 day C hlorophenols ^ 1 day Phenoxy herbicides or chlorophenols > 3 0 days 1.0 ( 0 .3 - 3 .1 ) 1.3 ( 0 .4 - 4 .1 ) 0 ( --)* 2.0 (0 .5 -8 .0 ) 1.5 (0 .6 -3 .7 ) 1.1 ( 0 .4 - 3 .0 ) 1.4 (0 .3 -6 .1 ) 2.7 (0 .7 -9 .6 ) 'N o soft tissue sarcom a cases were exposed to chlorophenols so confidence interval could n o t be calculated. Table V E stim ated relative risks for exposure to chlorinated phenoxy com pounds o r ch lo ro p h en o ls, 5 y ears' latency period, m atched pairs, (95% confidence intervals in parentheses)' Exposure S T S vs pop. cont. S T S vs cancer cont. M L vs pop. cont M L vs cancer cont. PH or CP > 1 day PH 1 day CP > 1 day PH or CP > 3 0 days 0.7 (0 .2 - 2.6) 0.8 (0 .2 - 3.7) 0 2.0 (0 .4 -1 0 .9 ) 1.7 ( 0 .4 - 1.0) 2.5 (0 .5 -1 2 .9 ) 0 2 .0 (0 .4 -1 0 .9 ) 0.9 ( 0 .4 - 2.4) 0.8 ( 0 .3 - 2.2) 1.2 ( 0 .3 - 5.4) 3 .0 (0 .6 -1 4 .9 ) 2.6 ( 0 .8 - 8.8) 1.8 ( 0 . 5 - 6.0) 3.0 (0 .2 -4 4 .9 ) 3.0 (0 .5 -1 7 .0 ) 'A b b r e v ia tio n s : S T S =* so ft tissu e s a rc o m a . M L TM m a lig n a n t ly m p h o m a , p o p . = p o p u la tio n , c o rn . = c o n tro ls . P H = p h e n o x y h e rb ic id e s. C P = c h lo ro p h e n o ls. on, it * is, ion ' !e r or tro- ed) ted for l et ai ied l in la latmd ate la for the ant rhe her Cl: een een ma :ers ienion *85) in n jA 7,8: of the sue the ires im. >83, Table VI E stim ated relative risks for sm oking tobacco or drinking pathological review of soft tissue sarcoma diagnoses (Alve- alc o h o l. 5 vi--: s ' laten cy , cases vs p o p u la tio n c o n tro ls. (9 5 % co n fid e n c e . intervals in parentheses)' gard & Berg, 1989), 5% of the soft tissue sarcomas reviewed were re-diagnosed as non-sarcomatous tumours. This paper S o ft tissue sarcoma M alignant lym phom a referred to two other studies in which 7% of "soft tissue vs population controls vs population controls sarcomas' and 6% of 'bone and soft tissue sarcomas' were Current sm oker Past sm oker Current drinker 2.8 ( 0 .9 - 9.0) 1.1 ( 0 . 3 - 3.8) 2.3 ( 0 .6 - 8.9) 2.2 (0.7 -6 .7 ) 2.2 (0.7 -7 .1 ) 0.6 (0.2 -2 .0 ) non-sarcomatous tumours respectively. It therefore seems possible that, in our 30 cases diagnosed as soft tissue sar coma. one or two cases may have been wrongly diagnosed. It Past drinker 2.9 (0 .4 -2 5 .0 ) 0.4 (0 .0 4 -3 .3 ) is unlikely that this would make a significant difference to the 'S m o k er = sm oking as m uch as one cigarette p er d ay fo r as long as 6 m o n th s, d r in k e r = c o n s u m in g m o re th a n 100 g ra m s o f a lc o h o l p e r y ear. main conclusion. One possible source of bias in the comparison between cases and population controls was the fact that the cases were drawn from a population o f public (non fee paying) patients at six Melbourne hospitals, whereas the population 1984; Smith & Pearce, 1986; Pearce et al., 1986, 1987), controls were drawn from all registered voters in the State of ! Kansas, USA (Hoar et al., 1986), Western Washington State. Victoria (all social classes). Most patients with cancer in USA (Woods et al., 1987) and Northern Italy (Vineis et al., Victoria would attend a Melbourne hospital at some stage in 1986). None of the main results o f these studies showed a their disease. Matching for statistical division of residence statistically significant association except for the study by helped to eliminate a possible selection bias introduced by Hoar et al. (1986) which reported a significant association the hospital locations and catchment areas. However there between non-Hodgkin's lymphoma and phenoxy herbicides remained a potential selection bias due to possible social (odds ratio 2.2, 95% Cl: 1.2-4.1). Hoar et al. found no class differences between public (non-fee paying) patients and increased risk of soft tissue sarcoma or Hodgkin's disease. the population controls. The largest case control study, by Woods et al. (1987), One measure of social class in Australia is educational reported an odds ratio of 0.9 (95% Cl: 0.5-1.5) for non- level. Cases had slightly more education than their matched Hodgkin's lymphoma and phenoxy herbicides and no in cancer controls, although the difference was not statistically creased risk of soft tissue sarcoma (actual odds ratio not significant [P = 0.08 for age of leaving school, P = 0.4 for reported). highest qualification ever obtained, Wilcoxon signed rank In the present study considerable efforts were made to test). Cases had significantly less education than their match 1 obtain the most accurate exposure data possible by the use of ed population controls (P = 0.006 for age of leaving school. face-to-face interviews and by obtaining data from the sub P --0.02 for highest qualification ever obtained). However jects themselves, rather than next-of-kin of deceased subjects. there was no association found between definite exposure to The matching of cases and controls was maintained through chlorinated phenoxy compounds or chlorophenols (categor out the analysis to maximise the power to detect increased ised as none, 1--30 days, > 3 0 days) and age o f leaving relative risks if they existed. The opinions expressed by sub school (dichotomised as ^ 15 years and > 15 years) jects as to the causes of their cancer indicated a lack of any (P = 0.92, chi square test for trend) or between definite recall bias. exposure and highest qualification ever obtained (dichomo- Two weaknesses in the study are its small size and the tised as primary or secondary school only vs trade certificate, | relatively low response rates. Larger sample sizes would have diploma or degree) (P = 0.62, chi square test for trend). If it j made the study impractical. The rarity of soft tissue sarcoma can be assumed that educational level is a good measure of meant that it took over 5 years to accrue 30 living soft tissue social class, then it can be concluded that the social class sarcoma patients who would consent to be interviewed. It differences between cases and controls did not affect the was considered undesirable to obtain larger numbers o f cases relative risks concerning exposure to chlorinated phenoxy by interviewing relatives of dead patients as the data would compounds or chlorophenols. be considerably less reliable. The sample sizes were sufficient This study has found no statistically significant association to detect a relative risk of 5 at the 0.05 level of significance between exposure to phenoxy herbicides or chlorophenols with power of 90% for soft tissue sarcoma and 99% for and the development of soft tissue sarcoma and lymphoma. malignant lymphoma. However it is a relatively small study and the findings should The response rates were low compared to the Swedish be viewed as one piece of evidence in the large and growing studies by Hardell and his colleagues but not very much literature concerning this question. lower than in the New Zealand case-control studies (79% to 88%) (Smith et al., 1983, 1984; Pearce et al., 1986, 1987). It Note: Details of this study are included in a thesis by one of ' is understandable that response rates will decrease with the us (J.G.S.) submitted to the University of Sydney. amount of effort required. Face-to-face interviews are more demanding than postal questionnaires which were used in Sweden or telephone interviews which were used in New Zealand. Age group ( < 65 vj J 65 years) and place of W e are m ost grateful to the staff o f the V ictorian C ancer Registry w ho carried out the selection and m atching o f cases and controls and correspondence w ith hospitals, in p articular M s Helen H andsjuk, D r residence (city vr country) were not significantly associated G ra h a m G iles, M s Vicki H iggins, M s R o se a n n e E vans a n d M s with response rate. Response rates for cancer patients varied A liso n D o d d s. T h e c o o p e ra tio n o f th e D ire c to rs o f M ed ical Services, 1 with different hospitals and gradually decreased over the m edical reco rd s sta ff a n d o th e r s ta ff a t th e follow ing h o sp ita ls is period o f the study, probably because of waning enthusiasm by hospital staff involved in sending out letters and following up non-responders. This is unlikely to have introduced a bias with respect to the exposures of interest however. much appreciated: Peter M acCallum C ancer Institute, A lfred H os p ita l, P rin c e H e n ry 's H o s p ita l, S t V in c e n t's H o s p ita l. A u s tin H o s pital an d Royal M elbourne H ospital. T he assistance o f D r J.D . M athew s in the preparation o f the protocol, and the cooperation of cancer patients and healthy m en, w ho agreed to be interview ed, are It proved to be too difficult to arrange for pathological- g ratefu lly ackno w led g ed . T h e a n c illa ry expenses o f th e stu d y w ere review o f soft tissue sarcoma specimens. In a recent study a su p p o rte d by a g ra n t from the H e a lth D e p a rtm e n t o f V icto ria. R eferences ADENA, M.A. Si WILSON, S.R. (1982). G eneralised L in ea r M o d els in Epidem iological Research. Case-control Studies. T h e In stat F ound atio n fo r Statistical D ata A nalysis: Sydney. ALVEGAr D. T.A. & BERG. N.O. (1989). H isto p a th o lo g y peer review o f high-grade soft tissue sarcom a: the S candinavian Sarcom a G ro u p experience. J. Clin. O ncol., 7, 1845. 19198 BRESLOW, N.E. 4 DAY, N.E. (1980). S ta tistica l M eth o d s in C ancer Research. Volume -T he Analysis o f C ase-control Studies. In tern a tional Agency for Research on C a n cer Lyon. CASTLES. I. (1986). A u stra lia n S ta n d a r d C la ssifica tio n o f O ccu p a tio n s. First edition. A ustralian B ureau o f Statistics: C an b erra. ERIKSSON, M .E , HARDELL. L,, BERG. N.O., MOLLER, T. 4 AXELSON. 0.(1981). Soft-tissue sarcom as and exposure to chem ical substances: a case-referent study. Br. J. Ind. M ed., 38, 27. ERIKSSON, M.. HARDELL, L. 4 ADAM I. H.-O. (1990). E x p o su re to dioxins as a risk factor for soft tissue sarcom a: a population-based case-control study. J. N a tl Cancer Inst., 82, 486. FIN GERHUT. M.A.. HALPERIN, W.E., MARLOW, D.A. 4 7 o th e rs (1991). C ancer m ortality in w orkers exposed to 2,3,7,8-tetrachlorodibenzop-d io x in . N ew Engl. J. M ed., 324, 212. HARDELL, L. 4 SANDSTROM, A. (1979). C a se -c o n tro l stu d y : so ft-tissu e sarcom as and exposure to phenoxyacetic acids o r chlorophenols. Br. J. Cancer, 39, 711. HARDELL, L., ERIKSSON, M ,, LENNER, P. 4 LU N D G REN . E. (1981). M alignant lym phom a and exposure to chem icals, especially organic solvents, chlorophenols an d phenoxy acids: a case-control study. Br. J. Cancer, 43, 169. HARDELL, L. 4 ERIKSSON, M. (1988). T he asso ciatio n betw een soft tissue sarcom as and exposure to phenoxyacetic acids. Cancer, 62, 652. HOAR. S.K.. BLAIR. A., HOLMES, F.F. 4 4 o th e rs (1986). A g ric u ltu ra l herbicide use and risk o f lym phom a and soft-tissue sarcom a. J A M A , 256, 1141. LYNGE, E. (1985). A fo llo w -u p stu d y o f c a n c e r incid en ce a m o n g w orkers in m anufacture o f phenoxy herbicides in D enm ark. Br. J. Cancer, 52, 259. PEARCE. N.E., SMITH. A.H.. HOWARD, I.K .. SHEPPARD. R.A.. GILES H.J. 4 TEAGUE, C-A. (1986). N o n -H o d g k in 's ly m p h o m a and exposure to phenoxyherbicides, chlorophenols, fencing w ork, and m eat w orks em ploym ent: a case-control study. Br. J. Ind. M ed 43, 75. PEA RCE. N.E., SH EPPA R D , R.A ., SM ITH . A.H. 4 TEA G U E, C.A (1987). N o n -H o d g k in 's lym phom a an d farm ing: an expanded case-co n tro l stu d y . Int. J . Cancer, 39, 155. SM ITH. A.H., FISHER, D.O., GILES. H J. 4 PEARCE. N. (1983). The N ew Z ealan d soft tissue sa rco m a case-control study: interview findings concerning phenoxyacetic acid exposure. Chemosphere 12, 565. SM ITH. A.H., PEARCE. N.E.. FISHER, D.O., GILES, H.I., TEAGUE, C.A. 4 HOWARD, I.K . (1984). S oft tissue sa rco m a a n d exposure to phenoxyherbicides an d chlorophenols in N ew Z ealand. J. Natl Cancer Inst., 73, 1111. SM ITH, A.H. 4 PEARCE, N.E. (1986). U p d a te o n so ft tissue sarcom a a n d phenoxyherbicides in N ew Z ealan d . Chem osphere, 15, 1795. VINEIS, P., T ER R A C IN I, B., C IC C O N E. G. 4 8 o th e rs (1986). P henoxy herbicides an d soft-tissue sarcom as in fem ale rice w eeders. A p o p u latio n -b a sed ,case-referen t study. Scand. J. W ork Environ. H ealth, 13, 9. W OODS, J.S., POLISSAR, L.. SEVERSON, R.K ., HEUSER, L.S. 4 KUL- A N D E R , B.G. (1987) S o ft tiss u e s a rc o m a a n d n o n - H o d g k in 's lym p h o m a in relation to phenoxyherbicide an d chlorinated phenol exposure in w estern W ash in g to n . J . N a tl C ancer Inst., 78, 899. WORLD HEALTH ORGANISATION (1977). International Classification o f D iseases. 1975 Revision, Volum e l (9th ed). W orld H ealth O rganisation: Geneva. 19139 T a b le 1. A nnual ag e-ad ju sted in cid en ce rates fo r n o n -H o d g k in 's lym phom as*in the entire stu d y area and in tw o subareas defined on the basis o f m easurem ents o f levels o f 2.4-D and 2.4.5-TP: 1985-1988 BRIEF COMMUNICATION Incidence Rates of Lymphomas and Soft-Tissue Sarcomas and Environmental Measurements of Phenoxy Herbicides Paolo Vineis* Fabrizio Foggiano, Martine Tedeschi, Giovannino Ciccone S ev er a l stu d ies {1-6) h a v e s u g g e s te d an association betw een phenoxy-her' icid e exposure and the occurrence o f jft-tissu e sarcom as and n o n -H o d g k in 's lym p h om as, but other stu d ies (7,8) h ave n ot confirm ed such an association. D ata are particularly con flictin g in the case o f soft-tissue sarcom as, whereas a recent e d ito r ia l (9) has s u g g e s te d that the ev id en ce for an association betw een phenoxy-herbicide exposure and the o c cu rren ce o f n o n -H o d g k in 's lym p h om as is stronger. W e report here the results o f an in vestigation into the occurrence o f m a lig nant lym phom as and soft-tissue sarcom as in an Italian com m u n ity w here p h en oxy herbicides-- ie, 2,4-d ich lorophenoxyacetic acid (2,4-D ) and 2,4,5trichlorophenoxyacetic acid (2,4,5-T ) and their propionic d erivatives (2 ,4 -D P and 2,4,5-TP. respectively) and 4chloro-2-m ethylphenoxyacetic acid-- have been extensively used. Methods T h e study area is represented by the w in ces o f N ovara and V ercelli (north ern Italy; total population, 9 0 0 0 0 0 ). In th ese provinces, rice grow in g is the m ain agricultural activity, and p h enoxy h erb icid es have been u sed in rice w ee d ing since 1950. A reat Entire study area (N = 253) Category A+ (N = 24) Category B (N = 30) O ther m unicipalities not included in category B Incidence rates (95ifc C l) Men W om en 8.8 (7.4-10.4) 5.8 (4.7-7.0) 11.7 (6.5-19.3) 6.2 (2.8-11.8) 18.2(11.1-28.0) 7 .4 (3 .5 -1 3 .6 ) 8.1 (6.7-9.7) 5.7 (4.7-6.9) * P e r 100 0 0 0 , ages 15 th ro u g h 7 4 years. +N = No. of cases. Thirty-tw o m unicipalities w here phenoxy herbicides have been m easured in the soil or w ater. {Thirteen m unicipalities along a canal with higher levels o f phenoxy herbicides. W e have identified all cases o f H od gk in 's d isease [International C las sification o f D iseases (IC D) code 201], n o n -H o d g k in 's ly m p h o m a (IC D c o d e s 200 and 202), and soft-tissue sarcom a (IC D co d e 171) n ew ly d iagn osed in 1985 through 1988 in m en and w om en ag ed 15 through 7 4 years liv in g in the tw o provinces. C ases were identified through a su rvey con d u cted in the four h ospitals o f the tw o p rovin ces and in e x traprovincial institutions to w hich resi dents o f the tw o provinces could be referred. After prevalent cases were dis ca r d e d , 4 3 c a s e s o f H o d g k in 's d ise a s e , 141 c a se s o f n o n -H o d g k in 's lym p h om a, and 30 cases o f soft-tissue sarcom a were id en tified in m en , and 2 0 ca ses o f H o d g k in 's d isea se, 112 c a se s o f n onH o d g k in 's lym p h o m a , and 19 c a ses o f so ft-tissu e sarcom a w ere id en tified in w om en. G as chrom atographic analyses o f soil and water contam inated by 2,4-D and 2 ,4 ,5 -T P w ere perform ed in 1974 through 1975 in several areas o f the tw o provinces. The results o f these analyses h a v e b een p u b lish ed (10). T h e sam e an alyses have been used in the present study for assessm ent o f exposure. W ater w as sam pled every 1 or 2 w eek s during the 2-year period, w hereas the soil w as sam pled tw o to five tim es per year. A reas w here exposure occurred are separated into tw o categories according to tw o different criteria. C ategory A in cludes areas w here 2,4-D or 2,4,5-T P has been detected in w ater and/or so il in at least one o f the m easurem ents (32 m unicipalities); category B includes areas w here the highest levels o f water contam ination (2 ,4 -D , 0.0 7 to 0 .46 ppm; 2,4,5-T P , 0 .1 0 to 0 .1 9 ppm) w ere found (13 m unicipalities located alon g a canal, R oggia M ora). C ategories A and B par tially overlap. A ge-adjusted incidence rates have been com puted by m eans o f the direct standardization m eth od (11), w ith the w orld population used as a reference (12). N in e ty -fiv e p ercen t c o n fid e n c e in tervals (C Is) o f the rates have been com puted according to the P oisson d istrib u tio n (13). T h e in c id e n c e rate ratio (R R ) w as com puted as a m easure o f association: the corresponding 95% CIs w ere based on the form ula exp(ln R R 1.96 S D (ln R R )] ( / / ) . Results In T able 1 the age-adjusted incid en ce rates o f n on -H od gk in 's lym ph om as are reported for the en tire study area and in the tw o subareas p reviou sly d efin e d in R eceived Septem ber 21, 1990: revised N ovem ber 29. 1990; accepted D ecem ber 11, 1990. Supported by the European Econom ic Com munity (Europe Against Cancer Program contract 9000171/1990), by A ssociazione Italiana per le R icerche sul Cancro, and by Lega Italiana per la Lotta Contro i Tum ori. Sezione Provinciale di A lessandria. Unit o f Cancer Epidem iology. D ipartim ento di Scienze Biom ediche e O ncologia U m ana, Univer sit di Torino, Italy. W e thank P. Boffetta and B. T erracini for their kind cooperation and useful com m ents. *Correspondence to: Paolo Vineis. M D . Unit of C ancer Epidem iology, D ipartim ento di Scienze Biom ediche e O ncologia U m ana. U niversit di Torino, via Santena 7,1-10126 Torino, Italy. categories A and B. T he in cid en ce rate has ciea riy increased am on g m ales in the areas in category A (p h en oxy her bicides m easured at any level) and more m arkedly in the areas that m eet the criterion for category B (m unicipalities along the canal with higher levels o f contam ination). The rate ratio com par ing the latter area with the rem aining m unicipalities w as 2.2 (95% C l, 1.4 to 3.5) am ong m ales. A slight increase am ong fem ales was also observed. The annual incidence rates (per 100 000) for H o d g k in 's d isease in the entire study area w ere 3.3 (95% C l, 2.4 to 4.4) for m ales and 1.5 (95% C l, 0.9 to 2 .3 ) for : fem ales. For soft-tissue sarcom as, these rates w ere 1.8 (95% C l, 1.2 to 2 .6 ) for m ales and 0 .9 (95% C l, 0.5 to 1.4 ) for fem ales. A bsolute numbers were too sm all to allow m eaningful interarea com parisons. Discussion i W e found a higher incidence rate o f ; n o n -H o d g k in 's ly m p h o m a s in m a le s livin g in a rice-grow ing area w here 2,4D and 2 ,4 ,5 -T P h ave b een id en tified in the so il and the water. T he rate in the m ost polluted m unicipalities w as two tim es h igher than that in the rest o f the territory. T h is observation is con sistent w ith p rev io u s reports (3,4,14). N o a s sociation betw een exposure to phenoxy herbicides and developm ent o f H odg kin's d isease or soft-tissue sarcom as has been detected, but the absolute num bers o f cases, particularly soft-tissu e sar com as, w ere sm all. A ll o f the hospitals that serve the tw o p rovin ces h ave b een su rv ey ed , so that case ascertainm ent is lik ely to be com plete. Prevalent cases have been dis carded. The incidence rates that w e have estim ated are com parable to those reported from ca n cer reg istries in W estern E u rop e (12). A n a ly tic m ea su re m ents in soil and water w ere available for the years 1974 through 1 9 7 5 , ie, 10 to 11 y ea r s b e fo r e c a n c e r in c id e n c e w as m easured; this allow s for a reasonable latent period. D efin ition o f exposure categories w as done a priori on the basis o f the analytic m easurem ents. D esp ite the w eakness o f the approach (ecoiogic analysis), the reported as sociation betw een the developm ent o f n o n -H o d g k in 's ly m p h o m a s a n d p h en oxy-herb icide contam ination is o f inter est because o f the availability o f objective m easurem ents o f the relevant ch em ic a ls in the so il and w ater. It is m ore lik ely that the ex c ess num ber o f n o n -H o d g k in 's lym p h om as in the area o f higher p ollu tion , if it is real, is at tributable to occupational rather than to environm ental exposure. A case-control study on lym phom as and occupational exp osu res is currently planned in this area. References (/) Hardell L, Sandstrom A: C ase-co n tro l study: Soft-tissue sarcom as and exposure to phenoxyacetic acids or chlorophenols. Br J C ancer 39:711-717. 1979 (2) E r ik s s o n M , Hardell L , B e r g N O , et al: Soft-tissue sarcom as and exposure to chem i cal substances: A case-referent study. Br J Ind M ed 38:27-33, 1981 (3) H a r d e l l L, E r ik s s o n M , Le n n e r P, e t a l : M alignant lym phom a and exposure to ; chem icals, especially organic solvents. I chlorophenols and phenoxy acids: A casecon tro l study. B r J C a n c e r 4 3 :1 6 9 -1 7 6 , 1981 (4) H o a r SK. B l a ir A. H o l m e s FF. e t a l : Agricultural herbicide use and risk o f lym phom a and soft-tissue sarcom a [published erratum appears in JA M A 256:3351, 1986]. JA M A 256:1141-1147, 1986 (5) L y n g e E: A follow -up study o f cancer in cidence am ong w orkers in m anufacture of phenoxy herbicides in Denm ark. Br J Cancer 5 2 :2 5 9 -2 7 0 .1 9 8 5 (6) W o o d s JS. Po l issa r L. S ev e r so n RK. et a l : Soft-tissue sarcom a and non-H odgkin's lym phom a in relation to phenoxy herbicide and chlorinated phenol exposure in w estern W ashington. JN C178:899-910, 1987 (7) S m it h A H , P e a r c e N E. F is h e r D O , e t a l : Soft tissue sarcom a and exposure to phenoxyherbicides and chlorophenols in N ew Zealand. J N C I73:1111-1117, 1984 (8) Pea r c e NE. Sh eppa r d RA, S m ith AH. et a l : N o n -H o d g k in 's ly m p h o m a an d farm ing: An expanded case-control study. Int J Can cer 39:155-161. 1987 (9) Blair A: H erbicides and n o n -H o d g k in 's lymphoma: New evidence from a study of Saskatchew an farm ers. J Natl C ancer Inst 8 2 :5 4 4 -5 4 5 ,1 9 9 0 (10) E n t e N a z io n a l e R ise Inquinam ento delle acque. del teneno e delle risaie del com prensorio Dora Baltea-Ticino-Po. M ilano. Italy: 1978 ( 11) R o t h m a n KJ: M o d e m E p id e m io lo g y . B o s ton: Little, Brown, 1986 (12) M u ir C, W a t e r h o u s e J, M a c k T . e t a l : C ancer Incidence in Five Continents, vol V. IA RC Sci Publ No. 88. Lyon: LARC, 1987 (13) H a e n sze l W . L o v el a n d DB, S ir k e n MG: Lung-cancer m ortality as related to residence and sm o k in g h istories. I. W h ite m ales. J N atl C ancer Inst 28:947-1001. 1962 (14) W ig l e DT, S em enctw RM . W il k in s K , et a l : M ortality study of C anadian m ale farm operators: N on-H odgkin's lym phom a m or tality and agricultural practices in Sas katchewan. J Natl Cancer Inst 82:575-582, 1990 Make Tracks.. ... to your nearest mailbox and send for the latest copy of the free Consumer Information Catalog. It lists about 200 free or low-cost government publications on topics like health, nutrition, careers, money management, and federal benefits. Just send your name and address to-. Consumer Information Center Department MT Pueblo, Colorado 81009 A public service of the US. General Services Administration. 19201 Original article Annals o f Oncology 3 /Suppl. 3): S85-SH 8, 1992. O 1992 K lu w e r A ca d e m ie Publishers. P rim ed in the Netherlands. Risk factors for adult soft tissue sarcoma in northern Italy S. Franceschi & D. Serraino Epidemiology Unit, Aviano Cancer Centre, Italy Summary. The role of several potential risk factors in the etiology of soft-tissue sarcoma (STS) was examined in a hos pital-based case-control study, conducted in the Friuli-Vene zia Giulia region, northeast Italy, between 1985 and 199 L. A total of 93 STS cases (53 males and 40 females, median age: 52 years) and of 721 controls (371 males and 350 females, median age: 54 years) were interviewed. Significant increased risks were associated with a history of herpes zoster infection (odds ratio (OR): 2.3,95% confidence interval (CI): 1.1-4.9), chicken-pox (OR: 2.1, 95% Cl: 1.2--4.1) and mumps (OR: 2.0, 95% Cl: 1.1-3.8). None of the other medical conditions investigated - socio-economic and anthropometric indica tors, tobacco smoking, consumption of alcoholic beverages, coffee and tea - seemed to' affect STS risk. No risk elevation was found in subjects employed in agriculture (OR - for > 10 years employment - 0.8, 95% Cl: 0.4-1.5), nor in those who reported exposure to pesticides or herbicides (OR =* 0.4, 95% Cl: 0.1-1.2). Workers who reported exposure to chemi cal agents or to benzene or other solvents for more than 10 years had, respectively, a 1.8-fold (95% Cl: 0 .7 -4 .4 ) and a 2.2-fold (95% Cl: 0 .9-5.5) higher risk of developing STS. Key words: case-control study, risk factors, soft-tissue sar coma In tr o d u c tio n S oft-tissue sarcom as (STS) are a group o f rare m alig nancies, derived from m esenchym al tissues oth er than b on es and cartilage, w hich accoun t for less than 2% o f the overall ca n cer m ortality in w estern co u n tries, in hi d in g Ita ly [1 --2J. D u e to th e rarity o f S T S a n d t o t h e h e terogen eou s cell types, few ep id em iologic studies have in vestigated th e etio lo g y o f STS [3]. M ost o f the ep id em iologic research has focu sed on occu p ation al risks, n am ely o ccu p a tio n in agriculture and exp osu re to p henoxyherbicides and ch lorinated p henols [5 -7 ], but in con sisten t results have b een re ported. A sm all fraction o f STS seem s to b e attribut able to g en etic susceptibility [3, 8 -1 0 ], heavy external ionizing radiation [3, 11, 12] and im m u n osu p p ression [1 3 -1 5 ]. T h e role o f other p otential risk factors is n ot know n, although som e etiologic h yp oth eses have b een m ade, in particular o n the role o f in fectio u s agen ts [3]. In this paper, the results o f a ca se-con trol study [4,5] o n the role o f several occupational and n o n -o ccu p ational factors in the etiology o f ST S are rep orted . years before interview, who had been admitted as in-patients or referred for follow-up to the out-patient clinics of the Aviano Cancer Center or to the general hospitals in the study area. The present analysis is based on 93 cases (53 men and 40 women), aged 16 to 79 years (median age: 52 years), interviewed before June 1991. According to the classification of Enzinger and Weiss [16|, the most frequent cell type was sarcoma of the fibrous tissue (43%), followed by myomatus sarcomas (17%), liposarcomas (13%) and various other cell types (27%). Kaposi's sarcoma linked to human immuno deficiency virus infection and visceral sarcomas were excluded. Controls The control group consisted of 721 patients (371 men and 350 women), aged 17 to 79 years (median age: 54 years), admitted to the aforementioned hospitals for a wide spectrum of diseases. Specifi cally excluded from the comparison group were patients whose cause of hospital admission was of malignant disorders, conditions related to alcohol and tobacco consumption, as well as any disease which might have resulted in diet modification, e.g. disorders of re spiratory and digestive tracts, cardiovascular diseases, diabetes, etc. Haematologic, allergic or autoimmune diseases were also excluded. Twenty-eight percent of the controls were admitted for traumatic conditions (mainly fractures and sprains), 32% for musculoskeletal diseases (mainly low-back pain and disk disorders), 17% for minor conditions requiring surgical operations and 23% for other illnesses such as disorders of ear, nose, throat, skin or teeth. M aterials and m ethods Questionnaire Since June, 1985, we have been conducting a hospital-based casecontrol study on risk factors for SIS in the Friuli-Venezia Giulia region. Northeast Italy. Cases The cases were histologically confirmed STS. diagnosed within two The questionnaire included several socio-economic indicators like education, occupation, social class (based on the head of the house hold's occupation [17|, sibship size, birth order) and life style habits, such as smoking, frequency of consumption of alcoholic and methyixanthine-containing beverages and frequency of weekly con sumption of selected food indicators. Information was obtained about acquired disorders potentially related to the immune system - chronic infectious diseases, auto- 19202 86 immune diseases, allergic conditions, common childhood infections, immunization practices, tonsillectomy and radiation exposure for diagnostic and therapeutic purposes. Age at starting and stopping employment in 17 industries or occupations, the subject's role in the industry or occupation in terms of direct involvement in production aspects, and history of exposure to 15 occupational agents or groups of agents were also investigated. Data analysis The effect of several potential risk factors on STS risk was estimated by means of odds ratios (ORs) adjusted for age and sex [18|, and their 95% confidence intervals (CIs) [19|. R esults N o n e o f the s o c io -e c o n o m ic in d icators in vestigated in the p resen t stud y se em ed to affect substantially th e risk o f STS, sin ce cases and con trols w ere sim ilarly distrib uted as regards education, social class, sibship size and p ond eral in d ex (T able 1). C ases an d con trols also sh o w ed sim ilar habits o f to b a cco u se an d a lco h o l con sum ption (not show n). T ab le 2 illustrates th e distrib ution o f STS ca ses and con trols accord in g to m ed ical history. A sim ilar pro portion o f cases and con trols reported a history o f chronic infectious d iseases, w hile a positive association em erged for ch ick en -p ox (O R - 2.1, 95% C l: 1 .2 -4 .1 ) a n d m u m p s ( O R - 2 .0 , 9 5 % C l: 1.1--3 .8 ) a n d a h is to r y o f h erp es zoster in fection (O R - 2 .3 , 95% C l: 1 .1 -4 .9 ). N o n e o f the other m ed ical con d ition s investigated was associated w ith th e prob ability o f d ev elo p in g STS, in- Table 2. Distribution of 93 cases of soft-tissue sarcoma and 7">i controls, according to history of selected diseases and medical pro cedures. Pordenone, Italy, 1985-1991. Condition* Cases (n - 93) # (%) Controls MH-OR (95% ci)1' (n - 721) * (%) Chicken-pox No Yes Mumps No Yes Measles No Yes Chronic infectious diseases No Yes Herpes-zoster No Yes Allergic conditions No Yes Past exposure to therapeutical radiations No Yes Tonsillectomy No Yes 19 (32) 40 (68) 16 (25) 49 (75) 11(17) 54 (83) # 84 (91) 8(9) 82 (88) U (12) 72 (77) 21 (23) 91 (98) 2 (2) 69 (75) 23 (25) 233 (47) 263 (53) 211 (39) 327 (61) 121 (24) 384 (76) 1 2.11 (1.15-4.U) 1 1.98 (1.06-3.76) l 1.68 (0.82-3.58) 644 (90) 1 73 (10) 0.79 (0.33-1.81) 680 (94) 1 40 (6) 2.26 (1.06-4.94) 565 (79) 1 153 (21) 1.13 (0.65-1.97) 697 (97) 1 23 (3) 0.73 (0.12-3.30) 565 (79) 1 151 (21) 1.23 (0.70-2.16) Table l. Distribution of 93 cases of soft-tissue sarcomas and 721 controls, according to socio-economic and anthropometric indica tors. Pordenone, Italy, 1985-1991. Cases (n - 93) * (%) Controls (n - 721) * (%) MH-OR (95% Cl)* Education (years) <5 6-8 >9 Xj for trend Social class V-IV (lowest) III II-1 X f for trend Other Sibship size" 1-2 3-4 >5 X f for trend Body'mass" index <23 24-26 >21 X 7 for trend 54 (58) 14(15) 25 (27) 55 (59) 27 (29) 7(8) 4(4) 19 (20) 35 (38) 39 (42) 30 (33) 35 (38) 27 (29) 425 (59) 143 (20) 153(21) 1 0.73 (0.32-1.52) 1.30 (0.96-4.02) 1.04, p - 0.31 446 (62) 211 (29) 34(5) 30(4) 1 1.08 (0.64-1.84) 1.72 (0.65-4.46) 0.86, p - 0.35 - 152(21) 240 (33) 327 (46) 1 1.14 (0.60-2.20) 0.79 (0.39-1.61) 0.27, p - 0.60 265 (37) 240 (33) 215 (30) l 1.07 (0.60-1.95) 0.91 (0.49-1.68) 0.01, p - 0.99 " Odds ratio (OR) and 95% confidence interval (Cl), adjusted for age and sex according to the Mantel-Haenszel (MH) procedure. b For some items, the sum does not add up to the total because of missing values. J For some items, the sum does not add up to the total because of missing values. h Odds ratio (OR) and 95% confidence interval (Cl), adjusted for age and sex according to the Mantel-Haenszel (MH) procedure. elud in g exp osu re to therapeutical radiations and tonsil lecto m y (T able 2). A s co n ce rn s o ccu p a tio n a l risk factors, n o significant a sso cia tio n o f STS risk w ith em p lo y m en t in agriculture (O R for > 1 0 years = 0 .8 , 95% C l: 0 .4 -1 .5 ), or exp o sure to h erbicides a n d /o r pesticides (O R = 0.4, 95% C l: 0 .1 -1 .2 ) em erged (Table 3). C onversely, an approxi m ately tw o -fo ld in crease in STS risk w as n oted am ong su bjects w h o reported m ore than 10-year ex p osu re to chem ical agents (other than p esticides or h erbicides) ( O R - 1.8, 95% CL 0 .7 -4 .4 ) or to b en zen e or other so lv en ts (O R = 2 .2 , 95% : 0 .9 -5 .5 ) (T able 3). A n alysis according to strata o f STS histologic sub group d id n ot su ggest any d iscrep an cy w ith the overall risk estim ates (not show n). D iscu ssion T h e fin d in gs o f the p resen t stu d y su ggest that p ast viral in fection s m ay increase the risk o f STS. E p iso d es of h erp es-zoster infection w ere m ore com m on am ong ST S c a ses than am ong co n tro ls, particularly in the three years p reced in g STS d iagnosis. A sim ilar relationship w as rep orted in other stu d ies o n m alignant lym p h om as, w ith th e greatest risk near the tim e o f lym p h om a diag- . ' 19203 l a a n d 7-21 led ical pro- (9 5 % CI)h 5 -4.11) 6 -3 .7 6 ) 2 -3 .5 8 ) 3 -1.81) 6-4 .9 4 ) 5 -1.97) 1 -3 .3 0 ) 1 -2 .1 6 ) `- " c a u s e of jj u ste d for :d u re . d tonsilg n ific a n t riculture )r expoA, 95% a p p r o x ii am ong osu re to b icid es) jr other gic sub: overall ast viral odes of am ong le three n sh ip m om as, a diag- Uihte J. O d d s ratio o f so ft-tissu e sa rc o m a fo r se le c te d o c c u p a tio n s and e x p o s u re s .3 P o rd e n o n e , Ita ly 1 9 8 5 - 9 1 . O ccupation o r exposure O d d s r a tio 1*- ( 9 5 % C l) b y d u r a tio n o f occupation o r exposure <10 y ears >10 years X j (trend) A griculture F u rn itu re / upholstery M echanics L ivestock/m eat processing Chem ical agents M etal d u st D yes/paints W ood dust B enzene/solvents H erbicides/ p e s tic id e s 0.4 (0 .1 -1 .7 ) 0.2( 0 .0 - 1 .5 ) 0.7 (0 .1 -3 .3 ) 1.4 (0 .2 -7 .0 ) 1.1 ( 0 .2 - 5 .6 ) 1.8 ( 0 .6 -5 .0 ) 0.9 (0 .3 -2 .9 ) 0.8 (0 .3 -2 .3 ) 0.5 (0 .1 -2 .2 ) 0.7 (0 .2 -2 .5 ) 0.8 (0 .4 -1 .5 ) 0 .79; p - 0.37 1.1 (0 .4 --2 .7 ) 0 . 1 1: p - 0 .7 4 0.9 (0 .3 -2 .4 ) 0.19: p - 0 .6 6 1.4 (0 .6 -3 .0 ) 1.8 (0 .7 -4 .4 ) 0.9 (0 .3 -2 .6 ) 0.9 (0 .2 -2 .7 ) 0.9 (0 .3 -2 .4 ) 2.2 (0 .9 -5 .5 ) 0.96: p - 0.33 2.13: p - 0.15 0.13: p - 0.72 0.07; p - 0.79 0.09; p - 0.77 1.90; p - 0 .1 7 0.4 (0 .-1 .2 )' 3.14; p - 0.08 J O nly occupations and expo su res w ith at least 5 exposed cases. " N on-exposed subjects as reference category. A djusted for age in q u in q u en n ia an d sex by th e p ro ced u re o f V lantei-H aenszel. nosis [20, 21], in dicating that first occu rren ce o f h erp es zoster infection m ay b e favored by the early p hases o f STS, p ossib ly on accou n t o f an early im pairm ent o f the im m une system . C om m on ch ild h ood in fection s w ere m ore frequently reported by cases than co n tro ls, in particular a h istory of chicken-pox and m um ps. A lth ou gh n on e o f such viruses have b een d irectly lin k ed w ith STS o r w ith oth er m alignancies, viruses have lo n g b een su sp ected to play so m e role in the origin o f h u m an sarcom as [2 2 -2 5 ]. A t tention has been recently focu sed on the association b e tw e e n c y t o m e g a lo v ir u s a n d K a p o s i's s a r c o m a , a ty p e o f ST S w h ich is o v e r re p r ese n ted a m o n g th e im m u n o su pp ressed patients [26]. T h e association b etw een past in fectiou s d iseases and STS, if it exists, m ay h ave im p ortan t im p lica tio n s fo r u n d erstan d in g ST S e tio lo g y , an d it sh o u ld b e fu rth er in vestigated in stud ies in w h ich th e m ed ical h istories can be confirm ed through h ospital records, and, possibly, by m eans o f serologic m arkers. In agreem ent w ith o th er w ork sh ow in g n o a sso cia tion b etw een STS risk and ex p osu re to radiations, neither diagnostic n or therapeutic exp osu re w as asso ciated w ith STS risk in the p resen t study [27, 28]. S im i larly, n o n e o f the s o c io -e c o n o m ic and a n th ro p o m e tr ic indicators w as foun d to affect STS risk, n or w ere ciga rette sm oking, alcoh ol, co ffee and tea co n su m p tion . T h ese findings are co n sisten t w ith a few p reviou s in vestigations [2 7 -2 9 ] w hich w ere also negative as con cerns these factors, w ith the excep tion o f a case-con trol stud y in w hich the risk o f STS w as directly a sso cia ted w ith adult b od y w eigh t [29]. A s regards the association betw een occup ation and STS risk, the results o f the p resen t stud y p rovide further evid en ce against the possib ility that em p loy m ent in agriculture an d , in particular, ex p o su re to p esti cid es or h erb icides play an im p ortan t role in the etio lo - gy o f STS. A lthough the study was based only on 93 c a s e s o f S T S , a n d fo u n d n o sign ifican t a sso c ia tio n , it w as ab le to exclu d e, at the usual level o f statistical sig n ifican ce, an in creased STS risk from m ore than 10year ex p osu re to herb icides or p esticides greater than 20% . T h is may, h en ce, be an useful contribution, par ticu larly in co n sid era tio n o f the rarity o f the d isea se and o f the recent d eb ate on the topic. T h e results o f so m e ep id em io lo g ic stu d ies seem to indicate that farm ers have elevated risks for a few non-epithelial can cers, in clu d in g STS [for a review, se e 3 0 and 31|. H ow ever, the ev id e n c e rem ains in con clu sive. A risk o f 0 .9 w a s fo u n d in a co h o rt o f S w ed ish agricultural and forestry w orkers p otentially exp osed to p henoxy acid h e r b ic id e s [3 2 ], and a ca se-co n tr o l stu d y co n d u c te d in N ew Z ealan d w as also negative [33], STS risks low er than u n ity w ere fou n d am on g agricultural, gardeners and forestry w orkers in a study based o n 15 can cer registries in E n glan d and W ales [34], C on versely, in this study, w orkers w h o rep orted to have b een exp osed for m ore than 10 years to u nsp eci fied chem ical agents or to b en zen e or oth er solvents had a nearly tw o-fold , albeit nonsignificant, in creased risk o f ST S. A lth ou gh b en zen e exp osu re w as co n sist en tly a sso cia ted w ith an elev a ted risk o f a cu te leu k em ia [35], the few stu d ies w hich have investigated the p o ten tial role o f ex p o su re to b e n z e n e o n th e risk o f d e v e l op in g ST S sh o w ed very m od est, n onsign ificant risk ele vations [27, 36]. W eaknesses and strengths o f hospital-based casecon trol studies have long b een d iscu ssed [18]. T h e g eo grap h ic area h erein exam in ed is n ot covered by a can cer registry, thus STS p atien ts in clu d ed in the stu d y grou p m ay n o t b e rep resen tative o f all STS ca ses o c c u r ring in the p op u lation . T h e collab oratin g h osp itals, how ever, in clu d ed the m ajority o f d iagn ostic and thera p eu tical facilities for STS availab le in the area an d, therefore, m ost STS cases sh ou ld have b een referred there. F urtherm ore, a good com parability b etw een cases and con trols sh ould have b een achieved by inter view ing con trols com ing from the sam e catchm ent area o f STS cases and adm itted to hospitals for a w ide sp ec trum o f acu te con d itions. A m ore im portant lim itation o f these data co m es from the p ow er o f the study, w hich was low for m any o f the variables exam in ed , thus h am pering the interpreta tion o f th e findings. M ost o f all, as regards the pur p orted role o f past in fectiou s d iseases w hich em erges from th e p resen t study, the p oten tial role o f recall bias is o f c o n c e r n , particularly sin c e o n ly in terview -d ata w ere available and no validation on past m edical records or serology has b een p ossib le. T h e perfor m a n ce o f th e in terview s in a h o sp ita l settin g sh o u ld have assu red, how ever, a m ore com p lete ascertainm ent o f m ed ical history and a clo ser com parability betw een ca n ce r c a se s an d co n tro ls than th o se o b ta in a b le in a co m m u n ity settin g [37], In c o n clu sio n , a lso bearing in m ind such lim itations and th e difficulties o f studying ep id em iologicaily the etio lo g y o f rare and h eterogen ou s d iseases like ST S, the JiLc oC f6 88 present study investigated a few risk factors that were not previously considered. Some of the reported asso ciations, like the increased risk related to history of past infectious diseases and exposure to chemical agents and benzene, may be useful in providing leads for further investigation. Acknowledgements The contribution of the Italian Association for Cancer Research, Milan, is gratefully acknowledged. We thank Mrs. Tiziana Angelin, Maria Grazia Valentini and Dema Gerdol for interviewing cases and controls, Mrs. Ilaria Calderan for editorial assistance. References 1. L e v i F, M a is o n n e u v e P, F ilib e rti R , L a V e c c h ia C , B o y le P. C an cer incidence an d m ortality in E u ro p e. Soz P raeventivm ed 1989; 34 (suppl. 2). 2. L a V ecchia C , N egri E , D ecarli A . Fasoli M , an d C islaghi C . C a n cer m o rtality in Italy: an overview o f age-specific an d agestandardized trends from 1955 to 1984. Tum ori 1990; 7 6 :8 7 166. 3. T ucker M A , F raum eni JF. Soft-tissue. In Schottenfeld D , F raum eni JF J r (eds): C an cer epidem iology and prevention. W B Saunders C om pany: Philadelphia, PA 1982; 8 2 7 -3 6 . 4. S errain o D , F ran cesch i S, T alam int R , F rustaci S, L a V ecchia C . N o n -o ccu p atio n al risk factors for ad u lt so ft-tissu e sarco m a in N o rth e rn Italy. C a n c e r C au ses a n d C o n tro l 1991; 2: 1 5 7 -6 4 . 5. S errain o D , F ranceschi S, L a V ecchia C , C a rb o n e A . O ccu p a tion an d so ft-tissu e sarco m a in N o rth ern Italy. C a n cer C auses an d C ontrol, in press. 6 . V in eis P, T e rra c in i B , C ic c o n e G et l. P h e n o x y h e rb ic id e s an d soft-tissue sarco m as in fem ale rice w eeders. A p o p u latio n based case-referent study. Scand J W ork E nviron H ealth 1987; 1 3 :9 -1 7 . 7. H o a r S K . B la ir A , H o lm e s F. A g r ic u ltu re h e r b ic id e u s e a n d risk of lym phom a an d soft-tissue sarcom a. JA M A 1986; 256: 1141-7. 8. Strong LC. T h eo ries o f pathogenesis: m u tation an d cancer. In M ulvihill JJ, M iller RW , Fraum eni JF J r (eds): G enetics of hum an cancer. N ew York: R aven Press 1977; 4 0 1 -1 5 . 9. B aird PA , W orth A J. C ongenital g eneralized fibrom atosis: an autosom al recessive condition? C lin G en et 1976; 9 :4 8 8 -9 4 . 10. Li FP, F rau m en i JF. R h ab d o m y o sarco m a in children: epi dem iologic stu d y an d identification o f a fam ily c an ce r syn drom e. JN C I 1969; 43: 1365-73. 11. K im J H , C h u F C , W o o d w ard H O e t aL R a d ia tio n in d u c e d so fttissue a n d b o n e sarcom a. R adiol 1978; 1 2 9 :5 0 1 -8 . 12. C zesin K, W ronkow ski Z . Second m alignancies o f th e irradi ated area in p atients treated for uterine cervix cancer. G ynecol O ncol 19 7 8 ;6 :3 0 9 -1 5 . 13. H o o v er R , F rau m en i JF Jr. R isk o f can cer in renal tran sp lan t re c ip ie n ts. L a n c e t 1 9 7 3 ; ii: 5 5 -7 . 14. K inlen L J, Sheil A G R , Peto J et al. C ollab o rativ e U n ited K ingd o m -A u stralasian stu d y o f c an ce r in p atien ts tre a te d w ith im m unosuppressive drugs. B r M ed J 1979; 2: 1461-6. 15. P enn I. C an cers com plicating organ tran sp lan tatio n . N Engl J M ed 1990; 323: 1767-9. 16. E n zin g er F M , W eiss SW. Soft-tissue tu m o rs. St. L ouis, C V M osby 1983. 17. O ffic e o f p o p u la tio n c e n s u s a n d su rv e y s . C la s s if ic a tio n o f occupation 1980. H M SO 1980. 18. M antel N , H aenszel W. Statistical asp ects o f th e analysis of d ata from retrospective studies o f disease. JN C I 1959; 2 r 719-48. 19. C ornfield J. A statistical problem arising from retrospective studies. In N eym an J (ed): P roceeding o f the third Berkeley S y m p o siu m , IV, B erkeley: U n iv ersity o f C a lifo rn ia Press 1956. 2 0 . C artw right RA , M cK inney PA, O 'B rien C et al. N o n -H o d g k in s ly m p h o m a: case-co n tro l ep id em io lo g ical stu d y in Y orkshire. L eukem aR es 1988; 12:81-8. 21. F ranceschi S, S erraino D. Bidoli E et al. T h e epidem iology of non-H o d g k in 's lym phom a in the n o rth -east o f Italy: a hospitalbased case-control study. L eukem ia Res 1989; 1 3 :4 6 5 -7 2 . 2 2 . R o u s P. A s a rc o m a o f th e fow l tra n s m is ib le b y a n a g e n t s e p a r ab le from the tu m o r cells. J E xp M ed 1911; 1 3 :397-411. 23. M orton D L. Soft-tissue sarcom as. In H olland JF, Frei E (eds): C an cer m edicine. Philadelphia: L ea and Febiger 1973; 1845-61. 24. H irshaut Y. Pei DT, M arcove RC et al. Seroepidem iology of h u m a n sa rc o m a an tig en (S 1). N E n g l J M e d 1974; 2 9 1 :1 1 0 3 -7 . 2 5 . K a rp a s A , M c G e o c h V, T u c k e rm a n F e t al. V ira l-c o d e d in fo r m a tio n in h u m a n r h a b d o m y o s a r c o m a c e lls. L a n c e t 1974; ii: 1330-1. 2 6 . G ira ld o G , B eth E , H u an g E S. K ap o si's sa rc o m a an d its rela tio n sh ip to cytom egalovirus (C M V ): III. C M V D N A an d C M V e a r ly a n tig e n s in K a p o s i's s a rc o m a . In t J C a n c e r 1 9 7 8 ; 26: 2 3 -9 . 27. W oods JS, P olissar L, S everson R K et al. S oft-tissue sarcom a a n d n o n -H o d g k in 's lym phom a in relatio n to p h enoxyherbicide and chlorinated phenol exposure in w estern W ashington. JN C I 1987; 78:899-910. 2 8 . K a n g H , E z in g e r F, B reslin P e t al. S o ft-tissu e sa rc o m a and m ilitary service in V ietnam : a case-control study. JN C I 1987; 7 9 :6 9 3 -9 . 29. Z ahm SH , B lair A , H olm es FF, Boysen C D , Robe! RJ, Frau m eni JF Jr. A case-control study of soft-tissue sarcom a. A m J E pidem iol 1989; 1 3 0 :6 6 5 -7 4 . 3 0. B lair A , M alk er H , C a n to r KP, B u rm eister L , W iklund K. C a n cer am ong farm ers. A review. Scand J W ork E nviron H ealth 1 9 8 5 ;1 1 :3 9 7 -4 0 7 . 31. B lair A . Z ahm SH . H erbicides an d c a n c e r a review and discus sion o f m thodologie issues. In: R ecent results in cancer re search, 1990. B e rlin-H eidelberg: Springer-V erlag 1990; 120: 132-45. 3 2. W iklund K, H olm L E . S oft-tissue sarco m a risk in Sw edish agri cultural and forestry w orkers. J N atl C ancer Inst 1986; 76: 2 2 9 -3 4 . 33. Sm ith A H , Pearce N E , F isher D O , G iles H J, Teague CA, H ow ard JK . Soft-tissue sarcom a and exposure to phenoxyherbicides an d chloro p h en o ls in N ew Z ealan d . J N atl C ancer Inst 1984; 73: 1111-7. 34. B alarajan R, A cheson E D . Soft-tissue sarcom as in agriculture and forestry w orkers. J E pidem iol C om m unity H ealth 1984; 3 8 :1 1 3 -6 . 3 5 . M c C ra w D S , J o y n e r R E , C o le P. E x c e s s le u k e m ia in a re fin e ry population. J O ccup M ed 1985; 2 7 :2 2 0 -2 . 36. E riksson M , H ardell L, A dam i H O . E xposure to dioxins as a risk factor for soft-tissue sarcom a: a population-based casecon tro l study. J N atl C a n cer Inst 1990; 8 2 :4 8 6 -9 0 . 37. K elly JP, R o sen b erg L , K au fm an DW, S h ap iro S. R eliability of p erso n al interview d a ta in a h o sp ital-b ased case-co n tro l study. A m J E pidem iol 1990; 131: 7 9 -9 0 . C orrespondence to: Dr. S. Franceschi Div. E p id em io lo g y C en tro di R iferim ento O ncologico V. P e d e m o n ta n a O c c . 33081 A viano, Italy t ; C- ' *,? \J IN THE CIRCUIT COURT OF THE NINTH JUDICIAL CIRCUIT OF FLORIDA, IN AND FOR ORANGE COUNTY. CASE NO: Cl 89-8657 DIVISION: 32 THOMPSON KIMBERLY MOYER, as Personal Representative of the Estate of ROBERT W. MOYER, II, deceased. Plaintiff, v s. DOW CHEMICAL COMPANY, et al., Defendants. ___________________________________________ / NOTICE OF TAKING DEPOSITION *DUCES TECUM TO: TO ALL ATTORNEYS ON ATTACHED MAILING LIST PLEASE TAKE NOTICE that the undersigned attorneys will take the deposition of: NAME AND ADDRESS: David Eaton, Ph.D. c/o Alan Wagner, Esq. Carlton, Fields, Ward, et al., First Florida Bank Building, P. O. Drawer 190 Tallahassee, FL 32302 DATE AND TIME: December 15, 1992 at 9:30 a.m. PLACE OF TAKING: Larsen & Smith, Inc., Court Reporters, 1325 4th Ave., Suite 1740, Pudget Sound Plaza, Seattle, WA. 98101 (206/623-6717) upon oral examination before Larsen & Smith, Inc., Court Reporters, Notary Publics, or any other officer authorized by law to take depositions in the State of Washington. The oral examination is being taken for the purpose of discovery, for use at trial, or for such other purposes as are permitted under the applicable Statutes or Rules of the Court. DUCES TECUM See Subpoena marked as Exhibit "A" and attached hereto. 19206 I HEREBY CERTIFY that a true copy of the foregoing has been furnished to the attorneys listed on the attached mailing list by U. S. Mail, this the > day of October, 1992. SCHULER, WILKERSON, HALVORSON & WILLIAMS, P.A. Attorneys for Plaintiff Barristers Building - Suite 4-D 1615 Forum Place West Palm Beach, FL. 33401 (407) 689-8180 cc: Court Reporter BY: Richard D. Schuler FBN 158226 19207 MOYER V . DOW CHEMICAL C O . . et al LIST OF COUNSEL ROGER LUTZ, ESQUIRE Holland & Knight Attorneys for HELENA CHEMICAL COMPANY P. 0. Box 1526 800 North Magnolia Avenue Penthouse A Orlando, FL 32802 (407) 425-8500 Fax: (407) 423-3397 and ROGER LUTZ, ESQUIRE Lutz, Webb, Bobo & Baitty Attorneys for HELENA CHEMICAL COMPANY 1 Sarasota Tower, 5th Floor 2 North Tamiami Trail Sarasota, FL 34236 (813) 951-1800 Fax: (813) 366-1603 ALAN WAGNER, ESQUIRE Carlton, Fields, Ward, Emmanuel, Smith & Cutler, P.A. Attorneys for DOW CHEMICAL COMPANY & OCCIDENTAL 215 So. Monroe Street, Suite 500 First Florida Bank Building Tallahassee, FL 32301 (904) 224-1585 Fax: (904) 222-0398 DANIEL C. JOHNSON, ESQUIRE Carlton, Fields, Ward Emmanuel, Smith & Cutler, P.A. Co-Counsel for DOW CHEMICAL COMPANY & OCCIDENTAL Post Office Box 1171 Orlando, FL 32802 (407) 849-0300 Fax: (407) 648-9099 PAUL T. REID, ESQUIRE Popham, Haik, Schnowbrich & Kaufman, Ltd. Attorneys for MONSANTO COMPANY and SOUTHERN MILL CREEK PRODUCTS C O . , INC. 4100 One Centrust Financial Center 100 S.E. 2nd Street Miami, FL 33131 (305) 530-0050 Fax: (305) 530-0055 19208 NLM Search - Dioxins and human cancer risk, 11/15/92 1 AN 92404676. 9211 A. AU Nessel-C-S. Butler-J-P. Post-G-B. Held-J-L. Gochfeld-M. Gallo-M-A. Tl Evaluation of the relative contribution of exposure routes in a health risk assessment of dioxin emissions from a municipal waste incinerator. 90 J-Expo-Anal-Care-Environ-Epidemiol. 1991 Jul. 1(3). P 283-307. 2 AN 91087887. 921OU. AU Fingerhut-M-A. Halperin-W-E. Marlow-D-A. Piacitelli-L-A. Honchar-P-A. Sweeney-M-H. Greife-A-L. Dill-P-A. Steenland-K. Suruda-A-J. Tl Cancer mortality in workers exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin [see comments] CM Comment in: N-Engl-J-Med. 1991 Jan 24. 324(4). P 260-2. Comment in: N-Engl-J-Med. 1991 Jun 20. 324(25). P 180912. SO N-Engl-J-Med. 1991 Jan 24. 324(4). P 212-8. 3 AN 92016944. 9210U. AU Manz-A. Berger-J. Dwyer-J-H. Flesch-Janys-D. Nagel-S. Waltsgott-H. Tl Cancer mortality among workers in chemical plant contaminated with dioxin [see comments] CM Comment in: Lancet. 1991 Nov 30. 338(8779). P 1393. Comment in: Lancet. 1991 Dec 21-28. 338(8782-8783). P 1592. 90 Lancet. 1991 Oct 19. 338(8773). P 959-64. 4 AN 92300798. 9208A. AU Paustenbach-D-J. Wenning-R-J. Lau-V. Harrington-N-W. Rennix-D-K. Parsons-A-H. Tl Recent developments on the hazards posed by 2,3,7,8-tetrachlorodibenzo-p-dioxin in soil: implications for setting risk-based cleanup levels at residential and industrial sites. Ex. 5 . Osta Witness T X JOANNE LEATiOTA 192C9 RF REVIEW ARTICLE: 230 REFS. 90 J-Toxicol-Environ-Health. 1992 Jun. 36(2). P 103-49. 5 AN 92202116. 9206A. AU Ames-B-N. Gold-L-S. Tl Natural chemicals, synthetic chemicals, risk assessment, and cancer. SO Int-Symp-Princess-Takamatsu-Cancer-Res-Fund. 1990. 21. P 303-14. 6 AN 92198578. 9206A. AU Johnson-E-S. Tl Human exposure to 2,3,7,8-TCDD and risk of cancer. RF REVIEW ARTICLE: 121 REFS. 90 Crit-Rev-Toxicol. 1991. 21(6). P 451-63. 7 AN 92016958. 9205U. AU Saracci-R. Kogevinas-M.Bertazzi-P-A.Bueno-de-MesquitaB-H. Coggon-D. Green-L-M. Kauppinen-T. L'Abb'e-K-A. Littorin-M. Lynge-E. et-al. Tl Cancer mortality in workers exposed to chlorophenoxy herbicides andchlorophenols [see comments] CM Comment in: Lancet. 1991 Nov 30. 338(8779). P 1392-3. 90 Lancet. 1991 Oct 26. 338(8774). P 1027-32. 8 AN 92139450. 9204A. AU Smith-A-H. Patterson-D-G-Jr. Warner-M-L. MacKenzie-R. Needham-L-L. Tl Serum 2,3,7,8-tetrachlorodibenzo-p-dioxin levels of New Zealand pesticide applicators and their implication for cancer hypotheses. SO J-Natl-Cancer-lnst. 1992 Jan 15. 84(2). P 1048. 9 AN 92134764. 9204A. AU Paustenbach-D-J. Sarlos-T-T. Lau-V. Finley-B-L. JeffreyD-A. Ungs-M-J. Tl The potential inhalation hazard posed by dioxin contaminated soil. 19210 90 J-Air-Waste-Manage-Assoc. 1991 Oct. 41(10). P 1334-40. 10 AN 92054513. 9201 A. AU Bertazzi-P-A. Tl Long-term effects of chemical disasters. Lessons and results from Seveso. 90 Sci-Total-Environ. 1991 Jul 1. 106(1-2). P 5-20. 11 AN 92046162. 9201 A. AU Keenan-R-E. Paustenbach-D-J.Wenning-R-J. Parsons-A-H. Tl Pathology rvaluation of the Kociba et al. (1978) bioassay of 2.3.7.8- TCDD: implications for risk assessment. SO J-Toxicol-Environ-Health. 1991 Nov. 34(3). P 279-96. 12 AN 92006613. 9112A. AU Bock-K-W. Tl Roles of UDP-glucuronosyltransferases in chemical carcinogenesis. RF REVIEW ARTICLE: 163 REFS. SO Crit-Rev-Biochem-Mol-Biol. 1991. 26(2). P 129-50. 13 AN 91335193. 9109A. AU Gough-M. Tl Human health effects: what the data indicate. RF REVIEW ARTICLE: 92 REFS. 90 Sci-Total-Environ. 1991 May 1. 104(1-2).P 129-58. 14 AN 91305603. 9108A. AU Tollefson-L. Tl Use of epidemiology data to assess the cancer risk of 2.3.7.8- tetrachlorodibenzo-p-diox in. RF REVIEW ARTICLE: 51 REFS. 90 Regul-Toxicol-Pharmacol. 1991 Apr. 13(2). P 150-69. 15 AN 91290820. 9108A. AU Conolly-R-B. Andersen-M-E. Tl Biologically based pharmacodynamic models: tools for toxicological research and risk assessment. RF REVIEW ARTICLE: 52 REFS. 90 Annu-Rev-Pharmacol-Toxicol. 1991. 31. P 503-23. 16 AN 90134015. 9003A. AU Fries-G-F. Paustenbach-D-J. Tl Evaluation of potential transmission of 2,3,7,8tetrachlorodibenzo-p-diox in-contaminated incinerator emissions to humans via foods. RF REVIEW ARTICLE: 144 REFS. 90 J-Toxicol-Environ-Heaith. '1990. 29(1). P 1-43. 17 AN 90189211. 9004A. AU Eriksson-M. Hardell-L. Adami-H-O. Tl Exposure to dioxins as a risk factor for soft tissue sarcoma: a population-based case-control study [see comments] CM Comment in: J-Natl-Cancer-lnst. 1990 Nov 21. 82(22). P 1785-6. 90 J-Natl-Cancer-lnst. 1990 Mar 21. 82(6). P 486-90. 18 AN 89391595. 891OA. AU Fitzgerald-E-F. Weinstein-A-L. Youngblood-L-G. Standfast S-J. Melius-J-M. Tl Health effects three years after potential exposure to the toxic contaminants of an electrical transformer fire. 90 Arch-Environ-Health. 1989 Jul-Aug. 44(4). P 214-21. 19 AN 90035237. 8912A. AU Anderson-L-M. Jones-A-B. Miller-M-S. Chauhan-D-P. Tl Metabolism of transplacental carcinogens. RF REVIEW ARTICLE: 111 REFS. 90 lARC-Sci-Publ. 1989. (96). P 155-88. .20 AN 90094743. 9002A. AU Dunnette-D-A. Tl Assessing risks and preventing disease from environmental chemicals. 90 J-Community-Health. 1989 Fall. 14(3). P 169-86. 21 AN 89321231. 8908A. AU Nuzzo-F. Stefanini-M. Tl Monitoring of genetic instability in subjects with increased risk of cancer. SO Ann-lst-Super-Sanita. 1989. 25(1). P 81-9. 22 AN 89072219. 8901 A. AU Gough-M. Tl Science policy choices and the estimation of cancer risk associated with exposure to TCDD. SO Risk-Anal. 1988 Sep. 8(3). P 337-42. 23 AN 88252393. 8808A. AU Hubert-Habart-M. Tl [Is dioxin carcinogenic? A re-evaluation of the risk (editorial)] SO Bull-Cancer-(Paris). 1988. 75(4). P 339-40. 24 AN 88239759. 8807A. AU Olson-L-J. Anderson-H-A. Jones-V-B. Tl Landspreading dioxin-contaminated papermill sludge: a complex problem. SO Arch-Environ-Health. 1988 Mar-Apr. 43(2). P 186-9. 25 AN 88077739. 8802A. AU Thomas-T-L. Tl Mortality among flavour and fragrance chemical plant workers in the United States. SO Br-J-Ind-Med. 1987 Nov. 44(11). P 733-7. 26 AN 88069559. 8712A. AU Smith-A-H. Tl Infant exposure assessment for breast milk dioxins and furans derived from waste incineration emissions. SO Risk-Anal. 1987 Sep. 7(3). P 347-53. 27 AN 87205291. 8706A. AU Shu-H-P. Paustenbach-D-J. Murray-F-J. Tl A critical evaluation of the use of mutagenesis, carcinogenesis, and tumor promotion data in a cancer risk assessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin. SO Regul-Toxicol-Pharmacol. 1987 Mar. 7(1). P 57-88. 28 AN 87192346. 8706A. AU Sielken-R-L-Jr. Tl Quantitative cancer risk assessments for2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD). SO Food-Chem-Toxicol. 1987 Mar. 25(3). P 257-67. NLM Search (1987-present) 12/9/92 Key words: Sarcoma and herbicides 1 C / AU ^ TI SO 92322548. 9209A. Franceschi-S. Serraino-D. Risk factors for adult soft tissue sarcoma in northern Italy. Ann-Oncol. 1992 Apr. 3 Suppl 2. P S85-8. 2 AN AU yy / Tl (v RF SO 92315979. 9209A. Ibrahim-M-A. Bond-G-G. Burke-T-A. Cole-P. Dost-F-N. Enterline-P-E. Gough-M. Greenberg-R-S. Halperin-W-E. McConnell-E. et-al. Weight of the evidence on the human carcinogenicity of 2,4-D. REVIEW ARTICLE: 49 REFS. Environ-Health-Perspect. 1991 Dec. 96. P 213-22. 3 //AU ^ Tl SO 92215709. 9206A. Smith-J-G. Christophers-A-J. Phenoxy herbicides and chlorophenols: a case control study on soft tissue sarcoma and malignant lymphoma. Br-J-Cancer. 1992 Mar. 65(3). P 442-8. 4 AN AU Tl / 7 cm J SO 92016958. 9205U. Saracci-R. Kogevinas-M. Bertazzi-P-A. Bueno-de-Mesquita-B-H. Coggon-D. Green-L-M. Kauppinen-T. L'Abb'e-K-A. Littorin-M. Lynge-E. et-al. Cancer mortality in workers exposed to chlorophenoxy herbicides and chlorophenols [see comments] Comment in: Lancet. 1991 Nov 30. 338(8779). P 1392-3. Lancet. 1991 Oct 26. 338(8774). P 1027-32. 5 AN 92162959. 9205A. AU Serraino-D. Franceschi-S. La-Vecchia-C. Carbone-A. Tl Occupation and soft-tissue sarcoma in northeastern Italy. SO Cancer-Causes-Control. 1992 Jan. 3(1). P 25-30. 6 AN 92160755. 9204A. AU Holly-E-A. Aston-D-A. Ahn-D-K. Kristiansen-J-J. V? -1- TI Ewing's bone sarcoma, paternal occupational exposure, and other factors. 90 Am-J-Epidemiol. 1992 Jan 15. 135(2). P 122-9. 7 AN 91335193. 9109A. 7 AU Gough-M. Tl Human health effects: what the data indicate. RF REVIEW ARTICLE: 92 REF9. 90 9ci-Total-Environ. 1991 May 1. 104(1-2). P 129-58. 91223003. 9106A. Green-L-M. A cohort mortality study of forestry workers exposed to phenoxy acid herbicides. 90 Br-J-Ind-Med. 1991 Apr. 48(4). P 234-8. 9 /\N i AU Tl 90 91197712. 9105A. Coggon-D. Pannett-B. Winter-P. Mortality and incidence of cancer at four factories making phenoxy herbicides. Br-J-Ind-Med. 1991 Mar. 48(3). P 173-8. 91186874. 9105A. 9chwartzbaum-J-A. George-9-L. Pratt-C-B. Davis-B. An exploratory study of environmental and medical factors potentially related to childhood cancer. Med-Pediatr-Oncol. 1991. 19(2). P 115-21. 91140653. 9103A. Vineis-P. Faggiano-F. Tedeschi-M. Ciccone-G. Incidence rates of lymphomas and soft-tissue sarcomas and environmental measurements of phenoxy herbicides. J-Natl-Cancer-lnst. 1991 Mar 6. 83(5). P 362-3. 12 .A N *1 e AU > Tl 90 91211646. 9105A. Vineis-P. 9ettimi-L. 9eniori-Costantini-A. [Exposure to agricultural chemicals and oncogenic risk] Med-Lav. 1990 9ep-Oct. 81(5). P 363-72. -2- 19218 13 AN 91147259. 9104A. AU Johnson-C-C. Feingold-M. Tilley-B. Tl A meta-analysis of exposure to phenoxy acid herbicides and chlorophenols in relation to risk of soft tissue sarcoma. SO Int-Arch-Occup-Environ-Health. 1990. 62(7). P 513-20. 14 M l 90386163. 9010A. AU Pearce-N. Reif-J-S.' Tl Epidemiologic studies of cancer in agricultural workers. RF REVIEW ARTICLE: 87 REFS. SO Am-J-Ind-Med. 1990. 18(2). P 133-48. 15 AN 90201594. 9005A. AU Johnson-E-S. Tl Association between soft tissue sarcomas, malignant lymphomas, andphenoxy herbicides/chlorophenols: evidence from occupational cohort studies. RF REVIEW ARTICLE: 124 REFS. SO Fundam-Appl-Toxicol. 1990 Feb. 14(2). P 219-34. 16 ^N 90352515. 9009A. 'AU Wingren-G. Fredrikson-M. Brage-H-N. Nordenskjold-B. Axelson-O. Tl Soft tissue sarcoma and occupational exposures. SO Cancer. 1990 Aug 15. 66(4). P 806-11. 17 AN 89106120. 8903A. AU Uehara-Y. Murakami-Y. Sugimoto-Y. Mizuno-S. Tl Mechanism of reversion of Rous sarcoma virus transformation by herbimycin A: reduction of total phosphotyrosine levels due to reduced kinase activity and increased turnover of p60v-src1. SO Cancer-Res. 1989 Feb 15. 49(4). P 780-5. 18 AN 91046840. 9012A. AU Kelly-S-J. Guidotti-T-L. Tl Phenoxyacetic acid herbicides and chlorophenols and the etiology of lymphoma and soft-tissue neoplasms. RF REVIEW ARTICLE: 89 REFS. 192 3- - 90 Public-Health-Rev. 1989-90. 17(1). P 1-37. 19 AN 89171721. 8905A. AU Bond-G-G. Bodner-K-M. Cook-R-R. Phenoxy herbicides and cancer: insufficient epidemiologic evidence for a causal relationship. RF REVIEW ARTICLE: 63 REF9. 90 Fundam-Appl-Toxicol. 1989 Jan. 12(1). P 172-88. 20 AN X AU V Tl 90 89295246. 8908A. Ciccone-G. Vineis-P. Inter-rater agreement in the assessment of occupational exposure to herbicides. Med-Lav. 1988 9ep-Oct. 79(5). P 363-7. 21 M* V$ 90 89157076. 8904A. Wiklund-K. Dich-J. Holm-L-E. 9oft tissue sarcoma risk in Bwedish licensed pesticide applicators. J-Occup-Med. 1988 Oct. 30(10). P 801-4. 22 m 88156355. 8804A. AU Axelson-O. Tl Pesticides and cancer risks in agriculture. RF REVIEW ARTICLE: 88 REF9. 90 Med-Oncol-Tumor-Pharmacother. 1987. 4(3-4). P 207-17. 23 AN / / Tl RF 90 88001947. 8711A. Lynge-E. 9torm-H-H. Jensen-O-M. The evaluation of trends in soft tissue sarcoma according to diagnostic criteria and consumption of phenoxy herbicides. REVIEW ARTICLE: 38 REF9. Cancer. 1987 Oct 15. 60(8). P 1896-901. 24 AN 87199597. 8706A. Q AU Woods-J-9. Polissar-L. 5everson-R-K. Heuser-L-9. Kulander-B-G. Tl 9oft tissue sarcoma and non-Hodgkin's lymphoma in relation to phenoxyherbicide and chlorinated phenol exposure in western / Washington. 1S218 4- - 90 J-Natl-Cancer-lnst. 1987 May. 78(5). P 899-910. 1 25 AN Y AU 4 Tl 90 88046789. 8711 A. Xue-S-Z. Health effects of pesticides: a review of epidemiologic research from the perspective of developing nations. Am-J-Ind-Med. 1987. 12(3). P 269-79. 26 AN 87206089. 8706A. AU Vineis-P. Terracini-B. Ciccone-G. Cignetti-A. Colombo-E. Donna-A. Maffi-L. Pisa-R. Ricci-P. Zanini-E. et-al. Tl Phenoxy herbicides and soft-tissue sarcomas in female rice weeders. A population-based case-referent study. 90 9cand-J-Work-Environ-Health. 1987 Feb. 13(1). P 9-17. ( 192^9 1 HUMAN HEALTH EFFECTS OF INCREASED HERBICIDE USE: PERCEIVED v. ACTUAL RISKS Synthetic chemicals used to control unwanted plant growth have become an indispensable part of commercial agriculture, forest management and home gardening. The use of synthetic organic chemicals as herbicides is a relatively recent pheno menon; prior to the early 1940's, most herbicide application was in the form of inorganic salts such as lead arsenate. Since the development and marketing of the phenoxy acid herbicides in 1944, herbicide use in the U.S. has grown more rapidly than any other class of chemical pesticide. In 1963, annual sales of herbicides were on the order of 100 million pounds; by 1975 sales increased to 645 million pounds, surpassing the total usage of insecticides in the United States. By 1964, nearly twice as much of the crop area in the United States was treated with herbicides as with insecticides or fungicides. Although the ecological and health risks associated with widespread insecticide use became a prime issue of the environmental movement in the 1960's, herbicides as a chemical class were generally viewed by both scientists and the general public as rather innocuous compounds. Indeed, classical measures of toxicity such as acute lethality (LD,-q ) are univer sally much lower for herbicides than for many widely used insecticides. The widescale use of the now notorious herbicide mixture, "Agent Orange," in the jungles of Viet Nam brought to focus potential adverse health effects that might result from exposure to the phenoxy acid herbicides 2,4-D and 2,4,5-T. Veterans returning from Viet Nam began to attribute a wide variety of ill effects to their exposures to Agent Orange years earlier. Their suspicions of toxicity from Agent Orange were fueled by the discovery that many batches of this herbicide mixture were conta minated with the highly toxic chemical, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). TCDD is considered the most toxic man-made chemical known, and is capable of producing cancer and birth defects in experimental animals at very low doses compared to many other known carcinogens and teratogens. Moreover, in contrast to the active herbicide components which are readily eliminated from the body and are rapidly broken down in the environment, TCDD is highly persistent and tends to accumulate both in the environment and the human body. Citizen complaints of ill health effects from herbicide spraying began to appear in he early 1970's, particularly in the forest lands of the Pacific Northwest, where aerial application of phenoxy acid herbicides is widespread. Of particular concern were reports of increased occurences of miscarriages and birth defects among rural residents of these areas. Based on a pre liminary finding of increased miscarriage rates in the Alsea, Oregon area which appeared to correlate wth 2,4,5-T application, 19220 El . C\ .jfcts.. Witness JOANNE LEATCIA 2 the EPA ordered an emergency suspension of most uses of 2,4,5-T. However, 2,4-D use was not affected by this ruling. this regu latory action, hotly contested by the manufacturers of 2,4,5-T, has transformed what was perceived to be a scientific question to a highly emotional debate between industry and citizens, with scientists mostly caught in the middle. The dilemma facing many scientists, particularly those trained in toxicology, is the inability of traditional scientific methods to discern between actual versus perceived risks with the degree of certainty that the scientific community demands before recognizing such risks as "fact". Traditional regulatory approaches governing the' sa-fety of chemicals have tended to rely upon the addage of "innocent until proven guilty", accepting as proof only those data which meet the predefined criteria of "statistical significance". To the layperson, however, an n of 1 is generally adequate proof that x caused y, and there is little appreciation for the "scientific method". How then is the toxi cologist to deal with the public's questions regarding the health effects of chemicals, when, in the eyes of the layperson, circum stantial evidence abounds that x caused y, yet scientifically there is no statistical proof that such a relationship exists. The case of herbicides is an excellent example with which to study this dilemma. What are the perceived risks of herbicides? Obviously this depends upon the individual's experiences and understanding of chemical hazards. There is probably no other class of chemicals for which a greater range in perceived risks exist than the phenoxy acid herbicides. In a recent book entitled "A Bitter Fog", author Carol Van Strum graphically depicts her perception of the damage that herbicides have on human health and the environment. Speaking of an incident in which her children were apparently directly sprayed by a roadside crew with a combination of 2,4-D and 2,4,5-T she states "The children did no more fishing that spring day in 1975. They were too upset, and their eyes, noses and mouths burned. I brought them to the house and washed them, changed their clothes, gave them soda to rinse their mouths, but the bitter taste and burning wouldn't go away. By evening, all were sick with nausea, vomiting, cramps, headaches, diarrhea. One of the boys had an uncontrollable bloody nose...". She continued, "The day after the spray truck came, we went back to the river and saw that it had changed. The leaves of alder and willow on both sides of the river wilted, their new growth in unnatural spirals. The water had an oily scum on it. Floating or beached on the shore were dead crayfish and trout, and the sodden bodies of two merganser ducklings. Caddis larvae and snails that usually traced patterns in the silt on the river bottom were dead. Songbird nests along the shore were empty, except for one, in which a hermit thrush still sat, dead." Later, she describes a second episode: "Within 24 hours we were sick again, with the same symptoms as before. This time, there was a further complication, uterine bleeding in both myself and my daughter, who had not yet begun to menstruate at that 19221 3 time.... In the next weeks we had an epidemic of deformities among the chicks, ducklings, and' goslings that hatched on the farm: crossed beaks, stunted or missing wings, toes or whole feet on backward, and stunted growth." Numerous magazine and newspaper articles have also described the perceived impacts of herbicide use on human health and the environment in a similarly graphic manner. Thus, there are numerous anecdotal reports describing the ability of herbicides to produce a wide range of adverse health effects, including neurological effects, birth defects, reproductive disturbances, cancer and a variety of 'subjective' effects such as nausea, headache and malaise. If, in fact, these reports are true, how could such highly toxic, dangerous chemicals be allowed for such widespread use for nearly 40 years? This, of course, is the question many citizens have been asking of the EPA, the chemical industry and the scientific community after hearing of such reports. Many in the scientific community, upon hearing of such reports, have responded with a great deal of skepticism, citing the bulk of the scientific literature on phenoxy acid herbicides that suggests that such widespread, severe toxicologic responses could not occur at normal application rates of phenoxy acid herbicides. However, extrapolation of laboratory data to predict what might happen in the "field" is difficult, at best, and scientists should recog nize the limitations of classical laboratory studies for predic ting actual effects. What is known about the actual health risks associated with long-term, low level exposure to dioxins and/or phenoxy acid herbicides? The answer to even this question is exceedingly complex. There are some data in laboratory animals which indicate that 2,4,5-T can cause certain birth defects at rela tively high doses - that is, at doses which are sufficiently high enough to produce evidence of toxicity to the mother (generally greater than 50 mg/kg/day). Whether the birth defects which result are a result of the chemical per se, or are an indirect effect of maternal toxicity, is unclear. TCDD, on the other hand, is an extremely potent teratogen, capable of inducing cleft lip and palate in fetuses at doses in the order of several nanograms/kg/day, or one one-millionth the effective dose for 2,4,5-T listed above. However, extrapolation of such results to humans is extremely difficult because of large differences in species susceptability to TCDD. There is a five thousand fold difference in the acute lethality of TCDD between guinea pigs and hamsters. We have little idea where humans fit on this scale, although monkeys appear to be relatively sensitive to both the acute toxic and teratogenic effects of TCDD. A second con founding factor in predicting human effects from exposure to phenoxy acid herbicides is the wide range in levels of TCDD contamination of different herbicide formulations. Certainly the single most important factor in assessing risk from a chemical is to establish the dose, or amount, of exposure. In virtually every human instance of herbicide exposure there has been no information as to the level of dioxin contamination, or an accurate indication as to the extent of exposure to the herbicide 1 Qo o o 4 itself. Factors such as the amount of body surface area exposed, the percent of the dose deposited on the skin that is actually absorbed, the concentration of herbicide in the spray, the con centration of herbicide in the air, the duration of breathing such vapors, and the frequency and duration of exposure over time are paramount to an accurate estimation of exposure. Several epidemiologic studies on workers exposed occu pationally to phenoxy acid herbicides have identified an increased risk of developing certain types of cancer (soft tissue sarcoma, stomach cancer and lymphomas). These studies have been challenged by other scientists because of mis-identification of pathologic tumor type and other procedural problems. There have been several other epidemiologic studies which have not found any correlation between herbicide exposure and an increased risk for cancer. Thus, there appears to be inadequate and conflicting data in the scientific literature regarding the risk of developing cancer from phenoxy acid herbicide exposure. Recent studies on residents of Seveso, Italy, where a chemical manufacturing plant explosion resulted in extraodinarily high levels of contamination of the environment with TCDD, revealed an apparent increase in abnormal birth outcomes in women residing in the area, as well as a dramatic decline in the birth rate in the year following the accident. Whether the decline in birth rate was a result of an increase in contraceptive use (unlikely), an increase in elected abortions because of the incident, an actual toxicological effect from the dioxin (increase in miscarriages/spontaneous abortions) or some other factor is not known. Because the exposure to dioxin (TCDD) in this episode was several orders of magnitude above that which would be encountered in other environmental exposures, such as from the application of herbicides in forest management, or from living in the dioxin contaminated environment of Times Beach, Missouri, it is difficult to extrapolate the human exposure data from Seveso to these comparatively low level situations. We continue to find new areas that are contaminated with dioxin, and this has generated a substantial amount of concern among the general public, the scientific community and regulatory agencies. The levels of dioxin reported at most of these sites are generally in the order of a few parts per billion (ppb). One ppb is roughly equivalent to a single drop of pure chemical diluted in 13 thousand gallons of water! Never-the-less, such low levels of TCDD are capable of producing biologic effects in some experimental animals if present in the diet for a prolonged period of time. The perceived risks of dioxin in the environment are very great. Whether the actual risks approach the perceived risks is an unanswered and much debated question. The concern over phenoxy acid herbicides and dioxin has grown to include all herbicides in the eyes of many people. Application of any herbicide is often greated with resistence by nearby residents, regardless of the nature of the chemical being 19223 used. While the magnitude of the concern is often out of pro portion with the risk, it is important that scientists maintain an open mind and consider all possibilities of detrimental effects of herbicide use, recognizing the limitations of current scientific methodologies to accurately predict such effects. The lessons learned from asbestos, ionizing radiation, DDT, thali domide and numerous other substances, where traditional scienti fic methods failed to adequately predict the risks to human health and/or the environment before damage was done, should be kept in mind by the scientist who readily dismisses perceived risks as impossible because they are not consistent with the available scientific data. However, it is equally important that scientific data not be used out of context to project irrational fears of potential adverse health effects from chemical applica tion. Many of the perceived risks of herbicides are the result of unwarranted extrapolation of a single, unconfirmed laboratory report as pertinent to any potential human exposure situation. The harm generated by such predictions may be more damaging than the chemical itself. The mental anguish and physical stress placed on a pregnant woman who is told that herbicides will be sprayed in her neighborhood is probably more harmful to the developing child, if the woman perceives that the chemical is likely to harm her baby, than the chemical itself. Scientists must make an effort to put risks in perspective, utilizing all available scientific information but recognizing the limitations and deficiencies in each study. More emphasis should be placed on establishing relative risk rather than absolute risk. The general public often does not understand the significance of, for example, an increased risk of cancer of one in ten thousand if chemical x is present as a contaminant in the diet at 1 ppm over a lifetime of exposure. They are more likely to understand the risk if the same risk were expressed in relat ive terms; for example, as being roughly equivalent to the increased risk of developing cancer from eating x ounces of charcoal broiled steak over a lifetime, or from sitting in the sun for x many hours. Although such comparisons are crude estimations at best, the magnitude of the error is not likely to be greater than the error of determining absolute risk. The risks to human health that might result from increased herbicide use in "low-till" farming are probably minimal when put in perspective with other risks encountered daily. However, such an opinion must be qualified for each circumstance: the nature of the chemical; who is at risk for exposure (the applicator, the farmers family, nearby neighbors, etc), and at what level; what other sources of exposure they might encounter; what is known about the toxicologic effects of the chemical, etc. There will undoubtedly be controversy over the use of herbicides for many years to come. Hopefully, the scientific community will respond with reasoned concern, utilizing science to its fullest, yet recognizing its limitations, in an attempt to put the risks of herbicide use in perspective. 8 n x . f ost/it'W of fat* fataby of C o "^ 0/Lr*y / ? V Y - - t'- f USc. o f p h < M o \(y CfCiJ{ h<Lrbi < 5 0 - C2oia/ Csh .tnt c a , / ihMs (\4e^h^\y C o~1~ah%t*) &n1~ 1151' fc 'A c.*f>-Za5<^ *" c2)C*j\t *J y3ws*n /n g^Jl 161 - `t'C.OQ idfa*ttf'iL$. C o o ^ /h i/utr*/' ^<~ o c c C^/at-a^no 1 6 2 . - Q p u t/n J *** Rc^hcb hcbhtQ. b-P'>nS / < ? 7 - C.l\i<M JLcsLm<x fegato h l & J 1^,4 a i H C O-Q 1*16*11 - R jL f> 0 $ b y O i'to .h i 4 9 , ( - ( z . c f c i /*i Y Cfc A /a/iyi an t . o o f t j K ' J / f ? - - . & i o r i - e A n , i R ^ n a o ^ o k L a .6 V <T-T r {'j u ^ J o ^ J i n ^ J y / f ? - /-Uv *f >>4 '* ^ c a e /u '^ /f? - j . } q o * $ b i d o n o . - o**J " 0 / /^ '^ P JC U L i'c. h e J * * * J *~ m ?- 7Y ' C I t i * - T ff/x <* W P S >j uM - C 0 h < jr )r t ,z iu r t 0 '1 <* r' ^ Q ( f t n ( r 0 l) M in in g $ / ? 7 - 5<Lux. j ' i f J ' y Q c a c J 1 1 1 b - c j g j foauit. ' / j e J Q r^ n jc : /lVa.Js P&adiy fo 1 1 Y - fcP /J $5*1 z. s R p f R lo* - Witness L L 22L .-----JOANNE LEATiOTA 1922!i - ffl *% < * *'< f o ' c c i h h t c i g L S / *i f l / s < m . j O ' 4 7 ^ ^ ' T P f l /5 i t i x j J L M J J ~ f i V i O y S u l p x S i t i O K o f l ^ f 1322B 339 e u S f 5THCC* S a n F R A N C ISC O . Ca l i f o r n i a A iO -*-287e Fa C & ' m i l C <a (S 772--26 ' T LE MONC (AiS^_77fc > 6 0 0 0 5 2 5 u n i v EOS'TY a v e n u e Pa l C a .T C . Ca '. 'F C R n i a 9 a 3 C - - ' S 0 6 FACS-M.a (iS, 32A-o3fi r t L C R N Q N C |a i I 3 26 - 7 6 0 C HELLER, EHRMAN. WHITE & McAU LIEFE a t t o r n eys A PAttTNCRSmO INClUOlNC RROFCSSlONAL CORPORATIONS 6 1 0 0 COLUMBIA CENTER- 701 FIFTH AVENUE SEATTLE. WASHINGTON 9 8 l 0 - 7 0 9 6 T E LE P H O N E (9 0 6 1 7 - O S O O - FA C S IM ILE (2061 M 7 - 0 6 * 9 November 10. 1989 13 0 0 s. w. r.^Th. a v e n u e P O S T la.n o .. OftEOON 9 7 2 0 * 5 e f a c s im i l e ISC3I 2*.-C5C T E l E P h O n C 15C3 227-7AOO 5 5 5 S G c'Th r t C w C R STfiC7 LOS a n C E l ES, Ca l ' f & R nia & C C ? i- 2 3 0 6 f a c s i m i l e (2.31 fe`A - i 6 6 e TC*Rh ONE (2iji 6 8 fr -0 2 0C L Y N D A L. B R O T H E R S Risk Assessment and Communication Conference Work Group Dear Members: Thank you for attending the meeting in our offices at Heller, Ehrman on November 2, 1989. As promised, enclosed please find a copy of the articles contributed by Chuck Kleeberg. In addition, this letter confirms our next meeting for January 19, 1990 at 5:00 p.m. at the Heller, Ehrman offices. I have contacted Dave Eaton, who has enthusiastically agreed to attend and to talk with us on risk and risk assessment. I look forward to seeing you all on the 19th. If anybody has any information they would like to distribute to the group, please feel free to send it to me at the above address. Also, bring ideas for our next meeting at which we will focus on risk communication. I have also enclosed a copy of the work group mailing list, subject to changes and additions as we go forward. Very truly yours cc: David Eaton, Ph.D Enclosures 1 v 'cT ^ -' C u ^ J s (<=~ / COMBINED PUBLIC BULLETIN BOARDS * * * * * * * * * * * * * * * * ** * v-* * * * * ** * * * * 4 * * T h i s are a combines GENERAL, EMERGENCY, and LE GI SL AT IV E b u l l e t i n s . .. ;S>can < - >MA IN <R>ead i E nt er <D'-el ete HAZARD VS. OUTRAGE; NOW THE PUBLIC SEES HEALTH R I S K iCl'-hoos - T e s e n t a t i on by P e t e r M. Sandman, P h . D ., R u t g e r s U n i v e r s i t y To the ASTHO Annual M e e t i n g , A p r i l 1 9 9 9 . Vail , Colorado. N ot es by: -A. Lament C a r s t e n s , S u p e r v i s o r , Community L i a i s o n S e c t i o n . Dept, of H e a l t h , M a i l Stop: E T - 3 6 . Ol ympia, Wa. 9 8 5 0 4 y11i,,.i-fii T. iTteSsaoer of sk ' - ! h." i-i r- ! i i nvo] f wp ".y.ir-e! i i>a1: 1 . 1 . How does one ge t th e p u b l i c to ta k e r e a l r i s k s s e r i o u s l y and r e a c t when t r u l y t h r e a t e n e d ? How can one c o n v i n c e a p a t h e t i c or diempowered people to be m o ti va te d to take s e 1 -f- pr ot ec t i ve ri\s as ur es ? 2 How does one calm people down? How can one help pe opl e to deal with a n x ie tie s provoked by inadequate information and avo id h y s t e r i a when the r i s k i s not s e r i o u s ? A n x i e t y --p ro du c in q s i t u a t i o n s may not be ha za rd ou s: ha za rd ou s s i t u a t i ons may not pr oduce an;; i e t y . R i s k o f t e n i s viewed as the pro du c t of two f a c t o r s : magnitude - how bad what happens i s X p r o b a b i l i t y - how o f t e n i t happens However, t h i s f o r m u l a m i s s e s an e s s e n t i a l i n g r e d i e n t in p u b l i c r e s p o n s e . That i n g r e d i e n t i s o u t r a g e . To un d er stan d p u b l i c response, i t would be c l e a r e r to c a l l the product of magnitude k probatai1 i t y , " h a z a r d . " R i s k then becomes the sum of hazard p l u s outrage. The p u b l i c aind s c i e n t i s t s d i f f e r i n t h e i r d e f i n i t i o n . S c i e n t i s t s use the word " r i s k " ^g_&^hazard. The p u b l i c responds to a p e r c e i v e d r i s k a s an o u t r a g e . TfTe-'sci ant i f i c mind may say something i s not dangerous but the public mind. i f i t finds something d i s g u s t i n g , respo n ds with a sense of alarm ana outrage. The sense of o u t ra g e i s r e a l and must be d e a l t with. In the lite ra tu re of public r is k assessment, twenty-si u v ar i ab 1 es s ys t ema t i c a 11 y t uc n up*. E 1 even a f t hes e pr ec ur s or s or measures of p u b l i c h e a l t h o u t r a g e r e a c t i o n rare d i s c u s s e d here. Di s t i n o t io n s people use in a s s e s s i n g r i s k , from low -risk' to hi Ch r is k , include: 19228 The cJi stinction b et we en risks that are vo l un t a r y and those which are coerced. This is as true + o f~ c o mm un it ie s as for individuals. Coercion is seen as an assault which creates risk to those receiving-the force. A natural vs. a m a n - c r e a t e d risk. Radon is viewed as an act of God. A tanker oil spill, is seen as an act of man and triggers greater outrage. The familiar vs. the exotic or unfamiliar. Radon col lects in the basement, a fa miliar location. Watching on TV an EPA c le an up by men in s p a c es u i t s c r eat es a qreater 1s v e 1 of anx i e t y . No t m eiTio r a b 1e v s . m e m o r ab 1e . M e m o r ab 1e r e 1a te s to how easy it is to c o n j u r e u d t h o u g h t s of h ow bad things could be or how thing s cou 1d go wrong . N on--memorab 1e e 1ates to those s it ua t i o n s we have lived in without a sense of apprehensi o n . Memorable reactions can be stimulated by imagination or associations as well as real events. Talk of building waste i nc in e ra t o rs c o n j ur e s up in many minds thoughts of be lching industrial smokestacks. Not dreaded vs. a s ens e of dread. AIDS and cancer cause dread. A number of AIDS or cancer dea th s will cause more alarm in a comm uni ty than an equal number of dea th s from other diseases. A 5 5- qal lo n drum marked "mater ial s'1 will cause less anxiety than the same drum and content marked "waste." Know-able vs. not knowable. "Knowabi 1 ity" r e lat es to a number of facto rs such as sc ie nt i f i c knowledge, fami 1iarity, and detectability. Beliefs also may be treated as fact. Debate as to what is knowable may lead to conflict. Fear and ou tr age ina y inc r e as e a s d eb a te d e vslops r s g ar d in q what a risk level re al ly is. The scien ti fic c a p a c i ty to d"tect mi nute amc>unts of sub stances inc re ase s th is d e b a te . R a t es in disc u.s s in g r is k a s ses s m e n t t a ke s a di fferent meaning from rates in standard settings. An agenc y arguing t h a t views of risk are ex a g g e r a t ions may be viewed by the public as d i sinter ested and thereby increase fear and outrage. Co ntr oll ed by the individual (self) vs. c on tro ll ed byothers. Control is determined by who implements, visualize the difference between sharpening the butcher knife and cutti ng the rib roast vs. holding the roast while so meo ne else cuts. In many c o nf li cts between government agencies and communities, the communi ties view themselves as holding the meat while the agency wields the knife. An agency sends mixed, c o n t r a d i c t o r y me ss age s when it tries to control or iimpose and at the same time try to say "don't worry." An agency which has a hi s t o ry of adverse impact on communities and people must acknowledge that hi st or y if it is to reduce dread and other anxietyP r o d uc ing fe e 1 in q s . ^ qqq Being knowle dg eab le is a part of the sense of control and voluntary participation. For an agency to share control means that it must a l l ow the c o m m u n i t y to im plement Isey c o n tr o 11 in q fa c tor s su c h as doi ng th e s tud y t a d e te r m in e need , *desi rabi 1 ity , et al . Fair vs. u n f a i r . One eleme nt of public p e rc e p t io n of fairness is the perception as to distribution of risks and benefits. The belief is that benefits should benefit those sharing the risk arid not skip the r i s k .b e a r e r s . For in s t a n c e , o utr ag e may increase over a. g a rba ge incinerator proposal which includes bringing garbage from outsid e the area. Pu bl i e p e rc e p t i o n of f a i r n e s s may also have a "my risk vs. your risk" element. Mora 1.1y irrel evant vs. mora 1 1v r a 1e v a n t . 0eve 1opi ng public vaiues that some factor,, for example, environmental c o n t a m i n a t i o n , is w r o n g -- EVIL.-- g i ve s moral re 1s 'ance to the issue. Howev er , scme moral 1 y re 1ev-ant. issues have less tradeoffs. The more incensed the public becomes, the higher its sense of risk. When the public become s very angry , it may demand a zero - r isk: so 1u.tion . If the public ag en cy shows no signs of ge t t i ng to sera r is I;, the pub 1 ic b ec ofnes an q r y ag a in. Trusted vs. not trusted. Health agenci es are t r ust ed by the public when they deliver a warning, when they say that something is dangerous. <"They are the e x p e r t s . ") The hea.l th agency may not be trust ed when it says somethi ng is safe. (G th e r s are sayi nq it is not s a f e . > Distrust may be one of the c o mm u ni ty 's chief m e a s u r e s of the size of risk. Good pro ce ss vs. bad process. The proc es ses of interaction between the governmental agency and the comm uni ty may contri bute to the commLini ty 's s e n s e of r i sk a . Openness vs. secrecy. O p e n n e s s reduces a public sense of risk. A g e n ci e s should tell what they will do and what happened. The public is more interested in being told than in what they are told. B. Ack n ow l ed ge prior mi sd e e ds vs. denial and stonewalling. The American public is tolerant of repentant si nner s but dist ru sts those who 1eqal ist ical 1 y de n y thc^t sin has occurred. "We goofed," is a necessary defense for public eqenci es. C. Caring communi cati on vs. jargon and a r g u m e n t . What the speaker termed "technobabble" is distancing language. Five conclusions were presented: 1. Risk means outrage, not Harare!, to people. The public view or risk r e l a t e s to their sense of outrage, not to health impact. 2. Mo one learns much about hazard w hi le ex pe ri e nc i n g a. high level of outrage. S u b s t a n t i v e issues are not the person's isues when -feelings are running strong. Substantive d a t a , such as data on hazard, becomes amm;ini ti on by l o o k i n g at d a ta as s o m e t h i n g to d i st or t in o r de r to pr o ve this c o n c 1u s io n s of o u tr ag e . T h e o n 1y valid, relevant data is that whi ch s upports the ou trag s . 3. Outrage s i m u 11 a n e o u s 1y is a: A . D i s t r a c t i o n -f r om h a z a r d . B, Pr ot ect io n o-f the values we really care about. Outrag e -factor s a r e T-ac t o r s pe a p 1e va 1u e -- r e a 1 1y c ar e a u o u t . T hie y r e -f1ec t th ing s so c ie ty want s taken ser ious 1y . 4. Where hazard is high, the job is to make ou t r a ge higher. For example, wh ile 8 0 - 9 0 % o-f the hazard o-f smoking is to the smokers and on l y 1 0-20'/. to non-smokers, the outrage o-f the passive sm ok e r is r e s p o n s i b l e -for most control measures. MADD is ano ther examp le of translating a serious hazard into serious outrage and getting progress on the problem. 5. Where the hazard is lew, the job of risk c o mm u n ic ati on is not to explain that the risk is low. R a t h e r , the job of com mu n ic a ti on is to keep the level of outra ge low. The job is to find ways to keep the issues at the lowrisk, n o n -ou tra ge sta ge of the eleven f ac tor s listed above. The c o mm un ic a ti on must soeak to the vari a b 1es of o u t r a q e , not just to the haz a r d s . Dr. Denman Scott, A S T H G p r e s id e n t -e l e c t , observ ed thsit the process works. When syringes washed up on the Rhode Island shoreline, his first words were "it's disgusting." He then could say further, "It's disgusting but, incidentally, you will not get AIDS if you step o n one of these n e ed l e s on the beach." JL3 O O 3 1 1- 4 - V xir S eattle C ity L ight M em orandum DATE: June 20, 1989 TO: Distribution List FBOM: Msy Cerstle vis JuliaGreenl) SUBJECT: Electric MagneticField Communications Conference <r j u The attached report tumarlze* highlights from the Electric Magnetic Fields (EMF) Communications Conference sponsored by the Western Systems Coordinating -Council held June 7-9. Of particular Interest are the consents on risk perception and the Importance of addressing the political and societal issues in planning for utility projects. The pivotal role of public Information and Involvement was stressed throughout the conference. I hope that this report will be useful to you in the planning and decision-making process. Julia Greenlee vould be glad to answer any questions or to discuss this report further if you are interested. In light of the articles in the New Yorker magazine the EMF issue may be in for Increased attention and scrutiny. JG:ph Attachment Distribution List Macdonald Saven Taha Garman lockey Valters EMF Workgroup cc: Gerstle Bailey Nelsen Forslund Greenlee --- Klein Lock Sugiyama 531-1(1-811 19222 Quest Ion regarding: o ODC hormone promotion o Cell communication effect o Melatonin and breast cancer o Tumor cell growth o Decreased cancer-killIn ability Other Effect Issues o Reproduction - Electric blanket use - C h i c k embryos (pulse magnetic fields) o Melatonin - Mood, behavior - Seasonal breeding o Cellular effects - Calcium (primarily electric fields) - Communication between cells o Behavior, learning EPA's Position Potential for cancer and behavior changes. Utilities should exercise caution before allowing any increase in exposure. Hew York Studies Several areas of potential concern. Conclnsions o Some nonhazardous field effects and potential for effects on human health or performance. o Several epidemiological studies suggest 60-hertz fields may be weak cancer promoters. o Limited evidence for effects on reproduction, o More long-term research needed. o Trend developing to recommend limiting chronic field exposures. -2- 19233 Morgan's research Involved the following schedule of activities: o Performing scoping studies - o Developing and testing questionnaires o Developing and testing common materials o Performing public perception studies The studies included face-to-face interviews, perception focus groups, and decision focus groups. According to the early results of the research: e o Public perception of the funding source is critical to credibility. o Relative risk is not often realised by public or an especially compelling argument. o Attention to an issue Implies risk. o Reed doubted. This was highlighted as a significant public perception that, was also emphasized by Herbert Thler. Thier encouraged utilities to thoroughly test need within their own organizations and not to approach the public until they can convincingly deaonstrate need. What people need to know, according to Morgan, are basic facts, basic principles, and one or two advanced facts. Granger Morgan is In the process of developing the next step in the research that will focus on strategies and outcomes based on the results of the research on public perception. 8-- arr of Presentation bv Dr. lerbert Thler Dr. Herbert Thier, from the University of California, Berkeley, gave an excellent presentation on Science Education as a precursor to risk communications. Useful EXF risk communication aids should: o Ala toward clear objectives o Define overall risk communication strategy o Establish content o Develop and integrate detailed risk messages and evaluations Correlation is not a synonym for causation; it i s a s o m e w h a t hard distinction for the public to make. 5- r, o u / f /! 19234 Quantitative risk assessment - risk hazard prioritize regulatory action. The acceptance of risk la a political question that requires political action; perception Is reality. In dealing with public perception of EXF, it is Important to define what la the message and who conveys It. Frequently you cowaunicate risk; the public hears certainty. Public perception deteralnes acceptance of risk. Factors that Influence perception are: Leas Risky More Risky Voluntary Controlled Beneficial Familiar Controlled by aelf Fair Mot dreaded Mot memorable Chronic Diffused in time Insediate Rot fatal Involuntary Uncontrollable Mot beneficial Unfamiliar Controlled by others Unfair Dreaded Meaorable Acute ` Focused in time and space Delayed Fatal You need to establish ongoing interaction between you and the community. The utilities' goal is to lnfora the public; the decision making is the prerogative of the public. According to Thier, the only thing that works with the public is the argument of actions taken versus words or comparisons. Have you as a utility minimized the hazards and made improvements? Put another way, the question to ask is "What have you done to make things safer?" Risk management decisions Involve: o Risk identification o Risk analysis o Risk comparison o Resource allocation 6- - .8 3 3 5 When you send the solution Instead of alternatives to the public, the solution becomes the definition of the probleu. You set up a power situation that results In only tap reactions-- either subalt or oppose. You rule out possible alternatives and force the public Into an adversarial relationship. The attention needs to be focused on fundanental Interests. o Take tine to build an understanding In the c o m unity about what the problen Is and go out to the c o m unity early. o Don't trivialize risk. o Moral categories nean nore than risk data. o Policy decisions are seen in terns of being either risky or safe. o Equity and control issues underly nost risk controversies. o Risk decisions are better when the public shares the power. o Risk conaunlcations are easier when emotions are seen as legitlnate. o Public lnvolvenent nust be thoroughly Integrated into declslon-naklng process. There nust be a result iron participating. The public lnpacts decision naklng and the-Utility learns something. Different foruns nust be provided for different public sectors. Public bearings are the least effective way to resolve Issues. Public lnvolvenent nust be done before public hearings. Giant foruns are not effective consensus-building devices. o The role of public lnvolvenent is to build and naintaln relationships with the public. It Is the accumulative effect of relating to the public consistently over tine. Media Relations o Envlronnental risk Is not a big story o Politics are nore newsworthy than science o Reports cover viewpoints, not truth o The risk story is sinpllfied to a dichotomy o Reporters try to personalize the risk story o Claims of risk are usually nore newsworthy than clains of safety o Reporters do their Job with limited expertise on subject and tine 8- - 19236 R IS K CO N FEREN CE WORK GROUP Ms. Lynda L. Brothers Heller, Ehrman, White & McAuliffe 6100 Columbia Center 701 Fifth Avenue Seattle, WA 98104-7098 (206) 447-0900 Ms. Zimmie Caner 1711*5 East Republican Seattle, WA 98112 (206) 464-7366 Mr. John Giese John Giese & Associates 13201 Linden Avenue North Seattle, WA 98133 (206) 368-8326 Dr. Bob Kitchell 1145 Broadway Seattle, WA 98122 (206) 329-1760 (Polyclinic) Mr. Chuck Kleeberg Seattle-King County Health Department 201 Smith Tower Seattle, WA 98104 (206) 296-4722 Ms. Paula Stewart 4003 134th Avenue Southeast Bellevue, WA 98006 (206) 442-0574 (206) 643-8609 (h) Dr. Lucy Sutphen Virginia Mason Center for Women's Health M.S. H8-E 1100 Ninth Avenue P.O. Box 900 Seattle, WA 98111 (206) 223-7500 Ms. Jean Wooldridge Cancer Information Service Fred Hutchison Cancer Research Center 1124 Columbia Street, MP 243 Seattle, WA 98104 (206) 467-4682 19238 E N V H 567 Environmental Carcinogenesis Winter quarter, 1993 Instructor: David L. Eaton, Ph.D. Environmental Health and Environmental Studies 685-3785 I. Basic Concepts of Chemical Carcinogenesis A. Epidemiology and the Etiology of Cancer - What role do chemicals play? (1)* B. The multi-stage nature of carcinogenesis (1) II. Biochemical and molecular events in chemical carcinogenesis A. Biotransformation of chemical carcinogens (1) discussion focus: aromatic amines; PAHs B. genotoxicity (1) discussion focus: N-nitroso compounds; aa toxin C. protooncogenes and signal transduction (1) discussion focus: G-proteins, protein kinase C, and phorbol esters D. tumor suppresor genes (1) discussion focus: P53 E. hormones, hormone receptors & growth regulation (1) discussion focus: estrogens; TCDD F. apoptosis in carcinogenesis (1) discussion focus: experimental methods to assess apoptosis G. role of cell proliferation in carcinogenesis (1) discussion focus: Saccharin; nitriloacetic acid H. chemo- and dietary interventions and anticarcinogenesis (1) discussion focus: phenolic antioxidants as chemoprotectants III. Experimental models in chemical carcinogenesis A. initiation-promotion regimens (1) discussion focus: Mouse skin papillomas B. preneoplastic lesions/hyperplastic foci (1) discussion focus: Solt-Farber protocol C. anomolies in chemical carcinogenesis 1. peroxisomal proliferation; male rat specific nephropathy (1) discussion focus: Trichloroacetic acid; 2,3,5-Trimethylpentane 2. B6C3F1 mouse hepatomas; transgenic animals (1) discussion focus: chlorinated hydrocarbons (DDT, dieldrin) 3. solid state (e.g. fiber) carcinogens (1) discussion focus: asbestos IV. Assessment of chemical carcinogens and cancer risk A. mathematical modeling- linearized multistage , PBPK and MVK models (1) B. NTP/NCI Bioassay program - experimental design & limitations (1) E x . , . Date; Witness 6 ^ O JOANNE LEATIOTA 1 9 2 3 9 ^Number of lectures shown in parentheses (Tentative schedule). FORMAT - Each 90 minute session will be divided approximately equally into two parts - a "didactic" lecture by the instructor or guest lecturer on the scheduled topic, and a group discussion on the assigned reading, lead by one of the students in the class. The discussion should go beyond the material presented by the instructor, and should be focused on the assigned reading (review article or book chapter), plus one additional recent "original research" article on the specific discussion topic(s), selected and distributed at least one week in advance by the student responsible for leading the discussion. Generally, the discussion focus will be on a specific chemical which produces or acts via the general processes discussed in the lecture. The discussion leader should familiarize herself/himself with other literature on the topic, and be able to demonstrate a relatively high level of understanding of the topic. A bibliography of recent articles (dated 1991 or later) and a 2-3 page summary of the topic of discussion must be turned in within one week following the discussion. A student will be responsible for only one discussion session during the quarter. GRADING: Grades will be based on two written exams and the performance of the student as a discussion leader. The mid-term will be a take-home exam, whereas the final will be an in class, closed book exam. Each exam and the discussion leader responsibilities will be worth 33% of the grade. 19240 Readings List for ENVH 567 - Environmental and Occupational Carcinogenesis (* indicates required reading) I. Basic Concepts of Chemical Carcinogenesis *Fraumeni, J. F., Jr. Epidemiology of Cancer. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 171-181. `Willett, W.C. Diet and human cancer. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 191-198. Perera, F. P., Santella, R. , Brandt-Rauf, P., Kahn, S., Jiang, W. and Mayer, J. Molecular epidemiology of lung cancer. O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 219-236. `Barrett, J. C. Relationship between mutagenesis and carcinogenesis. O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 101-112. `Knudson, A. Genetic events in human carcinogenesis. O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 17-25. `Weinstein, I. B., Borner, C. M., Krauss, R. S. et al. Pleiotropic effects of protein kinase C and the concept of carcinogenesis as a progressive disorder in signal transduction. \ N : O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 113-124. Thorgeirsson, S.S. Silverman, J.A. Gant, T.W. and Marino, P.A. Multidrug resistance gene family and chemical carcinogens. Pharmacol. Ther. 1991. 49(3). P 283-92. 19241 Barrett, J.C. Shelby, M.D. Mechanisms of human carcinogens. Prog. Clin. Biol. Res. 1992. 374. P 415-34. Eriksson, L. C. and Andersson. G. N. Membrane biochemistry and chemical hepatocarcinogenesis. Crit. Rev. Biochem. Mol. Biol. 1992. 27(1-2). P 1-55. Bannasch, P. Hacker, H.J. Klimek, F. Mayer, D. Stumpf, H. and Zerban, H. Cytochemical, microbiochemicai and molecular genetic analysis of chemical carcinogenesis. Prog. Histochem. Cytochem. 1991. 23(1-4). P 45-60. Ramel, C. Pollution, carcinogenesis and cancer prevention. ActaOncol. 1991. 30(6 Spec No). P 27-33. *Harris, C.C. Chemical and physical carcinogenesis: advances and perspectives for the 1990s. Cancer Res. 1991 Sep 15. 51(18 Suppl). P 5023s-5044s. Trush, M.A. Kensler, T.W. An overview of the relationship between oxidative stress and chemical carcinogenesis. Free Radic. Biol. Med. 1991. 10(3-4). P 201-9. Pitot, H.C. Dragan, Y.P. Facts and theories concerning the mechanisms of carcinogenesis. FASEB J. 1991 Jun. 5(9). P 2280-6. Biotransformation and Carcinogenesis *Guengerich, F.P. Metabolic activation of carcinogens. Pharmacol. Ther. 1992. 54(1). P 17-61. Tsuchida, S. and Sato, K. Glutathione transferases and cancer. Crit. Rev. Biochem. Mol. Biol. 1992. 27(4-5). P 337-84. Stiborov'a, M. Frei, E. Schmeiser, H.H. and Anzenbacher, P. The role of peroxidases in the activation of chemical carcinogens. Drug.Metabol. Drug. Interact. 1991. 9(3-4). P 177-90. Kawajiri, K. Fujii-Kuriyama, Y. P450 and human cancer. Jpn. J. Cancer. Res. 1991; 82(12). P 1325-35. *Miles, J.S. and Wolf, C.R. Developments and perspectives on the role of cytochrome P450s in chemical carcinogenesis. Carcinogenesis. 1991; 12(12). P 2195-9. Bock, K.W. Roles of UDP-glucuronosyltransferases in chemical carcinogenesis. Crit. Rev. Biochem. Mol. Biol. 1991. 26(2). P 12950. Gonzalez, F.J. Crespi, C.L. Gelboin, H.V. DNA-expressed human cytochrome P450s: a new age of molecular toxicology and human risk assessment. Mutat. Res. 1991 Mar. 247(1). P 113-27. Genotoxicity, DNA adducts *Wood, M. L. and Essigmann, J. M. Molecular mechanisms by which carcinogen-DNA adducts cause mutations. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 137-152. Elia, M.C. DeLuca, J. G. and Bradley, M.O. Significance and measurement of DNA double strand breaks in mammalian cells. Pharmacol. Ther. 1991. 51(3). P 291-327. Poirier, M.C. and Weston, A. DNA adduct determination in humans. Prog. Clin. Biol. Res. 1991. 372. P 205-18. *Hsia, M.T. Carcinogen-macromolecular adducts as biomarkers in human cancer risk assessment. Biomed. Environ. Sci. 1991 Jun. 4(1-2). P 104-12. *Lutz, W.K. Dose-response relationships in chemical carcinogenesis: from DNA adducts to tumor incidence. Adv. Exp. Med. Biol. 1991. 283. P 151-6. Ashby, J. Determination of the genotoxic status of a chemical. Mutat. Res. 1991 Jun. 248(2). P 221-31. Michejda, C.J. Carcinogenic consequences of DNA alkylation. Contrib. Gynecol. Obstet. 1991. 18. P 71-8. Oncogenes and tumor suppresor genes 19243 *Bourne, H., Goretzki, P., Landis, C., Lyons, J., McCormick, F., Pace, A. and Wong, Y. G-protein oncogenes. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 327-333. Cooper, C.S. The role of non-ras transforming genes in chemical carcinogenesis. Environ. Health. Perspect. 1991 Jun. 93. P 33-40. Brandt, and Rauf, P.W. Advances in cancer biomarkers as applied to chemical exposures:the ras oncogene and p21 protein and pulmonary carcinogenesis. J. Occup. Med. 1991 Sep. 33(9). P 951-5. *Burns, P. A. Brown, K. Bremner, R. Clarke, M. and Balmain, A. Molecular alterations in oncogenes or tumor suppressor genes during chemical carcinogenesis. Prog. Histochem. Cytochem. 1991. 23(1 4). P 100-6. Anderson, M.W. You, M. Reynolds, S.H. Proto-oncogene activation in rodent and human tumors. Adv-Exp-Med-Biol. 1991. 283. P 235-43. *Stanbridge, E. J. Functional evidence for human tumour suprresor genes: Chromosome and molecular genetic studies. IN: C a n c e r S u r v e y s , Vol. 12: T u m o u r S u p p r e s s o r G e n e s , T h e Cell C y c l e a n d C a n c e r . , A. J. Levine, Editor; Cold Springs Harbor Laboratory Press, 1992; pp. 5-24. Polakis, P. and McCormick, F. Interactions between p21ras proteins and their GTPase activating proteins. C a n c e r S u r v e y s , Vol. 12: A. J. Levine,T u m o u r S u p p r e s s o r G e n e s , T h e C e l l C y c l e a n d C a n c e r , Editor; Cold Springs Harbor Laboratory Press, 1992; pp. 25-42. *Weinberg, R.A. Oncogenes, tumor suppressor genes and cell transformation: Trying to put it all together. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 1-16. Cell Proliferation, growth regulation and apoptosis *Sharp., P.A. Regulation of transcrption and oncogenic transformation. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 335-343. Sigal, S.H. Brill, S. Fiorino, A.S. Reid, L.M. The liver as a stem cell and lineage system. Am. J. Physiol. 1992 Aug. 263(2 Pt 1). P G13948. *Melnick, R.L. Does chemically induced hepatocyte proliferation predict liver carcinogenesis? FASEB-J. 1992 Jun. 6(9). P 2698706. *Bowden, G.T. Krieg, P. Differential gene expression during multistage carcinogenesis. Environ. Health. Perspect. 1991 Jun. 93. P 51-6. Tennant, R.W. Elwell, M.R. Spalding, J.W. Griesemer, R.A. Evidence that toxic injury is not always associated with induction of chemical carcinogenesis. Mol. Carcinog. 1991. 4(6). P 420-40. *Butterworth, B.E. Goldsworthy, T.L. The role of cell proliferation in multistage carcinogenesis. Proc. Soc. Exp. Biol.Med. 1991 Nov. 198(2). P 683-7. Cohen, S.M. Analysis of modifying factors in chemical carcinogenesis. Prog. Exp. Tumor. Res. 1991. 33. P 21-40. Naccarato, R. Farinati, F. Hepatocellular carcinoma, alcohol, and cirrhosis: facts and hypotheses. Dig. Dis. Sci. 1991 Aug. 36(8). P 1137-42. Cope, F.O and Wille, J.J. Carcinogenesis and apoptosis: Paradigms and paradoxes in cell cycle and differentiation. In: A p o p t o s i s : T h e M o l e c u l a r B a s i s o f C e l l D e a t h . C u r r e n t C o m m u n i c a t i o n s in C e l l M o l e c u l a r B i o l o g y , Vol 3. L. D Tomei and F. O. Cope, Editors, Cold Spring Harbor Laboratory Press, 1991, pp.61-86. *Ledda-Columbano, G.M and Columbanao, A. Apoptosis and Hepatocarcinogenesis. In: A p o p t o s i s : T h e M o l e c u l a r B a s i s o f C e l l Vol 3. L. DD e a t h . C u r r e n t C o m m u n i c a t i o n s in C e l l M o l e c u l a r B i o l o g y , Tomei and F. O. Cope, Editors, Cold Spring Harbor Laboratory Press, 1991, pp. 101-119. . 3n Ao * ** v \ `Roberts, A.B. and Sporn, M.B. Mechanistic interrelationships between two superfamilies: The steroid/retinoid receptors and transforming growth factor-(3. In: C a n c e r S u r v e y s vol. 1 4 : G r o w t h R e g u l a t i o n b y N u c l e a r H o r m o n e R e c e p t o r s , M.G. Parker, Editor; Cold Springs Harbor Laboratory Press,1992, pp. 205-220. de The, H. and Dejean, A. Retinoic acid receptors and oncogenesis. IN: J. Brugge, T.O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 377-382.. `Green, S. Nuclear receptors and chemical carcinogenesis. Trends Pharmacol. Sci. 1992 Jun. 13(6). P 251-5. J1L aaJ 9r-*1 &a- Risk Assessment, Regulatory Process Huff, J. Cirvello, J. Haseman, J. and Bucher, J. Chemicals associated with site-specific neoplasia in 1394 long-term carcinogenesis experiments in laboratory rodents [published erratum appears in Environ Health Perspect 1991 Nov;95:213] Environ. Health Perspect. 1991 Jun. 93. P 247-70. Green, T. Species differences in carcinogenicity: the role of metabolism and pharmacokinetics in risk assessment. Ann. 1st Super. Sanita. 1991. 27(4). P 595-9. Luebeck, E.G. Moolgavkar, S.H. Stochastic description of initiation and promotion in experimental carcinogenesis. Ann. 1st Super Sanita. 1991. 27(4). P 575-80. Auletta, A.E. Brown, M. Wassom, J.S. Cimino, M.C. Current status of the Gene-Tox Program. Environ .Health Perspect. 1991 Dec. 96. P 33-6. Gold, L.S. Slone, T.H. Manley, N. B. Bernstein, L. Target organs in chronic bioassays of 533 chemical carcinogens. Environ. Health Perspect. 1991 Jun. 93. P 233-46. *Huff, J. Haseman, J. Rail, D. Scientific concepts, value, and significance of chemical carcinogenesis studies. Annu. Rev. Pharmacol. Toxicol. 1991. 31. P 621-52. Huff, J. Bucher, J. Yang, R. Carcinogenesis studies in rodents for evaluating risks associated with chemical carcinogens in aquatic food animals. Environ. Health. Perspect. 1991 Jam 90. P 127-32. `McConnell, E.E. NTP carcinogens-interpretational problems. Mutat-Res. 1991 Jun. 248(2), P 233-7. Nesnow, S. Agarwal, S.C. Lambert, G, R. Earley, K. Gupta, R.C. Interspecies sensitivity to chemical carcinogens. Prog. Clin. Biol. Res. 1992. 374. P 381-97. `Ames, B. N. and Gold, L S. Mitogenesis, mutagenesis and rodent cancer tests. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 125-136. Experimental Models of Carcinogenesis Farber, E. Hepatocyte proliferation in stepwise development of experimental liver cell cancer. Dig. Dis. Sci. 1991 Jul. 36(7). P 973-8. Curtis, S, B. Mechanistic models. Basic Life Sci. 1991. 58. P 36782; discussion 382-6. DiGiovanni, J. Modification of multistage skin carcinogenesis in mice. Prog-Exp-Tumor:Res. 1991. 33. P 192-229. Adams, J.M. and Cory, S. Transgenic models of tumor development. Science. 1991 Nov 22. 254(5035). P 1161-7. Moser, A., Dove, W., Drinkwater, N. and Pitot, H. Genetic models and multistage carcinogeneiss in the gastrointestinal system of the mouse and rat. IN: O r i g i n s o f H u m a n C a n c e r : A C o m p r e h e n s i v e R e v i e w , J. Brugge, T. Curran, E. Harlow and F. McCormick, Editors, Cold Springs Harbor Laboratory Press, 1991; pp. 601-608. Physiologically Based Pharmacokinetic Models Blancato-J-N. Physiologically-based pharmacokinetic models in risk and exposure assessment. Ann-lst-Super-Sanita. 1991. 27(4). P 601-8. Hetrick-D-M. Jarabek-A-M. Travis-C-C. Sensitivity analysis for physiologically based pharmacokinetic models. J-PharmacokinetBiopharm. 1991 Feb. 19(1). P 1-20. *Leung-H-W. Development and utilization of physiologically based pharmacokinetic models for toxicological applications. J-ToxicolEnviron-Health. 1991 Mar. 32(3). P 247-67. Gregory-A-R. Uncertainty in health risk assessments. Regul-ToxicolPharmacol. 1990 Apr. 11(2). P 191-200. *Paustenbach-D-J. Important recent advances in the practice of health risk assessment: implications for the 1990s. Regul-ToxicolPharmacol. 1989 Dec. 10(3). P 204-43. Watanabe-P-G. Schumann-A-M. Reitz-R-H. Toxicokinetics in the 1 C" C 8 evaluation of toxicity data. Regul-Toxicol-Pharmacol. 1988 Dec. 8(4). P 408-13. Cancer Prevention, anticarcinogenesis Wattenberg, L.W. Inhibition of carcinogenesis by minor dietary constituents. Cancer-Res. 1992 Apr 1. 52(7 Suppl). P 2085s-2091s. Smith, P.J. Carcinogenesis: molecular defences against carcinogens. Br. Med. Bull. 1991 Jan. 47(1). P 3-20. Anomolies in Carcinogenesis 1. Peroxisomal Proliferation *Green, S. Peroxisome Proliferators: A model for receptor mediated carcinogenesis. In: C a n c e r S u r v e y s vol. 1 4 : G r o w t h R e g u l a t i o n b y N u c l e a r H o r m o n e R e c e p t o r s , M.G. Parker, Editor; Cold Springs Harbor Laboratory Press, 1992, p-. 221-232. Tanaka-K. Smith-P-F. Stromberg-P-C. Eydelloth-R-S. Herold-E-G., Grossman-S-J. Frank-J-D. Hertzog-P-R. Soper-K-A. Keenan-K-P. Tl Studies of early hepatocellular proliferation and peroxisomal proliferation in Sprague-Dawley rats treated with tumorigenic doses of clofibrate. Toxicol-Appl-Pharmacol. 1992 Sep. 116(1). P 71-7. Sharma-R. Lake-B-G. Foster-J. Gibson-G-G.Microsomal cytochrome P-452 induction and peroxisome proliferation byhypolipidaemic agents in rat liver. A mechanistic inter-relationship. BiochemPharmacol. 1988 Apr 1. 37(7). P 1193-201. *Butterworth-B-E. Loury-D-J. Smith-Oliver-T. Cattley-R-C. The potential role of chemically induced hyperplasia in the carcinogenic activity of the hypolipidemic carcinogens. Toxicol-Ind-Health. 1987 Jun. 3(2). P 129-49. Goldsworthy-T-L. Popp-J-A. Chlorinated hydrocarbon-induced peroxisomal enzyme activity in relation to species and organ carcinogenicity. Toxicol-Appl-Pharmacol. 1987 Apr. 88(2). P 225- 1 Q9ZI9 i <D h* 33. *Hertz-R. Arnon-J. Hoter-A. Shouval-D. Bar-Tana-J. Clofibrate does not induce peroxisomal proliferation in human hepatoma cell lines PLC/PRF/5 and SK-HEP-1 .Cancer-Lett. 1987 Mar. 34(3). P 263-72. 2. Male Rat Specific Nephropathy *Lehman-McKeeman-L-D. Caudill-D. Biochemical basis for mouse resistance to hyaline droplet nephropathy: lack of relevance of the alpha 2p.-globulin protein superfamily in this male rat-specific syndrome. Toxicol-Appl-Pharmacol. 1992 Feb. 112(2). P 214-21. Olson-M-J. Johnson-J-T. Reidy-C-A. A comparison of male rat and human urinary proteins: implications for human resistance to hyaline droplet nephropathy. Toxicol-Appl-Pharmacol. 1990 Mar 1. 102(3). P 524-36. Thompson-K-L. Rosner-M-R. Regulation of epidermal growth factor receptor gene expression by retinoic acid and epidermal growth factor. J-Biol-Chem. 1989 Feb 25. 264(6). P 3230-4. 3. B6C3F1 mouse hepatomas Wolff-G-L. Gaylor-D-W. Blackwell-B-N. Moore-G-E. T. Bladder and liver tumorigenesis induced by 2-acetylaminofluorene in different F1 mouse hybrids: variation within genotypes and effects of using more than one genotype on risk assessment. J-Toxicol-EnvironHealth. 1991 Jul. 33(3). P 327-48. Candrian-U. You-M. Goodrow-T. Maronpot-R-R. Reynolds-S-H. Anderson-M-W. Activation of protooncogenes in spontaneously occurring non-liver tumors from C57BL/6 x C3H F1 mice. CancerRes. 1991 Feb 15. 51(4). P 1148-53. *Buchmann-A. Bauer-Hofmann-R. Mahr-J. Drinkwater-N-R. Luz-A. Schwarz-M. Mutational activation of the c-Ha-ras gene in liver tumors of different rodent strains: correlation with susceptibility to hepatocarcinogenesis. Proc-Natl-Acad-Sci-U-S-A. 1991 Feb 1. 88(3). P 911-5. Bauer-Hofmann-R. Buchmann-A. Wright-A-S. Schwarz-M. Mutations o in the Ha-ras proto-oncogene in spontaneous and chemically induced liver tumours of the CF1 mouse. Carcinogenesis. 1990 Oct. 11(10). P 1875-7. *Drinkwater-N-R. Hanigan-M-H. Kemp-C-J. Genetic determinants of hepatocarcinogenesis in the B6C3F1 mouse. Toxicol-Lett. 1989 Dec. 49(2-3). P 255-65. Specific Carcinogenic substances Huff-J-E. Salmon-A-G. Hooper-N-K. Zeise-L. Long-term carcinogenesis studies on 2,3,7,8-tetrachlorodibenzo-p-dioxin and hexachlorodibenzo-p-dioxins. Cell-Biol-Toxicol. 1991 Jan. 7(1). P 67-94. Tricker-A-R. Preussmann-R. Carcinogenic N-nitrosamines in the diet: occurrence, formation, mechanisms and carcinogenic potential. Mutat-Res. 1991 Mar-Apr. 259(3-4). P 277-89. Gelderblom-W-C. Marasas-W-F. Vleggaar-R. Thiel-P-G. Cawood-ME. Fumonisins: isolation, chemical characterization and biological effects. Mycopathologia. 1992 Feb. 117(1-2). P 11-6. Magos, L. Epidemiological and experimental aspects of metal carcinogenesis: physicochemical properties, kinetics, and the active species. Environ-Health-Perspect. 1991 Nov. 95. P 157-89. 1 " A m erica's Epidemic of Chem icals and C ancer" - M vth o r F act?1 D avidL. Eaton, PhD . Associate Professor o f Environmental Heath and Environmental Studies University o f Washington Seattle, WA There is no debate that cancer is a devastating and deadly disease. One in three people living in the U.S. today will contract some form of cancer in his or her lifetime, and one in four will die from it, if current rates continue. However, the contention that we are in the throes of an unprecedented cancer epidemic, and that this "epidemic" is caused in l^rge part by cancer-causing pesticides and industrial chemicals is simply not supported by the scientific data available on cancer incidence and etiology (causes). Are cancer rates increasing in epidemic proportions? It is certainly true that both the total number of people and the fraction of all deaths attributable to cancer have increased dramatically in the past 50 years. However, cancer is largely a disease o f old age, and thus it is necessary to adjust such statistics for changes in the age distribution of a population over the years. A 1988 report from the National Cancer Institute states that "the age adjusted mortality rates for all [types of] cancers combined, except lung cancer, has been declining since 1950 for all individual age groups except 85 and above". Statistics from the American Cancer Society yield the same conclusion. The reason for excluding lung cancer in this evaluation is to examine cancer trends without the strong influence that the dramatic increase in lung cancer incidence (and deaths) have on total cancer rates. The available scientific evidence would support the supposition that we are in an "epidemic" of lung cancer. For most of the first half of this century, lung cancer mortality was not even in the "top five" types o f cancer-related deaths. Lung cancer is now the leading cause of cancer-related deaths in both men and women. There is abundant scientific evidence to demon strate that 85 - 90% of all lung cancers in men, and perhaps 70% in women, is directly attributable to smoking, a habit which increased steadily in popularity from the early 1900's through the 1950's. Per capita consumption of cigarettes increased 5-fold in men from 1900 to 1960, and with it a concomitant increase in lung cancer. The risks o f several other types o f common cancers are also increased by smoking (e.g. cancers of the bladder and esophagus). Approximately one-third of all cancer deaths could be eliminated by the elimination of smoking from our society. The sup position that many o f these smoking-related cancers may be related to carcinogenic pesticides and chemicals in tobacco is grossly misleading- while the statement in its absolute sense is likely to be true, the source of these carcinogenic pesticides and chemicals is mother nature, not the chemical and agricultural industry. What proportion of cancers can be related to environmental pollution from synthetic pesticides and industrial chemicals? The often-cited statistic that "70 to 90% of all cancers are caused by environmental influences and are hence theoretically preventable" is frequently incorrectly interpreted as meaning that chemical pollution is responsible for 70 - 90% of cancers. The original source of this statement stems from studies on human cancer incidence reported in the late 1960's. These studies suggested that most cancers could not be directly traced to genetic or hereditary factors, and therefore it was concluded that the majority of cancers must have an "environmental" etiology. In this context, the term "environmental" includes not only chemicals in the environment, ^This article has been prepared for the fourth edition of an undergraduate textbook entitled "Environmental Science: A Framework for Decision Making", by Daniel D. Chiras; Benjamin Cummings Publishing Company, Menlo Park, California, 1990. -1- WExr.t.n3es4s.-rZ'n-.hr>briL t JOANNE LEATIOTA 52 but lifestyle factors such as smoking and alcohol, dietary factors such as the proportion of fat and fiber in the diet, "natural" carcinogens that occur in nearly all foods, cancer-causing viruses such as hepatitis B, and occupational exposures to substances such as asbestos. O f the variety o f environmental factors other than smoking, dietary factors are now generally thought to represent the largest source of cancer risk, perhaps related to 30-40% of all cancers. Although synthetic chemicals such as industrial pollutants and pesticides present in trace amounts in our food supply may contribute to dietary risk, recent studies have suggested that this contribu tion is trivial relative to other "non-pollutant" factors. For example, the risk of breast cancer in women, second only to lung cancer in incidence and mortality, is significantly increased by high fat diets, and the amount of fiber in the diet substantially influences the risk of colon cancer, a major site of cancer in both men and women. Recently scientists have found that by far the largest source of exposure to cancer-causing chemicals comes not from industrial pollution, but from chemicals that occur naturally in our diet. All plants produce toxic chemicals as a means of protection against insects, fungi and animal predators. It has been estimated that we ingest in our diet about 10,000 times more of "nature's pesticides" than man-made chemical residues. Many of these chemicals are potent mutagens and carcinogens, and are frequently present at levels thousands of times higher than the trace levels of synthetic pesticide residues and industrial chemicals sometimes found in food crops. For example, natural carcinogenic chemicals are found in mushrooms, parsley, basil, fennel, pepper, celery, figs, and mustard, to name a few. The vast majority of natural chemicals present in foods (tens of thousands) have never been tested to determine if they might be carcinogenic. There is also sub stantial scientific evidence that cooking of foods, especially meats, forms highly mutagenic and perhaps carcinogenic chemicals from natural precursors such as certain protein components. Taken together, the dietary risk factors from natural sources, often present in relatively high amounts, are far more important than the pesticide residues and industrial chemicals that can often be detected at exceedingly small concentrations in our diets. Unfortunately, because of the rela tively high exposure to carcinogens from natural sources, the complete elimination of synthetic industrial chemicals from our diet, if it were possible, would not likely have any significant bene ficial effect on cancer incidence and mortality. There is little question that extensive exposure to some industrial chemicals can increase the risk o f certain types o f cancer. Occupational exposures to potentially carcinogenic chemicals can be a significant risk factor for cancer, and every effort should be made to identify and reduce workplace exposure to these chemicals. However, the reliance upon a 1978 report of the Department of Health Education and Welfare which stated that "20 to 38% of all cancers could be attributed to occupational exposure to just 6 industrial chemicals" is no longer valid, as even the authors of this report acknowledge that this was a substantial overestimation of the contribution of occupational exposures to total cancer incidence. Most authoritative sources now estimate that occupational exposures account for no more than 5 to 10% of all cancers, and many of these cancers are a result o f extensive exposures to asbestos and a few other industrial carcinogens that were commonplace in the 50's and 60's, but have since been greatly reduced (although certainly not eliminated). O f course, 5% is too high, and every effort should be made to reduce this contribution further. Finally, recent advances in the understanding of the biology of cancer suggest that "spontaneous" or "background" alterations in DNA may explain much of the etiology of cancer. The use of modem techniques in molecular biology have revealed that DNA is inherently unstable, can be altered by normal errors in DNA replication, and is subject to extensive damage from processes associated with normal cellular metabolism. Considering that within our life span, our cells undergo about 10 million billion (1016) cell divisions, and that spontaneous errors in this process which lead to mutations and cancer accumulate with age, it is not surprising that cancer 19253 -2- WORLD HEALTH ORGANIZATION INTERNATIONAL AGENCY FOR RESEARCH ON CANCER LONG-TERM HAZARDS OF POLYCHLORINATED DIBENZODIOXINS and POLYCHLORINATED DIBENZOFURANS JOINT NIEHS / IARC WORKING GROUP REPORT IARC, LYON JUNE, 1978 IARC INTERNAL TECHNICAL REPORT No. 78/001 19254 1ARC MONOGRAPHS ON THE EVALUATION OF THE CARCINOGENIC RISK OF CHEMICALS TO HUM ANS In 1971, the International Agency for Research an Cancer (IARC) initiated a p r o g r a m e on the evaluation of the carcinogenic risk of chemicals to humans involving the production of critically evaluated monographs on individual chemicals. The role of the monograph programe is to aollect a n available relevant experi mental and epidemiological data about groups of chemicals to which humans are known to be exposed, to evaluate these data in terms of hunan risk with the help of inter national working groups of acknowledged experts in chemical carcinogenesis and related fields, and to publish and disseminate the conclusions of those working groups as a series of IARC Monographs. Some Inorganic Substances. Chlorinated Hydrocarbons. Aromatic Amines. N-Nitroso Compounds, ano Natural Products Some Inorganic and Organometaluc Compounds C ertain Polycyclic Aromatic Hydrocarbons and Heterocyclic Compounds Some A romatic Amines. Hydrazine and ~ Related Substances. .V-Nmtoso Compounds and Miscellaneous Alkylating Agents Some O rganochlorine Pesticides Sex Hormones Some Anti-thyroid and Related Substances. Nitrofurans and Industrial C hemicals Some Aromatic Azo Compounds Some A ziridines. V-, 5- and O-Mustards and Selenium Some Naturally Occurring Substances Cadmium. Nickel. Some Epoxides. Miscellaneous Industrial Chemicals and G eneral Considerations on Volatile A naesthetics Some Carbamates, Thiocarbamates ano Carbazides Some Miscellaneous Pharmaceutical Substances Asbestos Some Fumigants, the Herbicides 2.4-D and 2.4.5-T , Chlorinated Dibenzodioxins and Miscellaneous Industrial Chemicals Some Aromatic A mines and Related Nitro Compounds-- Hair Dyes. Colouring Agents and Miscellaneous Industria; Chemicals Some ;V-Nitroso Compounds Volume 1. 1972; 184 pages USS 4.20: Sw. fr. 12.-- Volume 2. 1973; 181 pages USS 3.60: Sw. fr. 12.-- Volume 3. 1973: 271 pages USS 5.40: Sw. fr. 18.-- Volume 4. 1974: 286 pages USS 7.20: Sw. fr. 18.-- Volume 5. 1974; 241 pages USS 7.20; Sw. fr. 18.-- Volume 6. 1974; 243 pages USS 7.20: Sw. fr. 18.-- Volume 7. 1974; 326 pages USS 12.80: Sw. fr. 32.-- Volume 8. 1975 ; 357 pages USS 14.40: Sw. fir. 36.-- Volume 9. 1975; 268 pages USS 10.80; Sw. fr. 27.-- Volume 10. 1976: 353 pages USS 15.00: Sw. fr. 38.-- Volume 11. 1976; 306 pages USS 14.00: Sw. fr. 34.-- Volume 12. 1976; 282 pages USS 14.00: Sw. fr. 34.-- Volume 13. 1977; 255 pages USS 12.00: Sw. fr. 30.-- Volume 14, 1977 ; 106 pages USS 6.00 : Sw. fr. 14.-- Volume 15, 1977; 354 pages USS 20.00; Sw. fr. 50.-- Volume 16. 1978; 400 pages USS 20.00: Sw. fr. 50.-- Volume 17, 1978; 365 pages USS 25.00; Sw. fr. 50.-- Available from WHO, Distribution and Sales Service, j z -lj. Geneva z/, Switzerland; or fron the United Nations Bookshop, New York, NY 10017, USA (retail only) ; or from the Franklin Institute Press, Benjamin Franklin. Padcway, PA 19103, USA. 1925-5 LONG-TERM HAZARDS OF POLYCHLORINATED DIBENZODIOXINS and POLYCHLORINATED DIBENZOFURANS JOINT NIEHS / (ARC WORKING GROUP REPORT This rep o rt rep resen ts the views and opinions o f an US National I n s titu te o f Environmental Health S cien ces/In tern a tio n a l Agency fo r Research on Cancer ad hoc Working Group which met during 10-11 January 1978 in Lyon, France to review the h isto ry o f human exposure to these chem icals, to c o lla te cu rren t inform ation, and to plan and fo r e c a s t needed new d ir e c tio n s . IARC, LYON JUNE, 1978 CONTENTS LIST OF TABLES ........................................ 4 PARTICIPANTS .......................................... 5 JOINT NIEHS/IARC WORKING GROUP REPORT ...................... 7 I. BACKGROUND ...................................... 7 Purpose of the NIEHS/IARC Meeting ..................... 7 Report Contents ................................... 7 Introduction to the Problem.............. i.......... 8 II. CHEMICAL ASPECTS .................................. 9 III. TOXICOLOGICAL ASPECTS .............................. 9 Animals ......................................... Chloracne ...................................... Hepatotoxicity .................................. Hypoplasia of the lymphoid tissues ................... General debilitation and wasting .................... Embryotoxicity and teratogenicity of 2,3,7,8-tetra-CDD .... Chick oedana disease .............................. Other effects ................................... 9 14 14 14 15 16 15 15 Enzyme Induction .................................. 15 Hepatic Cytosol Binding Protein....................... 19 Structure-Activity Relationships ...................... 20 Pharmacokinetics .................................. 21 Mutagenicity ..................................... 21 Carcinogenicity ................................... 22 Humans .......................................... 23 Toxic effects in humans ........................... 23 Human exposure to PCDDs and PCDFs .................... 27 IV. EXPOSURE EPISODES CONSIDERED IN DETAIL BY THE NIEHS/IARC ad hoc WORKING GROUP 3i Phenoxyacids exposure .............................. 51 Hexachlorophene exposure ............................ 53 Phenoxyacids and chlorcphenols exposure in forestry workers and in wood industry .............................. 33 2,3,7,8-Tetra-CDD exposure during 2,4,5-T or polychlorinated phenols production and following industrial accidents .... 34 V. RECOMMENDATIONS ................................... 43 VI. REFERENCES ....................................... 45 Part I - References cited in the text ......... 45 Part II - Vforking papers ...... 57 3 Number 1 2 3 4 5 6 7 LIST CF TABLES Title Possible number of iscmers for polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans Estimated single oral lethal dose50 _ 3 o values of certain polychlorinated dibenzo-p-dioxin iscmers Sunmary of the toxic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin Evaluation of reported embryotoxic and terato genic effects induced by 2,3,7,8-tetrachlorodibenzo-p-dioxins in rats and mice Ongoing long-term carcinogenicity testing of chlorinated dibenzo-p-dioxins Toxic effects of 2,3,7,8-tetrachlorodibenzo-pdioxins in humans Epidsniological studies on 2,3,7,8-tetrachlorodibenzo-p-dioxin Figure 1 General formulae for polychlorinated dibenzo-pdioxins and polychlorinated dibenzofurans Participants in the NXEHS/IARC Wbrking Group on the Lonq-Term Hazards of Polychlorinated Dibenzodiaxins / Polychlorinated Dibenzofurans Lyon, France, 10-11 January 1978 Matters Dr A.U. Arstila, Professor and Chairman, Department of Cell Biology, University of Jyvskyl, Vapaudenkatu 4, 40100 Jyvskyl, Finland Dr 0. Axelson, Department of Occupational Medicine, Eegional Hospital, S-581 85 Linkping, Sweden (Vice-Chairman) Dr P.J. Baxter, Employment Medical Advisory Service, Health & Safety Executive, Baynards House, 1 Chepstow Place, London W2 4TF, United Kingdon (Rapporteur-Epidem iology) Dr F. Berrino, Servizio di Epidsniologia, Istituto Nazionale per lo Studio e la Cura dei Tumori, Via Venezian 1, 20133 Milano, Italy (R apporteu r-Epidem iolo g y ) Dr L. Bisanti, Uffici Regione, Ospedale Desio, Milano, Italy Dr R. Frentzel-Beyme, Institut fr Dokumentation, Information und Statistik, Deutsches Krebsforschungszentrun, Im Neuenheimer Feld 280, 6900 Heidelberg 1, Federal Republic of Germany Dr A. Hay, Department of Animal Physiology and Nutrition, University of Leeds, Kirkstall Laboratories, Vicarage Terrace, Leeds IS5 3HL, United Kingdom Professor L. Jirasek, Universit Karlova FVL, 2. Dermatologicka Klinika, U. Nanocnice 2, 128 08 Prague 2, Czechoslovakia Dr G. May, IA Welbeck Road, Bolsover, Derbyshire S44 6DH, United Kingdom Dr J.A. Moore, Associate Director, Research Resources Program, National Institute of Environmental Health Sciences, PO Box 12233, Research Triangle Park, North Carolina 27709, United States of America (Chairman) Dr G.F. Peruzzo, Caimissario Straordinario, Uffici della Regione, Via S. Carlo 4, Seveso (Milano), Italy Professor F. Pocchiari, Director General, Istituto Superiore di Sanit, Viale Regina Elena 299, 00161 Rana, Italy 5 i ;q j> ''J Dr A. Poland, Assistant Professor of Oncology, McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, Wisconsin 53706, United States of America (Rapportenr/Toxicology) Professor C. Rappe, Department of Organic Chemistry, University of Urea, S-901 87 Umea, Svreden (Rapporteur/Chem istry) Dr V. RiihimSki, Department of Industrial Hygiene and Toxicology, Institute of Occupational Health, Haartmaninkatu 1, 00290 Helsinki 29, Finland Professor I.J. Selikoff, Mount Sinai School of Medicine, Environmental Sciences Laboratory, Cunmings Basic Sciences Building, 10 East 102 Street, New York, N.Y. 10029, United States of America Dr R.R. Suskind, Director, Institute of Environmental Health, Kettering Laboratory, University of Cincinnati Medical Center, 3223 Eden Avenue, Cincinnati, Chio 45267, United States of America Dr J.G. Vos, Head, Department of Oncology, Laboratory of Pathology, Rijks Instituut voor de Volksgezondheid, Postbus 1, Bilthoven, The Netherlands Dr N. Wald, University of Oxford, Department of the Regius Professor of Medicine, Radcliffe Infirmary, Oxford QX2 6HE, United Kingdom Secretariat Dr H. Bartsch, Unit of Chemical Carcinogenesis, IARC Dr J.E. Huff, Unit of Chemical Carcinogenesis, IARC (Secretary) Dr T. Kuroki, Unit of Chemical Carcinogenesis, IARC Dr R. Montesano, Unit of Chenical Carcinogenesis, IARC Dr J. Parizek. Division of Environmental Health, WHO, Geneva Dr R. Saracci, Unit of Epidemiology and Biostatistics, IARC (S ecreta ry) Dr L. Tcmatis, Chief, Unit of Chenical Carcinogenesis, IARC (Secretary) Mr J.D. Wilboum, Unit of Chemical Carcinogenesis, IARC a. 6 Joint NIEHS/IARC Working Group Report on the Long-Term Hazards of Polychlorinated Dibenzodioxins / Polychlorinated Dibenzofurans I. BACKGROUND In February 1977, an IARC ad hoe Working Group met to evaluate the carcinogenic risk to humans of certain fumigants, the herbicides 2,4-D and 2,4,5-T, chlorinated dibenzo-p-dioxins, and miscellaneous industrial chemicals (IARC, 1977). What became clear both during the meeting and subsequently, while preparing IARC Monograph Volume 15 for publication, was the need not only for better coordination of ongoing and projected epidemiological efforts concerning the chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans, but also for a worldwide action plan which would serve to harmonize the various activities already in progress or in preparatory stages. Purpose of the NIEHS/IARC Meeting During 10-11 January 1978, in Lyon, France, a joint US National Institute of Environmental Health Sciences/Intemational Agency for Research on Cancer ad hoc Working Group considered and discussed the feasibility of coordinating epidaniological studies on the long-term hazards associated with the chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans (PCDDs and PCDFs). Eighteen invited scientists, from eight countries, presented intro ductory working papers summarizing the most up-to-date and relevant infor mation available iron their individual programmes (References - Part II) . Report Contents This report represents the collective views and scientific opinions of the Working Group. The greater part of this document comprises epidaniological studies related to episodes of human exposure. The review begins, however, with a brief section concerning possible routes of human exposure, an overview of the pertinent chemical characteristics, and the salient toxicological properties of the structurally similar PCDDs/PCDFs. The Working Group report ends with recaanendations for future activities. Introduction to the Problem Human exposure in the workplace can occur when chlorinated dibenzo-pdioxins (PCDDs) are formed during the production of certain ccnpounds such as the herbicide 2,4,5-trichlorcphenoxyacetic acid (2,4,5-T), the fungicide pentachlorcphenol, and the germicide hexachlorophene (IARC, 1977). The dioxins, inpurities/contaminants associated with these end-products, result most often from treatment of chlorinated benzenes at elevated temperature and pressure under alkaline conditions. Via the widespread use of these ccnmercial products (Gosselin e t a l . , 1976; IARC, 1977) the general population may also became exposed. In recent years, out-of-control chemical reactions during the produc tion of 2,4,5-trichlorophenol have proceeded to the explosive stage thereby exposing persons to toxic levels of PCDDs. Fran certain of these accidents, moreover, not only did those occupationally involved receive dangerous exposure but also those inhabiting the surrounding areas received uncarmon risk. The structurally-related chlorinated dibenzofurans (PCDFs) are contaminants found in sane polychlorinated biphenyl ccnpounds - Aroclor, Clophen, Phenoclor (IARC, 1978a). These unwanted - and until recently undetected and unrecognized impurities are more toxic and represent a potentially greater environmental hazard than the ccnpounds they contaminate (Blair, 1973; Huff & Wasson, 1973, 1974a,b; IARC, 1977, 1978a; NIEHS, 1973; Ramel, 1978). For instance, 2,3,7,8-tetra-CDD has been recently found in beef fat iron cattle grazed on 2,4,5-T-treated rangeland and in breast milk from women living in areas where 2,4,5-T is used on rangeland or in forest areas (Meselson e t a t . , 1978). Various investigators (Buser & Bosshardt, 1978; Buser e t d l . , 1978b,d; Olie e t a l . , 1977) have also discovered PCDDs and PCDFs in fly ash and flue gas from municipal incinerators, and in dust from fungicidetreated wood (Levin e t a l . , 1977). *. 1 8 19262 II. CHEMICAL ASPECTS The chlorinated dibenzo-p-diaxins (PCDDs) and chlorinated dibenzofurans (PCDFs) are two series of tricyclic aromatic compounds which exhibit similar chemical and physical properties (IAEC, 1977, 1978a; NIEHS, 1973). The basic two-dimensional structures as shewn in Figure 1 have eight possible points of chemical addition. Fran the monochloro to the octachloro deri vatives, a variety of positional isaners are possible - 75 PCDDs and 135 PCDFs (Table 1). The extreme toxic potency of seme of these compounds, as well as the large number of potential isaners, warrant analytical methods exhibiting high sensitivity and specificity to monitor the environment. A desirable detection limit of one part per trillion (ppt or a picogram/gram sample) is being reached with current methodology. Essential requirements included efficient clean-up, good separation and selectivity, ultra-sensitive quantification, and validation (Buser, 1977; Buser & Rappe, 1978). III. TOXICOLOGICAL ASPECTS Animals The prototype and most extensively studied isaner of the PCDDs and PCDFs is the 2,3,7,8-tetrachlorodibenzo-p-diaxin (2,3,7,8-tetra-CDD), perhaps the most potent man-made toxin presently known. The comparative oral lethal dose values for various PCDDs in mice and guinea-pigs (Table 2) show clearly that the 2,3,7,8-tetra-CDD and the 1,2,3,7,8-penta-CDD isomers are the most toxic (McConnell e t a l ., 1978a). Toxic effects induced by PCDDs and PCDFs vary quantitatively and qualitatively among different species; however, within a single species the untoward consequences are markedly similar for all PCDDs and PCDFs that have been studied. For example, the toxic effects induced by 2,3,7,8-tetra-CDD which are most often observed in mice, guinea-pigs, and monkeys, are illustrated in Table 3. In a brief report, daily oral intake (12-61 days) of <1 ygAg tw 2,3,7,8-tetra-CDD was stated as being lethal to young male rhesus monkeys (McNulty, 1977) ,whereas McConnell e t a l . (1978b) reported an ID in fanale rhesus monkeys as <70 ygAg bw. 50 9 Figure 1 General formulae for polychlorinated dibenzo-j;-dioxins and polychlorinated dibenzofurans Cly CHLORINATED DIBENZO-p-DICKINS Empirical formulae: Molecular weight ranges: Cl2H7-0C1l-82 218-460 CHLORINATED DIBENZCFURANS C12H7-0C1l-8 202-444 TABLE 1 Possible number of isomers for polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans Chlorine atoms 1 2 3 4 5 6 7 8 TOTAL PCDD isaners 2 10 14 22 14 10 2 1 75 PCDF isaners 4 16 28 38 28 16 4 1 135 11 19265 TABLE 2a Estimated single oral lethal doses0 - 3 0 values of certain polvdilorinated dibenzo-p-dicocin isctrers Chlorine isomer 2,8 2,3,7 2,3,7,8 1,2,3,7,8 1,2,4,7,8 1,2,3,4,7,8 1,2,3,6,7,8 1,2,3,7,8,9 1,2,3,4,6,7,8 l-N02-3,7,8 l-NH2-3,7,8 l-N02-2,3,7,8 l-NH2-2,3,7,8 Guinea-pigs yg/kg bw y mole/kg bw >300,000 >1,180 29,444 120.41 2 0.006 3.1 0.009 1,125 3.15 72.5 0.185 70-100 0.178-0.255 60-100 0.153-0.255 >600 >1.400 >30,000 >90 >30,000 >99 47.5 0.129 194.2 0.576 Mice yg/kg tw y mole/kg kw -- >3,000 >10 283.7 0.88 337.5 0.94 >5,000 >14 825 2.11 1,250 3.19 >1,440 >3.67 -- -- -- >2,000 >5.4 >4,800 >14.2 ^Fran McConnell e t a t . (1978a) eThe ID5 0 - 3 0 was calculated by the Spearman-Karber rrethod (Finney, 1964). Estimated range due to variability in replicates. 12 '') y ; ( ) 1 TABLE 3a Surroary of the toxic effects of 2 , 3 , 7 , 8-tetracMorodibenzo-p-diaxin Thymus involution ' - - Spleen reduction (white pulp) Bone marrow hypoplasia Liver megalocytosis/degeneration Bile duct hyperplasia Testicular degeneration Renal pelvis hyperplasia Urinary bladder hyperplasia Adrenal cortical atrophy (Zona Glomerulosa) Haemorrhage intestinal adrenal Ascites Cutaneous lesions Mice Guinea-pigs Monkeys (female) +++ ++'+ +++ ++ + 4 ++ +++ - + - + +++ ++ +++ ++ - + - ++ - - ++ ++ -- ++ ++ -- + +++ ^Frcsn Moore (1978) NA = not applicable 13 132G7 A preliminary report on 2,3,7,8-tetra-CDF concerning the toxicity of PCDFs (Moore e t a l . , 1976), and a report on PCDFs (Oishi e t a l . , 1978) have been published. The toxic syndrome produced by PCDDs and PCDFs may be divided into seven categories: Chloracne - 2,3,7,8-Tetra-CDD and the tri-PCDF and tetra-PCDF were found to be active skin irritants and to induce acneform lesions in the skin of rabbit ears (Kimmig & Schulz, 1957a) . Classical chloracne is a hallmark of PCDDs exposure in humans and an analagous hyperkeratosis and the modulation of sebaceous structures to keratin cysts has been cbserved in monkeys, rabbits, and hairless mice. Chloracne or acneform dermatitis is a carmen occupational dermatitis characterized by canedones, keratin cysts, pustules, papules and abscesses. In 1957, Kimmig & Schulz (1957b) found that 2,3,7,8-tetra-CDD was the agent responsible for causing occupational chloracne in employees of chlorophenol-producing factories. Further, in 1971 PCDDs were implicated as causing chloracne in male workers in a plant producing 2,4-D and 2,4,5-T (Bleiberg e t a l . , 1964; Poland e t a l . , 1971). Chloracne may appear weeks or months after the initial exposure to PCDDs and PCDFs. s ' . ' L r r Hepatoxicity -AIhe degree of hepatic-involvement appears to be dose-dependent, and the severity of the changes produced varies between species (Gupta e t a l . , 1973). Hepatic necrosis produced by 2,-3,7,8-tetraCDD is probably a contributing cause of death in rats and rabbits, while hepatic necrosis and liver insufficiency are less extensive in mice and are minimal in comparison in guinea-pigs and monkeysJ(Buu-Ho! e t a l . , 1972a; Harris e t a l . , 1973; Jones, 1975; Jones & Butler, 1974; McConnell e t a l . , 1978b; Schulz, 1968; Schwetz e t a l . , 1973; Vos e t a l . , 1973, 1974). Hepatic porphyrin accumulation has been cbserved in mice, rats, and chickens. Hypoplasia of the lymphoid tissues - Particularly involved are the cortical cells of the thymus and this hypoplasia has been cbserved in mice, rats, guinea-pigs, and monkeys. The most significant findings in both mice and guinea-pigs treated with sublethal doses of 2,3,7,8-tetra-CDD were in the lymphoid system, resulting in suppression of cell-mediated inmunity, particularly in young animals (Vos, 1977; Vos e t a l . , 1973) . Lew levels of 2,3,7,8-tetra-CDD that did not produce overt clinical or * pathological changes still reduced host defences: 1 yg/kg tw given orally once weekly for 4 weeks to mice before infection with Salmonella increased mortality and decreased the time fran infection to death (Thigpen e t a l . , 1975) . The increased mortality may be caused by the endotoxin content of these gram negative bacteria, since 2,3,7,8-tetra-CDD markedly increases the susceptibility of mice to endotoxin (lipopolysaccharide of Escherichia col i) (Vos e t a l . , 1978) . Treatment of female mice and rats with 2,3,7,8tetra-CDD during the latter half of gestation and in the postnatal period resulted in a severe depletion of lymphocytes in the thymic cortex of the offspring (Vos & Moore, 1974) . Cellular inmunity was impaired. Haematological changes in mice, rats and guinea-pigs treated with 2,3,7,8-tetra-CDD included lymphopenia and thrombocytopenia (Weissberg & Zirikl, 1973; Zinkl e t a l . , 1973); manifest also is an increased suscepti bility to infection concomitant with the suppression of cell-mediated immunity. For 9 months, female rhesus monkeys received a diet containing 500 parts per trillion 2,3,7,8-tetra-CDD; within 6 months, the monkeys became anaemic and after 9 months pancytopnie (Allen e t a l . , 1977) . The marked thrombocytopenia was associated with widespread haemorrhage. Death occurred in five of the eight monkeys between months 7 and 12 of the experi ment at total exposure levels of 2-3 yg 2,3,7,8-tetra-CDD/kg bw. At autopsy, in addition to extensive haemorrhage, there was a distinct hypocellularity of the bone marrow and lymph nodes. Hypertrophy, hyperplasia and metaplasia of the epithelium in the bronchial tree, bile ducts, pancreatic ducts, salivary-gland ducts and palpebral conjunctivae were observed. Squamous metaplasia and keratinization of the sebaceous glands and hair follicles were present in the skin. Death was attributed to canplications from the severe pancytopenia (Allen et a l . , 1977). General debilitation and wasting - Animals that receive a toxic or lethal dose of PCEDs or PCDFs exhibit a chronic and progressive weight loss with parallel mobilization of peripheral fat, increased serum triglyceride 15 19229 levels, and development of a fatty liver. Death due to PCDDs or PCDFs intoxication is delayed, as exemplified by an elapsed time period of 6 to 8 weeks following administration of a lethal dose and eventual death (Allen e t a l . , 1977; McConnell s t a l . , 1978a,b). Embryotoxicity and teratogenicity of 2,3,7,8-tetra-CDD - In repeated or single doses of 2,3,7,8-tetra-CDD to mice, as little as 1-10 \ig/kg bw cause increased frequencies of cleft palate and kidney abnormalities (see Table 4) (Courtney & Moore, 1971; Neubert & Dillmann, 1972; Neubert e t a l . , 1973; Smith s t c l . , 1976). In rats, embryo-lethal effects occur under experimental conditions (Sparschu s t a l . , 1970, 1971), and kidney anomalies (Courtney & Moore, 1971), intestinal haemorrhages and general oedema can be produced in the foetuses (Khera & Ruddick, 1973). Few follow-up studies of the effects of prenatal exposure on postnatal functions have been published. In mice, foetal kidney abnormalities caused by 2.3.7.8- tetra-CDD may progress into a hydronephrosis during the postnatal period (Moore s t a l . , 1973). Chick oedema disease - Hydropericardium, ascites, subcutaneous oedema, liver necrosis and death were described in 1957 following the accidental administration of toxic fats in the feed of broiler chickens (ascites have been observed also in mice) (IAKC, 1977). The toxic material was later identified in ccnmercial oleic and stearic acids produced from inedible tallcw recovered from animal hides; trichlorophenols and pentacMorqphenols had been used in the curing of the hides. The oedema causative was termed toxic fat and more specifically chick oedema factor and was characterized by X-ray crystallography as 1,2,3,7,8,9-hexa-CDD (Cantrell s t a l . , 1969) . 2.3.7.8- Tetra-CDD, hexa-CDD, and octa-CDD have been also identified in several carmercial fatty acids (Firestone, 1973). Daily doses of 10 or 100 ug hexa-CDD/kg tw, or of 1 or 10 yg 2,3,7,8-tetra-CDD/kg bw, produced a positive response in the chick oedema bioassay; 0.5% octa-CDD in the diet had no effect (Schwetz s t a l . , 1973). Similar oedematous effects were observed in rats, pigs, dogs, and monkeys, but not in guinea-pigs (Vos, 1978). Decreased serium albumin may be associated with the oedema; oedema has been also shewn to occur in chicks dosed orally with 2,3,7,8-tetra-CDF (McKinney e t a l . , 1976). 16 i H1 CD to ~1 TABLE 4" E v alu atio n o f rep o rte d em bryotoxic and te ra to g e n ic e f f e c ts induced by 2 ,3 ,7 ,8 -tetrach lo ro d ib en zo -p -d io x in s in r a ts and m ice S p e c ie s S train Em bryotoxic/ te ra to g e n ic e ffe c t Dose (p g /k g bw) low est, t e s t e d ` ~EDso`! Rat Mouse CD CD-I DBA/2J C57BI./6J NMRI In testin al haem orrhage K idney abnorm ality C left p alate K idney abnorm ality C left palate K idney abnorm ality C left palate K id n ey abnorm ality C left palate 0.125 0 .5 1? 3 1 3 3 3 3 3 9 15 5 0 .5 ? >1 >3 1-3 >3 >3 >3 <3 6.5 <9 40 15 Period of d o sin g (days) R o u te R eference o ra l S p a rs c h u et a l . , 1971 6-15 s.c . C o urtney & M oore, 1971 6-15 6-15 6-15 6-15 6-15 6-15 6-15 9-13 13 11 s.c . s.c s.c . s.c . s.c . s.c . o ra l oral o ra l o ra l M M 1 <1 M II II II H H II H N eubert & D illjra n n , 1972 ii M H N e u b e rt et'al., 1973 ii 1 n c,F ra ii N e u b e rt at a/. (1973) \ l i e low est dose w ith which an em bryotoxic o r te ra to g e n ic e f f e c t d e te c ta b le a t b ir th has been produced i s in d ic a te d . S in ce som etim es o n ly one dose le v e l was te s te d , t h is does n o t n e c e s s a rily re p re s e n t th e low est dose from which an e f f e c t could r e s u lt. cED5 0 - d o s e r e q u i r e d t o p r o d u c e a n e m b ry o to x ic e f f e c t i n 50% o f a n im a ls Other effects - In one or more species of laboratory animal, bone marrow hypoplasia, testicular degeneration, renal pelvis and urinary bladder hyperplasia and haemorrhage in the intestines and adrenal have been observed. Enzyme Induction 2.3.7.8- Tetra-CDD and other halogenated dibenzo-p-dioxins and dibenzo- furans stimulate a number of enzyme activities, most notably in the liver (IARC, 1977) . 2,3,7,8-Tetra-CDD is a potent inducer of hepatic and renal microsomal drug metabolizing enzymes (Beatty & Neal, 1976; Buu-Hol e t a l . , 1971, 1972b; Fowler e t a l . , 1975; Greig, 1972; Greig & DeMatteis, 1973; Hook e t a l . , 1975a,b; Lucier et a l . , 1973, 1975a,b; Woods, 1973). Intoxication with 2 , 2 , 1 , 8-tetra-CDD results in a marked increase in the cellular smooth endoplasmic reticulum content of hepatic and renal cells (Fowler e t a l . , 1973, 1975). 2,3,7,8-Tetra-CDD can simultaneously activate and suppress certain microscme-associated foreign-conpound and steroid- hormone-metabolising enzyme systems (Hook e t a l . , 1975a) as well as increase the activity of both renal and hepatic glutathione-S transferase (Kirsch e t a l . , 1975). 2.3.7.8- Tetra-CDD is the most active of the PCDDs in inducing hepatic 6-aminolevulinic acid (AIA) synthetase and aryl hydrocarbon hydroxylase (AHH) in chick embryo liver preparations (Poland & Glover, 1973a,b) . The PCDDs that induce ALA synthetase in chick embryo have two common properties: (1) halogen atoms occupy at least 3 of the 4 ring positions (2,3,7,8) , and (2) at least one free, non-halogenated carbon atom is unoccupied (Poland & Glover, 1973a). The available toxicological data (Schwetz e t a l . , 1973) indicate that those PCDDs that are lethal at lew doses, teratogenic or produce acne also induce ALA synthetase; those PCDDs that are not toxic generally do not induce ALA synthetase. The structureactivity relation of PCDDs to induce AHH in chick embryos was identical to that in inducing ALA synthetase (Poland & Glover, 1973a) . Mixed function oxidase enzyme systems of mouse strains 'non-responsive' to other aromatic hydrocarbons were induced by single doses of 2,3,7,8tetra-CDD, as evidenced by increases in hepatic monooxygenase activities >. 18 19272 and in concentrations of cytochrome P-448 (Chhabra et a l . , 1974; Kouri & Nebert, 1977; Nebert et a l . , 1975; Poland et a l . , 1974). Genetic resis tance to induction of AHH by 3-methylcholanthrene in DBA/2J mice was over come by treatment with 2,3,7,8-tetra-CDD. This result conflicts with the hypothesis that induction of AHH activity is a consequence of the formation of cytochrome P-448 (Chhabra et a l . , 1976) . A component that has a high binding affinity for 2,3,7,8-tetra-CDD was found in mouse liver cytosol (Poland et a l . , 1976a). ' 2.3.7.8- Tetra-CDD induces AHH even in 'poorly responsive' strains of mice (Niwa et a l . , 1975), not only in liver but also in lung, kidney and colon. The DBA/2N strain, which responds only weakly to the sarcanatogenic action of 3-methylcholanthrene, becomes highly susceptible after treatment with 2,3,7,8-tetra-CDD (Kouri, 1976; Kouri & Nebert, 1977). 2.3.7.8- Tetra-CDD is approximately 30,000 times more potent than 3-methylcholanthrene in inducing AHH activity in rat liver (Poland & Glover, 1974). McConnell e t a l . (1978a) reported that the comparative toxicity of 13 PCDDs in mice and guinea-pigs (see Table 2) supports the idea that the relative potency (or rank order) of a congener to produce one toxic response is a good indicator of its relative potency (or rank order) to produce other toxic manifestations. Hepatic Cytosol Binding Protein A macrcmolecular binding species has been characterized in the hepatic cytosol fraction of mouse and rat liver which has the in v itv o binding properties predicted for the receptor for the induction of AHH activity based on the in vivo biology (Poland e t a l . , 1976a). Namely: (1) 3H2,3,7,8-tetra-CDD binds to this cytosol protein reversibly with a high affinity (Kd = 0.27 nM) comparable to the ED for hepatic AHH induction 5 (ED^ in mice = 1 nmole/kg fcw); (2) the binding affinity of halogenated dibenzo-p-dioxins and dibenzofurans for this protein in v itv o corresponds to their potency to induce hepatic AHH activity in the chicken embryo; (3) other compounds, such as the polycyclic aromatic hydrocarbons, which induce AHH activity and cytochrome P^-450 also compete for this cytosolic 19 19273 binding protein, but compounds which induce other types of microsanal monooxygenase activities (e.g., phenobarbital) and steroids fail to bind. Thus, this cytosolic binding protein may be the receptor for the induction of AHH activity. Structure-Activity Relationships The pathologic effects produced by the toxic PCDD and PCDF isomers are similar to those of 2,3,7,8-tetra-CDD for a given species, differing only in the intensity of the toxic effect produced by a'given isaner. The toxic PCDDs have chlorine atoms in at least three of the four lateral ring positions (2,3,7 and 8) with at least one unsubstituted ring position (the octa-CDD is comparatively inactive). To the extent that the toxicity has been determined, a similar structure-toxicity relationship has been observed for the PCDFs. The structure-activity relation established for PCDDs and PCDFs for the induction of hepatic aryl hydrocarbon hydroxylase (AHH) activity and for binding to the hepatic cytosol binding species has been extended to other classes of chlorinated aranatic compounds. 3,4,3'41-Tetrachloroazcxybenzene (TCAOB) and 3,4,3',41-tetrachloroazobenzene (TCAB) are potent acnegens formed as trace contaminants in the synthesis of 3,4-dichloroaniline or herbicides based on this compound (Poland e t a t . , 1976b). At high doses in animals, TCAB is reported to produce thymic involution and liver damage similar to 2,3,7,8-tetra-CDD (Du Pont Inc., 1978). Both TCAOB and TCAB are potent inducers of hepatic AHH activity and bind to the hepatic cytosol binding protein with a high affinity. Congeners such as 3,5,3'5'-tetrachloroazoxybenzene and -azobenzene fail to induce AHH activity, fail to bind to the hepatic cytosol species, and fail to produce chloracne. Of 16 halogenated biphenyl ccrnpounds tested, only 3,4,3'4'-tetrachloro-, 3,4,5,3'4'5'-hexachloro- and 3,4,5,3'4'5'-hexabramo-biphenyls induced hepatic AHH activity and bound to the hepatic cytosol binding species (Poland & Glover, 1977). The 3,4,314'-tetrachlorobiphenyl has been reported to produce chloracne. McKinney e t a l. (1976) found that of five hexachlorobiphenyls tested in chickens, 3,4,5,3'4'5'-hexachlorobiphenyl was by far the most toxic, and the only one that induced signi ficant chick oedema and involution of the thymus. 20 Pharmacokinetics In the rat, following acute or chronic administration, 2,3,7,8-tetraCDD is accumulated primarily in the liver and to a lesser extent in the fat, and largely eliminated unmetabolized in the faeces with a whole bodyhalflife of about 3 waeks. Sane pharmacokinetic experiments suggest the formation of a polar metabolite appearing in the urine, but direct attempts to demonstrate metabolism with hepatic microscmes in v itr o have been nega tive (IARC, 1977). Mutagenicity Only four dioxin isomers have been evaluated for mutagenicity: the 2,7-di-, 2,3,7,8-tetra, and octa-CDDs as well as the unsubstituted dibenzo-p-dioxin (Wasson e t a l . , 1978). 2,3,7,8-Tetra-ODD increased the reversion frequency to stxeptarycin independence in Escherichia c o li Sd-4. In Salmonella typhimurivm, frameshift mutations in strain TA1532, but not base substitutions in strain TA1530, were induced by toxic concentrations of 2,3,7,8-tetra-CDD (Hussain e t a l . , 1972). In plate assays, the response was positive with S. typhimurivm TA1532, doubtful with TA1531 and TA1534, and negative with G46 and TA1530 (Seiler, 1973). McCann (1978) tested 2,3,7,8-tetra-CDD in S. typhimurivm, both with and without metabolic activation, using a spot-test and the standard plate test with strains UU532, T3U535, ,mi537 and TA1538; all these tests were negative. Octa-CDD was non-mutagenic in S. typhimurivm strains G46, TA1530 and TA1531, and doubtful results were obtained with strains TA1532 and TA1534 (Seiler, 1973). Metabolic activation systems were not included in any of these microbiological assays. Inhibition of mitosis and chromosomal abnormalities (dicentric bridges and chromatin fusion with formation of multinuclei or a single large nucleus) were observed in endosperm cells of the African blood lily (Haemantkvs Kqtherinae Baker) treated with 2,3,7,8-tetra-CDD in the presence or absence of 2,4,5-T (Jackson, 1972). 21 No chranoscmal aberrations ware observed in bone-marrow cells of male rats treated with 2,7-di-CDD, 2,3,7,8-tetra-CDD, or dibenzo-p-diaxin by oral intubation, with 2,3,7,8-tetra-CDD by intraperitoneal injection or orally (Green & Moreland, 1975). However, when Osborne Mendel rats of both sexes were treated twice weekly for 13 weeks with 2,3,7,8-tetra-CDD, a significant but weak increase in the number of chremoscme aberrations in bone marrow cells was reported (Green e t a l . , 1977) . 2,3,7,8-Tetra-CDD did not induce dominant lethal mutations in Wistar rats after oral administration to males for 7 days (Khera & Ruddick, 1973) . Carcinogenicity Two reports indicate that chronic administration of low levels of 2.3.7.8- tetra-CDD to rats is associated with an increased incidence of neoplasia (IARC, 1977; Kociba e t a l . , 1978; Van Miller e t a l . , 1977). Groups of 10 male Sprague-Dawley rats were fed a diet containing 2.3.7.8- tetra-CDD for 78 weeks in the following amounts (figures in parentheses are approximate weekly doses) : 0,1 part per trillion (0.0003 yg/kg kw), 5 ppt (0.001), 50 ppt (0.01), 500 ppt (0.1), 1 part per billion (0.4), 5 ppb (2.0), 50 ppb (24), 500 ppb (240), and 1000 ppb (500) . The three highest dose levels (50, 500, and 1000 ppb) were toxic and killed all animals by the fourth week, of the six remaining test groups, the overall incidence of neoplasms was 23/60 (38%); none occurred in the 1 ppt group. In the 5 ppt group, 5/10 animals had 6 neoplasms (ear-duct carcinoma, lymphocytic leukemia, adenocarcinoma, malignant histocytama (with metastases), angiosarcoma, Leydig-cell adenoma); the following groups also showed neoplasms: 50 ppt, 3 observed in 3/10; 500 ppt, 4 observed in 4/10; 1 ppb, 5 observed in 4/10; 5 ppb, 10 observed in 7/10. Neoplasms ware not observed in the controls (Van Miller e t a l . , 1977). Groups of 100 Sprague-Dawley rats (50 males and 50 females) received for two years diets containing 0, 22, 210, and 22,000 parts per trillion (ppt), equivalent to 0.0, 0.001, 0.01, and 0.1 yg 2,3,7,8-tetra-CDD/kg bw/day. Continuous ingestion of 0.001 yg/kg bw/day did not cause any-chsnicallyrelated changes in tumour incidence or toxicity; feeding with 0.01 yg/kg 22 192'" B bw/day induced an increased incidence (P<0.05) of hepatocellular hyperplastic nodules (female: 18/50 versus 8/86), of focal alveolar hyperplasia in the lungs, and of urinary excretion of porphyrins (female). Dietary intake of 0.1 ug/kg bw/day caused an increased incidence (P<0.05) of hepatocellular carcinomas (female: 11/49 versus 1/86) and squamous-cell carcinomas of the lung (female: 7/49 versus 0/86), of the hard palate/nasal turbinates (male: 4/50 versus 0/85; female: 4/49 versus 0/86), and of the tongue (male: 3/50 versus 0/85) . Further increased ware adenoma of the adrenal cortex (male) and-hepatocellular hyperplastic nodules (female). At this dose, certain age-related lesions were reduced (males: acinar adenoma of the pancreas; females: granulosal cell neoplasm of the ovary, benign and malignant tumours of the mammary gland, pituitary adenoma, and benign tumours of the uterus). Also chronic administration of 2,3,7,8-tetra-CDD caused multiple toxicologic effects,including increased mortality, decreased body weight gain, slight depression of certain hematologic parameters, increased urinary excretion of porphyrins and delta-aminolevu linic acid, increased serum levels of alkaline phosphatase, glutamyl transferase and serum glutamic pyruvic transaminase, and morphologic changes primarily of the hepatic, lymphoid, respiratory and vascular tissues of the body (Kociba e t a l . , 1978). These two reports shew that chronic administration of 2,3,7,8-tetraCDD causes an increased incidence of neoplasms, but not whether 2,3,7,8tetra-CDD acts as an initiator or promoter. This consideration is particularly important because unequivocal evidence is lacking that 2,3,7,8-tetra-CDD is a mutagen or is metabolized, and no evidence is available that 2,3,7,8-tetra-CDD and/or metabolite(s) bind covalently to macrcnplecules. As summarized in Table 5, at least 24 long-term carcinogenicity studies using mice and rats are currently in progress (IABC, 1978b) . Humans Toxic effects in humans - Toxicity due to 2,3,7,8-tetra-CDD has been reported after (i) occupational exposure during the industrial synthesis of 2,4,5-trichlorophenol(TCP) and 2,4,5-T, (ii) exposure in factories and in 23 TABLE 5 Ongoing long-term carcinogenicity testing of chlorinated dibenzo-p-dioxins1 Caipound Number of studies Route Species Dibenzo-p-dioxin 2,7-di-CDD 2,3,7-tri-CDD 2,3,7,8-tetra-CDD 1,2,3,6,7,8-hexa-CDD 1,2,3,7,8,9-hexa-CDD 1,2,3,4,6,7,8,9-octa-CDD 4 Oral (diet) :Mouse, rat Skin Mouse Mouse (with EMBA2) 4 Oral (diet) Mouse, rat Skin Mouse Mouse (with EMBA) 3 Oral (gavage) Mouse, rat Skin Mouse 5 Oral (diet) Rat Oral (gavage) Mouse Oral (gavage) Mouse, rat Skin Mouse 3 Oral (gavage) Mouse, rat Skin Mouse 2 Oral (gavage) Mouse Skin Mouse (with 1,2,3, 6,7,8-hexa-CDD) 3 Oral (diet) Mouse, rat Skin Mouse ^ t a taken fran Information Bulletin on the Survey of Chemicals Being Tested for Carcinogenicity, No. 7 (IARC, 1978b). Details of these studies as well as the authors/institutes are available in this source reference document. 2EMBA = dimethylbenzanthracene 24 the surrounding environment due to accidents occurring during the synthesis of TCP, and (iii) exposure to herbicides and other materials containing 2,3,7,8-tetra-CDD. Exposed subjects have been found to develop a wide variety of lesions and symptoms (Table 6). For instance, a typical exposure victim experiences a number and a variety of clinical signs and symptoms: e.g., early exposure symptoms may include a burning sensation of the eyes, nose, and throat followed by headache, dizziness, nausea, and vomiting. Sane days later, severe itching, redness, swelling of the face, more marked over the eyelids, nose and lips, may develop. Within the initial weeks' post-exposure, inflamed nodules as well as pustules appear cm the face, forearms, shoulders, neck, and trunk, leading then to comedones and cysts. After a month or two, acneform eruptions anerge and the skin becomes hyperpigmented. At about the same time, aching muscles, mainly in the thighs and chest area, become manifest and aggravated on exertion. Insomnia, extreme irritability and loss of libido also occur during this stage. Other than the consistently found clinical feature of acne, other findings in humans may include: neuromuscular symptoms (weakness and pain with nerve conduction abnormalities), porphyria cutanea ta rd a , hepatic dysfunctions, hyperlipidaemia, cutaneous hyperpigmentation and hirsutism, chronic eye irritation, emotional disorders, and neurcpsychiatric syndromes. Chloracne, one of the most constant and prominent features of 2,3,7,8tetra-CDD exposure, has been described as a refractory acne characterized by inclusion cysts, comedones and pustules, with eventual scarring of the skin, more frequently originating on the face and sometimes spreading to other parts of the body. Many patients also have blepharoconjunctivitis and irritation of other mucous membranes. Sometimes the chloracne is preceded by erythematous and oedematous skin lesions. The latent period between exposure and the appearance of clear signs of chloracne ranges from a few weeks to several months (May, 1973) . An important and unique episode revolves around three scientists who were self-exposed to 2,3,7,8-tetra-CDD: 'A' heated trichlorophenol in an alkaline solution, 'B' heated potassium trichlorcphenate, and 'C' worked in the same laboratory as 'B' and used a diluted solution of the synthesized dioxin (Oliver, 1975). 'A' and 'B' developed chloracne eight weeks after 25 19273 TABIiJ 6 Toxic e ffe c ts o f 2 ,3 ,7 ,8 -tetrach lo ro d ib en zo -p -d io x in s in hunans EFFECTS KliFi-MSNCES Dermat o l o q i c a l c h lo ra c n e porphyria cutanea tard a hyperpigm entation and h irsu tism In te rn a l l i v e r damage*2 ra is e d se ru n h e p a tic enzyme le v e ls d iso rd e rs o f f a t m etabolism " " carbohydrate m etabolism cardiovascular d iso rd ers urinary tra c t " resp irato ry " p a n c re a tic " N euroloqical polyneuropathies (p erip h eral n e u ritis ) low er ex trem ity weakness s e n so ria l im pairm ents (s ig h t, h earin g , a n e ll, ta s te ) P sychiatric n eu ra sth e n ic o r d e p re ssiv e syndromes Bauer e t a l . , 1961; B ert e l a l . , 1976; B leib erg e t a l . , 1964; C a r te r e t a l . , 1975; Dugois e t a l , , 1968; F ire s to n e , 1978; G ia n o tti, 1977; Goldmann, 1972, 1973; Jen sen & W alker, 1972; J i r S s e k e t a l , , 1973, 1974; KiJimig f, S c h u lz , 1957a; May, 1973; O l i v e r , 1975; P o lan d a t a l . , 1971; S u s k in d , 1951 B le ib e r g e t a l . , 1964; J i r S s e k e l a l . , 1973, 1974, 1976 B le ib erg e t a l . , 1964; O liv e r, 1975; Poland e t a l . , 1971 Bauer e t a l . , 1961; B leib erg e t a l . , 1964; Dugois e t a l . , 1968; Goldmann, 1972, 1973; J ir S s e k e t a l . , 1973, 1974; May, 1973; Ton T h a t, 1977 B ert e t a l . , 1976; B le ib erg e t a l . , 1964; J i r l s e k e t a l . , 1973, 1974; May, 1973; P o lan d e t a l . , 1971; Ton T h a t, 1977 J i r l s e k e t a l . , 1973; O liv e r, 1975; P oland e t a l . , 1971 Goldmann, 1972, 1973; J ir S s e k e t a l . . 1973, 1974; Poland e t a l . , 1971 Goldmann, 1972, 1973; J ir S s e k e t a l . , 1973, 1974 C a r te r e t a l . , 1975; Goldmann, 1972, 1973 Bauer e t a l . , 1961; Goldmann, 1972, 1973 Goldmann, 1972, 1973 Goldmann, 1972, 1973; J ir S s e k e t a l . , 1973, 1974 Bauer e t a l . , 1961; F ire s to n e , 1978; Goldnann, 1972, 1973; e t a l . , 1973, 1974; O liv e r, 1975; Poland e t a Z .,1 9 7 1 Goldnann, 1972, 1973; O liv e r, 1975; Poland e t a l . , 1971; Ton T h a t, 1977 JirS se k Bauer e t a l . , 1961; F ir e s to n e , 1978; Goldmann, 1972, 1973; J i r S s e k e t a l . , 1973, 1974; O l i v e r , 1975; P o lan d e t a l . , 1971 ^M ild f i b r o s i s , f a t t y c h a n g e s , h a a n o f u s c in d e p o s it io n and parenchym a1- c e 11 d e g e n e r a tio n w ere o b s e rv e d in a few c a s e s . exposure, whereas 'C' showed no evidence of the characteristic acneform lesions. Delayed symptoms, probably due to 2,3,7,8-tetra-CED, developed approximately two years later and 'B' and 'C' showed: personality changes (mainly loss of energy and drive); impairment of vision, taste and muscular coordination; sleep disturbances; gastrointestinal symptoms; and hirsutism. 'A' experienced none of these adverse effects. All three exhibited hypercholesterolaania (>300 mg/100 ml) . Human exposure to PCDDs and PCDFs - Major sources of human exposure to PCDDs and PCDFs include: - exposure in the workplace; - exposure in factories and in the surrounding environment from industrial accidents; - exposure to contaminated materials, wastes, or food in the general environment; and - exposure in Vietnam and other intensive herbicide spraying operations. (Prior and subsequent to use as a defoliant, 2,4,5--T was used in weed-killing and forest-thinning operations in the United States of America and elsewhere). Occupational exposure may occur: - in manufacturing plants producing chlorinated phenols (tri-, tetra-, and pentachlorophenols), or phenoxy acid herbicides (2,4-D, 2,4,5-T), or PCBs; - in factories utilising these chemicals for the production of other substances (hexachlorophene iron 2,4,5-trichlorophenol); - in factories manufacturing or repairing transformers and capacitors or having heat exchange or heat hydraulic systems containing PCBs; and - in the use processes of these chemicals under various occu pational conditions such as spraying of herbicides, using chlorinated phenols for a variety of applications (especially wood preservative), sawing or otherwise processing treated wood, and using hexachlorophene in sanitary occupations. 27 The burning of materials impregnated with oarenercial 2,3,4,6-tetrachlorophenates yielded 150-1000 yg of mixed PCDDs/g chlorophenate. Although only found as a minor constituent, 2,3,7,8-tetra-GDD has been quantified at levels exceeding 10 yg/g chlorophenate (Rappe e t a l . , 1978b). Pyrolysis of a technical grade PCBs-mixture yielded many PCDF isaners; the total yield could be as high as 3-25%. One of the resin constituents is 2,3,7,8-tetra-CDF, the most toxic PCDF-iscmer. Consequently, uncontrolled burning of PCBs can be an important environmental source of the hazardous PCDFs, and operations such as welding or soldering electrical equipment containing PCBs, or using casting waxes in foundries, may possibly entail a significant exposure (Buser e t a l . , 1978a). A PCB used in a heat exchange system for two years contained approximately 1.25 ppm of 2,3,7,8-tetra-CDF and a total of 15 ppm of PCDFs (Buser e t a l . , 1978c). Apart iron accidents such as the one which occurred in Seveso in 1976, general environmental exposure may originate from herbicide spraying and waste disposal. An outbreak of PCDDs poisoning in humans, horses, and other animals occurred in Missouri in 1971 (Beale e t a l . , 1977; Carter e t a l . , 1975) following the spraying of contaminated oil for dust control in horse arenas. Another possible source of exposure to PCDDs and PCDFs pollution are waste oils, and possibly other waste, when burned both in industrial and municipal incinerators. Under simulated environmental conditions, the combustion of a standard 2,4,5-T formulation led to formation of small amounts of PCCDs and PCDFs (Ahling e t a l . , 1977). Recently, Buser & Bosshardt (1978) quantified that the total amount of PCDDs and PCDFs in fly ash from an industrial incinerator heating facility was 0.2 ppm and 0.1 ppn, and in the fly ash from an industrial heating facility in Switzerland was 0.6 ppm and 0.3 ppm. More than 30 individual, PCDDs can be identified in the fly ash, but the known highly toxic PCDD iscmers are only minor consti tuents (Buser e t a l . , 1978b). The number of PCDF iscmers was larger, but, in this case, the known highly toxic iscmers are major constituents (Buser e t a l . , 1978d). 28 2 J An additional potential source of human exposure has been revealed recently. Beef fat taken iron cattle grazed on 2,4,5-T-treated rangeland > contained 2,3,7,8-tetra-CDD; of the 11/14 positive samples, the four with the highest levels had 12, 20, 24, and 70 ppt (Meselson e t a l . , 1978). In another study, 3/24 samples of beef fat contained 3-4 ppt 2,3,7,8-tetra-CDD (Kocher e t a l . , 1978). Moreover, in a preliminary report of an ongoing enlarged study of women living in areas where 2,4,5-T is used on rangeland, 4/18 breast milk samples contained approximately i ppt each (Meselson e t a t . , 1978) . Workers in wood processing industries are exposed to wood dust contain ing preservatives as well as accompanying impurities and degradation products. Wood dust fran a saw-mill in which a 2,3,4,6-tetrachlorophenol formulation was used as a fungicide was found to contain 1-10 ppm PCDFs and <0.5 ppm PCDDs (levin e t a l . , 1977). In 1968, more than 1200 persons in South-west Japan were intoxicated by consuming a commercial rice oil contaminated with 1000 ppm PCBs. Nagayama e t a l . (1976) analyzed the rice oil (Yusho oil) and found 5 ppm PCDFs, the major constituents of which were tetra- and penta-CDFs. Buser e t a l . (1978c) have recently shewn that 2,3,7,8-tetra-CDF was the main PCDF in the Yusho oil (0.45 ppm). The high level of PCDFs was caused by leakage fran heat exchangers containing PCBs contaminated with PCDFs. Workers manufacturing 2,4,5-trichlorophenol(TCP) or 2,4,5-T during normal production operations and/or following explosions taking place in these plants, may have been exposed to a variety of polychlorinated chemicals whose type and quantity depends on the particular chemical processes in use and on the phase of the reaction in which the accident took place. Apparently, for instance, PCDFs are produced in the earlier stages when mainly tetrachlorobenzene is present and PCDDs are produced nearer the end of the reaction when primarily trichlorophenol is extant. The concentration of chlorophenols and chlorobenzenes, for instance, was probably higher in the 1963 episode in The Netherlands, where the accident took place at the beginning of the reaction, than in the 1968 episode in the United Kingdom, or the 1976 accident in Italy, where the explosion took place at the end. 29 192C3 A thorough description of the reported industrial accidents and other cases of intoxications observed in exposed workers is given in IARC Monographs' Volume 15 (IARC, 1977). In this present report, only those episodes specifically discussed at length by the NIEHS/IARC ad hoc Working Group will be considered. A sequential perspective of the known accidents is given here below (see also Table 7, pages 35-37). In 1949, the first reported cases of industrial poisoning due to the formation of 2,3,7,8-tetra-CDD in uncontrolled exothermic reactions occurred during the manufacture of 2,4,5-trichlorophenol at a 2,4,5-T-producing factory in Nitro, West Virginia; 288 persons ware affected (see Firestone, 1978; Suskind, 1978). In November 1953, an accident occurred in Ludwigshafen, Federal Republic of Germany, during the manufacture of TCP (Goldman, 1972, 1973; Hofmann, 1957; Thiess & Goldman, 1978); 53 workers were affected by chloracne. Five cases of chloracne were reported following an industrial acci dent in an Italian TCP-producing factory (Hofmann & Meneghini, 1962). In 1963, an accident occurred at the 2,4,5-T-producing factory in The Netherlands; approximately 50 persons vrere affected by chloracne (Dalderup, 1974; Hay, 1976). In 1966, Dugois e t a l . (1968) observed 21 cases of chloracne after an accident in a French factory producing TCP in the Grenoble region. In 1968, an accident occurred at the TCP-producing factory at Bolsover, Derbyshire, United Kingdom (Jensen & Walker, 1972; May, 1973). Within the next 7 months, 79 workers developed chloracne. In 1971, 2,3,7,8-tetra-CDD contaminated waste oil caused an outbreak of poisoning in humans, horses and other animals (Carter e t a l . , 1975; Kimbrough e t a l . , 1977). In July 1976, an accident at the TCP-producing factory in Meda, Italy (Bert et a l . , 1976) resulted in the contamination of a large, densely populated area, including the towns of Seveso, Mada, Cesano Maderno and Desio. t 1 30 19 204 Key information concerning the effects of human exposure to herbicides in Vietnam, especially to the so-called "Agent Orange"1 (a 50:50 mixture of the -butyl esters of 2,4-D and 2,4,5-T, containing up to 30 mg/kg or more 2,3,7,8-tetra-CDD), may be found in, among others, the foliating references: Carmittee on the Adverse Effects of Herbicides in Vietnam, 1974; IARC, 1977; Meselson e t a t . , 1972; Ton That, 1977; Westing, 1978. IV. EXPOSURE EPISODES CONSIDERED IN DETAIL BY THE NXEHS/IARC ad hoc WORKING GROUP The text below is a condensed version of the discussions which took place during the two-day NIEHS/IARC ad hoc Working Group Meeting: Phenoxyacids exposure Cohort studies on herbicide sprayers have been conducted or are being planned in Scandinavian countries. The more advanced of these concerns Swedish railroad workers with exposure to a variety of herbicides. These people exhibited a significantly increased tumour incidence (apparently dose-dependent) and tumour mortality (Axelson & Sundell, 1974). The excess of tumours was found particularly among workers with exposure to amitrole (aminotriazole) whereas those exposed to phenoxyacids (2,4-D and 2,4,5-T) shewed only a slightly increased excess of cancers. The study was small sized, carprising 2978 person-years at observation in the total cohort and with 18 deaths versus 20.54 expected. The original conclusion from this study was that amitrole exposure may have caused an excess of tumours, whereas there was probably no pertinent increase in tumour incidence associated with exposure to phenoxyacids. The study has been recently reanalysed using a case-control approach, and through stratification on amitrole when considering the effect from phenoxyacids, and v ice versa. The results shew a possible and previously masked tumour-inducing effect also from phenoxyacids. 1Mone than 2,000,000 gallons of Agent Orange remaining from military defoliant operations were destroyed by incineration on board ship in the Pacific Ocean, 120 miles iron Johnston Island. Temperatures not lewer than 1250C were used. The steel containers are to be melted (Walsh, 1977). 31 19205 Another retrospective cohort mortality study, conducted in Finland on workers of five companies involved in spraying 2,4-D and 2,4,5-T on brushwood, did not show ary increase in mortality. During 1955-1971, in the younger group of workers (under 45 years of age), however, 4 cancer deaths were observed versus less than 2 expected (no statistically significant difference). A prospective follow-up study for the period 1972-1976 revealed fewer cancer-related deaths than expected in all age groups. Clinical and anamnestic investigations were also performed shewing a picture of acute complaints during and following the spraying operations: headache, transient dizziness, fatigue, abdominal complaints, skin and mucous irritations and a few cases of persistant papulae. It is interesting to note that sane samples of 2,4,5-T used in Finland for spraying operations contained from 0.04 to 0.07 ppm 2,3,7,8tetra-CDD. This may imply that the 2,-4,5-T produced before 1965 contained lower levels of 2,3,7,8-tetra-CDD impurities than has been observed subse quently from 2,4,5-T used elsewhere. However, the analytical methods used in the determination of 2,3,7,8-tetra-CDD in the Finnish samples were antiquated in view of the rapid advancements made in analytical methodology and thus a confirmatory analysis of the samples using modem techniques was made by Rappe e t a l . (1978a). Five samples of 2,4,5-T ester dating from 1962 to 1967 were analysed for HDDs and PCDFs; levels of 0.1-0.95 ppm 2,3,7,8-tetra-CDD and 0.1-0.15 ppm tetra-CDF were fcjund (Rappe e t a l . , 1978a). A second feasibility study was designed and conducted in Finland to determine whether one could obtain anamnestic, clinical, haematological, and immunological.information on Finnish railway and forestry workers who had been exposed for several years to herbicides emtaining 2,3,7,8-tetra-CDD. The exposed group consisted of 30 men who were control-matched to persons of the same age caning from the same district with similar job backgrounds and living conditions. Of the tests made, including liver and immunofunction studies, no differences were observed between the exposed and control groups. Although the population study is too small to make any general conclusions as to the potential adverse health effects from long-term exposure to phenoxyacids, the investigation shows that follow-up is possible 32 19206 on the health status of persons exposed for several years to 2,3,7,8-tetraCDD-containlng phenoxyacids, and then to conpare the exposed group results to carefully matched controls. The study is in progress. Hexachlorophene exposure An excess of malformations, sane severe, has been reported among children whose mothers were employed as nurses in a hospital. The mothers were exposed to hexachlorophene soap during early pregnancy, and the hypo thesis of a causal-relation between such an exposure and the occurrence of the malformations has been advanced (Hailing, 1977a); five severe and six slight malformations were observed in 65 children in the exposed group, whereas only one slight malformation was observed in the 68 children of the unexposed group. This report has been followed by another study (to be published) of a similar group of children of exposed mothers in capari son to children of unexposed mothers. Again a high frequency of malfor mations among the offspring of exposed mothers was reported (Hailing, 1977b) Frcm these studies, however, it is not clear if exposure to other hazardous chemicals could be excluded. A retrospective-prospective study is being initiated among long-tem workers of an United States' factory which used Seveso-produced 2,4,5-trichlorophenol in the manufacture of hexachlorophene. However, in the case of hexachlorophene, polychlorinated xanthenes have been also identified as contaminants at a higher level than 2,3,7,8-tetra-CDD (Baughman, 1974). Phenoxyacids and chlorophenols exposure in forestry workers and in wood industry At the Regional Hospital in Umea (Northern Sweden) , Hardell (1977) observed that several patients suffering from mesenchymal tumours reported occupational or other exposures to phenoxyacids. More specifically, 87 mesenchymal tumours were diagnosed during the years 1970 through 1976. Of these cases, 55 were men (more than expected) and 43 of these 55 men were known by profession: 19 ware either forestry workers, farmers and forestry workers, or workers in sawmills and paper-mills where exposure to chloro phenols is cannon. Based on the official statistics of Sweden, one can calculate apprcximatively the expected fraction of tumours within these trades, resulting in an expectancy of approximately 11 cases versus the 19 observed. 33 1920 n The Mount Sinai School of Medicine in New York is currently planning a field survey on long-term health effects on wood-preservative workers in Arkansas and Oklahoma in the United States of America. 2,3,7,8-Tetra-CDD exposure during 2,4,5-T or polychlorinated phenols production and following industrial accidents The following cases, as well as others that have not been considered at the ireeting, have been described in the IABC Monographs' Volume 15 (IARC, 1977) and only some of the previously available information will be given here. The ongoing epidemiological studies on these cohorts are summarized in Table 7. As each individual cohort has a relatively small number of person-years, a study has been proposed in which all the cohorts would be pooled and the mortality compared to that expected from the national statistics. The first reported cases of industrial poisoning due to the formation of 2,3,7,8-tetra-CDD in uncontrolled exothermic reactions occurring during the manufacture of TCP were seen in 1949 at a 2,4,5-T-producing factory in Nitro, West Virginia, USA; 228 persons were affected. These individuals were studied by Suskind & Ashe (1978) and followed-up for a period of four years. Symptoms included chloracne, nausea, vomiting, headaches, severe muscular aches and pain, fatigue, emotional instability and intolerance to cold. Laboratory findings shewed raised total lipids and raised prothrombin time. Among those affected were not only workmen, but also laboratory personnel, medical personnel, and even the Safety Director who visited the areas of exposure. Several wives who had never visited the plant also developed acne, usually at the sane time as their employee husbands. It is important to note that information on the toxicity as well as on the stability (long persistence and slew degradability) of 2,3,7,8tetra-CDD was not available at the time the first reports of acute toxicity were observed, and no measures were taken to decontaminate the factories and control the residue levels of 2,3,7,8-tetra-CDD. In fact, the presence of 2,3,7,8-tetra-CDD or other PCDDs and PCDFs was not even suspected. Workers, therefore, were probably exposed for a considerable time to 2,3,7,8-tetra-CDD following the first observations of acute toxic effects. 34 19238 CD tO CO CD T7U3J 7 E pidaniological stu d ie s on 2 ,3 ,7 ,8 -le tr. rhlorixlibonxo-p-djuxin Hunan e x posure e p iso d e s caused by a c c id e n ta l o r o c c u p a tio n a l e x p o su re o c c u rrin g i n f a c t o r i e s producing 2,4,5-trichlorophenol and/or 2,4,5-trichlorophenoxyaoetic acid Town and C ountry Year of a c c id e n ta l o r occu p a tio n a l exposure Nufiber of exposed persons C lin ic al follow-up and cohort stu d ies Observatio n s todate Proposed actions Ccmnents and reoonm endations o f th e NiniS/IARC .</ l/.v- Working Group (a) A ccid en tal ex p o su res1 N itro , West V irg in ia , United S ta te s of America 1949 I.udwigaha fe n / R hein, Federal Republic of Germany 1953 Amsterdam, `Hie N eth erlan d s 1963 228 Only 36 in d iv id u a ls No s i g n i f i c a n t lo n g -term A m ortality cohort e ffo rts should be nude to trace w ere fo llo w e d , fo r d is e r v a t io n s ire y e t a v a i l stu d y i s being th e in d iv id u a ls who were a period of 4 able (s a ir follow l-up pi annul. exposed, p o s s ib ly through Utc years. in d iv id u a ls had sypt.cms o f employment can**nsation f i l e s . severe in terco stal n eu ritis). 75 A m o r t a lit y c o h o rt 6 C ancer d e a th s wore o b s e r To c o n tin u e Lim stu d y h as been con ved : 3 o f stcxnach cancer fo l lOW-up. d u c te d 24 y e a rs in th e age group 60-69 a f te r the accid en t; (s ig n ific a n tly hig h er than a l l exposed workers ex pected), 2 o f o a t-c e ll were traced fo r a carcinoma o f tile lung and to t a l o f 1525 1 o f co lo n ic adonocarciiXJiw person-years of observation. H ie r e s u l t s o f Ux* oont inuing follow-U|> stu d y s Ix x j U I p e r io d ic a lly bo rule a v a i la b le . 106 P re liiu in a ry o b s e r Of 8 d e a tiis , 5 o r 6 were A fu ll rojxirt is v a tio n s on m o r ta lity from c a r d io v a s c u la r d is e a p 1an n u l fo r 1978 a re a v a i la b le ; 93 s e s - 5 p ro b a b ly fr a n Hiyo- exposed workers have c ard in al in fa rc tio n s; the been traced up to nyocardial in farctio n 1977. death r a te seems h ig h e r in th e most h e av ily exfiosed workers a ffe c te d by ch lo r- a cn e . One w orker, who d id not develop chloracne, died fra p an creatic ciircirwna 14 m onths a f t e r Ukj a c c i d e n t; hfjwever, lie cun- p la in e d o f abduidnal |>ain p rio r to the accid en t. E f f o r ts should tx> made to l ra c e a l l th e w orkers who were exposed. UU1) 'moiu d o ta i I s aixxit tiicso a c c id e n ts a r e y iv e n in th e t e x t , s t a r t i i v i on [ y e 34, and in IAIiC (1977). CUX)\ Table 7 - continued Bolsover, D erbyshire, U n ite d Kingdom 1968 Meda, B rianza d i Seveso, Ita ly 1976 90 Of th e w orkers One w orkers i s known to have An e p id e m io lo g ic a l E f f o r ts sh o u ld be made t o t r a c e b e lo n g in g t o th e o r i d ie d from c o ro n a ry thrombo c o h o rt stu d y w i l l be a l l th e w o rk ers who were g in a l c o h o r t, 50% a re s i s . planned when th e exposed. s t i l l enployed and re s u lts o f the pre ro u tin ely undergo se n t c lin ic a l follow c lin ic a l and labora up in v e stig atio n tory in vestigations. boexire a v a ila b le . ? 2 A c l i n i c a l fo llo w -u p The f i r s t c l i n i c a l o b se rv a Among o th e r ongoing An in v e s t ig a ti o n s h o u ld be made o f th e in h a b ita n ts tio n s showed a high in c i investigations, a of the p o ssib le te ra to g e n ic and fr e n th e more c o n ta dence o f 2 , 3,7,8-tetra-C D D - long-term m o rbidity o th e r ad v erse re p ro d u c tiv e and m inated areas is dependent s k in le s i o n s among stu d y h a s been i n i t i developm ental e f f e c t s from being c a r r ie d o u t. c h ild re n liv in g in th e more a te d . 2,3,7,8-tetra-C D D exposure. A At th e tim e o f th e contam inated a re a. A lso, a m o rtality re g is te r and popula accident, liv in g in nunber of persons residing tion file s , su itab le for ade Zone A was a popula in th e zo n e surrounding the quate d e scrip tio n s o f m orbidity tio n o f 730 p eo p le, more contam inated a re a s had d ata, should be e sta b lish e d . w ith 200 o f th o se skin lesio n s, but w ith Further, the canoer re g istry being between 0-14 poorly d efin ed c lin ic a l th a t already covers a nearby y e a rs o f a g e . From f e a tu r e s ; th e s e le s i o n s may province, should be extended to th e se 730 p e rso n s, be th e r e s u l t o f 2 ,3 ,7 ,0 - include the Seveso area. 623 (125 betw een 0-14 tetra-C D D e x p o su re . More y ears o f age) were ov er, both c lin ic a l and follow ed-up. In Zone la b o ra to ry fin d in g s su g g est B liv ed a population an increased incidence of o f 4732 p e o p le , w ith h e p a t ic s u ffe re n c e among 1293 between th e ages a d u lts liv in g in th e h ig h ly 0-14 y e a r s . An accu c o n ta m in ated a r e a s . A p r e ra te oount o f those lim inary m o rtality study follow ed in Zone B is su g g ests th a t althouqh th e n o t a v a ila b le , b u t o v e r a ll r a te s from 2 Seveso 943 y o u n g sters b e t towns a re n o t d if f e r e n t from ween 0-14 y e a rs o f a nearby a re a , an in c re a se age a re in clu d ed in o f d eath s from liv e r c irrh o the follow -up. s i s and leukemia was noted. 2No r e l i a b l e in fo rm a tio n was a v a i la b le t o th e Working Group a b o u t th e a c t u a l nunber o f exposed p e rs o n s ; th e e n t i r e a r e a h as a p o p u la tio n o f approxim ately 200,000 people. Table 7 continued (b) O ccupational exposure C zechoslovakia 1965-1969 80 A fo llo w -u p h a s been F iv e p a t i e n t s d ie d d u rin g The follow -up w ill Hie r e s u lts o f th e co n tin u in g made r e g u la r ly on 55 th e f i r s t 5 y e a r -fo llc w -u p c o n tin u e . A 1 0 -y e a r fo llo w -u p stu d y sh o u ld w orkers o f th e o r i p e rio d : 2 d ie d from bron r e p o r t w ill be p r e p e rio d ic a lly be made a v a ila b le . g in a l c o h o rt. The chogenic carcincxna and 1 pared during 1978. workers lo s t to from a ra p id ly developing follow -up a re m ainly a ty p ic a l a th e ro sc le ro s is . fo r e ig n e r s who l e f t One f u r t h e r d e a th , p ro b a b ly the country. from liv e r c ir r h o s is , has r e c e n t ly o c c u r re d . Many o f th e w orkers a r e now s u f f e r ing from h ypertension and show s ig n s o f h y p e rlip i- daemia aixl h y p e rc h o le ste ro - laesnia a s w ell as p re d ia b e tic synptans and sig n s. OJ Ct D 'CO CD r' In Novenber 1953, an accident occurred at a TCP factory in Ludwigshafen, Federal Republic of Germany; 75 workers were exposed to the reaction products during the accident and the subsequent cleaning work; most were affected by chloracne, 42 severely: 21 of the 42 suffered consequent damage to internal organs or disturbances of the nervous system The most relevant features were polyneuritis, sensorial impairments and liver damage. The son of one of the workers developed chloracne following contact with his father's workclothes. An additional case of poisoning occurred 5 years later in a worker who was involved in repair work (welding one of the auto claves) at the contaminated site; subsequently this worker died with necrotic pancreatitis. All the 75 workers could be traced in a cohort study 25 years later; mortality was compared both with the regional and national mortality and with a control group of workers iron the same factory. In the exposed group, 17 deaths were observed (11 to 25 expected depending on the control population); 6 from cancer (4 or less expected), 5 from cardiovascular diseases (as expected), 2 from suicides (0.2-0.6 expected), 1 from liver cirrhosis, 1 from a urogenital tract disease, and 2 from traffic accidents. Of the 6 cancer deaths, 3 ware from stomach cancer and occurred in the age group 60-69, which is significantly more than expected. This effect was more pronounced when a minimum observation period of 10 years was considered, and more so for stomach cancer after a 20-year observation period. (Use of a long observation period permits a better estimation of the relation between exposure and disease when long "latency" periods are involved, as for cancers). A similar accident occurred at the 2,4,5-T-producing factory in The Netherlands in 1963; 106 men are likely to have been exposed to 2,3,7,8-tetra-CDD (and possibly PCUFs) as they were in the building where the accident occurred during the subsequent period, March-July 1963. These workers may be divided into three groups: A: factory employees (exposed during all phases of the cleaning and reconstruction activities) - 44 men; 38 1 192^2 B: clean-up crew (non-factory personnel, exposed during the months May-July) - 18 iren; and C: plumbing, painting and insulating personnel (non-factory workers, exposed during the months May-July) - 44 men During the explosion in Amsterdam, 2 operators were present in the building. With the exception of these men, who were probably exposed primarily by the inhalation route, all other workers used safety masks and were probably exposed by dental contact. Chloracne was the most common and prominent lesion observed following 2,3,7,8-tetra-CDD exposure (26/44 examined in the A, 10/16 in the B, and 8/16 in the C groups) . Liver function tests (thymol turbidity and serum glutamic pyruvic trans aminase and serum glutamic oxaloacetic transaminase) did not indicate liver damage. A few men complained of fatigue. The latent period between exposure and the appearance of chloracne, characterized by comedones, pustules and cysts on the face and sometimes on other parts of the body, was approximately 4-6 weeks. Sometimes, erythematous and oedematous skin lesions, possibly due to phenolic compounds, were noted one day after exposure. No skin lesions (chloracne) were reported in family members of the employees. Of the A group, 3/44 workers have died, 1 of pancreas carcinoma in 1964, 1 of nyocardial infarction at the age of 69, and 1 of a traffic accident. Of the 17/18 traced individuals in the B group, 4 have died (3 from sudden death, probably of nyocardial infarction, at ages 41, 53, and 65, and 1 of ill-defined pulmonary or cardiac cause at age 44). Only 32/44 of the C group were traced and 1 of them died of sudden death, probably nyocardial infarction, at age 50. Jircisek e t a l. (1973, 1974, 1976) studied 55 subjects of a cohort of 78 workers from a factory in Czechoslovakia producing 2,4,5-T and pentachlorophenol, who ware affected by chloracne. The workers were probably exposed during the years 1965-1968 to unknown levels of 2,3,7,8-tetra-CDD. A significant symptom was the disturbance in porphyrins metabolism with signs of porphyria cutanea tarda. Many workers also suffered severe hepatic and neurological damage and had raised blood levels of cholesterol and total lipids. In the first years of follow-up, 5 deaths were observed, 2 from 39 bronchogenic carcinoma at age 47 and 59 (only 0.12 lung cancer deaths were expected from national mortality statistics), 1 iron atypical athero sclerosis at age 57, and 1 fran the acute occupational intoxication. One further death, probably from liver cirrhosis, has recently been observed and 50 workers are still being followed. Mariy workers continue to shew lipaemia and hypercholesterolaania and have developed prediabetic changes and hypertension. The workers lost to follow-up are mostly foreigners who returned to their country of origin. In July 1976, an accident in a TCP-producing factory in Meda, Italy resulted in the contamination of the towns of Seveso, Meda, Cesano Mademo, Desio, Nova Milanese and possibly others of a large, densely populated area called the Brianza di Seveso with a total population of about 220,000 inhabitants. A retrospective health survey conducted on the workers of the ICMESA factory revealed that two workers among those directly involved in the production of 2,4,5-trichlorophenol suffered from skin lesions previous to July 1976. These facts may be indicative of a previous 2,3,7,8-tetra-CDD exposure. A large and intensive follow-up programme is in progress in the Seveso area; data were presented about the first dermatological, neurological, clinical and laboratory findings, as well as about foetal damage. Dermato logical examinations, including the screening of 32,000 children, resulted in the finding of hundreds of children with skin lesions, 135 of which were eventually identified as chloracne; 59 cases were fran the more contaminated area (A zone). However, precise description of diagnostic criteria and accurate anamnestic information, including, among others, data on places of residence, were not included. This information is needed before evaluating the meaning of the chloracne incidence data by territorial areas and efforts should be made obtain it. Chloracne cases were not systematically searched for in adults. Neurological examinations showed both signs of subclinical neurological damage and cases of clinically detectable polyneuropathy in adults (34/338 persons fran A zone and 8/185 from the surrounding areas). 40 i 19234 In about 30% of 1654 adults fran both A zone and nearby zones, hepato megaly was reported to be present on clinical investigation. No information, however, is given on the criteria by which the hepatomegaly was evaluated. The same percentage of abnormalities was observed in one or more liver tests (mainly glutamyl transferase and serum glutamic pyruvic transaminase and serum glutamic oxaloacetic transaminase). Extensive monitoring of various haematological parameters from 25,000 persons is new in progress and the data are expected to became available during 1978. Results of this investi gation should allow a better understanding as to whether or not these alter ations are caused by exposure to 2,3,7,8-tetra-CDD and PCDDs. So far, imnunological investigations, cytogenetic research, and embryomorphology analysis on cases of therapeutical abortions have not given abnormal results. However, the study of the frequency of congenital malformations and spontaneous abortions seems to be very difficult to accom plish. The importance of a systematic study on this subject has been stressed by the Working Group, because the questions on the teratogenicity of 2,3,7,8-tetra-CDD in humans cannot be solved by any other study. A preliminary mortality study has been possible in 2 of the 11 towns of the Seveso area. The 1975-1977 mortality rates from liver cirrhosis and leukemia in these two limited zones of the Seveso area were compared to that of a nearby area (province of Varese). Within an overall mortality which did not appear different, an increase of deaths from liver cirrhosis and from leukemia was noted. The validity and significance of these observations should be carefully evaluated. V . RECCMMENDATICNS A chemical analytical programme should be encouraged to identify and standardize analytical methods used, to catalogue products contaminated with PCDDs and PCDFs, and to establish a register of known PCDD and PCDF standard samples. The Working Group unanimously recommended that a system be developed for an international exchange of information and research coordination on the health effects of chlorinated dibenzo-p-dioxins/chlorinated dibenzofurans. Further, and in particular, the following three areas were indi cated as deserving special attention: 1. Development of sensitive and reliable measurements of body burden; improved, more sensitive, and less expensive analytical technology should be generated. Standardized sampling (air, water, soil, tissues and body fluids, environmental), preparation, clean-up, and analysis should be promoted; 2. Development of cannon protocols for clinical examinations (including reproductive experience) and, in particular, design of a protocol for basic core information; and 3. Development of an international registry of exposed persons to serve as a basis for long-term follow-up: this was considered especially important because the relatively small size of popula tions involved in individual exposure episodes are an obstacle to risk assessment and thus pooling of data is almost a necessity. The Working Group recommended that IAEC explore the technical and financial feasibility, possibly in cooperation with other national and/or international organizations, of setting-up such a permanent mechanism which could include as a minimum starting programme: a) a periodical, perhaps biennial, meeting to up-date the progress of ongoing studies, and to identify needed developments; b) an exchange of relevant information accruing during the interval between meetings; and c) a register of chloracne cases, containing simple, essential infor mation in a standardized form. This PCDDs/PCDFs exposed-related sign may help to better define an exposed cohort population suitable for long-term follow-up. 1S2S6 V I . REFERENCES Part 1 : References cited in the text Ahling, B., Lindskog, A., Janssen, B., Sundstran, G. (1977) Formation of polychlorinated dibenzo-p--dioxins and dibenzofurans during combustion of a 2,4,5-T formulation. Chemosphere, _8, 461-468 Allen, J.R., Barsotti, D.A., van Miller, J.P., Abrahamscn, L.J. & Lalich, J.J. (1977) Morphological changes in monkeys consuming a diet containing low levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Food Cosmet. Toxicol., 15, 401-410 Axelson, 0., Sundeil, L. (1974) Herbicide exposure, mortality and tumor incidence. An epidemiological investigation on Swedish railroad workers. Wbrk-environ. Health, 11, 21-28 Bauer, H., Schulz, K.H. & Spiegelberg, U. (1961) Berufliche Vergiftungen bei der Herstellung von Chlorphenol-Verbindungen. Arch. Gewerbepathol. Gewerfaehyg., 18, 538-555 Baughman, R.W. (1974) Tetrachlorodihenzo-p-dioxins in the Environment. High Resolution Mass Spectrometry at the Picogram Level (Thesis, Harvard University) Beale, M.G., Shearer, W.T., Karl, M.M. & Rcbsan, A.M. (1977) Long-tenn effects of dioxin exposure. Lancet, i, 748 Beatty, P.W. & Neal, R.A. (1976) Induction of DT-diaphorase activity of rat liver by 2,3,7,8-tetrachlorodibenzo-p-dioxin (Abstract No. 232) . Toxicol, appi. Pharmacol., 37, 189 Bert, G., Manacorda, P.M. & Terracini, B. (1976) I controlli sanitari: la sanit incontrollata. Sapere, 796, 50-60 Blair, E.H., ed. (1973) Chlorodioxins - Origin and Fate. Advances in Chemistry Series 120, Washington DC, American Chemical Society, 141 pages Bleiberg, J., Wallen, M., Brodkin, R. & Applebaum, I.L. (1964) Indus trially acquired porphyria. Arch. Dermatol., 89, 793-797 Buser, H.-R. (1977) Determination of 2,3,7,8-tetrachlorodibenzo-p-dioxin in environmental samples by high-resolution gas chromatography and lew resolution mass spectrometry. Anal. Chem., 49, 918-922 Buser, H.R. & Bosshardt, H.-P. (1978) Polychlorierte Dibenzo-p-dioxine, Dibenzofurane und Benzole in der Asche karnnunaler und industrieller Verbrennungsanlagen. Mitt. Gebiete Lebensmittelunters. Hyg., 69, 191-199 45 Buser, H.R. & Rappe, C. (1978) Identification of substitution patterns in polychlorinated dibenzo-p-dioxins (PCDDs) by nass spectrcmetry. Cheroosphere, 1_, 199-211 Buser, H.R., Bosshardt, H.-P. & Rappe, C. (1978a) Formation of poly chlorinated dibenzofurans (PCDFs) fron the pyrolysis of PCBs. Chemosphere, 1_, 109-119 Buser, H.R., Bosshardt, H.-P. & Rappe, C. (1978b) Identification of poly chlorinated dibenzo-p-dicKin iscmers found in fly ash. Chemosphere, 7, 165-172 Buser, H.R., Rappe, C. & Gara, A. (1978c) Polychlorinated dibenzofurans (PCDFs) found in Yusho oil and in used Japanese PCB. Chanosphere (in press) Buser, H.R., Bosshardt, H.-P., Rappe, C. & Lindahl, R. (1978d) Identifi cation of polychlorinated dibenzofuran iscmers in fly ash and PCB pyrolyses. Chanosphere (in press) Buu-Ho, N.P., Hien, D.-P., Saint-Ruf, G. & Servoin-Sidoine, J. (1971) Proprits cancrariimtiques de la ttrachloro-2,3,7,8 dibenzo-pdicKine. C.R. Acad. Sei. (Paris), 272, 1447-1450 Buu-Hol, N.P., Chanh, P.-H., Sesqu, G., Azum-Gelade, M.C. & Saint-Ruf, G. (1972a) Organs as targets of 'dioxin' (2,3,7,8-tetrachlorodibenzo-pdioxin) intoxication. Naturwissenschaften, 59, 174-175 Buu-Hol, N.P., Chanh, P.-H., Sesqu, G., Azum-Gelade, M.C. & Saint-Ruf, G. (1972b) Enzymatic functions as targets of the toxicity of 'dioxin' (2,3,7,8-tetrachlorodibenzo-p-diaxin). Naturwissenschaften, 59, 173-174 ~ Cantrell, J.S., Wbb, N.C. & Mabis, A.J. (1969) Identification and crystal structure of hydropericardium-producing factor; 1,2,3,7,8,9-hexachlorodibenzo-p-diaxin. Acta CcystaTIpgr., Sect. B25 (1), 15-156 Carter, C.D., Kimbrough, R.D., Liddle, J.A., Cline, R.E., Zack, M.M., Jr, Barthel, W.F., Koehler, R.E. & Phillips, P.E. (1975) Tetrachlorodibenzo-diaxin: an accidental poisoning episode in horse arenas. Science, 188, 738-740 Chhabra, R.S., Tredger, J.M., Philpot, R.M. & Fouts, J.R. (1974) Selection of inducers: an important factor in characterizing genetic differences to induction of aryl hydrocarbon hydroxylase in strains of mice. Life Sei., 15, 123-130 Chhabra, R.S., Tredger, J.M., Philpot, R.M. & Fouts, J.R. (1976) Relation ship between induction of aryl hydrocarbon hydroxylase and de novo synthesis of cytochrome P-448 (Pi-450) in mice. Chem.-biol. Interact., 15, 21-31 ---- 1----------- 46 13228 Committee on the Effects of Herbicides in Vietnam (1974) The Effects / of Herbicides in South Vietnam, Part A, Summary and Conclusions, ^ Washington DC, National Acadeny of Sciences Courtney, K.D. & Moore, J.A. (1971) Teratology studies with 2,4,5-trichlorophenoxyacetic acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Pharmacol., 20, 396-403 Dalderup, L.M. (1974) Safety measures for talcing down buildings contami nated with toxic material. H. T. soc. Geneeskd., 52, 616-623 Dugois, P., Amblard, P., Aimard, M. & Deshors, G. (1968) Acn chlorique collective et accidentelle d'un type nouveau. Bull. Soc, frang. Derm. Syph., 75, 260-261 Du Pont Inc. (1978) Personal communication to Dr A. Poland. Finney, D.J. (1964) Statistical Method in Biological Assay, New York, Hofner, pp. 524-531 Firestone, D. (1973) Etiology of chick edema disease. Environ. Health, 5, 59-66 Firestone, D. (1978) The 2,3,7,8-tetrachlorodibenzo-parg-dioxin problem: a review. In: Ramel, C., ed., Chlorinated Phenoxy Acids and their Dioxins: Mode of Action, Health Risks and Environmental Effects. Eool. Bull. (Stockholm), 27, 39-52 Fcwler, B.A., Lucier, G.W., Brown, H.W. & McDaniel, O.S. (1973) Ultrastructural changes in rat liver cells following a single oral dose of TCDD. Environ. Health Perspect., 5_, 141-148 Fowler, B.A., Hook, G.E.R. & Lucier, G.W. (1975) Tetrachlorodibenzo-pdioxin induction of renal microscmal enzyme systems. Toxicol, appl. Pharmacol., 33, 176-177 Gianotti, F. (1977) Chloracn au ttrachloro-2,3,7,8-dibenzo-p-dioxine chez les enfants. Ann. Dermatol. Venereol., 104, 825-829 Goldmann, P.J. (1972) Schwerste akute Chlorakne durch TrichlorphenolZerSetzungsprodukte. Arbeitsmed. Sozialmed. Arbeitshyg., 1_, 12-18 Goldmann, P.J. (1973) Schwerste akute Chloracne, eine Massenintoxikation durch 2,3,6,7-Tetrachlordibenzodioxin. Der Hautarzt, 24, 149-152 Gosselin, R.E., Hodge, H.C., Smith, R.P. & Gleason, M.N. (1976) Clinical Toxicology of Commercial Products, 4th ed., Baltimore, Williams & Wilkins Co., 1800 pages Green, S. & Moreland, F.S. (1975) Cytogenetic evaluation of several dioxins in the rat (Abstract No. 99). Toxicol, appl. Pharmacol., 33, 161 47 19239 Green S., Moreland, F. & Sheu, C. (1977) Cytogenetic effect of 2,3,7,8tetrachlorcriibenzo-p-dioxin on rat bone marrow cells. FDR By-Lines, No. 6, 292-294 Greig, J.B. (1972) Effect of 2,3,7,8-tetrachlorodibenzo-l,4-dioxin on drug metabolism in the rat. Biochem. Pharmacol., 21, 3196-3198 Greig, J.B. & De Matteis, F. (1973) Effects of 2,3,7,8-tetrachlorodibenzop-dicocin on drug metabolism and hepatic microsctnes of rats and mice. Environ. Health Perspect., 5_, 211-219 Gupta, B.N., Vos, J.G., Moore, J.A., Zinkl, J.G. & Bullock, B.C. (1973) Pathologic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin in laboratory animals. Environ. Health Perspect., 5, 125-140 Hailing, H. (1977a). Misstdnkt samband me1Ian hexaklorofenexposition och missbildningsbflrd (Suspected link between exposure to hexachlorophene and birth of malformed infants). T.Hkartidninqen, 74, 542-546 Hailing, H. (1977b) Personal cannunication to Dr 0. Axelson. Hardell, L. (1977) Maligna mesenkymala tumflrer och exposition fftr fenoxisyror - en klinisk observation (Malignant mesenchymal tumors and exposure to phenoxy acids - a clinical observation). LSkartidningen, 74, 2753-2754 Harris, M.W., Moore, J.A.,Vos, J.G. & Gupta, B.N. (1973) General biologi cal effects of TCDD in laboratory animals. Environ. Health Perspect., .5, 101-109 Hay, A. (1976) Toxic cloud over Seveso. Nature (Lond.), 262, 636-638 Hofmann, H.T. (1957) Neuere Erfahrungen mit hoch-toxischen Chlorkohlenwasserstoffen. Naunyn-Schmiedeberq's Arch. exp. Pathol. Pharmakol., 232, 228-230 Hofmann, M.F. & Meneghini, C.L. (1962) A preposito delle follicolosi da idrocarburi clorosostituiti (acne clorica). G. Ital. Derm., 103, 427-450 Hook, G.E.R., Haseman, J.K. & Lucier, G.W. (1975a) Induction and suppres sion of hepatic and extrahepatic microsomal foreign-ccnpound-metabolizing enzyme systems by 2,3,7,8-tetrachlorodibenzo-p-diaxin. Chem.-biol. Interact., 110, 199-214 Hook, G.E.R., Orton, T.C., Moore, J.A. & Lucier, G.W. (1975b) 2,3,7,8TetrachloroLhenzo-p-dioxin-induced changes in the hydroxylation of biphenyl by rat liver microsames. Biochem. Pharmacol., 24, 335-340 Huff, J.E. & Wassam, J.S. (1973) Chlorinated dibenzodiaxins and dibenzofurans. Environ. Health Perspect., 5, 283-312 48 19300 Huff, J.E. & Wasson, J.S. (1974a) Health hazards from chemical impurities: chlorinated dibenzodioxins and chlorinated dibenzofurans. Int. J. Environ. Stud., 6_, 13-17 Huff, J.E. & Wasson, J.S. (1974b) Hazardous Contaminants : Chlorinated Dibenzodioxins and Chlorinated Dibenzofurans. In: Savage, E.P., e d . , Environmental Chemicals; Human and Animal Health, II, 175-197 Hussain, S., Ehrenberg, L . , Lfroth, G. & Gejvall, T. (1972) Mutagenic effects of ICE on bacterial systems. Arribio, 1, 32-33 IARC (1977) IARC Monographs on the Evaluation of *the Carcinogenic Risk of Chemicals to Man, 15, Sane Fumigants, the Herbicides 2,4-D and 2.4.5- T, Chlorinated Dibenzodioxins and Miscellaneous Industrial Chemicals, Lyon, 354 pages IARC (1978a) IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, 18, Polychlorinated Biphenyls and Polybrcminated Biphenyls, Lyon (in press) IARC (1978b) Information Bulletin on the Survey of Chemicals Being Tested for Carcinogenicity, No. 7 , Lyon, 460 pages Jackson, W.T. (1972) Regulation of mitosis. LEI. Cytological effects of 2.4.5- trichlorophenoxyacetic acid and of dioxin oontaminants in 2.4.5- T formulations. J. cell. Sei., 1 0 , 15-25 Jensen, N.E. & Walker, A.E. (1972) Chloracne: three cases. Proc. roy. Soc. M e d . , 65, 687-688 Jirsek, L . , Kalensky, J. & Kubeck, K. (1973) Acne chlorina and porph yria cutanea ta rd a during the manufacture of herbicides. Cs. Dermatol., 48, 306-317 Jirsek, L . , Kalensky, J . , Kubeck, K . , Pazderov, J. & Lukas, E. (1974) Acne chlorina, porphyria cutanea tarda and other manifestations of general intoxication during the manufacture of herbicides. II. Cs. Dermatol., 49, 145-157 Jirsek, L., Kalensky, J., Kubeck, K . , Pazderov, J. & Lukas, E. (1976) Chlorakne, porphyria cutanea tarda und andere Intoxikationen durch Herbizid. Hautarzt, 27, 328-333 Jones, G. (1975) A histochemical study of the liver lesion induced by 2,3,7,8-tetrachlorodibenzo-p-diaxin (dioxin) in rats. J . Pathol., 116, 101-105 ---------- Jones, G. & Butler, W.H. (1974) A morphological study of the liver lesion induced by 2,3,7,8-tetrachlorodibenzo-p-diaxin in rats. J . Pathol., 112, 93-97 49 193o i . Khera, K.S. & Ruddick, J.A. (1973) Polychlorodibenzo-p-diaxins: perinatal effects and the dominant lethal test in Wistar rats. In: Blair, E. H . , ed., Chlorodioxins - Origin and Fate. Advances in Chemistry Series 120, Washington DC, American Chemical Society, pp. 70-84 Kimbrough, R.D., Carter, C.D., Liddle, J.A., Cline, R.E. & Phillips, P.E. (1977) Epidemiology and pathology of a tetrachlorodibenzodiaxin poisoning episode. Arch, environ. Health, 32, 77-86 Kirnnig, J. & Schulz, K.H. (1957a) Berufliche Akne (sog. Chlorakne) durch chlorierte aromatische zyklische ther. Dermatologia, 115, 540-546 Kirrmig, J. & Schulz, K.H. (1957b) Chlorierte aromatische zyklische ther als Ursache der sogenannten Chlorakne. Naturwissenschaften, 44, 337-338 Kirsch, R . , Fleischner, G . , Kamisaka, K. & Arias, I.M. (1975) Structural and functional studies of ligandin, a major renal organic anion binding protein. J. clin. Invest., 5 5 , 1009-1019 Kocher, C.W., Mahle, N.H., Humnel, R.A., Shadoff, L.A. & Getzendaner, M.E. (1978) A search for the presence of 2,3,7,8 tetrachlorodibenzo-pdioxin in beef fat. Bull, environ. Contain. Toxicol., 1 9 , 229-236 Kociba, R.J., Keyes, D.G., Beyer, J.E., Carreon, R.M., Wade, C.E., Dittenber, D.A., Kalnins, R.P., Frausan, L.E., Park, C.N., Barnard, S.D., Hummel, R.A. & Huniston, C.G. (1978) Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dicxin in rats. Tcxicol. appl. Pharmacol, (in press) Kouri, R.E. (1976) Relationship between levels of aryl hydrocarbon hydroxylase activity and susceptibility to 3-methylcholanthrene and benzo (a)pyrene induced cancers in inbred strains of mi c e . In: Freudenthal, R.I. & Jones, P.W., eds, Polynuclear Aromatic Hydro carbons: Chemistry, Metabolism and Carcinogenesis, Vol. 1, New York, Raven Press, pp. 139-151 Kouri, R.E. & Nebert, D.W. (1977) Genetic regulation of susceptibility to polycyclic hydrocarbon-induced timers in the m o u s e . In: Hiatt, H.H., Watson, J.D. & Winsten, J.A., eds, Origins of Hunan Cancer, Book B, Cold Spring Harbor, N.Y., Cold Spring Harbor laboratory, pp. 811-835 Levin, J.O. & Nilsson, C.A. (1977) Chrcmatographic determination of poly chlorinated phenols, phenoxyphenols, dibenzofurans and dibenzodioxins in wood-dust fron worker environments. Cheroosphere, , 443-448 Lucier, G. W ., McDaniel, O.S., Hook, G.E.R., Fowler, B.A., Sonawane, B.R. & Faeder, E. (1973) TCDD-induced changes in rat liver microsomal enzymes. Environ. Health Perspect., 5_, 199-209 50 19302 Lucier, G. W . , McDaniel, O.S. & Hcok, G.E.R. (1975a) Nature of the enhance ment of hepatic uridine diphosphate glucuronyltransferase activity by 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Biochem. Pharmacol., 24, 325-334 Lucier, G . W . , Sonawane, B . R . , McDaniel, O.S. & Hook, G.E.R. (1975b) Post natal stimulation of hepatic microsanal enzymes following administration of TCDD to pregnant rats. Chem.-biol. Interact., 1 1 , 15-26 May, G. (1973). Chloracne from the accidental production of tetrachlorodibenzodioxin. Br. J. ind. Me d . , 30, 276-283 McCann, J. (1978) Unpublished data as reported in Wasscrn, J.S., Huff, J.E. & Loprieno, N. (1978). A review of the genetic toxicology of chlori nated dlbenzo-p-dicocins. Mutat. Res, (in press) McConnell, E., Moore, J.A., Haseman, J. & Harris, M. (1978a) The conparative toxicity of chlorinated dibenzo-p-diaxin in mice and guinea-pigs. Toxicol, appl. Pharmacol, (in press) McConnell, E.E., Moore, J.A. & Dalgard, D.W. (1978b) Toxicity of 2,3,7,8- tetrachlorodibenzo-p-diaxin in rhesus monkeys (Maoaaa mulatto.) following a single oral dose. Toxicol, appl. Pharmacol., 4 3 , 175-187 McKinney, J.D., Chae, K . , Gupta, B.N., Moore, J.A., Goldstein, J.A. (1976) Toxicological assessment of hexachlorobiphenyl iscmers and 2,3,7,8tetrachlorodibenzofurans in chicks. I. Relationship of chemical parameters. Toxicol, appl. Pharmacol., 36, 65-80 McNulty, W . P . (1977) rhesus monkeys: 108-109 Toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin for brief report. Bull, environ. Contain. Toxicol., 18, Meselson, M.S., Westing, A.H. & Constable, J.D. (1972) Background material relevant to presentations at the 1970 Annual Meeting of the AAAS. American Association for the Advancenent of Science, Herbicide Assess ment Canmission, 92nd Congress, 2nd Session, US Congressional Record, 118, S3227-S3233 Meselson, M . , O'Keefe, P.W. & Baughman, R. (1978) The evaluation of possible health hazards frcm TCDD in the environment. In: Presentation for Symposium on the Use of Herbicides in Forestry, Arlington, Virginia (in press) Moore, J.A. (1978) Personal coimunication. Moore, J . A . , Gupta, B. N . , Zinkl, J.G. & Vos, J.G. (1973) Postnatal effects of maternal exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) . Environ. Health Perspect., 7, 81-85 51 _LJWWu - 3 Moore, J.A., Gupta, B.N. & Vos, J.G. (1976) Toxicity of 2,3,7,8-tetraehlorodibenzofuran - preliminary results. In: National Conference on Poly chlorinated Biphenyls, 1975, Chicago, EPA-560/6-75-004, Washington DC, Environmental Protection Agency, pp 77-80 Nagayama, J., Kuratsune, M. & Masuda, Y. (1976) Determination of chlori nated dibenzofurans in Kanechlors and "Yusho oil". Bull, environ. Contain. Toxicol., 15, 9-13 Nebert, D . W . , Robinson, J.R., Niwa, A., Kumaki, K. & Poland, A.P. (1975) Genetic expression of aryl hydrocarbon hydroxylase activity in the mouse. J. cell. Physiol., 85, 393-414 Neubert, D. & Dillmann, I. (1972) Snbryotoxic effects in mice treated with 2,4,5-trichlorcphenoxyacetic acid and 2,3,7,8-tetrachlorodibenzo-pdioxin. Naunyn-Schmiedeberg's Arch. exp. Pathol. Pharmakol., 272, 243-264 Neubert, D., Zens, P . , Rothenwallner, A. & Merker, H.-J. (1973) A survey of the snbryotoxic effects of ICED in mamnalian species. Environ. Health Perspect., 5_, 67-79 NIEHS (National Institute of Environmental Health Sciences) (1973) Con ference on Chlorinated Dibenzodioxins and Dibenzofurans, April 1973, Research Triangle Park, North Carolina. Environ. Health Perspect., 5, 1-313 Niwa, A., Kumaki, K., Nebert, D.W. & Poland, A. (1975) Genetic expression of aryl hydrocarbon hydroxylase activity in the mouse. Arch. Biochem. Bioplys., 166, 559-564 Oishi, S., Merita, M. & Fukuda, H. (1978) Comparative toxicity of poly chlorinated biphenyls and dibenzofurans in rats. Toxicol, appl. Pharmacol., 43, 13-22 Olie, K . , Vermeulen, P.L. & Hutzinger, 0. (1977) Chlorodibenzo-p-dicodns and chlorodibenzofurans are trace components of fly ash and flue gas of same municipal incinerators in The Netherlands. Chsnosphere, 8, 455-459 Oliver, R.M. (1975) Toxic effects of 2,3,7,8-tetrachlorodibenzo 1,4dioxin in laboratory workers. Br. J. ind. M e d . , 3 2 , 49-53 Poland, A. & Glover, E. (1973a) Chlorinated dihenzo-p-dicodns: potent inducers of -aminolevi.il inic acid synthetase and aryl hydrocarbon hydroxylase. II. A study of the structure-activity relationship. Mol. Pharmacol., % 736-747 Poland, A. & Glover, E. (1973b) 2,3,7,8-Tetrachlorodibenzo-p-dioxin: a potent inducer of -aminolevulinic acid synthetase. Science, 179, 476-477 52 . -1 Poland, A. & Glover, E. (1974) Comparison of 2,3,7,8-tetrachlorodibenzop-dioxin, a potent inducer of aryl hydrocarbon hydroxylase, with 3-methylcholanthrene. Mol. Pharmacol., 10, 349-359 Poland, A. & Glover, E. (1977) Chlorinated biphenyl induction of aryl hydrocarbon hydroxylase activity: a study of the structure-activity relationship. Mol. Pharmacol., 13, 924-938 Poland, A.P., Smith, D., Metter, G. & Possick, P. (1971) A health survey of workers in a 2,4-D and 2,4,5-T plant with special attention to chloracne, porphyria autanea ta rd a , and psychologic param eters. Arch, environ. Health, 22, 316-327 Poland, A.P., Glover, E . , Robinson, J.R. & Nebert, D.W. (1974) Genetic expression of aryl hydrocarbon hydroxylase activity. Induction of mono-oxygenase activities and cytochrome Pj-450 formation by 2,3,7,8tetrachlorodibenzo-p-dioxin in mice genetically 'non-responsive ' to other aranatic hydrocarbons. J. biol. Chem., 249, 5599-5606 Poland, A.P., Glover, E. & Kende, A.S. (1976a) Stereospecific, high affinity binding of 2,3,7,8-tetrachlorodibenzo-p-diaxin by hepatic cytosol. J. biol. Chan., 251, 4936-4946 Poland, A., Glover, E . , Kende, A.S., DeCamp, M. & Granasmenico, G.M. (1976b) 3,4,3',41-Tetrachloro azoxybenzene and azobenzene: potent inducers of aryl hydrocarbon hydroxylase. Science, 194, 627-630 Ramel, C., ed. (1978) Chlorinated Phenoxy Acids and their Dioxins: Mode of Action, Health Risks and Environmental Effects. Ecol. Bull. (Stockholm), 2 7 , 302 pages Rappe, C . , Buser, H.R. & Bosshardt, H.-P. (1978a) Identification and quanti fication of polychlorinated dibenzo-p-diaxins (PCDDs) and dibenzofurans (PCDFs) in 2,4,5-T-ester formulations and herbicide orange. Chemosphere, (in press) Rappe, C., Marklund, S., Buser, H.R. & Bosshardt, H.-P. (1978b) Formation of polychlorinated dibenzo-p-diaxins (PCDDs) and dibenzofurans (PCDFs) fcy burning or heating chlorophenates. Chemosphere, _3 269-281 Schulz, K.H. (1968) Zur Klinik und Atiologie der Chlorakne. Arbeitsmed. Sozialmed. Arbeitshyg., 3, 25-29 Schvretz, B.A., Norris, J.M., Sparschu, G.L., Rowe, V.K., Gehring, P.J., Onerson, J.L. & Gerbig, C.G. (1973) Toxicity of chlorinated dibenzop-diaxins. Environ. Health Perspect., 5_, 87-99 Seiler, J.P. (1973) A survey an the mutagenicity of various pesticides. Experientia, 29, 622-623 53 Smith, F . A . , Schwetz, B.A. & Nitschke, K.D. (1976) Teratogenicity of 2.3.7.8- tetrachlorodibenzo-p-diaxin in CF-1 mice. Tcocicol. appl. Pharmacol., 38, 517-523 Sparschu, G.L., Dunn, F.L. & Rowe, V.K. (1970) Teratogenic study of 2.3.7.8- tetrachlorodihenzo-p-dicadn in the rat. Toxicol, appl. Pharmacol., 17, 317 Sparschu, G.L., Dunn, F.L. & Rewe, V.K. (1971) Study of the teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-diaxin in the rat. Food Cosinet. Toxicol., 9_, 405-412 Suskind, R.R. (1951) Hie chemistry of the human sebaceous gland. I. Histocheraical observations. J. invest. Dermatol., 1 7 , 37-54 Suskind, R.R. (1978) Personal caiitiunicaticn. Suskind, R.R. & Ashe, W.F. (1978) Unpublished reports. Personal corrmunication. Thiess, A.M. & Goldmann, P. (1978) Ober das Trichlorphenol-Dioxin-Unfallgeschehen in der BASF A G vom 13. Novanber 1953. In: Vortrag. auf dim IV. Msdichem-Kongress, Haifa, 1976 (in press) Thigpen, J. E . , Faith, R.E., McConnell, E.E. & Moore, J.A. (1975) Increased susceptibility to bacterial infection as a sequela of exposure to 2.3.7.8- tetrachlorodibenzo-p-diaxin. Infect. Imnun., 12, 1319-1324 Ton That, T. (1977) Pathologic humaine at animale de la dioxins. Rev. Med. (Paris), 1 8 , 653-657 Van Miller, J.P., Lalich, J.J. & Allen, J.R. (1977) Increased incidence of neoplasms in rats exposed to low levels of 2,3,7,8-tetrachlorodibenzo-p-diaxin. Chemosphere, , 537-544 Vos, J.G. (1977) Dtmune suppression as related to toxicology. CRC Crit. Rev. Toxicol., 5, 67-101 Vos, J.G. (1978) 2,3,7,8-Tetrachlorodibenzo-pgra-dioxin: effects and mechanisms. In: Ramel, C . , e d . , Chlorinated Phenoxy Acids and their Dioxins: Mode of Action, Health Risks and Environmental Effects. Ecol. Bull. (Stockholm), 27, 165-176 Vos, J.G. & Moore, J.A. (1974)' Suppression of cellular iirmunity in rats and mice by maternal treatment with 2,3,7,8-tetrachlorodibenzo-pdiaxin. Int. Arch. Allergy appl. hrmunol., 47, 777-794 Vos, J. G . , Moore, J.A. & Zinkl, J.G. (1973) Effect of 2,3,7,8-tetrachlorodibenzo-p-diaxin on the imrtune system of laboratory animals. Environ. Health Perspect., 5_, 149-162 54 Vos, J.G., Moore, J.A. & Zinkl, J.G. (1974) Toxicity of 2,3,7,8-tetrachlorodibenzo-p-diaxin (TCDD) in C57B1/6 mice. Toxicol, appl. Pharmacol., 29, 229-241 Vos, J.G., Kreeftenberg, J.G., Engel, H.W.B., Minderhound, A. & Van Noorle Jansen, L.M. (1978) Studies on 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced iirmune suppression and decrease resistance to infection:endotoxin hypersensitivity, serun zinc concentrations and effect of thymosin treatment. Toxicology (in press) Walsh, J. (1977) Odyssey of Agent Orange ends in the Pacific. Science, 197, 966-967 Wasson, J. S . , Huff, J.E. & Loprieno, N. (1978) A review of the genetic toxicology of chlorinated dibenzo-p-dioxins. Mutat. Res, (in press) Weissberg, J.B. & Zinkl, J.G. (1973) Effects of 2,3,7,8-tetrachlorodi benzo-p-dioxin upon hemostasis and hematologic function in the rat. Environ. Health Perspect., _5, 119-123 Westing, A.H. (1978) Ecological considerations regarding massive environ mental contamination with 2,3,7,8-tetrachlorodibenzo-para-dioxin. In: Ramel, C., ed . , Chlorinated Phenoxy Acids and their Dioxins: Mode o f Action, Health Risks and Environmental Effects. Ecol. Bull. (Stockholm), 27, 285-294 Woods, J.S. (1973) Studies of the effects of 2,3,7,8-tetrachlorodibenzo-pdioxin on manmalian hepatic ^-aminolevulinic acid synthetase. Environ. Health Perspect., 5_, 221-225 Zinkl, J. G . , Vos, J.G., Moore, J.A. & Gupta, B.N. (1973) Hematologic and clinical chemistry effects of 2,3,7,8-tetrachlorodibenzo-p-diqxin in laboratory animals. Environ. Health Perspect., 5_, 111-118 55 i g q r,7- _Z_ j Part II - Working papers (For more details, contact authors) Arstila, A.CJ. & Sorvari, T.E., A n epidemiological feasibility study on the effects of 2,3,7,8-tetrachlorodibenzo-pgra-diaxin (TCDD) o n Finnish forestry, road and railway workers: preliminary findings. Axelson, 0., A review on Swedish epidemiologic studies with relations to chlorinated dibenzodiaxins. Axelson, 0. (1978), Aspects on confounding in occupational health ,epidemiology. Scand. J. Work Environ. Health, 4 85-89 Berrino, F., Preliminary report on mortality in the Seveso area, 1975-77. Buser, H.R., Bosshardt, H.-P. & Rappe, C., Formation of polychlori nated dibenzofurans (PCFs) fran the pyrolysis of PCBs. Dew Chemical Catpany, Preliminary assessment of chronic toxicity study and three-generation reproduction study of 2,3,7,8-tetrachlorod ibenzo-p-dioxin. Frentzel-Beyme, R., Coordination of epidemiological studies on the long-term hazards of chlorinated dibenzodiaxin. May, G., T.C.D.D. in the U.K. Moore, J.A., Comparative toxicity of dibenzodiaxins and dibenzofurans. Pocchiari, F . , Accidental TCDD contamination in Seveso (Italy): Epidemiological aspects. Poland, A., Chlorinated dibenzo-p-diaxins and dibenzofurans: review of toxicology and enzyme induction and consideration of mechanism of toxicity. Rappe, C., Chemical aspects of polychlorinated dibenzodiaxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) Riihim!ki, V., Asp, S., SeppSlMinen, A.M. & Hembe r g , S., Symptomatology, morbidity and mortality experience of chlorinated phenoxyacid herbicide (2,4,-D; 2,4,5-T) sprayers in Finland. A clinical and epidemiological study. Selikoff, I.J., Epidaniological evaluation of the public health signi ficance of polychlorinated dibenzofurans. Suskind, R.R., Chloracne and associated health problems in the manu facture of 2,4,5-T. Thiess, A.M. & Frentzel-Beyme, R . , Mortality study of persons exposed to dioxin after an accident which occurred in the BASF on 13th November 1953. Vos, J.G., Sterringa, T.J., Zellenrath, D., Dalderup, H.J. & L.M . , TCDD accident at a chemical factory in The Netherlands.57 57 I9 3 i/8 xG. J. M. to vapors '.I. 1. (1975c). vapors of n c , J. M. sponse to ;n g , J. M. ; to vapors 1 N G , J. M. i to vapors .I N G . J. M. espunse to UNO, J. M. esponse to 9. 1 N E R . D . J., studies. IX. rm acol. 36. , AND .an response N , L. J.. A N D and human -456. N. L. J.. A N D and human -of threshold ues upon the r. h i d . H y g . ring repeated T O X IC O L O G Y A N D A P P L IE D P H A R M A C O L O G Y 3 6 , 4 9 1 - 5 0 1 (1] The D ose-D ependent Pharm acokinetic Profile of 2,4,5-Trichlorophenoxy A cetic Acid Following Intravenous A dm inistration to Rats M. W . Sauerhoff, W . H . Br a u n , G . E. Blau, and P. J. G ehring P harm acokineticsM etabolism Group, D ow L epetit. U .S.A ., and C om putations R esearch L aboratory. D ow C hem ical. U .S .A ., and T oxicology Research L a b o ra to r y , H e a lth a n d E n viro n m e n ta l R e se a rc h , D ow C h em ica l. U .S .A ., M idland, M ichigan 48640 R eceived Ju ly 3 ,1 9 7 5 ; accepted January, 2 ,1 9 7 6 The Dose-Dependent Pharmacokinetic Profile of 2,4,5-Trichlorophenoxy Acetic Acid Following Intravenous Administration to Rats. S a u e r h o f f , M . W ., B l a u , G . E ., B r a u n , W . H . , a n d G e h r in g , P . J . ( 1 9 7 5 ) . Toxicol. Appl. Pharmacol. 36, 491-501. The clearance of 2,4,5-trichlorophenoxy acetic acid (2,4,5-T) from plasma and its elimination from the body o f rats were determined after single intravenous doses o f [1'iG]2,-:-,5-T. Dosedependent pharmacokinetics were found. Urinary excretion of un changed 2,4,5-T accounted for approx. 95% of the li'' activity excreted from the body. The results show that the distributk. .'.id elimination o f 2,4,5-T are markedly altered when larger doses were v.. .ministered. The pharmacokinetic data indicate that the statistical pnyection o f resuk o f experiments with large doses o f 2,4,5-T, to predict the hazard o f - ;>osuto small at aunts, is no* justified because the capability of the body : handle the compound has been altered. 2.4.5-Trichlorophenoxyacetic acid (2,4,5-T) is - plant growth regulator and herbicide. The absorption, metabolism, and excretion of 2,4,5-T have been studied in man and experimental animals. Five human male volunteers ingested a sin- 'e oral dose of 5 mg/kg without presenting detectable clinical effects (Gehring et a l, 1913). The ranee of 2,4,5-T from the plasma as well as excretion from the body occurred by an apparent first-order rate process with a half-life of 23 hr. Essentially allnf the 2,4,5-T was absorbed into the body and excreted unchanged in urine. Erne (1966) demonstrated that orally administered salts of both 2,4,5-T and 2,4-dichIorophenoxyacetic acid (2,4-D) were readily absorbed in rats and eliminated primarily by the. kidneys. The half-life of clearance from plasma of both 100 mg/kg of 2,4,5-T and 50-200 mg/kg of 2,4-D was approx 3 hr. The elimination half-life of 2,4-D from tissues ranged from 5 to 10 hr. Grunowefa/. (1971) reported that 7 days after oral administration of 50 mg/kg of 2,4-5-T to rats, only 69% of the dose was recovered, all in the urine. The main constituent in. urine was unchanged 2,4,5-T. Approximately 15-30% of the dose recovered in urine was found as conjugates of 2,4,5-T. Courtney (1970) found approx 90% of orally administered 2,4,5-T (50-100 mg/kg) unchanged in the urine of rats within 72 hr. Piper et al. (1973) studied the clearance of 2.4.5-T from plasma and its elimination from the body of rats after single oral doses of [UC]2,4,5-T. The half-life values for the clearance of 2,4,5-T from plasma of rats given C opyright 1976 b y A cadem ic Press, Inc. Ai| rig h ts o f re p ro d u c tio n in an y fo rm reserv ed . Printed in G reat B ritain 491 i: 492 SAUERHOFF ET AL. doses o f 5, 50, 100, and 200 mg/kg were 4.7, 4.2, 19.4, and 25.2 hr, respectively. Halflife values for elimination from the body were 13.6,13,1,19.3, and 28.9 hr, respectively. These results establish dose-dependent pharmacokinetics. In interpreting the data of Piper et al. (1973) it is not possible to state whether dose-dependence is attributable to saturation of some excretory mechanism or to absorption. To resolve questions on absorption and subsequent eliminaton of 2,4,5-T additional pharmacokinetic studies were conducted to determine the rate of 2,4,5-T clearance from plasma, the volume of distribution, metabolism, and routes and rates of 2,4,5-T elimination from rats following a single iv dose of either 5 or 100 mg/kg of [14C]2,4,5-T. METHODS Anim als. Sprague-Dawley (Spartan substrain) rats weighing 180-220 g were housed in metal metabolism cages designed for the separate collection of urine and feces. The rats were acclimated in their individual metabolism cages for 4 days prior to dosing. Food and water were available ad libitum to all rats throughout the acclimation and experi mental periods. The cages were maintained in an air-conditioned room at a temperature of 24C with a 12-hr light-dark cycle. Jugular cannulas were surgically implanted in all rats while anesthetized with methoxyflurane 72 hr prior to administration of the dose (Harms and Ojeda, 1974). Dosage solution. [Ring-U-14C]2,4,5-T (4.3 mCi/mmol, Mallinckrodt Chemical) was administered to rats as the sodium salt in an aqueous solution. A stoichiometric equivalent of NaOH was added and mixed with the 5- and 100-mg/kg dosing solutions, converting the free acid of 2,4,5-T to the soluble sodium salt. The two dosing solutions were prepared equimolar with respect to sodium by adding NaCl to the 5-mg/kg dosing solution. The infrared spectrum of the sodium 2,4,5-trichlorophenoxyacetate (Na-2,4,5-T) obtained as a NUJOL mull was identical in all essential details to the spectrum of.Dow Analytical Standard Na-2,4,5-T. An excess of HC1 was added to an aliquot of the dosing solution and an additional infrared spectrum of the recovered free acid was obtained. This was compared to the spectrum of Dow Analytical Standard 2,4,5-T. The spectra of these materials were identical in all essential details. Thus, the addition of NaOH to 2,4,5-T did not cause chemical decomposition of the starting material. Unlabeled Dow Analytical Standard 2,4,5-T (greater than 99 %pure) and [14C]2,4,5-T were prepared to contain 1 mg of 2,4,5-T/1.5 ml for the 5-mg/kg dosing solution and 20 mg of 2,4,5-T/1.5 ml for the 100-mg/kg dosing solution. Each solution contained approx. 5 pCi/1.5 m l o f [14C]2,4,5-T. Doses o f 2,4,5-T were prepared to administer 5 mg/kg and 100 mg/kg of the free acid of 2,4,5-T. Dosing solutions were infused through ajugular cannula using a constant delivery syringe pump at the rate of 1 ml/min. After delivery of the dosing solution, the cannula was washed with physiological saline. The dose was determined gravimetrically by weighing the dosing syringes prior to and immediately following delivery of the dose. Five-microliter aliquots of the dose solutions were counted to quantitate the amount of 14C activity administered to each rat. The radiochemical purity of the dosing solution was determined using two thin-layer chromatography systems. Approximately 100,000 dpm were spotted on (1) alumina - (0.25-mm thickness, Brin' (50); and (2) silica gel (0.2 (90): methanol (5).'glacial The R / value of 2,4,5-T area of each plate was s< activity in an Aquasol-wa to be 99.0% radiochemic: M etabolism and balanc of [14C]2,4,5-T. Urine wa; both dose levels. Urine tr 12, 36, 60, and 84 hr at tl mg/kg dose. Following the last colh sanguinated. The carcasse fat, and skeletal muscle v washed with water and : activity. Blood plasma study. Fo centrations of [14C]2,4,5-' or 100 mg/kg of [14C]2,4, cutting the tail vein with tube. The tubes were imr centrifuge. The tubes wen was drained into preweit Plasma samples were colh dose, and at 0.5, 1, 2, 4, 8 Analysis o f urine fo r 2, were spotted on silica gel; (90):ethanol (5).'glacial ? respectively. Standards of Plates were scanned on a : [14C]2,4,5-T was marked,, sequent 14C analysis. The the same manner. The per [14C]2,4,5-T was calculate Radioactivity analysis. Liquid Scintillation Sysiei rection. The validity of tl ['~C]-toluene as the interi urine, plasma, and cage w; liver samples were prepan Biological Material Oxidi Carbon dioxide formed d> methoxyethanol (10 ml) a: 1 Rr measured as the ratio c tanee from the origin to the so 19310 ^T^WWW.T "WA V'' S***''***** Tywyy r-\ --let. 'if^v***-'*"*'*<**^*AiaUgJ>aa *.y...v :ctively. Haif( respectively, .i the data of .ctributable to i-T additional learance from es of 2,4,5-T f [14C]2,4,5-T. sere housed in feces. The rats dosing. Food m and experia temperature implanted in ion o f the dose `_'h 'cal) was stc. aometric sing solutions, King solutions -mg kg dosing i (Na-2,4,5-T) drum of Dow ;of the dosing 'was obtained. The spectra of t of NaOH to F id [I4C]2,4,5-T j solution and(:on contained5 administer 5 ifused tlirouyh iml/min. After eal saline. Th.e | prior to and j of the dose ;stered to each Two thin-layer n (!' alumina t PHARMACOKINETICS OF 2,4,5-T 493 (0.25-mm thickness, Brinkman) and developed with glacial acetic acid (50): methanol (50); and (2) silica gel (0.25-mm thickness, Brinkman) and developed with chloroform (90): methanol (5):glacial acetic acid (5). Each plate was developed two dimensionally. The R f l value of 2,4,5-T on both silica gel and alumina plates was 0.5. The diagonal area of each plate was scraped in 1-cm sections and the scrapings counted for 14C activity in an Aquasol-water gel scintillation cocktail. The dosing solutions were shown to be 99.0% radiochemically pure. Metabolism and balance study. Two rats of each sex received either 5 or 100 mg/kg o f [14C]2,4,5-T. Urine was collected every 12 hr throughout the experimental period at both dose levels. Urine traps were immersed in Dry Ice baths. Feces were collected at 12, 36, 60, and 84 hr at the 5-mg/kg dose, and every 24 hr through 168 hr at the 100- mg/kg dose. Following the last collection interval, the rats were killed by decapitation and ex sanguinated. The carcasses were skinned, and the liver, kidneys, a sample of perirenal fat, and skeletal muscle were removed from each animal. Each metabolism cage was washed with water and acetone. The cage wash was subsequently counted for l4C activity. Blood plasma study. Four rats of each sex were used to characterize the plasma con centrations of [14C]2,4,5-T. Two male and two female rats were adminis'. d either 5 or 100 mg/kg of [14C]2,4,5-T. Whole blood samples (approx. 70 pi) were lAiuinc.by cutting the tail vein with a scalpel and collecting the blood in a heparinized capillary tube. The tubes were immediately sealed at one end and centrifuged in a hematocrit centrifuge. The tubes were broken close to the plasma red cell interface and the plasma was drained into prc-.>eighed liquid scintillation vials for subsequent 14C counting. Plasma samples were .ollected at 0.5, 1, 2, 4, 8, 12, 20, 28. and 36 hr after the 5-mg/kg dose, and at 0.5, 1, 2, 4, 8, 12, 24, 36, 48, 60, and 72 hr after the 100-mg/kg dose. Analysis o f urine fo r 2,4,5-T metabolites. One hundred-microliter aliquots of urine were spotted on silica gel and alumina thin-layer plates and developed with chloroform (90):ethanol (5):glacial acetic acid (5), and glacial acetic acid (50):methanol (50), respectively. Standards of [I4C]2,4,5-T were chromatographed with the urine sample Plates were scanned on a Panax Thin Layer Scanner. The area on the plates containing [l4C]2,4,5-T was marked, scraped, and transferred to liquid scintillation vials for sub sequent 14C analysis. The remaining areas on the tic plates were scraped and handled in the same manner. The percentage of.tota! l4C activity on each tic plate associated with [UC]2,4,5-T was calculated. . ; .. ~ Radioactivity analysis. All samples were-counted' in a Nuclear Chicago Mark II Liquid Scintillation System utilizing external standard ratios to determine quench cor rection. The validity of the quench correction curve was checked periodically using [uC]-toluene as the internal standard. Water and Aquasol were added to aliquots of urine, plasma, and cage wash, and analyzed for l4Cactivity. Feces, kidney, carcass, and liver samples were prepared as 33 %aqueous homogenates and oxidized in a Beckman Biological Material Oxidizer. Perirenal fat and skeletal muscle were oxidized directly. Carbon dioxide formed during the oxidation was trapped in 5 M 2-aminoethanol in 2- nicthoxyethanol (10 ml) and counted for14C activity following addition of ascintillation 1 H f m easured as the ratio o f the d istance from the origin to the top o f the sp o t divided by the d isIttfke from th e o rig in to th e so lv e n t fro n t. V- r- ,-.J- . >, . v'>,. ....... - 494 SAUHRHOFF T A L. cocktail (10 ml). The cocktail contained 4 g of 2,5-diphenyloxazole (PPO) and 0.1 g of 1:4-di(2-(5-diphenyloxazolyl)benzene) (POPOP) per liter of 1:1 (v/v) toluene:2methoxyethanol. Recovery of J4C activity from spiked samples was 95 + 5%. Gas chromatography-mass spectrometry o f2,4,5-T. Urine samples (5 ml) were acidified with 1 ml of 1.0 n HC1 and extracted twice with 5 ml of diethyl ether. The ether extracts were methylated with diazomethane, evaporated to dryness, and redissolved in 0.5 ml of hexane. A 2-/tl aliquot of the hexane was injected into a LKB 9000 gas chromato graph-mass spectrometer (gc-ms). The gc column was glass 6 ft x 2.0 mm (i.d.), and packed with 10% OV-1 on Chromosorb W 80-100 mesh. The tnje 268, 270, and 272 peaks were monitored for sample quantitation. Recorded peaks were symmetrical; therefore peak heights and not peak areas were used for sample quantitation. lationship is necessary to ch linear parameter estimatioi .(Wagner, 1973) which is the Michaelis-Me substrate concentration: V,, where the enzyme is saturate * t i r B. The pharmacokinetic pa from plasma at both dose le RESULTS The concentration of 2,4,5-T in the plasma of rats as determined by 14C activity is shown in Fig. 1 on a semilog-a-ri-t-h-mic--p-l-ot-o-f concentra<tion versus time following iv nonlinear re- j .... R ate C onstants (-) [14C]2,4,5-T from Pla Single Dose (mg/kg) 5 0.16 100 0.030 . * 0.13 *Two male and two fei * Calculated by linear n c Maximum elimination 1973): c, - c2 ci - c 3 V Jkmr. Fig. 1. Concentration of 2,4,5-T in the plasma of rats as a function of time following a single iv dose of [MC]2,4,5-T at 5 mg/kg (B) and 100 mg/kg (A) as determined from the concentration of 1JC activity. Data for each dose level were obtained from four rats. the linear portion of the pla? cally different from the valut were observed in the clearai Projection of the blood p allows calculation of the initi Dividing the initial plasma into the dose in micrograms { per kilogram. Vd increased 1 and 100-mg/kg dose levels, Table 2 shows the excreta 100 mg/kg of [14C]2,4,5-T. E in urine and feces during su UC activity from the bodycorresponding half-life valu fitted by linear regression f Mo sex differences were obs i I m: :* .1 0) and 0.1 g /) toluene:2- Were acidified ether extracts solved in 0,5 tas chromatomn (i.d.), and 270, and 272 : symmetrical; uion. lationship is necessary to characterize the pattern for clearance with 100 mg/kg. Non linear parameter estimation can be used to fit the data of curve A to the equation , (Wagner, 1973) d c/d t Vm.c k,, I + c which is the Michaelis-Menten equation for a one-enzyme reaction where c is the substrate concentration: Vm= 16.6 1.82 /ig/hr/ml corresponds to the maximum rate where the enzyme is saturated and k m= 127.6 25.9 tg/e is the substrate concentration at \ V m. The pharmacokinetic parameters describing the rate of clearance of [UC]2,4,5-T from plasma at both dose levels are presented in Table 1. The rate constant k describing i:C activity is >.e follow in g iv >: the 5-m g/kg ! nonlinear re- TABLE 1 R ate C onstants ( k ) a n d H alf-L ife Values ( / +) for th e C learance of [l4C]2,4,5-T from P lasma and Volumes of D istribution F ollow ing a Single I ntravenous D ose of 5 and 100 mg/kg1' Dose (mg/kg) k (h r'1) h V* (hr) (ml/kg) ] rr : ! i'o ig a single iv on.^iitration of l'C 5 0.16 tO-Ol" 4.33 0.27 190 8 100 0.030 0.00811(0-36 hr) 23.1 4.9 (0-56 hr) 23J 10 0.13 0.03c (36-72 hr) 5.30 1.23 (36-72 hr) " Two male and two female rats at each dose level. 6 Calculated by linear regression analysis. c Maximum elimination rate constant calculated by the following equations (Wagner, 1973): Ci - ci + Km1n(c,/c2) = Vm(r2- /,) Ci - c3+ Km\n {cjc3) = Vm(t3 - l,) VJkm= k (maximum elimination rate constant). the linear portion of the plasma curve at the 100-mg/kg dose (36-72 hr) was not statisti cally different from the value when rats were given 5 mg/kg (0-36 hr). No sex differences werejabservedin the clearance of 14C activity from plasma. Projection of the blood plasma concentration curves back to the y intercept, t = 0, allows calculation of the initial concentration of 2,4,5-T in the plasma prior to excretion. Dividing the initial plasma concentration of [l tC]2,4,5-T in micrograms per milliliter into the dose in micrograms per kilogram gives the volume of distribution (V d), milliliters per kilogram. Vd increased with dose from 190 8 ml/kg to 235 10 ml/kg for the 5and 100-mg/kg dose levels, respectively. Table 2 shows the excretion of 14C activity by rats following single iv doses of 5 and 100 mg/kg of [14C]2,4,5-T. Excretion is expressed as a percentage of the total dose found in urine and feces during successive time intervals. The rate constants for excretion of UC activity from the body of rats given a single iv dose of 5 and 100 mg/kg and the corresponding half-life values (Aj.) are presented in Table 3. The constants were calcu lated by linear regression from data collected through the S4-hr collection interval. Mo sex differences were observed in the excretion of UC activity from the body. ';.i :,f ;. : :v ; i ; i: ri S '* ' ! :i 496 SAUERHOFF E T A L. (ooo/) Uh eNnO 8 8 d "l r+- + i1 +i 1 m 'eSn. no O--OONON NO ood o +1 +1 +1 +1 in ~ vs en mp dddo (iNn (N +1 OS Cft ^V- ^-H^co oo O --1 <N (N --' O O O O +i +i +i +i +i +i +| T f N CO f 'l (N O o R2 OOo +1 +1+1 +1 Mn - no --o . no o sen. +1 OO 0n0 (N--^(N mO O -CN<-n4- n'VCTO=? 7 8 ? fi VO 8 ^ 8 15 <N scooon IL COO mo o dd* d mT+1f 1 S+Oi 1 +1 o+n1 ri d d d <CNO d +1 00 ri h as s s s Tf-ifN-OO +! +r+i +i +i +i +i tN N^ ONn QO-MP^ <ns to<***OO n +) O s co o<Hz Wo>0 Q- 1 R ate Cor E xcretion Dose ! (mg/kg) i ------- ;5 ! oo Mean + T h e e lim in a tio n o f J4 elim ination rate constat? 84-hr tim e interval at th: are differences in the ar group. A dditionally, urh biphasic w ith a slow er e Eighty-four hours after i initial d o se rem ained in th T o assess the overall r studies were h om ogen ize | senate. In ad dition , the c activity for the 5- and 100 S D ). Tissue to plasm a ratios adm inistration o f the d os tissues and the tissue/plasi the tw o d ose levels w ere ; results, h o w e v e r , indicate t K idney to p lasm a ratio: adm inistration o f 100 mg plasm a and k id n ey is presc the 4-hr interval. The kidni j tion o f the d ose. A t lo w p la: ! T h e p e r c e n ta g e o f 14C a c i ch rom atograp h y system s is in u rin ea p p e a re d a s [I4C]2,` i of l4C a c tiv ity a p p e a r e d a t t | the m aterial in urine w ith a 1 analyzed by g c-m s. T he an on the specific activity o f the , interval. T he p ercen tage o f i was 96.0 1 .0 ,9 7 .0 1 .4 , a intervals, respectively. Ther 14  ir*iItUm'*'*^*-*'*-,^*`M-r*.\. &/f f1 f i T ;1 P H A R M A C O K IN E T IC S O F 2,4,5-T TABLE 3 R ate C onstants ( k ) an d H alf-life Values (/+) for the E xcretion o f [I4C]2,4,5-T from the Body o f R ats F ollow ing a S ingle Intravenous D ose of 5 and 100 mg/kg D ose (m g/kg) ka ( h r 1) tt (hr) 5 0.065 0.020 100 0.063 0.020 (0 -8 4 hr) 0.010 0 .0 0 5 (84-168 hr) Mean SD for four rats. 10.7 11.0 69.3 497 t - T. / . i--' 4T A T h e e lim in a t io n o f 14C a c t iv ity fr o m th e b o d y a p p e a r s to b e d o s e -d e p e n d e n t. T h e e lim in a tio n r a te c o n s t a n t s a n d t v a lu e s fo r u r in a r y e lim in a t io n o f 14C a c tiv ity o v e r a 84-hr tim e in terval at th e tw o d o se levels are n ot sta tistica lly d ifferent. H o w ev er, there arc d iffe r e n c e s in th e a m o u n t o f 14C a c tiv ity e x c r e te d in fe c e s b e tw e e n e a c h d o s a g e group. A d d it io n a lly , u r in a r y e lim in a t io n o f 14C a c tiv ity a t th e h ig h d o s e a p p e a r s to b e b ip h a sic w ith a s lo w e r e lim in a t io n o f 14C a c tiv ity p a st th e 8 4 -h r c o lle c tio n in te r v a l. Eighty-four h ours after ad m in istration , approx. 3.7% as com pared w ith 0.6% o f the initial d o se rem ain ed in th e b o d y o f ra tsa t th e 1 0 0 -a n d 5 -m g /k g d o se g ro u p s, resp ectively. T o a s s e s s t h e o v e r a ll r e c o v e r y o f 14C a c tiv ity , th e c a r c a s s e s o f rats in th e e x c r e tio n studies w e r e h o m o g e n iz e d a n d th e 14C a c tiv ity d e te r m in e d in a liq u o ts o f th e h o m o gen ate. In a d d it io n , th e c a g e w a s h w a s c o u n te d fo r 14C a c tiv ity . T o t a l r e c o v e r y o f 14C activity fo r th e 5- an d 1 0 0 -m g /k g d o se level w as 9 9 .5 + 5.1 % a n d 9 6 .0 3.0 % (m ean SD). Tissue to p lasm a ratios w ere evaluated at the 5- and 100-m g/k g d ose levels 8 hr after a d m in istr a tio n o f th e d o s e . T h e c o n c e n tr a tio n o f 14C a c tiv ity in p la s m a a n d se le c te d tissues and th e tissu e/p la sm a ratios are p resen ted in T a b le 4. T issu e to p la sm a ratios at the tw o d o s e le v e ls w e r e n o t s ta tis tic a lly d iffe r e n t w ith e x c e p tio n o f th e k id n e y . T h e results, h o w e v e r , in d ica te th a t th e liver to p la sm a ra tio is h igh er at th e 1 0 0-m g/k g d o se. K id n ey to p lasm a, ratios w ere a lso evalu ated at selected tim e in tervals after the iv a d m in istr a tio n o f 1 0 0 m g /k g o f [ 14C ]2 ,4 ,5 -T . T h e c o n c e n t r a tio n o f 14C a c tiv ity in plasm a a n d k id n e y is: p r e s e n te d in T a b le 5. K id n e y to p la s m a r a tio s w e r e le ss th a n 1 a t the 4 -h r in te r v a l. T h e .k id n 'e y t o p la s m a r a tio in c r e a s e d w ith tim e f o llo w in g a d m in is tr a tion o f th e d o s e . A t lo w p la s m a c o n c e n t r a tio n s o f 14C a c tiv ity , th e r a tio w a s 4 .9 5 0 .5 9 . T h e p e r c e n ta g e o f ,4 C a c t iv ity in u r in e a s s o c ia te d w ith [14C ]2 ,4 ,5 - T in tw o th in -la y e r c h r o m a to g r a p h y s y s te m s is p r e s e n te d in T a b le 6. B e tw e e n 9 6 .7 a n d 9 9 .3 % o f 14C a c tiv ity in urine a p p e a r e d a s [ 14C ] 2 ,4 ,5 - T a t th e 5 - m g /k g d o s e le v e l w h ile b e tw e e n 9 3 .8 a n d 9 7 .1 % of 14C a c tiv ity a p p e a r e d a t th e 1 0 0 -m g /k g d o s e le v e l. In o r d e r to c o n fir m th e id e n tity o f the m a te r ia l in u r in e w ith a s im ila r Rf v a lu e to th a t o f [ l4C ]2 ,4 ,5 - T , u r in e s a m p le s w er e analyzed by g c -m s. T h e a m o u n t o f 2 ,4 ,5 -T in each urine sam p le w as calcu lated based on the s p e c ific a c t iv ity o f th e d o s e a n d p e r c e n ta g e o f d o s e e x c r e te d d u r in g e a c h c o lle c tio n interval. T h e p ercen ta g e o f c a lcu la ted 2 ,4 ,5 -T a t th e h ig h d o se as d eterm in ed by g c -m s "as 96.0 + 1.0, 97 .0 1.4, and 100.3 7.5 for the 0 -1 2 , 12-24, and 2 4 -3 6 hr co llectio n intervals, r e s p e c tiv e ly . T h e r e fo r e , u n c h a n g e d [ 14C ]2 ,4 ,5 - T a c c o u n t e d fo r g r e a te r th a n t W! i: : l J I- r\ sJasi^viii. --J.. -- SA U E R H O FF r /IL . TABLE 4 14C TC o n c e n t r a t i o n o f A c t iv it y in S e l e c t e d ^ a n d P lasma 8 H R AFTER TH E A D M IN ISTR A TIO N O F 5 A N D 100 [ C J-,4 ,5 -* Dose (mg/kg) Tissue Concentration (j`g/g) Tissue/PIasma ratio 5 Liver 0 .9 4 0.12 Kidney Brain 9 .2 6 2.23 __ h Fat 0 .4 3 0 .0 7 Muscle 0 .3 2 0.02 Plasma 3.85 0.85 1 0 0 Liver 157 9 Kidney 2 0 5 15 Brain 6.6 0 .7 Fat 4 0 .0 8 .8 Muscle 53.0 9 .4 Plasma 416 36 0 .2 4 0.03 2 .4 0 1.23 0.11 0 .0 3 0.08 0.02 0.38 0.07 0.49 0 .1 0 0.0 1 6 0 .0 4 0 0.1 0 0 .0 3 0 .1 2 0.03 - E xpressed as m icrogram equivalents of 2,4,5-T per g ram ; m ean SD of fo u r rats. S am ple below detection lim it. TABLE 5 C oncentration of i4C A ctivity in K idney and P lasma at SELECTED TlME INTERVALS AFTER THE INTRAVENOUS ADMINISTRATION o f 100 m g /k g [14C ]2 ,4 ,5 -T 0 T im e interval (hr) Plasm a concentration K idney concentration K idney/Plasm a ratio 4 415 1 9 297 1 1 0.72 + 0.02 8 359 20 281 17 0.78 0.03 16 315 2 1 297 29 0.94 0.04 32 142 2 1 198 1 8 1.40 + 0.11 64 25 9 122 + 45 4.95 0 .5 9 . E xpressed as m icro g ram equivalents o f 2,4,5-T p er g ram ; m ean + SD o f fo u r rats... TABLE 6 P ercentage of U rinary 14C A ctivity as [ I4C ]2 ,4 ,5 -T 0 System Dose (mg/kg) 0-12 Time interval (hr) 12-24 Silica gel a i 2o 3 Silica gel a i 2o 3 5 97.0 1.0 99.3 1.2 5 99.3 0.6 98.7 2 .3 100 93.8 5.5 94.7 3 .3 100 94.8 2.5 97.1 2 .5 . M ean SD o f fo u r rats from each excretion interval. 24-36 99.0 1.7 96.7 3 .5 95.0 5 .2 96.4 3 .9 '-.Ai- - , ** <hV*\i ' ' r JO 1: 1.7 3-5 s > PHARMACOKINETICS OF 2,4,5-T 499 9 3 .8 % o f th e 14C a c tiv ity in u r in e a s d e te r m in e d b y t h in -la y e r c h r o m a to g r a p h y a n d g c-m s at b oth d ose levels. ! DISCUSSION G r e a te r th a n 9 3 .8 % o f th e 14C a c t iv ity e x c r e te d in u r in e w a s u n c h a n g e d [ 14C ]2 ,4 ,5 -T . It can be inferred that this relation ship also exists in plasm a. T herefore, clearance o f I 14C a c tiv ity fr o m p la s m a a n d u r in e r e p r e s e n ts c le a r a n c e o f 2 ,4 ,5 - T to a g o o d fir st a p proxim ation. T he clearance o f 2,4,5-T from p lasm a w as clearly d ose-d ep en d en t since there w ere m arked differences in th e sh ap es o f the p lasm a clearance curves at the twodose levels. It h as b een, su g g ested th a t 2 ,4 ,5 -T is cleared from th e p la sm a by th e k id n ey v ia th e o r g a n ic a c id s e c r e to r y p a th w a y . H o o k et al. (1 9 7 4 ) h a v e s h o w n th a t k id n e y s lic e s in vitro c o n c e n tr a te [ 14C ]2 ,4 ,5 - T . T h e 2 ,4 ,5 - T p la s m a -t im e p r o file in d ic a te s th is a c tiv e process m ay be saturated at 100 m g/k g because clearance from plasm a occurs at a m axim um rate during the first 36 hr and then d ecreases accord in g to a first-order p ro cess. C learance during the first 36 hr can also be described by a first-order process. T h e rate co n sta n t an d h a lf-life v a lu es are sh o w n in T a b le 1. T h e rate c o n sta n t c h a ra c terizin g the first-order p rocess for 3 6 -7 2 hr is the sam e valu e as the rate co n sta n t valid for the entire clearance curve at the low d o se level. T h is pattern exhibited at the high d ose with decreasing su bstrate co n cen tra tio n s is ch aracteristic o f a saturable active tran sp ort process. A dd itional evidence in su pp ort o f the con cep t o f saturability o f the excretory m ech an ism o f 2 ,4 ,5 -T c o m e s from a n a ly sis o f k id n e y tissu e/p la sm a ra tio s. E ig h t h o u r s a fte r adm inistration o f 5 m g/k g the k id n ey/p lasm a ratio is 2.40 w hile at 100 m g/k g the ratio is 0 .4 9 . K id n e y to p la s m a r a tio s a t s e le c te d in te r v a ls f o llo w in g a d m in is tr a tio n o f 100 m g/kg dem onstrate that as the p lasm a con centration o f 2,4,5-T decreases, the ratio in creases. T h e kidney can n ot concentrate 2,4,5-T as effectively at the higher p lasm a co n centrations associated w ith adm inistration o f 100 m g/kg. H ow ever, as plasm a co n centrations o f 2 ,4,5-T decrease, the k idn ey can con centrate 2,4,5-T . T h e p lasm a con cen tration d ata p resen ted in F ig. 1 reveal ad d ition al d o se-d ep en d en t effects. P rojection o f the curve b ack to the y-in tercep t (tim e = 0), allow s calcu lation o f the co n cen tra tio n o f 2 ,4 ,5 -T in p la sm a p rior to excretion . T h is co n cen tra tio n is d e pendent o n the d istrib ution o f 2 ,4 ,5 -T to tissu es and its su b seq u en t entry in to the cells a n d /o r tis s u e s p a c e s . T h e Vd o f 2 ,4 ,5 - T is s ig n ific a n tly la r g er a t th e h ig h d o s e . I f firstord er p r o c e s s e s w e r e o p e r a tiv e , th e n th e Vd w o u ld n o t d iffe r a t th e se tw o d o s e le v e ls . A com p arison o f liver tissu e/p lasm a ratios 8 hr after the iv ad m in istration o f 5 an d 100 m g /k g o f [ 14C ]2 ,4 ,5 - T r e v e a ls th a t th e r a tio s a re 0 .2 4 a n d 0 .3 8 , r e s p e c tiv e ly . L iv e r tissue, th erefore, is exp osed to a h igher co n cen tration o f 2 ,4 ,5 -T than w ou ld h ave b een predicted assu m in g the first-ord er k in etics o f a 5-m g/k g d o se o f 2,4,5-T . W h ile th ese differences are n o t sta tistica lly d ifferen t, a trend is in d icated . Exam ination o f the excretion data (T able 2) reveals tw o significant features. A t the 1 0 0 -m g /k g d o s e le v e l, a g r e a te r fr a c tio n , o f it h e d o s e a s 14C a c tiv ity is e x c r e te d in th e feces. T h is su ggests that at h igh d o se levels b iliary excretion o f 2 ,4 ,5 -T a n d /o r d egrad a tion p rod u cts p lays a m ore im p o r ta n t ro le in the overall elim in a tio n o f 2 ,4 ,5 -T fr o m the b od y . L. !; !; ) i;:: r 1 it I- \\ f. 7;: A tr 7i7. <o7<u> i r-rt.^irfn 500 SAUERHOFF E T AL. 1 At the 100-mg/kg dose level, there appears to be a second slow phase of 14C activity excreted in urine between 84 and 168 hrs. In rats given 5 mg/kg, 0.6% of the dose remained in the body after 84 hr. In those given 100 mg/kg, 3.7% of the original dose remained and was gradually eliminated. This leads to two alternative conclusions: (1) at 100 mg/kg, a larger fraction of the dose has gained entry into compartments, tissues, and cells from which it is slowly released, and/or (2) a fraction of 2,4,5-T has been metabolically converted in the body to products which are less readily excreted. This latter hypothesis is supported by data presented in Table 6. Between 96.7 and 99% of 14C activity in urine appeared as [14C]2,4,5-T at the low dose while between 93.8 and 97% of 14C activity appeared as [14C]2,4,5-T at the high dose. While these are not signifi cantly different, a trend is indicated. There are direct toxicological implications for the dose-dependent elimination of 2.4.5- T from plasma as well as from the body. Systemic toxicity of a drug or foreign compound is often a function of the concentration and duration of that drug in plasma. If the drug or foreign compound is eliminated at a slower rate from the plasma, and/or different metabolites are formed at a high dose, then the ability of the drug and/or metabolite to induce toxicity at the high dose will be greater. Therefore, more severe toxicological manifestations than would have been predicted will often occur at the high dose level where nonlinear kinetics are operative. However, toxic effects have not been observed with a single iv dose of 100 mg/kg of 2,4,5-T. There appear to be significant differences between the results of this study and that of Piper et al. (1973). Piper et al. administered 2,4,5-T orally, while in this study 2,4,5-T was administered by the iv route. Piper demonstrated dose-dependent elimination of 2.4.5- T from plasma. For any given oral dose, semilogarithmic plots of elimination versus time in plasma gave a linear relationship. At increasing doses, however, the elimination ti was longer which is a criterion of dose-dependent kinetics (Levy, 1968). Clearance curves did not show the classical Michaelis-Menten graphical representa tion of nonlinear kinetics as is shown in this study. One explanation for the data is that the study was not conducted over a sufficiently long time interval. An alternative is that oral absorption of 2,4,5-T is playing some undefined role in altering the clearance of 2,4,5-T from plasma. S' ` The pharmacokinetic data presented in this report indicate that the statistical projection of results of experiments with large doses of 2,4,5-T to predict the hazard of exposure to small amounts is not justified, because the capabili ty of the body to handle the compound has been altered. Ii Hook, J. B., Bailie, M. D., transport of 2,4,5-trichlorc 209-218. Levy, G. (196S). Dose depen Principles in D rug Evalnatic York. Piper, W. N ,, Rose, J. Q., Let phenoxyacetic acid (2,4,5-T Pharmacol. 26, 339-351. W a g n e r , J. G. (1973) Proper which are useful in pharma ' REFERENCES. : Courtney, D. K. (1970). 2,4,5-T in the rat: Excretion pattern, serum levels, placental transport and metabolism. In Pesticides Symposia, pp. 277-283. Hales and Assoc., Florida. E rne, K. (1966). Distribution and elimination of chlorinated phenoxyacetic acid in animals. Acta Vet. Scand. 7, 240-256. G ehring, P . J ., K r a m e r , C . G ., Schw etz, B . A ., R ose, J . Q ., and R ow e, V . K . (1 9 7 3 ). T h e fa te o f 2 ,4 ,5 -trich Io ro p h e n o x y acetic acid (2 ,4 ,5 -T ) fo llo w in g o ral a d m in istra tio n to m an. Toxicol. AppL Pharmacol. 2 6 , 3 5 2 - 3 6 1 . G ru no w , W., Boerne, C., and Budezles, B. (1971). Renal, excretion of 2,4,5-T by rats.Fd. Cosmet. Toxicol. 9, 667-670. H arms, P. J., and O jeda, S. R. (1974). A rapid and simple procedure for thechroniccannulation of the rat jugular vein. J. Appl. Physiol. 36, 391-392. 18 19318 -'*%*r*&K&vj&?lt*r-.-*.--. ***% & 4 Ijvny dosC [d o se js:(D {ssues, ^m ciai latter `of l i C et 97% vsiuniti- " ition of ^foreign .plasm a, r, a n d /o r and/or ;e severe ;r at the have not a of .Iv 2,4,5-T /nation of 'im ination .ever, the ivy. 196$). epresentaaua is that ernative is c clearance istical proi hazard ot iv to handle PHARMACOK1NETICS OF 2,4,5-T 501 Hook, J. B., Bailie, M. D ., J ohnson, J. T., and G ehiung, P. J. (1974). In vitro analysis o f transport o f 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) by rat. Fd. Cosmet. Toxicol. 12, 209-218. L e v y , G. (1968). D ose dependent effects in pharmacokinetics. In Importance o f Fundamental Principles in Drug Evaluation. (D . H. Tedeschi and R. S. Tedeschi, Eds.). Raven Press, N ew York. P iper, W. N ., R ose, J. Q., L en g, M. L . , and G eh ring.P . J. (1973). The fate o f 2,4,5-Jrichlorophenoxyacetic acid (2,4,5-T) following oral administration to rats and dogs. Toxicol. Appl. Pharmacol. 26, 339-351. W a g n e r , J. G. (1973) Properties o f the Michaelis-Menten equations and its integrated form which are useful in pharmacokinetics. J. Pharm. Biopharm. 1, 103-121. F i. I '* '. Ip i r '>-r> \ Vf i -7 S!.' !: -'1l r . i-'.*&.' ransport animals. 973). The \ to man. y rats. Fd. nutation 18 193 J ami-s C . G arrison, Teresa U . Bisee, Paui. Peterson, and Edwin M . U yeki. C hanges in I lep ato cy le Ploidy in R esponse lo C h ro m iu m , A nalyzed by C o m p u te r-A s s is te d M i c r o s c o p y .......................................................................................................... W. H. Lawrence, a . Lassi.o, J. E. T urner, Z. Feng, and S. E. Bond. S o m e N ovel In hibitors o f Platelet A ggregation: A cu te T oxicity in M ice an d Its R ela tio n s h ip lo in I "tiro E llicacy a n d T o x ic ity . II. N ip e c o to y la m in o a lk a n e a n d N i- p c c o lo y lp ip c r a /in c C o n g e n e r s .................................................................................................... Gary Ai.ieniiy, Richard M. Sharpe, and Paui. M. D. Foster. Changes in S e rto li C e ll F u n c tio n in V itr o I n d u c e d b y N i t r o b e n z e n e ............................................. James W. Oi.diiam, John R. Carson, Alan G ill, Robert F. Preston, and Debra D. Sedor. G e n e tic T o x ic ity o f a P h a r m a c o lo g ic a lly A c tiv e G r o u p o f u /V /io -(A ry lu lk y n y l) p h c n o x y p r o p a n o la m in e s ..................................................................... Carole A . Kimmei. and Elaine Z . Francis. P ro c e e d in g s o f t h e W o r k s h o p o n th e A cceptability an d Interp retatio n o f D erm al D evelopm ental T oxicity Studies Hiroshi Yamaiiciii.T osiiikazii Kaise, KeikoTakaiiashi, and Yukio Yamamlira. T o x ic ity a n d M e ta b o lis m o f T r i m e l h y la r s i n e in M ic e a n d H a m s te r s . . Man ion R . Frierson, Frances A . Mielacii, and D . M . Kociiiiar. C o m p u te r - A u to m ated S tru c tu re E valuation (C A SE) o f R etin o id s in T eratogenesis B ioas sa y s ............................................................................................................................................................ F. Gagne, M . Marion, and F. Denizeaij. M e ta l H o m e o s ta s is a n d M e ta llo th io - n e in I n d u c tio n in R a in b o w T r o u t I le p a lo c y te s E x p o s e d lo C a d m i u m .............. An n o u n c e m e n t s ................................................................................................................................................ 346 356 364 376 386 399 408 429 438 > I i JNDAMI N I Al A N |) AIN'I II I) I OXIi'OI <KiY 14, 2 19 -2 .1 4 ( 1990) REVIEW Association between Soft Tissue Sarcomas, Malignant Lymphomas, and Phenoxy Herbicides/Chlorophenols: Evidence from Occupational Cohort Studies Iik k 'S . J o h n s o n T p u ie m io lo i'y R ranch. N a tio n a l in stillin ' o jlln v ir o n in e n ta l H ea lth .Sciences. `.() . t t o \ I 2 2 .i.i. R e s e a r c h I r ia n x l c 1`a r k . N o r t h C a r o l in a 2 7 7 0 (J R e c e iv e d M a y J , iV S V ; a c c e p t e d S e p t e m b e r i t , /(AY9 A.sm k w u o ii b e tw e en S o li-'T issu e S a rc o m a s. M a lig n a n t L y m p h o m a s , a n d I'h c n o x y sA h lo ro phenols: L vidcnce from O ccupational C oh o rt Studies. Jo h n s o n . b. S. (1990). T u n d a m T o x ic o l. 14,21 9 -2 .1 4 . S o m e c a s e - c o n tr o l s tu d ie s h a v e r e p o rte d a s ig n ifican t a sso c ia tio n lx.iw ccn occupational use o f phenoxy herbicides and chlorophenols and soft-tissue sarcom as and m alig nant lym phom as. H ow ever, persons w ho spray or apply these substances are concom itantly c*|X)scd to o th er p otentially carcinogenic chem icals a n d oncogenic viruses, w hich have been found o r suspected to play a role in the etiology o f these tum ors. N o study has thoroughly controlled for these other exposures, som e o f which have been show n lobe independently associ ated with these tum ors even alter controlling for exposure to phenoxy acids o r chlorophenols. O n the o th er h an d , it has been fo u n d th at a n o bserved risk from exp o su re to p henoxy herbicides disappeared on controlling lo r som e o f these co n co m itan t exposures in the rare instance (his was attem pted Also, on several occasions, an association has been observed with occupations in which exposure to phenoxys and chlorophenols m ay occur, b ut not with the com pounds them selves. A ccordingly, a detailed review o f the evidence from occupational cohort studies was c o n d u c te d , to see if it c o rro b o rates th at fro m c a se -c o n tro l studies. It w as fo u n d that the evidence does not unequivocally incrim inate phenoxys and chlorophenols as a cause o f these tum ors. The results obtained with cohort studies o f sprayers and applicators d o not corroborate the a s so c ia tio n re p o rte d a m o n g th is o c c u p a tio n a l g ro u p , in c a s e - c o n tr o l stu d ie s. It is p o ssib le th at I *th e s u sp ected a s so c ia tio n m ay w ell kd u e . p a rtly o r w holly, to o n e o r m o re o f the o th e r c o n c o m i tant exposures. 1low ever. in view o f the fact that the m ajority o f the cohorts need further follow u p to he in fo rm a tiv e , it is c o n c lu d e d th a t f u rth e r s tu d ie s o f th ese c o h o rts a re req u ire d b efo re it can be determ ined w hether or not these tum ors are caused by exposure to phenoxy acids and c h lo ro p h e n o ls . *I'msut-u.-iyui'ioxKoi<y T h e phenoxy herbicides and chlorophenols have exlcnsive use in agriculture, forestry, co n struction, and oth er industries. C o m m er cial-grade p rep aratio n s o f these co m p o u n d s are co n tam in ated by the highly toxic dioxins and furans, som e isom ers o f w hich arc know n to be p o ten t carcinogens in laboratory a n i m als (K o cib a, 1984; N atio n al Toxicology P ro g r a m , 198 2 ; P ilo t t-l a /., 1 9 8 0 ). T h e r e a r e rep o rts that in h u m an s, exposure to phenoxy acid herbicides an d ch lo ro p h en o ls increases risk fo r so ft-tissu e sa rc o m a s (S I'S) a n d m alig nan t ly m phom as (M L ). T h e potential lor w idespread h u m an exposure to these sub sta n ces a n d th e ir c o n ta m in a n ts is signilicanl. H en ce, w hat role if an y , they play in the c a u sa tio n o f th ese tu m o rs, is o f im p o rtan c e. A large n u m b er o f studies o f different de signs have previously reported on the rela tionship betw een m alignant lym phom as ( H o d g k in 's d ise a se , n o n - H o d g k in 's ly m p h o m as) and farm ing o r agricultural w ork or ex p o su re to pesticides (B inaschi cl til., 1985; Brace, 1986; B row nson an d Rcif, 1988; 219 1 )2 7 2-0590/90 $.1.00 i '.il> )itl;lil ' I4'n> > ilic Sx tcly o f I o n c o lo g y All n g lii x o l ic |)iix lu i ito ti m tiny fo rm ti-n t u d 220 i:k ic s. Jo h n s o n B u e sch in g a n d W o llsta d l, 1984; B u rm e iste r, c h e m ic a ls a n d a n im a l o n c o g e n ic viru ses) is 1981; B u rm e iste r et til., 1983; C a n to r, 1982; resp o n sib le. D clzcll an d G ru lferm an , 1985; D uhrow cl a l, O ne specific exposure w hich has received 1988; l-'asal cl a l . 1968; G a lla g h e r cl al.. c o n s id e ra b le a tte n tio n a s a p o te n tia l c a n d i 1984; G iles cl a l. 1984; M ilitan t, 1971; d ate is ex p o su re to p henoxy acid herbicides P earce cl al.. 1985; P riester a n d M aso n , 1974; o r ch lo ro p h cn o ls. T h ere are case rep o rts on S a flla s c l a l.. 19 8 7 ; S c h u m a c h e r , 1 9 8 5 ; S c h u - th e d e v e lo p m e n t o f n o n - H o d g k in 's ly m n iach er an d D clzcll, 1988; S tark cl a l, 1983; p h o m a (N H L ) follow ing exposure specifi W eisen b u rg e r cl al.. 1986; W ik lu n d cl a l. cally to p h en o x y acid h erb icid es o r c h lo ro 1988; W illiam s cl al.. 1977). M ost have p h cn o ls (B ish o p an d Jo n e s, l9 8 l;O ls s o n a n d sh o w n an increased risk (B inasehi cl al., B ran d t, 1981) a n d th is possible asso ciatio n 1985; Brace, 1986; B row nson an d Reif, 1988; has been e x am in ed in n o less th an eight c a se - B uesching a n d W o llstad l, 1984; B urm eister, co n tro l stu d ies o f M L (C a n to r cl al., 1985; 1 9 8 1; B u r m c is lc r c l a l . 1 9 8 3 ; C a n to r , 1 9 8 2 ; E v e re tt c l a l.. 1 9 85; H a rd e ll c l a l . 1 9 8 1 ; H o a r D u h ro w cl a l . 1988; G ile s cl ill., 1984; P carcc cl a l . 1986; H o a r Z a h rn cl al.. 1988a; H o a r c l a l., 19 8 5 ; S a ld a s c l a l.. 1 9 8 7 ; S c h u m a c h e r , Z a h m c l a l . 1 9 8 8 b ; P e a rc e c l a l . 1 9 8 6 , 1987; 1985; S tark cl al., 1983; W eisen b u rg er cl al., W o o d s cl a l, 1987). T h re e o f these (E verett 1986; W ik lu n d cl al.. 1988; W illiam s cl al.. cl al., 1985; H ard ell cl al., 1981; H o a r cl al., 1977). The excess risk w as statistically sig 1986) rep o rted a statistically significant asso n ific an t in all b u t lo u r stu d ie s (B race, 1986; c ia tio n , w ith e stim a te d relativ e risk s o f 1.9 for D u h ro w cl al.. 1988; S tark cl al., 1983; W il e x p o su re to p en tac h lo ro p h c n o l, 6 .0 for e x p o liam s cl al.. 1977). O l'ih esc, th e o b served risk su re to p h en o x y acid s a n d c h lo ro p h cn o ls, w as associated w ith farm in g only, in the a n d 2.6 for ex p o su re to 2,4 -d ich lo ro p h en o x y study group as a w hole o r a subgroup(s) o f acid, respectively. In tw o studies, the excess it (B in aseh i cl al., 1985; B ro w n so n a n d R eif, risk o b se rv ed w as n o t sta tistically sig n ifican t 1988; B uesching an d W ollstadl, 1984; B ur (C an to r cl a l, 1985; H o ar Z ah m cl a l, m cislcr, 1981; D u h ro w cl al., 1988; G iles cl 1988b). In th e re m a in in g th re e stu d ies, a sta al.. 1984; P earce cl al.. 1985; S ch u m ach er, tistically significant asso ciatio n w as observed 1985; S tark cl al.. 1983; W ik lu n d cl a l. betw een M L a n d sp ray in g o f herb icid es o r o c 1988), since p esticide/herbieide use w as not cu p atio n s in w hich exposure to phenoxy investigated specifically. Seven studies inves acid s/ch lo ro p h cn o ls m ight occur, but no as tigated peslieide/herbicide use (B race, 1986; sociation w as observed for exposure to the B u rm eister cl al.. 1983; C a n to r, 1982; S alilas c o m p o u n d s them selves in th e e n tire study c l a l . 19 8 7 ; S c h u m a c h e r a n d D c lz c ll, 1 9 8 8 ; p o p u la tio n ( P e a rc e <7 a l . 1 9 8 6 , 1987; W o o d s W e is e n b u rg e r c l a l , 1 9 8 6 ; W illia m s c l a l . <7 a l . 1987). 1977). N o association w as rep o rted in tw o o f T h ere are also case rep o rts o f soft-tissue them (B race, 1986; S ch u m ach er an d D clzcll, sarco m as follow ing exposure to phenoxy 1988). O n e study observed a significant asso acids o r ch lo ro p h cn o ls (C ook, 1981; Finger- ciatio n w ith pesticides, but d id not investigate h u t <7 a l. 1984; H ardell, 1977; Ilo n c h a r an d farm in g as a risk facto r (W eisen b u rg er cl a l . H a lp e rin , 1981; H o p e cl al.. 1984; Jo h n so n 1986). T h re e stu d ie s fo u n d sta tis tic a lly sig ct a l . 1 9 8 1; M o se s a n d S e lik o ll, l 9 8 l ; S a r m a nificant associations w ith b o th farm in g an d an d Jacobs, 1982) an d an excess o f STS has general h erbicide ex p o su re, separately (B u r been seen in th e Seveso region o f Italy follow m c is lc r (7 <//.. 1 9 8 3 ; C a n to r , 1 9 8 2 ; S a ftla s cl in g a n in d u s tr ia l re le a se o f 2 ,3 ,7 ,8 - ic lra c h lo . a l , 1987). 1'hese fin d in g s a re c o n s iste n t w ith ro d ib e n z o -p -d io x in (T C D D ) (P u n to n i cl a l , an elevated risk o f ly m p h o m a am o n g farm ers 1986). Since A gent O range, a m ixture o f the an d agricultural w orkers. T hey leave o pen, phenoxy acids 2,4-dichlorophenoxyacetic how ever, the guest ion o f w hich specific ex p o acid (2,4-D ) an d 2,4,5-lricltlorophenoxyaces' Q su re (s ) ( fr o m a m o n g th e v a r io u s a g r ic u ltu ra l lic a c id (2 ,4 ,5 -T ), w a s u se d in th e V ie tn a m s o n tissue: s a r c o m a , m a l ig n a n t i .y m im io m a 221 w ar, m ilitary personnel have been studied for th e o c c u r r e n c e o f S T S a n d M L ( A n d e r s o n <7 a !.. 1985; B reslin c l a l.. 1988; Kelt el it!.. 1987a,b; G o u n an d K uller, 1988; H olm es, 1986; K ang el a l.. 1986; K ogan an d C lapp, 1985, 1988; l.a tlm ip el til.. 1983; L a lh ro p el a l.. 1984; L aw rence el a l.. 1985; C en ters for D isease C o n tro l V ietnam E xperience Study, 1987; W olfe et a t.. 1984; W olfe a n d M ichale k , 19 8 5 ) w ith g e n e ra lly in c o n c lu s iv e re su lts . H ow ever, the relevance o f studies o f m ilitary perso n n el is lim ited ; th e g ro u p s u n d e r study are to o young, su llicient lim e since first ex p o sure (latency) has not accrued, an d exposure is in a d e q u a te ly ch a ra c te riz e d (ex cep t in th e stu d ies by L alh ro p el al. (1984) an d W olfe and M ichalek (1985)]. A ccordingly, the w eight o f evidence for investigating the risk o f STS from exposure to phenoxy acids or c h lo ro p h e n o ls is d eriv ed m ain ly fro m a large num ber o f m ore conventional case-control stu d ies(B alarajan an d A cheson, 1984; Eriks son cl a l.. 1981; G reenw ald et a l.. 1984; H ar den, 1981; H ardell and E riksson, 1988; H ar den and S andslrom , 1979; H o ar el a l.. 1986; H o ar Z ahrn el a l., 1988a; K ang el a l.. 1987; S m ith el a t.. 1982, 1983, 1984; S m ith a n d Pearce, 1986; Sobol el a l.. 1986; V ineis el a l.. 1987; W oods el a l., 1987) an d studies using proportional m ortality ratio (P M R ) analysis o fro u tin cly collected death certificates (G al lagher an d T hrellall, 1984; M ilh am , 1982). T h e c a s e -c o n tro l stu d ie s o f S I'S c o m p rise 12 d is tin c t se ries, re p o rte d in 16 p u b lic a tio n s . O n e scries (Sobel el a l.. 1986) w as anecdotal a n d o f lim ite d v a lu e . O f th e 11 r e m a in in g s c ries, all b u t o n e e x am in ed risk asso ciated w ith exposure to phenoxy acids or chlorophenols p er se. T h e single stu d y w hich d id n ot (B alar ajan and A cheson, 1984) recorded a statisti cally significant risk lor the o ccu p atio n al group o f farm ers, farm m anagers, an d gar d eners. O f th e 10 stu d ies w h ich d id , a sta tisti cally significant association betw een S I S an d these substances w as observed in only 4 series (E riksson el a l.. 1981; H ardell an d Eriksson, 1988; H ardell an d S an d slro m , 1979; V ineis el a l.. 1987), w ith estim ated relative risks o f 5.1, 3.3, 6.2, a n d 15.5 respectively, for e x p o sure to phenoxy acids o r chlorophenols; none o f these three series reported a significant as sociation w ith farm ing o r agriculture. Two o th e r series reported statistically significant asso ciatio n s w ith o ccu p a tio n s in w hich ex p o sure to phenoxys an d chlorophenols m ight occu r (railw ays, m eal w orks, tannery, lum ber grading) hut no such relationship was o b served lo r the co m p o u n d s them selves (Sm ith e l a l .. 19 8 4 ; W o o d s c l a l.. 1 9 8 7 ). T h e sig n ifi cance o f the finding by W oods el al. (1987) o f a statistically significant increased risk o f STS o f 7.2, d u e to chlorophenol exposure am ong perso n s in the U n ited S lates w ith S can d in a vian n am es, is difficult to assess in view o f the very m any co m p ariso n s m ade in this study an d the la d that this was observed neither for M L n o r betw een S I S an d phenoxy acids. W ith regard to the tw o P M R studies, the study by G allagher and T hrclfall (1984) did n o t show an y significant risk o f ST S for any o ccu p atio n , although significant associations w ere observed betw een saw m ill and p ulp w orkers an d N H L . T he oth er by M ilham (1982) o b tain ed a statistically significant risk o f 1.52 fo r S T S a m o n g fa rm e rs (n o t o th erw ise s p e c ifie d ), b u t it w a s a ls o n o te d th a t th e 15 o ccu p atio n s w ith the highest P M R s for S I S did not include those w ith exposure to phe noxy acids o r chlorophenols. T his study also o b s e rv e d a s ta tis tic a lly s ig n ific a n t risk o f 1.9 4 for H o d g k in 's disease a m o n g p a p c r/p u lp m ill w orkers. T h e evidence from the few case-co n tro l stu d ie s lin k in g M L a n d ST'S w ith e x p o su re to phenoxys an d chloro p h en o ls has m et w ith controversy (Coggon and A cheson, 1982; R oyal C om m issio n , 1985). T h is controversy has centered around the quality and detail o f exposure in fo rm atio n an d th e possibility o f b ias in th e a sc e rta in m e n t o f exp o su re. Bias cou ld be intro d u ced from interview ers o b tain in g exposure history unequally in cases an d co n tro ls (interview er bias) or from u n equal recall o f exposure betw een case and control q u estio n n aire respondents (recall bias). W hile the existence o f these types o f bias can be suspected (C oggon an d A cheson, 1982) p ro o f o f th eir existence has been 222 l-KIC S. JOHNSON t a b i .i : I ( 'O I I O K I S i U D II S O I W o K K I K N I N M A N I JI-AI T l I K IN C /t-O K M I l | A N N O C l A N I S : O l i s i K V M ) Ni i m i h -.h o i J J j a i i i s / N i i m i i i k O I { A SI s Sludy T oinl deaths Total eunccis S IS NIM 1. D u l d e ru p u n d Z e lle n ru ih 2. I'a /d c m v a <'/<//. .V I h c is s u n d I rc n t/irl-D c y m c 1' 4. L eim en and S/adow ski" 5. O ll, H o ld e r a n d O ls o n '' (>. C o o k i ' l u l 1' 1. C o o k (7 <// ' X. O ll i / u / /' 9. Zack and Suskind 10. Z aek a n d C ia lle y I I . C ogt'o n <7 /.' 12. Hond<y<j/. 2,4-1) plant lolul eohori I.V l .y n g e '' IMienoxy d c p a n m c n i Lola! cohort 25 6 21 41 11 4 29K .170 12 5K 1019 6K III K 2 7 17 l 1 61 Kl 9 9 297 17 26 17 208 00 00 00 00 00 10 15 15 10 10 0 -> 02 02 10 5 (X) HD 0 0 0 0 0 0 1 1 l 0 1 l 1 ' T h e se Iw o s tu d ie s w ere Ira n i th e su m e p lu m . h l liew lour studies were Irom the sam e plum . 1T his w usu m ixed eolioil of.sprayers a n d w orkers in m unurueiuring. All three eases o f M l. were from m an u factu r ing. 1-igure in p a ren th eses includes NI II. a n d I ID co m b in ed . r Blanks indicate d ata not available. difficult to establish. Sim ilarly, P M R analyses sutler from certain types o f bias (Johnson, 1986; W ong and D ccoutlc, 1982; W ong cl a i, 1985) w h ich m a k e s it im p ru d e n t to in te rp re t as causal the relatively sm all excess risk o b served a m o n g farm ers in the study by M ilham (1982). C o h o rt stu d ies o f ad eq u ate size are m ore definitive, an d have the advantage over caseco n tro l stu d ies th at bias in th e ascerta in m en t o f ex p o su re is m in im iz e d , sin ce ex p o su re sta tu s is d e te rm in e d before th e o u tc o m e o f in te r est is k n o w n (M a u s n e ra n d B alm , 1974). l-urIh crm o rc, in o ccu p atio n al settings, exposure sta tu s is m o re readily a n d reliably ascer tained. T herefore, cohort studies o f occupa tionally exposed groups should provide the best evidence w ith w hich to evaluate causa tion, since experim ental studies can n o t be d o n e in h u m an s. A c c o rd in g ly , it is o f in te r e s t to re v ie w c r iti cally the evidence for the association betw een STS o r M L and phenoxys and chlorophcnols available from these types o f studies. If (he es tim ates o f increased risk derived from cohort studies corroborate those from casc-conlrol stu d ies, th e ease for cau sality is g reatly strengthened, as these tw o types o f studies are the prim e analytic designs em ployed in epi dem iology. T h e re is ad d ed significance to (his exercise, since the p u rp o rted increased risk o f these tu m o rs d u e to phenoxy/chlorophenol exposure in case-co n tro l stu d ies originates largely from exposure d u rin g the use o f these com pounds during spraying or application. T he m uch higher exposure d u rin g the m an u facture and form ulation o f these com p o u n d s ( P a tte r s o n c l a i . 19 8 9 ) h a s f e a tu re d v e ry little in c a se-co n tro l studies, possibly because fewer persons are exposed this way than d u r ing th e ir use. This review se p ara te s e v a lu a tion o f co h o rts involved in the use o f these substances (w hich has m ore direct relevance to case-co n tro l studies) from that o f cohorts SOIT TlSSUli SARCOMA, MAl.KiNANT I.YMIMIOMA 22.1 iarm :2 ('OIH)K I S lUD ii SOI W oKK I KS IN MANl HAC'I I <KIN< i/l'OKM I U ATINti l`l AN I S I <IK W lllCII Ri l a h v i R isk s (R R s) ok C o n i m i n < i In ii k va i s W i ki Ciivi n Study 7. C o o k <7 al. 8. O il r ia l. 11. C oggon c i at. 12. R ond 47al. lolui c o h o rt" 11. ly n g e l olui cohort* I'h e m u y d e p a rtm e n t S IS C o n fid en ce RR interval 1.1 RR 2.4 1.9 0.4 1.8 (1.2) M l. C o n lid e n c e interval 0 .8 -5 .6 0.6-4.5 0 .0-1.! IK ) C o n fid en ce RR interval 0 0 -5 .6 0.9 0 .0 5.1 0.1 0.0-1 6 2.7 0 .0 -1 4 .7 " A rela tiv e risk o f .1.1. w h ich w as sta tistic a lly s ig n ilic a n l. w as o b ta in e d for all h c in o p o ie tic /ly iiip h u lic ca n ce rs com bined am ong w orkers in the 2,4-1) plant. * lig u re in parenlh cscsd en o tes M l. unsjH xilicd. involved in m an u factu re an d fo rm u latio n . A fo rm al m clan aly sis will n o t be c o n d u e ie d as the data are too scanty, an d the definition o f exposure in m ost instances w as not co m p ara ble from stu d y to study. C O H O R T S FRO M M A N U FA C TU R IN G A N D FO R M U LA TIN G PLAN TS T a b le s I a n d 2 s u m m a riz e 13 c o h o rt s tu d ies (rep resen tin g n in e d istin ct co h o rts) o f w orkers in m an u factu rin g /fo rm u latin g p lan ts (R o n d cl til., 1988; C oggon cl ul.. 1986; C o o k cl al., 1980, 1986; D ald e ru p a n d Z ellenralh, 1983; L ehnert and Lynge, 1985; O il cl ul.. 1980, 1987; P azd ero v a-V ejlu p k o v a cl ul.. 1981; l.c h n e rl a n d S zadkow ski, 1985; T h iess cl ul.. 1982; Z ack a n d G alley , 1983; Z u ck a n d S u s k in d , 1980). A ll 13 s tu d ie s e x cept that by Lynge (1985) w ere m ortality s tu d ie s . O l 't h e 12 m o r ta lity s tu d ie s , 7 d id n o t record an y d ea th from S I S (R o n d cl ul.. 1988; C oggon cl ul., 1986; D ald eru p a n d Z cllen rath , 1983; L eh n ert a n d O tl cl ul., 1980; P azd ero v a-V ejlu p k o v a ci ul., 1981; L eh n ert a n d S zadkow ski, 1985; T h iess cl ul., 1982) and the rem aining 5 reported one death each (C ook cl ul.. 1980. 1986; O il cl ul.. 1987; Zack and G alley, 1983; Zack and Suskind, 1980). In all I 2 studies, the expected n u m b e r o f d e a th s from S I S w as less th an one. Broadly speaking, abo u t 180 cancer deaths sh o u ld o cc u r in a c o h o rt in o rd er to expect o n e death from soft-tissue sarcom a (R iggan c l u l., 19 8 3 ). A s se e n in f a b l e I , w ith o n e e x c e p tio n (C o g g o n c l u l.. 1986), n o n e o l 't h e 12 m ortality studies recorded total cancer deaths an y w h ere close to this figure, liven in the stu d y by C oggon cl ul. (1986) th e to tal n u m ber o f cancer deaths am ong w orkers engaged in m an u factu rin g /fo rm u latin g w as probably less th a n 18 0 , s in c e th e 2 9 7 c a n c e r d e a th s re corded in the cohort included sprayers as w ell. N o death from m alignant lym phom as was re c o rd e d in 7 o f th e 12 m o rta lity s tu d ie s (C ook cl ul.. 1980; D ald eru p a n d Z elle n ralh , 1983; L eh n ert a n d O il cl ul.. 1980; P azdcrov a-V ejlu p k o v a cl ul.. 1981; L eh n ert an d S zadkow ski, 1985; T h iess cl ul.. 1982; Z ack an d G alley, 1983). Based on the studies w ith av ailab le d a ta (B ond cl ul.. 1988; C oggon cl ul.. 1986; C o o k cl ul.. 1986; O il cl ul.. 1987) ab o u t 100 d eath s (total) should occur in an o ccu p atio n al co h o rt in o rd er to expect one d eath from N H L . flic highest n u m b er o f lotal d eath s recorded in these 7 studies w as only 58 (Z ack a n d G alley, 1983). In ad d itio n , about 30 and 60 deaths from cancers should o ccu r in o rd er to expect o ne d eath from ihiiiI lo d g k in 's ly m p h o m a ( N H L ) a n d I lo d g k in 's 224 I KK' S. JOHNSON .J,, ^ OJ disease (IID ), respectively (R iggan cl a l. 198.1). R ased o n th is c rite rio n also , ii is seen from T able I Ihal d ie sizes o f these 7 co h o rts w ere clearly in ad eq u ate to investigate any but extrem ely large risks from M l., as the highest total n u m b er o f d eaths from cancers recorded in a n y o f th e m w a s o n ly 17 (L e h n e r t a n d Szadkow ski, 1985). O fth c rem ain in g 5 stu d ies w ith at least o n e M L d e a th recorded (B o n d cl a l , 1988; C'oggon cl it/.. 1986; C o o k c l ill.. 1986; O il cl til., 1987; Z ack a n d S u sk in d , 1980), all ex cep t th a t by C'oggon cl ul.. show ed al least a ro u n d a tw ofold no n sig nificant increased risk o f N lll. o r III). |N o ic th a t th e stu d y by O n cl til. (1 9 8 7 ) is a n u p d a te o f th e s tu d y b y C o o k cl til. (1 9 8 6 ).] In o n e o f these studies (B ond cl t i l. 1988) the increased risk o f .1.1 ( fo r a to ta l o f liv e d e a th s ) o f t u m ors o f the hem o|X )ielic/lym phalic system s am o n g w orkers in a 2,4-D plant was statisti cally significant, b ut risks lor ly m p h o m as w ere not given separately. T w o o f the live d eaths were due to N H L , one d u e to I ID , an d two d u e to leukem ia. T he D anish cancer incidence study by Lyngc (1985) reported an excess o f STS am ong cohort m em bers from tw o plants w hich m an u factu re d p h en o x y acids as well as a variety o f o th e r chem icals. Live cases o f STS o ccu rred in the co h o rt as a w hole an d the rel ative risk (R R ) for ST S w as 2.72 [95% confi d en ce in terv al (C l), 0 .8 8 -6 .1 4 ]. W h en a 10y ear latency w as allow ed for, the R R w as 1.67 (9 5 % C l, 1.0 - 9 .1 9 ) . T h e d u r a t i o n s o f e m p lo y m e n t fo r th e liv e c a s e s w e re .1, 1, 0 . 5 , 1 0 , a n d 90 m o n th s. L our o f the live STS eases had p o tential for exposure, three o f them having w orked in the general ship p in g d ep artm en t, w here p resu m ab ly all o th e r ch em icals p ro d u ced by the factory w ere h an d led as well, an d the fourth in the phenoxy herbicides packaging dep artm en t. T his cohort w as not com plete for persons hired before the m id1950s. T h e p re d o m in a n t |)h cn o x y herbicides produced by this factory w ere 2-m elhyl4-ch lo ro p h cn o x y acctic acid o r <(2-m clhyl- 4-chlo ro p h eiio x y p ro p io n ic acid (M C P A / M C PP) an d , to a lesser ex ten t, 2,4-dichlorophcnoxyacelic acid and 2,2,4-dichlorophe- noxypropionic acid. Sm aller am o u n ts o f 2 ,4 ,5 -1 (acid an d esters), as well as hcxachloro p h en e, w ere also pro d u ced . 2 ,4 ,5 -T richlorophenol (T C P ) was produced lor a very lim ited period (1951-1952). but only one ol the S TS e a s e s h a d b e e n e m p lo y e d d u r in g its p r o d u ctio n . I he factory also produced large am o u n ts o f other chem icals such as aniline dyes, organic an d inorganic dyes, and pig m en ts, a n d fo rm u lated various insecticides. It is to be n o ted th a t o f the fo u r ST S having possible exposure, four different histologic types w ere involved (liposarcom a, fibrosar com a, derm atolibrosarcom a o r hem angio pericytom a, and leiom yosarcom a). O nly one o f the four d id not o ccu r in an in ternal organ; based o n D an ish n a tio n a l d a ta (L ynge cl al., 1 9 8 7 ) o n e w o u ld h a v e e x p e c te d 1.4. Bight cases o f m alig n an t ly m p h o m a (one in a fem ale) w ere observed in the en tire D an ish co h o rt, an d the n o n sig n ifican t relative risk s w ere 1.1 a n d 0 .8 fo r m a le s a n d fem ales, respectively. N o case o f M L o ccurred am o n g w orkers specifically engaged in the m anulactu re/fo rm u lalio n o f phenoxy acids, even th ough 0.9 w as expected. A statistically sig nifican t relativ e risk o f 2.91 for M L w as o b served am ong w orkers engaged in the p ro cessing o f products o th er th an phenoxy acids. C O H O R T S O F U SERS (SPR A Y ER S, A PPLICA TO RS, A G R IC U L T U R E / F O R E S T R Y W O R K E R S , E TC .) A to ta l o f 16 o c c u p a tio n a l c o h o r t s tu d ie s o f users fro m ' 12 d istin c t c o h o rts h av e been co n d u cted to d ate (A xelson an d Sundell, 1974; A xelson cl al., 1980; B arlhel, 1976, 1 9 8 1 ; B la ir c l a l.. 1981; C'o g g o n ct a t., 1986; G reen, 1986; flogsledl and W esterlund, 1980; M a c M a h o n ct at.. 1988; M o rg an cl al., 1980; R iih im ak i cl al.. 1982, 1983; R oyal C om m ission, 1985; W ang and M aeM ahon, 1979; W iklund cl t il. 1987, 1988) (T ables 1 a n d 4). T w o rep o rts o f the sam e study (M ac M a h o n cl al.. 1988; W ang an d M a eM ah o n , 1979) did not provide data on STS and M L, even th o u g h ihe u p d ate (M aeM ah o n cl al.. S O IT TISSUI-: S A R C O M A , M A I IC iN A N l I.Y M R IIO M A 225 T A IM .Ii .1 COIIOK I S I t IDII SOI l I.SI.KS (Sl'HAVl kS, APPI I(A IO KS): OllSt KVt I) NUMIMK Ol l ) l A | IIS/NOMtiJ K Ol C a.SI S Sillily filial deaths Total cancers 1. K iih in ia k i <7 a l " 2 . (Incidence) K iihiniaki <7 a t* .1 . (In c id en c e ) B a rth e l 4, (Incidence) 1 logstedt a ul. 5. Circen <7 al. 6 . (Incidence) A xelson a n d SuD dcll'' 7. Axelson <7 / * 8 . (Incidence) M organ <7 a t 9. W ang and M acM ahon* 10. M a c M a h o n c l a l.1 11. B lair e l at. 12. ( in c id e n c e ) B a rth e l 13. (In c id e n c e ) W iklund *7 a l.'1 |4 . (Incidence) W iklund e l a l J 15. C 'oggon e i u l ? 16. R o y al C o m m is s io n R e p o r t ' 144 29 HO 10 20 * < 62 311 I0K2 * .7K . 10.19 692 26 26 30 K 5 IK K 2 6 .15 1.1 47 2 .12 K4 169 297 169 U' J S a m e c o h o rt. * T his was a m ixed cohort o f sprayers m id m anufacturing workers. 1Figure in parentheses d enote M l. unspecified. STS 0 0 0 0 0 0 0 0 0 0 0 1 6 1 1 N ili 0 0 0 0 0 0 0 1 0 > 21 0 id III) 0 0 0 0 0 0 1 (1 0 > II 0 1988) w as large en o u g h lor eases o f b oth dis lilied: on e case o f STS and two cases each o f eases lo h av e o ccu rred ; it w as rep o rted N H L an d H O w ere observed, w hich w ere not th o u g h th a t tu m o rs o f th e h em o p o ietic/ly m - significantly in excess. O f the o th e r 13 studies, phalic system s w ere n o t in excess. The study the results lor o n e cohort lor S IS an d M L h y B a rth e l ( 1 9 8 1 ) w a s s u b je c t lo s e le c tio n b ia s w e re r e p o r te d s e p a ra te ly ( W ik lu n d </ n l . as o n ly 70% ol th e stu d y p o p u la tio n w as id en - 1987, 1988). N o o ccu rren ce o f S I S w as re- TA III.U 4 C o i to t i I . S m m r s o t lt.si k.s (S i' k a y i k s. A t'i'i i( 'a i o k s ) i-o k W ittc ii Kt t a iiv i R isk s o k C o N i - t o m i i; In i i k v a is W t k i: ( iv i.n M ill. Ill) Study C o n iid c n c c C o n fid en ce ( onlidcncc KR interval RR interval RR interval 13. W ik lu n d *7 al. 0 .9 0 ..1 - 2 . 0 14. W ik lu n d r / a l 1 . 0 0 .6 - 1 .5 1 . 2 0 .6 - 2 . 2 16. R o y a l C o m m is s io n R e p o r t" 1.4 " l'Kurc in p u iviilluiscsiL'iiolcs M l. uilsjicL'ilial. 226 I RK' S JOHNSON p o lle d in n in e stu d ies (A xelson a n d S undell. 1974; A xelson cl u l . 1980; B arlhel, 1976; B lair ct ri/.. I98.V. C irecn , 1986; I lo g sie d l a n d W eslerlund, 1980; M organ e tu i, 1980; R uin- n u t I, i </ ill.. 1 9 8 2 . 1 9 8 2 ). In g e n e ra l, o v e r 100 ineidenl eases o f can cer sho u ld occur in o rder to c x |ic d one ease o f STS; to expect one ease each o f N lll. an d III), a total o f at least 40 an d 70 eases o f cancer, respectively, should o c c u r (C u tle r a n d Y o u n g , 1 9 7 5 ; W a te r h o u s e c l ill.. 1982). U sin g these crite ria , a n d th o se e s ta b lis h e d a b o v e lo r m o r ta lity s tu d ie s , it is seen front T able 2 that none o f the nine stu d ies w ith o u t a ease o f STS w as sullicicntly large to e x a m in e risk o f S 'l'S . In n o n e o f th e re m ain in g three studies w ith a recorded case(s) o f S I S was any statistically significant excess o f th e d ise ase o b se rv e d (C 'oggon c l u l , 1986; Koyul C om m ission, 1985; W iklund ci u l. 1988). Six eases o f S I S w ere recorded in the largest co h o rt (W iklund cl u l. 1988) an d the relative risk w as 0.89; 72% o f the co h o rt had p o tential lor exposure to phenoxy acids. H ow ever, the m ax im u m follow -up period w a s o n ly 18 y e a rs, a n d it is n o t k n o w n w h a t p ro p o rtio n o f the c o h o rt h ad a lim e since lirsl e x p o s u re lo n g e r th a n 18 y e a rs. T h u s, it is difficult to assess w hether the latency period w as adequate. Sim ilarly, w ith respect to m alignant lym p h o m as, only live stu d ies rep o rted an y d eath from , o r any case o f lym phom a (A xelson cl u l. 1980; Barlhel, 1981; M organ d a / .. 1980; R oyal C o m m issio n , 1985; W ik lu n d cl ul.. 1987). In n o n e o f these live stu d ies w as a sig nificant excess o f m alignant ly m p h o m a re p o rted , alth o u g h in at least tw o o f th em (A x elson d a l , 1980; M organ cl u l. 1980) a ease o ccu rred w hen n o n e w as ex p ected . Six stu d ies w ere obviously not large en o u g h to o b serve a case o f N lll. o r III) (A xelson and Sundell, 1974; Barlhel, 1976; G reen, 1986; Ilogsiedl and W cslcrlund, 1980; R iihim aki cl i l l , 1982, 198.2). Two stu d ie s d id n o t in d ic a te the im ilings for these m in o rs (M acM ah o n cl ul.. 1988; W ang an d M acM ahon, 1979) and a fu rth er o n e (B lair cl ill., 1982) m en tio n ed only that there was no excess. N o case o f M I. occurred am o n g sprayers in the m ixed cohort o f sprayers and production w orkers by Coggon cl ul. (1 9 8 6 ; p erso n al c o m m u n ic a tio n , 1988). A N IM A l. DATA A b rie f referen ce will lie m a d e to th e a n i m al data, to exam ine to w hat degree they com plem ent those lor hum ans. 2,4-D ichlorophcnol and 2,4,5-trichlorophcnol each acted as a skin tu m o r p rom oter w hen given alter a single in itialin g dose o f 0.2% d im clh y lb cn z[tt]an lh racen e (D M B A ) in b en zen e (B o u lw ell a n d B osch, 1959). In 2year feeding studies in m ale and fem ale rats an d m ice fed d iets co n tain in g 2,4-dichlorophenol, (here was no evidence o f carcino genic activity (N ational T oxicology Program , 1988a). 2,4,6-T richlorophenol has been reported to cause an increased incidence o f reticulum cell sarco m a an d h e p ato m a in m ice (In n cs ct ul.. 1969). It w as also fo u n d to be c a rc in o genic in m ale rats, in ducing ly m p h o m as or leukem ias, an d to be carcinogenic in m ale an d fem ale m ice, inducing hepatocellular carcinom as or adenom as (N ational C ancer Institute C arcinogenesis T echnical R eport, 1979). Pure pentachlorophenol (PC P) did not show evidence o f carcinogenicity in rats even at th e highest dose o f 20 m g/kg, although there w as suggestion that the anim als could h av e to lerated a h ig h er dose (S chw eiz cl ul.. 19 7 8 ). A n e o p la s tic ctl'ecl o f P C P w a s n o t o b se rv ed in m ice o b se rv ed fo r 18 m o n th s (In n e s cl u l. 1969). In a N atio n al T oxicology P ro g ram (N T P ) stu d y , it w as c o n c lu d e d th at there w as evidence o f carcinogenic activity for m ale m ice fed d iets co n tain in g technicalgrade PC P, as show n by increased incidences o f adrenal m edullary and hepatocellular neo plasm s. T h ere w as also an indication o f carci n ogenic activity for fem ale m ice exposed to technical-grade PC P, as show n by increased incidences o f hcm angiosarcom as and hepa tocellular neoplasm s. A carcinogenic activity w as observed for m ale m ice exposed to a s o n Tissui-: s a r c o m a , m a l ig n a n t l y m p h o m a 227 purer grade o f P C I\ D ow icidc h t'-7 , as show n by increased incidences o f adrenal m edullary and hepatocellular neoplasm s, an d lor fem ale m ice also, as show n by in creased incidence o f adrenal m edullary and hepatocellular neoplasm s and hem angiosnrcom as (N ational Toxicology Program , 1988b). M ale rats fed a d iet co n tain in g a high dose o f 2,4-dichlorophenoxyacelic acid show ed a statistically significant increased incidence o f b rain tu m o rs (H azleto n L aboratories, 1986). H ow ever, an expert panel review ing this study concluded that lor various reasons, there w as insufficient evidence to conclude that these tu m o rs w ere related to 2,4-D expo sure, particularly as an increase o f these tu m o rs w as n ot observed in fem ale rats (C ana d ian C entre for T oxicology, 1987). U n fo rtu nately, o th er studies o f 2,4-D have been lim ited by insufficient sam ple si/e o r inad e qu ate reporting, and an unequivocal d em o n stration o f carcinogenic activity has not been evident (H azleton Laboratories, 1984; Inter national A gency for R esearch on C ancer, 1977). S im ilarly, studies o f the carcinogenic ity o f 2 ,4,5-T a n d M C PA in an im als have g en erally b een sp arse a n d sull'cr fro m th e sam e lim itations as the 2,4-D studies (In ter national Agency for R esearch on C ancer, 1977, 1983). In su m m a ry , w hile th ere is sign ifican t e v i dence that chlorophenols have lum or-prom o lin g an d -in itiatin g cflects in an im als, a sim ilar effect has not been observed for any o f the phenoxyacelic acid herbicides, partly because m any o f the studies o f this group o f co m p o u n d s w ere lim ited, because o f sm all num bers o f anim als used. T h e currently available d a ta are in ad eq u ate to fully assess the carcinogenicity o f these co m p o u n d s in a n im a ls; th u s it is n o t k n o w n w h e th e r o r n o t p h en o x y s are carcin o g en ic in an im als. S U M M A R Y A N D D ISCU SSIO N cu m u lalcd sufficient person-years o f observa tio n for ad eq u ate ev alu atio n o f risk o f soltlissue sarcom as, since the expected num ber o f d e a th s fro m S T S in each o f th e m is less th a n o n e. It is estim a te d th at if the expected n u m b e r o f d e a th s fro m S I S in a study is on e. a n u n d erly in g risk o f d isease will have to be at least livefold for it to be d elected at th e 95 level o f significance an d w ith a statistical p o w e r o f 0 .8 ( B e a u m o n t a n d H rcslow , 1 9 8 1). A n in c re a se d risk o f S I'S w as rec o rd e d in th e single can cer incidence study by l.ynge (1985). H ow ever, in that study, three o f the fo u r ca se s o f S I'S w ith p o te n tia l for e x p o su re to phenoxy acid herbicides w orked in the general shipping departm ent, w here presum ably they had potential for co n co m itan t ex posure to a large n u m b er o f chem icals o th er th an phenoxys o r chlorophenols, w hich were also p ro d u c e d in th e factory. It is to be noted a m o n g th ese stu d ie s th a t a case o f S I'S o c cu rred in th ree c o h o rts (rep o rted in live stu d ies) even th ough n o n e w as expected. S im ila rly , 9 o f th e 13 u s a b le s tu d ie s o f sp rayers/applicators w hich provided infor m a tio n o n S I'S specifically d id n o t re p o rt a case o f the disease, and are therefore not ad e q u ate at the present tim e to evaluate risk o f S I'S. O n th e o th e r h a n d , in n o n e o f th e re m ain in g lo u r co h o rts w hich w ere sufficiently large o r w hich recorded a case(s) o f S I S. was any statistically significant excess o f the d is ease observed. Sim ilarly, in co n trast to the sit u ation w ith m anufacturing cohorts, n o case o f STS o ccu rred w hen n o n e w as exp ected in these cohorts o f sprayers and applicators. It is w o rth recollecting th at o n ly fo u r o f the ten series o f case-co n tro l studies (and a possi ble liltli) w hich have specifically investigated the role o f phenoxys and chlorophenols have delected an y statistically significant associa tio n b etw een S I S a n d these c o m p o u n d s , in the stu d y g roup as a w hole o r an y subg ro u p o f it. Soil- T issue S u ra m in s M aliym inl l.ym /i/um ius N o n e o f th e 12 m o r ta lity s tu d ie s o f m a n u Light o f the tw elve m o rtality stu d ies o f facturing cohorts conducted to date have ac- m anufacturing cohorts have not accum u- 228 r.RIC' S. JOHNSON lulcd sullicicnt yours o f observ atio n to ad e q u a te ly e v a lu a te risk o f n o n - H o d g k i n 's ly m p h o m a a n d H o d g k in 's d is e a s e . In n o n e o f th e rem aining four studies w hieh w ere sullicienlly large o r w hich recorded a case(s) o f M l. in the c o h o rt, w as a statistically signifi can t elevated risk o f th e disease observed. H o w e v e r, o n e c a s e o f H o d g k in 's d is e a s e w a s recorded in o n e co h o rt w hen n o n e w as ex pected, an d a t least aro u n d a tw ofold excess o f N H L o r H D (w hich w as not statistically significant) w as observed in tw o coh o h s. Sim ilarly, in the sp ray cr/ap p licato r co horts, o f the six stu d ies w hich w ere sullicicnlly large o r w hich recorded a case(s) o f M L, n o n e show ed a statistically signilicant increased risk o f th e disease. Bight studies w ere n o t sufficiently large to ev alu ate risk from M L. N evertheless, a case o f M L was re co rded in tw o o f them w hen n o n e w as ex pected. A s w ith soft tissue sarco m as, it is w o rth rec ollecting here also th at only three o f the eight case-co n tro l studies (and a possible fourth) w hich have specifically investigated the role ofphenoxys and chlorophcnols have detected an y statistically signilicant association be tw een M L an d these c o m p o u n d s, in the study g ro u p a s a w h o le o r an y su b g ro u p o f it. O n e im p o rtan t issue th at needs to be ad dressed additionally, for a com plete and thor o u g h review o f th is lite ra tu re , is th e p ro b lem o f concom itant exposure to o th e r chem ical or biological agents w hich are potentially carci nogenic. B earing in m in d th at the o nly evi dence from case-co n tro l studies linking these co m p o u n d s to cancer occurrence in hu m an s is based o n e x p o su re d u rin g th e use o f these c o m p o u n d s, it is also p e rtin e n t to n o te th a t persons w ho occupationally use phenoxys an d chlorophcnols (sprayers, applicators, farm ers) are invariably exposed to several o th er chem icals or oncogenic viruses conconiilanlly. T hus, unless these oth er polenHally carcin o g en ic ex p o su res w ere adeq u a tc ly c o n tr o lle d f o r in th e s e s tu d ie s , it _ c o u ld e q u a lly b e a r g u e d th a t a n y in c re a s e d risk observed in o ccu p atio n al g ro u p s w hich C /^ h a v e used p h c n o x y s/c h lo ro p h e n o ls is d u e not to them , but to one o r m ore o f these other concom itant exposures. T h e follow ing are consistent w ith this hy pothesis. I. In relation to m alig n an t lym phom as, o u t o f a total o f nine studies review ed w hich have investigated both phenoxy acids/chloroph cn o ls o r herbicides, as well as so m e o f the oth er con co m itan t exposures in any detail (B u rm cister cl al.. 1983; C a n to r, 1982; C a n to r ct al,, 1985; E v erett ct ut.. 1985; H ard ell cl ill., 1981; H o a r cl a!., 1986; H o a r Z a lin i cl nl., 1988a; H o a r Z a lim cl ill., 1988b; S aftlas cl ill., 1987; W o o d s cl at.. 1987), w ith o u t ex ception, a statistically significant increased risk o f M L w as o bserved in all o f these stu d ies lor one o r m ore concom itant exposures other than phenoxy herbicides, such as nonplicnoxy herbicides, insecticides, fungicides, sol vents, fertilizers, etc. T h is consistency w as not seen even w ith the can didate phenoxys/ c h lo r o p h e n o ls th e m s e lv e s . F u r th e r m o r e , it app ears that in nearly all o f these instances, the highest statistically significant relative risk recorded w as for o n e o r m ore o f these co n co m itan t exposures rather than for expo sure to phenoxy acids, chlorophcnols, o r her bicides. S im ila rly w ith S I'S, few stu d ie s h a v e in v e s tigated o r provided d ata on the role o f any o f these oth er exposures individually (E riksson clnl., 1981; Ilardell and E riksson, 1988; H ar d en an d S a n d slro m , 1979; I lo a r cl al., 1986; 1lo a r Z alim cl al.. 1988a; K an g cl al.. 1987; W o o d s cl til.. 1987). T w o o f th em (H o a r cl al.. 1986; H o a r Z a h m cl ul,, 1988a; K an g cl a l., 19 8 7 ) d id n o t o b s e rv e a n y s ta tis tic a lly sig nificant association w ith phenoxys o r cliloroplicnols, b u t on e recorded a statistically sig nificant association w ith exposure to insecti cides (H o a r Z alim cl nl.. 1988a). 2. N o stu d y has given the risk o f M L o r S I'S fro m e x p o s u re to p h e n o x y s a n d c h lo r o phenols alter exhaustively controlling for these co n co m itan t exposures, w hich m ay be ex trem ely difficult to achieve, as it requires a p riori know ledge of, a n d d a ta o n , all such p o ssib le ex p o su res, it is n o te w o rth y th a t in s o n iissiii-: s a r c o m a , m a l ig n a n t i .y m im io m a 22M ihc stu d y by W o o d s cl al. (1 9 8 7 ) w hen o th e r variables an d sonic co n co m itan t chem ical exposures w hich w ere found lb be signifi can tly associated w ith ML. in the stu d y w ere en tered in a logistic regression m odel, the risk lor exposure to phenoxyacetic acids only w as 0.85. 3. F o r both ML. an d STS, so m e ca se -c o n trol stu d ies have rep o rted a statistically sig n ifican t asso ciatio n w ith o c c u p a tio n s in w hich exposure to phcn o x y s/ch lo ro p h cn o ls m ight occur, o r even w ith spraying o f herbi cides, but no such association w as observed w ith the co m p o u n d s them selves. F u rth e r m ore, although an association w as observed betw een M l- an d saw m ill o r p a p e r/p u lp m ill w orkers in the tw o P M R studies, n o such as sociation w as observed for eith er o th er occu p atio n , o r all o ccu p atio n s c o m b in ed , in w hich exposure to phenoxys and chlorophenols m ight o ccu r (G allagher and T hrelfall, 1984; M ilh am , 1982). S im ilarly, in the PM R s tu d y by M ilh a m (1 9 8 2 ) th e 15 o c c u p a tio n s w ith th e h ig h est risk o f S I'S d id n o t in c lu d e o ccu p atio n s in w hich exposure to these co m pounds occurs, and no association betw een STS a n d all o ccu p atio n s (co m b in ed ) in w hich exposure to phenoxys an d chlorophenols m ight occur w as observed. 4. In the D anish cancer incidence study o f m a n u fa c tu rin g c o h o rts by l.y n g e (1985) in w h ich it w as p o ssib le to se p a ra te p h e n o x y -e x poscd w orkers from those engaged in the p ro d u ctio n o f o th er chem icals, the statistically significant excess risk o f M L o b tain ed for the factory w as confined to w orkers engaged in the m anufacture not o f phenoxys, but o f th e s e o t h e r c h e m ic a ls . In th a t s a m e s tu d y it is n o te d a lso th a t th e ex cess o f S I'S o c c u rre d am ong w orkers in the shipping dep artm en t o f the factory where exixisure to com pounds other than phenoxys/chlorophenols occurred. 5. It is n o tew o rth y th at in n o n e o f th e lo u r cohort stu d ies o f users (i.e., sprayers an d a p plicators) o f reasonable si/e (B ail lid , 1981; O oggon cl ul.. 198b; R oyal C o m m issio n , 1985; W ik h m d cl at.. 1987, 1988), w as an y significant risk o f STS o r M l. observed. T his is n o t su p p o rtiv e o f th e c a se -c o n tro l stu d ies w hich h av e fo u n d significant risk in asso cia tion w ith the use o f phenoxy acids an d chlo ro p h e n o ls. It is im p o rta n t to n o te th at since p erso n s w ho use phenoxys and chlorophenols invari ably are exposed to oth er candidate expo su re s sim u lta n e o u sly , it is o b v io u s th at s tu d ies in w hich risks o f S I'S a n d M l. have been given for exposure to phenoxys an d ch lo ro phenols m ainly through their use (case-co n trol studies, occupational cohort studies o f users) m ay not be sufficient by them selves to ev alu ate risk associated w ith these c o m pounds, unless these concom itant exposures have been adeq u ately controlled for in the a n aly sis, o r th e a n aly sis is restricted to a study g ro u p o r a subgroup w ithin (he data w hich used phenoxy herbicides o r chlorophenols exclusively. F ith cr o f these tw o alternatives has been difficult to achieve, since the form er requires a priori know ledge o f and d ata on al least all the relevant co n co m itan t o ccu p a tional exposures, som e o f w hich have already been show n to have independent eliologic roles in the occurrence o f M l. an d STS, while a n a d eq u ate sam p le size m ay be a lim iting factor for the latter even if such a group or subgroup can be identified. T h e value o f ex isting case-co n tro l studies, and cohort stu d ies o f users, m ay therefore be restricted be c a u s e o f th e s e fa c to rs . 1'a r c n lh c lic a lly , th e re is need to investigate Ihc role o f these o th er co n co m itan t exposures in tile etiology o f these tu m o rs m ore thoroughly th an has been d o n e in the past. F u tu re epidem iological studies should be directed tow ard further fol low -up a n d /o r pooling together o f o ccu p a tional co h o rt studies o f w orkers exclusively or p red o m in an tly engaged in the m anufacture o f phenoxy acids and chlorophenols or of other suitable uniquely exposed groups. C O N C l.U S IO N It is c o n c lu d e d fro m th is review th at th e w eight o f evidence does not unequivocally in crim inate phenoxy acids o r chlorophenols as a cause o f m alignant lym phom as an d soli-lis- 230 liRIC S. JOHNSON sue sarco m as in h u m an s, a t least from expo su re d u rin g th e H.vcof th ese c o m p o u n d s . T h is d istin ctio n betw een risk arising from expo sure d u rin g use an d risk arising d u rin g the m anufacture o f these com pounds m ay be o f im portance in the future, as the intensity o f ex p o su re is c o n sid erab ly d ifferen t in th e tw o groups. Studies in w hich these co m p o u n d s have been purported to be im plicated have been plagued w ith the intractable problem o f co n co m itan t exposures, an d it seem s plausi ble th at so m e o f these o th er exposures, w hich w ere not all ad eq u ately co n tro lled for in these studies, co u ld w ell ac c o u n t for at least a sig nificant part, if n o t all, o f the ap p aren t associ ation. O n the oth er hand, the vast m ajority o f occupational cohort studies have not accu m ulated sufficient person-years o f observa tion to date, yet even in th e presence o f this deficiency, cases o f M L an d STS have oc c u rre d w hen n o n e is ex p ected . S im ilarly , a l th o u g h th ere is at p resen t n o co n clu siv e ev i dence that phenoxy acids cause c an cer in an i m als, th ere is sig n ifican t ev id en ce th a t so m e o l'lh e c h lu rn p h e n o ls do. It is th erefo re p o ssi ble that the failure to d em o n strate conclusive evidence o fcarcin o g en icily in m ost o f the a n im al an d h u m an studies m ay be the result o f lim iting factors in these studies rath er than tru e lack o f th is ell'ect. It is p ru d e n t th erefo re to w ail for fu rth er an im al an d h u m an data, particularly to w ail until existing occu p a tional cohorts have been com bined an d /o r observed for longer periods, before a clear, decisive, an d unam b ig u o u s ev aluation can be m ade on the potential carcinogenicity o f these very im p o rtan t com pounds. lU il'L R L N C ES ANOl KSON. I t. A .. llANKAHAN, t.. IV. Jl-NSI N, M ., i.AHKIN, I).. Yli K. W .. ANO Wl-IOMAN, IV ( l lJH5). H 'tw ii/n u i I ic lth w i IW i'iiiii M tir la lily S lu tly W iscon sin Division o f I Icnllh. M adison. A x i i .s o n . (>.. a n d S liN lil l l , 1.. ( IV74). H e rb ic id e e x p o sure. m ortality and lum or incidence. An epidem iolog ical in v e stig a tio n o i S w ed ish r a ilro a d w o rk ers. H 'ttrk A J-.uvm./i H e a lth l l , 2 l - 2 H. ,, A |S I 1>N, ( ) -Sl'NIIl I I . I ... ANIII KSSON, K ., lilll INIi, I It kin i m l , t '.. a n d K l in i i. II. ( IVXOI. I lerh icid e C v \ 'x p o sn re a n il lu m o r m o rla lily . A n u p d a te d c p id e m io - fO CD logic investigation on Sw edish railroad w orkers. S c u n d .J . W ork E nviron. Ile a llit 6 73-79. Ba i a r a ja n . R .. a n d A r m .s o n , I; D. (1984). S oft-tissue sarcom as in agriculture and forestry workers. J. fc/ddem iol. C o m m u n . H ealth 38, 113-116. Ba k t iii-:!.. li. (1 9 7 6 ). H ig h in c id e n c e o f lu n g c a n c e r in p e rs o n s w ith c h ro n ic p ro fe ssio n a l e x p o s u re to |>esiicidcs in agriculture. ICrkrund. A im . O r#. 146, 2 6 6 274. Ba k t ii!:!., li. (1 9 8 1 ). In cre a sed risk o f lu n g c a n c e r in p es ticide-exposed m ale agricultural w orkers. J. 7 bxicol. hnviro n . H ea lth 8 , 1027-1040. Bh a ijm o n i . J. J., a n d Br h s io w , N. E. (1981). Pow er considerations in epidem iologic studies o f vinyl chlo ride workers. A m er. J. I.p u lc in m l 1 14, 725-734. Bin a s c iu , S ., S a n t o r o , CL, a n d In i k o i n i , U . (1 9 8 5 ). H y p o th e sis o f o c c u p a tio n a l e tio lo g y o f H o d g k in 's a n d n o n -H o d g k in 's ly m p h o m a : A c a s e - c o n tr o l s tu d y , ( i ! u d . M e d . Imv. 7 , 1 8 1 -1 8 6 . Bis h o p , ( '. M , a n d Jo n i:s , A. H . (1 9 8 1 ). N o n -H o d g k in 's ly m p h o m a o l'lh e scu lp in w o rk e rs e x p o sed to d i oxins. l.ancel 2, 369. Bi .a ik , A ., G r a u m a n , D . J ., L u b in , J. H , a n d F k a u m i-n i, J . F . (1 9 8 3 ), L u n g c a n c e r a n d o th e r c a u se s o f d eath am ong licensed pesticide applicators. J. N ad. C ancer Insl. 71, 31-37. Bo n d . G . G , W i i i i r s i k o i .m , N. ! !., K o u s i t, G J ., M( l.AKI-.N, li. A., l.IPPS, T . E., AND C(X)K, R. R. (1988). C ause specific m ortality am o n g em ployees e n gaged in the m anufacture, form ulation, or packaging o f 2,4-dichlorophcnoxyucciic acid an d related salts. lin t J. Ittd M ed 45,98-105. U o u i w n i , R. K., a n d B o sc u , I). K. (1959). T he tu m or-prom oting action o f phenol and related com p o u n d s fo r m o u se sk in . C a m e r R ex. 1 9 , 4 13 - 4 2 4 . Bract, K. I). (1986). A gricultural chem ical use, type of larm prixluction. an d farm cou n ty c ancer m ortality in the rural U nited Slates: A n ecological m odel. I has. Ahxtr. In f [//) 46, 2879. Br i m in , P.. K a n o , 11. K ., l i t : , Y., Bu k i . V., a n d S in p a k d , B. M .( 1988). P ro p o rtio n a te m o rta lity stu d y o f IJ.S. A rm y an d U.S. M arine C orps V eterans o f the V ie tn a m W a r. J. O c c u p . M e d . JO, 4 1 2 -4 19. Bk o w n n o n , R. O ., a n d R t:i i J. (1 9 8 8 ). A c a n c e r registry-hased study o f o ccu p atio n al risk lor ly m phom a, m ultiple m yelom a and leukem ia. Ini. J. ilp td e m io i 17,27-32. B u t s o UNO, I). P ., a n d W o i .i s i a d i , l.. (1 9 8 4 ). C a n c e r m orlalily am ong farm ers. J. N ad. C ancer last. 72,503. lU m M H SH K , I.. 1\ (1981). C an cer m o rlalily in Iowa farm ers, 1971-78. J. N a il C ancer Inxi 66,461-464. Bu k m i t s i i r , 1.. F ., F v i K i r n , Ci. 1).. V a n Lii-r , S. F,, a n j ) Isa ( s o n , P. (1 9 8 3 ). S elected c a n c e r m o n a iity a n d farm practices in Iowa. A m er. J. I-./ndem iol. 118, 7 2 77. C 'anadian ( V ntre for Toxicology (1987). ilx p e r t P a n el R e jn tr i o n ( 'a r c in o g e n i e n y o j 2 .4 -D . G u e lp h , O n ta rio , C an i o k , K. P .( 1982). I a rm in g an d m orlalily from non- SOIT TISSIJI: SARC OMA, MALIGNANT LYMPHOMA 231 H odgkin's lym phom a: A c a se-co n tro l siudy. h it. J. L V n , M. J., N a ir n , J. R., ( 'ouh in , I). M ., a n d A d in a . C ancer 29, 239-247. M . A. (1987b). M ortality am ong A ustralian conscripts C a n i 'OR, K ., l :v i ;R i; r i , G ., Bi a i k , A ., G ib s o n , R ., o f th e V ie tn a m co n flict era . II. C a u se s o f d e a th . A n ie r So j u m a n , I... a n d Is a c s o n , P. (1 9 8 5 ). P u n n in g a n d J E p id em io l 125,878-884. n o n - H o d g k in 's ly m p h o m a . (A b s tra c t) o f M e e tin g o f P iN iit K i u n , M ., H a i r i .r in , W . L.. H o n o i a k . P. a .. the Society for Lpidcm iologic Research, . in te r J. E p i SM IHI, A. B., CiRO III, I). II., AND RlfS.SI.lt , W. O. dem iol. 122, 535. (1984). An evaluation olT cports o f dioxin exposure (e n te rs lor D isease C on tro l V ietnam Lx|>ericncc Study and soft-tissue sarcom a pathology am ong chem ical (1987) . Post service m ortality am o n g V ietnam veter w orkers in the U nited States. S e u n d J W ork E nviron ans. J. Anier. M ed. Assoc. 257, 790-795. H ealth 10,299-303. CiK.ciO N, 1).. a n d A c tiiiso N . L. 1>. (1 9 8 2 ). D o p h e n o x y CiAU.AOMiK, R. P., a n d T iik i 1 1 At l . W . J. (1984). C an- herbicides cause cancer in m an? E aneet I, 1057-1059. O'iXiCiON, D-, P a NNI-'IT, B., WlNTliK, P. D ., A n i f s o n , L. D ., a n d bONSAi.t., J. (1 9 8 6 ). M o rta lity o f w o rk e rs exposed to 2-m cihyi-4-chiorophcnoxyucciic acid. S i uruJ. J. W o rk E n v iro n . H e a lth 1 2 ,4 4 8 -4 5 4 . Co o k , R. R. (1981). D ioxin, ch lo racn e, a n d soft-tissue sarcom a, l.uncct 1,618-619. C o o k , R. R ., Bo n d , G . ( i . , ( )i s o n , R . A ., O n . M . G ., a n d C o n d c k , M . R. (1986). (Evaluation o f the m ortal ity experience o f w orkers exposed to the chlorinated dioxins. C h em o sp h ere 15, 1969-1976. Co o k , R. R., T o w n s in d , J. ( . . O n , <., a n d S itv i R .sn iN, L. Ci. (1 9 8 0 ). M o rta lity ex |> e n e n ce o f e m ployees exposed to 2,3.7.8-lclruchlorodihcn/o-/M Jioxin ( I ( l)D ). J. O ccup. M e d 22, 530-532. C O T liK , S. J , AND YOIJNO, J. 1... JK. (L d s.) (1 9 7 5 ). T h ird N a tio n a l C ancer Survey: Incidence D ata. N a tional C ancer Institute M onograph 41, D IIL W Publi cation (N IH ) 75-787. IJ.S. D epartm ent o f H ealth, Ld- c c ra n d occupational exposure to chlorophenols. I a n cet 2,48. CiAl.l A C IIIK , R. P ,, T tlK I.il At I , w . J .. J i t I Ril.s, I'., Ba n d , P. R ., S p in i i .i i, J .. a n d C o i d m a n . A. J. (1984). C ancer an d aplastic anem ia in British C olum bia farm ers. J. N a tl. C'an cer In st 72, 1311-1315. Cill t:S, G .G ., l.U KISS, J. N .. BAtKIt . M . J ., I.OWI N illA i . R. M .. a n d Pan ton, J. (1984). M yeloproliferative an d lym phoprolilerative disorders in . T asm ania, 1972-80: O ccupational and fam ilial aspects. J N a tl C ancer Inst. 72, 1233-1240. G o iin . R.. a n d K d i i i r , I (1988). V ietn am M ilitary Service an d risk o f selected cancers. A bstract o f M eet ing o f the Society for Lpidcm iologic Research. . in ter J. E p id em io l. 128,901. G h i i .n , L. M . (1 9 8 6 ). M o r t a l i t v . i n a l y u x o f O n t a r i o H y d ro F o restry tr a d e s m e n C o h o rt IV 5 0 -IV X 2 . H ealth Services D epartm ent, H ealth and Safely Division, O n tario Hydro. u c u lio n , a n d W elfare, P u b lic I le a llh S erv ice, N a tio n a l CiKI 1 NWAI D, P ., K o VASZNAV. B .C 'lit I INS. 13. N ., AND Institutesol H ealth. T ill KKiAl ft l , Ci. (1 9 8 4 ). S a rc o m a s o f so ft-tissu e s a lte r D a IDI KO I\ I.. M ., ANI) Z lllI N K A U I , 1). (198.3). |)i- V ietnam service. J. N a tl. C a n c e r In st. 73, 11 0 7 -1 109. oxin exposure: 20 year follow -up. L a n c e t 2, 1134- H a Rd m i , L. (1977). P hcnoxyacetic acids an d can cer-- 1135. A clinical ot>scrvaiion (Letter). I.a k a r lid n in y c n 74, Di iV.l t I.. I:.. AND CiKtit t I.KMAN, S. (1 9 8 5 ). M o rta lity 2753-2754. am ong w hite an d non-w hite farm ers in N orth C aro H a r d i 1 1 , I . (1981). Relation of soft-tissue sarcom a, lina. 1976-1978. A n te r.J . E p id e m io l 12 1 ,3 9 1 -4 0 2 . m alignant lym phom a am i colon cancel to phenoxy D i n m o w , R .. P a ij i.s o n , J . ( )., a n d W in d i a n , R . W. acids, chlorophenols a nd o ther agents. S c a n d J H o ik (1988) . P arm ing and m alignant lym phom a in H an E nviron. H ea lth 7, 119-1)0. cock C ounty, O hio. lin t. J. h id M ed. 45, 25-28. * H a r d i 1 1 , I... a n d L k ik sso n , M . (1988). T h e associa (ERIKSSON, M ., H a KDI I.I., I.., UlK<i, N. ()., M OI l.t:K, I'., tion betw een soft tissue sarcom as and exposure to phe- a n d A x to son, O . (1981). Soil tissue sarcom as an d e x noxyucclic aciils, A new case-referrent study. < in n e r posure to chem ical substances: A case-relcrcnl study. 62, 652-656. Itrii. J. hut. M e d 38, 27-33. liv t K t ii 'i, Ci., Bt Am , A ., ( 'a n i o k , K ., G ih s o n , R ., a n d Van Lit.k , S. (1985). (Environm ental chem ical expo H a r d i i i , 1... I k ik s s o n . M .. 11 n n i r . IV. a n d I.ii n d o k i N, L. (1 9 8 1 ). M a lig n a n t ly m p h o m a a n d e.\()ositre to chem icals. csjKvially organic solvents, chlo ro p h e sures as risk factors for leukem ia a n d n o n llo d g k in 's nols aiul phenoxy acids: A case-co n tro l study, lir a J lym phom a. A bstract o f M eeting o f the Society for Lpi- ( a n c e r'4 3 , 1 6 9 - 176. dcm iologic Research. A nier. J E tn d e m io l 122, 535536. 1'a.s a i., L., J a c k s o n , L. W ., a n d K i a u iu k , M . R. H a r d i i I , I... AND.Sa n d s I H U M , A. ( 1979). C ase-co n tro l stu d y : S o il-tissu e s a rc o m a a n d ex|>o.sure to p h e n o x y acetic acids o r chlorophenols. lin t J ( 'ancer 39, 711- (1968). Leukem ia and lym phom a m ortality and farm 717. residence. A nier. J E p id e m io l 87, 267-274. I'ii'iT ', M . J., A d i-n a , M . A , C o iu iin , D. M .. a n d D u n n , H a /lc to n (.alxirutorics (1984). t h e V h a rm a co kin etu E valuation o f `- 2 . 4 - D u h l o r o p h e n o w a c e t u a e td M. ( 1987a). M ortality am o n g A ustralian conscripts o f {2 .4-D ) m th e M o u se, f inal R eport, Project No. 2184- th e V ie tn a m c n n llic l era. I. I X a th fro m all causes. 104. Industry Tusk fo rc e on 2.4-D, llu/.lcton futroru- A n te r.J E p id em io l 125,869-877. lories A m erica, Inc., V ienna, VA. 232 ERIC S. JOHNSON H aileto n la b o ra to rie s (1986). 2,4 -D u h lo ro p h en o .xya ceh c A c td i 'inut R eport. ( o m b in e d T o x icity a n d O ncogem a t y S tu d y in R a is. Voi. I. Project 2184-103. Indus try (ask I'orce, H u /leto n L.aboralorics A m erica, Inc., V ienna. VA. llOAK, S. K ., BtAIR. A., H o i MI S, I'. I-., BOYSliN, C . I).. RojJI I , R. J . I l o o v i K, R ., AND i-KAUMILNI, J. J , J k . (1986). A gricultural herbicide use an d risk ol' lym p h o m a a n d s o il-tissu e s a rc o m a . J . A m e r . M e d . /I.v.wk\ 256, 1141-1147. I I o a k Z a iim , S., Bi a ir , A ., H o i m i s , E. E., Bo y s iin . C. D . a n d R oitl:l . R. J. ( 1988a). A case-control study o |'s o il-tiss u e s a rc o m a a n d U o d g k in 's d is c a s e . E a rm in g and insecticide use. S ta n d . J. W o rk E nviron. H ealth. 1 4 ,2 2 4 -2 3 0 . Ho a r Z a iim , S.. W i im n k u r g i r , I). I)., Ra b iin t . IV A . S a a i . R. C .. C a n i o r , K . IV, a n d Bi a ir , A. (1988b). A case-control sludy o f non -llo d g k in 's lym phom a an d agricultural factors in Eastern Nebraska. A bstract o f Society for Epidem iologic Research. A n w r. J. E p id em io l. 128,901. N o g s o o r .C ., a n d W j s i i ki d n d , B. (1980). M ortality o f forest w orkers exposed to phenoxy acid prepara tions. I.u k a rtid n in g e n 77, 1828-1831. l io i M l s , A. P. (1986). Il cv/ J ir g m ia \ ic ln a m -E r a IV/cra n s M o rta lity S iu d v. W est Virginia H ealth D epart m ent, C harleston. IlD N t IIA K , ]V A ., AND H a | 1*1 K IN , W . I:. (1 9 8 1 ). 2,4,5I . tnclilorophenol and soft-tissue sarcom a. I.aneel I, 268-269. I lo p i., W ,, R ijssi i i , W ., a n d W i i s s . S. (1 9 8 4 ). P o r phyria cutanea tarda and sarcom a in a w orkers ex posed to 2,3,7.8-leiruchlorodiix:n/.odioxin-- M issouri. J A m e r . M e d .l.vwx' 25 1 , 1 5 3 4 -1 5 3 5 . International Agency for Research on C ancer (1977). I A R C M onographs on th e .valuation o f th e C arcino g e n ic R is k o f C h e m ic a ls lo U n m a n s Vol. 15. .Some E u m ig a n ls, d ie H e rb ic id e s 2.4-1) a n d 2 .4 .5 - T. ( 'hlorin a ied D ih cn zo d io sm s a n d M iscellaneous Industrial C h em ica ls International Agency for Research on C ancer, l.yon. International Agency for Research on C ancer (1983). A R C ' M ono g ra p h s on th e E va lu a tio n o f d ie ( areinog e n u R isk o l ( h c m ic u is to H u m a n s Vol. 30. M isc e lla neous Pesticides. International Agency for Research on C ancer, l.yon. In n i s , J. R . M .. I J i i a n d . B. M .. V a i i k io , M . ( i . , P i;iK in n 1 1 , 1... 1 is im i in . I , I I ak i . I R ., P a i i o n a , A. J., Ba d s, R. R., I ai k , II. I... C ar t , J. J., Ki i in . M ., M i u in i . i , I., a n d P i-:i i k .h, J. (1 9 6 9 ). B ioassay o f pesticides and industrial chem icals for tum origenicity in m ice: A p r e lim in a ry Hole. J N a ll. C 'a m i r I n s t. 4 2 , 1 1 0 1 - 1 1 14. Jo h n s o n , li. S. (1986). PM R an d relative risk (l etter). lin t J h id M i d 43, 214-216. Jo h n s o n . I'. li,, K o g i i k , M . A., a n d Br o w n . S. M . (19811. Soil-tissue sarcom as and chlorinated phenols. I.a n t'd 2, 40. K a n g , H ., Bk lsj.jn , JV. E u :. Y ., a n d Sij i:i*a r d , B. (1986). Soft-tissue sarcom a and m ilitary service in V ietnam : A case com parison groupanalysis o f hospital patien ts. / ( h r u p . M M 28, 1 2 15 -1218. K a n g , I I. , I-n / ig i r . E., Bk i si in , IV, E i u , M .. I n : , Y ,, a n d Siiiil'AK D, B. (1 9 8 7 ). S o ft tissu e s a re c o m a a n d m ilitary service in V ietnam : A case-control study. J. N atl. ( a m e r Inst. 79, 693-699. K o ciha, R. (1984). Evaluation o f the carcinogenic and m utagenic potential o f 2,3,7,8-T C D D and other chlo rinated dioxins. In B io lo g ica l M ec h a n ism s o f D io xin A c tio n . (A . P o la n d a n d R . L>. K im b ro u g h , lid s.), B a n b u ry R e p o rt 18. C o ld S p rin g I la rb o r l a b o ra to ry , C old Spring H arbor, NY. K<x ;a N, M ., a n d C l a r p , R. (J9 8 5 ). M o rta lity a m o n g V ietnam veterans in M assachusetts, 1972-1983. Soci ety for Epidem iologic Research A bstracts. A m er. J. E p id e m io l 122, 523. K o g a n , M ., a n d C i .a i'p, R . ( 1988). S o ft tissu e sa rc o m a m ortality am ong V ietnam veterans in M assachusetts, 1972 to 1983. Int. J. E p id e m io l. 17, .39-43. l.A IIIROI*, ( i. !>., W o t II., W . I I., AND A l HANISI, R. A. ( 1983). [P roject R a n c h H a n d II) A n E p id e m io lo g ic I n v e stig a tio n !/ H e a lth E ffe c ts in A ir E o ree P e r s o n n e l fo l low ing E xposure to H erbicides. B aseline M o rta lity S lu d y R esu lts. Prepared for The Surgeon C en erai, U.S. A ir Eoree, W ashington, IX '. Epidem iology D ivi sion. D ata Sciences D ivision, USAI-' School o f Aero.space M edicine. B rooks A ir E oiee Base. i'X . l . A MIRO!*. Ci. D ., W o i I I . W . I I.. AND Al HANISI', R. A. ( 1984). A ir E tn e e H e a lth S lu d y . (P roject R a n c h H a n d H). A n E p idem iologic In vestigation of H ea h li Eljccis m . Iir E o re e P e r s o n n e l fo llo w in g E x p o s u r e t o H e r b i cides. B aseline M o rb id ity S tu d y R esults. Prepared tor T h e Surgeon C e n e rai, U.S. A ir I'orce, W ashington, IX '. USAE School o f Aerospace M edicine, Aerospace M edical Division (AESC), B rooks A ir Eoree Base, T X . I .AW|(| N th , C. I-.. R i ll IV , A. A., Q u k k i n i o n , IV. C i u i n w a i d . IV. Pa g i , W . E,, a n d K u n i z , A. J. (1985). M ortality patterns o f New York V ietnam vet erans. A m er. J Pub. H ea lth 75, 277-279. El UNI R I VON C ., AND S/ADKOWSKl, D. (1 9 8 5 ). Zui llum unkun/L T ogcnilal von 2.3,7,8-T C D D (-1 In fa llv e rsic h e ru n g s re th ilic h e B eu rleilu n g -). A r b e its m e d . S o z ia l m e d . P r e v e n ttv m c d . ME 2 2 5 - 2 3 2 . I y n g i , I . (1985). A follow -up o f cancer incidence am ong w orkers in m anufacture o f phenoxy herbicides in D e n m a rk . U rti .J C a n c e r 5 2 , 2 5 9 -2 7 0 , l.YNGI . E.. SlO K M , i l . H ., AND Jl.NSI N, (). M . (1987). E valuation o f trends in soft-tissue sarcom a according to diagnostic criteria in consum ption o f phenoxy her bicides. C a n c e r 6 , 1 8 9 6 - 1901. M ai M a h o n , B., M o n so n , K. R ., W a n g , II. IE , a n d / h i n g . T. ( 1988). A second follow -up o f m ortality in a cohort o f pesticide applicators. J. O e a tp M ed. 30, 429-432. M a h s n i K, J. S-, a n d Ba h n . A. K . (E d s.) ( 1974). lp id e - SOM TISSIJI: SARCOMA, m a i .ic n an t l y m p h o m a 233 m to lo yy. A n ln ln n lu c to r y L ew . Saunders, Philadel phia/! .ondon/T oronto. M il.H a m , S. (1971). L eukem ia a n d m ultiple m yelom a in farm ers. A m er. J. H pidem ial. 94, 307-310. M il.Ha m , S. (1982). 1lerbicidcs, occu p atio n , a n d cancer. T u n e d 1, 1 4 6 4-1465. M o r g a n , O. P., L in , L. I.. a n d Sa ik a i y , II. H. (1980). M orbidity and m ortality in workers occupationally ex posed to pesticides. A rch. T.nvtron. ( 'ontam . Toxicol. 9, 349-382. M o s i:s , M .. a n d Si i ik o i i , 1. J. (1 9 8 1 ). S o ft-tissu e s a rc o m as, phenoxy herbicides and chlorinated phenols. L a n c e t I, 1370. N ational C ancer Institute C arcinogenesis T echnical R e port (1979). S e rie s No. 155 H io u ssu yo j 2,4.6-T richloro p h e n o lfo r P a ssib le ( 'o rcin o yen icitv. ('A S N o. 88-062, N C I-C G -T R -155. U.S. IX p artm cn t o f H ealth, lulucation, and W elfare, Public H ealth Service, N ational Institutes o f H ealth. N ational Toxicology Program (1982). T ech n ica l R eport Series Series N o 201. C areitutyenesis Itio u ssu y o f 2 .2 .7 .ti - 'L e tr a e h U tr o d ib e n z o - p - d io x ia ( ( 'A S N o . 1 7 4 6 0 1-6} in S w is s -W e b s te r m ic e ( h e r n ia l S tu d y ). U.S. D e p artm en t o f H ealth and H u m an Services, Public I leallh Service, N ational institutes o f H ealth. N at io n a I fo x icology P ro g ra m ( 1 9 8 8 a). / et h n ii a ! R e p o r t o n th e T o x ico h tfty a m i ( a rei/u t^en esis S tu d ie s o f 2.4Ih ch ln ro p h a n tl (( A S N O . I20-S.12) in T .I44/N R uts a n d H ot .IT , M ice N I P I R 353, N il! Publication 882808. U.S. D epartm ent o f H ealth and H u m an Ser vices. Public H ealth Service. N ational Institutes o f H ealth. N ational Toxicology Program ! 1988b). T echnical R eport on th e T oxicology a n d ( areitutyenesix S tu d ies o f Pentachlo ro p h en o l (C A S N O . H7-R6-5) in H 6 C JL , M ice. N T P T R 349, N IH Publication 88-2804. U.S. D epart m ent o f H ealth and H u m an Services, Public H ealth Service, N ational Institutes o f H ealth. O i s s o n , II . a n d B k a n i>i , L. (1 9 8 1 ). N o n - H o d g k in 's lym phom a o f the skin and occupational exposure to herbicides. L a n cet 2, 579. O i l , (., l l o i Dl-K, B. B., A N D O l s o n , R . D . (1 9 8 0 ). A m ortality analysis o f em ployees engaged in the m an u facture o f 2.4,5-lrichlorophcnoxyucctic acid. J. (h e a p . M ed 22,47-50. O n . M . G ., O t s o n , R . A .. O k >k , R. K ., a n d Bo n d , ( i.G . (1987). C ohort m ortality study o f chem ical workers with potential ex|)osuie to the higher chlori nated dioxins. J. O ccup. M e d 2 9 ,422-429. P a i o k s o n , D . CL, J r ., l-iN o i-K tiin , M . A ., R o b i k i s , D . W ., Nlil D tlA M , L. I... S w i.t NI:Y, M . IL , M a k i o w , D . A ., A n d k i.w s , J. S ,, J r ., a n d H a i p i r i n . W . L. (1989). Levels ol (tolychlorinuicd dilx*n/.o-/M lioxins u n d d ih en /o fu ra n sin w orkers e x cised to 2,3.7,8-letrach lorodibcn/.o-/i-dioxin. A m e r. J. h id . M e d . 16, 1.35146. P a /.DI KOVA-Vl J l U P K O V a , J .. N l MtOVA, M ., Plt'KOVA, J.. AND JiRASt.K, L. (1981). T h e d e v e lo p m e n t and prognosis o f chronic intoxication by tetrachlorod ihen/o-/M lioxin in m en. A rch. L m ir o n H e a lth i t , 5II. Pi ARCI., N . I\., S ill PPAKD, R. A ., Sm IIII, A. II.. AND I I.AOlii:, C . A. (1987). N on-I lo d g k m 's ly m p h o m a and farm ing: An expanded case-control study. Int. J C ancer.39, 155-161. P i a k o :. N . L . . Sm i i i i , A . 11., a n d I i s m k . ! ) . ( ) . ( 1985). M alignant lym phom a and m ultiple m yelom a linked w ith agricultural occupation in a New Z ealand cancer registry-based study. A a u r . J. L '/n d en a o / 121, 2252.36. IY a k c i;, N . 1:., S m i i i i , A. I I . . H o w a r d , J. K .. S im * PARI), R . A ., ( il l LS. II. A ., AND I I AOOI . ( '. A. ( 1986). N o n -I lo d g k in 's ly m p h o m a a n d c \|> o su rc U> p h en o x y herbicides, chlorophenols, fencing work, and meal works em ploym ent: A case-control study, itrit J hut. M e d 4.3,75-83. P ltO I, II. ('., (iOlDSW ORIHY. I.. C a M I'IH II, II. A., a n d Po l a n d , A. (1 9 8 0 ). Q u a n tita tiv e e v a lu a tio n o f the prom otion by 2,3,7,8-lclrachlorodihcn/:o-/j-din\in o f hepatocarcinogcncsis from diclhylniirosam inc. ( a n cer R es. 40, .3616-3620. P h il s II r , W . A ., a n d M a s o n , I . J .( 1974). H u m a n c a n cer m ortality in relation to poultry population, by c o u n ty , in 10 s o u th e a ste rn stales. J. N a tl C a n c e r In st. 5 .3 ,4 5 -4 9 . Pm n io n i, R .. M i Rt o , 1 .. 1 INI, A .. M l a / / a , I.., a n d Sa n II, L. (1986). S o il-tissu esa rc o m a s in Scvcso. L a n c e t 2, 52.5. R io o a n . W . B., V a n B k ik io i n , J .. A t o u a v i i i a . J. I-.. Bi.a o iiil r , J .. a n d M a s o n , T . J . (198.3). ( ..V ( a n c e r M a r ta h ty R a te s a n d 'Trends. /9 5 0 -/9 7 9 . Vols. i-iv. N ational C ancer Institute, linvirom nental Protection Agency. Riiiiim a k i. V., As p , S., a n d H i rnmi r< S. (1982). M or tality o f 2,4-dichlorophcnoxyacclic acid an d 2,4,5trichlorophcnoxyacclic acid herhiciile applicators in Liniand. Sea n d . J W o rk T nviron. Ilc td fh 8, 37-42. R iiiiim a k i, V.. Asp, S., Po k k a i a . 1... a n d H i rnmi r o . S. (1983). M ortality and cancer nuirhidity am ong chlorinated phcnoxyacid applicators in Liniand. ( hem o sp h e re 12, 779-784. Royal C om m ission on the Use an d litfects o f C hem ical Agents on A ustralian personnel in V ietnam . (1985). T in a I R ep o rt. Vol. 4. C a n cer. P arliam en tary Pajx'r 291/1985. Saj i i as, A . I '., Bi a ir . A ., C a n I o k . K. P ., H a n k a iia n . I... a n d A n d i kso n , H. A. (1987). ( an cer a n d o lh er causes o f death am ong W isconsin farm ers. A m e r J. Ind. M ed. I I , 119-129. S a k m a , P. R ., a n d J ac oils, J. (1 9 8 2 ). 1 h o rac ic sol'l-itssue sarcom a in V ietnam veterans ex cised to Agent O r ange. N . I .n y l J. M e d . .306, 1 109. S nniM A niL K , M . ('. (1985). Panning occupations and m ortality from non-H odgkin's lym phom a in U tah. A case-control study. J O ccup. M ed . 27, 580-584. 234 !:KK' S. JOHNSON S m iiM A t ill R. M ( \ , AND O i l / I I I , I-!. (1 9 8 8 ). A d e a th certificate case-co n tro l study o f n o n -H odgkin's lym phom a and occupation in m en in N orth Carolina. . i n n r . J . n il. M e d . 1.1, 3 1 7 -3 3 0 . S m w t i / , B. A .. Q u a s i , 1 . 1-.. Ki:i:t i :r , P. A .. H u m is io n .C . Cl , A N D K<h jh a . K. J. (1978). R esults o f iwoycar toxicity a nd reproduction studies o n pcntachlorophenol in rals. In P cn la ch lo ro p h en o l: C h e m istry . Pharn n u nlogy. a nt! E n viro n m en ta l i'o xicn ln g y(K . R. Rao, I'd.), pp 301-309. P lenum , New Y ork. SMIMI, A. 11.. IIS IIIK , ( f i l l s . H . J.. A N D Pl A R O , N. (1983). T he New Zealand soil-tissue sarcom a casecontrol study: Interview findings concerning phenoxyacelicacid exposure. C h em o sp h ere 12, 565-571. S m IIII, A. It , I lSIll K. D O . , Pl AKO . N ., AND r i AUtii:, ( . A. (1982). D o agricultural chem icals cause sofi-lissue aicorna? Initial lindings o f a c ase-co n tro l slu d y in N ew Z e a la n d . ( o n i n i u n H e a lth S t u d . 6 , 114 --119. S m i i i i , A. I I., a n d Pi a k < i:, N . P, (1 9 8 6 ). U p d a te o n soft-tissue sarcom a aiul phenoxyherhicides in New Zealand. ( ite m o sp h e re 15, 1795-1798. S m i i i i . a . I I.. P i a r o ., N. I-., J i.m h .r , O. O ., G h i s . H. J.. ItA O u r, C A., a n d IIo w a k d , J. K. (1984). Soft-tissue sarcom a an d cxjHisurc to phenoxyherhi cides and chlorophenols in New Z ealand. / N a t! C antet /m l 72, 1III-1II7. S u m i . W ., Bo n d , C . Ci,. S k o w k o n s k i, B. J .. Br o w n so n , P. J., a n d C o o k , R. R. (1986). S oil-tissue sar com a: C ase-control sludy. J. O ccu/r M e d 28, 804808. S i a k k . A.. I lo ll-. M . A N D S i o n i , R. (1983). C ancer incidence an d dealh from all causes in a cohort o f larm ers. Abstract o f M eeting o fih e Society for fp id cm iologic R esearch, lin e r J E p id e m io l. 118,431. I l l l i s s , A. M .. I HI N l / I l-B l VMI . R ., A N D I.INK. R. (1982) M ortality sludy o f |>eisons ex|wised lo dioxin in a irichlorophenol-process accideni itiai occurrcil in th e HASP A ( i o n N ovem l>er 17. 1953. .1m e t J b id . Med 3, 179-189. VlNl is. P.. I'l K R A i IN I, B.. C lU ONI . (.. ( 1( iNI.'l 11, A., C o i o M iio , li.. D o n n a , A,. M a i-i-i. I ... P is a . R .. R h < I. P.. Z a n in i, li., a n o C'o m iia , P. (1 9 8 7 ). P h e n o x y h e rb i cides an d soft-tissue sarcom as in fem ale rice-w ceders: A population-based case-control study. S ta n d . J IliwA -.n u ro n Ih -n ll/i 1 3 ,9 -1 7 . W a n o . II II.. a n d M a( 'M a h o n , B. (1979). M o rtality o f ixrslicide applicators. J. ( h e a p . M ed . 21, 7 4 1 -744. W a i i R i i u u s i , J. ( '., M u i r , C ., S i i a n m u o a r a i a n a m . K.. a n d P o w t 1 1 , J. (la ls.H 1982). ( tim e r I n c id e n c e in / i r e i 'ontm ents. Vol. IV. JA l(( 'S cicnlilic P ublications No. 42. International Agency for Research on Cancer, I yon. W i isi n u u r o i r . D. I ) , A r m i | At i| , J. ( ) ,, a n d P u r l u o . D. I (1986). I nM ionm enlal epidem iology of n o n -H odgkin's lym phom a (N H 1.) in Pastern N e braska (A bstract). I'ed. /Vor. 45,6 4 0 . W ik i u n d , K .. D i m , J., a n d H o i m , 1.. (1 9 8 7 ). R isk o f m alignant lym phom a in Swedish pesticide appliers. Jlril. J. ( 'inn e r 56, 5 0 5 -508. W ik i u n d , K ., D i m , J., a n d H o i .m , 1.. (1 9 8 8 ). S o ft-tis sue sarcom a risk in Sw edish licensed pesticide applica to rs. J . O c c u p . M e d . 3 0 , 8 0 1- 8 0 4 . W ik i u n d , K ., L in i :i o r s , B .. a n d H o i .m , l.. (1 9 8 8 ). Risk o f m alignant lym phom a in Sw edish agricultural and forestry workers, itrii. J. b u t. M ed. 45, 19-24. W lll.lA M S, K. K., S l I .iil NS. N. 1.., AND CiOI DSMIIII, J. R. (1977). A ssociations o f cancer site an d type with occupation and industry from the Third N ational C ancer Survey Interview. J. N u ll ( tim e r Inst. 59, 1147-1185. W o i .i i ., W . | | ,, a n d M k 'I ia i.i k , J. li. (1 9 8 5 ). P ro je c t R aneh H a n d U . A n E pidem iologic Investigation oj l/e a llh E ffects in A ir fo rc e P ersonnel follow ing E xp o su re lo H erbicides. M o riu lily U pdate-- / 985. Prepared fo r th e S u rg e o n G e n e ra l, LJ.S. A ir T o rce, W a sh in g to n . IX '. iipidcm iology D ivision, D ata Sciences Division, USAT School o f A erospace M edicine. B rooks Air fo rc e Base. T X . W o r n : , W . I I., M io iA i.h K , J. K., A i .b a n i-s i:, R. A., ei al. (1984). Project R a n eh H a n d H. A n E p id em io lo g ic Investigation o f H ealth E ffects in A ir fo rc e P ersonnel follow ing E x/ntsurc to H erbicides. M o rta lity U pdate- I9,S4. P repared for ihe Surgeon G eneral, U.S. Air fo rce, W ashington, IX '. Iipidcm iology D ivision. D ata Sciences D ivision, U S A f School o f A erospace M edi cine. Brooks Air fo rc e Base, TX. Wo n < x n i i I ) w o i i n / i:, I*. (1 9 8 2 ). M e th o d o lo g ic a l issues involving Ihe standardized m ortality ratio and p roportionate m oriulily ralio in occupational studies. J (h ea p . M ed. 24, 299-304. W l)N (i. ( )., M o RI.AN, R. W ., KIII II I KS. I... AND I.ARso n , S. R. (1985). C om p ariso n o f SM R , PM R , and PC 'M k in a co hort o f u n io n m em bers p otentially ex- l^osed to diesel exhaust em issions. B rit. J. In ti M ed . 42, 449-460. W o o d s , J. S.. P o t issa r, I... Sj vi r so n . R. K ., lit usi r , I . S., a n d K m a n d i R, B. ( i . (1 9 8 7 ). S o ft-tissu e s a r com a and n o n -H odgkin's lym phom a in relation to phenoxy herbicide and chlorinated phenol exposure in W estern W ashington. J. N atl. C ancer Inst. 78, 899910. Z a c k , J. A., a n d G ai i i.y, W . R. (1983). A m ortality sludy o f w orkers em ployed at the M onsanto com pany plant in N itro, West Virginia. E n viro n . S et. R es. 26, 575-591. /A t k , J. A., a n d Su sk in d . R. R. (1980). T he m ortality experience o f workers exposed lo leirachlnrodibcn/odioxin in a irichiorophcnol process accident. J. O ccup M ed 22, 11-14. CO CO '>.J oq H IN O A M I N I A l A N I I A I T I 11 l> l< >\K O l (Xi Y 14, 2.VS 242 (!*)*>()) 1 Assessing the Efficacy of Azaprophen and Physostigmine as a Pretreatment for Soman-Induced Incapacitation in Guinea Pigs by Response-Surface Modeling' CiiKisCii N N i N i is ,* Wai iik II. Ca r h :ii, Jr.,* I.akkii. W. Harris,! Richard A . Cakciiman,* lit i anor D . Cami*ih:i i Riissi i i M . B o v t i..* Brian (J . "I ai.iiot,! and K k hard I*. S o i ana! .S'. . \ n n v M i 'i l t a i l l i c w i i n fi h t s i l i t t l e o f < h c ii i u til I t c f a i w . . U u 'i th v n I ' n n i n y U in tu u l. M u n h i / u l . in n ! * M i'J u ttl ( ollcyt' o J l'iii;tn iii/\ ttiiin u i C'o n tm o n w cu h h I 'm v a stly . H ultniotul. I u yin u i K a v t v c t U i t t i U i i n J 7. / `AS'V. a a v p f a J iiifiu s l 25. /f/.vy Assessing the I '.llicao o f A / a p ro p lu n and INtysosligm ini' as a Prchvalm cm I'm S o m a n -In duced Incapacitation in ( iuinea 1`igs l>> Kes|X>nse-Suiiace M odeling. ( ii n n i m ,s. ( .. C a k 11 k . W. H.. J r .. I I a k k i s , |,, W.. ( a k < i i m a n , K. A.. C a m im u i i . !. I).. H i m i . K. M.. I \i h o i . 11. ( ..a n d So a n a . K. I*. ( l `>lMI). I ittti/u in .I/>/>/. h>\m >l. 14, 245-242. iMiysosiigmine (IN IY ) has the advantage over pyridostigm ine o f m in im izin g O P-im luccd incapacitalion Ivcausc it l>cnctratcs into the ( N S. I iowever. physostigm ine is behavioral!) toxic at relali\cl\ low concen trations it is anticipated that this could lx* olixcl by a cholinolylic to prevent lx`havmral delicti due to the carbamate pret real mem alone. I he thcrujKuiiceliicacy ofphy sosligminc/azaprophen pieireatmenl therapy w ascvalualed in soman-challenged guinea pigs. Kesjxm xcsurlace method ology was em ployed to desert lx* the relationship o f the preheat m cnl com bination with duration o f incapacitation. 1he significance o f (he com bination relative (o IN IS ' alone was evaluated in addition to dose com biu al ions that yiehl opt until lim e to recovery. A nalysis o f the lilted i espouse surlace indicated that com bination piclrcalm eni with these com|x>unds signilicanlly reduces the time to recovery tiller som an challenge versus pieireatmenl with IN I Y alone. iv x is -mcu Ol I..Xt.ol..g> P h y so stig m in e (P IIY ), an /V-m clhyl curham ulc an d an in h ib ito r o f acetylcholinesterase (A C hl:), in ad d itio n lo being evaluated as a p reirealnienl against nrganophosphatc expo sure, has been used in Ihe treatm en t o f P ar k in s o n 's d ise a se . S in e e P H Y is b e h a v io r a lly toxie a t relatively low eo n cen lralio n s (H arris cl al., 1984b), it is a n tic ip a te d Ibis c o u ld be ollset by a cholinolylic lo prevent behavioral deficit d u e to the carb am ate prelreatm en t alo n e (M ew aldl and (th o n eim , 1978; H arris cl ul.. 1984a). A /ap ro p b en (A ZA ; 6-mclhyl-6-azal>icyclo| 3 .2 .1)oelan-.)-ol-2.2-diphenylproprionale) is a r e c e n tly s y n th e s iz e d n o v e l c h o lin o ly lic . It has structural features o f both atropine and u p ro p h en (C arroll cl ul.. 1987), b o th o f w hich have been used successfully in co n ju n ctio n w ith carb am ate to antagonize intoxication by potent organophospborus anticholinesterase a g e n ts(I-leisher an d H arris, 1965; l.eadbcalcr d a l.. 1985). A p relrca lm cn l dosage com bination ol PH Y and cholinolylic w hich m inim izes the dosage o f PI IY could provide a prelreatm ent 1 I III!, w ork w.is M i|)|H .ik\l in p.n 1 hy ilic II.S. A m iy M c.lic;il Kl'v . i i i ll .hhI D r v d .ip iu c iil ( u u l 11 i.u u i. ( \ m IruLl N o. D A M D I7-K 6(-<>2<>2. w hich is w ith o u t a behav io ral d ecrem en tin g ell'eet. B ecause it p e n e tra te s in to th e ( N S. PH Y has Ihe advantage over pyridostigm ine o f m inim izing post-agent incapacitation 245 (I272-(I.`W()/`)I)$.1.1X1 (u|>> lig h t" l*M lti> the Society uf lm n.d..g> A ll lights >l n-jn<klui lnm in jo ) Ini in i i v i ->1 K' <LD CO i ' OD Public Health Rev 1989/90; 17; 1-37 Phenoxyacetic acid herbicides and chlorophenols and the etiology o f lymphoma and soft-tissue neoplasms Shona J. Kelly and Tee L. Guidotti ABSTRACT....................................................................................... 2 INTRODUCTION............................................................................. 2 Phenoxyacetic acid herbicides and chlorophenols .................... 4 Biological effects............................................................................. 7 Problems in study design............................................................. 10 SOFT TISSUE SA RC O M AS....................................................... 14 Manufacturing workers............................................................... 18 Agent orange................................................................................. 19 Other studies............................................. 20 LYM PHOM AS............................................................................... 21 DISCUSSION ................................................................................. 26 Recommendations....................................................................... 29 REFERENCES ............................................................................... 31 M rs. K elly w as fo rm e rly R e se a rc h A sso ciate an d D r. G u id o tti is P ro fe sso r o f O ccu p atio n al M edicine, O ccupational H ealth P ro g ram , U niversity of A lberta Faculty o f M edicine, 13-103 C linical Sciences Bldg., E d m o n to n , A lb erta T6G 2G 3, C anada. C orrespondence to D r. G uidotti. M rs . K e lly 's p re s e n t a d d r e s s is O c c u p a tio n a l M e d ic in e P r o g r a m , U n iv e rs ity o f B ritish C o lu m b ia, M ath er B uilding, 5804 Fairview C r., V ancouver, B.C., C anada. S uppo rted by the T rip artite Fund for O ccupational H ealth at the U niversity of A lberta and by gran ts from A lberta E nvironm ent, the A lberta O ccupational H ealth and Safety H eritage G rant P rogram , the Industrial V egetation M anagem ent A ssociation of A lberta, and the A lberta O ccupational H ealth and Safety H eritage G rant Program . 1990 Public H ealth R eview s P rim ed in Israel 2 Kelly S, Guidoni TL. Phenoxyaceiic acid herbicides ABSTRACT The phenoxyacetic acid herbicides and the chlorophenols are compounds of economic importance. The herbicide 2,4-D is widely used in agriculture, industry, and the home. Recently, concern has arisen over their safety as a result of studies linking these compounds with soft-tissue sarcomas and non-Hodgkin's lymphomas. We reviewed the available literature in order to advise a provincial government regulatory body and investigated mthodologie issues by examining the pattern of reported cases in Alberta. We conclude that, taking into account the serious limitations on methodology in the available data, the evidence for a causal association is strongest for non-Hodgkin's lymphomas and probably reflects either a weak effect or, possibly, a confounding exposure associated with the use of 2,4-D. Given the worst-case assumptions, however, the potency of 2,4-D as a carcinogen is probably weak. Its intrinsic toxicity is less than that of alternative herbicides and the hazard posed by its use is probably much less than either the use of chemical alternatives or manual cleaning of vegetation, which carry a high risk of occupational injuries. INTRODUCTION Government and the agricultural community have become increasingly concerned about the safety and health aspects of the use of herbicides. This concern followed the demonstration that farmers were among those at elevated risk for certain types of cancers. This observation stands in contrast to the general observation that farmers have a decreased overall mortality experience. Subsequently, one group of herbicides, the phenoxyacetic acids, has generated particular concern as a result of recent research. The carcinogenicity of phenoxyacetic acid herbicides and chlorophenols remains an unresolved issue, however, due in part to difficulties in the classification of the cancers of concern, problems intrinsic to the study of a rare disease, and an uncertain toxicological data base. These chemicals are widely used in agriculture and industry. This report reviews the evidence for carcino genicity of these compounds, as reflected in rates of soft-tissue sarcoma (STS) and non-Hodgkin's lymphoma (NHL) among occupationally-exposed users. Nosological problems with STS and OD CO o Public Health Reviews 1989/90; 17 o Kelly S, Guidoni TL. Phenoxyaceiic acid herbicides 3 limitations in study design are primary reasons why many studies have yielded inconclusive results. In this paper we review the studies on STS and NHL and offer recommendations for further study and public advisories. Table 1 lists the ICD-Oncology (1) classification numbers for the histologic types of greatest interest to this discussion. We reviewed the world literature on this topic to provide an authoritative basis for regulation for Alberta Environment, the regulatory agency responsible for pesticide use and distribution in this western Canadian province where consumption of the herbicide is relatively heavy. Table 1 Classification o f histologic subtypes o f soft-tissue sarcomas and lymphomas (1) 0000 oo1cc u> lC D No. S o ft Tissue Sarcom as 880 885-888 889-892 904 912-916 918-920 921-924 925 926 938-948 954-957 958 M alignant L ym phom as 959-63 964 956-966 969 970 * N ot otherw ise specified. D escription soft tissue tum ors and sarcom as, NOS* fibrom alous neoplam s lipom atous neoplasm s m yom atous neoplasm s synovial neoplasm s blood vessel tum ors osteom as, osteosarcom as chondrom atous neoplasm s giant cell tu m o rs m iscellaneous bone tum ors gliom as nerve sheath tum ors g ran u lar cell tu m o rs and alveolar soft p art sarcom a lym phom as, N O S* or diffuse reticulosarcom as H o d g k in 's d ise a se lym phom as, n o d u lar o r follicular m ycosis fungoides Public Health Reviews 1989/90: 17 4 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides Phenoxyacetic acid herbicides and chlorophenols Phenoxyacetic acids and the structurally-related chlorophenols are pesticides widely used for the control of weeds and fungi, and for wood preservation. (See Table 2 for names, abbreviations, and uses.) Commercial production of some of these pesticides may yield a number of contaminants, among them dioxins. Table 3 lists the known toxicity levels of these contaminants and the pesticide active ingredients, and are expressed by the LD5o. Toxicity in rats ranges over 6 orders of magnitude. Table 4 lists the pesticides in which dioxin contaminants are known to occur. Contamination of herbicide preparations by dioxins greatly complicates toxicological assessments because the outcomes of concern are similar. 2,3,7,8-tetrachloro-dibenzo-/-dioxin (TCDD), one of the most studied dioxins, has received the most attention because of its toxicity in animal studies. In the early 1970s detection limits for the analysis of TCDD were in the parts-per-million range (2). Since the level of contamination in most herbicide products varied considerably, early reports quantifying dioxin levels must be interpreted cautiously. Even so, gross contamination of up to 47 ppm (3) and 100 ppm (4) has been reported. The carcinogen nitrosodimethylamine was also found to be a contaminant in MCPA at levels of from 200-600 micrograms to 2-18 mg/kg (3). Contamination levels and product formulations vary among contries and even among parts of the same country because the products are "designed" for specific regional conditions. Production quality control and analytical methods for the detection of contamination have improved dramatically in the last decade. Current allowable levels of contamination by TCDD are 0.1 mg/ kg of the technical material, a standard instituted in Canada in 1975 and in the US in 1971 (2). A factor further complicating the study of this group is that they have often been formulated with other herbicides, including other members of their chemical group. In 1953, 64% of the products available in Canada containing 2,4,5-T also contained 2,4-D (5). Recollection by study subjects of products used in the past is confused by a plethora of similar-sounding names and by changes in product concentration using the same or similar names by either the same manufacturer or a competitor. 2,4,5-T is no longer manufactured and is not sold in Canada or the US. H-* CD Public Health Reviews 1989/90; 17 CO CO Kelly S, Guidoni TL. Phenoxyacetic acid herbicides O ceu 5 Xc5 Q 5 -O <u -C CS <3 *e*u &ueat O H C<u -Ci, X3 c <3 3 E<3 C s: o >eI*u. O -C) Xld 3u .1o2 eu !2 ue JS O3 e T<G30 %eu o (A co00 EUO tC->U JS H CU 04) > eI 'S <N .g oc t> O P O p d 3 *0 p. C uoC CX *2,2 oo !uu ou. JS JS Zo.? *5 G XS uu. ? JS O. JtJ> O ai ? S s .y "O 5o Mo s s o 'G<d u `o5 'aoeku. S x-5 >X. Xoc ,,o %CU. C^N jes 2 --XODc g-- -GCPSUx_xoc4eoto*)u o Xu .y C i ? w g 1 3 r r* rVs X3oeoOcl*u Xo c Vit rn </ X j4Cd> X G ^Bi CN rs a S 3 o3 .Ca s(>4> PU S0 re=u 1X g. A J>S. $S u GA ^ - 2o sO ?v j= 2V Oj o `<Gd u Goid ^ ^*oG x5gCU Xeu S 11 xh .y g v -2 V Ci <N XeOGOu xXuoGUoeuOu Xo 3 ci eVs o X VA E OE (J X -oz o j> 4s. X ofc- ou . adI- <ob Xso Xo ogc. jd X CS *3 T3 3 a, eu 4 y <+N PQ UX Xoeu cu C<u 2-O xy b V) yS 8E CU CS Xeouu .x2 9 JOS XO a < Pu H fc fl u #. 2 yS uft. M rh Hu Publie Health Reviews 1989/90; 17 6 Kelly S, Guidoni TL. Phenoxyaeelic acid herbicides Table 3 A n im a l toxicity o f the dioxin contaminants, phenoxyacetic acid herbicides, and chlorohenols (79,80) H om ologue nam e d i-d io x in s tri-d io x in s penta-dioxins hexa-dioxins hepta-dioxins octa-dioxin 2,4-D dicam ba d ic h lo rp ro p fenoprop M CPA m ecoprop p e n ta c h lo ro p h e n o l 2,4,5-T tetrachlorophenol tric h lo ro p h e n o l Isom er 2,7 2,8 2,3,7 te tra -d io x in s 2 ,3 ,7 ,8 1,2,3,4 1,2,3,7,8 1,2,4,7,8 1,2,3,4,7,8 1,2,3.6,7,8 1,2,3,7,8.9 1,2,3,4.6.7,8 acid sodium salt butyl ester dim ethylam ine salt iso octyl ester isobutyl ester butoxyethanol ester LDso (m g /k g body weight) > 2,000 300 30 0 .0 0 2 -5 > 1.000 0 .0 0 3 -0 .3 3 8 1 .1 -5 0 .0 7 -.8 0 .0 7 -1 .3 0 .0 6 -1 .4 > 0.6 > 1,000 699 997 695 949 896 618 850 2,900 800 1,000 700-800 930 . 50-140 500 470 1,000-3,000 to CO CjO Public Health Reviews 1989/90; 17 ro Kelly S, Guidoni TL. Phenoxyaceiic acid herbicides Table 4 D io x in contamination o f som e pesticides P e stic id e Known contam inants 2,4-D fenoprop h e x a c h lo ro p h e n e p e n ta c h lo ro p h e n o l 2.4,5-T te tra c h lo ro p h e n o l Inclusati di-, tri-, and tetrad io x in s (but not 2,3,7,8-T C D D ) 2,3,7,8-T C D D 2,3,7,8-T C D D hexa-, hepta-, octadioxins 2,3,7,8-T C D D hexa-, hepta-, octadioxins di-, tri-, tetradioxins 7 Occupations in which exposure to the phenoxyacetic acid herbicides and chlorophenols are likely to occur are listed in Table 5. It is clear from the diversity of these occupations that the pesticides have wide and important use in the economy. A recent federally-sponsored study of the effect of a withdrawal of 2,4-D on the Canadian economy suggests that the cost would fall between $130 and $200 million per year (Canadian dollars), mostly incurred by the agricultural sector. Much of the cost would result from substituting other methods of brush and weed control, including manual removal and more expensive alternative herbicides (6). 2,4-D is no longer protected by patent and is much less expensive than commercial alternatives which are so protected (7). 2,4-D has also been more extensively studied than alternative herbicides, many of which are known to be more acutely toxic (8). Biological effects of phenoxyacetic acid herbicides and chlorophenols A recent field experiment conducted at the University of Guelph is useful in separating out "dislodgeable residues"from total recoverable 2,4-D in grass (9). Within hours of application less than 4.5% of the Public Health Reviews 1989/90; 17 8 Kelly S, Guidoni TL. Phenoxyacetic acid herbicida 2,4-D applied could be dislodged. This declined to less than 0.02% over 7 days even in dry conditions that would maximize retention. We estimate from the data that there is a m axim um half-life of about 19 hours for the dislodgeable residues. Thus, in the absence of rain only about l%of the application would be "dislodgeable"after 2 days. This estimate would be modified downward for different (more dilute) commercial formulations, rainfall, or mowing. Pharmacokinetic data on the rate of absorption and the turnover of Table 5 Occupations associated with possible pesticide exposure to CO o-> agriculturalists new pesticides under development crop farmers and farm workers herbicides, insecticides, fungicides insecticides, litter from treated wood chips nursery and related workers herbicides, insecticides, fungicides farm machinery operators herbicides, insecticides, fungicides chemical processing operations all pesticides flour and grain milling seed treatments, fungicides sawmills fungicides wood treating occupations fungicides wood treating occupations wood preservatives laboring--wood fungicides pulp and paper fungicides textile and rope finishing and calenderingfungicides leather manufacturing and processing fungicides rubber processing, fabricating & repair fungicides, curing compounds petroleum industry slimicides railway section and track workers herbicides, wood preservatives electrical power: erect, install, repair herbicides other construction trades herbicides air pilots all pesticides air transport operating occupations all pesticides grain handling seed treatments, fungicides polyvinyl adhesives manufacture preservative application photographic solutions preservatives Public Health Reviews 1989/90; 17 Kelly S. Guidoni T L Phenoxyacetic acid herbicides 9 specific herbicides and their solvent carriers are also important in interpreting possible health effects. 2,4-D is poorly absorbed through the skin. The metabolic fate of a subcutaneously injected ester of 2,4-D has been studied in the rat (10). 95% of the administered dose was excreted in the urine as free acid within 48 hours. Studies on farmers going through their work routine in Saskatchewan using dermal and clothing pads have provided data on absorption and excretion (11). The amount of 2,4-D metabolite in urine was significantly correlated with the amount of herbicide sprayed. The time required for excretion of the urinary metabolite to return to background concentration was related to the amount sprayed and the number of exposures. Chromosomal aberrations, sister chromatid exchanges, and micro nuclei are morphological changes in cells recognized to be early indications of genotoxicity. Mutagenic studies in microbial and yeast systems have yielded conflicting results (3). Davring and Hultgren (12) demonstrated that an ester of 2,4,5-T acts as a mitotic poison or clastogen in bone marrow cells from male mice. A risk assessment conducted in 1981 by the US Environmental Protection Agency (13) concluded that there was "substantial toxicological and epidemiolo gical evidence that TCDD is a potent carcinogen". Shu et al. have concluded that the data overwhelmingly support a role for TCDD as a tumor promotor rather than as an initiator (14). There are conflicting reports of chromosomal changes in humans exposed to phenoxyacetic acid herbicides. Yoder et al. described increased frequency of chromatid lesions in lymphocytes from agri cultural workers during a season of work involving exposure to pesticides including 2,4-D and 2,4,5-T (15). Hogstedt et al., on the other hand, reported no such increase in a group of Swedish male workers exposed during the spraying of a number of pesticides, including 2,4-D, MCPA, and mecoprop(l). Crossen et al. observed a larger than normal number of sister chromatid exchanges in lymphocytes from workers exposed to 2,4-D and 2,4,5-T in New Zealand (17). Other than studies evaluating chromosomal changes in lympho cytes, experimental testing of the effects of phenoxyacetic acid herbicides on the immune system is limited. In screening a number of pesticides, Vos et al. reported that 2,4,5-T had a histologically visible effect on the tissues of the immune system (18). Blakley administered sublethal levels of the /i-butyl ester of 2,4-D dermally to female CD-I Public Health Reviews 1989/90; 17 IO Kelly S. Guidoni TL. Phenoxyacetic acid herbicides mice (19). There was minimal clinical or pathological alteration and no effect on antibody production but the treatment did enhance B-lymphocyte proliferative responses. Subacute oral doses of the same pesticide failed to demonstrate immune effects (20). A recent report by Hoffman et al. demonstrated an immunosuppressive effect of TCDD in humans which was consistent with evidence from animal studies showing T4 lymphocyte depletion and a decreased T4/T8 ratio (21). Thunberg (cited in Axelson, 3) has drawn attention to similarities between the effects of vitamin A deficiency and TCDD exposure, namely, epithelial lesions and infections, acne and keratosis, failure of normal growth, immunosuppression, reproductive abnormalities, teratogenesis, and lesions in the skeleton and nervous system. Problems in study design Epidemiological studies of pesticides share common methodological problems. Subjects are exposed to the substance at widely varying exposure levels and exposed intermittently and often concurrently with occupational exposures to other potentially toxic substances. Any risk assessment exercise must therefore interpret the available studies with caution. Six methodological problems are particularly significant in the study of phenoxyacetic acid herbicides and chlorophenols:1 1) A fundamental requirement of these studies is to document the nature and duration of exposure. None of the available studies has determined exposure levels accurately. Except in the rare case of relatively well-documented exposure of production workers, most studies rely on either a detailed recall of distant events (sometimes from a deceased subject's relatives) or obtaining death certificate analyses with implicit assumptions regarding the extent of exposure based on occupational title or employment records. Recall is obviously subject to bias. Some retrospective studies have obtained information about exposure by asking interview questions about events that happened more than 20 years before the time of the interview. Pesticide formulations and application techniques have changed with time. Application work tends to be seasonal and information about exposure can be extremely inaccurate because detailed records are rarely kept. Some investiga- Pubiic Health Reviews 1989/90; 17 Kelly -i'. Guidoni TL. Phenoxyacetic acid herbicides II tors have attempted to verify responses by checking supplier sales records. Even this approach cannot be regarded with confidence because, in general, only recent records can be confirmed. 2) The limitations of using death certificates as the sole source of data for cause of death or the decedent's occupation has been well documented (22, 23). The disease itself may be under- or overreported on death certificates (23) and the occupation coded on the form will be the last occupation or that of longest duration. This occupation is not necessarily that of most interest toxicologically. Mortality rates may differ widely among occupational groups for reasons of selection that are not related to the chemical exposure and these differences often persist well past retirement (24). A chemical exposure profile can only be inferred from death certificates. Studies that rely on death certificates alone may seriously underestimate risk because of the lack of pathological confirmation of the malignancy and omission of incident cases in subjects dying of other causes. However, available populationbased studies of mortality or disease frequency by occupation often omit essential data needed to derive refined risk estimates. They aggregate occupations so as to dilute the experience of exposed groups with that of unexposed groups (25). 3) There is considerable uncertainty over the latency period one should assume before the effects of interest appear. Different investigators have applied inconsistent criteria in selecting esti mates of latency in the analysis of their data. Although 20 years is commonly cited as a reasonable latency period for exposureinduced cancer, it may be too long for hematopoietic and lymphatic malignancies and is likely to be too short a time to detect any peak in excessive incidence of carcinomas. The minimum appropriate latency period for STS and other types of sarcoma is unknown. 4) The power of a study depends upon the numbr of subjects and the magnitude of the risk involved (26). A rough estimate of the proportion of the population exposed to phenoxyacetic acid herbicides and chlorophenols on a regular basis can be made by examining the proportion of farmers in the general population. This varies from almost 0% in urban populations to 11% in Alberta and 20% in some rural areas of the US (27). If a power of 80% is considered to be the acceptable minimum, either large Public Health Reviews 1989/90; 17 12 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides numbers of study subjects or significant health effects are required. Figures 1and 2 are power curves showing the relationship between number of cases and the assumptions regarding effects size. Figure 1 is based on the assumption that 10% of the population is regularly exposed; Fig. 2 is based on a corresponding figure of 20%. Groups of 50 or fewer subjects cannot reliably detect a risk ratio less than 3 by these criteria, even when there is a high ratio of controls to cases in a traditional case-control study model. 30 25 20 Power (%) Fig. I. Power functions: 20% of population exposed to the risk and a case-control ratio of 1:3. CD Public Health Reviews 989/90; 17 CD oo CJ\ Kelly S, Guidoni TL. Phenoxyacetic acid herbicides 13 Power Fig. 2. Power functions: 10% of population exposed to the risk and a case-control ratio of 1:3. 5) In a cohort study the treatment of subjects lost to follow-up can alter the study's findings (28). The common practice of assuming that those lost to follow-up are in fact alive at the end of the study will underestimate mortality. Many investigators prefer this conservative approach as one less likely to yield aspurious association. 6) The age distribution of subjects chosen for the study can radically affect the outcome. Young people are generally omitted from case-control studies because their exposure is assumed to have been low. This may be a false premise because they may have been exposed prenatally or in their home environment. Laundry, for example, has proven to be an extremely efficient method for the redistribution of pesticide residues (29). The public's perception of pesticides has altered radically in recent years. As little as 10years ago parents may not have considered pesticides to be as dangerous to their children as they do now. Public Health Reviews 1989/90; 17 14 Kelly S, Guidotti TL. Phenoxyacetic acid herbicides Although these methodological problems are difficult to assess or to overcome, their individual and collective effects are likely to under estimate the risk associated with exposure to the phenoxyacetic acid herbicides. Thus, demonstration of an elevated risk in such studies should be considered as an improbable outcome suggestive if it is observed. The risk estimates themselves, in this view, serve not so much as measurements of the magnitude of risk but as indicators of an elevated risk of unknown but larger magnitude. SOFT TISSUE SARCOMAS The epidemiological study of soft-tissue sarcomas (STS) typifies may of the problems inherent in occupational epidemiology. The nature and duration of exposure are difficult to quantify, death certificate entries on occupation cannot give an adequate exposure profile, latency is unk'own, and the treatment of subjects lost to follow-up can radically alter results. There are also problems specific to studies of STS such as the low overall incidence in the populations and the definition and identification of STS. Table 6 summarizes the results of the STS studies discussed. The soft-tissue sarcomas present a difficult problem. Two disease classification systems are used for neoplasms. One is based on anatomic side (ICD-9) and the other on tissue type (ICD-O). Death certificates use ICD-9 only, but cancer registries use both. Because soft-tissues are scattered throughout the body they are difficult to identify in the anatomic (ICD-9) system. Data from the Alberta Cancer Board demonstrate that only 45% of the soft-tissue sarcoma cases can be identified as such. In other geographic areas as few as 30% of cases can be identified. The tumors themselves are rare and histologic differentiation of the neoplasms is difficult but improving with new technology as more specific cell markers are developed. The study of soft-tissue sarcomas presents several unique problems: I) The STS are classified awkwardly in the International Classifi cation of Diseases (ICD) system. This system is organ-based. For example, soft-tissue sarcomas arising in the mesenchyme of the stomach would be classified under stomach rather than under ICD 171: Malignant Neoplasm of Soft and Other Connective Tissue. If ICD 171 is the only category used for patient identifi cX> Public Health Reviews 1989/90; 17 CO 00 Kelly S. Guicioui TL. Phenoxyacetic acid herbicides 15 cation, 50-75% of cases may be missed (19). This can be confirmed by an examination of data from the province of Alberta. (We are indebted to Ms. S. Fincham and Dr. M. Koch for making these data available to us.) Figure 3 presents ten years of histologically selected cases from the Alberta Cancer Board as originally coded. Figure 3 shows all cases coded in 1CD-9 171 against those coded otherwise. Only 45% of all histologically identified cases of soft-tissue sarcoma were coded as 1CD-9 category 171.1CD-9 171 cannot, because of its restrictive criteria, be considered a representative sample of all cases. & IC O 9 17 1 only Fig. 3. Soft-tissue sarcoma ICD-O subclasses, showing the percentage of cases in each histological (ICD-O) classification that were classified by the turnover registry under ICD-9 category 171 compared to those classified in all other ICD-9 categories. 2) These tumors are rare. Again, referring to Alberta data, the combined annual incidence rate is only 3.89 per 100,000 population if determined from ICD-9 or 8.5 per 100,000 population if calculated from ICD-O categories (30). It is difficult to recruit sufficient patients for a case-control study and cohorts would have to be extremely large to identify even a five-fold increase in risk. There is also a bimodal distribution that includes a peak in Public Health Reviews 1989/90; 17 16 Kelly S, Guidoui TL. Phenoxyacetic acid herbicides ICD-0 Categories Fig. 4. The percentage of each ICD-O classification of soft-tissue sarcomas classified by the tumor registry under ICD-9 category 171, compared to those classified in all other ICD-9 soft-tissue sarcoma categories showing similar distributions. early childhood and differing rates between males and females, females being underrepresented in exposed occupations. These features further reduce the number of cases available for the study of occupational associations of these tumors (31). 3) In spite of the common use'of the term "soft-tissue sarcoma," there is considerable variation in its operational definition. Some pathologists include extra-skeletal osteomas in the group but others do not. Gliomas may or may not be included and some researchers exclude visceral sarcomas. Mesothelioma, a visceral sarcoma, is usually classified separately but there may be overlap in pathological definition. 4) Histological differentiation within the STS group is difficult and requires classification by a pathologist familiar with these subgroups (22,31,32). Woods et al. reviewed all STS cases in one study (32). He ultimately retained the original diagnosis in 45% of the cases, changed the histologic diagnosis within the general category of STS in 34%, and rejected 21% as not STS or of c) O D Public Health Reviews 1989/90: 17 &:> Kelly S, Guidoni T L Phenoxyacetic acid herbicides 17 indeterminate pathology. Such detailed typing is usually restricted to research institutions. An association between sarcomas and pesticide exposure was first suggested in 1974 by Axelson et al. (33,34). They found a significantly higher tumor incidence and mortality rate from malignancies in Swedish railroad workers exposed to the herbicide amitrole, a suspect animal carcinogen that is chemically unrelated to the phenoxyacetic acids. In the analysis, it was found that the affected workers had also been exposed to 2,4-D and 2,4,5-T and to diuron and monuron, two urea herbicides unrelated to the phenoxyacetic acids. In 1977 Hardell described a group of STS patients who gave histories of occupational exposure to phenoxyacetic herbicides (35). Axelson subsequently re studied Hardell's cohort and suggested that the phenoxyacetic acid herbicide exposure might have been responsible for results in the earlier study but that a ten-year latency may be needed for the disease to appear (34). This conclusion generated a flurry of studies in Sweden and abroad (Table 6). The initial findings by Hardell generated two case-control studies. In the first, Hardell mailed a questionnaire to STS patients and to controls drawn from the general population of northern Sweden (36). In the analysis he examined several possible confounders, among them exposure to DDT, diesel oil, and tobacco smoking. They found that Table 6 Su m m a ry o f studies on soft tissue sarcomas C o h o rt type Positive and stai, signif. Study outcom e Positive, but N egative not stat. signif. m anufacturers forestry and agriculture V ietnam veterans pesticide exposed 33-34*.35,36*,37* 56,57 42,43,44,45.46,47 32,59*,62*,63 51.52.54.55* 60-61 * H istologically confirm ed STS cases. - Indicates sam e study o r p o p u latio n described in different reports. Public Health Reviews 989/90: 17 18 Kelly S. Guidoni TL. Phenoxyacetic acid herbicides exposure to phenoxyacetic acids or chlorophenols yielded a relative risk (RR) of 5.7 (95% confidence interval 2.9, 11.3). Exposure to phenoxyacids alone yielded an RR of 5.3 (2.4, 11.5) and to chlorophenols alone, an RR of 6.6 (95% confidence interval by testbased method: 1.4,31.1). These are very high estimates of relative risk. In the second study, Eriksson et al. (37) identified cases of STS from southern Sweden. They regarded persons in contact with cutting oils and in the shoe and leather industry as unexposed and used them in the control group. A relative risk of 6.8 (95%confidence interval 2.6,17.3) was found for exposure to pehnoxyacids and 3.3 (1.3, 8.1) for exposure to chlorophenols. This estimate is remarkably consistent with the estimate in the second Hardell study (36). Both the second Hardell (36) and the Eriksson et al. (37) studies examined all histologically diagnosed cases of STS (not just those under ICD 171) and excluded from analysis all who had sustained pesticide exposures within the previous 5 years only or for a duration of less than 1day. Both investigators supplemented the questionnaire information with telephone interviews but only for pesticide exposure data. This raised the question of bias, which they attempted to control by using another group of cancer patients as a second set of controls, reasoning that recall bias from concern over health would generally be found among cancer patients. Use of this second control group did not uncover evidence of bias. Manufacturing plant workers Following the study of Eriksson et al. (37) a series of case reports appeared in the literature describing patients with STS who had worked in pesticide production plants (38-41). Subsequent review of the histology reclassified two cases as carcinomas, not sarcomas. One case report described the occurrence of STS in a father and son, both of whom worked in the same plant, and raised the possibility of a genetic predisposition to the effects of exposure. However, the histologic types were very different (Schwannoma and lipoma) and the two had very different exposure histories (40). Several studies of US production plant cohorts (42-47) found no statistically significant excess of STS. Three of the studies were small (61, 121, and 204 persons) and had a very limited statistical power to detect increases. All had assumed a sufficient latency between Public Health Reviews 1989/90; 17 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides 19 exposure and outcome of study. Two of the studies used chloracne as an indicator of exposure. Chloracne in man usually occurs as comedomes in a malar distribution and on the pre- and post-auricular portions of the face and head and is often accompanied by hirsutism, melanosis, or secondary infection (48, 49). The threshold level for development of chloracne varies considerably among individuals (49), making it a poor indicator of exposure. Agent Orange Forty million liters of a 50:50 mixture of 2,4-D and 2,4,5-T named Agent Orange was applied to the tropical jungle in Southeast Asia as a defoliant by the US military in the Vietnam War. The herbicide was applied to clear vegetation from around bases and camps and along communication lines, and t'o expose operational staging areas. Contamination by 2,3,7,8-TCDD at levels of 0.02-54 micrograms/g have been reported in the Agent Orange used at the time (46). Agent White and Agent Blue were two other defoliant sprays used by the US military. Agent White contained picloram, which has been demon strated to be a carcinogen in rats and mice; Agent Blue contained salts of arsenic (50). Up to 2 million American military personnel and many more Vietnamese could have been exposed to one or more of these mixtures. In a study of 1,024 men who sprayed the herbicides in Vietnam ("Operation Ranchhand" personnel) there were no statis tically significant excesses of cancer in a baseline mortality study (51). Despite the numbers, the power of the study was limited because of the rarity of STS and the low numbers of expected cases. Kang et al. (52) gave the results of a hospital-based comparison group study of veterans who served between 1964 and 1975 and who were diagnosed in a Veteran's Administration (VA) Hospital as having STS (1CD 171). The controls were selected from the same hospital population. Military service information was abstracted from the US National Personnel Records Center as a surrogate for exposure data. No significant association between STS and previous military service in Vietnam was observed (OR = 0.83, 95% confidence interval 0.63, 1.09). This study must be interpreted with caution and has been highly controversial because personal exposure data were missing and outcome measurements were uncertain. In a follow-up study to examine potential bias in identifying STS, Kang et al. (54) stratified Public Health Reviews 1989/90; 17 20 Kelly S, Guidot TL. Phenoxyacetic acid herbicida the data on hospital type and found no statistically significant risks. Many cases of STS may have been missed because only ICD 171 was used to identify the cases. During the Vietnam era there was poor, and sometimes biased, military record keeping (53), a problem aggravated by a fire in the central records depository that destroyed many records in the early 1980s. It is probably impossible to establish satisfactory estimates of individual exposure. Outcome measurements may be equally difficult to obtain. Patients in VA hospitals also may not contribute a representiative sample of the total veteran population. Unfortunately, case ascertainment was poor because many hospitals refused to particupate in Kang's study. The only Vietnam veteran study to use histologically-identified STS was that by Greenwald et al. (55), who used interviews to collect data on 151 live and 130 dead cases, including military service, occupatioal history (emphasizing exposure to herbicides), specific chemical exposures, family history, medical history, smoking, and alcohol abuse. They found no significant association with any particular type of herbicide exposure. Some of their cases may not have had a sufficient latency period (15-20 years) for cancer to develop if the latency for STS is comparable to that for other cancer types. Other studies Three other studies of largely agricultural populations have reported detecting a significant excess of STS. Balarajan (56) examined STS among farmers, agricultural workers, and related occupational groups in England and Wales. One of the four occupational subgroups in the study (including farmers, farm managers, and market gardeners) showed a relative risk of 1.7 (95% confidence interval, 1.00, 2.88). Unfortunately, only 70% of their cancer patient registrations had an occupation listed and it is not knownwhethertherewould havebeenaselectivebiasinrecordingoccupation. Vineis et al. (57) identified STS cases through hospital pathology departments in 3 provinces of Italy. Cases were verified histologically but visceral cases were excluded. Ascertainment was poor and introduced the possibility of bias by excluding cases with character istics not shared with respondents (58). Vineis et al. found an increased risk of 2.7 (not significant at p = 0.05) among female rice weeders. El Zayadi et al. (59) found that 10 out of 14 patients with hepatic Public Health Reviews 1989/90; 17 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides 21 angiosarcoma in Egypt had a history of direct, chronic, recurrent exposure to agricultural pesticides. The types of pesticides were not described. Most other studies of STS have found no association between phenoxyacid herbicide or chlorophenol use and soft-tissue sarcoma. Smith et al. (60, 61) did not find significant results in a case-control study of all cancers registered in the New Zealand Cancer Registry. Hoar et al. (62) used histologically verified cases and found no increase in risk for STS among white male residents of Kansas who gave a history of work with herbicides including phenoxyacid herbicides and chlorophenols. Wiklund and Holm (63) studied a cohort of Swedish men who had been employed in agriculture or forestry in 1960 and found no significantly increased risk either in the whole cohort or in any of several subgroups. In Washington state Woods et al. (32), reported that study subjects with any past occupational exposure to phenoxyacetic acid herbicides had an estimated RR of 0.80 (95% confidence interval 0.5, 1.2). They used histologically verified cases of STS. In 1976 there was an industrial accident and explosion at a chemical plant in Seveso, Italy, exposing many local residents to trichlorophenols and dioxins. Local health authorities set up a health monitoring program and a cancer registry. By crudely comparing contaminated and uncontaminated areas, based on TCDD levels in soil, there appears to be a clear difference in the incidence of STS in 1980-81 (48). Unfortunately there are no comparably ascertained STS incidence rates from before the accident by which to establish any trend. Puntoni et al. suggest that it is highly unlikely that this increase could be due to the accident, however, as the latency period would then be only four years (64). LYMPHOMAS Lymphomas, like soft tissue sarcomas, arise from tissue of mesenchy mal origin. They are divided into Hodgkin's disease (HD) and nonHodgkin's lymphoma (NHL) on histologic criteria. NHLs are commonly divided further in the 1CD-9 into prolymphocytic lympho sarcoma, malignant lymphoma, lymphosarcoma, and reticulosarcoma (1). This classification scheme is considered obsolete by clinical Public. Health Reviews 1989/90; 17 22 Kelly S. Guidoni TL. Phenoxyacetic acid herbicida investigators, who have replaced this crude system with much more detailed nosologies based on morphology and immunological typing. Cancer registries use both systems. Studies on the association between lymphomas and the use of phenoxyacid herbicides or chlorophenols are summarized in Table 7. In other settings, risk for NHL is associated with immunological disorders or immunosuppressive treatment. The possibility of an association with immunostimulation has not been fully documented but it is of interest that 2,4-D is a stimulus to B-cell proliferation, given that NHLs are predominantly of B-cell origin (20, 31). Goldsmith and Guidotti (65) studied deaths due to lymphosarcoma in California males in 1971 and 1972. Based on 1970 census data they found a significant excess among farmers, farm managers, and farm laborers when compared to the expected number. In 1974, the American Cancer Society published a survey of malignancies in Iowa between 1969 and 1971. DeKraay (66) examined the distribution and suggested a link between pollution in drinking water on farms and lymphoma. In a death certificate study of Iowa farmers (67), eD CiO i-p"" o Table 7 Su m m a ry o f studies on lym phom a C ohort type Positive and slat, signif. Study outcom e Positive, but N egative n ot slat, signif. m anufacturers agriculture forestry and wood pesticide exposure 45.62,65,66,67,68 71,72,77/78 32,79*.80* 69 32,82+ 74 47 82+ 32,75-76-77 * H o d g k in 's d ise a se o n ly . + includes fishing. - indicates sam e study or population described in different reports. Public Health Reviews 1989/90; 17 Kelly S. Guidoni TL. Phenoxyacelic acid herbicides 23 Burmeister found a statistically significant SM R of 1.29 for lymphoma in spite of a significantly lower overall cancer rate. In a further casecontrol study (68), he found a statistically significant odds ratio of 1.26 for farming and NHL. In 1979, Hardell (69) noted that several patients with lymphoma had reported exposure to phenoxyacids and chlorophenols. In a casecontrol study (70), he sent self-administered questionnaires to patients with NHL of the histiocytic type and to controls and obtained an OR of 6.0 (95% confidence interval 3.7,9.7) for exposure to phenoxyacids or chlorophenols. He calculated an OR of 4.8 (2.9,8.1) for exposure to phenoxyacids only. Males in Wisconsin who had died of NHL, as recorded on their death certificates, were studied by Cantor (71), who found that farmers less than 65 years of age were more likely to have NHL than non-farmers (OR = 1.67,95% confidence interval 1.14,2.5). Reticulum cell sarcoma was more common among farmers under 65 years of age, with an OR of 2.7 (1.54, 4.5). Using computerized death certificates in Wisconsin, Saftlas et al. (72) examined causes of death in all persons whose "usual occupation" was described as farmer. Since farmers smoke less than non-farmers, smoking-related causes of death were removed from the analysis and a proportionate (corrected) mortality ratio (PCMR) was calculated on the remaining deaths. He found a statistically significant PCMR of 1.25 for the entire cohort of farmers. This could not be linked to specific types of farming when the degree of agricultural activity in each county was used as a surrogate for individual exposure. As with soft-tissue sarcomas a series of case reports on lymphoma in pesticide production workers appeared in the literature (73, 74). Unfortunately, most manufacturing cohort studies have focused on STS and did not consider NHL. Only two studies specifically examined the risk for lymphoma. One did not uncover a statistically significant excess. One of the largest studies of manufacturing cohorts is that by Bond et al., who were primarily investigating an alleged association between 2,4-D exposure and brain tumors suggested by animal studies (47). Among 878 chemical workers potentially exposed to 2,4-D between 1945 and 1982, they found two deaths from NHL, for an SMR of 391 (95% confidence interval: 44, 1411). Although elevated, this finding is not easily interpretable because of the small number involed and resulting wide confidence interval. Although it Public Health Reviews 1989/90; 17 24 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides cannot be considered evidence for an association, the study is suggestive. In other studies (42-44) the sample sizes have been too small to attain convincing power and the latency periods too short. Hoar et al. (62) studied white male residents of Kansas diagnosed with NHL. During a structured telephone interview participants were asked detailed questions about farming practices. Hoaret al. found no trend with years spent living or working on a farm but did find a significant trend between the risk of NHL and increasing years of herbicide use and number of days of herbicide exposure per year. With more than 20 days per year of exposure the OR was 6.0 (95% confidence interval: 1.9, 19.5). Subjects who reported usually mixing or applying the herbicide themselves showed an OR of 1.9 (95% confidence interval: 1.1,3.3) compared to those who had someone else perform this function (OR = 1.1; confidence interval not available). Farmers who did not use protective equipment with herbicide use had a higher odds ratio (OR = 2.3; 95% confidence interval: 1.0, 4.2). Higher risks for herbicide use were also seen among farmers who used backpack or hand sprayers (OR = 2.3; 95% confidence interval: 1.0, 5.2). They also examined the known associations affecting risk for lymphoma, including conditions resulting in immunosuppression, drugs, family history, cigarette smoking, coffee consumption, and ionizing radiation. Other than herbicide use, only family history had a significant OR of 2.3 (95%confidence interval: 1.6,3.2). Their sample of farmers had an unusually high percentage (77%) of people who "never used" herbicides. Farming practices in the "grain-belt" make this statistic highly improbable. However, the bias introduced by error in determining use of herbicides would tend to underestimate the risk calculated. This study has since been replicated in Nebraska. Pearce et al. have conducted several studies in new Zealand (75-77). I nal 985 case-control study (75) of lymphoma and multiple myeloma, they found an OR of 1.25 for agriculture or forestry workers, using current or most recent occupation on death certificate. For ICD category 202, the OR was 1.76. They then inteviewed lymphoma cases registered under ICD 202 and found no significant results for exposure to phenoxyacetic acid herbicides orchlorophenols(76). Meat packing workers exposed to 2,4,6-trichlorophenol and zoonotic viruses demonstrated an OR=5.7 (90% confidence interval 2.3, 14.3; p=0.03). There was an elevated risk for fencing work but in New Zealand only 5%of the pesticides used in this type of work are chlorophenols. They Public Health Reviews 1989/90; 17 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides 25 study was then expanded to include ICD category 200 (77). Neither the combined cohort nor the new classification showed a statistically significant excess associated with exposure to phenoxyacetic acid herbicides or chlorophenols. It is interesting to note that ICD category 200 had two results that were high but failed to achieve statistical significance for either railway workers who sprayed chemicals or for Ministry of Works employees who sprayed chemicals on public lands. Schumacher (78) studied white male Utah residents with a diagnosis of NHL occurring after their 35th birthday and who died before their 75th birthday. Occupations as reported on death certificates were analyzed for cases and for a series of controls with adenocarcinoma of the colon. Stratifying the data by year of diagnosis revealed an OR for any type of agricultural occupation of 6.6 (90% confidence interval 1.2, 3.57) for those diagnosed between 1952 and 1966; 3.1 (95% confidence interval: 1.2,8.1) for those between 1966 and 1971;and 1.0 by the period 1972-77. There may not have been a sufficiently long latent period between exposure and the latest time period studied. Studies of Hodgkin's disease (HD) do not appear to demonstrate an association with herbicide exposure but do suggest an association with exposure to wood dust. Milham and Hesser (79) analyzed death certificate occupations for 1549 white males aged 25 years or more who had died of HD in New York state excluding New York City in 1940-53 and 1957-64. Their study revealed a preponderance of men whose work entailed exposure to wood. Grufferman et al. (80) also found an excess risk of HD among woodworkers in the Boston Standard Metropolitan Statistical Area. They studied histologically confirmed HD cases and compared them to the 1960 and 1970 census data for the same area. As will be discussed below, this observation may provide a lead for investigating the mechanism of carcinogenicity. Mortality patterns for deceased members of the American Federa tion of Grain Millers Insurance Plan were studied by Alvanja et al. (81). Occupational information was taken largely from employment and union records. Questionnaires were sent to a sample. Alvanja et al. found a PMR=202 (p<.005) for disease of the lymphatic and hematopoietic system and a PM R=272 for other lymphatic neoplasms. Woods et al. conducted a case-control study in Washington state (32) to specifically examine the risks of using the phenoxyacetic acid herbicides or chlorophenols. Among study subjects with past occupational exposure to phenoxyacetic acid herbicides the estimated Public Health Reviews 1989/90: 17 26 Kelly S, Guidotti TL. Phenoxyacetic acid herbicides RR was 1.07 (95% confidence interval: 0.8, 1.4). For farmers the RR was 1.33 (95% confidence interval: 1.03,1.7) and for forestry herbicide applicators it was 4.8 (95% confidence interval: 1.2, 19.4). When the analysis was restricted to anyone with exposure to phenoxyacetic acid herbicides in any occupation for 15 or more years during a period 15 years prior to cancer diagnosis the RR was 1.71 (95% confidence interval: 1.04, 2.8). Occupation and risk of cancer in Denmark was studied by Olsen and Jensen (82) in a computerized death certificate study. They did not find a significant excess of NHL or HD in males or females employed in agriculture, forestry, or fishing. The inclusion of fishing, which has never been implicated as a suspect occupation, may have diluted the result. Wiklund and Holm (83) considered that increasing pesticide use would have resulted in increased risks in recent decades but found no statistically significant change in the RR for NHL or HD over 1961-79 in agricultural occupations. DISCUSSION Taking into account the limitations inherent in the methodology of all available studies, we conclude that the evidence for carcinogenicity of the phenoxyacetic acid herbicides, specifically 2,4-D, is strongest for NHL and weak for STS. If the association between the herbicide and NHL is causal, the potency of the chemicals as a carcinogen must be weak. There are only two studies of STS that show a convincing association with phenoxyacetic acid herbicides and chlorophenols, both from Sweden. Manufacturing plant cohorts which are assumed to be highly exposed did not show any association but, with two exceptions (45, 47), were too small in size to attain sufficient power. Under critical examination the association between STS and phenoxyacetic acid herbicide or chlorophenol exposure may be an irresoluble issue in the absence of epidemiologic evidence to support a conclusion. A study to be conducted in Canada has recently been announced. Canada has a climate and economic level similar to Sweden. This study will be coordinated by the University of Saskatchewan. Public Health Reviews 98990; 17 Kelly S. Guidoni TL. Phenoxyacetic acid herbicides 27 Studies of non-Hodgkin's lymphoma generally show a slightly increased risk. The effect of the phenoxyacetic acid herbicides would appear to be specific for NHL, since no effect has been demonstrated for Hodgkin's disease. Table 7 summarizes the data and reveals a pattern. Agricultural cohort studies all show a positive correlation, as do most studies in which pesticide exposure has been identified. In addition, studies that attempt to quantify pesticide exposure have tended to report higher risks (21, 62, 70). There are many reasons why risk estimates may be inconsistent among cohorts. Death certificate studies and occupational health monitoring data may dilute results by the inclusion of subgroups at low risk, depending upon the composition of the occupational category that includes agriculture (25). The codes for the lymphomas and 1CD 171 embrace a very wide variety of tumors. They are by no means equal in distribution or risk factors. The aggregation of diagnostic codes may be masking a highly specific effect limited to a rare tumor. Failure to distinguish among the many types of lymphoma may dilute risk estimates. Studies of non-Hodgkin's lymphomas present different but equally difficult classification problems than the soft-tissue sacromas, but show a clearer increase in risk. It cannot now be determined whether only one subtype of NH L or several show this increased risk. Cantor (71) found a higher risk for one subtype of NHL. Present systems of categorizing the lymphomas, especially, for epidemiological purposes are probably almost useless in the inter pretation of the carcinogenic behavior of chemical exposures because the immune system is so highly structured that it is likely that such effects are targeted in a very specific way, causing a particular type of lymphoma to the exclusion of others. We advocate a study in which cell surface antigen markers are used to specifically identify the lymphoma by a consistent typology, or the clinical system developed by the National Institutes of Health (84). We believe that terms such as lymphosarcoma or reticulosarcoma" are no longer useful. 1 he phenoxyacetic acid herbicides may be carcinogens specific to the immune system by a mechanism unique to lymphatic tissue. ICDD has been shown to be an immunosuppressant and 2,4-D an immunostimulant. TCDD or the herbicide itself might cause neo plasms in the immune and lymphatic system, perhaps by altering the Public Health Reviews I9H9/90; 17 28 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides surveillance function of lymphocytes. The data may suggest an interactive effect between phenoxyacetic acid herbicides or chlorophenols and something else in the agricultural or forestry environment. Hoar showed that farmers never exposed to herbicides had a risk close to that of non-farmers, suggeting that farming itself in the absence of exposure to herbicides confers no risk (62). Woods et al. showed that increased risks of NHL were observed among those with occupational exposure to organochlorine insecticides such as DDT and organic solvents, and to other chemicals typically encountered in agriculture, and the forestry and wood products industries (32). An important recent case-control study by Olsson et al. suggests that exposure to organic solvents may be a risk factor for NHL (85). If confirmed, this might shift attention to pesticide vehicles and diluents as well as their active ingredients. TCDD, a known contaminant in early 2,4,5-T formulations, is recognized to have an immunosuppressive effect (21) and appears to act as a promoter in carcinogenesis rather than as an initiator (14). Confounding or interacting exosures may also play a role in the potential for cancer risk associated with herbicide use. A putative risk factor participating in an interaction does not have to be a man-made chemical. Lymphomas can be induced in animals with plant lectins, which are carbohydrate-binding proteins that stimulate mitosis in lymphocytes and act as hemagglutinins. They have been identified in legume seeds (peas and beans), solanacia (potatoes and tomatoes), triticeae (grains), and several animal species. As noted, a similar process could be responsible for the association observed between Hodgkin's lymphoma and wood dust. Other substances in the natural environment are known to be initiators in carcinogenesis including aflatoxins, and some phenylpropenes (86) such as safrole. Several types of viral cancer have been identified in farm animals and this may also occur in humans. Two studies (67,71) of agricultural cohorts have identified increased risks of leukemia in groups exposed to grain, for NHL in groups exposed to milk production, and for both groups exposed to egg-laying chickens. Further investigation of an association between phenoxyacetic acid herbicides or chlorophenols and non-Hodgkin's lymphoma seems justified. Because of the many reasons why risk estimates may be inconsistent among cohorts, we cannot endorse the approach of meta analysis used by Bond et al. (87) of mapping probability densities from Public Health Reviews 1989/90; 17 Kelly S. Guidoni TL. Phenoxyacetic acid herbicides 29 published studies and drawing conclusions from pooled risk estimates. The quality, power, and consistency of method among the various available studies do not support such a sophisticated approach in our judgement. Targeted epidemiologic studies are needed that will quantify pesticide exposure and relate this to immunologic subtypes of non-Hodgkin's lymphoma. The potential for interactive effects between pesticides and other substances in the forestry or agricultural environ ment is a reasonable line of investigation. Future studies should address the possibility that phenoxyacetic acid herbicides, chlorophenols, or their contaminants act as promoters in carcinogenesis. A recent review of this same topic by Lilienfeld and Gallo, (88) which placed more emphasis on 2,4,5-T, has come to the same general conclusions as our interpretation. t Recommendations for the phenoxyacetic acid herbicides Government regulating agencies, school boards, parents, park super intendents, grounds maintenance personnel, and the agricultural community have become increasingly concerned about the safety and health aspects of herbicide use. This paper, for example, grew out of our own review of the toxicity of the phenoxyacetic acid herbicides prepared in part for the Pesticide Management Branch of Alberta Environment. It is our conclusion that the potential for carcinogenicity of2,4-D is real but very limited and not potent. Because 2,4-D quickly binds to the surfaces of plants and degrades very quickly in the environment there is no compelling reason to restrict its use on the basis of it being a widespread environmental hazard. It may present an occupational hazard in pesticide application but not of great magnitude and is not likely to be a significant environmental hazard. The evidence associating 2,4-D and the uncommon class of malignancies known as the "non-Hodgkin's lymphomas"is increasingly convincing in spite of methodological limitations, some of which we have documented here. However, 2,4-D is remarkably free of acute toxic effects. Even given the potential for increasing the rate of these uncommon cancers, its use presents a limited risk that can be controlled by the use of personal protection and sensible precautions. With respect to health concerns 2,4-D presents advantages over alternative herbicides although it is not a perfect solution to the problem of safe vegetation management. Public Health Reviews 198990; 17 30 Kelly S, GuiJoili TL. Phenoxyacetic acid herbicides The study by Hoar in Kansas (62) showed a clear decrease in the risk for NHL associated with the use of protective clothing. Agricultural use of 2,4-D can continue with minimal risk if farmers and applicators use personal protection to limit the extent and duration of exposure. There is no evidence that existing guidelines and recommendations are inadequate if followed. These include protective outer clothing, avoiding contamination of clothing, avoiding prolonged skin contact, respiratory protection until the herbicide application is completed, and uncontaminated or decontaminated air supplies in farm vehicles. Treatment of turf areas with 2,4-D for weed control can be performed safely in parks and playgrounds. Furthermore, most of the 2,4-D picked up on a child's clothing, for example, would not be directly applied to the skin. The efficiency of abrasion or rubbing as a means of picking up 2,4-D residue on skin or clothing would be minimal. The Guelph study used organic solvents and repeated abrasion (9). In practical terms the hazard would be vanishingly small after a day or two, even under the worst assumptions. Our assessment of the evidence to date serves to confirm that current recommendations to parks, school boards, and citizens concerned with lawn care are, we suggest, safe and present no substantial risk to children. Turf areas treated with 2,4-D are safe to reenter within 48 hours under dry conditions and within 24 hours after watering, rainfall, or a heavy dew. Should a child inadvertently wander into a site or play on treated grass before these reentry periods have expired there is no cause for immediate concern or medical examination. 2,4-D is low in toxicity and is by now well-studied. Alternative pesticides now available are more hazardous, more persistent in the enviromenl, and less well understood. Mechanical and manual removal of undesirable vegetation is very labor-intensive, expensive, and carries a high occupational health cost associated with ergonomic and safety hazards. Properly applied and with minimal precautions, 2,4-D is probably the safest alternative presently available for turf management in public areas, at least given the present level of documentation. Herbicides have widely replaced intensive manual control of unwanted vegetation for reasons of economy and safety. Concern over the safety of available herbicides has forced a rethinking of vegetation management by those in industry as well as by those criticizing current Public Health Reviews 1989/90; 17 Kelly S. Guidotti T L Phenoxyacetic acid herbicides 31 practices (89). The most intensive use of herbicides is to manage agricultural land and to control unwanted vegetation at industrial sites or corridors for transportation, energy, and communication facilities. Vegetation management in this setting is more than a matter of killing weeds safely. Its fundamental purpose is to provide cost-effective control of undesirable vegetation in a way that does not harm the environment and in a manner that protects the health and well-being of all affected. The parties primarily affected include applicators and field workers, community residents, farmers, and those whose safety is dependent on suppressing vegetation growth which may cause a fire hazard, a barrier to vision or visibility, destruction of facilities, impeded access to hazardous equipment, or entangled ground cover. Most approaches to managing vegetation without herbicides carry substantial safety and health implications: stoop labor to pull out weeds, widespread use of chainsaws, pulling stumps to remove trees, controlled burning, sending crews into inaccessible or dangerous terrain, and paving surfaces that would otherwise be left open. Changes in techniques of vegetation management would have an impact on operations in a number of industries besides agriculture and landscaping: power and telephone transmission lines, highway and road maintenance, park management, airport and other grounds management, railroad rights of way, and forestry, for example. A policy that encourages vegetation management by the means most appropriate to the application holds the most promise for protecting the health of the public. REFERENCES 1. World Health Organization. International classification of diseases for oncology. Geneva: WHO, 1976. 2. Interdepartmental Committee on Toxic Chemicals. Dioxins in Canada: the federal approach. Hull, Quebec: Environment Canada, December 1983. 3. Axelson O. The health effects of phenoxy acid hebicides. Recent Advances in Occupational Health 1984; 2: 253-267. 4. US Public Health Service. Toxicological profile for 2,3,7,8-letrachlorodibenzo-p-dioxin: draft for public comment. Agency for Toxic Substances and Disease Registry, US Public Health Service, 1988. 5. Plant Products Division, Dominion Department of Agriculture. List of Public Health Reviews I98V/90; 17 32 Kelly S, Guidotti T L Phenoxyacetic acid herbicides herbicidal products submitted for registration as of 1 May 1933. Ottawa, Ontario: Dominion of Canada, 1933. 6. Krystynak R. An economic assessment of the benefits of 2,4-D in Canada. Ottawa, Agriculture Canada, 1989. 7. McLaren C. Ban on use of 2,4-D assessed at millions. The Globe and Mail (Toronto), 3 November 1988, p A5. 8. Kelly S. Pesticide poisoning: An emergency treatment manual. Edmonton, Alberta, Alberta Environment, Pesticide Chemicals Branch, 1984. 9. Thompson DG, Stephenson GR, Sears MK. Persistence, distribution, and dislodgeable residues of 2,4-D following its application to turfgrass. Pestic Sei 1984; 15: 353-360. 10. Schulze GE, Blake JW , Dougherty JA. The metabolic fate of 2,4dichlorohenoxyacetic acid-n-butyl ester in the Wistar rat. Arch Toxicol 1985; 57: 231-236. 11. Grover R, Franklin CA, Muir NI, Cessna AJ, Riedel D. Dermal exposure and urinary metabolite excretion in farmers repeatedly exposed to 2,4-D amine. Toxicol Lett 1986; 33: 73-83. 12. Davring L, H ultgren K. Cytognie effects on in vivo bone-marrow cells of M us musculus induced by a commercial 2,4,5-T ester. Hereditas 1977; 85: 123-134. 13. US Environmental Protection Agency. Risk assessment on (2,4,5-trichlorophenoxy) acetic acid (2,4,5-T), (2,4,5-trichlorophenoxy) propionic acid and 2,3,7,8-tetrachloro-dibenzo-p-dioxin (TCDD). Washington, DC: US Government Printing Office, July 1981. 14. Shu HP, Paustenbach DJ, Murrary FJ. A critical evaluation of the use of mutagenesis, carcinogenesis, and tumor promotion data in a cancer risk assessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Regul Toxicol Pharmacol 1987; 7: 57-88. 15. Yoder J, Watson M, Benson WW. Lymphocyte chromosome analysis of agricultural workers during extensive occupational exposure to pesticides. Mutant Research 1973; 21: 335-340. 16. Hogstedt B, Kolnig A, Mitelman F, Skerfving S. Brief reports: cytogenic study of pesticides in agricultural work. Hereditas 1980; 92: 177-178. 17. Crossen PE, Morgan WF, Horan JJ, Stewart J. Cytogenic studies of pesticide and herbicide sprayers. NZ Med J 1978; 88: 192-195. 18. Vos JG, Krajnc EI, Beekhof PK, Van Logten ML Methods for testing immune effects of toxic chemicals: evaluations of immunotoxicity of various pesticides in the rat. In: Miyamoto J, Kearney PC, eds. Pesticide chemistry, human welfare and the environment. Vol. 3, Mode of action, metabolism and toxicology, p 497-504. Oxford: Pergamon Press, 1983. 19. Blakley BR, Schiefer BH. The effect of topically applied n-butylester of Public Health Reviews 1989/90; 17 Kelly S, Guidoni T L Phenoxyacetic acid herbicides 33 2,4-dichlorophenoxyacetic acid on the immune response in mice, J Appl Toxicol 1986; 6(4): 291-295. 20. Blakley BR. The effect of oral exposure to the n-butylester of 2,4dichlorophenoxyacetic acid on the immune response in mice. Int J Immunopharmacol 1986; 8 (1): 93-99. 21. Hoffman RE, Stehr-Green PA, Webb KB, et al. Health effects of long term exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. JAMA 1986; 255 (15): 2031-2038. 22. Fingerhut MA, Halperin WE, Honchar PA.et al. An evaluation of reports of dioxin exposure and soft tissue sarcoma pathology among chemical workers in the United States. Scand J Work Environ Health 1984; 10 (5): 299-303. 23. Percy C, Stanek E, Gloeckler L. Accuracy of cancer death certificates and its effects on cancer mortality statistics. Am J Public Health 1981; 71 (3): 242-250. 24. Finklestein MM, Liss GM. Selection bias in occupational case-control studies that use death registries to select subjects: a discussion and demonstration. Am J Industr Med 1987; 12:21-31. 25. Guidotti TL, Baser M, Goldsmith JR , Comparing risk estimates from occupational disease monitoring data. Public Health Rev 1987; 15: 1-27. 26. Schlesselman JJ. Case control studies: design, conduct, analysis. Oxford: Oxford University Press, 1982. 27. Delzell E, Grufferman S. M ortality among white and nonwhite farmers in North Carolina, 1976-1978. Am J Epidemiol 1985; 121 (3): 391-402. 28. Vena JE, Sultz HA, Carlo GL, Fiedler RC, Barnes RE. Sources of bias in retrospective cohort mortality studies: A note on treatment of subjects lost to follow-up. J Occup Med 1987; 29 (3): 256-261. 29. Rigakis KB, Kerr N, Crown EM, Eggertson B. Efficacy of selected laundry practices used by Alberta farm families in removing pesticide residues from clothing. Presented at the Association of College Professors of Textiles and Clothing -- Western Region Conference, 25 October 1985; Napa, CA. 30. Alberta Cancer Board. Data from the Cancer Registry of Alberta, 1987. 31. Schottenfeld D, Fraumeni JF. Cancer epidemiology and prevention. Philadelphia: WB Saunders, 1982. 32. Woods JS, Polissar L, Severson RK, Heuser LS, Kulander BG. Soft tissue sarcoma and non-Hodgkin's lymphoma in relation to phenoxyherbicide and chlorinated phenol exposure in Western Washington. J Natl Cancer Inst 1987; 78 (5): 899-910. 33. Axelson O, Sundell L. Herbicide exposure, mortality and tumor incidence: an epidemiologic investigation on Swedish railroad workers. Scand J Work Environ Health 1974; 11:21-28. Public Health Reviews 989/90; 17 34 Kelly S, Guidoni TL. Phenoxyacelic acid herbicides 34. Axelson O. Herbicide exposure and tumor mortality. Scand J Work Environ Health 1980; 6: 73-79. 35. Hardell L, Sandstrom A. Case-control study: soft-tissue sarcomas and exposure to phenoxyacetic acids or chlorophenols. Br J Cancer 1979; 39:711-717. 36. Hardell L, Eriksson M. Soft-tissue sarcomas, phenoxyherbicides, and chlorinated phenols. Lancet 1981; ii: 250. 37. Eriksson M, Hardell L. Berg NO, Moller T, Axelson O. Soft-tissue sarcomas and exposure to chemical substances: a case-referent study. Br J Industr Med 1981; 38: 27-33. 38. Cook RR. Dioxin, chloracne and soft tissue sarcoma. Lancet 1981; i (8220): 618-619. 39. Honchar PA, Halperin WE. 2,4,5-Trichlorophenol and soft tissue sarcoma. Lancet 1981; i (8214): 268-269. 40. Johnson FE, Kugler MA, Brown SM. Soft tissue sarcomas and chlorinated phenols. Lancet 1981; ii: 40. 41. Moses M, Selikoff IJ. Soft tissue sarcomas, phenoxy herbicides, and chlorinated phenols. Lancet 1981; i: 1370, 42. Cook RR, Townsend JC, Otl MG, Silverstein LG. Mortality experience of employees exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). J Occup Med 1980; 22 (8): 530-532. 43. Zack JA, Suskind RR. The mortality experience of workers exposed to TCDD in a trichlorophenol process accident. J Occup Med 1980; 22 (1): 11-14. 44. Ott MG, Holder BB, Olson RD. A mortality analysis of employees engaged in the manufacture of 2,4,5-trichlorophenoxyacetic acid. J Occup Med 1980; 22 (1): 47-50. 45. Ott MG, Olson RA, Cook RR, Bond GG. Cohort mortality study of chemical workers with potential exposure to the higher chlorinated dioxins. J Occup Med 1987; 29 (5): 422-429. 46. Sobel W, Bond GG, Skowronski BJ, Brownson PJ, Cook RR. A soft tissue sarcoma case control study in a large multi-chemical manufactur ing facility. Chemosphere 1987; 16(8/9): 2095-2099. 47. Bond GG, Wetterstroem NH, Roush GJ, McLaren EA, Lipps TE, Cook RR. Cause specific mortality among employees engaged in the manufacture, formulation or packaging of 2,4-dichlorophenoxyacetic acid and its related salts. Br J Industr Med 1988; 45: 98-105. 48. Merlo F, Puntoni R. Soft-tissue sarcomas, malignant lymphomas, and 2,3,7,8-TCDD exposure in Seveso. Lancet 1986; ii: 8521-8522. 49. Tindall JP. Chloracne and chloracnegens. J Am Acad Dermatol 1985; 13 (4): 539-558. Public Health Reviews 1989/90; 17 Kelly S. Guidoni TL. Phenoxyacetic acid herbicides 35 50. Committee on Veteran's Affairs. Hearing before the House Subcommittee on Medical Facilities and Benefits of the Committee on Veteran's Affairs. Washington, DC: US Government Printing Office 1981. 51. Buckingham WA Jr. Operation Ranch Hand: the Air Force and herbicides in South Asia 1961-71. Washington, DC: US Government Printing Office 1982. 52. Kang HK, Weatherbee L, Breslin PP, Lee Y, Shepard BM. Soft tissue sarcomas and military service in Vietnam: a case comparison group analysis of hospital patients. J Occup Med 1986; 28 (12): 1215-1218. 53. Christian RS, White JK. Battlefield records management and its relation ship with the Agent Orange Study. Chemosphere 1983; 12 (4): 761-768. 54. Kang H, Enzinger F, Breslin P, Feil M, Lee Y, Shepard B. Soft tissue sarcoma and military service in Vietnam: a case control study. J Natl Cancer Inst 1987; 79 (4): 693-699. 55. Greenwald P, Kovasznay B, Collins DN, Therriault G. Sarcomas of soft tissues after Vietnam service. J Natl Cancer Inst 1984; 73 (5): 1107-1109. 56. Balarajan R. Soft tissue sarcomas in agriculture and forestry workers. J Epidemiol Community Health 1984; 38 (2): 113-116. 57. Vineis P, Terracini B, Ciccone G, et al. Phenoxy herbicides and soft-tissue sarcomas in female rice weeders: a population-based case-referent study. Scand J Work Environ Health 1986; 13:9-17. 58. Sackett DL. Bias in analytic research. J Chron Dis 1979; 32: 63. 59. El Zayadi A, Khalil A, El Samny N, Hamza MR, Selim O. Hepatic angiosarcoma among Egptian farmers exposed to pesticides. Hepatogastroenterology 1986; 33: 148-150. 60. Smith AH, Fisher DO, Giles HJ, Pearce N. The New Zealand soft tissue sarcoma case-control study: interview findings concerning phenoxy acetic acid exposure. Chemosphere 1983; 12: 565-571. 61. Smith AH. Soft tissue sarcoma and exposure to phenoxy-herbicides and chlorophenols in New Zealand. J Natl Cancer Inst 1984; 73 (5): 1111-1117. 62. Hoar SK, Blair AA, Holmes FF, el al. Agricultural herbicide use and risk of lymphoma and soft-tissue sarcoma. JAMA 1986; 256(9): 1141-1147. 63. Wiklund K, Holm L. Soft tissue sarcoma risk in Swedish agricultural and forestry workers. J Natl Cancer Inst 1986; 76 (2): 230-234. 64. Puntoni R, Merlo F, Fini A, Meazza L, Santi L. Soft tissue sarcomas in Seveso. Lancet 1986; ii (8585): 525. 65. G o ld sm ith J R , G u id o tti T L . E n v iro n m en tal factors in the epidem iology of lymphosarcoma. Pathol Annu 1977; 12:411-425. 66. DeKraay WH. Pesticides and lymphoma in Iowa. J Iowa Med Soc 1978; 68 (2): 50-53. Public Health Reviews 1989/90; 17 36 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides 67. Burmeister LF. Cancer mortality in Iowa farmers 1971-8. J Natl Cancer Inst 1981; 66 (3): 461-464. 68. Burmeister LF. Selected cancer mortality and farm practices in Iowa. Am J Epidemiol 1983; 118(1): 72-77. 69. Hardell L. Malignant lymphoma of histiocytic type and exposure to phenoxyacetic acids or chlorophenols. Lancet 1979; i: S5-S6. 70. Hardell L, Eriksson M, Lenner P, Lundgren E. Malignant lymphoma and exposure to chemicals, especially organic solvents, chlorophenols and phenoxy acids: a case-control study. Br J Cancer 1981; 43: 169-176. 71. Cantor KP. Farming and mortality from non-Hodgkin's lymphoma: A case-control study. Int J Cancer 1982; 29: 239-247. 72. Saftlas AF, Blair A, Cantor KP, Hanrahan L, Anderson HA. Cancer and other causes of death among Wisconsin farmers. Am J Industr Med 1987; 11: 119-129. 73. Bishop CM, Jones AH. Non-Hodgkin's lymphoma of the scalp in workers exposed to dioxins. Lancet 1981; ii (8242): 369. 74. Olsson H, Brandt L. Non-Hodgkin's lymphoma of the skin and occupational exposure to herbicides. Lancet 1981; ii: 379. 73. Pearce NE, Smith AH, Fisher DO. Malignant lymphoma and multiple myeloma linked with agricultural occupations in a New Zealand Cancer Registry-based study. Am J Epidemiol 1983; 121 (2): 223-237. 76. Pearce NE, Smith AH, Howard JK, Sheppard RA, Giles HJ, Teague CA. Non-Hodgkin's lymphoma and exposure to phenoxy-herbicides, chloro phenols, fencing work and meat works employment: a case-control study. Br J Industr Med 1986; 43: 75-83. 77. Pearce NE, Sheppard RA, Smith AH, Teague CA. Non-Hodgkin's lymphoma and farming: An expanded case-control study. Int J Cancer 1987;39:155-161. 78. Schumacher MC. Farming occupations and mortality from non-Hodgkin's lymphoma in Utah. J Occup Med 1985; 27 (8): 580-584. 79. Milham S, Hesser JE. Hodgkin's disease in woodworkers. Lancet 1967; ii: 136-137. 80. Grufferman S, Duong T, Cole P. Occupation and Hodgkin's Disease; brief communication. J Natl Cancer Inst 1976; 57 (5): 1193-1195. 81. Alavanja MCR, Rush GA, Stewart P, Blair AJ. Proportionate mortality study of workers in the grain industry. J Natl Cancer Inst 1987; 78 (2): 247-252. 82. Olsen JH , Jensen OM. Occupation and risk of cancer in Denmark: An analysis of 93,810 cancer cases, 1970-79. Scand J Work Environ Health 1987; 13 (suppl 1): 1-91. 83. Wiklund K, Holm L. Trends in cancer risks among Swedish agricultural workers. J Natl Cancer Inst 1986; 77 (3): 657-664. Public Health Reviews 1989/90:17 Kelly S, Guidoni TL. Phenoxyacetic acid herbicides 37 84. Robb-Smith AHT. US National Cancer Institute working formulation of non-Hodgkin's lymphomas for clinical use. Lancet 1982; ii: 432-434. 83. Olsson H, Brandt L. Risk of non-Hodgkin's lymphoma among men occupationally exposed to organic solvents. Scand J Work Environ Health 1988; 14:246-251. 86. Guidotti TL, Binder S, Stratton JW , et al. Clove cigarettes: Development of fad and evidence for health effects. Curr Top Pulm Pharmacol Toxicol 1987; 2: 1-23. 87. Bond GG, Bonder KM, Cook RR. Phenoxy herbicides and cancer: Insufficient epidemiologic evidence for a causal relationship. Fund Appl Toxicol 1989; 12(1): 172-188. 88. Lilienfeld DE, Gallo MA. 2,4-D, 2,4,5-T, and 2,3,7,8-TCDD: An overview. Epidemiol Rev 1989; 11:28-38. 89. Mullison WR. An interim report summarizing 2,4-D toxicological research sponsored by the industry task force on 2,4-D research data and a brief review of 2,4-D environmental effects. The Technical and Toxicology Committees of the Industry Task Force on 2,4-D Research Data, July 1986. Public Health Reviews 1989/90; 17 Moa A H30W cusa Noiva ai sxiaiHxa l...- > i- ' THE HEALTH EFFECTS OF "AGENT ORANGE" AND POLYCHLORINATED DIOXIN CONTAMINANTS Technical Report Prepared by the Council on Scientific Affairs' Advisory Panel on Toxic Substances John R . Beljan, MD Nelson S. Irey, MD Wendell W. K ilgore, PhD Kazuo Kimura, PhD, MD Raymond R . Suskind, MD Jaroslav J. Vostal, MD, PhD R.H. Wheater, MS, Secretary D ept, of Environmental, Public and Occupational Health American Medical Association Chicago, Illinois 1 7 1982 October 1, 1981 F-453 H33 S,C-f_-Snvlroriaental Health U-:-.? --O-y of Washington Soa ft ji P Wash. 98195 193S0  Copyright 1981 American Medical Association 535 North Dearborn Street Chicago, Illinois 60610 All rights reserved Additional copies may be purchased from: Order Department OP-126 American Medical Association P.O. Box 821 Monroe Wisconsin 53566 NEA:81-634-126:11/812.5M i9351 CONTENTS P r e f a c e ............................................................................................................................................................... Introductory........................................................................................................................................................ 1 What is "Agent O range"?.................................................................................................................... 1 What is T C D D ? .....................................................................................................................................1 What Are the Current Social and M edical C on cern s.......................................................................... 3 T oxicological Evidence of a H ealth H azard........................................................................................... 5 T oxicity, Anim al .................................................................................................................................. 5 Laboratory Tests of T o x ic ity .............................................................................................. 5 D ioxin (T C D D )..........................................................................................................................5 P ic lo r a m ..................................................................................................................................... 8 T oxicity, H um an..................................................................................................................................... 8 2 ,4 - D ............................................................................................................................................. 8 2 ,4 ,5 - T ........................................................................................................................................9 S i l v e x ........................................................................................................................................ 10 T C D D ........................................................................................................................................ 10 M u ta g e n ic ity .............................................................................................................11 O n c o g e n ic ity .............................................................................................................12 T e r a to g e n ic ity ..........................................................................................................12 Reproduction and F ertility , A n im a ls ............................................................. 13 R eproductive E ffec ts in H u m a n s ..................................................................... 14 H ealth Surveys o f th e Exposed P op u lation s......................................................................................... 15 Of Monsanto E m p lo y e e s .................................................................................................................. 16 Of Dow C hem ical E m p loyees..........................................................................................................16 Of Seveso In h a b ita n ts........................................................................................................................17 Of BASF Em ployees; Ludwigshafen, G erm an y........................................................................ 17 Of NV Philips Em ployees; Netherlands ....................................................................................... 18 Of Swedish R ailw ay W orkers..........................................................................................................18 Of Other Swedish W orkers............................................................................................................... 18 O f Finnish R ailw ay and Forestry W orkers.................................................................................18 Of Dow /M onsanto C ohorts, S oft-tissu e S a r c o m a s ................................................................ 18 Of Other Industrial W o r k e r s ..........................................................................................................19 Of Vietnam War V e te r a n s ................................................................................................................ 21 Of "Dedrumming" P e r s o n n e l..........................................................................................................21 Environm ental F ate o f 2 ,4 -D /2 ,4 ,5 -T and T C D D ..............................................................................21 2,4,-D and 2 ,4 ,5 - T ............................................................................................................................. 21 T C D D ...................................................................................................................................................... 21 C ritical R eview s o f Prior S tu d ies............................................................................................................ 23 Of EPA's A lsea S tu d ie s .....................................................................................................................23 By Oregon State U n iv e r sity .............................................................................................. 23 By FIFRA's S cien tific Advisory P a n e l........................................................................... 24 By O t h e r s ................................................................................................................................24 Current and Proposed S t u d i e s .................................................................................................................. 24 International A gency for R esearch on Cancer ( I A R C )....................................................... 24 US Air Force E pidem iological Study............................................................................................25 P roject Ranch Hand P erson n el......................................................................................... 25 Armed F orces In stitute of P a th o lo g y ......................................................................................... 25 N ational Cancer Institute (N C I).................................................................................................... 25 N ational In stitute for O ccupational S a fety and H e a lth ....................................................... 25 US D epartm ent o f A gricu ltu re....................................................................................................... 25 i 19352 Veterans A d m in istration ................................................................................................................. 25 O t h e r .....................................................................................................................................................26 S u m m ary.......................................................................................................................................................... 27 R e c o m m e n d a tio n s....................................................................................................................................... 28 R e f e r e n c e s ..................................................................................................................................................... 29 Appendix Table 1 -- Occupational and Industrial A ccidental Exposures to TCDD in Plants Producing 2,4,5-trichlorophenol and/or 2 ,4 ,5 - T ......................... A1 PREFACE During the la tter stages of the US' involvement in Vietnam, herbicidal mixtures of 2,4,-D and 2,4,5-T--otherwise identified by the m ilitary as Agent Orange--were sprayed over certain areas of the Vietnamese jungle for the express purpose of defoliating the jungle and destroying one of the enemy's means of concealment. Similar spray programs have been used on occasion in the US as a means of forestry management. For the past 30 years, mixtures of 2,4-D and 2,4,5-T have been used extensively and uneventfully by the homeowner and farm er for ridding lawns and agricultural acreage of unwanted broadleaf vegetation. In any event, large numbers of persons have beenunintentionally exposed not only to varying amounts of 2,4,-D and/or 2,4,5-T, but also to a contaminant in the mixture--ie, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)--which may well be one of the most toxic substances known to man. Over the years, there have also been a number of industrial incidents wherein workers, as well as civilian populations, have been subjected to accidental exposure to these compounds. There are now pending a number of litigious actions and government regulatory responses that are based upon alleged adverse health effects from such exposures. The purpose of this report then is to review the medical evidence regarding the toxicity and long-term health effects of TCDD. In spite of the voluminous data on the biological effects of the phenoxy-type pesticides and their associated chlorinated dioxins, which oftentim es co-exist as contaminants of the pesticide formulations, there is still very little substantive evidence for many of the alleged claims that have been made against these compounds. The m ost serious of these allegations assert that Agent Orange, or compounds of a like nature, have caused malignant tumors, spontaneous abortions and birth defects. Although data from studies on experimental animals tend to support some of these claims, it is not certain that the animal data are extrapolatable to man. No laboratory animal can fully substitute for man; we must, therefore, depend on the results of ongoing epidemiologic studies on persons who are known to have been exposed. Thus far, one of the most extensive "human experimentsf is being conducted on the residents of Seveso, Italy. The whole story may not be known for another five years; but preliminarily, it appears that the dermatological effects were reversible; there has been no progression of neurological damage; differences in immune response betw een exposed and nonexposed persons have not been significant; and there has been no notable increase in infectious diseases. There may never be an entirely clear picture from Seveso of the e ffe c t of TCDD on reproduction or the fetus, because it is suspected that a number of women elected to have therapeutic abortions. No major pathology, however, was reported by the area physicians. In the case of industrial exposures, the epidemiological studies on these populations may be more fruitful. First, the workplace generally will have an unexposed control group that can be closely matched to the exposed group. Secondly, in many cases there will be medical records on the individuals and measurements, or close approximations, of the exposure concentrations. By being able to identify and locate earlier workers--by means of retirem ent, union or other employment records--the task of medical surveillance over extended periods of tim e will facilitate the discovery of long-term health effects from the exposure. There is a vast amount of literature available on the adverse health e ffe cts of 2,4-D, 2,4,5-T and TCDD; but one must view much of this information objectively, for iii O-' <J data--especially from the early years--may be devoid of adequate controls. Hopefully, the internationally coordinated studies of existing cohorts and the opportunity to share data from both past and future exposures, as proposed by the International Agency for Research on Cancer, will produce an irrefutable stand on the long-term consequences of exposure to TCDD. m iv O- fcv O' o HEALTH EFFECTS OF "AGENT ORANGE" AND POLYCHLORINATED DIOXIN CONTAMINANTS INTRODUCTORY What is "Agent Orange"? A gent Orange--or H erbicide Orange--was a label given by the US m ilitary forces to a 50:50 mixture of the n-butyl esters of 2,4-dichlorophenoxyacetic acid (2,4-D) and 2 .4 .5 - trichlorophenoxyacetic acid (2,4,5-T), together with a minor amount of the free acid 2,4,5-T (1% of th e total mixture) and varying am ounts (from 0.02 to 15 ppm, or a w eighted mean of 1.98 ppm) of the contam inant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). - 2 .4.5- T was an active ingredient in Agents Green, Pink, Purple and Orange, used by the m ilitary forces from 1962 thru 1970 to defoliate certain jungle areas of Vietnam. A pproxim ately 276,000 gal of Green, Pink and Purple--containing 64 kg (143 lb) of TCDD--w ere sprayed over South Vietnam prior to 1965 before th ey w ere replaced by A gent Orange. Another form ulation, H erbicide Orange II, sim ilar to A gent Orange (except that the isooctyl ester of 2,4,5-T was substituted for the n-butyl ester of 2 .4 .5 - T), was also sent to Vietnam; there is no available record to show how much of this form ulation was actually used or later disposed of (Young et al, 1978). According to Young et al (1978), about 25 m illion kilogram s (55,940,150 lb) o f 2,4-D and 20 m illion kilogram s (44,232,600 lb) o f 2,4,5-T --for a total o f 166 kg (368 lb) of TCDD--w ere distributed over approxim ately 1.4 m illion h ectares (3.5 m illion acres) o f Vietnam from January 1962 to February 1971. It is significant to n ote that alm ost 40% o f the entire amount o f TCDD that was released in Vietnam--or 64 kg (143 lb)--was contained in Agents Green, Pink and Purple, which m aterials w ere used on about 36.4 thousand hectares (90 thousand acres) from 1962 thru 1964; this was during a tim e when very few American troops were com m itted to the field. In addition to Agent Orange, other spray defoliants used in Vietnam included H erbicide White (about 5 m illion gallons) and Herbicide Blue (about 5,200 gallons). The active ingredient of Herbicide White was picloram (the triisopropanolamine salt of 4-am ino-3,5,6-trichloropicolinic acid), which is very persistent in the environment and highly carcinogenic in rats and m ice (Reuber, 1981). Herbicide Blue contained sodium ca co d y la te* (26%) and cacodylic acid (5%) (Young et al, 1978). Though the long-term e ff e c ts in humans o f either or both of the compounds is uncertain--on an acute basis, both sure rela tiv ely non -toxic--picloram and cacod ylic acid should be considered along with the rest of the chem ical toxicants encountered in Vietnam. What is TCDD? TCDD--or 2,3,7,8-tetrachlorodib en zo-p-dioxin--is only one of 75 various chlorodioxins; there are 22 positioned isomers of tetrachlorodibenzo-p-dioxin alone. The 2,3,7,8-tetrachloro isom er is considered to be the m ost toxic, while the 2,7-dichloro- and the octachloro- dioxins are the least toxic. It may form as a by-product of the synthesis of 2 .4 .5 - trichlorophenol (TCP), a precursor of 2,4,5-T, when 1,2,4,5-tetrachlorobenzene is subjected to alkaline hydrolysis at elevated tem perature and pressure. If the reaction * Sodium cacodylate is the sodium salt of cacodylic acid (hydroxydimethylarsine oxide). i . G G OG 6 tem perature is allow ed to rise above 180C , the sodium -2-hydroethoxide (a combination of the ethylene glycol solvent and caustic soda) decomposes exotherm ically and promotes the dimerization of sodium trichlorophenate to TCDD. The presence of TCDD as a contam inant of 2,4,5-trichlorophenol was disclosed by Kimmig et al (1957) and Schulz (1957) when workmen involved in the m anufacture of TCP developed chloracne. Halogenated dibenzodioxins are also apt to be created when certain chlorinated compounds in industrial and municipal w astes are incinerated (Anon, 1979; Bumb et al, 1980; Rappe et al, 1978; and Rawls, 1979), and they may be released by the burning of vegetation that has been treated with the chlorophenoxy herbicides and defoliants (Buu-Hoi et al, 1971). There is som e argument whether such release w ill occur under field conditions, especially if burning takes place one or more m onths after spraying (Norris, 1978). According to a recent report from the American Society of Mechanical Engineers (1981), however, municipal w aste incineration should pose no greater health risk from the polychlorinated dibenzo-p-dioxins than other forms of w aste disposed, ie, landfill. The report estim ates that the cancer risk will be increased by only one person in one million of the population. TCDD as an impurity o f 2,4,5-T was first revealed about 1959 by Tom ita et al (1959). It, as w ell as other, related dioxins, may also contam inate pentachlorophenol; hexachlorobenzene; 2,4,-dichlorophenyl-p-nitrophenyl ether; silvex; ronnel (fenchlorphos); erbon; trichloronate; hexachlorophene; 2,4,4'-trichloro-2'-hydroxydiphenyl ether and the polychlorinated biphenyls (PCB's) (Kimbrough, 1979). About 20 years ago, com m ercially available 2 ,4 ,5-T contained anywhere from 1 to 70 ppm of TCDD. When the industry becam e aware of the contam inant's existen ce and toxicity, production operations w ere m onitored and adtered to reduce the level. Manufactinring operations are now able to control the amount of TCDD in com m ercial 2,4,5-T form ulations to less than 0.01 ppm (with occasional batches as high as 0.05 ppm), a lev e l believed to be non-hazardous to humans and other organism s. D ata relatin g to an acceptable maximum level are presently under regulatory review by FIFRA (Federal Insecticide, Fungicide and Rodenticide Act); the recom m ended maximum concentration is now placed at 0.1 ppm (Council for Agricultural Science and Technology, 1978). TCDD was first noted for its ter a to g e n ic ity and fe to to x ic ity by C ourtney et ad (1970), who w ere testin g the biological a c tiv ity o f 2 ,4 ,5-T in m ice arnd ra ts--the sam ple they were using was later found to contain 30 ppm of 2,3,7,8-tetrachlorodibenzo-p-dioxin. The in cidence o f c le ft padate warn greater in both the C57BL/6 and AKR m ouse strains, while the C57BL/6 mouse and the rat had a higher incidence of cystic kidney. All doses given the rat led to gastrointestinal hemorrhage in the fetuses; also, the increased ratio o f fetad liver w eight to body w eight in the mouse suggested that TCDD was feto to x ic in this particular species. An early major review o f this subject wais the "Perspective on C hlorinated Dibenzo dioxins and Dibenzofurams" (Moore, 1973); unfortunately, little inform ation wais given on biological e ffe c ts of TCDD in man. Evidence for long-term adverse health e ffe c ts to man was still inconclusive in the extensive reviews by Young et al (1978) and Bovey et al (1980) and the International A gency for R esearch on Cancer (1977). 2 -- O' O O1 7 WHAT ARE THE CURRENT SOCIAL AND MEDICAL CONCERNS? Phenoxy herbicides cam e into com m ercial use as agricultural herbicides in the mid-40's with the introduction of 2,4-D , which was follow ed by 2 ,4 ,5-T in 1948. Both were used exten sively and uneventfu lly in the control o f woody and broadleaf veg eta tio n in croplands, forests, rights-of-w ay and turf until 1969, when public concern mounted over the human health hazards o f 2 ,4 ,5-T and/or TCDD. Consequently, in la te 1970, the Secretary of Agriculture suspended those uses of the herbicide where humans might encounter g rea test exposure to the TCDD. In 1971, the Environmental P rotection Agency (EPA) initiated can cellation proceedings against certain other uses of 2 ,4 ,5-T. EPA then withdrew its cancellation notice in 1974 and initiated a plan for the developm ent o f data that would be needed for their decision-m aking process. In April 1978, EPA issued a n o tice of R ebuttable Presum ption Against R egistration (RPAR)*. In March of 1979, the RPAR process was partially interrupted by an em ergency suspension to ban the use of 2 ,4 ,5-T in pastures, forests and rights-of-w ay home gardens, aquatic weeds, ditch banks and ornamental turf, excluding use on rangeland and rice; the suspension was prompted by reports of spontaneous abortions in humans in A lsea, Oregon, which abortions were allegedly caused by these herbicides (American Farm Bureau Federation, 1979 and US Environmental Protection Agency, 1979). Inasmuch as the EPA believed that silvex (or 2-2(2,4,5-trichlorophenoxy)propionic acid) was c lo se ly related to 2,4,5,-T and also apt to be contam inated with TCDD, the Agency suspended silv ex as w ell. Meanwhile, a number of Vietnam war veterans were reported to have had various m edical complaints (for instance, weight loss, liver dam age, recurrent skin rashes, deform ed offspring, stillbirths, cancer, sterility, personality changes and "other illnesses"), which they ascribed to their exposure to Agent Orange som e 8 to 18 years earlier. For exam ple, a helicopter pilot who had served in Vietnam , and who had been discharged in apparent excellent health, began to experience symptoms of bowel irregularity, sleeplessness, nausea, headaches, night sw eats and a general m alaise. He was diagnosed as "psychotic", then found to have terminal cancer of the colon, which m etastasized to the lym phatics. A class action suit was instituted in 1979 against five of the herbicide m anufacturers--Dow C hem ical, M onsanto, H ercules, Diamond Shamrock and ThompsonHayward--on his behalf and extended to "all Am erican servicem en whose health has been dam aged because o f co n ta ct with Agent Orange." The com panies in turn filed a thirdparty action against the US government, passing the responsibility for alleged harm to the government for its negligent misuse of the chem icals. A class action suit was likew ise filed against the Veterans Adm inistration and the US Department of D efense in January 1981. At the last count, som e 2.4 million Vietnam veteran s, including 1,200 in O peration Ranch Hand helicopter crew s, and 200 civilians who w ere involved in "dedrumming" and destroying about 40,000 55-gallon drums of surplus Agent Orange were presumably exposed. Other persons were also claim ed to be within 1/3 mile of the spray areas. The news media have stated that the potential number of "plaintiffs" will exceed 2.5 million people, from which about 40,000 veterans may possibly becom e sick or die as * RPAR is an intensive b en efit/risk review o f a currently registered p e sticid e to determ ine if new health e ffe c ts data and risk criteria can justify its continued registration and unrestricted use. 3 G"358 a result of Agent Orange. The reports go on to suggest that at least 2,000 children will su ffer "catastrophic" birth d e fe c ts. More them 1,250 law yers from 150 law firm s in the US are involved in the litigation to date (Elson, 1980). Victor Yannacone, the New York attorney who filed the first suit on behalf of the abovementioned helicopter pilot, has alleged that birth d efects, in addition to those already listed, include: urinary and genital tract deform ities, missing limbs, clubbed fe e t and mislinked blood vessels (Elson, 1980). Agent Orange litigation has not been confined to the m ilitary, for a number of women w ere believed to be at risk of m iscarriage a fter 13 spontaneous abortions occurred in Oregon's Alsea basin area. In this particular in stance, silvex was em ployed by the US Forest Service in their spray programs for this area. See the section below, "critical reviews of...A lsea studies," for additional discussion of the incident. Inasmuch as the phenoxy herbicides have been manufactured and used for the past 40 odd years, it is reasonable to expect that the above cases are not the only instances of frank exposure to these chem icals or their contam inants. The first of several accidental releases of TCDD, and other dioxins, attending the m anufacture of 2,4,5-trichlorophenol (TCP) or 2,4,5-T occurred in 1949. A consequence o f this industrial incident w as a su it--asking $1.7 billion--that has been filed against the Monsanto C hem ical Co. on behalf of 54 former Monsanto em ployees and representatives of eight deceased persons who worked in the Nitro, West Virginia plant. From 1948 to 1969, this plant manufactured 2,4,5-T; however, its major products w ere chem ical compounds for the rubber industry (eg, p-aminobiphenyl and 4-nitrobiphenyl, both of which are now known to induce bladder cancers in humans). The men are said to have suffered from kidney diseases, leukem ia and a variety of other ailm ents (Anon, 1981). To date, there are at lea st 579 workers who are known to have been exposed to TCDD, including 156 em ployees in the ICMESA (Industrie Chimiche Meda Societa Anonima) plant near Seveso, Italy. In July of 1976, TCDD w a s accid en tally released from the ICMESA trichlorophenol synthesis plant when a safety disk in a steam -heated reaction vessel ruptured. The plume of reactor contents, including TCDD, rose into the air and fell in a cone-shaped pattern about a m ile long and h alf-m ile w ide--an area o f about 2.02 km^ (about 500 acres or 0.8 sq mi). It has been estim ated that from 650 to 1,700 grams of TCDD were dispersed into the environment. Some of the environmental samples of contaminated grass contained as much as 15,840 ug/m , or more than 15 ppm, of TCDD. On the basis of soil sam ples, the territory was divided into three zones according to their relative degree of co n tam in ation : zone A--about 0.73 km^ (180 acres or 0.3 sq mi) with a population of 736 persons--ranged from a low that was n on -d etectab le to a high of 5,477 ug/m , or an average lev el of 235.5 ug/ni ; zone B--2.25 km 6 (556 acres or 0.9 sq mi) and 4,699 people--had an average of 3 ug/m ; whiLe zone R o f 12 km^ (2,954 acres or 4.6 sq mi) and 31,800 people was sporadically contam inated with an average of nbout 0.5 ug/rn (the low was undetectable and the high was less than 5 ug/m i . Serving as controls were some 180,000 persons who had been living under similar conditions outside the contaminated area (Homberger et al, 1979). 4 19359 Some o f the persons in the area becam e ill im m ediately, while others experienced chem ical bums of the skin, stom ach pains and internal hemorrhage within a few days of the accident. With the death of the animals and vegetation , the inhabitants were evacuated and the area was sealed o ff by o ffic ia ls, who fe lt that it would be ten or more years before the contam inant would have decom posed to a level that would be su fficien tly safe for resettlem ent (Rawls et al, 1976). Approximately 511 of the 736 evacuees from zone A were allowed to return to their living quarters by May 1977 (Homberger et al, 1979). From Vietnam have com e the observations of Dr. Ton-That Tung (Tung, 1971, 1973), that liver cancer incidence in Vietnam jumped from the eighth to the second most common form of cancer six years after the introduction of Agent Orange into that country in 1963. Abortions and congenital birth d efects were also said to be more frequent. Ford et al (1978) offered one possible explanation for this alleged rise in cancer incidence: that aflatoxin-contam inated cereal grains may have becom e more prevalent during the war. Unfortunately, these statem ents of Tung and others cannot be scientifically verified, and without the cooperation of the V ietnam ese governm ent, there is no likelihood of a long-term follow-up. One source of information about the consequences of TCDD on the Vietnamese people is the report of the National Academy of Sciences (1974). TOXICOLOGICAL EVIDENCE OF A HEALTH HAZARD T oxicity, Animal --Laboratory T ests o f T oxicity One of the common failings of animal toxicity assessm ents is the use of test doses that are much higher than those that may be encountered by humans. Such massive doses are intended to shorten the tim e for a response or, in the case of a carcinogenicity test, to increase the probable incidence of tumors within a sample of animals. Then too, toxic response is determ ined by the pharm acokinetics, pharm acodynam ics and adaptive considerations of the particular species of animal, and all of these factors d iffer--som etim es markedly--from one sp ecies to another. Animals w ith m etabolic rates that are higher than man's (eg, rats and mice) often w ill be relatively more sensitive than man. Thus, extrapolation of the toxicologic data from one animal species or pure-bred strain to another, including man, can be fraught with error and be m isleading. For exam ple, as Purchase (1980) points out in his literature review of 250 compounds that had been tested for carcinogenicity: 43 of these chem icals varied in their e ffe c ts on the rat and mouse--21 (8%) w ere carcinogenic only in the m ouse, while 17 (7%) w ere carcino genic only in the rat--and 5 (2%) varied among other sp ecies. Some 38% o f the 250 compounds w ere non-carcinogenic in both rat and m ouse and 44% w ere carcinogenic in both of these species. --D ioxin (TCDD) Two of the more pronounced biological effe c ts of many of the chlorinated dioxins are their tendency to cause acne (especially, in the rabbit and monkey) and the accum ulation o f fluid (ascites) in the pericardium and peritoneal cavity of chicks. This so-called "chick edema" disease was noted in la te 1957 (Sanger et al, 1958) when m illions of broilers in the eastern and m idwestern US died as a result of an apparent feed additive. The contam inant, or chick edem a factor, was finally traced to tallow s that had been recovered from hide stripping operations employing pentachlorophenol as a preservative 5 (Schm ittle et al, 1958); the contaminant was identified as a mixture of 2,3,7-trichlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzo-p-dioxin (Tomita et al, 1959). In general, chronic exposures lead to degeneration o f the liver and thymus--one sen sitive index of exposure is atrophy o f the thymus--porphyria cutanea tarda, altered levels in serum enzym es and w eight loss. Several animal sp ecies are also a ffe c te d by its fetotoxicity (National Research Council, 1977; Neubert et al, 1972, 1973). Delayed e ffe c ts from TCDD appear as pathological changes in the liver. Major organs to be a ffected are the liver, blood forming organs and reticuloendothelial system . Other than the usual progressive weight loss, the first clinical signs of toxicity, for the monkey at least, are marked by alopecia, facial edem a and a dry, scaly derm atitis over the rest of the body. . , TCDD by itself is extrem ely toxic; along with botulinus bacillus toxin and aflatoxin, it is one o f the m ost lethal organic ch em ical compounds known to man. In term s o f the acu te oral dose, or LDcg, its to x icity ranges from 0.6 ug/kg of body wt for the guinea pig (the m ost sensitive o f experim ental animals) to 115 ug/kg o f body wt for the New Zealand albino rabbit (Schwetz et al, 1973); single oral doses of 20 ug to 50 ug/kg/body wt have produced fatal liver necrosis in the rabbit (Schulz, 1957). Among primates, the LD5 o(raD for the Rhesus monkey is less than 70 ug/kg of body wt (McConnell et al, 1978). As for the rat, its oral LDgg (male) is 22 ug/kg vs the fem ale LDTM at 45 ug/kg (Schwetz et al, 1973), which is about 10,000 tim es less than the oral LDTM for 2 ,4,5-T. The feto to x ic dose, is about 1/400 of the above LDgg (or about 0.11 ug/kg) (O ffice o f Science and Technology, Panel on Herbicides, 1971). Death is probably due to h ep atic-cell necrosis. Liver lesions caused by TCDD in the rat were subjected to a morphological study (Jones et al, 1974). Parenchymal cell necrosis in the centrilobular zone with mononuclear cell infiltration was noted after the first week of exposure. By the second week there was dilatation of the sinusoids in the centrilobular zone, more severe inflam m ation and increased necrosis in the parenchym al cells; loss was not made up by hyperplasia of the live cells. Inflammation regressed after 3 to 6 weeks, but focal areas of necrosis remained around central veins and mural fibrosis of the central vein developed. No ill effects were discerned in young adult Sprague-Dawley rats that were fed either 0.001 or 0.01 ug of TCDD/kg 5 days/w k for 13 w eeks (Kociba et al, 1976); the slight increase in relative liver weight at the 0.01 level was considered to be a physiological adaptation rather than a toxic response. There were a number of changes in tissue and organ function at higher levels. The 1.0 ug/kg/day dose, for exam ple, caused som e m ortality (which was seem ingly more marked in fem ales), in activity, decreased food consumption and body w eight, icterus, elevated serum bilirubin and alkaline phosphatase, liver im pairm ent, lymphoid depletion of the thymus and other lymphoid organs, increased excretion of urinary porphyrins and delta-am inolevulinic acid, together with minor hem atopoietic changes (bone marrow production of red blood cells was not markedly inhibited). It was unclear whether the suppressed function of the reproductive organs would be a direct action of TCDD or the result of the debilitated condition of the anim al. In general, the findings o f this study supported prior studies of shorter duration. A 2-year chronic to x icity and oncogenicity study of TCDD was conducted on m ale and fem ale rats at three different dose levels: 0.1, 0.01 and 0.001 micrograms per kg per day (Kociba et ail, 1978). At the 0.1 ug level, the fat and liver of the fem ales contained 8,100 and 24,000 ppt of TCDD, respectively, at the end of two years. Some tumors (viz, 6 19381 hepatocellular carcinom as of the fem ales and squamous cell carcinom as of the lung, hard palate/nasal turbinates or tongue) were induced, while the strain-related tumors (ie, of the pituitary, uterus, mammary glands, pancreas and adrenals) were suppressed. Fewer aged m ales had chronic renal disease, and it was less severe . Other toxic signs included increased m ortality; decreased gain in body weight; depression of certain hem atologic parameters; increased urinary excretion of porphyrins and delta-am inolevulinic acid (ALA); increased serum a ctiv ities of alkaline phosphatase (AP), gam m a-glutam yl transferase (GGT) and serum glutam ic pyruvic transam inase (SGPT); as w ell as morphological changes in the hepatic, lymphoid, respiratory and vascular tissues. At the 0.01 ug lev el, there was slight poisoning: urinary excretion o f porphyrins (in fem ales) was increased, hepatocellular nodules were more numerous and there was an increased incidence of focal alveolar hyperplasia. No significant toxic effe c ts were noted at 0.001 ug. No increase in tumors was evident at either of the lower lev els. It should be noted that most of the above work used the Sprague-Dawley rat strain, a strain that characteristically has a high incidence of spontaneous tumors. As little as 5 ppt to 5 ppb in the rat d iet are said to induce various neoplasm s (Van M iller et al, 1977); lung and/or liver tumors have developed at the 1 or 5 ppb level. TCDD is a very potent inducer of aryl hydrocarbon hydroxylase (AHH), as w ell as the m ixed-function oxidases and the cytochrom e P^-450 (P-448) enzym es of the liver, lung, placen ta and kidney o f the m ouse and rat. In addition to alterin g norm al enzym e a c tiv ity , it may even p oten tiate the harm ful action o f other toxins or render an otherwise innocuous agent toxic. It is 30,000 tim es more powerful than 3-m ethylcholanthrene in inducing the activity of AHH in certain laboratory anim als (Poland et al, 1974). Mouse liver fractions, which had been pretreated with TCDD and subjected to 3-m ethylcholanthrene, showed a significantly greater mutagenic e ffe c t than without the pretreatm ent. A similar experiment using benzo(a)pyrene as the polycyclic hydrocarbon was unremarkable (Felton et al, 1975); one is left to wonder whether other organic compounds are likewise modified mutagenically by TCDD. The tissue-specificity of TCDD's induction property appears as part of the work of Berry et al (1977), wherein sam ples of liver, lung and placental tissue from TC DD -treated fem ale Sprague-Dawley rats were compared with tissue of untreated controls for the covalent binding capacity of benzo (a)pyrene to DNA. Although TCDD appears to have no profound direct toxicity to the adult kidney, its property as an enzym e inducer conceivably may convert harm less substances to nephrotoxins (Fowler et al, 1977). ^ C -la b eled TCDD was used to determ ine the tissue distribution and excretion of TCDD in the rat (Allen et al, 1975; Fries et al, 1975; Piper et al, 1973; and Rose et al, 1976). Excretion was largely via the fe c e s (over 50%), w ith sm all am ounts being elim inated in the urine and expired air. The compound accum ulated, for the m ost part, in the fat and liver with lesser amounts (som e 20% to 30% less), by decreasing order, in the spleen, bone, heart, lungs, testes, kidney, m uscle, skin, pancreas, brain, stom ach and adrenals. Accum ulation reached a steady sta te within 13 w eeks if adm inistration was continued for several months. TCDD was m etabolized only slightly by the rat. Most of it seem ed to rem ain unm etabolized in the liver--esp, in the m icrosom al fraction--o f the rat, m ouse, and monkey (M atthiaschk, 1978) and guinea pig (G asiewicz et al, 1979), which agrees with the. findings from an intraperitoneal injection of male m ice with an LDgg dose o f JH-TCDD (Vinopal et al, 1973). Tulp et al (1978) have suggested that the m etabolism of TCDD is unlikely to occur inasmuch as its m olecular structure hinders formation of the interm ediate 2,3-epoxide, which is the main route for other polychlorinated dibenzodioxins. More recent radiolabel studies by Ram sey et al (1979) have indicated 7 19362 that TCDD m etabolized slow ly in the rat liver to a variety of unidentified polar compounds, whose excretion rate in the bile was comparable to the authors' earlier disclosed rate of excretion in the feces. Poiger et al (1979) also have reported on the excretion of polar m etabolites of TCDD in rat bile. The hamster's rate of TCDD m etabolism was greater than that of the guinea pig's or rat's, which may help to explain the relative differen ces in toxicity betw een these species (Olson et al, 1980). The acute LD^q o f TCDD in the fem ale Rhesus monkey, while not determ ined accurately by McConnell et ad (1978), is still probably close to 70 ug/kg. This sp ecies thus appeaurs to be interm ediate in sensitivity betw een the guinea pig (0.6 ug/kg) and the rabbit (115 ug/kg) and more sensitive than the C57BL/6 mouse (284 ug/kg). The first toxic symptoms for the monkey were progressive w eight loss, follow ed by blepharitis--a m ost sen sitive marker for to x ic ity in this anim al--then skin lesions, that are not unlike those experienced by humans, and the loss of fingernails and eyelash es. The authors adso noted less reduction in spleen w eight than for other sp ecies, and the thymus appeared to be com pletely involuted. Increase in liver weight also was less marked than in rodents, and there were few er hepatic lesions. Perhaps this was due to less TCDD being retained in the monkey liver. Adult femade Rhesus m onkeys w ere also fed 500 ppt of TCDD for up to nine m onths (Allen et al, 1977; 1978), during which tim e they ingested about 2-3 ug of TCDD/kg/day. For the first three months there was acne, alopecia and periorbital edem a. Irregular menstruad cy c le s w ere accom panied by decreased 1 7-b eta-estrad iol auid progesterone levels. Early abortions and the inability to conceive occurred at the six-m onth point. D eath wais m ainly due to throm bocytopenia; in five out o f eigh t animads, death was preceded by pancytopenia. Microscopic examinations revealed suppressed hem atopoiesis o f the bone mairrow, hypocellularity o f the lymphoid tissu e, w idespread hem orrhage, cardiac hypertrophy, edem a and dilation, hyperplasia and metaplaisia o f adl ductal tissu e, hypertrophic gastritis w ith ulceration amd dilation o f the biliary sy stem . O ffspring w ere weak and died within a few w eeks, possibly from the additional dioxin in the m aternal m ilk. One test animad that is particularly sen sitive to CNS toxicants is the chicken. Sanderson et ad (1981) o f A ustralia treated both chicken eggs and hatchlings w ith "environmental levels" o f 2 ,4 ,5-T contam inated w ith 0.03 ppm of TCDD. Although the general a c tiv ity o f the animad was u n affected , there was a dose-related increase in jumping and a retardation in visual leaguing. --P icla ra m A review from the National Cancer Institute on the carcinogenicity of this compound (Reuber, 1981) has focused on two NCI studies and two Dow C hem ical Co. studies. The findings revealed that adl neoplamms, including m alignant ones, increased for both m ale and fem ale O sbom e-M endel rats. The malign amt neoplasm s co n sisted o f carcinom as--o f the adrenal, thyroid and pituitary glands--and sarcom as. M ice of both sexes developed neoplasm s o f the sp leen . Male rats adso had chronic renal d isea se, parathyroid hyperplasia and polyarteritis; both rats and m ice displayed testicular atrophy. T oxicity, Human --2,4-D Normadly, 2,4-D is not contam inated with TCDD. It o ften has been used amd te ste d in combination with 2 ,4,5-T, which can contain TCDD; hence, there is m erit to a lim ited discussion of how this compound is known to affect man. . 8 19353 Ingested 2,4-D is absorbed almost entirely from the gastrointestinal tract and clears from the body by a first-order rate process; the h alf-life is about 17.7 hrs. It is excreted in the urine mainly as unm etabolized 2,4-D (82.3%) and the 2,4-D conjugate (about 12.8%). Accumulation, in the event of continued intake of the compound, is unlikely (Sauerhoff, 1977). Three cases of polyneuritis have been attributed to rather m assive percutaneous absorption o f 2,4-D (Goldstein e t al, 1959). In each instance, the sym ptom s w ere progressive, with severe pain, paresthesia and paralysis. All of the victim s still suffered som e neurological damage after about three years from date of onset. No details were given on the com positions of the form ulations (ie, im purities and adjuvants). The lethal, oral dose (based on a 23-yr-old male) is probably over 80 mg/kg; thus, an 80 kg man would require at least 6.4 grams of the compound. A fatal suicide attem pt with an estim ated 6.5 g gave signs of convulsions and nonspecific engorgem ent with blood of the lungs, liver and brain (the tim e from ingestion to death was 18 hrs) (Nielson et al, 1965). In another incident (Berwick, 1970), a man accidentally ingested a 110 m g/kg dose of 2,4-D (isooctyl ester). Besides depression and irritation of the CNS, the victim suffered from m yopathy, GI irritation, nephropathy and to x icity to the liver; how ever, he recovered fully with no signs of peripheral neuropathy. Signs and symptoms of poisoning occur at an oral dose of around 3 to 4 grams (Hayes, 1963). There is one instance of a man who consumed 500 mg of purified 2,4-D each day for 21 days w ithout ill e ff e c ts . In another c a se, 18 iv doses--the last 12 o f which were 800 m illigram s or m ore--w ere delivered to a patient over the course o f 33 days for the treatm ent of coccidioidom ycosis; there were no side effects even after the 18th dose of 2,000 m g. The last injection of 3,600 mg produced acute illness--v iz, unconsciousness, tw itching of som e of the m uscles of the body, depression of the reflexes and involuntary urination--y e t, by the end o f 48 hours, th e subject had com p letely recovered . Myotonia soon after a heavy exposure of animals to 2,4-D is characteristic of a poisonous dose; this sign may also suggest intoxication in man, but it has not yet been reported (Hayes, 1963). The porphyria that has been observed among 2,4-D and 2 ,4 ,5-T workers by Poland et al (1971) was m ost likely due to chlorobenzene im purities, or TCDD. - 2 ,4 ,5-T A ttem pts to assess the health effe c ts of 2,4,5-T from toxicologic data that neither indicate nor account for the presence of TCDD will be virtually without m erit. If a report does not mention chloracne as being one of the clinical signs, then it is very unlikely that TCDD was present in such an amount as to cause organic damage. As part of an experim ent to determ ine the fa te of ingested 2 ,4 ,5-T in man, five m ale volunteers (from 31 to 58 yrs o f age) w ere given a single dose of 5 m g/kg o f body w t of 2 ,4 ,5-T (which contained less that 0.05 ppm of TCDD). There were no discernible clinical signs save for a m etallic taste, which persisted about 1-2 hours. Most of the 2,4,5-T (65%) w ent into the plasm a where it was bound reversibly with protein. Its concentration in plasm a reached a plateau a fter three days in those individuals who ingested the compound daily. Virtually till of the compound was excreted unm etabolized in the urine at a first-order rate (Gehring et al, 1973). 9 19234 2 .4 .5- T causes only a mild spastic condition and d ifficu lty in swallowing (Hayes, 1963). Based on the lack o f known illnesses from occupational exposures, the American Confer ence of G overnm ental Industrial H ygienists (ACGIH) has recom m ended that the threshold lim it value (TLV) for 2,^ ,5-T be se t at 10 m g/m (tim e-w eighted average for an 8-hr working day) or 20 m g/m 0 for a short-term exposure (ie, up to 15 min for no more than 4 tim es per day). --S ilv e x The toxicology of this compound is mentioned here because it was involved in the Alsea dispute, and it too can be contam inated with TCDD. Also known as fenoprop, 2-2(2,4,5-trichlorophenoxy)propionic acid is used on apples to intensify color and prevent fruit drop, to eradicate undesired plants in forests, to control aquatic weeds and to control a variety of weeds in turf and on industrial lands. Often used in combination w ith 2 ,4 ,5-T to enhance the herbicidal e ffe c t of the form ulation on refractory woody plants, this compound was em ployed in US Forestry Service spray programs o f the A lsea, Oregon aurea. Its use w as suspended by th e EPA a fter the announcement of a higher than usual number of spontaneous abortions in the region. The oral LD^g for rats is 650 mg/kg; m ice dosed w ith 15-46 m g/kg/day gave no signs of tumorigenicity (Hayes, 1975). Seven men and one woman ingested 1 mg of silvex /k g of body weight: Biom edical tests showed that ingested silvex was rapidly absorbed; peak blood plasm a concentration occurred within 2 to 4 hours. Elimination was rapid, primarily via the urine; within 24 hours a fter in gestion , 65% had been passed in the urine. The h a lf-life for urinary clearance averages about 16 hours. No clinically dem onstrable adverse e ffe c ts w ere observed (Sauerhoff et al, 1977). --TCDD TCDD is recognized as a potent chloracnegen in the monkey, rabbit (where it appears only on the ear), nude m ouse and man. Chloracne is typ ified by com edones--w ith or w ithout cysts and pustules--in a malar distribution, along with m elanosis and a secondary inflammation; the pre- and post-auricular forms of chloracne are som etim es accom panied by hirsutism in th ese areas. In more severe ca ses its location may becom e generalized. This disorder was first described by Herxheimer in 1899. Still la ter, Bauer (in 1918) and Teleky (in 1927) noted that it was due to derm al co n ta ct w ith certain chlorinated hydrocarbons. Chloracne has now become one of the more common forms of occupational derm atitis. A review on the subject was presented by Schwartz et al (1957) and Crow (1970), and more recently by Taylor (1979). The role of polychlorinated dioxins in chloracne cam e to light when Schulz (1957) and Bauer et al (1961) were investigating the disease among a group of German workers who had been handling technical grade 2 .4 .5 - trichlorophenol. Pure chlorophenol did not elicit in the rabbit the dermal response (ie, hyperkeratosis) of the impure compound (Kimmig et al, 1957); thus, they reasoned that an impurity, which was identified as a chlorinated dioxin, was responsible for the acne. The acnegenic property of the chlorinated dioxins, as well as other other chlorinated compounds, is apparently related to the degree of chlorination of the parent compound together with the relative m olecular positions of the chlorine atom s and their 10 sym m etry. Unchlorinated dioxin does not cause acne and it is non-toxic. Note too that 2,4,5-T per se does not cause chloracne; if the condition occurs upon exposure to a m ixture of 2,4,5-T, it is most likely that TCDD is responsible. The potent chloracnegens are, in general, polychlorinated arom atics, such as polychlorobiphenyls (PCB's), polychloroterphenyls, polychlorodiphenyloxide and chloronaphthalenes (or halow axes). Chloracne may appear 2 to 3 w eeks after the first exposure or after a delay of several m onths--and possibly even as long as two years (O liver, 1975). Mild chloracne clears up within several months; severe chloracne is persistent and has been known to last for as long as 15 years follow ing the cessation o f exposure to the causal agent (May, 1973). Persons m ost responsive to chloracnegens ate those who are prone to develop acne vulgaris. If there is no m edical history of chloracne, then the likelihood of a significant exposure to, or adverse health e ffe c ts from , TCDD is rem ote. H ence, chloracne is the clinical marker of TCDD exposure. In addition to its acnegenic properties, TCDD is also a very a c tiv e inducer of delta-am inolevulinic acid (d-ALA) synthetase and aryl hydrocarbon hydroxylase in both laboratory animals and man. The increase in GGT a ctiv ity and abnormal lipid lev els in eight men who contracted chloracne from occupational exposure to TCDD has been suggested as being due to the enzym e induction capability of TCDD (Walker et al, 1979). Quite likely, such induced d-ALA overstim ulates the production of porphyrin, which m ay explain the cause of porphyria cutan ea tarda (PCT) in certain experim ental animals and man (Crow, 1978). PCT has also been described as an interference with the uroporphyrinogen decarboxylase, which prevents conversion of uroporphyrinogen (UP) into protoporphyrinogen. The UP is converted instead to uroporphyrin, which in turn accum ulates in the liver until it is excreted in the urine. The disorder is enhanced clinically by alcohol, certain hormones and drugs, and a number of halogenated compounds (such as trichlorophenol) (Homberger et al, 1979). Seemingly, on the basis of tests w ith m ice, TCCD-induced porphyria does not occur in the presence of an iron deficiency (Sweeney et al, 1979). Other toxic e ffe c ts in man have been suggested by Reggiani (1977) and include: hyperpig m entation, hirsutism, liver damage (eg, mild fibrosis fatty changes, hem ofuscin deposition and parenchym al-cell degeneration), raised serum hepatic enzym e lev els, fat and carbohydrate m etabolism disorders, cardiovascular disorders, urinary and respiratory tract disorders, pancreatic disorders, polyneuropathies (eg, sensory impairments), weakness in lower extrem ities and neurasthenic or depressive syndromes. The m etabolism of TCDD in man is unknown, and for the present there is only lim ited inform ation available on the m etabolic pathways and m etabolites that may occur in other m am m als. M utagenicity: No clearly defined m utagenic e ff e c t has been observed in vitro with TCDD. The Am es te s t with Salm onella typhimurium (Gilbert et ad, 1980) and one strain of Escherichia coli (Hussain et al, 1972) was positive; however, the test with another strain of S typhimurium was negative. Hussain et al (1972), were presumably the first investigators to evaluate TCDD's m utagenic potentiad; they found that Escherichia coli (Sd-4), which m easures the reversion to streptom ycin independency, was positive (probably a baise-pair substitution m utation). Salmonella typhimurium (TA1530 and TA1532 strains), which m easures reversion to histidine protrophy, was positive (a fram eshift mutation) for TA1532 only at higher concentration s o f TCDD amd low er survived frequency; the te s t was n eg a tiv e for TA 1530. There could have been a weak prophage induction using _E coli (K39) c e lls , 11 193oB though the effect was difficult to separate from that of the dim ethyl sulfoxide solvent. By in feren ce, TCDD could be m utagenic if it w ere to in te rc a la te w ith DNA and lead to a fram e shift m utation; the intercalating effect with DNA was confirm ed by Kondorosi et al (1973) in th e ir study of TCDD's tran sfectio n of Q(beta) RNA. Seiler (1973) likewise perform ed the Ames test and found it to be positive with the TA1532 strain of typhim urium , y et his findings w ith o th er strain s--in addition to th e work of other research e rs--casts doubt on the reliab ility of the te s t resu lts. More recen tly , B ronzetti e t al (1980) experim ented w ith th e D7 strain of Saccharom yces cerevisiae to evaluate m itotic gene conversion and reverse m utation; both genetic e ffects w ere induced w ith the Sjq fraction, but there was no induction w ithout the fractio n . Their in vivo te st w ith m ale Swiss albino m ice injected1intravenously produced an effe c t within 20 days. O ther cytological changes from TCDD have been noted in the African blood lily (Jackson, 1972) and Drosophila (Davring e t al, 1971). Of more relevance to man are the in vitro studies on m am m alian cells--ie, HeLa; Balb-3T3, norm al mouse fibroblasts; SV101, virus (SV40)-transform ed 3T3 mouse fibroblasts; human foreskin fibroblasts and normal human lymphocytes (Beatty et al, 1975). There was no significant grow th inhibition in the cell cultures, nor w ere there discernible ultrastructural changes by electron microscopy. The authors did not dismiss the possibility that other test conditions and a higher level of organization than the single cell might reverse the reaction. A review of the genotoxicity of 2,4,5-T on anim als (Grant, 1979) points out th at the effects could be due to 2,4,5-T, TCDD or a combination of the two. Moreover, the emulsifiers and solvents of 2,4,5-T could be responsible for the chromosomal aberrations. O ncogenicity: DiGiovanni e t al (1977) found little or no tum or-causing or tum orprom oting pro p erties of TCDD, and B erry et al (1978, 1979) found a rem arkable inhibitory effect on skin tumors in mice. Toth e t ad (1978, 1979) found an increased incidence of liver tum ors in m ale Swiss (H/Riop) m ice th a t w ere fed b o th TCDD auid TCDD adm ixed w ith 2,4,5-trichlorophenoxyethanol (a derivative of trichlorophenol), but they were not convinced th a t TCDD was by itself a carcinogen. Even if TCDD were a weak tum or initiator, perhaps due to its relative chem ical stability w ithin th e organism , th e re is oth er d a ta of B erry e t al (1977) to show th a t TCDD facilitates the biotransform ation in rats of such tumor precursors as benzo(a)pyrene. According to the N ational Toxicology Program (NTP), TCDD when adm inistered by gavage was carcinogenic in the B6C3F1 mouse (NTP R ept 80-31; cited by Huff, 1980)--hepatocellular carcinom as w ere observed in both sexes of th e mouse but only fem ale mice and male rats developed follicular cell thyroid adenomas. Neoplastic nodules of the liver w ere noted in the fem ale rat. Likewise, derm al application of TCDD was carcinogenic in Swiss-W ebster mice (NTP R ept 80-32; cited by Huff, 1980); the females had integum entary fibrosarcom as, while results from the males were inadequate. (See also the section on "toxicity, animal," which discusses the toxicity and oncogenicity studies of Kociba e t al (1978).) T eratogenicity: 2,4,5-T containing 30 ppm of TCDD in two strain s of m ice (C57BL/6 and 12 AKR) and one strain of rats, produced increased incidences of cleft palate in mice and cystic kidney in the rat (Courtney et al, 1970); when comparably dosed with a sample containing only 1 ppm of TCDD, teratogenesis was not evident (Emerson et al, 1971). A gavage dose in the rat of 0.125 ug/kg/day from the 6th-15th days of gestation showed only a slight effect that was much pronounced at the level of 0.5 ug/kg/day (Sparschu et al, 1971). In utero exposure of mice to TCDD during the final half of the gestation period produced fetuses with poorly developed lymphatic systems and fatty infiltrates in the liver (Neubert et al, 1973). t Prenatally, TCDD causes hydronephrosis, which according to Gibson (1976), may be only a sign of delayed maturation and not permanent damage inasmuch as the test animals were taken by cesarean section. Gibson did find altered PAH transport from kidney slices of neonate animals even where there was no evidence of a lesion; the significance of these data on the adult form of the organism is unknown. TCDD also caused postnatal hydronephrosis in a suckling pup whose foster mother had been dosed beforehand with the compound (Moore et al, 1973). Reproduction and Fertility, Animals: Lamb and co-workers (1980) fed male mice (C57BL/6) varying dosages of a simulated Agent Orange mixture (using the free phenoxy acids instead of the butyl esters) in order to determine the effect of TCDD on the male's reproductive function. During the 8-wk period of dosing, the animals exhibited a doserelated liver and thymus toxicity and a significant reduction in body weight gain. Both the liver and thymus showed either significant or complete recovery when the test diet was terminated; there was also no significant change in sperm concentration, motility or percent of abnormal sperm either during or after the exposure. At the end of eight weeks, the treated males were mated to untreated virgin females; no significant decrease in fertility or reproduction nor evidence of germ cell toxicity or mutagenicity was evident, which is said to correlate well with the multigeneration studies of Murray et al (1979) and the dominant lethal assays of Khera et al (1973). Also, the paternal exposure did not appear to affect the survival of the offspring nor the development of the neonates. The three-generation reproduction study on the Sprague-Dawley rat led Murray et al (1977) to believe that fertility was only affected at the 0.01 and 0.1 ug/kg/day levels, not at 0.001 ug/kg/day; there were no other signs of organ damage. Seemingly, no embryotoxicity in this species occurs at or below 0.03 ug/kg/day according to Schwetz et al (1973). Dominant lethal studies of TCDD on male Wistar rats were negative, though the numbers of pregnancies were reduced; oral doses of 4, 8 and 12 ug/kg/day were given to groups of 20 animals for seven days prior to mating with untreated virgin females, and follow-up continued through seven sequential matings. Testes of the surviving males appeared normal histologically, except the inflamed state of the epididymides suggested an autoimmune reaction (Khera et al, 1973). Other adverse effects on the testes were shown in the 65-wk feeding studies on SpragueDawley rats by Van Miller et al (1977). TCDD levels of 0.05 to 1.0 ppm caused death within 4 weeks; the autopsied animals had a decided decrease in spermatogenesis. Male mice injected intraperitoneally with 0.4 mg/kg showed a 50% reduction in the rate of testicular DNA synthesis (Seiler, 1977). Other animals thus affected were the mouse and guinea pig (McConnell et al, 1976; 1978) and the chicken and monkey (Norback et al, 1973). These and other studies of the testes, wherein a weak dominant lethal effect was 1 .9 u 8 13 apparent, suggest that further research is needed, especially to determine if TCDD acts directly on the testes or by triggering a secondary response (Wassom et al, 1977/1978). Adult female Rhesus monkeys were fed 500 ppt of TCDD for up to 9 months (Allen et al, 1977; 1978), during which time they ingested about 2-3 ug of TCDD/kg/day. For the first three months there was acne, alopecia and periorbital edema. Irregular menstrual cycles were accompanied by decreased 17-beta-estradiol and progesterone levels. Early abortions and the inability to conceive occurred at the 6-month point. Offspring were weak and died within a few weeks, possibly from the additional dioxin in the maternal milk. Male Rhesus monkeys, fed for more than one year on a diet containing TCDD, displayed an absence of spermatids and mature spermatozoa (Allen et al, 1979). Reproductive Effects in Humans: There are some 287,000 defective births in the US each year, about 9% of all live births; another 560,000 infant deaths, spontaneous abortions, stillbirths and miscarriages cure due to defects in fetal development (March of Dimes Birth Defects Foundation, 1981). While about 20% of the defects are for genetic reasons and 3% to 5% cure the result of chromosomal aberrations, there are less than 10% that can be attributed to environmental agents--ie, radiation, infections, maternal metabolic imbalance, and drugs and other chemicals. The reasons for the remaining 65% to 70% of birth anomalies is indeterminate; in all likelihood, there may be a combination of any or all of these three factors (Wilson, 1973). A number of incidents over the past few years have focused on exposures of workers and the general public to lead, pesticides such as dibromochloropropane (DBCP) and chlordecone (Kepone) and the flame-retardant TRIS. Lead workers, for example, have shown evidence of deformed sperm (Lancranjan et al, 1975), while chemical and farm workers showed signs of oligospermia and azoospermia from their contact with DBCP and kepone (Whorton et al, 1977 and Cannon et al, 1978). The latest chemicals that appear to alter the quantity and quality of human sperm are toluenediamine and dinitrotoluene. Nine men who worked with these precursors of toluene diisocyanate (which is a polyurethane intermediate) experienced significantly reduced sperm counts and a reduced percentage of large forms of sperm (from CDC's Morbidity Mortality Wkly Rept 30:199 (1981)). And should the findings from the Boston Drug Surveillance Program (Jick et al, 1981) be substantiated, some common defects--such as Down's syndrome, limb deform ities, malignant brain tumors and hypospadias--and spontaneous abortions may be traced to certain vaginal spermicides that are used either pre- or post-conception. Other agents that have profound effects on human health are tobacco smoke and alcohol. In fact, many of the disabilities that have been claimed for slight exposures to 2,4,5-T, TCDD and silvex are typical of those caused by these two agents. Thus, any analysis of human toxicology or epidemiologic data must consider the smoking and drinking habits of the subjects. The Alsea II results, in particular, are meaningless without such information. Solid,, persuasive evidence relates smoking dose to the frequency of spontaneous abortions. The probability of miscarriage is almost double. Smoking also retards the rate of fetal growth and increases the risks of spontaneous abortion, fetal death and neonatal death in otherwise normal infants. Smoking during pregnancy may affect physical growth, mental development and behavioral characteristics of children at least up to the age of 11. Yet, evidence does not support a conclusion that maternal smoking increases the incidence of congenital malformations (US Surgeon General, 1979). An additional review on the subject is provided by Rylander et al (1981). 14 19 3 d 3 A very limited analysis was made of 43 cigarette smokers (matched against 43 nonsmokers) to see if there were any differences in sperm morphology between the two groups (Evans et al, 1981). Hie results confirm earlier work of a similar nature, indicating a higher frequency of morphologically abnormal sperm cells, which in turn may reflect genetic damage to the cells. Yet, a more detailed analysis of 344 males, extending over eight years, revealed no significant differences (at the 5% level) in sperm count, morphology or motility between non-smokers, light or moderate smokers and heavy smokers (Godfrey, 1981). And now that marijuana smoking and other illicit drugs have become more prominent and socially acceptable, perhaps as an aftermath of the Vietnam war, we should be aware of how they alter the reproductive functions of both men and women. Hembry, an infertility specialist and endocrinologist at Columbia University, reports that an average of 16 joints/day for four weeks lowers the sperm count significantly and increases the percentage of abnormally shaped sperm, not to mention the possible genetic damage that may occur if such sperm were to successfully produce a conception (Brody, 1981). The reproductive and other health hazards associated with alcohol consumption have been reviewed most recently by Eckardt et al (1981). There is currently no direct evidence of any causal connection between 2,4,5-T or TCDD and neural tube damage in the human fetus (Field et al, 1979; Hanify et al, 1981). The latter group--in their survey of 37,751 babies bom in Northland, New Zealand hospitals from 1960-77--did find a significant association between 2,4,5-T applications in the environment and occurrences of talipes, though TCDD exposure levels were unknown; the authors aver that the "causal relation remains to be established." Thomas (1980), in his examination of data from the Hungarian registry of birth defects, could detect no marked increase in the incidence rates of cleft palate and lip and cystic kidney disease during 1970-76, a time when there was a thirty-fold increase in the use of 2,4,5-T in Hungary. During this same period, the incidence rates of stillbirths, spina bifida and anencephalus were on the decline. Again, the extent of TCDD exposure was not clearcut, but it was assumed that the agricultural and forestry populations of Hungary--those most apt to have contact with the herbicide--were far more numerous than in Australia and the United Kingdom, and the amounts of TCDD in their 2,4,5-T formulations were no less than 0.1 ppm. Information about in vivo chromosomal aberrations due to TCDD in humans, for that matter, cytological effects of TCDD on man in general, is not only meager but inconclusive. HEALTH SURVEYS OF THE EXPOSED HUMAN POPULATIONS Most toxicologists will agree that the results obtained from their animal experiments only presume a like effect in humans. In the final analysis, the most reliable test for the human animal is man himself. And while intentional human experimentation is becoming less acceptable in our society, we are occasionally afforded human experience with a particular toxicant as the result of some untimely accidental exposure to it. Such an opportunity may occur in connection with the workplace, or it may involve a very diverse fraction of the general population. In either event, there will frequently be a definable cohort for an epidemiologic study. Even so, one must be mindful that the most carefully designed study will often lack proper controls, have incomplete exposure data or will fail to account for one or more confounding variables. 15 192*70 So it is with Agent Orange and the associated chlorodioxin. A number of exposures to TCDD among the industrial and general populations have occurred over a long period of time. A total of 14 accidents have been identified as occurring between 1949 and 1976, both here and abroad, in chemical plants that manufactured trichlorophenol. The data derived from the incidents are not entirely conclusive, yet many of the cases have been documented well enough to provide predictive evidence of the effects of TCDD on man. See Table 1 of the appendix for a summary. Of Monsanto Em ployees Of the 122 persons who were identified as having had chloracne following the TCP process accident at Nitro, West Virginia in 1949, 121 were selected for a cohort study (Zack et al, 1980). The results'of the standardized mortality analysis revealed that the 32 actual deaths (from all causes) were less than the 46.41 expected deaths; this SMR of 0.69 was the only statistically significant value among the collected data. Nor was there any apparent excess of deaths from malignant neoplasms or circulatory diseases. Of the nine deaths from malignant neoplasms, five were due to lung cancers--four of them in cigarette smokers--three were from neoplasms of the lymphatic and hematopoietic tissue; one was blamed on a rare form of skin cancer, a fibrous histiocytoma. Soon after the accident and on three occasions during 1949-53, Ashe et al (1949, 1950) and Suskind (1953) examined 12 of the workers who were severely affected with chlor acne; an additional 26 persons with chloracne, who were not connected with the accident, were examined in 1953. The most frequent clinical symptoms were acneform lesions, followed by severe muscle pains in the upper and lower extremities, shoulders and thorax; fatigue; nervousness and irritability; decreased libido; dyspnea; vertigo and intolerance to cold. In addition to one case of sensory loss in the foot, there were four instances of hepatomegaly and other signs of liver impairment. Upon re-examination of several of the above workers in 1953, there was marked improvement in the skin lesions and a general subsidence of the non-cutaneous symptoms. Some persons had persistent but inexplicable complaints of pains in the back and lower extremities, nervousness, excess fatigue and dyspnea. Another epidemiologic study of the 1949 accident will comprise a cohort of 437 persons; viz, the retirees and present and former Nitro employees, including those who were a part of the accident. The data are now being analyzed by Suskind and others, and their results will be published later. O f Dow C hem ical Em ployees Ott et al (1980) examined the mortality experiences of a group of 204 workers who had been exposed to 2,4,5-T during its production from 1950-71. Many of the workers were also potentially exposed to a number of other chemicals dinring this period. Only 19 episodes of acute exposure were noted during 1954-70, which consisted largely of irritant exposures to the eyes and skin. No cases of chloracne or porphyria cutanea tarda were evident, and mortality of this limited sample was stated to be comparable to that of US white males as well as to the "background mortality experience" at this location. A mortality study, undertaken on 61 Dow employees who had been engaged in the production of trichlorophenol and who had been accidentally exposed in 1964 to TCDD, did not disclose an elevated mortality rate; there were 4 actual deaths from all causes vs 7.8 expected. Overall cancer deaths were elevated (3 vs 1.5 expected) but no particular tumor type predominated. None of the findings was statistically significant (Cook et al, 1980). Of these workers, 49 developed chloracne. The three cancer deaths (identified as 116 C M- L c ? _> 1 an adenocarcinoma, a fibrosarcoma and a glioma with mtastass) were slightly above the expected number (3 vs 1.6), but they could not be ascribed to any predominant tumor type or particular organ/tissue site (Cook et al, 1980). The latest Dow study from an interview questionnaire of 715 wives of men who worked in the Michigan division revealed no statistically significant differences in occurrences of miscarriages, stillbirths and fetal deaths or malformations. The two groups (ie, 370 wives of men with exposure to TCDD and 345 controls, wives of men without possible exposure) were matched for maternal age, complications during pregnancy and labor, existing medical conditions, smoking and mother's occupation. One weakness in the study, however, may be that the exposed group had 737 conceptions vs 2,031 for the controls (Anon, 1981). O f Seveso Inhabitants An earlier section of this report described an explosive release of TCDD from a trichlorophenol reactor in the vicinity of Seveso, Italy. This is the largest single population--over 37,000 persons--to have been potentially exposed to varying doses of this compound. Two years after the incident occurred, the acute and mid-term health effects were assessed; the mild chloracne, which occurred mainly in a small group of children, healed quickly. Subclinical peripheral nerve impairment was reported; there was also some liver involvement, but without apparent functional disorders. Immunoresponse was not altered nor was susceptibility to infectious diseases increased (Reggiani, 1979). The long-term epidemiologic survey of the 220,000 residents of the Seveso area--conducted under the auspices of the NIEHS/IARC* working group--is expected to go on for at least five years. The most recent progress report of that group emphasizes the preliminary nature of all their findings and reiterates the above conclusions of Reggiani (Pocchiari et al, 1979, and Huff et al, 1980). Except for the skin, no organs or body functions were impaired. No derangement of gestation, no fetal lethality and loss, no gross malformations, no growth retardation at term and no cytogenetic abnormalities have yet occurred. Meanwhile, analysis of the livers of 144 sheep which died in the Seveso area following the accident has disclosed that tympanism was the cause of death, not TCDD (Anon, 1980). O f BASF Em ployees; Ludwigshafen, Germany Seventy-five workers in a trichlorophenol plant were exposed to reactor products during an accident and its subsequent cleanup. Most of them suffered chloracne; 42 suffered severely, and 21 of these persons also had systemic damage to internal organs or CNS disturbances. Polyneuritis, impaired senses and liver damage were most prominent. In 1978, a mortality study was conducted on the cohort: Of the 17 deaths (from 11 to 25 were expected, depending on the choice of control population), six were due to cancer (three from stomach cancer), five from cardiovascular diseases, one from cirrhosis of the liver and one from urogenital disease; all other deaths were either accidental or suicidal (Huff et al, 1980). No miscarriages or abortions were reported among the wives of exposed employees who were still at work in the plant (IARC, 1977). * NIEHS/IARC: National Institute of Environmental Health Sciences/International Agency for Research on Cancer , -, ;-'j rj 17 Of NV Philips Em ployees; Netherlands In this explosion of 1963 at a 2,4,5-T plaint, 106 workers were probably exposed to TCDD (or related polychlorodibenzofurans). The two operators were exposed to TCDD largely by way of inhalation. Other workers, who were either part of the regular adjoining work force or were part of the clean-up and rebuilding crews, were outfitted in protective gear; they may have had only dermal contact. Chloracne occurred in 44 persons. Liver damage was not apparent, but there were a few complaints of fatigue. Of the eight deaths, one was from a pancreatic tumor and one was due to a myocardial infarction (at age 69); five other deaths adso could have been from infarctions. 0 O f Swedish R ailway Workers The first study of this group, who had worked with both phenoxy acid (2,4-D and 2,4,5-T) and amitrole herbicides (Axelson et al, 1974), indicated that amitrole was probably solely responsible for the noted excess of tumors. A case-control analysis of the data has later revealed the possibility of a masked tumor-inducing effect of the phenoxy acids (Huff et al, 1980). Of O ther Swedish Workers Follow-up on an earlier line of investigation led Hardell and co-workers to additional cases of soft-tissue sarcomas, which again seemed related to occupational exposures to phenoxy acids or chlorophenols. Malignant histiocytic lymphomas* were clustered among a group of hospital admissions in Sweden from January to September 1978; 14 out of 17 of the patients were at the time in occupations that utilized phenoxy acids or chlorophenols (eg, forestry, wood and paper mills, and farming). Inasmuch as both 6-fold tumor increases (Hardell, 1977 and 1979) were attributed to impurities, such as TCDD, a case-control (case-referrent) study by Eriksson et al (1981) was launched; their objective was to confirm the previous findings and to clarify whether or not the cause was due to the impurities. The new risk ratio (5.7) for soft-tissue sarcomas from exposure to phenoxy acids or chlorophenols was of the same order of magnitude as before; furthermore, a risk ratio of 4.2 was obtained for other phenoxy acids (such as MCPA, 2,4-D, mecoprop and dichlorprop) which are free of dioxins. O f Finnish Railw ay and Forestry Workers A similar group of 30 workers in Finland, who were exposed to TCDD-containing herbicides, will be compared to a matched control group to see if there are any long term effects from their exposures. Thus far, no differences are apparent (Huff et al, 1980). O f Dow/M onsanto Cohorts (Soft-T issue Sarcomas) Researchers at the National Institute for Occupational Safety and Health (NIOSH) (Honchar et al, 1981) have released the results of an examination of four cohorts of chemical workers from Dow and Monsanto who were exposed to 2,4,5-T or 2,4,5-trichlorophenol and contaminated with TCDD. From the 105 deaths among these cohorts * Tumor sites included: frontal sinuses, bladder, ileum, retroperitoneal, parotid and submandibular areas, femur and groin. 18 were three from soft-tissue sarcoma (2.9% of the total); by comparison, the figure for similar deaths in 1975 among US males aged 20 to 84 yrs old was 0.07%. Casewise, one death (Monsanto) in 1978, at age 58, was from a malignant fibrous histiocytoma of soft-tissue origin, some 29 years after exposure to an accidental spill of TCP, which also resulted in chloracne; another Monsanto man, who had been involved in the synthesis of 2,4,5-T, was 49 at the time of death in 1972 from a generalized liposarcoma (he had no definite history of chloracne, yet his case cannot be discounted); the third case was a Dow TCP worker who died in 1975 at age 53 from fibrosarcoma--the subject had had facial dermatitis but no definite chloracne. Since the publication of the above NIOSH report, other cases of soft tissue sarcomas have been announced. One of these, a Dow worker (age unknown) who also was stationed in the TCP-production area when the outbreak of chloracne appeared in late 1963-64, died of a malignant fibrous histiocytoma; he, presumably, had had frank chloracne (Anon, 1981; Cook, 1981). Cook cites one common factor besides TCDD exposure for the four victims: they were cigarette smokers. Hardell and Eriksson (1981), on the other hand, saw no difference in the smoking habits of their subjects, and they refuted Cook's hypothesis that smokers who contracted chloracne from TCDD were at increased risk of these sarcomas. Their argument may be well founded when another disclosed case of soft-tissue sarcoma at Monsanto is considered (Moses et al, 1981). This man--a non-smoker--died at age 58 of a retroperitoneal neurogenic sarcoma (m a lig n a n t schwannoma*). He had worked for the company for 32 years as a truck driver, hauler and maintenance man; though he had not engaged in the production of TCP or 2,4,5-T, he could have been potentially exposed to either or both of the chemicals (probably for the first time around 1948). He had no history of chloracne. A father and son who were employed at the Sauget, Illinois chlorophenol plant of Monsanto were admitted almost simultaneously for diagnosis and treatment of soft-tissue sarcomas. The father, age 53, had a history of prolonged exposure prior to the tumor's development--a liposarcoma of the buttock and thigh. With resection and radiation therapy, the growth apparently was arrested. The 33-yr-old son, however, died shortly after the resection of an extensively metastasized fibrosarcomatous mesothelioma. Neither of the two men were known to have had chloracne (Johnson et al, 1981). O f Other Industrial Workers Porphyria cutanea tarda, of varying degrees of severity, and elevated urinary uroporphyrins were discovered among 11 out of 29 workers at Diamond Alkali's New Jersey plant. These workers, who were exposed to 2,4-D and 2,4,5-T in the manufac turing processes, were examined because of their chloracne symptoms (Bleiberg et al, 1964). The degree of hyperpigmentation was said to vary according to the severity of chloracne and was more prevalent in Negroes. The degree of hirsutism too was related to the severity of chloracne. The cause of porphyria cutanea tarda was assumed to be * According to the Task Force on Soft-Tissue Sarcoma, malignant schwannomas are said to comprise 4.9% of the soft-tissue sarcomas (Suit, 1978). 19 chemically induced as a result of damage or insult to the liver and not of genetic origin. The skin manifestations of PCT, ie, epidermal fragility and vesiculobullous eruptions, occurred on those areas of the face, ears and hands that were exposed to sun or pressure. A follow-up of Bleiberg et al was made in the same plant by Poland et al (1971) on 73 male workers. Moderate to severe chloracne was prevalent in only 13 out of 73 (18%) of the cases; again, it varied in degree and was accompanied by scarring, hyperpigmen tation, hirsutism and eye irritation. There was no evidence of porphyria cutanea tarda and only one case of uroporphyrinuria. The authors have inferred that porphyria cutanea tarda and chloracne are independent syndromes though they may have the same etiologic agent. There was either no acne or only minimal lesions in 82% of the employees. Occurrence of chloracne, likewise, was not significantly greater in Negroes than whites, and was seemingly not related to job location or duration of employment (hence, level of exposure); cardiovascular findings were unremarkable except for three persons with a history of myocardial infarction. Oliver (1975) has described three laboratory workers who were exposed to "pure" TCDD. Two of the three had typical chloracne. All had abnormal serum cholesterol levels but without porphyria. Two had gastrointestinal symptoms, one complained of blurred vision and impaired muscular coordination, and another complained of "neuralgia of the left thigh." Two years after the exposure, two of the men experienced personality changes, neurological disturbances and hirsutism. Jirasek et al (1973, 1974) have reported from Czechoslovakia on what is probably the most thoroughly documented group of exposed workers. From 1965-69, 78 persons were affected by their contact with the sodium salt of 2,4,5-T, the butyl ester of 2,4,5-T and pentachlorophenol; 76 of them developed acne, hypertrichosis and hyperpigmentation. Complaints included fatigue, weakness and pain in the extremities, sweating and headache; many had polyneuropathy. Clinical findings revealed hepatomegaly, porphyria cutanea tarda, and increases in ALA and alkaline phosphatase (Suskind, 1980). The tenyear follow-up of 55 workers by Pazderova-Vejlupkova et al (1980, 1981) emphasized the non-uniformity of TCDD's effects on individual organs and the development of illness that affects virtually every important organ system. Complete recovery is rare, except for anxiety and depression, and persistence of an impaired metabolism of fats and carbohydrates can be a major cause of premature arteriosclerosis. Children who were born to exposed individuals were not adversely affected. The Coalite episode, which occurred in Derbyshire, UK in April 1968, was first described by May (1973). This also involved an overheated reactor used to synthesize TCP; the explosive rupture of the kettle was accompanied by a detonation of the released glycol and chlorobenzene vapors. Seventy-nine cases of chloracne were initially recorded; many were severe, but they were resolved within four to six months. There was no evidence of depression, weight loss, malignancies or liver, kidney or cardiac impairment. The epidemiological survey of 1977-78, covered only 41 of the 90 exposed workers and revealed some abnormalities in blood chemistry and immune function. There were no chromosomal abnormalities in the circulating lymphocytes. The levels of serum cholesterol and triglycerides were elevated, while the gamma-glutamyl transpeptidase was abnormally high. Both IgD and IgM were lower than the controls. Twenty-nine of the original 41 workers were re-examined, but the immunoglobulins were not measured. The company is now reluctant to extend the investigation further, and it will no longer agree to medical examinations of the work force as long as the results are intended for publication (Hay, 1981). A group of 190 workers (80 current and 110 previously employed) of the Vertek Chemical 20 ,L *J\j> * yJ plant, Jacksonville, Arkansas, are being studied for possible adverse health effects from the plant's production of 2,4,5-T (Suskind, 1980). Results are presently unavailable. O f Vietnam War Veterans The Armed Forces Institute of Pathology is examining specimens of veterans' tissue to determine whether any of the illnesses in these men can be related back to their exposures to Agent Orange. By its very nature, the project can only reveal carcinogenesis and possibly the residual effects of acute toxicity. From past experience, toxic substances tend to damage specific organs or organ sites; thus, if TCDD is typical, one should expect a cluster of cases according to clinical symptoms and/or organ type. The only clustering thus far in the 152 cases that have been assessed occurs with the skin and subcutaneous tissues; the dominant diseases are epidermal inclusion cysts and chronic, non-specific dermatitis. There were no consistent patterns for all other dermatologic diagnoses and malignancies. If any malignancies were to have been induced by TCDD, they should be appearing by now. Yet, no unusual morphological features nor clustering of tumors by diagnosis or site implicate Agent Orange (Irey, 1981). Of "D edrum ming" Personnel As mentioned earlier, about 200 civilians were assigned by the military to destroy the contents of about 40,000 55-gallon drums (2.22 million gallons) of surplus Agent Orange at the close of the Vietnam war. Pre- and post-physical examinations, including X-rays, neurological examinations and extensive clinical chemical tests, were peformed on all workers along with environmental monitoring during this operation. Military personnel revealed no apparent physical effects (Young et al, 1978). ENVIRONMENTAL FATE OF 2,4-D , 2 ,4 ,5-T AND TCDD 2.4-- D and 2 ,4 ,5-T Environmentally, 2,4-D is regarded as a non-persistent material. At least 12 different reactions may contribute to its breakdown, such that its half-life in soil is less than 0.5 months. Phenoxy herbicides, as a class, undergo rapid decomposition in the soil; both 2.4- D and 2,4,5-T sure degraded to carbon dioxide, inorganic chlorides and water (Audus, I960). 2,4,5-T also is of minor ecological concern, though compared with 2,4-D it is slightly persistent; its half-life in soil when applied at customary rates varies from 2 weeks to 4 months. And since it is soluble in water only to the extent of 0.2 ppb, and is firmly bound to soil particles, its migration thru the soil should be minimal (Crosby et al, 1971; 1973). There were even no significant differences in the persistence and disappearance of these two herbicides when they were tested with various soil types, including the tropical soils of Vietnam (Blackman, 1974). TCDD TCDD is distributed equally among the fat and liver of organisms and to a lesser extent in the kidneys and skin. It is eliminated, for the most part, via the feces. Currently, the 21 1937G highest environmental residues of TCDD as measured by EPA ranged from 20-60 ppt in a small sample of fat from beef cattle that had grazed on 2,4,5-T-treated rangeland (Young et al, 1978). This finding conflicts with Kocher et al (1978), who detected only 3-4 ppt in range-fed beef. Mahle et al (1977) found no TCDD in the milk of cows that had grazed on 2,4,5-T-treated pasture or rangeland. Likewise, TCDD was undetectable (not more than 10 ppt) in human milk that was sampled from am area near West Texas, where 2,4,5-T herbicides were used repeatedly over a period of 20 years (Shadoff et al, 1977) . On the other hand, samples of human milk collected from Vietnam in 1970 and analyzed four years later contained 40-50 ppt of TCDD (Baughman, 1976). An extensive examination of 30 species of flora and fauna at the Eglin AFB test site, Florida, which had been exposed to soil levels of TCDD between*10 and 1500 ppt (mean = 165 ppt), revealed no abnormal histopathology nor other deleterious effects (Young et al, 1978) . TCDD breaks down rapidly on leaves of plants, in water and on the surface of soil, through the action of sunlight and the presence of am organic hydrogen-donor (eg, a petroleum hydrocarbon), the reaction is a progressive dechlorination of the aromatic ring. Both Crosby et al (1977) amd Nash et al (1978) confirm the extreme photosensitivity of TCDD in the presence of ultraviolet radiation and an organic hydrogen-donor, especially when the compound exists as a thin layer on plaint materiail and the like. It does persist in the soil amd in aquatic ecosystems, but--like 2,4,5-T--there is no evidence that leaching by water occurs. From experimental work with Seveso soils, it appears that TCDD cam be taken up by certaiin plants amd translocated from the roots to the foliage (Cocucci et al, 1979), the compound may then be photochemically degraded, released into the atmosphere by transpiration and/or metabolized by the plamt. This matter of translocation, however, is contrary to the earlier findings of Helling et al (1973) and Isensee et al (1971). TCDD generally has a half-life in soil no longer than one year (Kearney et al (1972), or about 230 days. Under certain conditions it may persist longer than this if it was initially present with high amounts of 2,4,5-T in the soil (Young, 1976). Though microbial degradation is not a major means of elimination from the environment, there are at least five strains of bacteria that are known to assimilate it (Matsumura et al, 1973). There is also a possibility that mutants of existing microbial strains, or even artificially created Pseudomonas strains (Chatterjee et al, 1981), will become increasingly effective scavengers in the future. When lake sediments containing TCDD were incubated in the laboratory, the compound was released from the sample mainly by evaporation. Its rate of metabolism was influenced by those conditions that also stimulated microbial growth in the sample (Ward et al, 1978). Due to the paucity of environmental sampling, it is not yet possible to say that TCDD is bioaccumulating; although, there is the possibility that levels of 0.3 ppt might be found along rights-of-way where periodic applications of 2,4,5-T and silvex have been made (American Farm Bureau Federation, 1979). There is no evidence that tetrachlorodioxins are formed in the environment (Sundstrom et al, 1979). The Sabine (Texas) River Authority recently removed 15 drums of chemical compounds, including 2,4,5-T and 2,4-D, from a waste disposal site that was located about 300 yards from the lake that furnishes about 20% of the drinking water to the city of Dallas and its suburbs. Buried in 1978, the waste does not appear to have left dangerous concentrations 22 of TCDD in the soil. EPA is conducting further sampling to ensure that there is no serious contamination (US Environmental Protection Agency, 1980). CRITICAL REVIEWS OF PRIOR STUDIES O f EPA's A lsea Studies In the Alsea basin of Oregon, 2,4,5-T and silvex were used on a seasonal basis to eradicate unwanted vegetation from the forest lands. In 1978, there was some concern that such spraying was responsible for a number of spontaneous abortions in the area. All told, 13 miscarriages were reported from nine women. As a result of much publicity and public concern, the EPA conducted a study--Alsea I--to assess the facts and to determine whether or not the spraying was indeed responsible. EPA concluded that there was no relationship between the spraying and the incidence of miscarriage; the rate was within the expected limits. The study was harshly criticised because the exposures were not documented and the statistical treatment was improper. Thus, Alsea II was begun in October of 1978, to study a much larger area and population than Alsea I and to compare miscarriages within the first 20 weeks of pregnancy with the spraying of 2,4,5-T. A high probability of an elevated miscarriage rate shortly after spraying was reported to the press, but this claim was proved to be without merit. --By Oregon State U niversity A critical analysis of Alsea It was conducted by an interdisciplinary team of Oregon State University scientists (Wagner et al, 1979). It did not support any of EPA's conclusions in the Alsea II study. The team contended that: there was no control population in the study; data on miscarriages were unreliable and inaccurate; medical practices were profoundly different in the three areas under consideration; information on use of 2,4,5-T was inaccurate and incomplete; variations in hospitalized miscarriages in all the areas were well within expectations; differences existed in topography, climate and demography; data collection was incomplete and failed to account for patterns of and differences in medical practice among areas. The panel termed the EPA's hospitalized spontaneous abortion indices (HSAD values to be unreliable because: there was substantial variation in the magnitude of the HSAI's for any one month among the six years and the peak HSAI in June was due to a single large excursion only in 1976, while the seasonal cyclic peaks were consistent with random variations. Furthermore, 23 19228 alleged increase in the use of 2,4,5-T was actually due to more complete collection of spray data, there was no statistically significant cross-correlation except for the apparent increase in the June HSAI--deletion of the single June 1976 HSAI value removed the significant correlation--and the HSAI increase in June 1976 was identified only with a location where human exposure was least likely to have occurred. Hence, according to the OSU panel, there was: no significant correlation between 2,4,5-T use and the HSAI, and EPA's study was seriously flawed because of incomplete and inaccurate data. --By FIFRA's S cien tific Advisory Panel This panel of the Federal Insecticide, Fungicide and Rodenticide Act reviewed the evidence and found neither immediate nor substantial threat to human health or the environment when silvex or 2,4,5-T was applied to rice, rangeland, orchards, sugarcane, and when used for certain non-crop purposes. --By Others Lamm (1979) also declared the Alsea II study to be epidemiologically unsound and the report useless. Both statistically and epidemiologically, the report does not support any relationship between herbicide spraying and miscarriages. Lamm (1980) has since claimed that only four out of the ten women who were cited by EPA in the Alsea II study underwent spontaneous abortions; the remainder were hospitalized for other reasons. New Zealand's Allan H. Smith (1979) termed the Oregon study of highly doubtful validity. Robinson (1979) of Austrialia emphasized that the EPA information in no way confirmed a hazard to human health from 2,4,5-T when it was used in a careful manner and according to label instructions on the container. A comprehensive review by the Australian Department of Health (1977) concluded that there was no evidence that 2,4,5-T caused human birth defects. CURRENT AND PROPOSED STUDIES International A gency for R esearch an Cancer (IARC) An ad hoc Working Group, comprised of representatives of the National Institute of Environmental Health Sciences (NIEHS) and IARC, met in Lyon, France in January of 1978 to discuss the feasibility of a coordinated epidemiological study on the long-term hazards of the chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans. A review of the meeting and an account of the long-term hazards are to be found in Huff et al (1980). An outcome of the meeting was the unanimous agreement for an exchange of information and the coordination of research activities. Three items were believed worthy of special attention: development of sensitive and reliable measurements of body burden, more sensitive and less expensive analytical methods and standardization of sampling procedures; 24 1 9 3 7 9 development of standard protocols for clinical examinations as well as basic information; and development of an international registry for long-term follow-up of exposed persons, including a registry of chloracne cases. IARC has contracted with the National Institute of Environmental Health Sciences (NIEHS) for NIEHS to establish and maintain an international registry of persons outside the US who have been exposed to phenoxy herbicides and their contaminants (personal communication, Suskind, 1981). US Air Force Epidem iological Study --P roject Ranch Hand Personnel This project will arrange for the examination of the 1,200 men who were involved in the actual handling and spraying of Agent Orange. They, and the control population of 20,000, are to be followed over the next 20 years (Holden, 1981). Armed Forces In stitu te o f Pathology (AFEP) Thru its Department of Environmental and Drug-Induced Pathology, AFIP is examining biopsy and autopsy tissue of all Vietnam veterans. To date, only 152 cases have been assessed; the findings have been presented earlier in this report under the heading "health surveys of exposed human populations." N ational Cancer Institute (NCI) The Epidemiology Branch of NCI is conducting a number of investigations where pesticide exposures are involved. In cooperation with the Federal Aviation Administration, a retrospective mortality study of 10,000 aerial pesticide applicators will be conducted; NCI will run a similar study on 4,500 pest control operators. The University of Minnesota, under contract to NCI, will conduct a case-control interview study of patients with diagnosed leukemia or non-Hodgkins lymphomas within the past year or over the next two years; some 300 cases of each disease and their matched controls will likely be included. A soft-tissue sarcoma study also has been proposed; this will be conducted jointly with the Armed Forces Institute of Pathology (Cantor, 1981). N ational Institute far O ccupational S afety and H ealth (NIOSH) NIOSH is establishing a registry of all US workers who may have had exposure to TCDD in the course of manufacturing 2,4,5-T. Results of the data analysis are not expected before about 1983 (Anon, 1980). US D epartm ent o f Agriculture Cancer deaths among Forestry Service workers will be compared with a cohort of other employees who are not considered at risk. V eterans Adm inistration An annotated review of the herbicide literature, including the phenoxy acids, their 25 derivatives and associated chlorodioxins, will be completed around October 1981. The work was performed under a contract awarded December 1980 to JRB Associates. A contract was awarded in May 1981 by the Veterans Administration to the University of California at Los Angeles for the design of an epidemiologic study of Vietnam veterans. Approximately 45,000 Vietnam veterans who expressed concern about the hazards of Agent Orange have been examined; data on 25,000 of these men have been placed in a special Agent Orange Registry, which may serve later to identify them and to provide medical information, as well as indications of health trends over the long term. The Chloracne Task Force was established to sift out those cases of dermatitis that either resemble or are truly chloracne. Cases of the former type will be re-examined by dermatologists who have an expert knowledge of the disease. Thus far, there have been only 700 cases of "skin conditions" out of the total of 3,500 claims that have been filed for damage from Agent Orange. O th ers Australia will fund a two-year study of its Vietnam veterans and their offspring to ascertain if there were any adverse effects of military exposure to Agent Orange. Some 60,000 veterans with an estimated 100,000 children will be involved (Anon, 1980). Canadian officials recently have become alarmed after the Canadian Wildlife Service found TCDD in herring gull eggs, a common fish-eating bird, from Lake Ontario and a part of Lake Michigan. These preliminary results have led them to believe there may be US industriad contamination of these waters. A joint US-Canadian effort to study the situation further is a likelihood (Anon, 1980, 1981). 26 19381 SUMMARY Agent Orange--as used in Vietnam--was a 50:50 mixture of the n-butyl esters of 2,4-dichlorophenoxyacetic acid (2,4-D) and 2,4,5-trichlorophenoxyacetic acid (2,4,5-T), together with a minor amount of the free acid 2,4,5-T (1% of the total mixture) and varying amounts (from 0.02 to 15 ppm, or a weighted mean of 1.98 ppm) of the contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Commerical formulations of 2,4,5-T, as well as silvex (or 2-2(2,4,5-trichlorophenoxy)propionic acid), may contain TCDD as a contaminant. At one time the amount of TCDD was as much as 70 ppm; present methods routinely produce levels below 0.01 ppm, with occasional excursions to no more than 0.05 ppm. TCDD may form as a by-product in the synthesis of 2,4,5-trichlorophenol (a precursor of 2,4,5-T), particularly if the alkaline reaction mixture is allowed to rise uncontrollably beyond 180C. It may also be produced by the pyrolysis of certain chlorinated compounds in industrial or municipal wastes and by the burning of vegetation that has been sprayed with the chlorinated phenoxy acid herbicides. Note that 2,4-D, a phenoxy herbicide closely related to 2,4,5-T, is not normally contaminated with TCDD. In addition to the marked variations in the sensitivity and susceptibility of animal species to all toxic substances, there are significant differences between some of the toxic effects of TCDD in experimental animals and the human experience; thus, the animal data cannot be translated directly to man. One of the more pronounced biological effects of TCDD, as well as a number of other chlorinated aromatic compounds, is a tendency to cause chloracne in certain animals and man. This particular skin disease is, in fact, regarded as the clinical marker for TCDD exposure. Systemic disorders in man from exposures to TCDD are unlikely to occur in the absence of chloracne. Such symptoms as impaired liver function, nephropathy, GI irritation, myopathy and neuropathy, including depression and irritation of the central nervous system, have been reported after exposure to large amounts of TCDD; however, these symptoms have not been progressive, and they have always cleared with time. Other toxic effects of TCDD in test animals appear as pathological changes in the liver, peripheral nerves, blood forming organs and reticuloendothelial system. TCDD is a very powerful enzyme inducer. In addition to altering normal enzyme activity, it may potentiate the harmful action of other toxins or even render an otherwise innocuous agent toxic. TCDD can serve as a promoter of carcinogenesis or as an inducer of cancer in some strains of rats and mice. The carcinogenicity is always accompanied by considerable systemic toxicity in contrast to some other chemical carcinogens. By itself, 2,4,5-T is not a carcinogen, nor has it been shown to cause genetic changes in any animal species. A consensus of studies using high doses of 2,4,5-T with 0.1 ppm or less of TCDD showed cleft palate in mice--but no other species--and embryotoxicity to the mouse, rat, hamster, sheep, monkey and rabbit. TCDD does induce genetic changes by the Ames test with S typhimurium and E coli but dominant lethal and cytogenetic evaluations in rodents do not confirm "that such changes will also occur in other animals, let alone man. 27 While 2,4,5-T and 2,4-D pesticides have been used in agriculture, forest management, and commerical and residential landscaping for over 30 years, there is still no conclusive evidence that they and/or TCDD are mutagenic, carcinogenic or teratogenic in man, nor that they have caused reproductive difficulties in the human. Both 2,4,-D and 2,4,5-T undergo rapid decomposition in the soil and are, therefore, of little environmental concern. TCDD does persist in soil longer than 2,4,5-T but, in general, its half-life in soil is no longer than one year; in the presence of ultraviolet light it breaks down rapidly when present as a thin film on plants, water and the surface of soil. RECOMMENDATIONS The Council on Scientific Affairs recommends that: (1) The above studies on exposed, or allegedly exposed, persons continue to be supported and, if feasible, enlarged to include the cooperative engagement of all internationally known exposure data, as recommended by the International Agency for Research on Cancer (IARC). (2) All physicians be alerted through AMA publications to the classical signs of chloracne and the possible signs and adverse effects of TCDD exposure. They should be encouraged to enlist in the present efforts to identify and treat those persons who have had serious exposures to TCDD, and to cooperate in the collection of vital information that is needed for the ongoing human epidemiologic studies. 28 REFERENCES* Allen JR, Barsotti DA, Lambrecht LK, Van Miller JP (1979): Reproductive effects of halogenated aromatic hydrocarbons on nonhuman primates. Ann NY Acad Sci 320:419-25. Allen JR, Barsotti DA, Van Miller JP, et al (1977): Morphological change in monkeys consuming a diet containing 500 ppt of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Food Cosmet Toxicol 15:401-10. Allen JR, Van Miller JP (1978): Health implications of 2,3,7,8-tetrachlorodi- benzo-p-dioxin and exposure in primates. Environ Sci Res 12:371-79 (from Chem Abstr 89:123996). .. Allen JR, Van Miller JP, Norback DH (1975): Tissue distribution, excretion and biological effects of (14C) 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Fd Cosmet Toxicol 13:501-05. American Farm Bureau Federation (1979): Scientific Dispute Resolution Conference on 2,4,5-T. Park Ridge, Illinois. American Society of Mechanical Engineers (1981): Study on State-of-the-Art of Dioxin From Combustion Sources. HOQ180, ASME, New York City (Cited from C&EN :22 (Jun 1981)). Anonymous (Feb 26, 1979): EPA disputes Dow's dioxin theory. C&EN :6. Anonymous (Feb 28, 1980): NIOSH registry on 2,4,5-T workers part of interagency study of dioxin. Occ Safety & Health Reptr :904. Anonymous (Apr 9, 1980): Australia funds check on Agent Orange. Chem Week :18. Anonymous (Oct 20, 1980): No tie between deaths, dioxins, Monsanto finds. C&EN :7. Anonymous (Dec 10, 1980): Minister's remarks raise dioxin concerns. Chem Week :17. Anonymous (Dec 17, 1980): More Seveso tests. Chem Week :21. Anonymous (Jan 5, 1981): Dow study finds 2,4,5-T no pregnancy risk. C&EN :6. Anonymous (Mar 1981): Second Dow sarcoma case found; company suggests smoking as a factor. Occ Safety & Health Reporter 10(41): 1348. Anonymous (Apr 1, 1981): Suit filed against Monsanto. Chem Week :19. Ashe WF, Suskind RR (Dec 1949 and Apr 1950): Reports on chloracne cases, Monsanto Chemical Co., Nitro, West Virginia. Reports of the Kettering Laboratory. Audus LJ (I960): Microbiological breakdown of herbicides in soils. In: Herbicides and the Soil. Woodford EK and Sugar GR (eds), Blackwell Sci Pub, Oxford, England, pp 1-9. Australian Dept of Health (1977): 2,4,5-T and Human Birth D efects. Report of the Division of Public Health, Department of Health, Australia. Axelson O, Sundell L (1974): Herbicide exposure, mortality and tumor incidence; an epidemiological investigation on Swedish railroad workers. Work-Environ Health 11:21. Bauer H (1918): Gewerbliche erkrankungen durch gechlflrte kflhlenwasserstoffe. Zbl Gew Hyg 6:100. Bauer H, Schulz KH, Spiegelburg U (1961): Berufliche vergisftungen bei der herstellung von chlorphenol-verbindungen. Arch Gewerbepath Gewerbehyg 18:538-55. Baughman RW (1976): Tetrachlorodibenzo-p-dioxins in the environment; high resolution mass spectrometry at the picogram level. Diss Abstr Int 36(7):3380B. Beatty PW, Lembach KJ, Holscher MA, Neal RA (1975): Effects of 2,3,7,8-tetrachloro- dibenzo-p-dioxin (TCDD) on mammalian cells in tissue cultures. Toxicol Applied Pharmacol 31:309-12. * Alphabetically by author 29 0O 1i>W 4 Berry DL, Slaga TJ, DiGiovanni J, Juchau MR (1979): Studies with chlorinated dibenzop-dioxins, polybrominated biphenyls and polychlorinated biphenyls in a two-stage system of mouse skin tumorigenesis; potent anticarcinogenic effects. Annals NY Academy of Sciences 320:405-414. Berry DL, Slaga TJ, Wilson NM, Zachariah PK, Namkung MJ, Bracken WM, Juchau MR (1977): Transplacental induction of mixed-function oxygenases in extra-hepatic tissues by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Biochem Pharmacol 26:1383-88. Berwick P (1970): 1,4-Dichlorophenoxyacetic acid poisoning in man. JAMA 214(6):1114-17. Blackman GF, Fryer JD, Lanf A, Newton M (1974): The effects of herbicides in South Vietnam; part B -- Working papers; persistence and disappearance of herbicides in tropical soils. Bleiberg J, Wallen M, Brodkin R, Applebaum K (1964): Industrially acquired porphyria. Arch Dermatol 89(6):793-97. Bovey RW, Young AL (1980): The Science of 2,4,5-T and Associated Phenoxy Herbicides. Wiley, New York. Brody JE (Mar 10, 1981): Sperm found to be especially vulnerable to environment. New York Times. Bronzetti G, Lee I, Zeiger E, Mailing H, Suzuki (1980): Genetic effects of TCDD in vitro and in vivo using D7 strain of S cerevisiae. Mutation Res 74:206-07. Bumb RR, Crummett WB, Cutie SS, et al (Oct 24, 1980): Trace chemistries of fire; a source of chlorinated dioxins. Science 210(4468):385-390. Buu-Hoi NP, Saint-Ruf G, Bigot P, Mangane M (1971) (in French): Preparation, properties and identification of dioxin (2,3,7,8-tetrachlorodibenzo-p-dioxin) in the pyrolysate of defoliants containing 2,4,5-T and its esters and in contaminated vegetation. Seances Acad Sei, Series D 273:708-11. Cannon SB, Veazy JM, Jackson RS, et al (1978): Epidemic kepone poisoning in chemical workers. Am J Epidemiol 107:529-37. Cantor K (1981): Report to the Veterans Administration's Advisory Committee on Health-Related Effects of Herbicides. Proceedings, 8th meeting (May 5, 1981), Washington, DC. Chatterjee DK, Kellogg ST, Furukawa K, Kilbane JJ, Chakrabarty AM (1981): Genetic approaches to the problems of toxic chemical pollution. Presented at Third Cleveland Symposium on Macromolecules; Recombinant DNA. Cocucci S, DiGerolamo F, Verderio A, Cavallaro A, Colli G, Gomi A, Invemizzi G, Luciani L (1979). Absorption and translocation of tetrachlorodibenzo-p-dioxin by plants from polluted soil. Experientia 35(4):482-84. Cook RR (1981): Dioxin, chloracne and soft-tissue sarcoma. Lancet L618-19. Cook RR, Townsend JC, Ott MG, Silverstein LG (1980): Mortality experience of employees exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). JOM 22(8) :530-32. Council for Agricultural Science and Technology (CAST) (Aug 1978): The phenoxy herbicides, 2nd ed. Report no. 77. Iowa State University, Ames, Iowa. Courtney KD, Gaylor DW, Hogan MD, Falk HL (1970): Teratogenic evaluation of 2,4,5-T. Science 168:864-866. Crosby DG, Moilanen KW, Wong AS (1973): Environmental generation and degradation of dibenzodioxins and dibenzofurans. Environ Health Perspect (Exptl issue 5):259-66. Crosby DG, Wong AS (1977): Environmental degradation of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Science 195:1337-38. Crosby DG, Wong AS, Plimmer JR, Woolson EA (1971): Photodecomposition of chlorinated dibenzo-p-dioxins. Science 173:748-49. Crow KD (1970): Chloracne. Trans St John's Hosp Dermatol Soc 56:79-99. Crow KD (1978): Chloracne; an up-to-date assessment. Ann Occup Hyg 21:297-98. 30 1 q 35 Davring L, Summer M (1971): Cytognie effects of 2,4,5-trichlorophenoxyacetic acid on oogenesis and early embryogenesis in Drosophila melanogaster. Hereditas 68:115-22. DiGiovanni JA, Viaje A, Berry DL, Slaga TJ, Juchau MR (1977): Tumor initiating ability of 2,3,7,8-tetrachlorodiobenzo-p-dioxin (TCDD) and Aroclor 1254 in a two-stage system of mouse skin carcinogenesis. Bull Environ Contam Toxicol 18:522-57. Eckardt MJ, Harford TC, Kaelber CT, Parker ES, et al (1981): Health hazards associated with alcohol consumption. JAMA 246(6) :648-66. Elson M (Oct 5, 1980): Lawyers map strategy here for vast Agent Orange suit. Chi. Trib. Emerson JL, Thompson DJ, Strebing RJ, et al (1971): Teratogenic studies of 2.4.5- trichlorophenoxyacetic acid in the rat and rabbit. Food Cosmet Toxic 9:395-404. Eriksson M, Hardell L, Berg NO, Mller T, Axelson O (1981): Soft-tissue sarcomas and exposure to chemical substances; a case-referrent study. Br J Indust Med 38:27-33. Evans HJ, Fletcher J, Torrance M, Hargreave TB (1981): Sperm abnormalities and cigarette smoking. Lancet 3s627--29. Felton JS, Nebert DW (1975): Mutagenesis of certain activated carcinogens in vitro associated with genetically mediated increase in monooxygenase activity and cytochrome P.-450. J Biol Chem 250:6769-78. Field B, Kerr C (1979): Herbicide use and incidence of neural-tube defects. Lancet K8130):1341-42. Ford RE, et al (Winter 1978): Mucotoxins, environmental contaminants in nature. Illinois Res :10--11. Fowler BA, Hook GER, Lucier GW (1977): Tetrachlorodibenzo-p-dioxin induction of renal microsomal enzyme systems; ultrastructural effects on pars recta (S^) proximal tubule cells of the rat kidney. J Pharmacol Exp Ther 203:712-21. Fries GF, Marrow GS (1975): Retention and excretion of 2,3,7,8-tetrachlorodibenzo-p-dioxin by rats. J Agric Food Chem 23:265-69. Gasiewicz TA, Neal RA (1979): 2,3,7,8-Tetrachlorodibenzo-p-dioxin tissue distribution, excretion and effects of clinical chemical parameters in guinea pigs. Toxicol Appld Pharmacol 51(2):329-39. Gehring PJ, Kramer CG, Schwetz BA, Rose JQ, Rowe VK (1973): The fate of 2.4.5- trichlorphenoxyacetic acid (2,4,5-T) following oral administration to man. Toxicol Applied Pharmacol 26:352-66. Gibson JE (1976): Perinatal nephropathies. Environ Health Perspect 15:121-30. Gilbert P, Saint-Ruf G, Poncelet F, Mercier M (1980): Genetic effects of chlorinated anilines and azobenzenes on Salmonella typhimurium. Arch Environ Contamn Toxicol 9:533-41. Godfrey B (1981): Sperm morphology in smokers. Lancet L948. Goldstein JA, Friesen M, Scotti TM, Hickman P, Hass JR, Bergman H (1978): Assessment of the contribution of chlorinated dibenzo-p-dioxins and dibenzofurans to hexachlorobenzene-induced toxicity, porphyria, changes in mixed-function oxygenases and histopathological changes. Toxicology and Applied Pharmacol 46:633-49. Goldstein NP, Jones PH, Brown JR (1959): Peripheral neuropathy after exposure to an ester of dichlorophenoxyacetic acid. JAMA 171:130-33. Grant WF (1979): The genotoxic effects of 2,4,5-T. Mutation Res 65:83-119. Hanify JA, Metcalf P, Nobbs CL, Worsley KJ (Apr 17, 1981): Aerial spraying of 2,4,5-T and human birth malformations; an epidemiological investigation. Science 212:349-50. Hardell L (1977): Malignant mesenchymal tumors and exposure to phenoxy acids; a clinical observation. Lkaartidningen 74:542-46. Hardell L (1977): Soft-tissue sarcomas and exposure to phenoxyacetic acids and 31 19306 cancer. Lkartidningen 74:2753. Hardell L (1979): Malignant lymphoma of histiocytic type and exposure to phenoxy- acetic acids or chlorophenols. Lancet 1:55-56. Hardell L, Eriksson M (1981): Soft-tissue sarcomas, phenoxy herbicides and chlorinated phenols. Lancet 1L250. Hardell L, Sandstrm A (1979): Case-control study; soft-tissue sarcomas and exposure to phenoxyacetic acids or chlorophenols. Br J Cancer 39:711-17. Hay A (1979): Accidents in trichlorophenol plants; a need for realistic surveys to ascertain risks to health. Ann NY Acad Sci 320:321-24. Hay A (1981): Dioxin hazards; secrecy at Coalite. Nature 290:729. Hayes WJ (1963): Clinical Handbook on Economic Poisons. US public Health Service, US DHEW. Revised Edition, Publication No. 476. US Govt Printing Office, Welsh DC. Hayes WJ (1975): Toxicology of Pesticides. Williams and Wilkins, Baltimore. Helling CS, Isensee AR, Woolson EA, Ensor PDJ, Jones GE, Plimmer JR, Kearney PC (1973): Chlorodioxins in pesticides, soils and plants. J Environ Quality 2:171-78. Herxheimer K (1899): Uber clorakne. Mnch Med Wschr 46:278. Holden C (1981): UCLA designing big Agent Orange study. Science 212:905. Homberger E, Reggiani G, Sambeth J, et al (1979): The Seveso accident; its nature, extent and consequences. Ann Occup Hyg 22:327-70. Honchar PA, Halperin WE (1981): 2,4,5,-T, trichlorophenol and soft-tissue sarcoma. Lancet I(8214):268-69. Huff JE (ed) (1980): Reproductive assessment and developmental toxicology. NTP Tech Bull 1(3):4. Huff JE, Moore JA, Saracci R, Tomatis L (1980): Long-term hazards of polychlorinated dibenzodioxins and polychlorinated dibenzofurans. Environ Health Perspect 36:221-40. Hussain SL, Ehrenberg G, Lofroth G, Gejvall T (1972): Mutagenic effects of TCDD on bacterial systems. Ambio 1:32-33. International Agency for Research on Cancer (1977): Some fumigants, the herbicides 2.4- D and 2,4,5-T, chlorinated dibenzodioxins and miscellaneous industrial chemicals. IARC Monograph 15: Evaluation of the Carcinogenic Risk of Chemicals to Man. Irey NS (Mar 1981): Agent Orange. Presentation before the joint Spring meeting of the American Society of Clinical Pathologists and the College of American Pathologists, San Diego. Isensee AR, Jones GE (1971): Absorption and translocation of root and foliage applied 2.4-dichlorophenol, 2,7-dichlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzo-p-dioxin. J Agric Food Chem 19:1210-14. Jackson WT (1972): Regulation of mitosis; cytological effects of 2,4,5-trichlorophenoxyacetic acid and of dioxin contaminants in 2,4,5-T formulations. J Cell Sci 10:15-25. Jick H, Walker AM, Rothman KJ, Hunter JR, Holmes LB, et al (1981): Vaginal spermicides and congenital disorders. JAMA 245:1329-32. Jirasek L, Kalensky J, Kubec K (1973): Acne chlorina and porphyria cutanea tarda during the manufacture of herbicides. Cesk Dermatol 48:306-17. Jirasek L, Kalensky J, Kubec K, Pazderova J, Lukas E (1974): Acne chlorina and porphyria cutanea tarda, other manifestations of general poisoning, during the manufacture of herbicides, n. Cesk Dermatol 49:145-57. Johnson FE, Kugler MA, Brown SM (1981): Soft-tissue sarcomas and chlorinated phenols. Lancet 11:40. Jones G, Butler WH (1974): A morphological study of the liver lesions induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. J Pathol 112:93-7. Kearney PC, Woolson EA, Ellington CP Jr (1972): Persistence and metabolism of 32 4 l oJ7 chlorodioxins in soils. Environ Sci Technol 6(12):1017-29. Khera KS, Ruddick JA (1973): Polychlorodibenzo-p-dioxins; perinatal effects and the dominant lethal test in the Wistar rats. In: Chlorodioxins--Origin and Fate. Advances Chem Series 120, Blair EH (ed), American Chemical Soc, Washington, DC, pp 70-84. Kimbrough RD (1979): The carcinogenic and other chronic effects of persistent halogenated organic compounds. Ann NY Acad Sci 320:415-18. Kimmig J, Schulz KH (1957): Berufliche Akne durch chlorierte aromatische zyklische Ather. Dermatologica 115:540-46. Kocher CW, Mahle NH, Hummel RA, Shadoff LA: (1978): A search for the presence of 2,3,7,8-tetrachlorodibenzo-p-dioxin in beef fat. Bulln Environ Contamn Toxicol 19:229-36. Kociba RJ, Keeler PA, Park CN, Gehring PJ (1976): 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD); results of a 13-week oral toxicity study in rats. Toxicol Appl Pharmacol 35(3):553-74. Kociba RJ, Keyes DG, Beyer JE, Beyer JE (1978): Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicology and Applied Pharmacol 46:279-303. Kondorosi A, Fedorcsak I, Solymosy F, Ehrenberg L, Osterman-Golkar S (1973): Inactivation of Q(beta) RNA by electrophiles. Mutation Res 17:149-61. Lamb JC (IV), Moore JA, Marks TA (1980): Evaluation of 2,4-dichlorophenoxyacetic acid (2,4-D), 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) toxicity in C57BL/6 mice; reproduction and fertility in treated male mice and evaluation of congenital malformations in their offspring. National Toxicology Program, Report NTP 80-44. Lamm SH (Aug 6, 1979): An epidemiologic assessment of the Alsea II report. Lamm SH (Dec 10, 1980): (Cited in "Toxic materials in brief") Toxic Materials News :339. Lancranjan I, Popescu HI, Gavanescu O, Klepsch I, Serbanescu M (1975): Reproductive ability of workmen occupationally exposed to lead. Arch Environ Health 30:396-401. Mahle NH, Higgin HS (1977): Search for the presence of 2,3,7,8-tetrachlorodibenzo-p-dioxin in bovine milk. Bull Environ Contamn & Toxicol 18:123-30. March of Dimes Birth D efects Foundation (1981): Facts/1981. March of Dimes Birth D efects Foundation, New York. Matsumura F, Benezet HJ (1973): Studies on the bioaccumulation and microbial degra dation of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Environ Health Perspect (Exptl issue 5):253-58. Matthiaschk G (1978): Survey about toxicological data of 2,3,7,8-tetrachlorodibenzo-p-dioxin. In: Dioxin; Toxicological and Chemical Aspects. Cattabeni F, et al (eds), SP Med Sci Books, New York, p 170. May G (1973): Chloracne from the accidental production of tetrachlorodibenzodioxin. Br J Ind Med 30:276-83. McConnell EE, Moore JA (1976): The comparative toxicity of chlorinated dibenzo-p-dioxin isomers in mice and guinea pigs. Toxicol Appl Pharmacol 37:146. McConnell EE, Moore JA, Dalgard DW (1978): Toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in Rhesus monkeys (macaca mulatta) following a single oral dose. Toxicol Appld Pharmacol 43:175-87. McConnell EE, Moore JA, Haseman JK, Harris MW (1978): The comparative toxicity of chlorinated dibenzo-p-dioxins in mice and guinea pigs. Toxicol Appld Pharmacol 44:335-56. Moore JA, Gupta BN, Zinkl JG, Vos JG (1973): Postnatal effects of maternal exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Environ Health Perspect (Exptl issue 5):81-5. 33 Moses M, Selikoff IJ (1981): Soft-tissue sarcomas, phenoxy herbicides and chlorinated phenols. Lancet L1370. Murray FJ, Smith FA, Nitschke KD, Humiston CG, Kociba RJ, Schwetz BA (1977): Three-generation reproduction study of rats ingesting 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol Appl Pharmacol 35(l):139-40. Murray FJ, Smith FA, Nitschke KD, Humiston CG, Kociba RJ, Schwetz BA (1979): Three-generation reproduction study of rats given 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the diet. Toxicol Appld Pharmacol 50:241-52. Nash RG, Beall ML Jr (1978): Environmental distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) applied with silvex to turf in microagroecosystem. Final Report EPA-1AG-D6-0054, US Dept of Agriculture. National Academy of Sciences (1974): Report of the Committee on the Effects of Herbicides in South Vietnam; part A -- Summary and Conclusions. National Academy of Sciences, Wash, DC. National Research Council (1977): Drinking Water and Health. Natl Acad Sci, Wash DC. Neubert D, Dillmann I (1972): Embryotoxic effects in mice treated with 2,4,5-trichlorophenoxyacetic acid and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Naunyn-Schmiedeberg's Arch Pharmakol Exp Pathol Z7Z(3):243-64. Neubert D, Zens P, Rothenwallner A, Merker HJ (1973): A survey of the embryotoxic effects of TCDD in mammalian species. Environ Health Perspect (Exptl issue 5):233-40. Nielson K, Kaempe B, Jensen-Holm J (1965): Fatal poisoning in man by 2,4-dichlorophenoxyacetic acid (2,4-D); determination of the agent in forensic materials. Acta Pharmacol Toxicol ZZ:224-34. Norback DH, Allen JR (1973): Biological responses of the nonhuman primate, chicken and rat to chlorinated dibenzo-p-dioxin ingestion. Environ Health Perspect (Exptl issue 5):233-40. Norris LA (Feb 1978): Post-symposium responses to questions submitted to Logan Norris, in Symposium on the Use of Herbicides in Forestry, Arlington, Virginia, pp 95-97. Office of Science and Technology (Mar 1971): Report on 2,4,5-T; a Report of the Panel on Herbicides of the President's Science Advisory Committee. US Govt Printing Office, Wash DC. Oliver RM (1975): Toxic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin in laboratory workers. Brit J Ind Med 3Z:49-53. Olson JR, Gasiewicz TA, Neal RA (1980): Tissue distribution, excretion and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the Golden Syrian hamster. Toxicol Appld Pharmacol 56:78-85. Ott MG, Holder BB, Olson RD (1980):A mortality analysis of employees engaged in the manufacture of 2,4,5-trichlorophenoxyacetic acid. J Occup Med 22(l):47-50. Pazderova-Vejlupkova J, Lukas E, Nemcova M, Pickova J, Jirasek L (1980): Chronic poisoning by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Pracov Lek 32:204-09. Pazderova-Vejlupkova J, Nemcova M, Pickova J, Jirasek L, Lukas E (1981): The development and prognosis of chronic intoxication by tetrachlordibenzo-p-dioxin in men. Arch Environ Health 36(1):5-11. Piper WN, Rose JQ, Gehring PJ (1973): Excretion and distribution of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Environ Health Perspect (Exptl issue 5):241-44. Pocchiari F, Vittorio S, Zampieri A (May 1979): Human health effects from accidental release of tetrachlorodibenzo-p-dioxin (TCDD) at Seveso, Italy. In: Health Effects of Halogenated Aromatic Hydrocarbons, Nicholson WJ and Moore JA (eds). Ann NY Acad Sci 320:311-20. Poiger H, Schlatter Ch (1979): Biological degradation of TCDD in rats. Nature (London) 281:706-07. 34 Poland A, Glover E (1974): Comparison of 2,3,7,8-tetrachlorodibenzo-p-dioxin, a potent inducer of aryl hydrocarbon hydroxylase, with 3-methylcholanthrene. Mol Pharmacol 10:349-59. Poland AP, Smith D, Metter G, Possick P (1971): A health survey of workers in a 2,4-D and 2,4,5-T plant; with special attention to chloracne, porphyria cutanea tarda and psychologic parameters. Arch Environ Health 22:316-27. Purchase 3FH (1980): Inter-species comparisons of carcinogenicity. Br J Cancer 41:454-68. Ramsey JC, Hefner JG, Karbowski RJ, Braun WH, Gehring PJ (1979): The in-vivo biotransformation of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the rat. Toxicol Appld Pharmacol 48(l)(part 2):A162. Rappe C, Marklund S, Buser HR, Bosshardt H-P (1978):, Formation of polychlorinated dibenzo-p-dioxins and dibnzofrans (PCDF's) by burning or heating chlorophenates. Chemosphere 7:269. Rawls RL (Feb 12, 1979): Dow finds support, doubt for dioxin ideas. C&EN 57:23-29. Rawls RL, O'Sullivan DA (Aug 23, 1976): Italy seeks answers following toxic release. C&EN :23. Reggiani G (1979): The estimation of the TCDD toxic potential in the light of the Seveso accident. Arch Toxicol (Suppl 2):291-302. Reggiani G (Jul/Aug 1977): Toxic effects of TCDD in man. NATO Workshop on Ecotoxicology, Guildford, England. Reuber MD (1981): Carcinogenicity of picloram. J Toxicol Environ Health 7:207-22. Robinson BD (1979): Safety aspects of 2,4,5-T herbicides. Technote No. 5/79. Dept, of Agriculture and Fisheries, Adelaide, Australia. Rose JQ, Ramsey JC, Wentzler TH, et al (1976): The fate of TCDD following single and repeated oral doses to the rat. Toxicol Appl Pharmacol 36:209-26. Rylander R, Vesterlund J (1981): Carbon monoxide criteria; w/reference to effects on the heart, central nervous system and fetus. Scand J Work Environ Health 7 (suppl 1):29-31. Sanderson CA, Rogers LJ (1981): 2,4,5-Trichlorphenoxyacetic acid causes behavioral effects in chickens at environmentally relevant doses. Science 211:593. Sanger VL, Scott L, Hamdy A, et al (1958): Alimentary toxemia in chickens. J Amer Vet Med Assoc 133:172-76. Sauerhoff MW (1977): The fate of 2,4-dichlorophenoxyacetic acid (2,4-D) following oral administration to man. Toxicol 8(1):3-11. Sauerhoff MW, Chenoweth MB, Karbowski RJ, Braun WH, et al (1977): Fate of silvex following oral administration to humans. J Toxicol Environ Health 3:941-52. Schmittle SC, Edwards HM, Morris DA (1958): A disorder of chickens probably due to a toxic feed; preliminary report. J Amer Vet Med Assoc 132:216-19. Schulz KH (1957): Klinische und experimentelle Untersurchungen zur Atologie der Chloracne. Arch Klin Exp Derm 206:589-96. Schwartz L, Tulipan L, Birmingham DJ (1957): Occupational Diseases of the Skin (3rd ed). Lea & Febiger, Philadelphia. Schwetz BA, Norris JM, Sparschu GL, Rowe VK, et al (1973): Toxicology of chlorinated dibenzo-p-dioxins. Environ Health Perspect (Exptl issue 5):87-99. Seiler (1973): A survey on the mutagenicity of various pesticides. Experientia 29:622-23. Seiler JP (1977): Inhibition of testicular DNA synthesis by chemical mutagens and carcinogens; preliminary results in the validation of a novel short-term test. Mutation Res 46:305-10. Shadoff LA, Hummel RA, Lamparski (1977): A search for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in an environment exposed annually to 2,4,5-trichlorophenoxyacetic acid ester (2,4,5-T) herbicides. Bull Environ Contamn Toxicol 18(4):478-85. 35 19; Smith AH (Aug 1979): Seasonal analysis of Oregon data on spontaneous abortions and 2.4.5- T spraying. (Presented at ANZSERCH, Annual Conference; Dunedin, New Zealand). Sparschu GL, Dunn FL, Rowe VK (1971): Study of teratogenicity of 2,3,7,8-tetrachloro- dibenzo-p-dioxin. Food Cosxnet Toxic 9:405-412. Suit HD (1978): Sarcoma of soft tissue. Cancer 28:284-95. Sundstrom G, Jensen S, Janson B, Erne K (1979): Chlorinated phenoxyacetic acid derivatives and tetrachlorodibenzo-p-dioxin in foliage after application of 2,4,5-tri- chlorophenoxyacetic acid esters. Arch Environ Contamn Toxicol 8(4):441-48. Suskind RR (1980): TCDD contamination in the United States; case study. Paper presented at the International Workshop on Areawide Chemical Contamination, National Academy of Sciences, Washington, DC. ' Suskind RR (1980): Clinical manifestations of 2,4,5-T exposures. Paper presented at the International Workshop on Chemical and Epidemiological Follow-up After Areawide Chemical Contamination, National Academy of Sciences, Washington, DC. Suskind RR (Jul 1953): A clinical and environmental survey; Monsanto Chemical Co., Nitro, West Virginia. Report of the Kettering Laboratory. Sweeney GD, Jones KG, Cole FM, Basford D, Krestynski F (Apr 20, 1979): Iron deficiency prevents liver toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Science 204:332. Taylor JS (1979): Environmental chloracne; update and overview. Ann NY Acad Sci 320:295-307. Teleky L (1927): Die pernakrankheit (chlorakne). Klin Wschr 6:845-48;897-901. Thomas HF (1980): 2,4,5-T use and congenital malformation rates in Hungary. Lancet II(8187):214-15. Tomita M, Ueda S, Narisada M (1959): Dibenzo-p-dioxin derivatives. 27. Synthesis of polyholodibenzo-p-dioxin; Yakugaku Zasshi 79:186 (from Chem Abstr 53:1315d). Toth K, Sugar J, Somfai-Relle S, Bence J (1978): Carcinogenic bioassay of the herbi cide, 2,4,5,-trichlorophenoxyethanol (TCPE) with different 2,3,7,8-tetrachloro- dibenzo-p-dioxin (TCDD) content in Swiss mice. Prog Biochem Pharmacol 14:82-93. Toth K, Somfai-Relle S, Sugar J, Bence J (1979): Carcinogenicity testing of herbicide 2.4.5- trichlorophenoxyethanol containing dioxin and of pure dioxin in Swiss mice. Nature 278:548-49. Tulp MThM, Hutzinger O (1978): Rat metabolism of polychlorinated dibenzo-p-diox- ins. Chemosphere (9):76l--68. Tung TT, An TT, Tam ND, et al (1973): Le cancer primaire du foie au Vietnam. Chirurgie 99:427-36. Tung TT, Tugen BQTKA, Tra DX, et al (1971): Clinical effects of massive and continuous utilization of defoliants on civilians. Vietnamese Studies 29:53-81. US Environmental Protection Agency (Mar 1979): Decision and emergency order suspending registrations for certain uses (2,4,5-T and silvex). Fed Reg 44:15874-876. US Environmental Protection Agency (Oct 1980): Toxics removed. EPA Jour 6(9):31. US Surgeon General (1979): Smoking and Health. USDHEW Public Health Service. Office of Smoking and Health. DHEW Publication No. (PHS) 79-50066. US Govt Printing Office, Wash DC. Van Miller JP, Allen JR (1977): Chronic toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Fed Am Soc Exp Biol 36:396. Van Miller JP, Lalich JJ, Allen JR (1977): Increased incidence of neoplasm in rats exposed to low levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Chemosphere 6:537-44. Vinopal JH, Casida JE (1973): Metabolic stability of 2,3,7,8-tetrachloro- dibenzo-p-dioxin in mammalian liver microsomal systems and in living mice. Arch 36 1 933. Environ Contamn Toxicol 1:122-32. Wagner SL, Witt JM, Norris LA, Higgins JE, Agresti A, Ortiz M (1979): A Scientific Critique of the EPA Alsea II Study and Report. Environmental Health Sciences Center, Oregon State University, Corvallis, Oregon. Walker AE, Martin JV (1979): Lipid profiles in dioxin-exposed workers. Lancet 1:446-47. Ward CT, Matsumura F (1978): Fate of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in a model aquatic environment. Arch Environ Toxicol 7:349-57. Wassom JS, Huff JE, Loprieno N (1977/1978): A review of the genetic toxicology of chlorinated dibenzo-p-dioxins. Mutation Res 47:141-60. Whorton D, Krauss RM, Marshall S, Milby TH (1977): Infertility in male pesticide workers. Lancet 111259-61. Wilson JG (1973): Environment and Birth D efects. Academic Press, New York, p 49. Young AL, Calcagani JA, Thalken CE, et al (1978): The Toxicology, Environmental Fate and Human Risk of Herbicide Orange and its Associated Dioxin. USAF Occupational and Environmental Health Laboratory, Technical Report No. TR-78-92, Final Report, Brooks Air Force Base, Texas. Young AL, Thalken CE, Arnold EL, et al (1976): Fate of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the Environment: Summary and Decontamination Recommendations. US Air Force Academy, USAFA-TR-76-18, Colorado. Zack JA, Suskind RR (1980): The mortality experience of workers exposed to tetrachlorodibenzodioxin in a trichlorophenol process accident. J Occ Med 22(1):11-14. 37 13332 Table 1 -- Occupational and Industrial Accidental Exposures to TCDD in Plants Producing 2,4,5-trichlorophenol and/or 2,4,5-T Date 1949 Site Injured/exposed Monsanto (Nitro, W Va) 122/250 1952-3 Boehringer (W Germany) 37/? 1953 BASF (Ludwigshafen, Ger) 55/75 1954 Boehringer (Hamburg, Ger) ?/31 1956 Rhone-Poulenc (France) 17/? 1956 Hooker (US) ?/? 1956 Diamond Alkali (N Jersey) 29/? 1963 Philips (Netherlands) 30/106 1964 Spolana (Cz echoslovakia) ?/78 Remarks Uncontrolled reaction. Some studies completed, others underway. 122 chloracne subjects used for cohort study: 32 deaths (all causes) vs 46.4 expd.; no excess deaths from malignant neoplasms or circulatory disease (Ashe et.al, 1949, 1950; Suskind, 1953; Zack et al, 1980). Occupational. Outcome unavailable (Bovey et al, 1980). Uncontrolled reaction. 17 deaths (11-25 expd.) (4 GI cancer -- 3 stomach, 1 colon -- and 2 oat-cell lung cancers). Continuing 25-yr follow-up. Most prominent were polyneuritis, impaired senses and liver damage (Huff et al, 1980; IARC, 1977). Occupational. Outcome unavailable (IARC, 1977; Young et al, 1978). Uncontrolled reaction. 17 chloracne cases, also hyperlipidema and hypercholesterolemia (Bovey et al, 1980; IARC, 1977). Probably occupational. Outcome unavailable (Young et al, 1978). Probably occupational. 29 chloracne subjects studied, 11 had porphyria cutanea tarda (Bleiberg et al, 1964; Poland et al, 1971). Uncontrolled reaction. 44 chloracne cases (42 severe, 21/42 had internal damage or CNS disturbances; viz, polyneuritis, impaired senses and liver damage); 8 deaths (6 possible myocardial infarctions); some fatigue; no liver damage. Full report planned (Huff et al, 1980). Occupational. 78 chloracne, 5 deaths (2 bronchogenic cancer, 1 atypical atherosclerosis, 1(?) liver Al f 1964 USSR 128/ 1964 Dow Chemical (Midland) /204 /61 1966 Rhone-Poulenc (France) ?/? 1968 Coalite (United Kingdom) ?/90 1974 Bayer (W Germany) 5/? 1972-3 Linz Works (Austria) 50/? 1976 ICMESA (Seveso, Italy) /156 cirrhosis, 1 industrial intoxication acci dent); many of remaining 50 studied over 10 yrs have hypertension, hypercholester olemia, prediabetes and hyperlipidemia; sig nificant amount of porphyria cutanea tarda and severe hepatic and neurologic damage. Study to continue (Huff et al, 1980; Pazderova-Velupkova et al, 1981). Occupational. 128 cases of skin lesions, 69 of 83 examined had "acne"; 18 with neurasthenic syndrome; many with liver impairment (esp, among those with severe skin lesions) (Bovey et al, 1980; IARC, 1977). Occupational. Mortality less than expected, comparable to US white males: from all causes, 11 vs 20.3; from malignant neoplasms, 1 vs 3.6; from cardiovascular diseases, 4 vs 9.1. No chloracne or PCT (Ott et al, 1980). Deaths (all causes) not elevated (4 vs 7.8 expd.); deaths (all cancers) elevated (3 vs 1.5 expd.) but no predominant type or site (Cook et al, 1980). Uncontrolled reaction. 21 chloracne cases (Bovey et ad, 1980). Uncontrolled reaction. 79 chloracne cases; 1 death (coronary thrombosis). Cohort study planned. Company refuses to divulge any more information (Bovey et ad, 1980; May, 1973, 1981). Occupational. Outcome unavaiilable (Young et al, 1978). Occupational. 5 chloracne cases; outcome unavailable (Young et al, 1978). Uncontrolled reaction. Over 500 residents treated for presumed toxic symptoms, 134 confirmed chloracne cases (Bovey et al, 1980). Long-term morbidity study underway -- overall mortality rate normal so fair. Cancer and mortality registries should be formed. A2  f\ r O rJ 19336 * u Toxicology polychlorinated Biphenyls (PCBs), pjbenzo-p-Dioxins (PCDDs), pibenzofurans (PCDFs), and Related Compounds: Environmental and Mechanistic Considerations Which Support the Development of Toxic Equivalency Factors (TEFs) Stephen^4fe, D.Phil. ' ABSTRACT Halogenated aromatic com pounds, typified by the polychlor inated d ib en zo -p -d io x in s (P C D D s), d iben zofu ran s (P C D F s), biphenyls (P C B s). and diphenylethers (P C D E s), are industrial compounds or byproducts w hich have been w idely identified in the environm ent and in c h e m ic a l-w a ste d u m p sites. H alo genated arom atics are invariably present in d iverse analytes as highly com p lex m ixtures o f isom ers and con gen ers and this complicates the hazard and risk assessm ent o f these com pounds. Several studies have confirm ed the com m on receptormediated m echanism o f action o f to x ic halogenated arom atics and this has resulted in the d ev elo p m en t o f structure-activity relationships for this class o f ch em icals. The m ost toxic hal ogenated aromatic is 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and based on in vivo and in vitro stu d ies the relative toxicities o f individual halogenated arom atics have been de termined relative to T C D D ( i .e . , toxic eq u ivalen ts). T he d e rived toxic equivalents can be u sed for hazard and risk as sessment o f halogenated aromatic mixtures; moreover, for more complex mixtures containing congeners for w hich n o standards are available ( e .g ., b rom o/ch loro m ix tu res), several in vitro or in vivo assays can be u tilized for hazard or risk assessm en t. Key Words: 2 ,3 ,7 ,8 -T C D D , h alo g en a ted arom atics, toxic equivalency factors. I. INTRODUCTION Polyhalogenated aromatic com pounds such as the polychlor inated biphenyls (PC B s), polychlorinated naphthalenes (PC N s), polychlorinated benzenes, polychlorinated phenols, polychlor inated terphenyls (P C T s), p olyb rom in ated b ip h en y ls (P B B s), and chlorinated p h en o ls, a n ilin es, and b en zen es are industrial chemicals which have been utilized for diverse com m ercial applications.17 For ex a m p le, P C B s w ere w id e ly used as heat transfer fluids, organic d ilu en ts, p lasticizers, lubricant inks, fire retardants, paint a d d itiv es, se a lin g liq u id s, im m ersion o ils , adhesives, dedusting agents, lam inating agents, w axes, and as dielectric fluids for capacitors and transformers. PC Ns and PCTs have enjoyed lim ited use as PCB replacement com pounds and PBB s w ere prim arily used as flame retardants. The m on o c y c lic h a lo g en a te d c o m p o u n d s are important in term ed iates in the synthesis o f a w id e array o f com m ercial chem ical products and the chlorinated phenols have been extensively used as fungicides. A num ber o f other com pounds, including the po lychlorinated dibenzo-p-dioxins (PC D D s), dibenzofurans (PC D Fs), alkylated PC D Fs, polychlorinated diphenyl ethers (PCDEs), polychlorinated hydroxydiphenyl ethers; polychlor inated azo- and azoxyb en zenes,- polybrom inated d ibenzo-pdioxins (P B D D s) and dibenzofurans (PB D Fs). and m ixed brom o/chloro dibenzofurans and dibenzo-p-dioxins have also been identified as (1) byproducts w hich form during the syn thesis o f the primary industrial halogenated aromatics (2) by products from other com m ercial processes, and (3) byproducts of com bustion. M ost m em bers o f this fam ily o f chem icals exhibit several com m on properties. T h ese com pounds tend to be highly li pophilic and the degree o f lip op hilicity is increased with in creasing ring chlorination. W ith few exceptions, the halogen ated arom atics w h ich contain m ore than one phenyl ring are also stable and resistant to breakdown by acids, bases, heat, and hydrolysis; these physicoch em ical characteristics are im portant factors w hich have contributed to the diverse industrial applications o f these com pounds. Not surprisingly, these prop erties also contribute to the w idespread environm ental problem s associated w ith certain m em bers o f this class o f ch em icals, particularly the P C B s, P C D D s, and PCDFs w hich have been identified in alm ost every com partm ent o f the global e c o sy s tem . 5-6-8'33 M o r eo v e r, b e c a u se o f the lio p h ilic nature o f the PCBs, P C D D s, and PC D Fs, these com pounds bioaccum ulate in the fo o d ch a in , and re sid u e s h ave b een d etected in fish, w ild life , and hum an a d ip o se tissu e , m ilk , and se ru m .34'62 Public and regulatory concern over the potential adverse human health effects and environm ental damage associated with exposure to h alogenated arom atics has been heightened by several in cid en ts in w h ich p oten tially higher le v e ls o f hum an exposure have been reported. T h ese include agricultural and occup ational ex p o su re s to h alo g en a ted arom atics.61-63'96 PCB fires,97 and se v er a l a ccid en ta l ex p o su res w h ich in clu d e the Y usho and Y u C h en g p o is o n in g s .98 IW the e x p lo sio n in S e v e s o , 105' 110 and the e x p o su re s a sso c ia ted with L ove C a n a l ," 1 chlorinated p h e n o x y h erb ic id e s used in V ietn am (A g e n t O r a n g e ),112' " 5 T im e s B e a c h , and the related ex p o su res in M is s o u r i," 7' 121 and the P B B fo o d ch ain con tam in ation in M ich i g a n .122' 127 T h e se in cid en ts, co u p le d w ith frequent press reports S te p h e n S a fe , D .P h il., Department of Veterinary Physiology and Phar macology, The Texas Veterinary Medical Center. College of Veterinary Medicine. Texas A&M University. College Station. TX 77843-4466. jCn fjc4 1990 / n Tb)( i cc/oj^cj 2 t C I)J V V 1 tilb'P r*? 51 V Critical Reviews In o f PCB fires, numerous sm all spills and accidents, and con tinuing disposal problem s, have generated considerable pres sure to develop and validate m ethods for the hazard and risk assessm ent o f toxic halogenated arom atics, with particular em phasis on the P C D D s, P C D F s, and PC B s. The problem s as sociated with this process are num erous and are d iscu ssed in this review article. II. FACTORS WHICH AFFECT THE HAZARD AND RISK ASSESSMENT OF TOXIC HALOGENATED AROMATICS A. Sources of Halogenated Aromatics The structures o f the p olycyclic halogenated arom atics (F ig ure 1) sh ow that all o f th ese co m p o u n d s have m u ltip le sites for halogen substitution in w hich there can be both d ifferen ces in substitution patterns and the d egree o f halogenation. T able 1 summarizes the number o f p ossible isom ers and congeners for the major classes o f p olycyclic halogenated aromatics. Thus, there are 209 possib le P C B s, P C D E s, and azo/azoxy b enzenes, 135 PCDFs, and 75 PC D D s and P C N s. T he number o f possible polychlorinated o -, m -, and p-terphenyls are 3043, 3155, and 1951, respectively, and the number o f possbile m ixed bromo/ chloro halogenated arom atics are exceedingly high (5020 p os sib le c o m p o u n d s).128 N ot surprisingly, the com position o f the com m erial PCB and P C N p rod u cts are h ig h ly c o m p le x . A recent stu d y 129 on the high resolution capillary gas chrom atographic analysis o f the com m ercial PC Bs revealed that 132 individual congeners were present in these m ixtures and the relative number o f congeners and their con cen tration s in the com m ercial products is d ep en dent on the degree o f chlorination o f the mixture. The com m ercial P C N m ixtu res are a lso c o m p le x ,130 h o w e v e r , the high resolution analysis and indentification o f the individual con geners has not been reported. T he halogen ated arom atics w hich are form ed as byproducts in the sy n th esis o f industrial ch em icals or as byproducts o f com bustion are, with few exceptions, com plex m ixtures o f isom ers and congeners. S om e o f the important industrial sources o f P C D D s and P C D F s are sum m arized in T ab le 2 and these include their form ations as byproducts in the productions o f P C B s ,3-5,6 ch lorin ated p h e n o ls and ch lorinated p h en o l-d eriv ed c h e m ic a ls ,131-132 h e x a c h lo r o b e n z e n e ,133 tech n ical h ex a ch lo ro c y c lo h e x a n e s ,134 and ch lo rid es o f iron, alu m in u m , and c o p p e r .135 P C D F s and P C D D s h ave a lso recently b een id en tified in efflu en ts, w a stes, and pulp sam ples from the pulp and paper in du stry, and in fin ish ed paper p r o d u c ts.136"142 In 1 9 7 7 , H u tzin ger and co-w ork ers first identified the pres en ce o f P C D D s and PC D Fs as trace com ponents in extracts o f fly ash and in flue gas em issions from m unicipal w aste incin erators in the N e th e r la n d s.143 A n a ly tic a l stu d ies in se v er a l lab oratories have sin ce identified PC D D s and PCDFs em issions from m u n icip a l incinerators in all parts o f the w o r ld .231144` 147 The levels o f these toxins which are em itted are highly variable and dependent on a number o f factors including the incinerator design operating conditions and the nature o f the w aste prod u c ts .148"151 S ev er a l stu d ies h ave a lso sh ow n the form ation o f FIGURE 1. Structures of the polycyclic halogenated aromatics. Table 1 Multiplicity of PCB, PCN, PCDE, PBB, PCDF, PBDF, PCDD, and PBDD Isomers and Congeners No. of Cl I 2 3 4 5 6 7 8 9 10 Total atoms PCBs/PBB 3 12 24 42 46 42 24 12 3 1 209 and PCDEs PCDFs/ 4 16 28 38 28 16 4 1 135 PBDFs PCDDs. 2 10 14 22 14 10 2 1 75 PBDDs and PCNs Table 2 Identification of PCDDs and PCDFs as Byproducts in Industrial Chemicals and Industrial Processes Industrial products containing PCDFs Phenoxy alkanoic acid compounds Chlorinated phenols Hexachlorobenzene Trichlorobenzene and tetrachlorobenzene Polychlorinated biphenyls Chlorides of iron. aluminum and copper Technical hexachlorocyclohexanes Pulp and paper industry Use Herbicides Wood preservative Fungicide and industrial waste Industrial use Industrial use Industrial use Insecticide Byproducts Ref. 131, 132 131. 132 133 303 3, 5, 6 135 134 136-- 142 52 Volume 21, Issue 1 T o x ic o lo g y pCDDs and PCDFs in home heating systems which use both coal and wood as fuel sources.152'133 Table 3 summarized the relative concentrations of PCDDs and PCDFs in soot from municipal waste incinerators and from home heating inciner ators and higher levels have generally been observed in the municipal waste incinerators. The "Trace Chemistries of Fire Hypothesis"143 suggested the PCDFs and PCDDs can arise from diverse combustion processes which include "natural" forest fires and volcanic eruptions. However, results from the analysis of dated aquatic sediment profiles have shown that there were time-dependent increases in the concentrations of PCDDs and PCDFs which correlated with the increased pro duction and use of chlorinated industrial compounds.29'33These results confirm that the production of industrial halogenated organic chemicals and their subsequent incineration contribute more to the accumulation of PCDDs and PCDFs in sediments than do natural combustion processes. The mechanisms associated with the heat-induced formation of PCDDs, PCDFs, and related compounds have been exten sively investigated.136"139 For example, in a flow reactor ex periment which utilized methane gas, hydrochloric acid, and oxyen, over 300 compounds were formed and these included alkanes, alkenes, alkynes, aromatic hydrocarbons, and chlor inatedaromatic hydrocarbons including PCDDs and PCDFs.160 In addition, the combustion or pyrolysis of a wide spectrum of chemicals can yield PCDDs and PCDFs and these include PCBs,161'163 chlorophenates165,166 and chlorinated phenol-de rived products,132 polyvinyl chloride and chlorinated al kanes,167168 and chlorinated benzenes.164 In addition, the py rolysis of brominated diphenyl ethers and biphenyls also gave complex mixtures of PBDDs and PBDFs as byproducts.169'171 Table 3 Concentration of PCDD/PCDF in the Soot of Home Heating (ppb)152'155 Compound TetraCDDs PentaCDDs HexaCDDs HeptaCDDs OCDD TetraCDF PentaCDF HexaCDF HeptaCDF 0CDF Municipal waste incinerator fly ash I9.06 37.25 115.49 275.88 598.69 79.45 120.30 116.34 108.24 42.90 Home heating Spot oil Soot coal/wood 3.90 0.41 3.06 1.37 1.05 28.91 16.55 6.24 1.78 0.33 1.54 5.13 5.20 2.76 2.37 50.80 30.04 11.67 3.23 0.53 2.3,7,8-TCDD 2.3.7,8-TCDF 0.60 2.47 0.10 0.21 1.07 1.92 Halogenated aromatic compounds (including several bromi nated chemicals) have been identified in the emissions from automobiles using leaded gasolines.172 Dibromoethane and dichloroethane are added to leaded gasolines and act as scav engers to prevent the deposition of lead compounds in engines. These compounds are not added to unleaded gasoline and not surprisingly no halogenated aromatics were emitted from au tomobiles using unleaded gasoline.173Mixed bromo/chloro dibenzofurans and dibenzo-p-dioxins have been identified as combustion byproducts from muncipal waste incinerators and this reflects sources of bromine in some waste streams. Rappe and co-workers have compared the GC-MS profiles of the PCDDs and PCDFs derived from several combusion sources.24,23 144 174 Their data indicate that for extracts from several diverse combusion samples there was a remarkable qualitative similarity between their GC-MS fragmentograms for various isomer groups. These results suggest that there are common pathways for the formation of PCDDs and PCDFs during the combusion or pyrolysis of organic materials. B. Environmental Impact of Halogenated Aromatics It is apparent from the discussion noted earlier that PCDDs and PCDFs are formed from diverse processes which can result in the introduction of complex mixtures of these compounds into the environment. The relative contribution of these sources and other unidentified sources to the overall input of PCDDs and PCDFs into the environment is unknown. In contrast, environmental contamination by the commercial PCBs is largely the result of careless use and disposal practices. PCBs, PCDDs, and PCDFs have been indentified as contaminants in almost every compartment of the ecosystem and, without exception, the PCB levels are significantly greater than the PCDDs plus PCDFs. In addition, these halogenated aromatics are routinely identified as complex mixtures of isomers and congeners in extracts from most environmental samples. PCBs, PCDDs, and PCDFs have been identified in ambient air samples from diverse locations.14"19,21'27,175'178 For exam ple, total PCDD and PCDF levels in air from rural locations in Germany or the U.S. are generally <0.1 to 1pg/m3, whereas these values can rise significantly depending on the degree of industrial and urban activities which generate combusion-derived PCDDs and PCDFs. It was also apparent from analysis of these atmospheric samples that they contained complex mix tures of isomers and congeners which varied in their relative concentrations. In one study, the atmospheric levels of several halogenated hydrocarbons were determined around Kobe, Japan177 and the average concentration of PCBs was 2800 pg/m3, whereas the PCDD and PCDF levels were 8.6 and 8.8 pg/m3, respectively. It was stated that the PCB levels were probably due to emissions from sites where commercial PCBs were spilled or dumped; in contrast, the atmospheric PCDDs and PCDFs were combustion-derived since the distribution of congeners in air samples and fly ash samples were similar. 1990 9 3 9 9 53 Critical Reviews In PCDDs and PCDFs have also been widely identified in sedi ment samples and the composition and dated profiles of these samples suggest that the contaminants are combusion-derived.29'33 The identification of PCBs in fish, wildlife, and human tis sues has now been reported for over 2 decades8'21-34'44 and the results of these analytical studies led to the ultimate ban on further use and production of these compounds. The precise congener compostion of PCB extracts from biota samples is variable and depends, in part, on which specific commercial PCB preparations were associated with a contaminated area.13 High resolution analysis of the PCBs in a composite human milk sample from Michigan has been reported.45 Not surpris ingly, the PCBs found in the human milk sample were highly complex and the congener composition and their relative con centrations did not resemble any of the commercial PCB prep arations. This fact raises obvious problems with regard to the hazard assessment of PCB mixtures and is discussed later in this review. PCDDs and PCDFs have also been widely identified in en vironmental biota and in human milk, serum, and adipose tissue. Rappe and co-workers46identified PCDFs in fat samples from snapping turtle (Hudson River) and grey seals from the Gulf of Bothnia, and PCDDs and PCDFs have been identified in fish and wildlife samples throughtout the global aquatic and marine environments.47'56 Congener-specific analysis has shown that the 2,3,7,8-substituted PCDDs and PCDFs were the major compounds present in most sample extracts, however, the rel ative concentration of the individual components were variable due to the different composition of the sources of the PCDD and PCDF pollutants. In a recent study,53 the analyses of PCDDs and PCDFs in fish samples from several of the Great Lakes has been reported. It was apparent from the results that the PCDD and PCDF levels were lowest in fish from Lake Superior and Lake Erie and highest in fish from Lake Ontario (Table 4). The results also showed that there were higher levels of total PCDFs than total PCDDs in all the fish extracts from Lake Erie, Lake Superior, Lake Huron, Lake Michigan, and Lake St. Clair. In addition, the predominant PCDF congeners were 2,3,7,8-TCDF, 1.2.3.7.8- pentaCDF, and 2,3,4,7,8-pentaCDF; the predomi nant PCDDs were 2,3,7,8-TCDD, 1,2,3,7,8-pentaCDD, and 1.2.3.6.7.8- hexaCDD. These compounds were also the major PCDD and PCDF congeners in the Lake Ontario fish extracts, however, in these extracts the total PCDD and PCDF levels were comparable. This was primarily due to the unusually high concentrations of 2,3,7,8-TCDD; these results suggested that Lake Ontario contains additional point sources of 2,3,7,8-TCDD which are probably associated with 2,4,5-trichIorophenol-derived chemical wastes which contain 2,3,7,8-TCDD as the major byproduct. PCDF and PCDD residues have been extensively identified in human adipose tissue, blood serum, and milk in individuals in many different countries.23-57'62 In most samples, the overall PCDD levels are higher than the concentration of PCDF, and OCDD is the dominant congener in most samples. Table 5 summarizes the PCDD and PCDF levels in human adipose tissue and milk samples from several countries. These data illustrate that only the toxic 2,3,7,8-substituted congeners are present in these samples and it is evident that the bioconcen tration potential of individual PCDD and PCDF congeners are very different in humans compared to fish (i.e., see Tables 4 and 3). Based on the distribution of PCDDs and PCDFs in human tissue, it is not possible to assign specific sources for these compounds, although it has been suggested that combustion processes are major contributors to the PCDDs and PCDFs which are bioavailabie.25However, it should also be noted that PCDDs and PCDFs have been reported in a variety of foods including fish, domestic food animals, cow's milk, butter, chickens, and eggs179 and these foods also contribute to the relative body burden of PCDDs and PCDFs in human tissues. ill. HALOGENATEO AROMATIC HYDROCARBONS: COMMON BIOCHEMICAL AND TOXIC RESPONSES AND MECHANISM OF ACTION The development of hazard assessment methodologies for a single toxin, a mixture of structurally related congeners, or a complete family of compounds such as the halogenated aro matic industrial compounds (Figure 1) requires an understand ing of the toxic and biochemical responses elicited by these chemicals and their mechanism of action. There are at least four possible mechanistic scenarios for the halogenated aromatic family of chemicals (Figure 1) and these include (1) members of the same class of compounds (e.g., PCDDs) elicit different responses via different mechanisms; (2) some (or all) members of the same class of compounds elicit similar responses via acommon mechanism; (3) some (or all) members of two or more different classes (e.g., PCDDs and PCDFS) within this family cause different responses via different mech anisms; (4) some members of all classes of the halogenated aromatic family of chemials elicit similar responses in animals and cells via the same mechanism of action. The successful development of methods for the hazard and risk assessment of the complex mixtures of halogenated aromatic compounds is directly dependent on the mechanisms of action of these com pounds. It is evident that if each individual congener or class of compounds cause different responses and act via indepen dent mechanisms, then the relative toxicities and carcinogen icities of every congener must be determined. This would be a formidable if not impossible task due to the highly complex mixtures of halogenated aromatics which are introduced into the environment and which have been detected in environ- 19400 54 Volume 21, Issue 1 -- Toxicology Table 4 Levels of 2,3,7,8-Substltuted PCOOs and PCDFs In Great Lakes Fish53 Congener 2.3,7,8-TCDF 1,2 ,3 ,7 ,8-PenuCDF 2 3 ,4 ,7 ,8-PentaCDF 1 2 .3 .4 .7 ,8 -HexaCDF 1,2 ,3 .6 .7 .8 -HexaCDF 1,2,3 .7 .8 .9 -HexaCDF 2,3 ,4 ,6.7,8 -HexaCDF 1.2 ,3 ,4 ,6 ,7 ,8 -HeptaCDF OCDF 2,3,7,8-TCDD 1.2 ,3 .7 ,8 -PentaCDD 1,2,3,4,7,8 -HexaCDD 1,2 ,3.6,7,8 -HexaCDD 1,2 ,3 ,7 ,8 ,9 -HexaCDD 1,2 ,3 ,4 ,6 ,7 ,8-HeptaCDD OCDD Lake Michigan 31--39 3.9--8.0 8.1-- 14 0.9--2.4 0.7-- 1.9 <0.08 0.3--4.0 0.5-- 1.7 <6.7 3.0--4.4 7.0-- 12 0.5-- 1.2 4.1-- 10 0.5-- 1.3 0.7-- 1.1 0 .8-- 1.2 Lake Erie 8 .8 -- 2 0 0.9--2.5 1 .2 -- 5.0 0.1--0.3 0.1--0.5 < 0.2 0 .2-- 1.2 0.1-- 2.4 < 0.2 1 .2 -- 3.2 1.8 --5.6 0.1--0.4 1.0 -4 .0 0.2--0.5 0.8-- 1.5 1. 1--5.1 Range of values (ppt) Lake Superior 5.7--22 0.6--2.4 1.0 --3.8 0.1-- 1.3 0.08--0.6 < 0.1 0.06--0.8 0.08-- 1.1 <1.7 0.3-- 1.4 0.7-- 4.2 0.1--0.5 0.4--2.0 0.1--0.7 0.4-- 1.2 0.4-- 1.2 Lake Huron 19--31 3.3--9.2 6.9-- 19 0 .6--3.0 0.5--2.1 < 0.1 0 .8--2 .8 0 . 1-- 1 .0 < 0 .2 4.5-- 12 5.3-- 15 0.4--0.9 2 .6 --5.2 0.5--0.8 0 .3 -- 1.3 0 .6-- 1.1 Lake Ontario 13-- 30 3.2--5.9 17--26 5.5--21 1.0 --3.0 < 0.2 0.6--2.3 0.4-- 1.5 < 1.8 37-- 45 7.4-- 10 0.3--0.7 3.4--6.3 0.3--0.6 0.5-- 1.7 0.9--1.8 mental and biological samples. However, there is strong evi dence that several compounds within each class of the halogenated aromatic family elicit comparable responses in cells, laboratory animals, and humans. The following discussion briefly summarizes the evidence which supports acommon mechanism of action for the toxic halogenated aryl hydrocarbons. A. Toxic Responses Several studies have reported that 2,3,7,8-TCDD and related toxic halogenated aromatics elicit a number of common toxic responses which include body weight loss, thymic atrophy, impairment of immune responses, hepatotoxicity and porphy ria, chloracne and related dermal lesions, tissue-specific hypoandhyperplastic responses, carcinogenesis, teratogenicity, and reproductive toxicity.180-186 Figures 2 and 3 illustrate the doseresponse toxic effects of selected representative halogenated aromatic hydrocarbons and show that selected congeners within eachclass elicit the same responses. The only major differences in these compounds are their relative toxic potencies. It should also be noted that the toxic effects of halogenated aromatics are dependent on the animal species or strain, their age, and sex. For example, the toxic halogenated aromatics such as 2,3,7,8-TCDD cause chloracne and related skin lesions in some strains of genetically inbred mice, rabbits, monkeys, and hu mans but not in other strains of mice, rats, guinea pigs, and hamsters. The interspecies differences in their response to hal ogenated aromatics is illustrated by the following LD values for 2,3,7,8-TCDD;187 188 in the guinea pig (0.6 to 2.0 (xg/kg), rat (22 to 45 (xg/kg), chicken (25 to 50 (xg/kg), monkey (70 (xg/kg), rabbit (115 (xg/kg), dog (100 to 200 |xg/kg), mouse (114 to 284 (xg/kg), bull frog (>1000 (xg/kg), and hamster (1157 to 5000 (xg/kg). B. Biochemical Responses One of the hallmarks of exposure to halogenated aromatics is their induction of both phase I and phase II drug-metabolizing enzymes in laboratory animals and mammalian cells in culture.180-186 2,3,7,8-TCDD and related toxic halogenated aro matics induce the cytochrome P-4501A1 and cytochrome P4501A2 hemoproteins and their associated microsomal mon ooxygenases, which include aryl hydrocarbon hydroxylase (AHH) and ethoxyresorufin O-deethylase (EROD) activities. In addition, several PCB and PBB congeners induce other forms of cytochrome P-450 (in the rat) and these include cytochromes P-450IIBI (cytochrome P-450b), P-450IIB2 (cytochrome P-450e), P-450IIA1 (cytochrome P-450a), and P450IIIA1 (cytochrome P-450p) and their associated monooxy genase enzyme activities.189-199 Toxic halogenated hydrocar bons also induce the phase II drug-metabolizing enzymes, glucuronosyl transferases, and glutathione S-transferases. Table 6 summarizes some of the major biochemical responses elicited by PCDDs, PCDFs, and PCBs/PBBs and these include the induction of diverse enzymes, the modulation of hormone re ceptor-binding activities, the alteration of thyroid hormone and vitamin A levels, and enzymes which metabolize steroid hor mones. However, not all of these responses have been observed for all classes of the halogenated aromatics. This is due to the extensive use of 2,3,7,8-TCDD as a prototype for investigating 1990 l/ini 55 Critical Reviews In Table 5 Levels of PCD D s and PCOFs In Human Tissues29 Congener Sample source and mean concentrations (ppt on fat basis) PCDFs Adipose tissue 2.3.7,8-TCDF 2,3,4,7,8-PentaCDF 1,2,3,4,7,8-HexaCDF 1,2,3,6,7,8-HexaCDF 2,3,4.6,7,8-HexaCDF 1,2.3,4,6,7,8-HeptaCDF OCDF Milk 2,3.7.8-TCDF 2,3.4,7.8-PentaCDF 1,2,3,4,7,8-HexaCDF 1,2.3,6,7,8-HexaCDF 2,3,4.6,7,8-HexaCDF 1.2.3.4,6,7.8-HeptaCDF OCDF Japan 9 25 15 14 8 nr nr FRG 2.6 22.9 8.2 6.6 3.3 6.4 22.8 Sweden 3,9 54 6 5 2 11 4 Sweden 4.2 21.3 4.7 3.4 1.4 7.4 3.2 FRG 0.9 44" 10 6.7 3.8 19.5 <1 Netherlands 2.8 79 8.9 10.3 6.4 39 ND PCDDs Adipose Tissue 2,3,7,8-TCDD 1,2.3.7,8-PentaCDD 1,2,3.6.7.8-HexaCDD 1,2,3,7,8,9-HexaCDD 1,2,3,4,6,7,8-HeptaCDD OCDF Milk 2,3,7,8-TCDD 1,2,3,7.8-PentaCDD 1.2,3,4,7.8-HexaCDD 1,2,3.6,7,8-HexaCDD 1,2.3,7.8,9-HexaCDD 1.2.3.4,6,7.8-HeptaCDD OCDD Japan 9 15 70 12 77 230 FRG <5 10.7 8.1 32.7 6.4 49.9 181.2 Sweden 3 10 15 4 97 414 Sweden 0.6 6.5 2.5 19 6.3 59.5 302 FRG 150 19.2 77 9.4 56 267 Netherlands 9.7 44 25 251 23 130 744 N ote: nr = not recorded; ND = not detected. FIGURE 2. Dose-response effects of 2,3,7,8-TCDD. 2.3.4,7.8-pentaCDF, 2,3,3',4,4',5-hexaCB, 1,3,6.8-TCDF. 3.3'4,4'-tetraCDE. and Arocior 1254 on the suppression of the splenic plaque-forming cell response to sheep red blood cells in C57BL/6 mice. the effects and mechanism of action of ail members of this large family of chemicals. C. Genetic Studies Poland, Nebert, and their colleagues have investigated the inducibility of hepatic microsomal AHH activities in geneti cally inbred aryl hydrocarbon (Ah) responsive and nonresponsive mice (typified by the C57BL/6 and DBA strains, respec tively) and their backcrosses.243'246Their results clearly showed at least one order of magnitude difference in the response of these strains to the induction of AHH activity by 2,3,7,8TCDD. In contrast, 3-methylcholanthrene was inactive as an FIGURE 3. Dose-response inhibition of body weight gain by 2.3.7.8-TCDD, 2.3,4,7,8-pentaCDF, 3,3'.4,4',5-pentaCB, 1,2,4,7,8-pentaBDD, and 2,3,3'. 4,4',5-hexa CB in immature male Wistar rats. inducer in the nonresponsive mice and genetic studies with inbred strains and their backcrosses showed that AHH induc ibility is inherited as a simple autosomal dominant trait in these strains. Based on these early studies, it was suggested that the induction of AHH activity by 2,3,7,8-TCDD may be a recep tor-mediated process and it was suggested that the decrease responsiveness of DBA/2 mice was due to a defect in the Ah 56 Volume 21, Issue 1 19402 Toxicology Table 6 Biochemical Responses Caused by Different Class of Halogenated Aromatics Response CYPIA1 induction CYP1A2 induction Modulation of steroidmetabolizing enzymes DT diaphorase induction Glutathione S-transferase induction UDP glucuronosyl transferase induction Epidermal transglutaminase induction Aldehyde dehydrogenase induction 8-Ammolevulinic acid synthetase induction Uroporphyrinogen decarboxylase inhibition Decreased EGF receptor binding Decreased progesterone receptor binding Increased Ah receptor binding activity Decreased estrogen receptor levels Decreased glucocorticoid receptor binding Modulation of thyroid hormone levels/function Decreased vitamin A levels Ref. PCDDs/PCDFs PCBs/PBBs 181.182.192.194, 195.198.199 181.182.192,194, 195,198,199 200--202.198,202 189--193, 196.198.199 189-- 193. 196,199 202 203,205 206 198 207--209 210,211 212 213 214 215 216.217 216,21--221 222,223.229 224,225 216,217, 219.220 226 227 224,225 228,229 230--234 235.236 237,238 239--242 receptor. This explanation of differential responsiveness in inbred strains of mice has been confirmed in a recent report which showed that the hepatic Ah receptor in DBA/2 mice has a lower affinity for [3H]-2,3,7,8-TCDD than the corresponding recep tor in C57BL/6 mice.247 Subsequent research with genetically inbred mice and their backcrosses has shown that for PCDDs, PCBs, and PCDFs several toxic responses including immunotoxicity, teratogenicity, body weight loss, hepatotoxicity, and porphyria segregate with the Ah locus.248"239 Moreover, some reports indicate that other genetic loci may be important forthe toxicity of 2,3,7,8-TCDD and related compounds.231233 D. Identification of the Ah Receptor Poland and co-workers260 first identified a cytosolic protein in C57BL/6 mouse liver which exhibited saturable and high affinity binding using [3H]-2,3,7,8-TCDD as a radioligand. Subsequent research in several laboratories have identified this receptor protein in both hepatic and extrahepatic tissues of several laboratory animal species, mammalian cells in culture, and in human organs and cultured cells.261 This research has helped to confirm the role of the Ah receptor in the mechanism of action of 2,3,7,8-TCDD and structurally related toxic hal ogenated aryl hydrocarbons. Some of the evidence which sup ports a receptor-mediated response mechanism for this family of compounds is briefly summarized below. It should be noted that the development of scientifically based toxic equivalency factors is directly related to the weight of the evidence which supports the proposed receptor-mediated mechanism of action. 1. Saturable Binding The synthesis of radiolabeled 2,3,7,8-TCDD with high spe cific activity triggered several important mechanistic studies; it was apparent that in the soluble fraction of hepatic and extrahepatic tissues from several species there is a protein which exhibited saturable binding with this radioligand.260"263 More over, several reports have also demonstrated that other ra diolabeled PCDD and PCDF congeners exhibited comparable saturable binding curves264 (see Figure 4). In addition, radiolabeled polynuclear aromatic hydrocarbons such as 3-methylcholanthrene, benzo[a]pyrene, and dibenz[a,h]-anthracene also exhibit saturable binding with this cytosolic receptor pro tein.263"267 2. High Affinity Binding The Kd values for the dissociation of the bound cytosolic receptor complex using both PCDD and PCDF radioligands varies from approximately 0.1 to 10 nM . This variability is due to a number of factors including the source of the cytosol (i.e., organ, species, sex, age, and strain of animal), the ex perimental conditions, and the receptor binding assay used in the studies. These variables, coupled with the apparent lower binding affinities of cytosols from human tissues and DBA/2 mice,247268 account for the wide range of the reported KD values. In addition, recent studies have reported that at least two factors, namely, the high level of nonspecific binding associated with many cytosol preparations and the degradation of unbound receptors during the assay, contribute to the under estimation of Kd values for the bound cytosolic Ah receptor complex.269270 If the above factors are taken into account, the actual Kd values may be as low as 10~12 M. 3. Tissue or Cellular Specificity and Correlation between Ugand Binding and Biological Response The responsiveness of a particular organ or cell to the effects of 2,3,7,8-TCDD and related compounds depends on the pres ence of a functional Ah receptor. This is illustrated by the effects of 2,3,7,8-TCDD on the induction of cytochrome P-4501A1 and associated monooxygenase enzyme activities in wild-type Hepa lclc7 cells and several mutant cell lines.271"273 In the wild-type cells, there are relatively high levels of the Ah receptor and 2,3,7,8-TCDD is an effective 1990 1 9 4 0 3 57 Critical Reviews In 2,3,7,8- [ 3 H] t c d d 2,3,7,8 - [ 3 H] TCDF 1,2,3,7,8 - [ 3 H] PCDD 1,2,3,7,8 - [ 3 H] PCDF FIGURE 4. Saturation binding of several 2,3.7.8-substituted radioligands to rat hepatic cytosol. (Courtesy of R. Rosengren. unpublished results.) inducer; whereas, class I variant cells, which contain relatively low levels of this receptor, are not inducible. Class II variant cells are also not inducible by 2,3,7,8-TCDD, however, the levels of the cytosolic Ah receptor in this cell line are com parable to those observed in wild-type cells. The nonrespon siveness of these cells is associated with a defect in the ac cumulation of transcriptionally active nuclear Ah receptor complexes. It is also evident from other studies that the pres ence of the Ah receptor is not sufficient for the induction of a biochemical response by 2,3,7,8-TCDD and related com pounds. For example, 2,3,7,8-TCDD did not induce AHH activity in MDA-MB-231 human breast cancer cells even though the Ah receptor complex accumulates in the nuclear fraction of these cells.276 Moreover, in genetically inbred haired and hairless mice, the effects of 2,3,7,8-TCDD are dependent on both the Ah and hr gene loci.253,277,278 A correlation between the formation of occupied nuclear receptor complexes and the expression of a biological response has also been reported in responsive target organs or cells. After administration of [3H]-2,3,7,8-TCDD to C57BL/6 mice, there was a time-dependent depletion of the cytosolic Ah re ceptor and this was paralleled by the appearance of the nuclear 2,3,7,8-TCDD:receptor complex; the formation of the nuclear receptor complex was followed by the subsequent induction of cytochrome P-4501A1 (or P|-450) mRNA and microsomal cy tochrome P-4501A l.279 In a recent study, it was reported that for selected radiolabeled PCDD and PCDF congeners there was a linear correlation between levels of occupied nuclear receptor complex vs. the magnitude of the induction (AHH) response in rat hepatoma H-4II E cells.280 These results clearly link responsivenes to 2,3,7,8-TCDD and related compounds to the accumulation of occupied nuclear 58 Volume 21, Issue 1 Toxicology ^ receptor complexes. Several groups have investigated the molecular biology of CYP1A1 gene expression181-182-281'24and their results clearly confirm the proposed receptor-mediated mechanismof action of 2,3,7,8-TCDD and related compounds, which is illustrated in Figure 5. Initial binding of the toxin to the cytosolic Ah receptor is followed by an activation or trans formation step and the subsequent accumulation of occupied nuclear receptor complexes. These nuclear complexes then in teract with specific DNA sequences or dioxin regulatory ele ments (DREs) which are located in the 5'-upstream region from the CYP1A1 gene. These interactions lead to the enhancement of CYP1A1 gene expression. It is assumed that many of the toxic effects elicited by halogenated aryl hydrocarbons are also the result of altered receptor-mediated gene expression, how ever, the molecular mechanisms of these responses are cur rently unknown. iv. DEVELOPMENT OF TOXIC EQUIVALENCY FACTORS FOR DIFFERENT CLASSES OF HALOGENATED AROMATICS A. Introduction The previous section of this review noted some of the evi dence which supports a receptor-mediated mechanism of action for 2,3,7,8-TCDD and related halogenated aromatic com pounds. One important additional criteria for receptor-mediated responses are the steroselective ligand-receptor interactions and the corresponding structure-activity relationships (SARs). Sev eral studies have reported the structure-receptor-binding rela tionships for different classes of halogenated aromatics in which the competitive in vitro binding ECjo (or IC) values for dif ferent congeners have been determined using [3H]-2,3,7,8TCDD as the radioligand and rodent hepatic cytosol as a source of the Ah receptor183"18J-260-283'289 (see Table 7). For the PCDDs, 2.3.7.8- TCDD was the most competitive ligand and the struc ture-binding relationships demonstrated that the most active compounds were the 2,3,7,8-substituted tetra- to hexaCDDs (see Figure 6). It was evident from the binding studies that (1) the most active congeners were substituted in all four lateral positions and the binding activity decreased with decreasing lateral substition and (2) for the 2,3,7,8-substituted congeners, there was a decrease in binding avidity with increasing non lateral chlorine substition. The structure-binding relationship for the PCDFs showed that substituents on all four positions in the ring contributed differentially to their binding avidities286-287 (Table 7). Like the PCDDs, the most active compounds were tetra- to hexaCDFs which are substituted in the lateral 2, 3, 7, and 8 positions (Figure 7). The dibenzofuran ring system pos sesses a single axis of symmetry and there are four different positions on each aromatic ring, namely, C-l (or C-9), C-2 (or C-8), C-3 (or C-7), and C-4 (or C-6). The most active congeners, namely, 2,3,4,7,8-pentaCDF (1.5 x 10-8 M ), 2.3.7.8- TCDF (4.1 x ICC8A/), 2,3,4,6,7,8-hexaCDF (4.7 x 10~8 A/), and 1,2,3,7,8-pentaCDF (7.5 x 10"8 A/) were all substituted in their lateral 2, 3, 7, and 8 positions. A comparision of the receptor binding EC values for the 2,3,4,7,8(1.5 x lO^A/), 1,2,4,7,8-(1.3 x 10-6 M ), and 1,2,4,6,8pentaCDF (>5 x 1CT6 M ) isomers demonstrates the impor tance of lateral chloro substituents since there is a decrease in receptor binding affinities with decreasing lateral substitution. Two pairs of PCDF isomers, namely, 1,3,4,7,8- (2.0 x 10-7 M) and 1,2,4,7,8-pentaCDF (1.3 x KC6 M)\ 2.3,4,7- (2.5 x 10-8 M ) and 2,3,4,8-TCDF (2.0 x 10~7 A/1 differ only with FIGURE 5. Proposed model for the mechanism of action of 2,3,7,8-TCDD and related toxic halogenated aromatic hydrocarbons. The initial formation of a cytosolic receptor complex is followed by an activation step, the formation of nuclear receptor complexes, and their interaction with specific nuclearbinding sites. This interaction contributes to an increase or induction of a 2,3,7,8-TCDD-inducibie gene such as CYP1A1. 1990 19405 59 Criticai Reviews In Table 7 Halogenated Aromatic Congeners as Competitive Ligands for the Rat Hepatic Cytosol Ah Receptor1*4,188 Congeners Receptor binding avidities (EC. A f) Congeners Receptor binding avidities (ECm. Af) PCDFs Dibenzofuran 2-MonoCDF 3-MonoCDF 4-MonoCDF 2,3-DiCDF 2,6-DiCDF 2,8-DiCDF 1,3,6,-TriCDF 1,3,8-TriCDF 2,3,4-TriCDF 2,3.8-TriCDF 2,6,7-TriCDF 2,3,4.6-TetraCDF 2,3,4,8-TelraCDF 2,3,6,8-TetraCDF 2,3,7,8-TetraCDF 1,2,4,8-TetraCDF 1,2,3,6-TetraCDF 1,2,3,7-TetraCDF l,3,4,7,8-PentaCDF 2,3,4,7,9-PentaCDF l . 2,3,7,9-PentaCDF 1.2,4,6,7-PentaCDF 1,2,4,7,9-PentaCDF 1,2,3.4,8-PentaCDF 1,2.3,7,8-PentaCDF 1.2,4,7,8-PentaCDF 2,3,4,7,8-PentaCDF 1,2,3,4,7,8-Hexa CDF 1,2,3,6.7,8-HexaCDF 1,2,4.6.7,8-HexaCDF 2,3,4,6,7,8-HexaCDF PCDDs 2,3.7,8-TCDD 1,2,3,7,8-PentaCDD 2,3,6,7-TetraCDD 2,3,6-TriCDD 1,2,3,4,7,8-FexaCDD 1,3.7,8-TetraCDD 1,2,4,7,8-PentaCDD 1,2,3,4-TetraCDD 2,3,7-TriCDD 2,8-TriCDD 1,2,3,4,7-PentaCDD 1,2,4-TriCDD OCDD 1-MonoCDD >io-3 2.8 x 1Or* 4.2 i 0.6 x IO"5 > io -J* 4.72 x 10* 2.46 x lCT* 2.57 x 10-* 4.40 x 10* 8.50 x 10-5 1.9 x IO"5 1.0 0.1 xlO * 4.5 x IO"7 3.5 x 10-7 2.0 x IO"7 2.2 x 10-7 4.1 0.6 x IO"* >io-5 3.5 x 10-7 1.1 x IO"7 2.0 x I0-7 2.0 x IO"7 4.0 x IO"7 6.77 x 10* 2.0 x 10-5 1.2 x 10-7 7.45 2.04 x 1(T* 1.3 x 10* 1.5 0.1 x lCT* 2.3 x icr7 2.7 1.0 x IO"7 8.3 x 10* 4.7 0.4 x 10* 1.0 x 10* 7.9 x Iff-* 1.6 x IO"7 2.2 x IO*7 2.8 x IO"7 7.9 x 10-7 1.1 x Iff-6 1.3 x IO* 7.1 x 10* 3.2 x IO"6 6.4 x IO"6 1.3 x IO-5 > 1.0 x IO"5 > 1.0 x IO-* Bromo/chloro dibenzo-p-dioxins 2,3.7,8-TBDD 2,3-Dibromo-7,8-dichlorodibenzo-p-dioxin 2,8-Dibromo-3,7-dichlorodibenzo-p-dioxin 2-Bromo-3,7,8-tri-chlorodibenzo-p-dioxin 2,4,6,8/1.3,7,9-TetraBDF 1,3,7,8-TetraBDD 1,2,4,7,8-PentaBDD 1,2,3,7,8-PentaBDD 2,3,7-TriBDD 2,7/2,8-DiBDD 2-MonoBDD PCBs 3.3',4,4',5-PentaCB 3,3',4,4',5,5'-HexaCB 3,3',4,4'-TetraCB 2,3,3',4,4'-PentaCB 2,3,4,4',5-PentaCB 2,3,3 ' ,4,4' ,5-HexaCB 2',3,4,4',5-PentaCB 2,3,3',4,4',5-HexaCB 2,3,4,4'-TetraCB Aroclor 1254 2,3,3',4,4',5,5'-HeptaCB 2,3',4,4',5-PentaCB 2,3',4,4',5,5'-HexaCB 2,2' ,4,4' ,5,5'-HexaCB 2,2',4,4'-TetraCB 2,3,4,5-TetraCB 1.50 x IO* 1.48 x IO-9 4.47 x IO-10 1.15 x 10* 9.30 x IO"6 2.00 x IO-9 1.70 x IO-1 6.61 x IO"9 1.17 x IO-9 1.55 x IO"6 2.95 x IO-7 1.2 x IO"7 Insoluble 4.3 x IO*7 7.1 x IO-6 4.1 x i r 6 7.1 x 10^ 1.4 x IO"5 5.0 x io* 2.8 x IO"5 6.0 x IO* Insoluble 9.1 x lO * 1.6 x IO"5 7.9 x IO"5 1.3 x IO* 1.4 x IO"* Note: The values for the PCDDs, PCBs and PCDFs were determined by sucrose density gradient analysis; the brominated compounds were determined using the hydroxylapatite assay. 19408 60 Volume 21, Issue 1 Toxicology and at least 2 meta positions (Figure 8). These congeners, namely, 3,4,4',5-tetra-, 3,3',4,4'-tetra-, 3,3,',4,4',5-penta, and 3,3',4,4',5,5'-hexachlorobiphenyl (CB), are approximate isostereomers of 2,3,7,8-TCDD in their coplanar conformations. A second set of PCB congeners, namely, the monoortho co planar analogs (Figure 9), exhibit lower competitive binding affinities for the Ah receptor (Table 7). Based on limited data, it was apparent that the structure-receptor-binding relationships for the PCBs and PBBs were comparable290 and only minimal results have been reported for other classes of halogenated aromatics. Cl Cl 1 2 , 3 . `*.6>7 ' 8 *h , P ta C D O Cl Cl 1,2 .3,4,6,7,8,9-octaC O O ^OURE 6. Structures of the 2.3,7.8-substituted PCDD congeners. ride I 'f i.J .3 ..7 .l-n *i*C O F 1.2.3.7.1.9-m laC O F 2,3.*,t.7.8ftiaCO F t.l.2.4.*.7.a-hptCDF 1 2 .3 .4 .7 J.9 .n a p ta C D F C c< 1.2 .3 .4 .6 .7 .8 .9 -o cta C O F FIGURE 7. Structures of the 2.3,7,8-substituted PCDF congeners. respect to their substitution of C-2 (or C-8) and C-3 (or C-7). Inboth cases, the C-3 (or C-7) -substituted compounds exhib ited relatively higher competitive binding affinities than the corresponding C-2 (or C-8) -substituted isomers. A compari son of the relative binding affinities of a series of C-l (or C9) and C-4 (or C-6) isomer pairs illustrates the higher binding activities of the isomer which retains the C-4 (or C-6) substi tuent. For example, the EC50 values for the 2,3,4,7-, 2,3,4,7,8-, 2,3,4,7,9-, and 2,3,4,6,7,8-substituted PCDFs were 2.5 x IO"8M , 1.5 x 10-* M , 2.0 x IO"7M , and 4.7 x 10-* Af, whereas the values for the coresponding C-l (or C-9) con geners (i.e., 1,2,3,7-TCDF, 1,2,3,7,8- and 1,2,3,7,9-pentaCDF, and 1,2,3,6,7,8-hexaCDF) were 1.1 x 10"7Af, 7.5 x NT* M, 3.4 x IO-7 A/, and 2.7 x IO-7 M , respectively. The structure-binding relationships for the PCBs283have shown that the most active compounds are substituted on both para o o FIGURE 8. Structures of the coplanar PCB congeners substituted in both p a ra and two or more meta positions. Cl CI Cl Cl Cl O Cl s^ A s^ C I Of C I^A ^C I Of O Cl 2`.3,4,4',5 2,3,4,4,,3 ci Cl O Cl 2,3,3;4,4` 2,3',4,4',5 2 ,3 ,3 `,4 ,< 5 2,3',4,4',5,5' 2,3,3',4,4,5I Cl . Ciy~h( ^ ^ ( ) ~ Lci LJ JO Cl 2,3t3, 4.< 5ts' FIGURE 9. Structures of the monoortho coplanar PCB congeners. Critical Reviews In The structure-binding relationships observed for the different classes of halogenated aromatics provide further support for a receptor-mediated mechanism of action for these compounds. Moreover, the results of several studies have demonstrated an excellent correlation between the structure-binding and struc ture-activity (biochemical and toxic) relationships for several classes of halogenated aromatics. 183' 186 This observation forms the mechanistic basis for the development of TEFs for indi vidual halogenated aromatic compounds. For the sake of con venience, most TEF schemes compare the relative toxicity of different compounds to the most potent member of this family of chemicals, namely, 2,3,7,8-TCDD. It will be noted in the following sections that the activity of a congener relative to 2,3,7,8-TCDD is dependent on several factors including the specific response, the species/sex/strain and age of the animals, the duration of the feeding study, and the route of compound administration. For some compounds, the TEF will be selected from a diverse spectrum of results whereas only minimal tox icologic data will be available for other congeners. D.L. Grant (Health and Welfare Canada) has suggested that when data are available from more than one response the TEF values should be derived from the following effects in descending order of priority: 1. Results obtained from long term carcinogenicity studies 2. Data from reproductive studies 3. The results of subchronic experiments which measure Ah receptor-mediated responses such as thymic atrophy, body weight loss, and immunotoxicity 4. Acute toxicity studies 5. In vivo or in vitro biochemical responses such as enzyme induction, receptor binding, etc. It is apparent from a review of the data that a complete set of data, including long-term carcinogenicity studies,291'294 are available only for 2,3,7,8-TCDD and therefore the assignment of TEF values for the halogenated aromatics requires a sub jective assessment of the results in which the response priorities noted above will be a contributing factor. B. PCDDs: Development of TEFs The relative acute and chronic toxicities of several PCDD congeners have been reported and it is apparent that the tetrato hexachloro 2.3,7,8-substituted analogs (Figure 6 ) are sub stantially more active than those congeners which contain less than four lateral substituents. 187188 For example, the acute LD50 values for 2,8-diCDD, 1,2,4,7,8-pentaCDD, and 1,3,6,8tetraCDD are, respectively, 8,470,000, >5000, and >2,987,000 |ig/kg in the mouse whereas the LDS0s for 2,3,7,8-TCDD were in the 114 to 284 p.g/kg range. Inspection of analytical results from most environmental samples (e.g., Tables 4 and 5) and human tissues indicates that the 2,3,7,8-substituted PCDDs are the major congeners present in these extracts. Therefore, based on the structure-toxicity relationships and the analytical data, TEFs have been developed only for the 2.3.7.8-substituted compounds. The ideal set of data from which to calculate the TEFs for PCDDs would be comprehensive cancer or multigenerational reproductive studies on all the 2,3,7,8-substituted compounds. Unfortunately, these data .are available only for 2,3,7,8TCDD291'294 and the results from these studies have been ex tensively utilized to regulate the dietary intake and permissable emission levels of 2,3,7,8-TCDD. However, since 2,3,7,8TCDD is a minor component of most environmental sources of PCDDs, TEFs for the other toxic congeners have been, de veloped from other toxicological endpoints. The results in Ta ble 8 summarize the various LDjo/EDsovalues for several toxic endpoints in several species for the 2,3,7,8-substituted PCDD congeners. These data illustrate that (1) 2,3,7,8-TCDD is clearly the most toxic halogenated aryl hydrocarbon, (2 ) the guinea pig is the most sensitive species to the effects of these toxins, and (3) the SARs are comparable within each animal species for every response. These results are not surprising and they are consistent with the proposed receptor-mediated mechanism of action for the PCDDs. The results in Table 9 summarize the relative biochemical and in vitro potencies of the 2,3,7,8-substituted PCDD con geners. These responses include receptor-binding affinities, keratinization, and the induction of AHH activity in rodents and in rodent cells in culture. The SARs for the biochemical responses elicited by 2,3,7,8-TCDD and related PCDDs are similar (but not identical) and the relative sensitivity of these effects are response- and bioassay-dependent. Figures 10 to 12 summarize the correlation between the --log EC50 values for AHH induction in rat hepatoma H-4-II E cells vs. the - log ED values for thymic atrophy, body weight loss, and hepatic microsomal AHH induction in the rat. The linear correlation coefficients for these responses were 0.73, 0.84, and 0.84 respectively. These data illustrate that the quantitative SARs for these congeners were comparable for several responses. Table 10 summarized recently proposed TEFs for the 2.3,7,8substituted PCDDs301'303 and compares these values with their relative potency ranges derived from the in vivo and in vitro responses summarized in Tables 8 and 9. The results show that the proposed TEFs overlap with the range of values for the relative in vivo and in vitro potencies of the 2,3,7,8-substituted PCDD congeners. Moreover, for most of these compounds, the assigned TEF values are conservative and correspond with the higher range of relative potencies. Some previous proposals for TEFs assigned values of zero for OCDD due to the low acute toxicity of this congener. However, recent studies by Couture and co-workers304 have reported that there was a marked increase in the absorption of this compound in rat liver and adipose tissue with n increasing number of doses. For example, animals given 65 doses of OCDD (5 doses/week for 13 weeks) exhibited marked increases in hepatic EROD activity and cytochrome P-450 levels and fatty vacuolizaton in the liver. The results of this study suggest 62 Volume 21, Issue 1 19408 J - Toxicology pablo 8 Comparative Toxic Potencies of the 2,3,7,8-Substltuted P C D D s187,1M Response Acute lethality Subchronic toxicities Body weight loss Thymic atrophy Immunotoxicity Teratogenicity Species (dose or cone.) Guinea pig (ptg/kg) Mouse (p.g/kg) Rat (p.g/kg) Rat (jimol/kg) Guinea pig (p-mol/kg) Rat (p.mol/kg) Mice (nmol/kg) Mice (p.g/kg) 2,3,7,8TCDD 0.6--2.0 114--284 22--45 0.05 0.0056 0.09 0.65, 0.77 3.4 1,2,3,7,8PentaCDD 3.1 337.5 0.62 -- 0.17 LD*/ED Values 1,2,3,4,7,8HexaCDD 1,2,3,7,8,9HexaCDD 1,2,3,6,7,8HexaCDD 72.5 825 60-- 100 >1440 70-- 100 1250 1.63 -- 1.07 -- -- -- -- 7.1 1,2,3,4,6,7,8HeptaCDD > 600 -- -- 85 OCDD -- >4 x to-6 -- >500 Table 9 Comparative Biochemical and In Vitro Potencies of 2,3,7,8-Substltuted PC D D s20-288,2**''300 Response System (conc./dose) 2,3,7,8TCDD 1,2,3,7,8PentaCDD EDm Values 1,2,3,4,7,8HexaCDD 1,2,3,7,8,9- 1,2,3,6,7,8HexaCDD HexaCDD Receptor binding AHH induction Mouse liver (AO Rat liver (AO Rat liver (p.mol/kg) H-4-II-E cells (AO Guinea pig liver (p.mol/kg) Chick embryo (nmol/kg) Keratinizations (nM) (XB/3T3 cells) 2.7 x IO-10 1.0 x IO"4 0.004 4.2 x IO-10 7.9 x 10-* 0.031 2.8 x 10-7 0.03 1.5, 7.2 x 10-" 22, 1.1 x 10-* 31, 2.1 x IO-9 189** 0.00028 0.2 0.9 0.05 0.1 1.4 X IO"9 128" 1.0 10 * Picomoles per piale. * Minimum dose effecting keratinization.299 1,2,3,4,6,7,8HeptaCDD OCDD > io-5 551s 2500s that if rats are exposed to multiple doses of OCDD, this com pound is approximately 1/100 to 1/1000 times less toxic than 2,3,7,8-TCDD. Hence, the value of 0.001 has been assigned to OCDD. C. PCDFs: Development of TEFs The results in Table 11 summarize the acute and chronic in vivo toxicities of all the 2,3,7,8-substituted PCDFs. It is evident from these results that 2,3,4,7,8-pentaCDF is the most highly toxic PCDF congener and for some responses (e.g., immunotoxicity) this compound is approximately equipotent with 2.3.7.8- TCDD.295 2,3,4,7,8-PentaCDF also exhibits a rela tively high competitive binding affinity for the Ah receptor (1.5 x 10~M compared to 1.0 x !0~A/ for 2,3,7,8-TCDD). The results also show that the remaining 2,3,7,8-substituted tetra-, penta, and hexaCDF congeners are less toxic than 2.3.4.7.8- pentaCDF, however, their relative toxicities are both species- and response-dependent. These differences in relative toxicities are probably related to several factors including phar macokinetic differences which govern their uptake, distribu tion, and retention in laboratory animals. In additon, there was not a direct correlation between the relative toxicities of the 1990 1 9 4 C 9 63 m a a sim su sm ........... ..... imi i i i a i r m u r T i r iiiMMMl ------------------------------------- Critical Reviews In -log E D j 0 (Thymic Atrophy) FIGURE 10. A plot of the -lo g EC values for AHH induction in rat hepatoma H-4-II E cells by several PCDD congeners and their corresponding -lo g ED values for thymic atrophy in the immature male rat.2*8 -log ED 5 g (AHH Induction) FIGURE 12. A plot of the -lo g EC values for AHH induction in rat hepatoma H-4-II E cells by several PCDD congeners and their corresponding - log ED values for hepatic microsomal AHH induction in the immature male rat.288 -log ED 5 0 (Body Wt. Loss) FIGURE 11. A plot of the -lo g EC values for AHH induction in rat hepatoma H-4-II E cells by several PCDD congeners and their corresponding -lo g ED values for inhibition of body weight gain in the immature male ra t : 2 ,3 ,7 ,8 -s u b s titu te d co n g en ers in the rat and their c o m p e titiv e A h recep tor-b in d in g a v id ities u sin g rat h ep atic c y t o s o l.287 H ow ever, for a broad spectrum o f PC DFs w hich contained both 2,3,7,8-su b stitu ted congeners and com pounds with less than four lateral substituents there w as a (qualitative) corre lation betw een their structure-binding and structure-toxicity re la tio n sh ip s.287 In a 13-w eek su b ch ron ic to x ic ity stu d y , P lu ess and c o -w o rk er s305,306 estim ated that the relative to x ic itie s o f 2 ,3 ,7 ,8 - T C D D :2 ,3 ,4 ,7 ,8 - p en taC D F : 1,2 ,3 , 7 ,8 -p e n ta C D F : 1, 2 ,3 ,6,7,8-h exaC D F were 1.0:0.4:0.01:0.10. These results were w ithin the ranges observed in the shorter term stud ies (see Table 11). T ab le 12 sum m arized the in vitro and b io c h e m ic a l e ffe c ts o f the 2,3,7,8-su b stitu ted PC D Fs. T he structure-activity rela tionships for these responses sh ow that the fully lateral-sub stituted (C14-C 1 6) com p oun ds are m ore a c tiv e 287 than co n g en ers with less than four lateral chlorine groups. The results also show that 2,3,4,7,8-p en taC D F is the m ost active 2 ,3 ,7 ,8 -su b stituted PC D F. The dose-response effects o f several structurally diverse PC DD and PCDF congeners on the induction o f hepatic m icrosom al AHH and EROD activities, thym ic atrophy, and b od y w eig h t lo ss in the rat h ave b een rep orted .287-288 F igures 13, 14, and 15 illustrate a plot o f the - l o g E C 50 v a lu es for A H H in d u ction in vitro, rat h ep atom a H -4-II E c e lls ) v s. the in vivo --lo g E D 50 valu es for th y m ic atrop h y, b o d y w eigh t lo ss, and hepatic m icrosom al A H H induction in the rat by the sam e com pounds. The relatively high linear correlation c o e f ficients betw een these responses dem onstrated that the SA R s for both PC D D s (Figures 10 to 12) and PC D Fs are com parable for d iv er se resp on ses in d ifferent b io sy ste m s. M o r e o v e r , c o m bination o f the results for both the P C D D s and PC D Fs also g a v e g o o d lin ear correlations b e tw e e n the in vitro induction 1 o n 64 Volume 21, Issue 1 Toxicology Table 10 A C o m p arison of the N A T O /C C M S and Nordic T E F Values and the Relative Potency R a n ge s for the 2,3,7,8-Substltuted P C D D s Congener TEF values NATO/CCMS 2.3,7,8-TCDD 1.2,3,7,8-PentaCDD 1,2,3,4,7,8-HexaCDD 1,2,3,6,7,8-HexaCDD 1,2,3,4,7,8,9-HexaCDD 1,2,3,4,6,7,8-HexaCDD OCDD 1 0.5 0.1 0.1 0.1 0.01 0.001 * Does not include receptor binding data. Nordic 1 0.5 0.1 0.1 0.1 0.01 0.001 Relative potency ranges In vivo toxicides In vitro results* 0.59--0.053 0.24--0.013 0.16--0.0152 0.14--0.016 0.0076 >0.0013 ___ 0.64--0.07 0.13--0.05 0.5--0.005 0.009 0.003 0.0006 Table 11 Com parative T oxic P oten cie s of the 2,3,7,8-Substltuted P C D F s 187,1*8,288,2*7,295,297 LOWED* Valoes Response Species 2^3,7,8- 1,2,3,7,8* 2.3.4,7,8* IJ^.4,7,8- 2,3,4,*,7,*. U3,7,S,7> 1,2,3,4,6,7,S 1.2,3.4,7 J > 2,7,8- (dosc/coee.) TCDF PentaCDF PentaCDF HexaCDF HexaCDF HexaCDF HexaCDF HeptaCDF HeptaCDF OCDF TCDD Acute lethality Subdvunic tonicities Body weight loss Thymic Atrophy Immunotoxicify Teratogenicity Guinea Fig Mouse (ng/fcg) Rat (M-g/tg) Rat (unol/kg) Guinea pig (pjnoi/kg) Rat (pjnol/kg) Mice (|imol/kg) Mice (pjnol/fcg) 5-- 10 >6000 >1000 3.20 3.60 0.014 0.22 2.64 0.0039 1.76 0.35 3--10 916 1.04 0.012 0.21 0.003 0.09 1.30 0.30 0.89 120 2.80 0.93 0.011 0.012 0.6--2.0 14--284 22--43 0.03 0.003* 0.09 0.0024 0.011 Table 12 Com parative Biochem ical and In Vitro Potencies of 2,3,7,8-Substltuted PCDFs2801288'287'295-2*7'21' ED* Values Response System Receptor binding Mouse liver (Kd. aAf) Rat liver m. AHH induction Rat liver (fimol/Vg) H-4-II E cells (AO Chick embryo (omol/kg) Mouse liver (ixmol/kg) Guinea pig (junol/fcg) Kentinization X B 3T3 cells <nAO 2,3,7,* TCOF 0.73 4.1 x 10-* 0.632 3.9 x 10-* 0.46 1,24,7,8. 24,4,7,8- 144.4,7.8- 24.4,6,7,*- 1,24,6,7,. PentaCDF PentaCDF HexaCDF HexaCDF HexaCDF 7.45 x IO-* 1.5 x 10-* 2.3 x IO*7 4.7 x 10-* 2.7 x 10-' 1.47 0.037 0.293 0.265 2.5 x ur* 2.6 x 10-" 3.6 x IO-10 6.9 x IO*10 1.5 x ir * 0.0039 0.0012 1,2J,7,8> HexaCDF 144,4,6,7,- 1,24,4,7,8.9- 2,3,7,$. HeptaCDF HepuCDF OCDF TCDD 0.27 l.o x i<r* 0.11 0.70 0.004 7.2 x IQ"11 0.2 0.00028 0.05 1990 1 9 4 1 1 65 Critical Reviews In -log EDS g (Thymic Atrophy) FIGURE 13. A plot of the -lo g EC values for AHH induction in rat hepatoma H-4-II E cells by several PCDF congeners and their corresponding -log ED values for thymic atrophy in the immature male rat.2" -" 7 RGURE 15. A plot of the -lo g EC values for AHH induction in rat hepatoma H-4-II E cells by several PCDF congeners and their corresponding -lo g ED values for hepatic microsomal AHH induction in the immature male rat.2-" 7 FIGURE 14. A plot of the -lo g EC values for AHH induction in rat hepatoma H-4-II E cells by several PCDF congeners and their corresponding r-atlomg6-E:mD values for inhibition of body weight gain in the immature male and in vivo toxic responses. This is further illustrated in Figures 16 and 17 in which comparable linear correlations are observed for the in vitro - log EC values (AHH induction) vs. - log EDjo values for body weight loss and immunotoxicity in the -log ED j 0 Body Wt. Loss (mol/kg) RGURE 16. A plot of the - log EC values for the AHH induction in rat hepatoma H-4-IIE cells by PCDD and PCDF congenes and their corresponding - log ED values for inhibition of body weight gain in the male guinea pig.29* 66 Volume 21, Issue 1 13412 fflirHim rii Toxicology be more appropriate. The 0.001 TEF assign to OCDF was derived from the data obtained after multiple administration of OCDD to rats.304 FIGURE 17. A plot of the -lo g EC values for AHH induction in rat bepatoma H-4-H E cells by 2,3,7,8-TCDD and related compounds and their cotresponding -lo g ED values for immunotoxicity in C57BL/6 _ JM.332.364.J68 ffUCC. male guinea pig and C57BL/6 mouse, respectively.295,298These results strongly support a common receptor-mediated mecha nismof action of the PCDDs and PCDFs; Safe and co-workers have also used these data and other results to support the use of the in vitro induction (AHH and EROD activities) responses as quantitative bioassays for the hazard assessment of toxic balogenated aromatic hydrocarbons307'311 Table 13 summarizes the proposed TEF values for the 2,3,7,8substituted PCDFs. Both the NATO/CCMS and Nordic values are identical except for the values of 0.05 and 0.01 which were assigned to 1,2,3,7,8-pentaCDF by the two groups, respec tively. Based on the result summarized in Tables 11 and 12, the 0.01 and 0.05 values may be inappropriate. For example, 2,3,4,7,8-pentaCDF is less than four times more potent than the 1,2,3,7,8-isomer as a teratogen in C57BL/6 mice.297 Based on this result and other comparable potency ratios of the two pentaCDFs, a more conservative TEF value for 1,2,3,7,8pentaCDF would be 0.10. Recent studies387 have determined the relative potencies of the 4 heptaCDF isomers as immunotoxins and as inducers of hepatic microsomal AHH and EROD activities in C57BLV6 mice. The ED values for the suppression of the splenic plaqueforming cell response to sheep red blood cells were 0.011 and 0.012 p.mol/kg for 1,2,3,4,6,7,8- and 1,2,3,4,7,8,9-heptaCDF, respectively; the corresponding ED value for 2,3,7,8-TCDD was 0.0024 p.mol/kg. These results suggest that the proposed TEF value of O.Ol301'303 may underestimate the relative toxicities of the heptaCDFs and a conservative value of 0.1 may D. TEFs for Other Halogenated Dibenzo-p-dioxins and Dlbenzofurans Recent studies have demonstrated that brominated/ chlorinated dibenzofurans and dibenzo-p-dioxins are also formed during combustion processes.128,169,170,311 These compounds have not been widely identified in the environment, however, analytical studies have focused on the detection and quanti tation of PCDDs and PCDFs and not their chlorinated/brominated or polybrominated analogs. Most in vitro studies with brominated PCDDs or PCDFs or the fully chlor inated analogs suggest that these compounds are approximately equipotent. A comparative study of the dose-response effects of several bromo/chloro and polybrominated dibenzo-p-dioxins in male rats is summarized in Table 14.289 These data show that the potencies of these congeners varies with the position and number of bromo substituents and the response which is measuresd. A direct comparision of the 2,3,7,8- and 1,2,3,7,8substituted bromo and chloro analogs shows that there are minimal differences in their toxic potencies (Tables 8 and 14). Interestingly, 1,3,7,8-tetraCDD was more toxic in the rat than the corresponding brominated analog, whereas the reverse or der of potency was observed for the 1,2,4,7,8-substituted bromo/chloro congeners.288,289 A recent study reported that 2,3,7,8-tetrabromodibenzo-p-dioxin was at least ten times less potent than 2,3,7,8-TCDD in a 28-day rat LD and a rabbit ear dermal comedogenicity study.312 The reduced potency of this and other brominated dibenzo-p-dioxin and dibenzofuran congeners needs to be assessed in additional studies. At pre sent, this author suggests that the same TEF values be used for both the chlorinated, brominated, and chloro/bromo (mixed) dibenzo-p-dioxins and dibenzofurans. E. PCBs: Development of TEFs Several studies have confirmed a correlation between the structure-binding and structure-activity relationships for the two major classes of PCBs which exhibit Ah receptor agonist activity 184,186.191.192.198.259.285.313-327 (Figures 8 and 9). The coplanarPCBs, 3,4,4',5-tetraCB, 3,3',4,4'-tetraCB, 3,3',4,4',5pentaCB and 3,3',4,4',5,5'-hexaCB which are substituted in both para, at least 2 meta, and no ortho positions are clearly the most toxic members of this class of halogenated aromatics. The rela tive toxic and biochemical potencies of the coplanar PCB con geners (class I) are summarized in Table 15 and it is apparent from these data that there are considerable variations within this group of compounds (note: very little data is available for the 3,4,4' ,5-tetraCB congener and this compound has been omitted from the table). The results obtained for the potencies of the co- 1990 1 9 4 1 3 67 Critical Reviews In Table 13 A Comparison of the NATO/CCMS and Nordic TEF Values and Relative Potency Ranges for the 2,3,7,8-Substltuted PCDFs TEF values Congener NATO/CCMS 2,3,7,8-TCDF 2,3.4,7.8-PentaCDF 1,2,3,7,8-PentaCDF 1,2,3,4,7,8-HexaCDF 2.3,4,6,7,8-HexaCDF 1,2,3,6,7,8-HexaCDF 1.2,3,7,8,9-HexaCDF 1,2,3,4,6,7,8-HeptaCDF 1,2,3,4,7,8,9-HeptaCDF OCDF 0.1 0.5 0.05 0.1 0.1 0.1 0.1 0.01 0.01 0.001 * Does not include receptor binding data. Nordic 0.1 0.5 0.01 0.1 0.1 0.1 0.1 0.01 0.01 0.001 Relative potency ranges In vivo toxidtles In vitro results* 0.17--0.016 0.8--0.12 0.9--0.018 0.18--0.038 0.097--0.017 -- -- 0.22 0.20 -- 0.43--0.006 0.67--0.11 0.13--0.003 0.2--0.013 0.1--0.015 0.048-0.037 -- -- -- -- Table 14 The In Vivo Biologic and Toxic Effects of PBDOs and Bromlnated PC D D s in the Immature Male Wlstar Rat2*9 Compound 2,3,7,8-TBDD 2,3-Dibromo-7,8- dichlorodibenzo-p-dioxin 2-Bromo-3,7,8- trichlorodibenzo-p-dioxin 1,2,3,7,8-PentaBDD 1,2,4,7,8-PentaBDD 1,3,7,8-TBDD Body weight loss 6.75 x io* 1.24 x 10~* 1.24 x 10-' 8.71 x 10-7 1.29 x io-5 2.52 x 10* I n vivo EDm values (mol/kg) Thymic atrophy AHH Induction 3.39 x io-* 7.25 x 10* 7.59 x io-10 4.90 x io-` 3.47 x 10* 2.46 x 10* 3.9 x io-7 6.17 x 10* 3.55 x io* 2.46 x 10* 1.95 x io-7 6.50 x IO* EROD Induction 3.55 x io-10 3.47 x io-10 1.15 x 10* 1.74 x 10* 2.19 x io-7 1.26 x 10* highly variable and dependent on both the species and the response. The data show that 3,3',4,4',5-pentaCB is clearly the most toxic coplanar PCB congener and the 2,3,7,8TCDD/3,3',4,4',5-pentaCB potency ratios were 66/1 (body weight loss, rat); 8.1/1 (thymic atrophy, rat); 10/1 (mouse fetal thymic lymphoid development); 125/1 (AHH induction, rat); 3.3/1 (AHH induction, rat hepatoma H-4-IIE cells), and 100/1 (chick embryo hepatocytes). Based on these toxicity data, a TEF value of 0.1 is appropriate for this congener. It is clear from the data in Table 15 that both the 3,3',4,4'-tetra- and 3,3',4,4',5,5'-hexaCB congeners are considerably less toxic than3,3' ,4,4' ,5-pentaCB, however, their relative potencies are highly variable. It is apparent from in vivo studies in the rat that 3,3',4,4'-tetraCB is >30 times less toxic than 3,3',4,4'5,5'hexaCB, whereas, in most of the in vitro assays, these com pounds exhibit similar potencies or the reverse order of potency is observed. The relatively low in vivo toxicity of 3,3',4,4'tetraCB in the rat may be due to rapid metabolism which has also been observed in male ICR mice;329-330 however, it should be noted that this congener has recently been detected in en vironmental and human samples331 and these results suggest that the rapid clearance from the rat may not be observed in all animal species. Based on the in vivo teratogenicity and toxicity in rats, the proposed TEF value for 3,3',4,4',5,5'hexaCB is 0.05 and a conservative value of 0.01 is assigned to 3,3',4,4'-tetraCB. These assigned TEFs will change as more quantitative toxicity data becomes available from future stud ies. Figure 9 illustrates the structures of all the monoortho co planar PCB congeners which can be derived from the four coplanar congeners. These compounds all exhibit Ah receptor agonist activity; however, only limited quantitative SARs have 68 Volume 21, Issue 1 19414 Toxicology Table 15 Comparative Toxic and Biochemical Potencies of Coplanar P C B s317,323-32* 32* Response :hronic toxicities Body weight loss Thymic Atrophy Bursal lymphoid development Thymic lymphoid development Immunotoxicity Teratogenicity AHH induction AHH induction AHH induction Receptor binding Target species/cell Rat (pmol/kg) Rat (p.mol/kg) Chick embryo (jig/kg) Mouse fetuses (AO Mice (p-moL/kg) Mice (pmol/kg) Rat (pmol/kg) H -4 -n E cells (AO Chick embryo hepatocytes Rat cytosol (AO 3,3,4,4'TetraCB >500 >500 50 3 x 10-7 500 3.5 x KT4 2.2 x 10-* 4.3 x 10-7 EDjoCjo Values 3,3',4,4',5PentaCB 3,3',4,4',5,5'HexaCB 3.3 0.95 4 2 x IO"9 0.50 2.4 x 1C-10 2.0 x 1CT9 1.2 x IO"7 15 8.9 300 2 x IO*7 .055-- . 11032* 1.10 6.0 x 10- -- Insoluble 2,3,7,8TCDD 0.05 0.09 2 x IO"10 .0024 .011 0.004 7.2 x 10-" 2.0 x 10-" 1.0 x ICC4 Table 16 Comparative Toxic and Biochemical Potencies of Monoortho Coplanar P C B s317,323"32* Response Target species/cell EDa/EC* Value! 2,3',4,4',J- 2 ^ J ',4 ,4 '- 2',3,4,4',5- 2,3,4,4',5 2,3,3',4,4',5 2,3,3',4,4',J- 2,3',4,4',S,S'- 2,3,3',4,4',5,5'- PentaCB PentaCB PentaCB PentaCB HexaCB HexaCB HexaCB HeptaCB 2,3,7,8TCDD Subchronic studies Body weight loss Thymic atrophy Bursal lymphoid development Immunotoxocity development Teratogenicity AHH induction Receptor binding Rat (iimol/kg) Rat (|unol/kg) Chick embryos (u-g/kg) Mice (pmol/Vg) Mice (pmol/kg) Rat (pjnot/kg) Rat hepatoma (cells (AT) Chick embryo hepatocytes (AO Rat cytosol (AO 1450 1550 >6000 750 1030 <6000 165 65 1.2 x IO"5 8.8 x 104 5.0 x 10-* 9.1 x 10* 4.3 x 10* 370 2790 130 3.9 x 10* 1.4 x 10* 180 200 30 9.7 x 10-7 1.0 x 10-' 4.7 x 10* 180 180 <6000 2 328 7 2.1 x 10* 1.4 x 10-5 7.1 x 10* 220 225 >6000 6 1.1 x 10* 5.0 x 10* >6000 1.3 x IO"3 1.6 x 10* 0.05 0.09 0.0024 0.011 0.004 7.2 x IO-11 2.0 x 10-" 1.0 x IO* these studies are summarized in Table 16. The relative poten cies of this subgroup of PCBs are highly variable. 2,3',4,4',5PentaCB is the major monoortho coplanar PCB present in Aroclors 1248 and 1254 and is routinely identified in environ mental and human extracts.13'21,44,45,331 The activity of this compound relative to 2,3,7,8-TCDD as an inducer of AHH activity is 0.004326 (chick embryo hepatocytes), 0.000008317 (rat hepatoma cells), and 0.000034323 (rat liver). In vivo studies in the rat show that the corresponding TEFs for 2,3',4,4',5-pentaCB are 0.000035 and 0.00006 for body weight loss and thymic atrophy, respectively.323 Some of the higher cholorinated hexaCB isomers are more toxic than 2,3',4,4',5-pentaCB and the differences in potency may be related to different rates of in vivo metabolism. For example, the 2,3,3',4,4',5-hexaCB is one of the most active monoortho coplanar PCBs in the rat and in the C57BL/6 mouse. In the latter species, the 2,3,7,8-TCDD/2,3,3',4,4',5-hexaCB po tency ratios for immunotoxicity and teratogenicity were 0.0012 and 0.00003, respectively.297,332 This represents a 40-fold dif ference in the relative potency of 2,3,3',4,4',5-hexaCB for 2 different responses in the same animal species. It is apparent from the data in Table 16 that this variability is also observed for other congeners. Based on the results obtained for one of the more active congeners, namely, 2,3,3',4,4',5-hexaCB, a 1990 M 1 5 69 Critical Reviews In TEF value of 0.001 for the monorotho coplanar PCBs would be highly conservative. Figure 18 illustrates the structures of several PCB congeners which represent six different structural groups: the coplanar PCBs (class I), monoortho coplanar PCBs (class II), monoortho co planar PCBs minus asingle p-chloro substituent (class III), dior tho coplanar PCBs (class IV), triorthocoplanar PCBs (class V), and tetraortho-substituted PCBs (class VI). The structure-immunotoxicity relationships for these compounds have been de termined inC57BL/6 mice by measuring theirdose-response ef fects on the splenic plaque-forming cell response to sheep red blood cells.332Previous studies have confirmed the immunotoxicity of class I and II structural groups258-333and this study also showed that class III and IV congeners were also active. Forex ample, the EDS0values for the immunosuppressive activities of 2,3,3',4,5,5'- and 2,3',4,4',5',6-hexaCB were 200 and 120 p.mol/kg, respectively. Previous studies have reported that, with the exception of 2,2',4,4',5,5'-hexaCB,184191 several diortho coplanar PCBs exhibit weak Ah receptor agonist activty i84.i9i.332.334 g asec on (he immunotoxicity results, the po tencies (relative to 2,3,7,8-TCDD) of the class III and IV PCBs would be 0.000012 and 0.00002, respectively. Since diortho coplanar PCBs are major components of the commercial prod ucts and extracts fromenvironmental samples, more researchon their relative toxicities should be carried out. However, at pre sent a provisional TEF value of 0.00002 is assigned. Based on the results presented in Tables 15 and 16 a proposed set of TEF values for the coplanar monoortho coplanar and dioortho co planar PCB congenes is summarized in Table 17. F. PBBs: Development of TEFs Several studies have demonstrated that SARs for PBBs are comparable to those reported for the corresponding PCBs.184,186191,192,334"345 The coplanar and monoortho co planar congeners all elicit a typical spectrum of Ah receptormediated responses which include a wasting syndrome, he- 3 .3 \4,4\5-pntaC 8 I 2 .3 .3 '.4 .4 '.S .h titC B II 2.3\4.4\5\6*hiaC B Mil 2 . 2 \ 4 , 4 ,t3 ,5 ,- H i C 8 lilb 2 ,3 \4 ,5 ,5 \S*h # ia C a IV 2.2*.4.4`,r .S -ri i C B V 2.2'.4.4*.6,6l* h * s a C B VI FIGURE 18. Classification of PCB congeners into six different structural groupings." 2 Table 17 Proposed TEF Values for the Coplanar and Monoortho Coplanar PCB Congeners Congener 1. Coplaner PCBs 3,3',4,4',5,5'-PenlaCB 3,3',4.4'.5,5'-HexaCB 3,3',4,4'-TetraCB 2. Monortho coplaner PCBS 2,3',4,4',5-PentaCB 2,3,3',4,4'-PentaCB 2',3,4,4',5-PentaCB 2,3.4,4',5-PentaCB 2,3,3',4,4',5-HexaCB 2,3,3',4,4',5'-HexaCB 2,3',4,4',5,5'-HexaCB 2,3,3 ' ,4,4' ,5,5' -HeptaCB 3. Dioortho coplanar PCBs TEF values 0.1 0.05 0.01 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.00002 ' Does not include receptor binding data. Relative potency range* 0.3--0.0006 0.1--0.0012 0.009--0.00008 0.0004--0.000006 0.0008--0.00006 0.00013--0.000018 0.00045--0.000074 0.0005--0.0000014 0.0004--0.0000065 0.0000055 No data available 0.00002 patoxicity, immunotoxicity and thymic atrophy, carcinogene sis, dermal toxicity, hepatotoxicity, and porphyria. In contrast to the PCBs, quantitative SARs are not available for individual PBB congeners. Therefore, it is recommended that the TEF values for the PCBs should also be utilized for the correspond ing PBBs (i.e., Table 17). G. Polychlorinated Diphenyl Ethers (PCDEs): Development of TEFs Only a limited amount of toxicologic data is available for the PCDEs and the results suggest that the SARs for the PCDEs and PCBs are comparable.346"348 Recent studies in my labo ratory have investigated the dose-response effects of several PCDE congeners on the suppression of the splenic plaque forming cell response to sheep red blood cells and the induction of hepatic microsomal AHH and EROD activities in C57BL/6 mice.349The immunotoxic potencies for these PCDE congeners (ED50 values for the suppression of PFCs/spleen and PFCs/106 cells) followed the respective order 2,3,3',4,4',5-hexaCDE (0.5 and 2.2 p.mol/kg) > 3,3',4,4',5-pentaCDE (8.8 and 5.1 p.mol/kg) > 2,3',4,4',5-pentaCDE (21.8 and 14.2 p.mo/kg) > 3,3',4,4'-tetraCDE (50.6 and 28.7 p.mol/kg) > 2,2',4,4',5,5'hexaCDE (81.2 and 56.5 p.mol/kg) > 2,2',4,5,5'-pentaCDE (258 and 228 (Jtmol/kg) > 2,2',4,4',5,6'-hexaCDE (>400 p.mol/kg for both responses). The potencies of the PCDE con geners as inducers of hepatic microsomal AHH and EROD activities were similar to their immunotoxicities. The structureactivity relationships for the PCDEs and PCBs differed sig nificantly. For example, the coplanar 3,3',4,4'-tetraCDE and 3,3',4,4',5-pentaCDE congeners were less immunotoxic than their monoortho 2,3',4,4',5-pentaCDE and 2,3,3',4,4',5hexaCDE analogs, respectively, and similar results were also observed for their enzyme induction potencies. For the cor- 70 Volume 21, Issue 1 19416 Toxicology responding PCB congeners, the coplanar compounds were sig nificantly more active than their monoortho analogs. In addi tion. two dioortho substituted compounds, 2,2',4,5,5'-pentaCDE and 2,2',4,4',5,5'-hexaCDE, were also immunotoxic at a dose of 400 p-mol/kg, whereas their PCB analogs were not active. These studies clearly demonstrate that, for the PCDE conge ners. increasing ortho-chloro substitution is less effective in reducing the activity of these compounds compared to the wellrecognized ortho effects reported for the PCBs. The ED50 val ues for the immunotoxicity and AHH induction activity of the coplanar and monoortho coplanar PCDEs in the C57BL/6 mice varied from 2.2 to 28.7 pmol/kg and 26.2 to 160 p.mol/kg, respectively. Since the corresponding ED50 value for the im munotoxicity of 2,3,7,8-TCDD was 0.0024 p.mol/kg, a TEF value of 0 .0 0 1 would be a conservative number for both the coplanar and monoortho coplanar PCDEs. H. Summary The development of TEFs for the toxic halogenated aro matics has been necessitated by the identification of complex mixtures of these compounds in almost every compartment of the global ecosystem. Although extensive toxicologic and car cinogenic data are available for one member of this family, namely, 2,3,7,8-TCDD, very little information is available for the remaining congeners. There is good evidence that the toxic halogenated aromatic hydrocarbons elicit common toxic re sponses through a common mechanism of action as illustrateed in Figure 5. Moreover, the SARs observed for the different classes of halogenated aromatics provide support for both the proposed receptor-mediated mechanism of action and the de velopment of TEFs for this family of chemicals. Based on the current data available, Table 18 summarizes the proposed TEFs for the halogenated aromatics. The TEF values for the PCDDs and PCDFs are similar to those described in the Nodic and NATO/CCMS documents and in a recent update by the Risk Assessment Forum of the U.S. Environmental Protection Agency;301' 303 however, values of 0.1 have been assigned to I,2,3,4,6,7,8- and 1,2,3,4,7,8,9-heptaCDF and 1,2,3,7,8 pentaCDF. The TEF values given to the PCBs, PCDEs, and brominated or bromo/chloro aromatic compounds have not pre viously been assigned by regulatory agencies. All the values reported in Table 18 are provisional and should be modified as more data become available. V. FACTORS WHICH EFFECT THE VALIDITY OF THE PROPOSED TEFs FOR 2,3,7,8-TCDD AND RELATED TOXIC HALOGENATED AROMATICS A. Interactive Effects -- Additivity The simple interactive effects of relatively potent Ah recep tor agonists and complex mixtures of these compounds have been investigated utilizing diverse Ah receptor-mediated re sponses. Pregnant mice were treated with different doses of 2,3,4,7.8-pentaCDF (0 to 30 (xg/kg) or 1,2,3,4,7,8-hexaCDF (0 to 300 p.g/kg) or a combination of the two toxins on days 10 to 13 of gestation.297 Based on probit analysis of the data, the combination of these two compounds as teratogens was additive. Similarly, the combination of 2,3,4,7,8-pentaCDF and 2,3,3',4,4',5-hexaCB (0 to 60 mg/kg) was also additive and this result contrasted to previous interactive studies with this PCB congener and 2,3,7,8-TCDD which gave apparent synergistic interactions for teratogenicity.350 However, based on the more comprehensive studies noted above, 297 it is likely that interaction of 2,3,7,8-TCDD and 2,3,3',4,4' ,5-hexaCB is also additive. Previous studies for several individual PCDD, PCDF, PBDD, and PCB congeners and a mixture of PCDFs have shown that there was linear correlation between their in vitro activity as inducers of AHH activity in rat hepatoma H-4-II E cells and their toxicities in at least three animal species287' 289-323 (see Figures 10 to 17). This research has confirmed the utility of the in vitro bioassay for quantitatively estimating the in vivo toxicities on individual congeners, and several reports also confirm the utility of the induction bioassay system for complex mixtures of halogenated aromatics. 307' 311-331 In addition, it has been shown that there was also a correlation between the in vitro induction activity of a mixture of six PCB congeners (2,3',4,4',5-pentaCB, 2,2',4,4',5,5'-hexaCB, 2,2',3,4,4',5'hexaCB, 2,3,3',4,4',5-hexaCB, 2,2',3,4,4',5,5'-heptaCB, and 2,2',3,3',4,4',5-heptaCB) and five PCDF congeners (2,3,7,8TCDF, 1,2,4,7,8-pentaCDF, 1,2,3,7,8-pentaCDF, 2,3,4,7,8pentaCDF, and 1,2,3,4,7,8-hexaCDF) and their corresponding in vivo toxicities in the rat.332-333 The PCB and PCDF mixtures were reconstituted from synthetic compounds and reconstituted to resemble the relative composition of PCBs and PCDFs iden tified in the liver of a victim of Yusho poisoning. Table 19 summarizes the observed and calculated EC50 values for in vitro induction of AHH and EROD activity by these two mix tures. The calculated values were derived from the known in vitro AHH induction activities of the individual congeners and their relative concentrations in the mixtures. The results showed that there was a good correspondence between the calculated and observed data and these studies suggest that the interactions of these compounds as inducers of AHH and EROD activities were additive. In additon, previous studies have demonstrated the linear correlations between the in vitro --log EC50 values for AHH induction vs. the - log ED50 values for in vivo toxicity for this mixture and other individual halogenated aromatic congeners287' 289 and this result indicates that the toxicities of the individual PCDF congeners in this mixture are also addi tive. Scientists from the New York State Department of Health have utilized a number of in vivo and in vitro bioassays for assessing the hazard of toxic halogenated aromatic mixtures associated with contaminants from PCB fires (e.g., PCDFs and PCBs) and extracts of samples from Love Canal.334' 361 For 1990 71 -thmw B iIITi.'-iTfW TIHiiMlW figitt Critical Reviews In Table 18 Proposed TEFs for the Toxic Halogenated Aromatics Congener PCDDs 2,3,7,8-TCDD 1,2,3.7,8-PentaCDD 1,2.3.4,7,8-HexaCDD 1,2,3.6,7,8-HexaCDD 1,2,3,7,8,9-HexaCDD 1,2,3,4,6,7,8-HeptaCDD OCDD PCDFs 2.3,7,8-TCDF 2,3,4,7,8-PentaCDF 1,2.3,7.8-PentaCDF 1,2,3,4,7,8-HexaCDF 2.3,4,6,7,8-HexaCDF 1,2,3,6,7,8-HexaCDF 1,2.3.7,8.9-HexaCDF 1,2,3,4,6,7,8-HeptaCDF 1,2.3,4,7,8,9-HeptaCDF OCDF Brominated and bromo/chloro Dibenzo-p-dioxin and dibenzofurens Same as described above in A and B TEF 1.0 0.5 0.1 0.1 0.1 0.01 0.001 0.1 0.5 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.001 Congener D. PCBs (a) Coplanar 3,3',4,4',5-PentaCB 3,3',4,4'-5,5'-HexaCB 3,3',4,4'-TelraCB (b) Monoortho coplanar 2,3',4,4',5-PentaCB 2,3,3',4.4'-PemaCB 2',3,4,4',5-PentaCB 2,3,4,4',5-PentaCB 2,3,3' ,4,4' ,5-HexaCB 2,3,3',4,4',5'-HexaCB 2,3',4,4',5,5'-HexaCB 2,3.3',4,4',5,5'-HeptaCB (c) Dioortho coplanar PCBs E. PBBs Same values as described above for the PCBs F. PCDFs Coplanar and monoortho coplanar congeners TEF 0.1 0.05 0.01 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.00002 0.001 example, the teratogenic and immunotoxic effects of an organic phase of a leachate (OPL) from Love Canal was investigated in genetically inbred mice.354-333 The OPL was a complex mixture of organic compounds which included chlorinated hydorcarbons, toluenes, benzenes, benzaldehyde, dibenzofurans, isomeric hexachlorocyclohexanes, and 2,3,7,8-TCCD (0.003% of the administered sample). The results of the immunotoxicity studies showed that in the Ah-responsive BALB/cByJ mice the immunosuppressive effects were primarily due to 2,3,7,8-TCDD and the interactive effects with the other organic constituents in the OPL were minimal. In contrast, the other chemical components in the OPL decreased the 2,3,7,8-TCDD-mediated Table 19 Summary of the Observed and Calculated ECso Values for the Induction of AHH and ERO D Activities in Rat Hepatoma H-4-II E Cells by PCB and PCDF Mixtures353 Treatment Reconstituted liver PCBs (observed) Reconstituted liver PCBs (calculated) Reconstituted liver PDCFs (observed) Reconstituted liver PCDFs (calculated) _________ EC (M) EROD AHH 6.86 X IO"5 7.01 X io-6 1.02 X ic -10 3.07 X io-10 m i X io-3 5.98 X io-6 3.23 X io-' 4.43 X 10-' immunotoxicity in the Ah-nonresponsive DBA/2J mice, and the thymic atrophy and hepatomegaly observed in these mice was associated not with 2,3,7,8-TCDD but the other organic components. In the teratogenicity studies, the development of fetal cleft palate and hydronephrosis in C57BL/6J mice treated with OPL was primarily due to the 2,3,7,8-TCDD in this mix ture and minimal interactions with the other chemicals were observed. However, the maternal toxicity observed in this study was primarily due to the other compounds. It was also con cluded that the effects of Love Canal soil extracts on the ma ternal health and fetal development in rats was also due to 2,3,7,8-TCDD which was present as a trace component of this mixture.336 Eadon and co-workers summarized the results of several studies on the biochemical and toxic effects of a complex mixture of PCDDs (minor) and PCDFs (major) from soot sam ples which originated from the PCB transformer fire in the Binghamton State Office building.337 Their in vivo data also showed a good correlation between bioassay-derived "2,3,7,8TCDD equivalents" in the extracts and the toxic equivalents calculated from the chemical analysis of these samples (see Section VI). Gierthy et al. have also investigated the in vitro activities of soot extracts from the Binghamton PCB fire using a keratinization assay with XB cells.360 The toxic equivalents data obtained fromthe bioassays were comparable'to the "2,3,7,8TCDD equivalents" which were calculated from the analytical (chemial) data using TEFs for the 2,3,7,8-substituted conge IS 41872 Volume 21, Issue 1 -- Toxicology ners. It was evident that the results not only confirm the utility of the TEF approach, but also indicate that the interactions of the PCDD/PCDF components in the soot mixture are primarily additive. B. Interactive Effects -- Nonaddltlvlty Safe and co-workers have reported the interactions of 2,3,7,8- TCDD and halogenated aromatic compounds which result in the partial antagonism of several biochemial and toxic re sponses elicited by 2,3,7,8-TCDD.293-361'368 The compounds which exhibit 2,3,7,8-TCDD antagonist activity generally ex hibit moderate competitive binding affinities for the Ah recep tor but are very weak Ah receptor agonists. For example, the ECjo value fr the competitive binding of 1,3,6,8-TCDF to the murine cytosolic Ah receptor was 1.25 x 10~7 M and the results of a double-reciprocal plot analysis of the saturation binding of [3H]-2,3,7,8-TCDD in the presence of different concentrations of 1,3,6,8-TCDF (1 to 100 nAf) suggested that 1.3 .6 .8- TCDF was a competitive anatagonist.293 In contrast, 1,3,6,8-TCDF was approximately 1.5 x 104 times less active than 2,3,7,8-TCDD as an immunosuppressive agent in C57BL/6 mice.193 Interactive studies showed that 1.3.6.8- TCDF partially antagonized 2,3,7,8-TCDD-mediated AHH and EROD induction activities in rat hepatoma H-4-II E cells362 and immunotoxicity in male C57BL/6 mice.293 The results in Table 20 illustrate the interactions of 1,3,6,8-TCDF with both 2,3,7,8-TCDD and 2,3,4,7,8-pentaCDF and the par tial antagonism of the decrease in the splenic plaque-forming cell response to sheep red blood cells caused by the two potent Ah receptor agonists. It should be noted that these effects are dependent on both the concentration of the agonist and antag onist, and in these studies the "window of antagonism" was observed at 1,3,6,8-TCDF/2,3,7,8-TCDD ratios between 1000/1 to 13,000/1. It is unlikely that the interactions of putative PCDD/PCDF antagonists and agonists will play a significant role in hazard or risk asessment of PCDDs and PCDFs since the relative levels of the potential antagonists are relatively low in extracts which contain PCDD and PCDF mixtures. The commercial PCB mixture Aroclor 1254 is also a weak Ah receptor agonist compared to 2,3,7,8-TCDD. For example, the ECjo/EDjovalues for the immunotoxicity of 2,3,7,8-TCDD in C57BL/6 mice and the AHH induction activity in rat hep atoma H-4-II E cells are 2.4 nmol/kg and 7.2 x 1CT11 M , respectively;364 the corresponding values for Aroclor 1254 are 350 nmol/kg and >10~3M , respectively. In contrast, the com petitive binding EC values for 2,3,7,8-TCDD and Aroclor 1254 differ by approximately 600-fold (1 x 10"* M and 6.0 x 10"6 M , respectively).283 The interaction of subeffective doses of Aroclor 1254 and an EDgo-ioo dose of 2,3,7,8-TCDD has demonstrated that Aroclor 1254 can partially antagonize several 2,3,7,8-TCDD-induced responses, including AHH and EROD induction in rat hepatoma H-4-II E cells and C57BL/6 mice, teratogenicity, and immunotoxicity in C57BL/6 mice.362-364,367 Table 21 summarizes the interactive effects on the splenic plaque-forming cell response to sheep red blood cells in C57BL/6 mice. Aroclor 1254 at doses from 5 to 150 p.mol/kg inhibited the immunosuppressive activity of 2,3,7,8TCDD (3.7 nmol/kg) and this inhibitory effect was partially reversed by a higher dose (11.2 nmol/kg) of 2,3,7,8-TCDD. The results of these and other interactive studies suggest that Table 20 Effects of 2,3,7,8-TCDD and 2,3,4,7,8-PentaCDF on the Splenic Plaque-Forming Cell (PFC) Response to Sheep Red Blood Cells in C57BL/6J Mice: interaction with 1,3,6,8-TCDF-295 Treatment Com oil 2,3,4,7,8-PentaCDF 2,3,4,7,8-PentaCDF + 1,3,6,8-TCDF 2,3,4.7,8-PentaCDF + 1,3,6,8-TCDF 2,3,7,8-TCDD 2,3,7,8-TCDD + 1,3,6,8-TCDF 2,3,7,8-TCDD + 1,3,6,8-TCDF 2,3,7,8-TCDD + 1,3,6,8-TCDF Dose _ 3.7 nmol/kg 3.7 nmol/kg 3.3 p.mol/kg 3.7 pmol/kg 6.6 timol/kg 3.7 nmol/kg 3.7 nmol/kg 6.6 p.mol/kg 3.7 nmol/kg 16.5 ixmol/kg 3.7 nmol/kg 49.3 (imol/kg Spleen cellularity x 107 _ 14.0 2.2 16.2 i 1.0 19.8 5.4 24.0 2.7 19.7 1.5 20.4 4.9 20.7 4.6 PFCs/spleen x I01 1.50 0.19 0.53 0.05 0.60 0.07 0.93 0.19b 0.29 0.08 0.55 * 0.14c 0.57 * 0.08c 0.59 0.06c * The control group contained 8 animals and all other groups contained 5 animals; data presented as means SO. b Significantly different (p <0.05 from 2,3,4,7,8-pentaCDF-treated mice. ` Significantly different (p <0.05 from 2,3,7,8-TCDD-treated mice. PFCs/104 viable cells 704 = 120 253 = 48 298 87 384 16b 125 * 46 284 * 81c 261 36c 216 = 29* 1990 1 9 4 1 9 73 Critical Reviews In Table 21 Effects of 2,3,7,8-TCDD, Aroclor 1254, and 2,3,7,8-TCDD Plus Aroclor 1254 on the Plaque-Forming Cell Assay In Mice Pretreated with Sheep Red Blood Cells364 PFCs/104 % Control Treatment (dose) PFCs/spleen x 10s viable spleen cells (PFCs/spleen x 10s) Com oil* 2,3,7,8-TCDD (3.72 nmol/kg)b (11.2 nmol/kg) Aroclor 1254 (5 funol/kg) (15 jxmol/kg)b (75 (imol/kg) (150 M-mol/kg) 2,3,7,8-TCDD (3.72 mnol/kg) plus Anoclor 1254 (5 p.mol/kg) 2,3,7,8-TCDD (3.72 nmol/kg) plus Aroclor 1254 (15 (imol/kg)*2,3,7,8-TCDD (3.72 nmol/kg) plus Aroclor 1254 (75 p.mol/kg) 2,3,7,8-TCDD (3.72 nmol/kg) plus Aroclor 1254 (150 p.mol/kg) 2,3,7,8-TCDD (11.2 nmol/kg) plus Aroclor 1254 (75 (imol/kg) 1.61 0.17 0.38 0.03c 0.20 0.03' 1.57 0.11 1.60 0.87 1.57 0.23 1.20 0.22 1.70 0.30 1.49 0.08 1.66 0.28 0.92 0.08' 0.58 0.10e 562 * 76 156 40e 62 -b 12' 561 140 465 114 594 35 394 35 491 62 480 48 558 52 # 372 81' 170 10e 100 24 12 98 100 98 75 106 93 103 57 36 * Eleven animals b Eight animals; all others, four animal/group. c Significantly different (p <0.01) from the com oil controls. Aroclor 1254 acts as a competitive antagonist. In addition, several other commercial Aroclors exhibit similar antagonist activities.368 As noted previously for 1,3,6,8-TCDF, the an tagonist activity of Aroclor 1254 is dependent on the antagonist/agonist ratios and their absolute concentrations. Table 22 summarizes the results obtained for the interactions of Aroclor 1254 and 2,3,7,8-TCDD in C57BL/6 mice; it is evident that Table 22 Aroclor 1254 as a 2,3,7,8-TCDD Antagonist In C57BL76 Mice -- Summary363-364-366 Response AHH induction EROD induction Thymic atrophy Immunotoxicity Teratogenicity % Maximum antagonism 20 23 0 100 80 Antagonist/agonist window 1,667-- 10,000/1 1,667-- 10,000/1 No antagonism observed 1,340--20,160/1 12,100/1 the % maximum antagonism is response-dependent and ratios of Aroclor 1254/2,3,7,8-TCDD, <20,000/1 and >1670/1, are required to observed partial antagonism. Analytical studies of extracts from human tissues have shown that the ratio of PCBs/PCDDs plus PCDFs is in the range within which PCBs partially antagonize the 2,3,7,8-TCDD-mediated responses in laboratory animal studies; however, the significance of PCB/PCDD plus PCDF interactions in humans is unknown. Nevertheless, it is possible that the TEFs assigned to individual PCDD and PCDF congeners are conservative values when these compounds are present in extracts which contain 103 to 104 times higher levels of PCBs. C. Responses Which Are Not Mediated through the Ah Receptor The TEF values for halogenated aromatics which are sum marized in Table 18 can be utilized for the hazard and risk assessment of those congeners which elicit Ah receptor-me diated responses. This sub-group of compounds (e.g., Figures 6 to 9) constitutes only a small fraction of the total number of 18470 74 Volume 2 , Issue 1 Toxicology halogenated aromatic congeners (e.g., see Table 1). This lim itation may not be important for PCDDs and PCDFs since most environmental samples primarily contain the 2,3,7,8-substituted compounds. In contrast, the "2,3,7,8-TCDD-like" PCBs are only a minor component of the total PCBs which are rou tinely identified in environmental extracts.4*-*3 Therefore, the proposed TEFs for PCBs do not account for the potential tox icity of the "non-2,3,7,8-TCDQ"-like congeners or their in teractive effects. Although it is unlikely that any of these PCBs elicit acutely toxic effects, there are several reports which show that some members of this class of PCBs elicit biochemical and toxic responses.184 For example, several PCB congeners resemble both phenobarbital (PB) and dexamethasone as in ducers of hepatic microsomal cytochrome P-450 isozymes191,197 (i.e., cytochromes b/e and cytochrome p, respectively). Both 2,2',4,4',5,5'-hexachloro- and hexabromobiphenyl have been characterized as PB-type inducers of hepatic drug-metabolizing enzymes.191-192-319-322-338-341 Moreover, like PB, 2,2',4,4',5,5'hexabromobiphenyl promoted diethylnitrosamine-initiated en zyme altered foci in Sprague-Dawley rats using a two-stage hepatocarcinogenesis protocol.371 In addition, a combination of both 2,2',4,4',5,5'- and 3,3',4,4',5,5'-hexabromobiphenyl gave an apparent synergistic interaction in the hepatocarcino genesis bioassay.372These data suggest that the corresponding PCB congeners may also exhibit comparable tumor-promoting activities, and it is clear that future studies should focus on the development of hazard/risk assessment approaches for those PCB/PBB congeners which are not covered in the current TEF approach. VI. APPLICATIONS OF TEFs FOR HAZARD ASSESSMENT OF TOXIC HALOGENATED AROMATICS TEFs are specifically designed to assess the potential hazard and risk of complex mixtures of halogenated aromatics. Uti lization of one or more of the TEF schemes301'303-337-373-374 requires high-resolution chemical analysis of specific analytes so that each individual congener can be unambigu ously identified and quantitated. Tanabe, Tatsukawa, and coworkers44-373'382 and oilier groups383'383 have utilized TEF ap proaches to determine the relative toxicities of halogenated aromatic mixtures and the relative contributions of the different classes of compounds. The concentration of a specific congener times the TEF gives the concentration in terms of "2,3,7,8TCDD equivalents". The results in Table 23382 summarize the concentrations of the individual congeners present in the PCDD, PCDF, and PCB fractions from Kanechlor 400, a commercial PCB mixture, and Yusho oil which was originally derived from Kanechlor 400. The concentration of "2,3,7,8-TCDD equiv alents" (in part-per-million) was determined for the major toxic PCDD, PCDF, and PCB congeners using the TEF values de- Table 23 2,3,7,8-TCDD Equivalents (ppm) In Kanechlor 400 and Yusho Oil382 Congener 2,3,7,8-TCDD 1,2,3,7,8-PentaCDD 1,2,3,4,7,8- HexaCDD I,2,3,6,7,8- HexaCDD 1,2,3,7,8,9- HexaCDD 1,2,3,4,6,7,8- HeptaCDD OCDD Kanechlor 400 2,3,7,8- TCDD Cone. eq. TEF (ppm) (ppm) 1.0 _ 0.5 -- 0.1 -- _ -- -- 0.1 0.1 0.01 0.001 -- -- Yusho Oil #2 2,3,7,8- Cone. (ppm) TCDD eq. (ppm) ND 0 0.00075 0.00038 0.00057 0.000057 0.038 0.0038 0.022 0.0022 0.190 0.00190 0.110 0.000110 Total 00 0.00747 2.3,7,8-TCDF 1,2,3,7,8-PentaCDF 2,3,4,7,8-PentaCDF 1,2,3,4,7,8- HexaCDF 2,3,4,6,7,8- HexaCDF I,2,3,6,7,8- HexaCDF 1,2.3,7,8,9- HexaCDF 1,2,3,4,6,7,8- HeptaCDF 1,2,3,4,7,8,9- HeptaCDF OCDF 0.1 0.1 0.5 0.1 0.1 0.1 0.1 0.1 0.1 0.001 1.7 0.23 0.68 0.37 0.033 0.16 0.006 0.130 0.038 0.2 0.17 0.023 0.34 0.037 0.69 0.76 1.4 1.2 0.0033 0.16 0.016 0.23 0.0006 ND 0.0130 0.20 0.0038 0.011 0.0002 0.076 0.069 0.076 0.7 0.12 0.016 0.023 0.020 0.0011 0.000076 Total 0.6069 1.025176 3,3',4,4',5-PentaCB 3,3',4,4',5,5'- HexaCB 3,3',4,4'-TetraCB 2,3,3',4,4'-PentaCB 2,3',4,4',5-PentaCB 2,3,3'.4,4',5- HexaCB 0.1 89 0.05 0.57 0.01 0.001 0.001 0.001 8,500 17,000 28,000 1,200 8.9 0.73 0.0285 0.03 85.0 12 17.0 32 28.0 37 1.2 3.2 0.073 0.0031 0.12 0.032 0.037 0.0032 Total 140.1285 0.2685 N o te: ND = not detected. Critical Reviews In rived in Table 18. In Kanechlor 400, the concentrations of "2,3,7,8-TCDD equivalents" for the PCB and PCDF fractions were 140.13 and 0.6069 ppm, respectively, and it was apparent that greater than 99% of the "dioxin-like" activity of the com mercial product was associated with the PCB fraction. In con trast, the concentrations of "2,3,7,8-TCDD equivalents" for the PCB and PCDF fractions in the Yusho oil were 0.2683 and 1.025 ppm, respectively. These results demonstrate that 79% of the "dioxin-like" activity in the Yusho oil was associated with the PCDF fraction and these calculations, correspond to previous in vitro and in vivo toxicology studies with Yusho oil fractions in which the PCDFs were shown to be more potent than the PCBs.352,386 Tanabe and co-worker derived their TEFs for individual PCDF and PCB congeners from their relative potencies as inducers of AHH activity in rat hepatoma H-4-II E cells in culture.286"288,317 Their results for the analysis of Yusho oil showed that the PCDF fraction contributed 77% of the "dioxin-like" activity of Yusho oil. Thus, the concentra tions of "2,3,7,8-TCDD equivalents" in Yusho oil were com parable using the TEFs in Table 18 or the TEF values derived from the AHH induction activities of the individual congeners. Therefore, these data not only confirm the utility of the TEF approach for the hazard assessment of complex mixtures of toxic halogenated aromatics but also show that the in vitro AHH induction assay provides a reliable bioanalytical method for quantitating the "dioxin-like" activity of these mixtures. Eadon and co-workers375 have investigated the toxicity of a complex mixture of PCDFs (Table 24) which were derived from soot samples formed during the PCB fire in the Bing hamton State Office Building. The total concentrationof "2,3,7,8TCDD equivalents" in the soot was 15.371 ppm using the TEF scheme summarized in Table 18; using slightly modified values for the TEFs, Eadon and co-workers357 estimated that the con centration of 2,3,7,8-TCDD equivalents was 19 ppm. The comparative toxicities of this mixture and 2,3,7,8-TCDD were determined in a series of in vivo studies in guinea pigs; the respective calculated "2,3,7,8-TCDD equivalents" concentra tions in the soot were 19, 21, 5, 18, 10, and 2 ppm for relative thymus weight decrease, % initial body weight decrease, the increase in serum triglycerides, the decrease in serum ALT, the formation of hepatocellular cytoplasmic inclusion bodies, and mortality (LD50). With the exception of acute toxicity data, it was apparent that there was an excellent overlap between the calculated and observed "2,3,7,8-TCDD equivalents" for these studies. Therefore, these results also validate the use of the TEF approach for 2,3,7,8-TCDD and related toxic halo genated aromatics. Several groups have utilized the TEF approach to determine the relative contribution of PCB, PCDD, and PCDF residues to the overall "dioxin-like" activity of halogenated aromatic con taminants in environmental and human samples.375'385 The re sults shown in Table 25 summarize the average concentrations (parts-per-thousand) of the toxic 2,3,7,8-substituted PCDDs and Table 24 2,3,7,8-TCDD Equivalents In Soot Samples Derived From the PCB Fire In Binghamton State Office Building" 7 Congener 2,3,7,8-TCDD 1,2,3,7,8-PentaCDD 1,2,3,4,7,8-HexaCDD 1.2,3,6,7,8-HexaCDD 1,2,3,7,8,9-HexaCDD 1,2,3.4,6,7,8- HeptaCDD OCDD Total 2,3,7,8-TCDF 1,2,3,7,8-PentaCDF 2,3,4,7,8-PentaCDF 1,2,3,4,7,8-HexaCDF 2,3,4,6,7,8-HexaCDF 1,2,3.6,7,8-HexaCDF 1,2,3,7.8,9-HexaCDF 1,2,3,4,6,7,8-HeptaCDF 1,2,3,4,7,8,9-HeptaCDF OCDF Total TEF 1.0 0.3 0.1 0.1 0.1 0.01 0.001 0.1 0.1 0.5 0.1 0.1 0.1 0.1 0.1 0.1 0.001 Cone. (ppm) 0.5 0.5 0.2 0.2 0.2 0.7 1.0 3.3 17 22 12 26 1.2 7.7 1.5 10 -- 13 110.4 2,3,7,8-TCDD eq. (ppm) 0.50 0.25 0.02 0.02 0.02 0.007 0.001 0.818 1.7 2.2 6.0 2.6 0.12 0.77 0.15 1.0 -- 0.013 14.553 PCDFs in human adipose tissue from 12 individuals and these levels are similar to those observed in human milk samples.57"61 The concentrations of "2,3,7,8-TCDD equivalents" in the PCDD and PCDF fractions were 12.01 and 8.38 ppt, respectively. High resolution analysis of these samples for the coplanar PCB con geners showed that the concentration of "2,3,7,8-TCDD equiv alents" was 41.0 ppt which was approximately two times higher than the concentration of the PCDDs plus PCDFs. The results of several studies by Tanabe and co-workers have reported that for extracts from diverse fish and wildlife samples the concen trationof "2,3,7,8-TCDD equivalents" for the coplanar PCBs is significantly greater than the corresponding values observed for the 2,3,7,8-substituted PCDDs plus PCDFs.375'384 These data indicate that future research on the potential hazards and risks associated with 2,3,7,8-TCDD and related compounds should not focus exclusively on the PCDDs an PCDFs but should also evaluate the contribution of toxic PCB congeners and other po tential toxic halogenated aromatics. VII. CONCLUSIONS 1. Toxic halogenated aromatics have been identified as com plex mixtures in almost every component of the global 19422 76 Volume 21, Issue 1 wnwwti Toxicology V i W i c O D Equivalents in Human Adipose Tissue Congener Wm * 7,8-TCDD ' i 'i ' t 7 g-PentaCDD ^ ! & ; 7.3-H acDD ! I J 6 .7.8-HexaCDD !*J 7.8,9-HexaCDD ^ .'4 .6 ,7 ,8- HeptaCDD 0CDD Total 23.7.8-TCDF j 2,3,7.8-PcntaCDF j '3 t4,7,8-PentaCDF j ' 2[3,4,7,8-HexaCDF 2 3 '4,6,7,8-HexaCDF 12^3,6,7,8-HexaCDF l'2,3,7,8,9-HexaCDF 1,2,3.4,6,7.8- HeptaCDF 1,2,3,4,7,8,9- HeptaCDF OCDF Total 3,3',4,4',5-PentaCB 3,3',4,4',5,5'-HexaCB 3,3',4,4'-TetraCB 2,3,3',4,4'-PentaCB 2,3',4,4',5-PentaCB 2,3,3',4,4' ,5-HexaCB Total TEF 1.0 0.5 0.1 0.1 0.1 0.01 0.001 0.1 0.1 0.5 0.1 0.1 0.1 0.1 0.1 0.1 0.001 0.1 0.05 0.01 0.001 0.001 0.001 Cone. (PP*) 3.7 6.4 3.9 34 5.7 33 510 3.1 0.5 11.0 5.6 1.4 5.3 -- 2.9 -- -- 330 90 350 2,3,7,8-TCDD eq. (ppt) 3.7 3.2 0.39 3.4 0.57 0.33 0.510 12.010 0.31 0.05 6.5 0.56 0.14 0.53 0.29 8.38 33.0 4.5 3.5 41.0 ecosystem including fish, wildlife, and human adipose tissue, serum, and milk. 2. 2,3,7,8-TCDD and related toxic compounds elicit com parable biochemical and toxic responses in human, lab oratory animals, and mammalian cells in culture. There is good evidence that the toxic halogenated aromatic hy drocarbons cause their responses through a common (Ah) receptor-mediated mechanism of action. 3. Not surprisingly, the activity of different halogenated aromatics is structure-dependent and several studies have reported the SARs among the different classes (e.g., PCDDs, PCDFs, and PCBs) of halogenated aromatics. 4. The existence of well-defined SARs for 2,3,7,8-TCDD and related compounds provides the mechanistic basis for the development of TEFs for halogenated aromatics. 5. The TEF values proposed in this review, and other similar TEFs, have been utilized to predict the toxicity of several different halogenated aromatic mixtures. The calculated "2,3,7,8-TCDD equivalents" using TEFs were similar to the observed in vivo toxicities of selected halogenated aromatic mixtures and these results validate the use of TEFs in the hazard and risk assessment of toxic halo genated aromatics. ACKNOWLEDGMENTS The financial assistance of the National Institutes of Health (P42-ES04917) and the Texas Agricultural Experiment Station is gratefully acknowledged. The author is a Burroughs Well come Toxicology Scholar. The invaluable assistance of Ms. Lorena Arguijo and Mr. Mark Harris in the preparation of this review is also appreciated. REFERENCES 1. Rappe, C., Buser, H. R., and Bosshardt, H.-P., Dioxins, dibenzofurans and other polyhalogenated aromatics: production, use, for mation and destruction. A n n . N .Y . A ca d . S c i., 320, i, 1979. 2. Hutzinger, O., Safe, S., and Zitko, V., The C hem istry o f P C B s, CRC Press, Boca Raton, FL. 1974. 3. Brinkman, U. A. Th. and De Kok, A., Production, properties and uses, in H alogenated B iphenyls, T erphenyls. N a p h th a len es, D lb en zodioxins a n d R elated P roducts, Kimbrough, R. D., Ed., Elsevier/North-Holland, Amsterdam, 1980, 1. 4. Pomerantz, I., Burke, J., Firestone, D., McKinney, J ., Roach, J., and Trotter, J., Chemistry of PCBs and PBBs, E nviron. H ealth P ersp ect., 24, 133, 1978. 5. Rappe, C. And Buser, H. R., Chemical properties and analytical methods, in H alogenated B iphenyls. T erphenyls. N a p h th a len es. D ibenzodioxins a n d R elated P roducts, Kimbrough, R. D., Ed., Elsevier/North-Holland, Amsterdam 1980, 41. 6. Ballschmiter, K., Rappe, C., and Buser, H. R., Chemical proper ties, analytical and environmental levels ofPCBs, PCTs, PCNs and PCBs, inH alogenatedBiphenyls, Terphenyls, Naphthalenes, D ibenzodioxins a n d R elated Products. 2nd ed., Kimbrough, R. D. and Jensen, A. A., Eds., Elsevier/North-Holland, Amsterdam, 1989,47. 7. De Voogt, P. and Brinkman, U. A. Th., Production properties and usage of polychlorinated biphenyls in, H a lo genated B iphenyls, Ter p h e n y ls, N aphthalenes. D ibenzodioxins, an d R ela te d P ro d u cts, 2nd ed., Kimbrough. R. D. and Jensen, A. A., Eds., Elsevier/NorthHolland, Amsterdam, 1989, 1. 8. Risebrough, R. W., Rieche, P., Herman, S. G., Peakall, D. B., and Kirven, M. N., Polychlorinated biphenyls in the global eco system, N ature, 220, 1098, 1968. 9. Fishbein, L., Chromatographic and biological aspects of polychlor inated biphenyls, J . C hrom atogr., 68, 345, 1972. 10. Ballschmiter, K., Buchert, H., and Bihler, S., Baseline studies of global pollution, Fresenius Z . A nal. C hem ., 306, 323, 1981. 11. Ballschmiter, K., Scholz, Ch., Buchert, H., Zell, M., Figge, K., Polzhofer, K., and Hoerschelmann, H., Studies of the global base line pollution, Z. F resenius A n a l. C h em ., 309, 1, 1981. 12. Wasserman, M., Wasserman, D., Cucos, S., and Miller, H. J., 1990 77 Critical Reviews In World PCBs map: storage and effects in man and his biologic en vironment in the 1970s. A n n . N .Y . A ca d . S c i., 320, 69, 1979. 13. Safe, S., Safe, L,, and MulUn, M., Polychlorinaged biphenyls: environmental occurrence and analysis, in P o lychlorinated B iphenyls (P C B s: M am m alian a nd E nviro n m en ta l T oxicology, Vol. l.SpringerVerlag, Heidelberg. 1987, 133. 14. Harvey, G. R. and Steinhauer, W. G., Atmospheric transport of polychlorobiphenyls to the North Atlantic, A tm os. E n viro n ., 8, 777, 1974. 13. Tanabe, S., Hidaka, H.,andTatsukaw a, R., PCBs and chlorinated hydrocarbon pesticides in Antarctic atmosphere and hydrosphere, C hem osphere, 12, 277, 1983. 16. Atlas, E. and Glam, C. S., Global transport of organic pollutants: ambient concentrations in the remote marine atmosphere. Science. 211, 163, 1981. 17. Stratton, C. L. and Sosebee, J. B., Jr., PCB and PCT contami nation of the environment near sites of manufacture and use. Bull. E nviron. Coraam . T oxicol.. 10, 1229, 1976. 18. Murphy, T. J., Pokojowczyk, J. C., and Mullin, M. D., Vapor exchange of PCBs with Lake Michigan: the atmosphere as a sink for PCBs, in P hysical B ehavior o f P C B s In th e G reat L a kes. Mackay, D., Paterson. S.. Eisenreich, S. J., and Simmons, M. S., Eds., Ann Arbor Science, Ann Arbor, MI, 1983, 49. 19. International Joint Commission, UC Great Lakes Water Quality -- Appendix E, Status Report on the Persistent Toxic Pollutants in the Lake Ontario Basin. 1977. 20. Krauss, P. P., Suns, K., and Buckley, E. H., Monitoring of PCBs in water, sediments and biota of the Great Lakes -- some recent examples, in P hysical B eh a vio r o f P C B s in th e G rea t L a kes. Mackay, D., Paterson, S., Eisenreich, S. I., and Simmons, M. S., Eds., Ann Arbor Science, Ann Arbor, MI, 1983, 367. 21. Gregor, D. J. and Gummer, W. D., Evidence of atmospheric transport and deposition of organochlorine pesticides and polychlor inated biphenyls in Canadian Arctic snow, E nviron. S ci. Technol.. 23, 561, 1989. 22. Rappe, C., Bergqvist, P. A., and Kjeiler, L. O., Levels, trends and patterns of PCDFs in Scandinavian environmental samples, C he m osphere, 18, 651, 1989. 23. Rappe, C., Nygren, M., Lindstrom, G., and Hansson, M., Diox ins and dibenzofurans in biological samples of European origin, C h e m osphere. 15, 1635, 1986. 24. Rappe, C. and Kjeiler, L.-O., PCDDs and PCDFs in environmental samples, air, particulates, sediments and soil. C hem osphere. 16, 1775, 1987. 25. Rappe, C., Andersson, R., Bergqvist, P. A., Brohede, C., Hans son, M., Kjeiler, L. O., Lindstrom, G., Marklund, S., Nygren, M., Swanson, S. E., Tysklind, M., and Wiberg, K., Overview on environmental fate of chlorinated dioxins and dibenzofurans. Sources, levels and isomertic pattern in various matrices, C hem os phere. 16, 1603, 1987. 26. Rappe, C., Kjeiler, L.-O., Bmckmann, P., and Hackhe, K.-H., Identification and quantification of PCDDs and PCDFs in urban air, C hem osphers, 17, 3, 1988. 27. Rappe, C., Marklund, S., Kjeiler, L.-O., and Lindskog, A., Long-range transport of PCDDs and PCDFs on airborne particles, C hem osphere. 18, 1283, 1989. 28. Norwood, C. B., Hackett, M., Prueil, R. J ., Butterworth, B. C., Williamson, K. J., and Naumann, S. M ., Polychlorinated dibenzop-dioxins and dibenzofurans in selected estuarine sediments, C he m osphere, 18, 553, 1989. 29. Czuczwa, J. M. and Hites, R. A., Environmental fate of combusiongenerated polychlorinated dioxins and fiirans. E nviron. S ci. T echnol., 18, 444, 1984. 30. Czuczwa, J. M. and Hites, R. A., Dioxins and dibenzofurans in air, soil and water, in D ioxins in the Environm ent. Kamrin, M.A. and Rodgers, P. W., Eds., Hemisphere, Washington, D.C., 1985, 85. 31. Czuczwa, J. M. and Hites, R. A., Airborne dioxins and dibenzo furans: sources and fates. E nviron. Sci. Technol.. 20, 195, 1986. 32. Czuczwa, J. M., McVeety, B. D., and Hites, R. A., Polychlori nated dibenzo-p-dioxins and dibenzofurans in sediments from Siskiwit Lake, Isle Royale, Science, 226, 568, 1984. 33. Czuczwa, J. M., Niessen, F., and Hites, R. A., Historical record of polychlorinated dibenzo-p-dioxins and dibenzofurans in Swiss lake sediments. C hem osphere, 14, 1175, 1985. 34. Sullivan, J. R., Delfino, J., Buelow, C. R., and Sheffy, T. B., Polychlorinated biphenyls in the fish and sediment of the Lower Fox River, Wisconsin, B ull. E nviron. C ontam . T oxicol.. 30, 58, 1983. 35. Paasivirta, J., Sarkka, J., Surma-Aho, K., Humpi, T., Kuokkanen, T., and Marttinen, M ., Food chain enrichment of orga nochlorine compounds and mercury in clean and polluted lakes of Finland, C hem osphere. 12, 239, 1983. 36. Zabik, M. E., Merrill, C., and Zabik, M. J., PCBs and other xenobiotics in raw and cooked carp, B ull. E nviron. Contam . T oxicol., 28, 710, 1982. 37. Stalling, D. L., Huckins, J. N., Petty, J. D., Johnson, J. L., and Saunders, H. O., An expanded approach to the study and mea surement of PCBs and selected planar aromatic environmental pol lutants, A n n . N .Y . A c a d .S c i., 320, 48, 1979. 38. Brunn, H. and Manz, D., Contamination of native fish stock by hexachlorobenzene and polychlorinated biphenyl residues, B ull. E n viron. C ontam . T oxicol., 28, 599, 1982. 39. Wickstrom, K., Pyysalo, H ,, and Perttila, M., Organochlorine compounds in the liver of cod (G adus M o rhua) in the northern Baltic, C hem osphere. 10, 999, 1981. 40. Olsen, P., Settle, H., and Swift, R., Organochlorine residues in wings of ducks in south-eastern Australia, A u st. W ildl. R es., 7,139, 1980. 41. Wharfe, J. R. and Van Den Broek, W. L. F., Chlorinated hydro carbons in macroinvertebrates and fish from the Lower Medway Es tuary, Kent, M ar. P ollut. B u ll., 9, 76, 1978. 42. Mowrer, J., Calambokidis, J., Musgrove, N., Drager, B., Beug, M. W., and Herman, S. G., Polychlorinated biphenyls in cottids, mussels and sediments in Southern Puget Bay, Washington, Bull. E n viro n . C ontam . T oxicol., 18, 588, 1977. 43. Castelli, M. G., Martelli, G. P., Spagone, C., Capellini, L., and Fanelli, R., Quantitative determination of polychlorinated biphenyls (PCBs) in marine organisms analysed by high resolution gas chro matography and selected ion monitoring, C hem osphere. 12, 291. 1983. 44. Tanabe, S., Kannan, N., Subramanian, A., Watanabe, S., and Tatsukawa, R., Highly toxic coplanar PCBs: occurence, source per sistency and toxic implications to wildlife and human, E nviron. P o l lut.. 47, 147, 1987. 45. Safe, S., Safe, L., Mullin, M., Polychlorinated biphenyls (PCBs): congener-specific analysis of a commercial mixture and a human milk extract, J . A gric. F ood C hem ., 33, 24, 1985. 46. Rappe, C., Buser, H. R., Stalling, D. L., Smith, L. M., and Dougherty, R. C., Identification of polychlorinated dibenzofurans in environmental samples. N ature, 292, 524, 1981. 47. Stalling, D. L., Smith, L. M., Petty, J. D., Hogan, J. W., John son, J. L., Rappe, C., and Buser, H. R., Residues of polychlor inated dibenzo-p-dioxins and dibenzofurans in Laurentian Great Lakes fish, in H um an a n d E nviro n m en ta l R isks o f C hlorinated D io xin s a n d R elated C om pounds, Tucker, R. E., Young, A. L., and Gray, A. P., Eds., Plenum Press, New York, 1983, 221. 78 Volume 21, Issue 1 Toxicology Ryan J. J.. Lau, p - PUon' J* c -> Lewis, D., 2,3,7,8Tetrachlorodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzofuran residues in Great Lakes Commercial and Sport Fish, in C hlorinated D ioxins a n d D ibenzofurans in the T otal E nvironm ent, Choudhary, G., Keith, L. H., and Rappe, C.. Eds., Butterworths, Boston, MA, 1983. 87. ?- 49. Van den Berg, M., Blank, F., Heeremans, C., Wagenaar, H., and Olle, K., Presence of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans in fish-eating birds from fish from the Netherlands. A rch E nviron. C ontam . T o xico l., 16, 149, 1987. jO Oehme, M., Furst, P., Kruger, C., Meemken, H. A., and Groebel, W ,, Presence of polychlorinated dibenzo-p-dioxins, dibenzo furans and pesticides in arctic seal from Spitzbergen, C hem osphere. 17, 1291, 1988. 31. Miyata. H., Takayama, K., Ogaki, J ., Kashimoto, T., and Fukushima, S., Polychlorinated dibenzo-p-dioxin in blue mussel from marine coastal water in Japan, B ull. E nviron. C ontam . T o xico l., 39, 877, 1987. 32. Miyata, H., Takayama, K., Ogaki, J., Mimura, M., and Kash imoto, T., Study on polychlorinated dibenzo-p-dioxins and diben zofurans in rivers and estuaries in Osaka Bay In Japan, T oxicol. E nviron. C h em ., 17,91, 1988. 33. Zacharewski, T., Safe, L., Safe, S., Chittim, B., DeVault, D., Wilberg, K., Bergqvist, P. A., and Rappe, C., Comparative anal ysis of polychlorinated dibenzo-p-dioxin and dibenzofuran congeners in Great Lakes fish by gas chromatography-mass spectrometry and in vitro enzyme induction activities. E nviron. Sei. T echnol.. 23, 730, 1989. 54. Bignert, A., Olsson, M., Bergqvist, P.-A., Bergek, S., Rappe, C ., de Wit, C., and Jansson, B., Polychlorinated dibenzo-p-dioxins (PCDD) and dibenzofurans (PCDF) in seal blubber, C hem osphere, 19, 551. 1989. 55. Bergqvist, P.-A., Bergek, S., Hallback, H., Rappe, C., and Slorach, S. A., Dioxins in cod and herring from the seas around Sweden, C hem osphere, 19, 513, 1989. 56. Koistinen J., Paasivirta, J., and Vuorinen, P. J., Dioxins and other planar polychloroaromatic compounds in Baltic, Finnish and Arctic fish samples, C hem osphere. 19, 527, 1989. 57. Beck, H., Eckart, K., Mathar, W., and Wittkowski, R., Depen dence of PCDD and PCDF levels in human milk on various param eters in the Federal Republic of Germany. Chem osphere. 18, 1063, 1989. 58. Nygren, M., Hansson, M., Sjostrom, M., Rappe, C., Kahn, P., Gochfeld, M., Velez, H., Gbent-Guenther, T., and Wilson, W. P., Development and validation of a method for determination of PCDDs and PCDFs in human blood plasma. A multivariate com parison of blood and adipose tissue levels between Vietnam veterans and matched controls, C hem osphere, 17, 1663, 1988. 59. Rappe, C ., Nygren, M ,, Buser, H., Masuda, Y., Kuroki, H., and Chen, P. H., Identification of polychlorinated dioxins (PCDDs) and dibenzofurans (PCDFs) in human samples, occupational exposure and Yusho patients. E nviron. Sei. R es., 26, 241, 1983. 60. Schecter, A., Ryan, J. J., and Constable, J. D., Chlorinated diox ins and dibenzofurans in human milk from Japan, India and the United States of America, C hem osphere, 18, 975, 1989. 61. Fingerhut, M. A., Sweeney, M. H., Patterson, D. G., J r., Piacitelli, L. A., Morris, J. A., Marlow, D. A., Hornung, R. W., Cameron, L. W., Connally, L. 8 ., Needham, L. L., and Hal perin, W. E., Levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the serum of U.S. chemical workers exposed to dioxin contaminated products: interim results, C hem osphere, 19, 835, 1989. 62. Goldman, L. R., Hayward, D. G., Flattery, J ., Handy, M. E., Patterson, D. G., J r., Needham, L. L., Siegel, D., Chang, R .,' Stephens, R. D., and Kizer, K. W., Serum, adipose and autopsy tissue and PCDF levels, in people eating dioxin contaminated beef and chicken eggs. C hem osphere, 19, 841, 1989. 63. Smith, A. H., Matheson, D. P., Fisher, D. O., and Chapman, C. J., Preliminary report of reproductive outcomes among pesticide applicators using 2,4,5,-T, N .Z . M ed. J ., 93, 177, 1981. 64. Smith, A. H., Fisher, D. O., Giles, H. J., and Pearce, N,, The New Zealand soft tissue sarcoma case-control study: interview find ings concerning phenoxyaceuc acid exposure, Chemosphere, 12, 565, 1983. 65. Smith, A. H,, Pearce, N. E., Fisher, D. O., Giles, H. J ,, Teague, C. A., and Howard, J, K., Soft tissue sarcoma and exposure to phenoxyherbicides and chlorophenols in New Zealand . J . N atl. C an cer In st., 73, 1111, 1984. 66. Suskind, R. R., Long-term health effects of exposure to 2,4,5-T and/or its contaminants, C hem osphere, 12, 769, 1983. 67. Susklnd, R. R., Chloracne, "the hallmark of dioxin intoxication", Scand. J . W ork E nviron. H ealth, 11, 165, 1985. 68. Suskind, R. R. and Hertzerb, V. S., Human health effects of 2,3,5T and its toxic contaminants, JA M A , 251, 2372, 1984. 69. Nelson, C. J., Holson, J. F., Green, H. G., and Gaylor, D. W., Retrospective study of the relationship between agricultural use of 2,4,5-T and cleft palate occurrence in Arkansas, Teratology. 19. 377, 1979. 70. Zack, J. A. and Suskind, R. R., The mortality experience of work ers exposed to tetrachlorodibenzodoxin in a trichlorophenol process accident, J. O ccup. M e d ., 22, 11, 1980. 71. Hardell, L., Relation of soft-tissue sarcoma, malignant lymphoma and colon cancer to phenoxy acids, chlorophenols and other agents, Scand. J. W ork E nviron. H ea lth , 7, 119, 1981. 72. Hardell, L., Epidemiological studies on soft-tissue sarcoma and ma lignant lymphoma and their relation to phenoxy acid or chlorophenol exposure, Urnea U niv. M ed . D iss.. New Series No. 65, University of Umea, Sweden, 1981, 139. 73. Hardell, L. and Bengtsson, N. O., Epidemiological study of so cioeconomic factors and clinical findings in Hodgkin's disease, and reanalysis of previous data regarding chemical exposure, B r. J . C an cer, 48. 217, 1983. 74. Hardell, L. and Eriksson, M., Soft-tissue sarcoma and exposure to dioxins. Lancet, 2, 868, 1986. 75. Hardell, L. and Sandstrom, A., Case-control study: soft-tissue sarcomas and exposure to phenoxyacetic acids or chlorophenols. B r. J . Cancer, 39, 711, 1979. 76. Hardell, L., Eriksson, M., Leaner, P., and Lundgren, E., Ma lignant lymphoma and exposure to chemicals, especially organic sol vents, chlorophenols and phenoxy acids: a case-control study, Br. J. Cancer, 43, 169, 1981. 77. Hardell, L., Johansson, B., and Axelson, O., Epidemiological study of nasal and nasopharyngeal cancer and their relations to phenoxyacid or chlorophenol exposure. A m . J . Ind. M ed., 3, 247, 1982. 78. Thiess, A. M., Frentzei-Beyme, R., and Link, R., Mortality study of persons exposed to dioxin in a trichlorophenol-process accident that occurred in the BASF AG on November 17, 1953, A m . J. Ind. M ed .. 3, 179, 1982. 79. Sterling, T. D. and Arundel, A. V., Health effects of phenoxy herbicides: a review, Scand. J . W ork E nviron. H ealth, 12, 161, 1986. 80. Hoffman, R. E., Stehr-Green, P. A., Webb, K. B., Evans, R. G., Knutsen, A. P., Schramm, W. F., Staake, J. L., Gibson, B. B., and Steinberg, K. K., Health effects of long-term exposure to 2,3,7,8-TCDD, JA M A . 255, 2031, 1986. 81. Hoar, S. K., Blair, A., Holmes, F. F., Boyse, C. D.., Robei, R. J., Hoover, R., and Fraumeni, J. F., J r., Agricultural herbicide 1990 1 9 4 S S Critical Reviews In use and risk of lymphoma and soft-tissue sarcoma, JA M A , 256, 1141, 1986. 82. Jlrasek, L., Kalensky, J., and Kubec, K.,Acne chlorina and por phyria cutanea tarda during the manufacture of herbicides, Cesk. D erm a to l.. 48, 306, 1973. 83. Cook, R. R., Townsend, J. C., Ott, M. G., and Silversteln, L. G., Mortality experience of employees exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), J. O ccup. M ed., 22, 530, 1980. 84. Axelson, O. and Sundell, L., Herbicide exposure, mortality and tumor incidence: an epidemiological investigation on Swedish rail road workers. W ork E nviron. H ealth, 11, 21. 1974. 85. Axelson, O., Sundell, L., Anderson, K., Edling, C., Hogstedt, C., and Kling, H., Herbicide exposure and tumor mortality: an updated epidemiologic investigation on Swedish railroad workers, S ca n d . J . W ork E nviron. H ea lth , 6, 73, 1980. 86. Flllpplni, G., Bordo, B., Crenna, P., Massetto, N., Musicco, M., and Boeri, R., Relationship between clinical and electrophysiological findings and indicators of heavy exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin, Scand. J. W ork E nviron. H ealth. 7. 257, 1981. 87. Eriksson, M., Hardell, L., Berg, N. O ., Moller, T., and Axelson, 0 . , Soft-tissue sarcomas and exposure to chemical substances: a casereferent study, B r. J . Ind. M e d ., 38, 27, 1981. 88. Flngerhut, M. A., Halperin, W. E., Honchar, P. A., Smith, A. B., Groth, O. H., and Russell, W. O., An evaluation of reports of dioxin exposure and soft tissue sarcoma pathology in U.S. chemical workers, in B iological M echanism s o f D io xin A ctio n , Banbury Report 18, Poland, A. and Kimbrough, R. 0 ., Eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1984, 461. 89. Flngerhut, M. A., Sweeney, M. H., Halperin, W. E., and Schnorr, T. M., Epidemiology of populations exposed to dioxins, in Solving H a za rd o u s W aste P roblem s: L ea rn in g F ro m D ioxins. Exner, J. H., Ed., ACS Symposium Series 338, American Chemical Society, Washington, D.C., 1987, 143. 90. Coggon, D. and Acheson, E. D-, Do phenoxy herbicides cause cancer in man?. Lancet, 1, 1057, 1982. 91. Gustavsson, P., Hogstedt, C., and Rappe, C ., Short-term mortality and cancer incidence in capacitor manufacturing workers exposed to polychlorinated biphenyls (PCBs), A m . J . In d . M e d .. 10, 341, 1986. 92. Bertazzi, P. A., Zochetti, C., Guercilena, S., Foglia, M. D., Pesatori, A., and Riboldi, L,, Mortalilty study of male and female workers exposed to PCBs. Helsinki, Finland, presented at Int. Symp. on Prevention of Occupational Cancer, 1981. 93. Bertazzi, P. A., Riboldi, L., Pesatori, A., Radice, L., and Zo chetti, C., Cancer mortality of capacitor manufacturing worker. A m . J . In d . M ed ., 11, 165, 1987. 94. Brown, D. P., Mortality of workers exposed to polychlorinated bi phenyls -- an update, A rch. E n viro n . H ea lth . 42, 333, 1987. 95. Brown, D. P. and Jones, M., Mortality and industrial hygiene study of workers exposed to polychlorinated biphenyls, A rch. Environ. H ealth. 36. 120, 1981. 96. Smith, A. H., Shindell, S., and Kurland, A. T., Epidemiology, in H um an H ea lth A sp ects o f E n viro n m en ta l E x p o su re to P olychlor in a ted D ibenzo -p -d io xin s a nd D ib en zo fu ra n s, UAREP 1988, 1-1. 97. des Rosiers, P. E., Chlorinated combustion products from fires in volving transformers and capacitors, C hem osphere, 16, 1881, 1987. 98. Chen P. H. and Hsu, S. T., PCB poisoning from toxic rice-bran oil in Taiwan, in P C B s a n d The E nironm ent, Vol. 3, Waid, J. S., Ed., CRC Press, Boca Raton, FL, 1987, 27. 99. Kashimoto, T. and Miyata, H-, Differences between Yusho and other kinds of poisoning involving only PCBs, in P C B s and The E nvironm ent, Waid J. S., Ed., CRC Press, Boca Raton, FL, 1987, 1. 100. Kashimoto, T., Miyata, H., Kunita, S., Tung, T., Hsu, S. T., Chang, K. J., Tang, S. Y., Ohi, G., Nakagawa, J., and Ya mamoto, S. I., Role of polychlorinated dibenzofurans in Yusho (PCB poisoning). A rch E n viro n . H ealth. 36, 321, 1981. 101. Chen, P. H., Wong, C. K., Rappe, C., and Nygren, M., Poly chlorinated biphenyls, dibenzofurans and quaterphenyls in toxic ricebran oil and in the blood and tissues of patients with PCB poisoning (Yu-Cheng) in Taiwan, E nviron. H ealth P erspect.. 59, 59, 1985. 102. Kuratsune, M., Yusho, in H alogenated Biphenyls, Terphenyls, N a p h th a len es. D ib en zo d io xin s a n d R elated Products, Kimbrough, R. D., Ed., Elsevier, Amsterdam, 1980, 287. 103. Kuratsune, M., Yusho with reference to Yu Cheng, in H alogenated Biphenyls, Terphenyls, Naphthalenes. Dibenzodioxins and R elated P roducts. 2nd ed., Kimbrough, R. D. and Jensen, A. A., Eds., Elsevier/North-Holland, Amsterdam, 1989, 381. 104. Rogan, W. J ., Yu Cheng, in H alogenated B iphenyls, T erphenyls, N a p hthalenes, D ib en zo d io xin s a n d R ela ted P roducts. 2nd ed., Kim brough, R. D. and Jensen, A.A., Eds., Elsevier/North-Holland, Am sterdam. 1989, 401. 105. Caramaschi, F., Como, G., Favaretti, C,, Glambelluca, S. E., Montesarchio, E., and Fara, G. M., Chloracne following envi ronmental contamination by TCDD in Seveso, Italy, Int. J . E p id e m iol.. 10, 135, 1981. 106. Mocarelli, P., Marocchi, A., Brambilla, P., Gerthoux, P. M., Young, D. S., and Mantel, N., Clincial laboratory manifestations of exposure to dioxins in children: a six-year study of the effects of an environmental disaster near Seveso, Italy, JAM A, 256, 2687, 1986. 107. Tenchinl, M. L., Crimuado, C., Pachetti, G., Mottura, A., Agosti, S. , and De Carli, L., A comparative cytogenetic study on cases of induced abortions in TCDD-exposed and nonexposed women. E n viron. M u ta g en ., 5, 73, 1983. 108. Reggiani, G., Anatomy of a TCDD spill: the Seveso accident. H a z a rd A sse ss. C hem . C urr. D e v ., 2, 269, 1983. 109. Puntoni R., Merlo, F., Finl, A., Meazza, L., and Sand, L., Soft tissue sarcomas in Sevesco, L ancet, 2, 525, 1986. 110. Regiani, G. M., The Seveso accident: medical survey of a TCDD exposure, in H a lo g en a ted B ip h en yls, T erphenyls. N aphthalenes. D ibenzodioxin a n d R ela ted P roducts, 2nd ed.. Kimbrough, R. D. and Jensen, A. A., Eds., Elsevier/North-Holland Amsterdam, 1989,445. 111. New York State Department of Health, Love Canal: P ublic H ealth Tim e B o m b . A S p ecia l R e p o rt to the G overnor a nd L egislature, Al bany, NY, 1978. 112. Constable, J. D. and Hatch, M. C., Reproductive effects of her bicide exposure in Vietnam: recent studies by the Vietnamese and others. T eratogen. C arcinogen. M u ta g en ., 5, 231, 1985. 113. Greenwald, P., Kovasznay, B., Collins, D. N., and Therriault, G., Sarcomas of soft tissues after Vietnam service, J . Natl. C ancer In st., 73, 1107, 1984. 114. Lawrence, C. E., Reilly, A. A., Quyickenton, P., Greenwald, P., Page, W. F., and Kuntz, A. J ., Mortality patterns of New York State Vietnam veterans, A m . J. P ublic H ealth, 75, 277. 1985. 115. Lathrop, G. D., Wolfe, W. H., Machado, S. G.. Michalek, J. E., Karrison, T. G., Miner, J . C., Grubbs, W. D., Peterson, M. R., and Thomas, W. F., An epidemiologic investigation of health effects in Air Force personnel following exposure to herbicides: first follow-up examination results, Vol. 1, National Technical In formation Service, Springfield, VA, 1987. 116. Dan, B. B., Vietnam and birth defects, JAM A, 252, 936, 1984. 117. Kimbrough, R. D., Falk, H., Stehr, P., and Fries, G., Health implications of 2,3,7,8-tetrachlorodibenzodioxin (TCDD) contami nation of residential soil, J . Toxicol. E nviron. H ealth. 14, 47, 1984. 118. Knutsen, A. P., Roodman, S. T., Evans, R. G., Mueller, K. R., Webb, K. B., Green P. S., Hoffman, R. E., and Schramm, 80 Volume 21, Issue 1 18426 Toxicology W. F., Immmune studies in dioxin-exposed Missouri residents: Quail Run. B u ll. E nviron. C om am . T oxicol.. 39, 481, 1987. 119. Webb, K., Ayres, S., Slavin, R., Knutsen, A., Roodman, S,, Gedney, W. B., Schramm, W., Hotchkiss, R. L., Miller R., and Donnell, H. D., Results of a pilot study of health effects due to 2,3,7.8-tetrachlotodibenzo-p-dioxin contamination -- Missouri, JA M A . 251, 1139, 1984. 120. Stehr, P. A., Stein, G., Falk, H., Sampson, E., Smith, S. J., Steinberg, K., Webb, K., Ayres, S., Schramm, W., Donnell, H. D., and Gedney, W. B., A pilot epidemiologic study of possible health effects associated with 2,3,7,8-tetrachlorodibenzo-p-dioxin contaminations in Missouri, A rch . Environ. H ealth. 41, 16, 1986. ' 121. Hoffman, R. E. and Stehr-Green, P. A., Localized contamination with 2,3,7,8-tetrachlorodibenzo-p-dioxin: the Missouri episode, in H alogenated Biphenyls, Terphenyls, Naphthalenes, D ibenzodioxins and R ela te d Products. 2nd ed., Kimbrough, R. D. and Jensen, A. A., Eds., Elsevier/North-Holland, Amsterdam, 1989, 471. 122. Carter L. J., Michigan's PBB incident: chemical mixup leads to disaster. Science, 192, 240, 1976. 123. Robertson, L.W. and Chynoweth, D. P., Another halogenated hydrocarbon. E nvironm ent, 17, 25, 1975. 124. Chen, E., P BB : A n A m erican Tragedy, Prentice-Hall, Englewood Cliffs. NJ, 1979. 125. Eggington, J., The Poisoning o f M ichigan, Norton, New York, 1980. 126. Reich, M. J ., Environmental politics and science: the case of PBB contamination in Michigan, A m . J . P ublic H ealth, TS, 302, 1983. 127. Jackson, T. F. and Halbert, F. L., A toxic syndrome associated with the feeding of polybrominated biphenyl contaminated protein concentrate to dairy cattle, J . A m . Vet. M ed. A ssoc.. 165, 437, 1974. 128. Buser, H.-R., Brominated and brominated/chlorinated dibenzodiox ins and dibenzofurans: potential environmental contaminants, Che- m osphere, 16, 713, 1987. 129. Schulz, D. E., Petrtick, G., and Duinker, J. C., Complete char acterization of polychlorinated biphenyl congeners in commercial Aroclor and Clophen mixtures by multidimensional gas chromatog raphy-electron capture detection. Environ. Sci. Technol., 23, 852, 1989. 130. Haglund, E. and Bergman, A., Synthesis of hexa and heptachlo- ronaphthaienes, C hem osphere. 19, 195, 1989. 131. Rappe, C., Marklund, S., Buser, H. R., and Bosshardt, H. P., Formation of polychlorinated dibenzo-p-dioxins (PCDDs) and di- benozofurans (PCDFs) by burning or heating chlorophenates, C h e m o sp h ere. 7, 269, 1978. 132. Ahllng, B., Lindskog, A., Jansson, B., And Sunstrom, G., For mation of polychlorinated dibenzo-p-dioxins and dibenzofurans dur ing combustion of a 2,4,5-T formulation, C hem osphere, 8, 461, 1977. 133. Villanueva, E. C., Jennings, R. W., Burse, V. W., and Kim- borough, R. D., Evidence of chlorodibenzo-p-dioxin and chloro- dibenzofuran in hexachlorobenzene, J. A gric. F o o d C hem ., 22, 916, 1974. 134. Scholz, B. and Engler, M., Determination of polychlorinated di benzo-p-dioxins and dibenzofurans in wastes of technical hexa- chlorocyclohexane, C hem osphere, 16, 1829, 1987. 135. Heindl, A. and Hutzinger, O., Search for industrial sources of PCDD/PCDFs. II. Metal chlorides, Chem osphere, 15, 653, 1986. 136. Clement, R. E., Suter, S. A., Reiner, E., McCurvin, D., and Hollinger, D., Concentrations of chlorinated dibenzo-p-dioxins and dibenzofurans in effluents and centrifuged particulates from Ontario pulp and paper mills, C hem osphere. 19, 649, 1989. 137. Beck, H ., Drop, A., Eckart, K., Mathar, W., and Wittkowski, R., PCDDs, PCDFs and related compounds in paper products, C he m osphere. 19. 655, 1989. 138. Axegard, P. and Renberg, L., The influence of bleaching chemicals and lignin content on the formation of polychlorinated dioxins and dibenzofurans, C hem osphere, 19, 661, 1989. 139. Kuehl, D. W,, Butterworth, B. C., DeVlta, W. M,, and Sauer, C. P., Environmental contamination by polychlorinated dibenzo-pdioxins and dibenzofurans associated with pulp and paper mill dis charge, B iom ed. E nviron. M a ss S p ectro m ., 14, 443, 1987. 140. Clement, R. E., Tashiro, C., Suter, S., Reiner, E., and Hollinger, D. , Chlorinated dibenzo-p-dioxins (CDDs) and dibenzofurans (CDFs) in effluents and sludges from pulp and paper mills, C hem osphere. 18, 1189,'1989. 141. Wiberg, K., Lundstrom, K., Glas, B., and Rapp, C., PCDDs and PCDFs in consumers' paper products, C hem osphere, 19, 735, 1989. 142. Kitunen, V. H. and SaUdnoja-Salonen, M. S., Occurrence of PCDDs and PCDFs in pulp and board products, C hem osphere, 19,721, 1989. 143. OUe, K., Vermeulen, P., and Hutzinger, O., Chlorodibenzo-pdioxins and chiorodibenzofurans are trace components of fly ash and flue gas of some municipal incenerators in The Netherlands, C he m osphere, 6, 455, 1977. 144. Olie, K., Berg, M. v. d., and Hutzinger, O., Formation and fate of PCDD and PCDF from combustion processes, C hem osphere, 12, 627, 1983. 145. Crummett, W. B. and Townsend, D. I., The trace chemistries of fire hypothesis: review and update, C hem osphere. 13, 777, 1984. 146. Tlernan, T, O., Taylor M. L., G arrett, J. H., VanNess, G. F., Soldi, J. G., Dels, D. A., and Wagel, D. J ., PCDDs PCDFs and related compounds in the effluents from combustion processes, C he m osphere, 12, 595, 1983. 147. Hutzinger, O., Blumich, M. J ., Berg, M. v. d., and OUe, K., Sources and fate of PCDDs and PCDFs: an overview, C hem osphere, 14, 581, 1985. 148. Clement, R. E., Tostne, H. M., Osborne, J., Ozvadc, V., and Wong, G., Gas chromatographic/mass spectrometric determination of polychlorinated dibenzo-p-dioxins and dibenzofurans in incinerator stack emissions and fly ash: a 13-test study, B iom ed. E nviron. M ass Spectrom .. 17, 81, 1988. 149. Nottrodt, A., Duwel, U., and BaUschmiter, K., Influence of formation of PCDD/PCDF in MWI plants -- results of investiga tions at several plants in the FRG, C hem osphere. 19, 309, 1989. 150. Takeshita, R. and Akimoto, Y., Control of PCDD and PCDF for mation in fluidized bed incinerators, C hem osphere, 19, 345, 1989. 151. Benfenari, E., Gizzi, F., Reginato, R., FanelU, R., Lodi, M., and Tagllaferri, R., Polychlorinated dibenzo-p-dioxins (PCDD) and po lychlorinated dibenzofurans (PCDF) in emissions from urban incin erator. Correlation between concentration of micropollutants and combustion conditions, C hem osphere, 12, 1151, 1983. 152. Thoma, H., PCDD/F concentrations in chimney soot from house heating systems, C hem osphere, 17, 1369, 1988. 153. Hutzinger, O. and Fiedler, H., Sources and emissions of PCDD/PCDF, C hem ospheres, 18, 23, 1989. 154. Aittola, J.-P., Viinikainen, S., and Roivainen, J., The emission of PCDDs/PCDFs and related compounds from co-combustion of RDF with peat and wood chips, C hem osphere, 19, 387, 1989. 155. Chiu, C., Thomas, R. S., Lockwood, J., LI, K., Halman, R., and Lao, R. C. C., Polychlorinated hydrocarbons from power plants, wood burning and muncipal incinerators, C hem osphere, 12, 607, 1983. 156. BaUschmiter, K., BraunmiUer, I., Niemczyk, R., And Swerev, M., Reaction pathways for the formation of polychloro-dibenzo-p- 1990 1 9 4 2 7 81 Critical Reviews In dioxins (PCDD) and dibenzofurans in combustion processes, II. Chlorobenzenes and chlorophenols as precursors in the formation of polychloro-dibenzo-p-dioxins-dibenzofurans in flame chemistry, C hem osphere, 17, 995, 1988. 157. Choudhry, G. G., Olle, K., and Hutzlnger, O ., Mechanisms in the thermal formation of chlorinated compounds including polychlor inated dibenzo-p-dioxins. in C hlorinated D ioxins a n d R ela ted C o m p o u n d . Im pact on the Environm ent. Hutzinger, O., Frei, R. W., Merian. E., and Pocchiari, E., Eds., Pergamon Press. Oxford, 1982, 275. 158. Choudhry, G. G., Broecke, J. A. V. D., and Hutzlnger, O., Formation of polychlorodibenzofurans (PCDFs) by the photolysis of polychlorobenzenes (PCBzs) in aqueous acetonitrile containing phe nols, C hem osphere. 12, 487, 1983. 159. Karasek, F. W. and Dickson, L. C., Model studies of polychlor inated dibenzo-p-dioxin formation during municipal refuse inciner ator, Science, 237, 754, 1987. 160. Eklund, G., Pederson, J. R., and Stromberg, B., Methane hy drogen chloride and oxygen from a wide range of chlorinated organic species in the temperature range 400C--950oC, Chem osphere, 17, 575, 1988. 161. Buser, H. R. and Rappe, C., Formation of polychlorinated diben zofurans (PCDFs) from the pyrolysis of individual PCB isomers, C hem osphere, 8, 157, 1979. 162. Buser, H. R., Bosshardt, H.-P., and Rappe, C., Formation of polychlorinated dibenzofurans (PCDFs) from the pyrolysis of PCBs, C hem osphere, 7, 109, 1978. 163. Buser, H. R., Bosshardt, H.-P., and Rappe, C., Identification of polychlorinated dibenzofuran isomers in By ash and PCB pyrolyses, C hem osphere, 7, 419, 1978. 164. Buser, H. R., Formation of polychlorinated dibenzofurans (PCDFs) and dibenzo-p-dioxins (PCDDs) from pyrolysis of chlorobenzenes, C hem osphere, 8, 415, 1979. 165. Zoller, W. and Ballschmlter, K., Formation of polychlorinated dibenzcdioxins and dibenzofurans by heating chlorinated phenols and chlorophenates at various temperatures, F resenius Z . A n a l. C hem ., 323, 19. 1986. 166. Rappe, C ., Marklund, S., Buser, H. R., and Bosshardt, H. P., Formation of polychlorinated dibenzo-p-dioxins (PCDDs) and di benzofurans (PCDFs) by burning or heating chlorophenates, C he m o sphere, 7, 269, 1978. 167. Ahling, B., Bjorseth, A., and Lune, G., Formation of chlorinated hydrocarbons during combustion of polyvinyl chloride, C hem o sp h ere. 7, 799, 1978. 168. Christmann, W., Kasiske, D., Kloppel, K. D., Partscht, H., and Rotard, W., Combustion of polyvinylchloride -- an important source for the formation of PCDD/PCDF, C hem osphere, 19, 387, 1989. 169. Thoma, H , Rist, S., Hauschulz, G ,, and Hutzinger, O., Polybrominated dibenzodioxins and -furans from the pyrolysis of some flame retardants, C hem osphere, 15, 649, 1986. 170. Hosseinpour, J., Schwind, K. H., and Thoma, H., Brominated/chlorinated dibenzo-p-dioxins and furans: synthesis of stan dards and detection in fly ash from a municipal waste incinerator, C hem o sp h ere, 19, 109, 1989. 171. Dumler, R., Thoma, H., Lenoir, D., and Hutzlnger, O., Thermal formation of polybrominated dibenzodioxins (PBDD) and dibenzo furans (PBDF) from bromine containing flame retardants, C hem o sp h e re. 19, 305, 1989. 172. Muller M. D. and Buser, H.-R., Halogenated aromatic compounds in automotive emissions from leaded gasoline additives. E nviron. S c i. T ech n o l., 10, 1151, 1986. 173. M arklund, S., Rappe, C., Tysklind, M., and Ergeback, K.-E., Identificaton of polychlorinated dibenzofurans and dioxins in exhausts from cars run on leaded gasoline, C hem osphere, 16, 29, 1987. 174. Marklund, S., KjeUer, L .-0., Hansson, M., Tysklind, M., Rappe, C. , Collazo, H., and Dougherty, R., Determination of PCDDs and PCDFs in incineration samples and prolytic products, in C hlorinated D ioxins a n d D ibenzofurans in P erspective, Rappe, C., Keith, L. H., and Choudhry, G., Eds., Lewis Publishers, Chelsea, MI, 1986, 79. 175. Eitzer, B. D. and Hites, R. A. Polychlorinated dibenzo-p-dioxins and dibenzofurans in the ambient atmosphere of Bloomington, In diana, E n viro n . S ci. T echnol., 23, 1389, 1989. 176. Eitzer, B. D. and Hites, R. A., Atmospheric transport of poly chlorinated dibenzo-p-dioxins and dibenzofurans, E nviron. Sci. T ech nol.. 23, 1396. 1989. 177. Nakano, T., Tsiyi, M., and Okuno, T., Level of chlorinated or ganic compounds in the atmosphere, Chem osphere, 16, 1781, 1987. 178. Tientan, T. 0.,W ageI, D. J ., Vanness, G. F., G an tt, J. H., Sloch, J . G., and Harden, L. A., PCDD/PCDF in the ambient air of a metropolitan area in the U.S., C hem osphere, 19, 541, 1989. 179. Beck, H., Eckart, K., Mathar, W., and Wittkowski, R,, PCDD and PCDF body burden from food intake in the Federal Republic of Germany, C hem osphere, 18, 417, 1989. 180. Poland, A., Greenlee, W. F., and Kende, A. S., Studies on the mechanism of action the chlorinated dibenzo-p-dioxins and related compounds, A n n . N .Y . A cad. S c i., 320, 214, 1979. 181. Whitlock, J. P., J r ., The regulation of cytochrome P-450 gene expresin, A n n u . R ev. P h arm acol. T o xico l., 26, 333, 1986. 182. Whitlock, J . P., J r ., The regulation of gene expression by 2,3,7,8tetrachlorodibenzo-p-dioxin, P harm acol. R e v ., 39, 147, 1987. 183. Poland, A. and Knutson, J. C., 2,3,7,8-Tetrachlorodibenzo-p-dioxin and related halogenated aromatic hydrocarbons: examination of the mechanism of toxicity, A n n u . R e v . P h a rm a co l. T oxicol., 22, 517, 1982. 184. Safe, S., Polychlorinated biphenyls (PCBs) and polybrominated bi phenyls (PBBs): biochemistry, toxicology and mechanism of action, C rit. R ev. T oxicol.. 13, 319, 1984. 185. Safe, S. H., Comparative toxicology and mechanism of action of polychlorinated dibenzo-p-dioxins and dibenzofurans, A nnu. R ev. P harm acol. T oxicol., 26, 371, 1986. 186. Goldstein, J. A. and Safe, S., Mechanism of action and structureactivity relationships for the chlorinated dibenzo-p-dioxins and related compounds, in H a lo g en a ted B ip h e n yls, T erphenyls, N aphthalenes, D ibenzodioxins a n d R ela te d P roducts, 2nd ed., Kimbrough R. D. and Jensen, A. A., Eds., Elsevier/North-Holland, Amsterdam, 1989, 239. 187. Kociba, R. J. and Schwetz, B. A., Toxicity of 2,3,7.8-tetrachIorodibenzo-p-dioxin (TCDD), D rug M eta b . R ev., 13, 387, 1982. 188. Kociba, R. J. and Schwetz, B. A., A review of the toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) with a comparision of the toxicity of the other chlorinated dioxin isomers. A ssoc. F o o d D rug O fficials Q. B u ll., 46, 168, 1982. 189. Ryan, D. E., Thomas, P. E., Korzenlowski, D., and Levin, W., Separation and characterization of highly purified forms of liver mi crosomal cytochrome P-450 from rats treated with polychlorinated biphenyls, phnobarbital and 3-methylcholanthrene, J . B iol. C hem ., 254, 1365, 1979. 190. Thomas, P. E-, Reik, L., Ryan, D. E., and Levin, W., Regulation of three forms of cytochrome P-450 and epoxide hydrolase in rat liver microsomes: effects of age, sex and induction, J . B iol. C hem ., 265, 1044, 1981. 191. Parkinson, A., Safe, S., Robertson, L., Thomas, P. E., Ryan, D. E., Reik, L. M., and Levin, W., Immunochemical quantitation of cytochrome P-450 isozmyes and epoxide hydrolase in liver mi- 82 Volume 21, Issue 1 Toxicology crasomes from polychlorinated and polybraminated biphenyls: a study of structure activity relationships, J. B iol. C hem .. 258, 5967, 1983. 192. Dannan, G. A.. Guengerich, F. P., Kaminsky, L. S., and Aust, S. D., Regulation of cytochrome P-450. Immunochemical quanti tation of eight isozymes in liver microsomes of rats treated with polybraminated biphenyl congeners, J . B iol. C hem ., 258, 1282, 1983. 193. Hardwick, J. P., Llnko, P., and Goldstein, J. A., Dose response for induction of two cytochrome P-450 isozymes and their mRNAs by 3,4,5,3',4',5'-hexachlorobiphenyl indicating coordinate regula tion in rat liver. M ol. P harm acol.. 27, 676, 1985. 194. Goldstein, J . A. and Llnko, P., Differential induction of two 2,3,7,8tetrachlorodibenzo-p-dioxin-inducible forms of cytochrome P-450 in extrahepatic verses hepatic tissues. M ol. P harm acol.. 25, 185, 1984. 195. Dees, J, H., Masters, B. S. S., Muller-Eberhard, U., and John son, E. F., Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxins and phenobarbital on the occurrence and distrbution of four cytochrome P450 isozymes in rabbit kidney, lung and liver. C ancer R es.. 42, 423, 1982. 196. Ozawa, N., Yoshihara, S., Kawano, K., Okada, Y., and Yoshimura, H., 3,4.5,3',4'-PentachIorobiphenyl as a useful inducer for purification of rat liver microsomal cytochrome P-448, B iochem . B iophys. R es. C om m on.. 91, 1140, 1979. 197. Scheutz, E. G., Wrighton, S. A., Safe S. H., and Guzelian, P. S., Regulation of cytochrome P-450p by phenobarbital and phenobarbital-like inducers in adult rat hepatocytes in primary monolayer culture and in vivo. B iochem istry, 25, 1124, 1986. 198. Yoshimura, H., Yoshihara, S., Koga, N., Nagata, K., Wada, I., Kuroki J., and Hokama, Y., Inductive effect on hepatic enzymes and toxicity of congeners of PCBs and PCDFs, E nviron. H ealth P ersp ect., 59, 113, 1985. 199. Keys, B., Hlavinka, M., Mason, G., and Safe, S., Modulation of rat hepatic microsomal testosterone hydroxylases by 2,3,7,8-tetrachlorodibenzo-p-dioxin and related toxic isostereomers. Can. J . P hys io l. P h a rm a co l., 63, 1537, 1985. 200. Moore, R. W., Potter, C. L., Theobald, H. M., Robinson, J. A., and Peterson, R. E., Androgenic deficiency in male rats treated with 2,3,7,8-tetrachiorodibenzo-p-dioxin, Toxicol. A ppl. P h a rm a co l., 79, 99, 1985. 201. Mebus, C. A., Reddy, V. R., and Piper, W. N., Depression of rat testicular 1-hyroxylase and 17,20-lyase after administration of 2.3.7.8- tetrachlorodibenzo-p-dioxin (TCDD), B iochem . P harm a c o l., 36. 727, 1987. 202. Okey, A. B., Enzyme induction in the cytochrome P-450 system, P harm acol. T her., 45. 241, 1990. 203. Kumaki, K., Jensen, N. M., Shire, J . G., and Nebert, D. W., Genetic differences in induction of cytosol reduced NAD(P): men adione oxidoreductase and microsomal aryl hydrocarbon hydroxylase in the mouse, J . B iol. C hem ., 252, 157, 1977. 204. Beatty, P. W. and Neal, R. A., Induction of DT-diaphorase by 2.3.7.8- tetrachlorodibenzo-p-dioxin (TCDD), Biochem . Biophys. Res. C om m un., 68, 197, 1976. 205. Yoshihara, S., Nagata, K., Yoshimura, H., Kuroki, H., and Masuda, Y., Inductive effect on hepatic enzymes and acute toxicity of individual polychlorinated dibenzofuran congeners in rats, Toxicol. A p p l. P ha rm a co l., 59, 580, 1981. 206. Kirsch, R., Fleischner, G., Kamisaka, K., and Arias, I. M., Structural and functional studies of ligandin, a major renal organic anion-binding protein, J . Clin. Invest., 55, 1009, 1075. 207. Ahotupa, M., Enhancement of drug-metabolizing enzyme activi ties by pure PCB congeners, B iochem . P harm acol., 30, 1866, 1981. 208. Laraartlnlere, C. A., Duringer, C. S., and Luder, G. W., Altered ontogeny of glutathione S-transferase by 2,4,5,2',4',5'-hexachlorobiphenyl, Toxicol. A p p l. P ha rm a co l., 51, 233, 1979. 209. Kohl), K. K,, Mukhtar, H., Bend, J. R., Albro, P. W., and McKinney, J. D., Biochemical effects of pure isomers of hexachlorobiphenyl-hepatic microsomal epoxide hydrase and cytosolic gluthathione S-transferase activities in the rat, Biochem . P harm acol.. 28. 1444, 1979. 210. Owens, I. S., Genetic regulation of UDP-glucuronosyltransferase induction by polycyclic aromatic compounds in mice: co-segregation with aryl hydrocarbon (benzo[a]pyrene) hydroxylase induction, J. B iol. C hem .. 252, 2827, 1977. 211. Thunberg, T., Ahlborg, U. G ., and Wahlstrom B., Comparison between the effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin and six other compounds on vitamin A storage, the UDP-glucuronosyltransferase and the aryl hydrocarbon hydroxylase activity in the rat liver. A rch . T oxicol.. 55, 16, 1984, 212. Goldstein, J. A,, McKinney, J. D,, L uder, G. YV,, Hickman, P,, Bergman, H., and Moore, J. A., Toxicological assessment of hexachlorobiphenyl isomers and 2,3,7.8-tetrachlorodibenzofuran in chicks, n. Effects on drug metabolism and porphyrin accumulation, Toxicol. A p p l. P harm acol.. 36, 81, 1976. 213. Puhvel, S. M., Ertl, D. C-, and Lynbert, C. A., Increased epi dermal transglutaminase activity following 2,3,7,8-tetrachlorodibenzo-p-dioxin: in vivo and in vitro studies in mouse skin, Toxicol. A p p l. P harm acol.. 73, 42, 1984. 214. Hempel, J., Harper, K., and Lindahl, R., Inducible (class 3) aldehyde dehydrogenase from rat hepatocellular carcinoma and 2,3,7,8tetrachlorodibenzo-p-dioxin-treated liver distant relationships to the class l and 2 enzymes from mammalian liver cytosol/mitochondria. B iochem istry. 28, 1160, 1989. 215. Poland, A. and Glover, E., Chlorinated dibenzo-p-dioxin: potent inducers of delta-aminolevulinic acid synthetase and aryl hydrocarbon hydroxylase. II. A study of the structure-activity relationship, M ol. P harm acol.. 9, 736, 1973. 216. DeVerneuil, H., Sassa, S., and Kappas, A., Effects of polychlor inated compounds, 2,3,7,8-tetrachlorodibenzo-p-dioxin, phenobar bital and iron on hepatic uroporphyrinogen decarboxylase, B iochem . J .. 214. 145, 1983. 217. Sinclair, P. R., Bernent, W. J ., Bonkovsky, H. L., Lambrecht, R. W., Frezza, J. E., Sinclair, J . F., Urquhart, A. J., and Elder, G. H., Uroporphyrin accumulation produced by halogenated bi phenyls in chick-embryo hepatocytes, B iochem . J ., 237, 63, 1986. 218. Jones, K. G, and Sweeney, G. D., Association between induction of aryl hydrocarbon hydroxylase and depression of uroporphyrinogen decarboxylase activity. R es. C om m un. C hem . Pathol. P harm acol.. 17, 631, 1977. 219. Sinclair, P. R., Bernent, W. J ., Bonkovsky, H. L., and Sinclair, J. R,, Inhibition of uroporphyrinogen decarboxylase by halogenated biphenyls in chick hepatocyte cultures: essential role for induction of cytochrome P-448, B iochem . J ., 222, 737, 1984. 220. Sassa, S., DeVerneuil, H., and Kappas, A., Inhibition of uropor phyrinogen decarboxylase activity in polyhalogenated aromatic hy drocarbon poisoning, in B io lo g ica l M echanism o f D ioxin A ction, Poland, A., and Kimbrough, R. D.. Eds., Banbury Report, No. 18, Cold Spring Harbor Press, Cold Spring Harbor, NY, 1984, 215. 221. Yao, C. and Safe, S., 2,3,7,8-Tetrachlorodibenzo-p-dioxin-induced porphyria in genetically inbred mice: partial antagonism and mech anistic studies, Toxicol. A p p l. P h a rm a co l., 100, 208, 1989. 222. Madhukar, B. V., Brewster, D.W., and Matsumura, F., Effects of in vivo administered 2,3,7,8-tetrachlorodibenzo-p-dioxin on re ceptor binding of epidermal growth factor in the hepatic plasma membrane of rat, guinea pig, mouse and hamster, P roc. N a tl. A cad. Sci. U .S A ., 81, 7407, 1984. 19429 1990 83 Critical Reviews In 223. Hudson, L. G., Toscano, W. A., Jr., and Greenlee, W. F., Reg ulation of epidermal growth factor binding in an human keratinocyte cell line by 2,3,7,8-tetrachlorodibenzo-p-dioxin, Toxicol. A ppl. P harm acol.. 77, 251, 1985. 224. Romkes, M. and Safe, S., Comparative activities of 2,3,7,8-tetrachlorodibenzo-p-dioxin and progesterone as antiestrogens in the famale rat uterus, Toxicol. A p p l. P harm acol., 92, 368. 1988. 225. Romkes, M., Plskorska-Ptlszczynska, J., and Safe, S., Effects of 2.3.7.8- tetrachlorodibenzo-p-dioxin on hepatic and uterine estrogen receptor levels in rats, T oxicol. A p p l. P harm acol., 87, 306, 1987. 226. Sloop, T. C. and Luder, G. W., Dose-dependent elevation of Ah receptor binding by TCDD in rat liver, Toxicol. A ppl. P harm acol., 88. 329, 1987. 227. Denomme, M. A., Lcece, B., LI, A., Towner, R., and Safe, S-, Elevation of 2,3,7,8-TCDD rat hepatic reeceptor levels by polychlor inated biphenyls: structure-activity relationships, Biochem . P har m a co l., 35, 277, 1986. 228. Sunahara, G. I., Luder, G. W., McCoy, Z., Bresnlck, E. H., Sanchez, E. R., and Nelson, K. G., Characterization of 2,3.7.8tetrachlorodibenzo-p-dioxin-mediated decreases in dexamethasone binding to rat hepatic cytosolic glucocorticoid receptor, M ol. P har m acol.. 36, 239, 1989. 229. Ryan, R. P., Sunahara, G. I., Luder, G. W., Birnbaum, L. S., and Nelson, K. G., Decreased ligand binding to the hepatic glu cocorticoid and epidermal growth factor receptors after 2,3,4,7,8pentachlorodibenzofuran and 1,2,3,4,7,8-hexachlorodibenzofuran treatment of pregnant mice, Toxicol. A p p l. P harm acol., 98, 454, 1989. 230. Bastomsky, C. H., Enhanced thyroxine metabolism and high uptake goiters in rats after a single dose of 2,3,7,8-tetrachlorodibenzo-pdioxin. E ndocrinology, 101, 292, 1977. 231. KelUng, C. K-, Menahan, L. A., and Peterson, R. E., Hepatic indices of thyroid status in rats treated with 2,3,7,8-tetrachlorodibenzo-p-dioxin, B iochem . P ha rm a co l., 36, 283, 1987. 232. Rozraan, K., Rozman, T., Scheufler, E., Pazdemik, T., and Greim, H., Thyroid hormones modulate the toxicity of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD), J. Toxicol. Environ. H ealth. 16. 481, 1985. 233. Pazdemik, T. L. and Roman, K., Effect of thyroidectomy and thyroxine on 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced immunotoxicity. Life S ci., 36, 695, 1985. 234. Henry, E. C. and Gasiewicz, T. A., Changes in thyroid hormones and thyroxine glucuronidation in hamsters compared with rats fol lowing treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin, Toxicol. A ppl. P harm acol.. 89, 165, 1987. 235. Gupta, B. W., McConnell, E. E., Goldstein, J. A., Harris, M. W., and Moore, J. A., Effects of a polybrominated biphenyl in the rat and mouse. I. Six-month exposure, Toxicol. A ppl. P har m acol.. 68, 1, 1983. 236. Van den Berg, K. J ., Zurcher, C., and Brouwer, A., Effects of 3,4,3',4'-tetrachlorobiphenyl on thyroid function and histology in marmoset monkeys, Toxicol. L ett., 41, 77, 1988. 237. Thunberg, T., Ahlborg, U. G-, and Johnson, H., Vitamin A (retinol) status in the rat after a single oral dose of 2,3,7.8-tetrachlorodibenzo-p-dioxin. A rch . T oxicol., 42, 265, 1979. 238. Brouwer, A., Hakansson, H., Kukler, A., Van den Berg, K. J., and Ahlborg, U. G., Marked alterations in retinoid homeostasis of Sprague-Dawley rats induced by a single i.p. dose of 10 p.g/kg of 2.3.7.8- tetrachlorodibenzo-p-dioxin, Toxicology, 58, 267, 1989. 239. Brouwer, A., Van den Berg, K. J., and Kukler, A., Time and dose responses of the reduction in retinoid concentrations in C57BL/Rij and DBA/2 mice induced by 3,4,3',4,-tetrachlorobiphenyl, Toxicol. A ppl. P harm acol, 78, 180, 1985. 240. Cecil, H. C., Harris, S. J., Bitman, J., and Fries, G. F., Poly chlorinated biphenyl-induced decrease in liver vitamin A in Japanese quail and rats. B ull. E nviron. C ontam . T o xico l., 9, 179, 1973. 241. Brouwer, A. and Van den Berg, K. J., Early and differential decrease in natural retinoid levels in C57BL/Rij and DBA/2 mice by 3,4,3',4'-tetrachlorobiphenyI, T oxicol. A p p l. P h a rm a co l., 73, 204, 1984. 242. Azais, V., Arand, M., Rauch, P., Schramm, H., Bellenand, P., Narbotuie, J. F., Oesch, F., Pascal, G-, and Robertson, L. W., A time-course investigation of vitamin A levels and drug metabolizing enzyme activities in rats following a single treatment with protypic polychlorinated biphenyls and DDT, T oxicology. 44, 341, 1987. 243. Poland, A. and Glover, E ,, Comparison of 2,3,7,8-tetrachlorodi benzo-p-dioxin, a potent inducer of aryl hydrocarbon hydroxylase, with 3-methylcholanthrene, M ol. P h a rm a co l., 10, 349, 1974. 244. Poland, A. and Glover, E., Genetic expression of aryl hydrocarbon hydroxylase by 2,3,7.8-tetrachlorodibenzo-p-dioxin: evidence for a receptor mutation in genetically non-responsive mice. M ol. P har m a co l.. 11, 389, 1975. 245. Poland, A. P., Glover, E-, Robinson, J . R-, and Nebert, D. W., Genetic expression of aryl hydrocarbon hydroxylase activity: induc tion of monooxygenase activities and cytochrome P,-450 formation by 2,3,7,8-tetrachlorodibenzo-p-dioxin in mice genetically "nonresponsive" to other aromatic hydrocarbons, J. B iol. C hem ., 249,5599, 1974. 246. Thomas, P. E., Kouri, R. E., and Hutton, J. J ., The genetics of aryl hydrocarbon hydroxylase induction in mice: a single gene dif ference between C57BL6J and DBA2J, B iochem . G en et., 6, 157, 1972. 247. Okey, A. B-, Vella, L. M., and H arper, P. A., Detection and characterization of a low affinity form of cytosolic Ah receptor in livers of mice nonresponsive to the induction of cytochrome Pi-450 by 3-methylcholanthrene, M ol. P h a rm a co l., 35. 823, 1989. 248. Poland, A. and Glover, E-, 2,3,7,8-Tetrachlorodibenzo-p-dioxin: segregation of toxicity with the Ah locus, M o l. P h a rm a co l., 17, 86, 1986. 249. Vecchi, A., Sironi, M-, Canegrati, M. A., Recchia, M., and Garattini, S., Immunosuppressive effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin in strains of mice with different susceptibility to in duction of aryl hydrocarbon hydroxylase, Toxicol. A p p l. P harm acol.. 68, 434, 1983. 250. Nagarkatti, P. S., Sweeney, G. D., Gauldle, J., and Clark, D. A., Sensitivity to suppression of cytotoxic T cell generation by 2.3.7.8- tetrachlorodibenzo-p-dioxin (TCDD) is dependent on the Ah genotype of the murine host, Toxicol. A p p l. P h a rm a co l., 72, 169, 1984. 251. Greig, J. B., Francis, J. E., Kay, S. J., Lovell D. P., and Smith, A.G., Incomplete correlation of 2,3,7,8-tetrachlorodibenzo-p-dioxin hepatotoxicity with Ah phenotype in mice, T oxicol. A p p l. P h a rm a co l.. 74, 17, 1984. 252. Jones, K. G. and Sweeney, G. D., Dependence of the prophyrogenic effect of 2,3,7,8-tetracholorodibenzo-p-dioxin upon inheritance of aryl hydrocarbon hydroxylase responsiveness, T oxicol. A p p l. P har m a co l., 53, 42, 1980. 253. Knutson, J. C. and Poland, A., Response of murine epidermis to 2.3.7.8- tetrachlorodibenzo-p-dioxin: interaction of Ah and hr loci. C eil. 30, 225, 1982. 254. Gasiewicz, T. A., Olson, J. R., Geiger, L. H., and Neal, R. A., Absorption, distribution and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in experimental animals, in H u m a n and E n vironmental Risks o f Chlorinated D ioxins and Related Compounds, Tucker, R. E., Young, A. L., and Gray, A. P., Eds., Plenum Press, New York, 1983, 495. 4 Volume 21, Issue 1 Toxicology Q ^pm an, D. E. and Schiller, C. M., Dose-related effects of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) in C57BU6J and DBA/2J mice, Toxicol. A p p l. P harm acol.. 78. 147, 1985. 236 Clark. D. A., Sweeney, G,, Safe, S., Hancock, E., Kilbum, D. G., and Gualdie, J., Cellular and genetic basis for suppression of cytotoxic T cell generation by haloaromatic hydrocarbons, lm m unopharm acology. 6, 143, 1983. 237. Luster, M. I., Hong, L. H., Boorman, G. A., Clark, G., Hayes, H. T ., Greenlee, W. F., Dold, K., and Tucker, A. N., Acute myelotoxic responses in mice exposed to 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD). Toxicol. A ppl. P h a rm a co l., 81, 156, 1985. 258. SUkworth, J. B., Antrim, L., and Kaminsky, L. S., Correlations between polychlorinated biphenyl immunotoxicity, the aromatic hyrocarbon locus and liver microsomal enzyme induction in C57BL/6J and DBA/2 mice, Toxicol. A ppl. P h a rm a co l., 75, 156, 1984. 259. Robertson, L. W., Parkinson, A., Bandiera, S., Lambert, I., Merrill, J ., and Safe, S., PCBs and PBBs: biologic and toxic effects on C57BL/6J and DBA/2J mice, Toxicology. 31, 191, 1984. 260. Poland, A., Glover E., and Kende, A. S., Sterospecific, high affinity binding of 2.3.7,8-tetrachlorodibenzo-p-dioxin by hepatic cy tosol: evidence that the binding species is the receptor for induction of aryl hydrocarbon hydroxylase, J . B iol. C hem ., 251, 4936, 1976. 261. Safe, S. H., The aryl hydrocarbon receptor, ISI A tla s Scl. P har m a co l.. 2. 78, 1988. 262. Gasiewicz, T. A. and Rucci, G., Cytosolic receptor for 2,3,7,8tetrachlorodibenzo-p-dioxin: evidence for a homologous nature among various mammalian species, M ol. P harm acol., 26, 90, 1984. 263. Denison, M. S., Wilkinson, C. F., and Okey, A. B., Ah receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin: comparative studies in mammalian and non-mammalian species, C hem osphere, 15, 1672, 1986. 264. Farrell, K., Safe, L., and Safe, S., Synthesis and aryl hydrocarbon receptor binding properties of radiolabeled polychlorinated dibenzofuran congeners. A rch. B iochem . B io p h y s., 259, 185. 1987. 265. Okey, A. B. and Vella, L. M., Binding of 3-methylcholanthrene and 2,3,7,8-tetrachlorodibenzo-p-dioxin to a common Ah receptor site in mouse and rat hepatic cytosols, E ur. J . B iochem ., 127, 39, 1982. 266. Okey, A. B., Dube, A. W., and Vella, L. M., Binding of benzo[a]pyrene and dibenz[a,h]anthracene to the Ah receptor in mouse and rat hepatic cytosols. C ancer R es.. 44, 1426, 1984. 267. Okey, A. B., Vella, L. M,, and Iverson, F., Ah receptor in primate liver binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin and carcino genic polycyclic aromatic hydrocarbons. C an. J . P hysiol. P harm a c o l.. 62, 1292, 1984. 268. Manchester, C. K., Gordon, S. K., Golas, C. L., Roberts, E. A., and Okey, A. B., Ah receptor in human placenta: stabilization by molybdate and characterization of binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin, 3-methylcholanthrene and benzo[a]pyrene. C an c e r R e s.. 47, 4861, 1987. 269. Bradfield, C. A., Kende, A. S., and Poland, A., Kinetic and equilibrium studies of Ah receptor-ligand binding: use of [l23I]-2iodo-7,8-dibromodibenzo-p-dioxin. M ol. Pharm acol.. 34, 229, 1988. 270. Bunce, N. J., Landers, J. P., and Safe, S., Kinetic models for association of 2,3,7,8-tetrachlorodibenzo-p-dioxin with the Ah re ceptor, A rch . B iochem . B iophys.. 267, 384, 1988. 271. Israel, D. I. and Whitlock, J. P., J r ., Induction of mRNA specific for cytochrome P:-450 in wild type and variant mouse hepatoma cells, J . B iol. C hem ., 258, 10390. 1983. 272. Israel, D. I. and Whitlock, J. P., Jr., Regulation of cytochrome Pi-450 gene transcription by 2,3,7,8-tetrachlorodibenzo-p-dioxin in wild type and variant mouse hepatoma cells, J . B iol. C hem ., 259, 5400, 1984. 273. Israel, D. I., Estolano, M. G., Galeazzi, D. R., and Whitlock, J. P., J r., Superinduction of cytochrome P,^t50 gene transcription by inhibition of protein synthesis in wild type and variant mouse hepatoma cells, J . B io l. C h e m ., 260, 5648, 1985. 274. van Gurp, J. R. and Hankinson, O., Isolation and characterization of revertants from four different classes of aryl hydrocarbon hydrox ylase-deficient Hepa-1 mutants. M ol. C ell. B iol.. 4, 1597 1984. 275. Hankinson, O., Anderson, R. D., Blrren, B. W., Sander, F., Negishl, M., and Nebert, D. W ., Mutations affecting the regulation of transcription of the cytochrome Pt-450 gene in the mouse Hepa1 cell line, J . B iol. C h e m .. 260, 1790, 1985. 276. Harris, M., Piskorska-Pllszczynska, J., Zacharewski, T., Romkes, M., and Safe, S., Structure-dependent induction of aryl hydrocarbon hydroxylase in human breast cancer cell lines and characterization of the Ah receptor. C a n c er R e s., 49, 4531, 1989. 277. Poland, A., Palen, D., and Glover, E., Tumor promotion by TCDD in skin of HRS/J hairless mice, N ature, 300, 271, 1982. 278. Poland, A., Knutson, J ., Glover, E., and Kende, A., Tumor promotion in the skin of hairless mice by halogenated aromatic hy drocarbons, in G enes a n d P ro tein s in O ncogenesis. Weinstein, I. B. and Vogel, H. J., Eds., Academic Press, New York, 1983, 143. 279. Tukey, R. H., Hannah, R. R., Negishl, M., Nebert, D. W., and Elsen, H. J., The Ah locus: correlation of intranuclear appearance of inducer-receptor complex with induction of cytochrome P-450 mRNA, C ell. 31, 275, 1982. 280. Zacharewski, T., Harris, M ., and Safe, S., Induction of cyto chrome P-450 dependent monooxygenase activities in rat hepatoma H-4QE cells in culture by 2,3,7,8-tetrachlorodibenzo-p-dioxin and related compounds: mechanistic studies using radiolabeled conge ners, A rch . B iochem . B io p h y s., 272, 344, 1989. 281. Gonzalez, F. J. and Nebert, D.W., P-450 genes: structure evolution and regulation, A nna. R ev. B io c h em .. 56, 945, 1987. 282. Gonzalez, F. J. and Nebert, D. W., Autoregulaton plus upstream positive and negative control regions associated with transcriptional activation of the mouse cytochrome Pi-450 gene. N ucleic A cid s R es., 13, 7269. 1985. 283. Jones, P. B., Galeazzi, D. R., Fisher, J. M., and Whitlock, J. P., J r ,, Control of cytochrome Pi-450 gene expression by dioxin, Science. 227, 1499, 1985. 284. Sogawa, K., Fujisawa-Sehara, A., Yamane, M., and Fujii-Kuriyama, Y., Location of regulatory elements responsible for drug induction in the rat cytochrome P-450c gene. P roc. N atl. A cad. Sci. U .S .A ., 83. 8044, 1986. 285. Bandiera, S., Safe, S., and Okey, A. B., Binding of polychlorinated biphenyls classified as either phenobarbitone-, 3-methylcholanthreneor mixed-type inducers to cytosolic Ah receptor, Chem . B iol. In ter act., 39, 259, 1982. 286. Bandiera, S., Sawyer, T., Romkes, M ., Zmudzka, B., Safe, L., Mason, G., Keys, B., and Safe, S., Polychlorinated dibenzofurans (PCDFs): effects of structue on binding to the 2,3,7,8-TCDD cyto solic receptor protein, AHH induction and toxicity, Toxicology, 32, 131, 1984. 287. Mason, G., Sawyer, T., Keys, B., Bandiera, S., Romkes, M., Piskorska-Pliszczynska, J ., Zmudzka, B., and Safe, S., Poly chlorinated dibenzofurans (PCDFs): correlation between in vivo and in vitro structure-activity relationships. Toxicology, 37, 1. 1985. 288. Mason, G., Farrell, K., Keys, B., Piskorska-Pliszczynska, J., Safe, L., and Safe, S., Ploychlorinated dibenzo-p-dioxins: quanti tative in vitro and in vivo structure-activity relationships. Toxicology, 41, 21, 1986. 289. Mason, G., Zacharewski, T., Denomme, M. A., Safe, L., and Safe, S., Polybrominated dibenzo-p-dioxins and related compounds: 19431 85 Critical Reviews In q u antitative in vivo and in vitro stru ctu re-activ ity relationships. T o x icology. 44. 245, 1987. 290. Safe, S ., B andiera, S ., Saw yer, T ., Z m udzka, B ., M ason, G ., Rom kes, M ., D enom m e, M . A ., S parling, J ., O key, A. B ., and F u jita , T ., Effects o f structure on binding to the 2,3,7.8-TC D D receptor protein and AHH induction- halogenated biphenyls. Environ. Health Perspect.. 61. 21, 1985. 291. K ociba, R. J ., Keyes, D. G ., Beyer, J . E ., C arreon, R. M ., W ade, C . E ., O ittenber, D . A ., K alnins, R . P ., F ranson, L . E ., P ark , C . N ., B ernard, S. D ., H um m el, R . A ., and H um iston, C . G ., Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodienzo-p-dioxin in rats, Toxicol. A ppl. Pharm a col.. 46. 279. 1978. "' 292. M urray, F. J ., Sm ith, F. A ., N itschke, K . 0 ., H um iston, C . G ., K ociba, R . J ., an d Schw etz, B . A ., Three generation reproduction study o f rats given 2 .3 .7 ,8 -tetrach lo ro d ib en zo -p -d io x in (TC D D ) in the diet, Toxicol. A ppl. Pharm acol., 50, 241, 1979. 293. National Toxicology Program , Carcinogenesis bioassay o f 2,3,7,8- tetrachlorodibenzo-p-dioxin (CA S N o. 1746-01-6) in Osbom e-M en- del rats and B6C3F1 mice (gavage study), DHHS Publication No. (NIH) 82-1765, National Technical Inform ation Service, Springfield, VA. 1982. 294. National Toxicology Program , Carcinogenesis bioassay of 2 3,7,8- tetrachlorodibenzo-p-dioxin (CAS No. 1746-01-6) in Swiss-W ebster mice (dermal study), DHHS Publication No. (NIH) 82-1757, Na tional Technical Information Service, Springfield, VA, 1982. 295. D avis, D. and Safe, S ., Im m unosuppressive activities o f polychlor inated dibenzofuran congeners: quantitative structure activity rela tionships and interactive effects, Toxicol. A ppl. Pharm acol.. 9 4 ,1 4 1 , 1988. 296. K ervllet, N . S ., B rau n er, J . A ., and M atlock, J . P ., Humoral immunotoxicity of polychlorinated diphenyl ethers, phenoxphenols, dioxins and furans present as contam inants o f technical grade pen- tachlorophenol. Toxicology, 36, 307, 1985. 297. B irnbaum , L. S ., H arris, M . W ., C raw ford, D. D ., and M or rissey, R . E ., Teratogenic effects o f polychlorinated dibenzofurans in com bination in C57B L/6n m ice, Toxicol. A ppl. Pharm acol., 91, 246. 1987. 298. H olcom b, M ., Y ao, C ., an d S afe, S ., The biologic and toxic effects o f polychlorinated dibenzo-p-dioxins and dibenzofurans in the guinea pig: quantitative structure-activity relationships, Biochem. Pharm a col.. 37, 1535, 1988. 299. K n u tso n , J . C . an d P o lan d , A ., K eratinization o f mouse teratoma ceil line XB produced by 2 ,3 ,7 ,8 -tetrach lo ro d ib en zo -p-dioxin: an in vitro model of toxicity, Cell, 22, 27, 1980. 300. B r a d la w , J . a n d C a s te rlin e , J . L ., Induction o f enzym es in cell cultures: a rapid screen for the detection o f planar polychlorinated organic compounds, J. Assoc. O ff. A nal. Chem ., 62, 904, 1979. 3 01. B e ilin , J . S . a n d B a rn e s , D. G . , Interim procedures for estim ating risks associated with exposures to m ixtures o f chlorinated dibenzo- p-dioxins and dibenzofurans (C D D s and CD Fs) and 1989 update. R isk Assessm ent Forum, U .S .-E .P .A ., 1989. 302. A hlborg, U. G ., H akansson, H ., W aern, F ., and Hanberg, A ,, N ordisk dioxinriskbedom ning. R ap p o rt firan en n o tdisk expengrupp. N ordiska M inisterradet, M iljorapport, 1988. 303. NATO/CCM S, North Atlantic Treaty Organization (NATOVcommittee on the challenges o f m odem society (CCM S), Report Number 176, August 1988. 304. C o u tu re , L . A ., Elw eli, M . R ., a n d B irn b a u m , L . S ., Dioxin-like effects observed in m ale rats follow ing exposure to octachlorodi- benzo-p-dioxin (O CD D ) during a 13 week study, Toxicol. A ppl. Pharm acol., 93, 31, 1988. 305. Pluess, N ., Polger, H ., H ohbach, C ., S uter, M ., and Schlatter, C ,, Subchronic toxicity of 2,3.7,8-pentachlorodibenzofuran(PeCDF) in rats, C hem osphere. 17, 1099, 1988. 306. Pluess, N ., Polger, H ., H ohback, C ., Suter, M ., and S chlatter, C . , Subchronic toxicity of some chlorinated dibenzofurans (PCDFs) and a mixture of PCDFs and chlorinated dibenzodioxins (PCDDs) in rats. C hem osphere, 17, 937, 1988. 307. Safe, S ,, D avis, D ., R om kes, M ., Yao, C ., Keyes, B ,, Piskorska- Pllszczynska, J ., F arrell, K ., M ason, G ,, Denom m e, M . A ., Safe, L ., Z m u d z k a, B ., and H olcom b, M ., Development and validation o f in vitro bioassay for 2 ,3 ,7 ,8-TCDD equivalents, C hem osphere, 19, 853, 1989. 308. Safe, S ., M ason, G ., Saw yer, T ., Zaharew ski, T ., H arris, M ., Y ao, C ., K eyes, B ., F arrell, K ., Holcom b, M ., D avis, D ,, Safe, L ., P isk o rsk a -P lisz c z y n sk a , J . , Leece, B ., D en o m m e, M . A ., H u tz in g e r, O., T h o m a , H ., C h ittim , B ., a n d M a d g e , J . , Development and validation of in vitro induction assays for toxic halogenated mixtures, Toxicol. Irui. H ealth, 5, 757, 1989. 309. Safe, S., Determination of the 2,3,7,8-TCDD toxic equivalent fac tors: support for the use of the in vitro AHH inductions assay, C h e m osphere, 16, 791, 1987. 310. Safe, S ., M ason, G ., F arrell, K ., Keys, B ., Piskorska-Pliszczyn s k a , J . , M a d g e , J . A ., a n d C h ittim , B ., Validation of in vitro bioassays for 2,3,7,8-T C D D equivalents, C hem osphere. 16, 1723, 1987. 311. Z acharew ski, T ., H arris, M ., Safe, S., T hom a, H ., an d H utzin- g e r, O ., Applications of the in vitro aryl hydrocarbon hydroxylase induction assay for determining "2,3,7,8-T C D D equivalents": pyrolyzed brominated flame retardants. Toxicology, 51, 177, 1988. 312. P in k erto n , M . N ., K ociba, R . J ,, Petrella, R. V ., M cA llister, D . L ., W illis, M . L ., Fulfs, J . C ., T hom a, H ., and H utziner, O ., A preliminary report on the investigation of the comparative toxicity of combustion products of high impact polystyrene with and without decabromodiphenyl-oxide/antimony trioxide as a flame retardant us ing 2,3,7,8-tetrabromodibenzo-p-dioxin and 2,3,7,8-tetrabromodibenzofuran as positive controls, Chem osphere, 18, 1243, 1989. 313. G o ld ste in , J . A., The structure-activity relationship of halogenated biphenyls as enzyme inducers, Ann. <V.Y. A cad. S ci., 320, 164, 1979. 314. Y o s h im u ra , H ., Y o s h ih a ra , S ., O z a w a , N ., a n d M ik i, M ., Pos sible correlation between induction modes of hepatic enzymes by PCBs and their toxicity in rats, Ann. N .Y . A cad. S c i.. 320, 179, 1979. 315. P o la n d , A. a n d Glover, E., Chlorinated biphenyl induction of aryl hydrocarbon hydroxylase activity: a study of the structure activity relationship, M o l. P harm acol.. 13. 924, 1977. 316. G o ld s te in , J . A ., H ic k m a n , P ., B e rg m a n , H ,, M c K in n e y , J . D ., a n d W a lk e r, M . P ., Separation of pure polychlorinated biphenyl isomers into two types of inducers on the basis of induction of cy tochrome P-450 or P-448, Chem . B iot. Interact., 17, 69, 1977. 317. S a w y e r, T . a n d S afe, S ., PCB isomers and congeners: induction of aryl hydrocarbon hydroxylase and ethoxyresorufin O-deeihytase ac tivities in rat hepatoma cells, Toxicol. Lett., 18, 87, 1982. 318. Parkinson, A., Cockerline, R ., and Safe, S., Induction of both 3methylcholanthrene- and phenobarbitone-type microsomal enzyme activity by a single polychlorinated biphenyl isomer, B iochem . P h a r m acol., 29, 259, 1980. 319. P a rk in s o n , A ., R o b e rts o n , L ., S a fe , L ., and S afe, S ., Polychlor inated biphenyls as inducers of hepatic microsomal enzymes: struc ture-activity rules, Chem . B iol. Interact., 30, 271, 1981. 320. S a fe , S ., R o b e rts o n , L . W ., S a fe , L ., P a rk in s o n , A ., B a n d ie ra , S ., S a w y e r, T ., a n d C a m p b e ll, M . A ., Halogenated biphenyls: molecular toxicology. Can. J. Physiol. P harm acol., 60, 1057, 1982. 321. P a rk in s o n , A., C o c k erlin e, R ., a n d S afe, S., Polychlorinated bi- 19432 36 Volume 21, Issue 1  -V -- Toxicology v'* . _vl som ers and congeners as inducers o f both 3-m ethylcholanh e and phenobarbitone-type microsomal enzym e activilty, Chem. gtol. Interact.. 29, 277, 1980. Parkinson, A ., R o b e rtso n , L ,, S a fe , L ., a n d S afe, S ., P olychlor inated biphenyls as inducers o f hepatic microsomal enzym es: effects f diortho substilion, C hem . B iol. Interact.. 3 1, 1, 1981. te e , B ,, Denomme, M . A ., T ow ner, R ,, Li, S. M . A ., and S afe S ., Polychlorinated biphenyls: correlation between in vivo and vitro quantitative structure-activity relationships (Q S A R s), J . T ox icol. Environ. Health. 16, 379, 1985. Nlkolaidls, . , Im m unotoxicity o f 2 ,3 .7 ,8-Tetrachlorodibenzo-pDioxin and Related Polychlorinated A rom atic H ydrocarbons in Birds, Ph.D. thesis, University o f Uppsala, Sw eden, 1989. B ru n s tro m , B., T oxicity o f co p ian ar polychlorinated biphenyls in avian embryos, Chemosphere. 19, 756, 1989. Y ao, C . C .,T h e Interactions o f 2 ,3 ,7 ,8-TCDD and M C D F inTC D D Induced Porphyria, Ph.D. thesis, Texas A&M University, College Station, 1989. j 27 R odm an, L . E ., Shedlofsky, S. L , Sw im , A. T ., and R obertson, w ., Effects of polychlorinated biphenyls as cytochrom e P-450 induction in the chick em bryo h epatocyte cu lture, A rc h . B iochem . Biophys.. 275, 252. 1989. 328 M a rk s , T . A ,, K im m el, G . A ., a n d S ta p le s , R . E ., Influence of symmetrical polychlorinated biphenyl isomers on em bryo and fetal developm ent in m ice, T oxicol. A p p t. P h a rm a col., 61, 269, 1981. 329. S afe, S ., M etabolism , uptake, storage and bioaccum ulation, in H alogenated Biphenyls, Terphenyls, Naphthalenes, D ibenzodioxins and Related P roducts, K im brough, R . D ., E d ., E Isevier/N orth-H olland, A msterdam, 1980, 81. 330. C levenger, M . A ., R oberts, S. M ., L attin , D. L ., H arb iso n , R . D ., and Jam es, R . C ., The pharmacokinetics of 2 ,2 ',5 ,5 '-tetrachlorobiphenyl and 3,3 ',4 ,4 '-tetrach lo ro b ip h en y l and its relation ship to toxicity, Toxicol. A ppl. P harm acol.. 100, 315, 1989. 331. K an n an , N-, T anabe, S ., O no, M ., and T atsu k aw a, R ., Critical evaluation o f polychlorinated biphenyl toxicity in terrestrial and m a rine mammals: increasing impact of non-ortho and mono-ortho co pianar polychlorinated biphenyls from land to ocean. A rch. Environ. Contam . Toxicol.. 18, 850, 1989. 332. D avis, D ., Immunotoxicity o f the Halogenated Aromatic Hydrocar bons -- OSARs and Interactive Effects, Ph.D . thesis, Texas A&M University, College Station, 1989. 333. S ilkw orth, J . B. and G rab stein , E . M ., Polychlorinated biphenyl im m unotoxicity: dependence on isom er planarity and Ah gene com plex, Toxicol. Appl. Pharm acol.. 65, 109, 1982. 334. S to n a rd , M . D. and G rieg, J . B ., Different patterns of hepatic microsomal enzyme activity produced by adm inistration o f pure hexachlorobiphenyi isomers and hexachlorobenzenes, Chem . Biol. In teract., 15, 365. 1976. 335. E cobichon, D . J ., H ansell, M . M ., an d Safe, S ., Isomerically pure brom obiphenyl congeners and h epatic m onooxygenase activities in the rat: influence o f position and degree o f brom ination, Toxicol. Appl. Pharm acol., 47, 341, 1979. 336. M oore, R . W ., Sleight, S. D ., an d A u st, S. D ., Induction o f liver microsomal drug-metabolizing enzym es by 2 ,2 ',4 ,4 ',5 ,5 '-h ex ab ro m obiphenyl, Toxicol. Appl. Pharm acol., 44, 309, 1978. 337. M oore, R . W ., Sleight, S. D ., an d A ust, S. D ., Effects o f 2,2'dibrom obiphenyl and 2 ,2 ',3,4,4',5,5'-heptabrom obiphenyl on liver m icrosomal drug-metabolizing enzym es, Toxicol. A ppl. Pharmacol., 48, 73, 1979. 338. R obertson, L , W ., P arkinson, A ., C am pbell, M . A ., and Safe, S . , Polybrom inated biphenyls as aryl hydrocarbon hydroxylase in ducers: structure-activity correlations, Chem . Biol. Interact., 42, 53, 1982. 339. R obertson, L. W ., A ndres, J . L ., Safe, S ., and Lovering, S. L ., Toxicity o f 3 ,3 ',4 .4 '- and 2,2',5,5'-tetrabrom obiphenyi: correlation o f activity with aryl hydrocarbon hydroxylase induction and lack o f protection by antioxidants, J. Toxicol. Environ. Health. 11,81, 1983. 340. R obertson, L . W ., P ark in so n , A ., an d Safe, S ., Induction o f both cytochrom es P-450 and P-448 by 2,3',4,4',5-pentabrom obiphenyl, a com ponent o f FireM aster*, Biochem . Biophys. Res. C om m on., 92, 175, 1980. 341. M oore, R . W ., D an n an , G . A ., an d A ust, S. D ., Structure-function relationships for the pharmacological and toxicological effects and m etabolism o f polybrom inated biphenyl congeners, in M olecular B a sis o f Environm ental Toxicity, Bhatnagar, R. S ., E d., Ann A rbor Science, Ann A rbor, M I, 1980, chap. 8. 342. R obertson, L . W ., P arkinson, A ., B andiera, S ,, and Safe, S ., Potent induction of rat liver m icrosomal drug-metabolizing enzym es by 2,3,3',4,4',5-hexabrom obiphenyl, a com ponent o f FireM aster*, Chem. Biol. Interact., 35, 13, 1981. 343. D annan, G . A ., M oore, R . W ., Besaw, L . C ., and A ust, S. D ., 2 ,4 ,5 ,3 ',4 ',5 '-H e x ab ro m o b ip h en y l is both a 3-m ethylcholanthreneand phenobarbital-type o f inducer o f microsomal drug-metabolizing enzym es, Biochem . Biophys. Res. Com m un., 85, 51, 1978. 344. D annan, G . A ., G uengerich, F. P ., K am insky, L . S ., and A ust, S . D ., Regulation o f cytochrom e P-450, J. Biol. Chem ., 258, 1282, 1983. 345. D annan, G . A ., Sleight, S. D ., F ra k er, P. J ., K rehbiel, J . D ., an d A ust, S. D ., Liver m icrosomal enzym e induction and toxicity studies with 2,4,5,3',4'-pentabrom obiphenyl, Toxicol. Appl. P har macol., 64, 187, 1982. 346. C arlso n , G . P ., S m ith , E . N ., an d Jo h n so n , K . M ., Induction o f xenobiotic m etabolism o f rat liver by chlorinated diphenyl ether iso mers, D rug Chem. Toxicol., 3, 293, 1980. 347. C hui, Y. C ., H ansell, M . M ., A ddison, R . F ., and L aw , F. C . P ., Effects o f chlorinated diphenyl ethers on the mixed function oxidases and ultrastructure o f rat and trout liver, Toxicol. Appl. Pharm acol., 81, 287, 1985. 348. Iverson, F ., N ew som e, H ., an d H lerlihy, L ., Induction of rat he patic m onooxygenase activity by polychlorinated diphenyl ethers. Food Cosmet. Toxicol., 25, 305, 1987. 349. Howie, L ., D ickerson, R ., D avis, D ., and Safe, S ., Immunosup pressive and m onooxygenase induction activities of polychlorinated diphenyl ether congeners in C 57B L/6N mice: quantitative structureactivilty relationships, Toxicol. A p p l. P harm acol., in press. 350. B lrnbaum , L. S ., W eber, H ., H arris, M . W ., L am b, J . C ., IV , and M cK inney, J . D .,T oxic interaction of specific polychlorinated biphenyls and 2,3,7,8-tetrachlorodibenzo-p-dioxin: increased inci dence o f cleft plate in m ice, T oxicol. A p p l. P h a n m c o l., 77, 2 9 2 , 1985. 351. B radlaw , J . A ., G arthofT, L. L ., H urley, N. E ., and Firestone, D ., Com parative induction o f aryl hydrocarbon hydroxylase activity in vitro by analogues o f dibenzo-p-dioxins. F ood Cosm et. Toxicol., 18, 627, 1980. 352. B andiera, S ., F arrell, K ., M ason, G ., Kelley, M ., Rom kes, M ., B an n ister, R ., an d S afe, S ., Com parative toxicities of the poly chlorinated dibenzofuran (PCDF) and biphenyl (PCB) mixtures which persist in Y usho victim s. C hem osphere, 13, 507, 1984. 353. S aw yer, T . W . an d S afe, S ., In vitro AHH induction by polyciorinated biphenyl and dibenzofuran mixtures: additive effects, C he mosphere. 14, 79, 1985. 354. S ilkw orth, J . B ., C utler, D. S ., A ntrim , L ., H ouston, D ., T um asonis, C ., an d K am insky, L . S.,Teratology of 2,3,7,8-tetrachIorodibenzo-p-dioxin in a com plex environm ental mixture from the L ove C anal, F u ndam . A p p l. T o xico l., 13, 1, 1989. 355. Silkw orth, J . B ., C u tler, D. S ., an d S ack, G ., Immunotoxicity of 19433 1990 87 ------ ~J*tn g n rttM iM m ii iiiiiiliaiffliaiiia^ \ Critical Reviews in 2.3.7.8- tetrachlorodibenzo-p-dioxin in a complex environmental mixture from the Love Canal, F undam . A ppl. Toxicol., 12, 303. 1989. 336. SU kw orth, i . B ., T um asonis, C ., B riggs, R . G ., N arang, A . S ., N arang, R. S ., R ej, R ., Stein, V. B ., M cM artin, D. N ., and K a m in sk y , L . S ., The effects of Love Canal soil extracts on maternal health and fetal development in rats, F undam . A ppl. Toxicol., 7, 471, 1986. 337. E adon, G ., K am insky, L . S ., Silkw orth, J . B ., AJdous, K ,, H llker, D ., O 'K e e fe , P ., S m ith , R., G le rth y , J . , H aw ley, J . , K im , N ,, a n d D e C a p rio , A ., Calculation of 2 ,3 ,7 ,8-TCDD equivalent con centrations of complex environmental contaminant mixtures, E nvi ron. H ea lth P ersp ect., 70, 2 21, 1986. 338. D e C a p rio , A ., M c M a r tin , D ., O 'K eefe, P ., R e j, R ., S ilk w o rth , J . B ., a n d K a m in sk y , L. S ., Subchronic oral toxicity of 2,3,7,8tetrachlorodibenzo-p-dioxin in the guinea pig: comparisons with a PCB-containing transformer fluid pyrolysate, Fundam . A ppl. Toxi co l., 6, 434, 1986. 339. S ilk w o rth , J . B ., M c M a r tin , D ., D e C a p rio , A ., Rej, R., O 'K eefe, P., a n d K a m in sk y , L. S ., Acute toxicity in guinea pigs and rabbits of soot from a polychlorinated biphenyl-containing transformer Are, Toxicol. A ppl. Pharm acol., 65, 425, 1982. 360. Gierthy, J . F ., C r a n e , D ., a n d F re n k e l, G. D., Application of an in vitro keratinization assay to extracts of soot from a fire in PCBcontaining transformer,, F undam . A p p l. Toxicol., 4, 1036, 1984. 361. G ie rth y , J . F . a n d C r a n e , D ., In vitro bioassasy for dioxin-like activity based on alterations in ephithelial cell proliferation and mor phology, F u n d a m . A p p l. T oxicol.. 5, 754, 1985. 362. Keys, B., Piskorska-Pllszczynska, J . , and Safe, S., Polychlori nated dibenzofurans as 2,3,7,8-TCDD anatgonists: in vitro inhibition of monooxygenase induction, Toxicol. L ett., 31, 131, 1986. 363. H a a k e , J . M ., S a fe , S ., M a y u ra , K ,, a n d P h illip s, T . D ,, Aroclor 1254 as an antagonist of the teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin, Toxicol. L ett., 38, 299, 1987. 364. B a n n is te r, R. D ., D av is, D ., Z a c h a re w sk i, T ., T iz a rd , I., a n d S afe, S ., Aroclor 1234 as a 2,3,7,8-tetrachloiodibenzo-p-dioxin an tagonist: effects on enzyme induction and immunotoxicity, Toxicol ogy. 46, 29, 1987. 365. B a n n is te r, R . a n d S afe, S ., The effects of receptor antagonist on the AHH induction activity of 2,3,7,8-TCDD in C57BLV6 and DBA/2 mice, C hem osphere, 16, 1739, 1987. 366. B iegel, L ., H a r r is , M ., D avis, D ., R o s e n g re n , R ., S a fe L ., a n d S afe, S ., 2,2',4,4',5,5'-Hexachlorobiphenyl as a 2,3,7,8-tetrachlorodibenzo-p-dioxin antagonist in C57BL/6J mice, Toxicol. A ppl. P ha rm a co l.. 97, 561, 1989. 367. Biegel, L ., Howie, L ., a n d S afe, S ., Polychlorinated biphenyl (PCB) congeners as 2 ,3 ,7 ,8-TCDD antagonists-teratogenicity studies, C he m osphere, 19, 955, 1989. 368. D avis, D . a n d S afe, S ., Dose-response immunotoxicities o f com mercial polychlorinated biphenyls (PCBs) and their interaction with 2.3.7.8- tetrachlorodibenzo-p-dioxin, Toxicol. L ett., 48, 35, 1989. 369. Noren, K., Changes in levels of organochlorine pesticides, poly chlorinated biphenyls, dibenzo-p-dioxins and dibenzofurans in human milk from Stockholm 1972-- 1985, Chem osphere, 17, 39, 1988. 370. W illia m s, D . T ,, LeBel, G . L ., a n d J u n k in s , E ., OrganohaJogen residues in human adipose autopsy samples from six Ontario munic ipalities, J . A sso c. O ffic. A n a l. C h em ., 71, 410, 1988. 371. Jensen, R. K., S le ig h t, S . D., G o o d m a n , J. I., A u st, S . D., a n d T ro s k o , J. E., Polybrominated biphenyls as promoters in experi mental hepatnctrcinogenesis in rats. C arcinogenesis, 3, 1183, 1982. 372. J e n s e n , R . K . a n d S leig h t, S. D ., Sequential study on the synergistic effects of 2,2',4,4',5,5'-hexabromobiphenyl and 3,3',4,4',5,5',- hexabromobiphenyl on hepatic tum or prom otion. Carcinogenesis, 7, 1771, 1986. 373. B eilin, J . S . an d B arnes, D. G ., Interim procedures for estim ating risks associated with exposures to mixtures of chlorinated dibenzop-dioxins and dibenzofurans (CD D s and CD Fs), Risk Assessm ent Forum. USEPA 625/3-87/012, 1987. 374. Van der H efjden, C . A ., K naap, A. G . A. C ., K ram ers, P. G . N ., an d V an L ogten, M . J ., Evaluation o f the carcinogenicity and m utagenicity of 2 ,3 ,7 ,8-tetrachlorodibenzo-1,4-dioxin (TC D D ): clas sification and no-effect level, Bilthoven, The N etherlands: State In stitute of National Health. Report DOC/LCM 300/292, 1982. 375. T anabe, S ., K annan, N ., W akim oto, T ., and T atsukaw a, R ., M ethod for the determ ination o f three toxic non-ortho chlorine sub stituted coplanar PCBs in environm ental sam ples at part-per-trillion levels, Int. J. Environ. Anal. C hem ., 29, 199, 1987. 376. K annan, N ., T anabe, S ., W akim oto, T ., and T atsukaw a, R ., C oplanar PCBs in A roclor and K anechlor m ix tu res, J . A sso c. O ff. Anal. C hem .. 70, 451. 1987. 377. T an ab e, S ., K an n an , N ,, O no, M ., an d T a tsu k a w a , R ., Toxic threat to marine mammals: increasing toxic potential o f non-ortho and m ono-ortho coplanar PCBs from land to ocean, Chemosphere, 18, 485, 1989. 378. K an n an , N ., T an ab e, S ., an d T atsu k aw a, R ., Toxic potential on non-ortho and m ono-ortho coplanar PCBs in com m ercial PCB prep arations, 2,3,7,8-TC D D toxicity equivalence factor approach. Bull. Environ. Contam. Toxicol., 41, 267, 1988. 379. K an n an , N ., T anabe, S ., and T atsu k aw a, R ., Potentially haz ardous residues o f non-ortho chlorine substituted coplanar PCB s in hum an adipose tissue, Arch. Environ. H ealth, 43, 11, 1988. 380. T anabe, S ., K annan, N ., S ubram an lan , A n ., W atanabe, S ., O no, M ., an d T atsukaw a, R ., Occurrence and distribution o f toxic coplanar PCB s in the biota, Chem osphere, 16, 1965, 1987. 381. T arh an e n , J ., K oistinen, J ., P aasiv irta, J ., V u o rin en , P . J ., Koiv u sa rri, J ., N u u ja, I., K an n an , N ., a n d T a tsu k a w a , R ., Toxic significance o f planar aromatic com pounds in the Baltic ecosystem -- new studies on extremely toxic coplanar PC B s, Chemosphere, 18, 1067, 1989. 382. T anabe, S ., K annan, N ., W akim oto, T ., T atsukaw a, R ., O kam ota, T ., an d M asuda, Y ., Isomer-specific determ ination and toxic evaluation o f potentially hazardous coplanar PCBs dibenzofurans and dioxins in the tissues o f "Y usho" PCB poisoning victim and in the causal oil, Toxicol. E nviron. C hem ., 24, 215, 1989. 383. H arris, H. J ., K ubiak, T. J ., an d T rick , J . A ., M icrocontaminants and reproductive impairment o f Forster's tern on Green Bay, Final report to US Fish and Wildlife Service, UW Sea G rant Institute, W is. Dept, of N at. Res. and Green Bay M etropolitan Sew erage District, 1985, 42. 384. H offm an, D. J ., R attner, B. A ., Silea, L ., D ocherty, D ., and K u b iak , T . J ., Embryotoxicity, tetratogenicity and aryl hydrocarbon hydroxylase activity in F orster's tem s on G reen B ay, Lake M ichigan, Environ. R es., 42, 176, 1987. 385. B ry an , A . W ., Stone, W . B ., and O lafsson, P . G ., Disposition of toxic PC B congeners in snapping turtle eggs: expressed as toxic equivalents o f TCDD, Bull. Environ. Contam . Toxicol., 39, 791, 1987. 386. K un ita, N ., K ashim oto, T ., M iy ata, H ., F u k u sh im a, S ., H ori, S ., an d O b an a, H ., Causal agents of Y usho, A m . J . In d . M edicine. 5, 45, 1984. 387. D ickerson, R . an d Safe, S ., The structure-dependent effects o f heptachlorodibenzofurans in male C 57B L/6 mice: im m unotoxicity and m onooxygenase enzym e induction, F undam . A p p l. T oxicol., in press. 19434 88 Volume 21, Issue 1 SCIENTIFIC SOLUTIONS A NEW REVIEW OF THE DIOXIN LITERATURE IN THE CONTEXT OF COM PENSATION FOR VIETN AM VETERANS By Richard W. Clapp and James R. Olson Ea r l y l a s t y e a r , a t a s k f o r c e o f en v iro n m en ta l scien tists p rod u ced a report (1) u p d atin g and su m m arizin g the scien tific literature on the health effects o f exposure to p h en o x y a cetic acid h erb icid es a n d /o r their associated contam inants (chlorinated dioxins). The work w as d on e at the request o f three veterans' organizations -- th e A m erican L eg io n , the V ietn am V eterans of A m erica, and the N ational Veterans Legal Services Project -- to b e in clu d ed in th e d eb a te in C o n g ress ab ou t com pensation o f V ietnam veterans for service-connected health problem s. The review w as d on e prim arily to see h o w m uch evidence had accum ulated that w ould support a veteran's claim that h is or her health problem m ight be related to th ese p a rticu la r ch em ical ex p o su res. It w as d o n e w ith the sam e overall goal as had been set for the D epartm ent o f V eterans A ffairs' (form erly the Veterans A dm inistration or VA) o w n scientific ad visory com m ittee, but the task force cam e to different conclusions and policy recom m endations. T he report b y the A g en t O range Scientific Task Force w as released in A pril, 1990, and im m ediately becam e a part o f the consideration o f the C ongressional Com m ittee on V eterans A ffairs in h earin gs it held on extend ing co m p en sa tio n to V ietnam veteran s su fferin g from a variety o f cancers and other health effects. The Secretary o f V eterans A ffairs, E dw in D erw insk i, p reviou sly had recom m en d ed ex ten d in g b en efits to V ietnam veterans w h o had b een d ia g n o sed w ith n o n -H od gk in 's lym p hom a. T he hearings in volved consideration o f this and several other cancers and non-m alignant diseases. O ne of the stron gest proponents o f extend ing com pensation to other d isea ses w as A dm iral Elm o Z u m w alt, w h o had been C h ief o f N aval O perations in Vietnam. H is son had been d ia g n o sed w ith both H o d g k in 's Disease and nonH od gkin's lym phom a and had died after m uch publicity about h is m ilitary service in Vietnam . THE DEBATE OVER THE EVIDENCE The debate about the degree of scientific evidence connecting health problem s to exposure to phenoxyacetic acid h erb icid es a n d /o r chlorinated d ioxin s con tinu es. N e w inform ation about occupationally exposed w orkers co n tin u es to a ccu m u la te. M ost recen tly , a can cer m ortality stu d y o f 5,172 w orkers exposed to TC D D found statistically sign ifican t excess m ortality from all cancers com bined, cancers o f the respiratory tract, and so ft tissue sarcom a, and concluded that the results w ere consistent with the status o f TO DD as a carcinogen. (2) N evertheless, the p ap er an d p u lp in d u stry sim u lta n e o u sly is ch allenging the d ioxin p oten cy index proposed by the Environm ental Protection A g en cy (EPA), and h as offered testim ony that dioxin effluent standards for paper and pulp m ills in G eorgia, M aine and several other states are too stringent. The follow in g su m m ary o f the Report o f the Task Torce on H um a n Health Effects Associated w ith Exposure to H erbicides and/or their A ssocia ted 'C o n ta m in a n ts (1) is offered as an exam p le o f h ow to reach defensible scientific conclusions and p olicy recom m endations, even under conditions of im perfect hum an ep id em iologic evidence and a politically charged environm ent. T h e task force on A g en t O range w as estab lish ed in 1989-'90 b y th e a fo re m e n tio n ed th ree v etera n s organ ization s. The m em bers o f the task force w ere invited to contribute to a sum m ary of the evidence of the health effects o f A g en t O range that w ou ld be used to support legislation introduced in the U.S. C ongress in the S pring o f 1990. T he p rim ary au th ors w ere tw o in dividu als w ith p revious experience in scientific stud ies o f A g en t O ran ge or d ioxin (Dr. O lson and Dr. Q a p p ), and other m em bers o f the task force and other scientists active in th is fie ld w er e ask ed to r e v ie w d ra ft rep orts and su ggest substantive and editorial changes. The w ork of . / , (^NEW SOLUTIONS Spring 1991 3 l J (/^ L Ajr % 19435 SCIENTIFIC SOLUTIONS p r o d u c in g th e fin a l r e p o r t w a s c o o r d in a te d b y th e health e ffe d w as "more lik ely than not" or "at lea st as N a tio n a l V eterans L egal S ervices P roject T h e full list o f lik ely as not" d u e to the exp osu re. T he latter criteria is m em bers included several other researchers fam iliar w ith im portant for disability com pensation purposes under the A g e n t O r a n g e a n d its h e a lth e ffe c ts or w ith o th er VA w here a statistically significant association w ill be environm ental d eem ed to ex ist if the ex p o su re s an d h ealth evidence for and against effects. T he issu e th e task "The issue the task force was asked to the association is a t least equally balanced. (4) force w as ask ed to a d d ress is o n e of the m o st co n troversial of con tem p orary p u b lic health issues: the health address is one o f the m ost controversial o f contemporary public health issues: the health effects o f Agent Orange, specifically its components 2,4,-D, 2,4,5-T (two phenoxyacetic T h is ap p ro a ch to interpreting the results o f sc ien tific stu d ie s in ord er to m ake recom m en d ation s to p o licy effe c ts o f A g e n t O ra n g e, acid herbicides) and the dioxin contam inants.tt m akers differs from the sp ec ifica lly its com trad itional ap proach . It p o n e n ts 2,4,-D , 2,4,5-T w a s p r im a r ily in (tw o phenoxyacetic ad d response to the general h e r b id d e s) an d th e d io x in con tam in an ts. T here h ave intent o f the C ongress and the courts to give the benefit of b e e n n u m e r o u s s d e n t if ic r e v ie w s o f th e v o lu m in o u s the d o u b t to the veteran. B ut it h as broader relevance to lite ra tu r e o n th is su b ject. S o m e w er e for reg u la to ry standard-setting or com pensation policies in other areas. p u r p o se s su ch as esta b lish in g a llo w a b le en vironm en tal G iven th e difficulty o f con d u ctin g scientific stud ies in the co n ta m in a tio n le v e ls or c o n d itio n s u n d er w h ich th e hum an p op u lation s sp ecifically affected b y som e toxic h erb id d e s m a y b e u sed in th e U .S. (2,4,5-T w a s banned e x p o s u r e , a n d th e d iff ic u lt ie s in e x tr a p o la tin g from m o re th an 10 y ea r s a g o ). L ik e w ise, th ere h a v e b een anim al experim ents to hum ans, it seem s hum ane to give several rev iew s o f the health e ffe d s o f A gen t O range the b en efit o f the d o u b t to th e victim s. T he further e x p o su r e in V ietn am v ete ra n s, so m e b y g o v ern m en t im plication o f such an ap proadi is that those in a position co m m issio n s in the U .S. an d A ustralia, w h ich u su ally to prevent toxic exposures, w hen som e sd en tific evidence foun d that the ev id en ce w as too w eak or contradictory to of potential health effects is available, should act early to con clu d e that an yth in g excep t a serious skin condition red u ce o r avoid su ch ex p o su res and thereby p reven t c a lle d c h lo r a c n e is c a u se d in h u m a n s b y e x p o su r e to d isea se an d eventu al co m p en sa tio n costs later. dioxin or A gent O range. T his latest review , com p leted in the Spring o f 1990, T he U .S. V eterans A dm in istration also h as conducted w as d irected prim arily at assessin g ad verse health effects re v ie w s o f th e literature on A g en t O range and has an in V ietn a m v etera n s w h ic h m a y b e a sso c ia ted w ith a d visory com m ittee o f sd en tists w h o se responsibility is e x p o s u r e to A g e n t O r a n g e , a p h e n o x y a c e tic acid o n -g o in g re v ie w o f n ew ev id e n c e. T h is ad v iso ry h erb icid e w h ich w a s con tam in ated w ith d ioxin . c o m m itte e h a d u s e d th e sta n d a rd in m a n y s d e n tific (R eferen ces in th is rep o rt to "D ioxin," "T C DD ," and c o m m itte e s w h ic h can b e su m m a rized a s a " b eyon d-a- "2,3,7,8 - TC D D " all refer to 2,3,7,8 - tetrachlorodibenzo - reason able-d ou bt" a p p ro a d i to the stu d ies it review ed . It p - d io x in .) T h e p u r p o s e o f th e n e w r e v ie w w a s to is a n e x tr e m e ly c o n s e r v a tiv e a p p r o a c h , w h ic h o th e r d eterm in e the exten t o f th e ev id e n c e in the literature in con trib u tors to this journal h a v e characterized as "safe support o f veterans' d a im s of a d verse health effects. b u t o v er ly critical." (3) T h e A g en t O range task force, on T h is rev iew in clu d ed all stu d ie s and p u b lication s th e o th er h a n d , took th e ap proach that there m ust b e review ed u p until 1990 b y the A dvisory C om m ittee on -i so m e e v id e n c e from r e le v a n t sc ie n tific stu d ie s th at E nvironm ental H azards of the D epartm ent of Veterans r e a c h e d th e le v e l o f s ta tis tic a l s ig n ific a n c e w id e ly A ffairs, and additional pertinent stud ies not review ed by a c c e p te d in th e fie ld (u s u a lly le s s th an a 5 p erc en t the ad v iso ry com m ittee. It w a s aim ed prim arily at the probability that the find in g w as d u e sim p ly to chance). p o licy d erisio n o f com p en sation for sp ecific illn esses B ut th e overall im p ression o f the full range o f stu d ies, b e c a u se th ey m ig h t b e se rv ic e-re la te d , and w a s n ot both those w hich sh ow ed a strong assod ation betw een sim p ly a n o th er co m p ila tio n o f all the articles in the exp osu re to the com p onents o f A gen t O range and som e sc ien tific literature. W hat fo llo w s is a sh orten ed and health effect and those w hich did not, should be that the adapted version of the task force report * 32 Spring 1991 N IWSOLUTIONS I 19436 SCIENTIFIC SOLUTIONS STANDARD USED BY THE AUTHORS T h is literature rev iew focu sed on ep id em io lo g ic stu d ies of exp osed h um ans and used the standard "significant statistical association" because that is the standard u sed b y the VA's a d v iso ry com m ittee. T he task force did n ot take this to m ean that every stud y or even the m ajority o f stud ies had to m eet som e m inim um test o f statistical sig n ifica n ce su ch as le ss than a 5 p ercen t probability that the result w a s d u e to chance. Likew ise, the task force d id n ot attem pt to com bine the results from all th e review ed stu d ie s into so m e overall resu lt that could b e subjected to th e sam e kind of statistical test. Instead, the task force took into account the aggregate ev id en ce o f a variety o f differen t stu d ies sh o w in g an a sso cia tio n b etw een p h en o x y a cetic acid h erb icid e (a n d /o r d ioxin ) and certain d iseases or conditions. The m em bers then w eigh ed the evid en ce in a subjective w ay, taking into account the strengths and w eaknesses o f the individual stud ies, and m ad e an overall determ ination of the likelihood that the association w as strong enough to be com pensable. There w ere several reasons for this approach. M ost im portant w as that in order for ep id em iologic stud ies to fin d sta tistic a lly sig n ific a n t a sso cia tio n s b etw een exp osu re and health effects lik e cancers, these stu d ies m u st h a v e su ffic ien t p o w e r and sen sitiv ity . T his gen erally requires large num bers o f exp osed subjects availab le for stu d y and fo llo w -u p for m any d ecad es (if n ot for the lifetim e o f the subjects). A major problem w ith the variou s ep id em iologic stu d ies o f p eop le exp osed to phenoxyacetic a d d h erb id des is that there have n ot been m any large populations w ith know n exposures available for stu d y and follow -up o ver lon g periods, or, w here such p o p u la tio n s ex ist, lik e V ietn am veteran s and the V ietnam ese, these have n ot been adequately studied. PRIMARY CONCLUSIONS OF THE AUTHORS A fter review o f the sa e n d fic literature, the task force con clud ed there is a sign ifican t statistical assod ation b etw een ex p o su re to p h en o x y a cetic acid h erb icid es a n d /o r th eir asso cia ted con tam in an ts (ch lorin ated d ioxin s) and the fo llo w in g diseases: non-H odgk in 's lym phom a, soft tissue sarcom a, skin disorders/chioracne, su b d in ica l h ep atotoxic effects (in clu d in g secondary coproporphyrinuria and chronic h epadc porphyria), and porphyria cutanea tarda (m ost likely on ly in individuals w ith in h erited u ro p o rp h y rin o g en d ecarb oxylase d efid en cy ). T he aggregate in terp retation of several sou n d stu d ies sh o w in g a sta tistica lly sign ifican t a sso cia tio n for each o f th ese co n d itio n s m akes this con d u sion inescapable. Further, the sc ien tific ev id e n c e su p p ortin g the existence o f a significant statistical assod ation betw een exp osu re to p henoxyacetic a d d h erb id d es a n d /o r their assoaated contam inants (chlorinated dioxins) and disease is at least as strong as the saen tific evidence o f a lack of the assod ation for the follow in g adverse health effects: H odgkin's d isease, neurologic effects, and reproductive and d evelop m en tal effects. For each o f th ese health effects, there are sou nd scien tific stu d ies sh o w in g statistically significant evidence o f an assod ation betw een exposure and effect. H ow ever, the assod ation is not as consistent for these effects in the various stud ies as for the first grou p o f health effects. There are other adverse health effects for w hich there is so u n d sc ie n tific e v id e n c e o f an a sso c ia tio n w ith exposure, but the evid en ce d oes not reach the level of form al statistical sign ifican ce. W ere the task force a p olicy-m ak in g b od y con cern ed w ith d isa b ility com p en sation for veteran s, it m igh t con clu d e that com p en sation w as d u e. H ow ever, the lim ited data availab le at this tim e sh o w an a sso d a tio n , b ut n ot a sign ifican t statistical association , b etw een exp osu re to phenoxyacetic acid herb icides a n d /o r d ioxin and the follo w in g d iseases: leu k em ias, cancers o f the kidney, testis, pancreas, stom ach, prostate, colon, hepatobiliary tract, and brain, p sy ch o so d a l effects, im m u n ological abnorm alities, gastrointestinal ulcer, and altered lipid m etabolism . Further research sh ould b e directed at these a sso cia tio n s, and p olicy-m ak ers, co g n iza n t o f the lim itations on find in g significant statistical assod ations and of the obligation to give the veteran the benefit of the d ou b t, w ould b e justified in v ie w in g these d isorders in a m ore con clu sive ligh t for the p u rp ose o f establishing entitlem ent to disability com pensation. F in ally, the e p id e m io lo g ic e v id e n c e on the association s b etw een exp osu re and the ad verse health effed s described ab ove is strongly supported b y a w id e range o f experim ental anim al stud ies. (5) E vid en ce in su p p ort o f the a b o v e co n clu sio n s is found in the sum m ary statem ents w hich follow . CANCER A fter review o f the saen tific literature, it is con d u d ed that there is a significant statistical a sso d a tio n betw een exp osu re to p henoxyacetic a d d h erb id d es a n d /o r their associated contam inants (chlorinated d ioxin s) and non- NEWSOLUTIONS Spring 1991.33 19437 r TABLE 1 - SELECTED EPIDEMIOLOGIC STUDIES: SOFT TISSUE SARCOMA AUTHORS (Year. R eferen ce) N O . CASES OR TYPE. C O H O R T SIZ E H a r d e ll St S a n d s tr o m c-c 52 EXPOSURE Phenoxy RR EST COMMENTS__ 53* Original study (1979) (a) H erbicides E r ik sso n , et aL c-c 110 Phenoxy 6.8* (1981) (b) H erbicides Sm ith, et al c-c 82 Phenoxy 1.3 (1984) (c) H erbicides Lynge cohort 4,459 Phenoxy 3.96* (1985) (d) H erbicides C o g g a n , et aL cohort 5,754 Phenoxy 1.04 (1986) (e) W ik lu n d St H o lm cohort 354,620 H erbicides M ixture 0.9 (1986) (0 H o lm e s, et aL PMR 41,059 V ietn am 4.29* PM R stu d y of (1986) (g) Agent Orange 615 deaths H oar; e t aL (1986) (h) o c 200 M ostly 2 ,4-D 1.4 N o d io x in Contam ination K a n g , e t aL c-c 234 V ietn a m 0.83 VA H osp ital (1986) (i) Agent Oranee C ases W ood s, et aL o c 128 Phenoxy 0.8 (1987) (i) H erbicides V in eis, et aL o c 68 Phenoxy 2.7* W om en o n ly (1987) (k) K o g a n St C la p p c-c 10 H erbicides V ietn am 5.16* SM O R w ith in (1988) 0) A gent Orange PMR Study E riksson, et al o c 237 Phenoxy 1.80* S p ecific for (1990) (m) H erbicides D ioxin CDC o c 346 V ietn a m 1.00 1.6 in I C orp s (1990) (n) A gent Orange W in gren et aL o c 96 Phenoxy 1.6 (1990) (o) H erbicides C lap p , et al. o c 17 V ietn am 3.08* C-C lin k age stu d y (in press) (p) A gent Orange F in g erh u t et aL cohort 5,172 Phenoxy 3.38 9.22* for th ose (1991) (q) em ployed > 1 year k 20 vrs. latency 'S tatistically Significant(p<.05) c-c m eans case control study PM R m eans Proportionate M ortality Ratio SM O R m eans Standardized M ortality O d d s Ratio N H L m eans N on-H odgkin's Lym phom a 34 Spring 1991 NEW SOLUTIONS 13438 SCIENTIFIC SOLUTIONS TABLE 2- SELECTED EPIDEMIOLOGIC STUDIES: NON-HODGKIN'S LYMPHOMA AUTHORS NO. CASES OR RR (Year. Reference) HardeU TYEL COHORT SIZE EXPOSURE. EST. COMMENTS c-c 169 Phenoxy 4.8* (1981) (r) Herbicides Bunneiste^ et al. cohort 6,402 Mixture 1.29 (1983) (s) Pearce, et aL c-c 83 Chlorophenol 1.4 ' (1986) (t) Hoaiv et aL c-c 200 2,4-D 1.6 6.0*for those (1986)(h) Primarily soravine >20davs/vr. Woods, et aL c-c 576 Phenoxy 1.07 1.71*for those (1987)0) Herbicides emdoved 15 vrs. Sc15vrs. latency Bond, et aL cohort 878 Phenoxy 3.1* NHL in a grouped (1988) (u) Herbiddes Category_____________________ Breslln, et aL PMR 24,235 Vietnam 2.10* Marines only (1988) (v) (deaths) Aeent Oranee CDC c-c 1,157 Vietnam 1.47* 125* in I Corps (1990) (n) Aeent Oranee Zahm, et aL c-c 201 2,4-D 1.5 3.3 in those (1990) (w) Primarily spravine >21 days/year Wigle, et aL cohort 69,513 Mixture 0.92 1.34in those on farms with (1990) (x) 250+ acres sprayed * HardeU, L 4cSuiditrom, A 'Cue-Control Study; Soft TissueSarcomas and Exposure to Phenoxyacetk Adds or Chlorophenol*" Sr / Cancer39(1979): 711717. b. Eriksson, M. HardeU, L, et aL "SoftTissue Sarcomas and Exposun to ChemicalSubstances: A Case-Referent Study,' Sr JInd M id 38(1981):27-33. c Smith, A H . Pearce, N.E, Flshe* D.O. et at 'Soft-Tissue Sarcoma and Exposure to Phenoxyherbiddes and Chlotophenols in New Zealand,* JNCI 736K1984): 1111-1117. d. Lyng* E. 'A Follow-UpStudyofCancerIncidenceAmong Workersin Manufacture of Phenoxy Herbicides in Denmark,' Sr/ Cancer5200(1985): 259-270. e. Coggon, D. Pannett, B. Winter P.D, et aL, "Mortality of Workers Exposed to 2 Methyl- 5 Chlorophaioxy AceticAdd,' Sand / WorkE m in H allh 12(1986): 448-1S4. f. Wiklund, K. and Holm, L-E "Soft-TissueSarcoma RiskInSwedish Agricultural and Forestry Worker*' JNCI76(1986):2229). g. Holmes,AJT. Bailey, C. Baron, EC, et aL 'Vletnam-Era VeteransMortality Study.' West Virginia Health Department, 1986, h. Hoar; SJC. Btai*A, Holmes, FA, et aL 'Agricultural Herbicide Useand Riskof Lymphoma and Soft-TissueSarcoma,' JAMA 256(9)(1986): 1141-1147. L Kang, H Jt, Weatherbce, L, Sreslin, P i. et aL "Soft Tissue Sarcomu and Military Service in Vietnam: A Case Comparison Croup Analysis of Hospital Patients,' J Occupations/Me*28(1986): 1215-1218. j. Wood* JA. Polissa* L, Severson, EK. et aL ^oft Tissue Sarcoma and Non-Hodgkins Lymphoma in Relation to Phenoxy Herbicides and Chlorinated Phenol Exposure in Weitern Washington,' [NCI 78(1987): 899-910. k. Vincis, P. TerradnL B, Cicome, C,etaL "PhenoxyHerbiddes and Soft-TissueSarcomu in Female RiceWeeders: APopulation-Based Case-Referent Study,' Sand I Work E m in H a ltk 13(1987): 9-17. L Kogan, Mi), and Capp E W. "Soft TissueSarcoma Mortalityamong Vietnam Veterans in Massachusetts. 1972-1983,' Int J Epi 17(1988):39-0. m. Eriksson, M, HardeU, L. rnd Adami, HO. "Exposure to Dioxins u a Risk Factor for Soft Tissue Sarcoma: A Population-Based Case-Control Study,' [N O 82(1990): 486-490. n. CDC *71 Association of Selected Cancers with service in the US Military in Vietnam.' VoLL Non-Hodgkins Lymphoma; VoLII, Soft-Tissue and Other Sarcomas. Atlanta, GA 1990. o. Wingren, G. Fredriksson, M, Noorlind Brag* H, et aL "Soft-TissueSarcoma and Occupational Exposure*' C tnar 66(1990): 806-811. p. Oapp, EW. Cupple* LA, Colton, T, and Ozonoft, D.M. 'Cancer surveillanceof veterans in Massachusett* 1982-88,' Int J Epi(1991). q. Fingerhut MA, Halpoin, W.E, Marlow, DA, et aL 'Cancer Mortality in Workers Exposed to 23,7,8-Tetrachlorodibenzo-p-Dloxin.' NewEngftnd Jtunal ef Medici 32409911:212-218. t HardelL L *Relation of Soft-Tissue Sarcoma, Malignant Lymphoma and Colon Cancer to Phenoxy Add* Chlorophenol* and Other Agent*' San d j Ybrk E m in H a ltk 7(1981): 119-1301 * Burmdste* LF. Everett, CD, VanLte SLF.and Isacson, P. "Selected Cancer Mortality and FarmPracticesin low*' Am / Epi 118(1983):72-77. L Pearce, NE, Smith, A H, Howard, JJC, et aL "Non-Hodgkins Lymphoma and Exj>oeure to Phenoxyherbidde* Chlorophenol* Fencing Work, and Meat WorksEmployment ACase-ControlStudy,' Br / India M id43(1986): 75-83. u. Bond, G C, Wettertroem, MH, Rousch, G.), et aL' Cause SpecificMortalityamong Employees Engaged in the Manufacture, Formulation, or Packagingof 2A-0lchlorophenoxyaceticAdd and Related Salt*' Br/ India Med 45(1988): 98. v. Breslln, P, Kan&H H , Le* Y, et aL 'Proportionate MortaUty Study of USArmy and USMarine Corps Veterans ofthe Vietnam War,'/Occ Med 30(5X1988): 412-419. w. Zahm, SH , Wdsburge* D.D, Babbitt, PA, et aL 'A Case-Control Study of Non-Hodgkins Lymphoma and the Herbidde 2A-Dichlorophenoxyacetic Add 04-0) in Eutem Nebrask*' Epidemiology 10(1990): 349356. x. WIgl* D.T, Semendw, EW, Wilkin* K, et aL 'Mortality Study of Canadian Male Farm Operator*' Non-Hodgkin's LymphomaMortalityand Agricultural Practices in Saskatchewan,* JNCI 82(1990): 575-582. NEWSOLUTIONS Spring 199135 19439 H odgkin's lym phom a and soft tissue sarcom a, and that and soft tissue sarcom a. T hese tw o cancers have been th e e v id e n c e s u p p o r tin g th e e x is te n c e o f su c h an found elevated in other studies o f veterans d on e by state association is at least a s stron g as the ev id e n c e o f a lack o f health dep artm ents in M assach usetts (7), W est V irginia (8) the association for H odgkin's disease. A representative and W isconsin (9). A t a m in im u m , th e C D C fin d in gs group of the stu d ies sh o u ld b e p u t in the w hich w ere review ed in con text of these other ord er to reach th is con clu sion is "There are other adverse health effects for stu d ies o f veterans, as w ell as stu d ies of su m m a rized in the accom pan ying tables. T he C en ters for D isease C ontrol (CDC) Selected C ancers S tu dy which there Is sound scientific evidence o f an association with exposure, but the evidence does not reach the level of form al statistial significance. Were the taskforce a policy agricu ltu ral w ork ers ex p o sed to h erb icid es, sev era l of w h ich are su m m a rized in the tables. (6), relea sed in A p ril, 1990, fou n d that veterans w h o served in V ietn am had a making body concerned with disability compensation for veterans, it m ight conclude that compensation was due." T h e C D C S elected C ancers Study and other p rop osed stu d ies of A gen t O range-exposed sig n ifica n tly in creased veteran s w ere the risk o f n on -H o d g k in 's subject of a C ongres lym ph om a over veterans w ho served elsew h ere d uring sion al h earing on July 11, 1989. T h is and su b seq u en t the sam e p eriod. T his stu d y actually m ay add su pport to hearings on the C D C d ioxin stu d ies w ere su m m arized in the ev id en ce that exp osu res in V ietnam cau sed increased a stron gly w ord ed report b y th e H u m a n R esources and risk o f n o n -H o d g k in 's ly m p h o m a , e v e n th o u g h its Intergovernm ental R elations Subcom m ittee o f the C om authors d en y that their data support the contention that m ittee on G overnm ent O p erations, H ou se o f R epresen A gen t O range w a s the cause. A lthough the stu d y d id not tatives (A ugust 9,1990). (10) focu s on an y direct m easure of A gent O range exposure, O ther conditions also w ere fou n d to b e associated and the p rop osed blood sam pling to d eterm in e d ioxin w ith exp osu re to p h en oxyacetic a d d h erbirides. A full le v els in p atients w a s n ev er carried ou t, the authors m ad e d iscu ssion o f the evid en ce for these a sso a a tio n s is found conclusions concerning A gent Orange exposure. in the report o f the task force and is n o t presented here. First, the C D C authors contend that V ietnam veterans C o p ies are availab le from the N a tio n a l V eterans L egal w h o served in m C orp s w ere m ost lik ely to b e exp osed to Services Project A g en t O range, and the results of their stu d y indicate that they w ere a t decreased risk of non-H odgkin's lym phom a. S econ d , the C D C stu d y found excessive risk o f nonH o d g k in 's ly m p h o m a in "blue-w ater" N a v y veteran s w h o m th e au th ors p resum e had n o exp osu re to A gen t O range. Finally, the C D C authors state that the responses to q u estion s ab out spraying, handling or b ein g around A gen t O range w ere sim ilar am ong veterans w ith nonH odgkin's lym phom a and veterans w ith no cancer of any type. Critics h ave challenged these C D C interpretations of th e data b eca u se th e stu d y did n ot contain a n y d irect m easu re o f exp osu re to A g en t O range. In ad d ition , som e indirect ex p o su re criteria n ot em phasized b y C D C d o lead to th e con clu sion o f ad verse health effects in exp osed veteran s. For exam p le, one of the h ea v ily sp rayed region s in V ietnam w a s I Corps. If the C D C data on v etera n s in I C o rp s an d III C orp s are taken together, th ey sh o w an increased risk o f both non-H odgkin's lym phom a CONCLUSIONS T his review of the sd en tific literature led the task force to the follow ing condusions: 1. There is a significant statistical assod ation betw een exp osu re to phenoxyacetic a d d h erb irides a n d /o r their associated contam inants (chlorinated d ioxin s) and nonH o d g k in 's lym p h om a, so ft tissu e sarcom a, sk in d isord ers/ch loracn e, su b clin ical h ep atotoxic effects (in clu d in g secondary coprop orp h yrin uria and chronic h ep atic porphyria), and porphyria cu tanea tarda (m ost lik e ly o n ly in in d iv id u a ls w ith in h e r ite d u rop orp h yrin ogen d eca rb o x y la se d eficien cy ). The aggregate interpretation o f several so u n d stu d ies sh ow in g a statistically significant a sso d a tio n for each of these conditions m akes this con d u sion inescapable. 2. The sdentific evidence supporting the existence of a sign ifican t statistical a sso d a tio n b etw e en exp osu re to p henoxyacetic ad d h erb icides a n d /o r their assod ated 36 Spring 1991 NEW SOLUTIONS 19440 SCIENTIFIC SOLUTIONS contam inants (chlorinated dioxins) is at least a s strong as E D IT O R 'S N O T E : O n J a n u a ry 2 9 , 1 991, th e U .S . the scientific ev id en ce o f a lack o f the association for the C ongress p assed H.R. 556, the A gen t O range A ct of 1991. fo llo w in g a d v e r s e h ea lth effects: H o d g k in 's d is e a s e , It p rovid es for com p en sating Vietnam veterans diagnosed neurologic effects, and reproductive and developm ental w ith non-H odgkin's lym phom a, soft tissue sarcoma, and effects. For each o f these health effects, there are sound ch loracn e. It w a s p a ssed b y the Sen ate and sign ed b y scientific studies sh ow in g statistically significant evidence President Bush in the follow in g tw o days. o f an association betw een exposure and effect. H ow ever, a statistically significant association is n ot a s consistently supported for th ese effects as for d ie first grou p o f health effects. - ACKNOW LEDGEM ENTS The authors w ould like to thank the reviewers fo r their helpful edmments. 3 . T h e a b o v e e ig h t a d v e r s e h e a lth e ffe c ts sa tis fy thReEFER E N C ES V A 's test for a n a sso c ia tio n w h ich q u a lifie s fo r d isa b ility 1. G app, R.W., Commoner, B., Constable, ]., Epstein, SS., com pensation, that is, w here the scientific evidence sh ow s Kahn, P.C, Olson, J.R, Ozonoff, D.M. "Report of the Task that a significant statistical association is "at least as lik ely Force on Human Health Effects Associated with Exposure a s not." (4). to Herbicides a n d /o r Their Associated Contaminants (chlorinated dioxins),* 1990. Available from National DISCUSSION Veterans Legal Services Project, 2001 S Street N.W., Washington, DC 20009-1125. The conclusions o f the A gen t O range Scientific Task Force w ere different from the VA A d visory C om m ittee and other scientific review s such as on e produced for the 2. Fingerhut, M.A., Halperin, W.E., Marlow, D.A., et al. 'C ancer Mortality in Workers Exposed to 2,3,7,8Tetrachlorodibenzo-p-Dioxin.' New England Journal of Medicine 324(1991X212-218. p u lp a n d p a p e r in du stry. (11) T h e task fo rce c o n d u c te d a 3. Moure-Eraso, R. and Tsongas, T. "Benzene and Cancer The com prehensive review o f epidem iological data available OSHA Standard, Workers' Compensation and Public as o f early 1990 w hich included som e p ertin en t peerreview ed publications not review ed at that tim e by the other tw o bodies. Furthermore, the review w a s focused Health Policy," New Solutions 1(2X1990k13-21. 4. 38 Code of Federal Regulations 1.17(d)(1); 54 Federal Register 40389 (Oct 2,1989). 5. Environmental Protection Agency. "Health Assessment on the issue o f identifying possible health problem s for Document for Polychlorinated Dibenzo-p-dioxin," EPA- w hich V ietnam veterans m ay receive com pensation and 600/8-84-014F, 1985. u sed an approach w hich varied from that u sed b y the VA A dvisory C om m ittee. The task force conclusions, if they w ere im plem ented as p olicy, a lso w ou ld h ave sig n ifica n t econ om ic 6. Centers for Disease Control. "The Association of Selected Cancers with Service in the US. Military in Vietnam." VoL I, Non-Hodgkins Lymphoma; VoL 0 , Soft-Tissue and Other Sarcomas. Atlanta, CA: 1990. 7. Kogan, M.D. and Gapp, R.W. 'Mortality among Vietnam im plications for the affected veterans and their fam ilies. Veterans in Massachusetts, 1972-1983." Office of the T h e actual n um ber o f veteran s w ith the d isea ses or Commissioner of Veterans Services Agent Orange Program, c o n d itio n s liste d in co n clu sio n 1, a b o v e, ca n n o t be estim ated w ith precision, b u t the overall cost to the U.S. Treasury w ould be relatively m od est b y com parison to 1985. 8. Holmes, A.P., Bailey, C., Baron, R.C., et al. 'Vietnam-Era Veterans Mortality Study." West Virginia Health Department, 1986. the total am oun t o f veterans benefits paid annually. 9. Anderson, H.A., Hanrahan, L.P., Jensen, M., et al. Since die task force report w as subm itted to C ongress "Wisconsin Vietnam Veteran Mortality Study: in M ay, 1990, th e S ecretary o f V eteran s A ffa ir s h as reco m m en d ed th at so ft tissu e sarcom a in V ietn am veteran s b e con sid ered a service-related con d ition . Earlier, h e had recom m ended com pensating V ietnam veterans Proportionate Mortality Ratio Results; Standarized Mortality Ratio Results. Final Report' Wisconsin Dept of Health and Social Services, Division of Health, Section of Environmental and Chronic Disease Epidemiology, 1986. 10. Human Resources and Intergovernmental Relations w h o had non-H odgkin's lym phom a, sayin g that it w as a Subcommittee. "The Agent Orange Coverup: A Case of d ecision n ot based on science but as a m atter o f public Flawed Science and Political Manipulation." Twelfth policy. The im plications o f these tw o policy d ecisions are that there m ay be a shift toward givin g the benefit of the d ou b t in areas o f scientific controversy to the victim s w ho Report by the Committee on Government Operations, together with Dissenting Views, 1990 11. National Council of the Paper Industry for Air and Stream Improvement, Inc. 'D ioxin: A Critical Review of its h a v e th ese tw o can cers. If this is so , then th e A g en t Distribution, Mechanism of Action, Impacts on Human O ran ge Scientific Task Force m ay h a v e p layed a role in Health, and the Setting of Acceptable Exposure Limits." justifying a n ew approach to such policy d ecision s. Technical Bulletin #524,1987. 37NEW SOLUTIIOONN5S Spring 119291 19441 rE n v ir o n m e n ta l H ea lth P e rs p e c tiv e s Vol. 70, p p . 2 21-227, 1986 ( \ Calculation of 2,3,7,8-TCDD Equivalent Concentrations of Complex Environmental Contaminant Mixtures by GeorgeEadon,* Laurence Kaminsky,* Jay Silkworth,* Kenneth Aldous,* David Hilker,* Patrick O'Keefe,* Robert Smith,* John Gierthy,* John Hawley,* Nancy Kim,* and Anthony DeCaprio* Sufficient toxicological data are now available to permit use of conventional risk assessment techniques to estimate the hazards associated with human exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8TCDD). However, many real-worid exposures involve complex mixtures of dibenzodioxins, dibenzofurans, and related compounds. Historical approaches to risk assessment on such mixtures have ranged from ignoring all compounds except 2,3,7,8-TCDD itself to assuming that all compounds have potencies equal to 2,3,7,8-TCDD. An alternative approach which uses existing literature data and analytical results to calculate the "2,3,7,8-TCDD equivalent" concentration of a mixture in order to "predict" its biological potency relative to 2,3,7,8-TCDD itself is advanced here. Previously reported in vivo acute and subchronic studies and some recently obtained analytical chemistry data are integrated here to clarify the utility of this important approach and to assess the uncertainties associated with its use. This predictive approach, and various conceptually similar ones, have now found wide applicability to the risk assessment process associated with exposure to complex mixtures of dioxins, dibenzofurans, and related compounds. Introduction Sufficient toxicological data are now available to per mit use of conventional risk assessment techniques to estimate the hazards associated with human exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) (1). In a few incidents, 2,3,7,8-TCDD has been the only congener present at toxicologically significant concen trations. More commonly, however, real-world expo sures involve complex mixtures of compounds with dioxin-like activity containing up to 75 different chlor inated dibenzo-p-dioxin (PCDD), 135 chlorinated dibenzofuran (PCDF), and 75 chlorinated biphenylene (PCBE) isomers and congeners. For example, com mercial polychlorinated biphenyls (PCBs) (2), emissions from garbage-burning resource recovery plants (3), and chemical waste incinerators U), foodstuffs (5), human milk (6), and pyrolyzed PCBs or chlorobenzenes (7,8) all consist of complex mixtures in which 2,3,7,8-TCDD is a minor component relative to the total PCDD, PCDF, and PCBE concentration. Such mixtures present a difficult problem for esti *Wadsworth Center for Laboratories and Research, New York State Department of Health, Albany, NY 12201. mating human health hazards for at least three reasons. First, analytical chemistry has not yet progressed to the point that reliable isomer-specific quantitation or even qualitative identification is possible for the full range of dioxin-like compounds. For example, the un labeled standards needed both for qualitative identifi cation and for determination of response factors are not generally available. In addition, only a very limited set of 13C- or 37Cl-labeled internal standards, which are necessary to correct for run-to-run variations in recov ery, have been prepared. Even modem high resolution capillary gas chromatographic techniques do not permit resolution of all possible congeners. Second, insufficient toxicological data are available to permit a rigorous risk assessment for most congeners. For example, two-year carcinogenesis data are available only for 2,3,7,8-TCDD itself, for 2,7-dichlorodibenzodioxin, and for a mixture of two 2,3,7,8-substituted hexachlorodibenzodioxins (hexaCDDs) (9). Single oral dose guinea pig LD50 values have been reported for 16 different dioxins and five different furans (10). Fewer than 25 of the 75 possible dioxin and 20 of the 135 possible furan congeners have been subjected to in vitro determination of biological activity (10). Finally, there exists a third confounding factor. The biological potency of the mixture may not 19442 . 222 EADO\' ET AL. r simply equal the summation of the activities of the in Table 1. Influence of structure and chlorination pattern on \ dividual congeners present, if synergistic of antago guinea pig oral LDM(male. Hartley, 200-250 g). nistic interactions among the mixture's components are significant. On February 5, 1981, a fire involving a transformer containing 60% Aroclor 1254 and 40% tri- and tetrachlorobenzenes extensively dispersed a fine oily soot throughout a 17 story office building in Binghamton, New York. Chemical analysis demonstrated the pres ence of substantial concentrations of PCDDs, PCDFs and PCBPs in the soot and on various surfaces (11-13). The necessity of establishing reoccupancy criteria for the building focused attention on the need to estimate the biological potency of this complex -mixture. Two conceptually opposed methods can be discerned in pre1981 governmental approaches to this problem. The commonly used early approach (probably encouraged by lack of analytical chemistry data) was explicitly or Compound LDjo, pg/kg 2,3,7,8-TetraCDD 2,3,7,8-TetraCDF 2,3,4,7,8-PentaCDF 1,2,3,7,8-PentaCDD 1,2.3,4,7,8-HexaCDD 1,2,3,7,8,9-HexaCDD 1,2,3,7.8-HexaCDD 2,3,4,6,7,8-HexaCDF 1,2,3,4,6,7,8-HeptaCDD 1,2,4,7,8-PentaCDO 2,3,7-TriCDD 2,8-DiCDD 1,3.6,8-TetraCDD 2.5" 5-10b <10" 3.1' 73c 60-100' 70-100' 120" <600' 1,125' 29,444' 730,000' >15,000,000" "Data of Silkworth et al. (19). bData of Moore et al. (20). Ratio LDm compound LDso 2,3,7,8-TCDD 1 2-4 <4 1.2 29 24-40 28-40 48 <240 450 1,200 300,000 600,000 implicitly to ignore all compounds other than 2,3,7,8TCDD itself. Alternatively, all compounds detected were assumed equal in potency to 2,3,7,8-TCDD (U). 'Data of McConnell et al. (2 1 ). dData of Kawamura et al. (22). *Data of McKinney et al. (23). Either approach ignores substantial biological evidence suggesting that the potencies of particular congeners vary widely, but are, in some cases, negligible when compared to 2,3,7,8-TCDD. Any realistic attempt to estimate a mixture's biological potency must therefore take into account the relative potencies of its individual components. The approach eventually developed to estimate the hazard associated with exposure to the Binghamton environmental mixture's biological potency relative to 2,3,7,8-TCDD itself, and thus verification of the TCDDequivalents calculation. The intent of this paper is to present an integrated picture of the previously reported biological experiments (Table 1) and some recently ob tained analytical chemistry data (Table 2) to clarify the utility of this important approach. mixture is a fairly general one. Published short-term biological data are used to estimate the ratio of the Calculation of T C O D Equivalents potency of a particular congener relative to 2,3,7,8TCDD. Then, the observed or estimated concentration from Chemical Data of that congener is multiplied by this ratio to estimate Chemical data can be used to estimate the 2,3,7,8- the "2,3,7,8-TCDD equivalent" concentration of that TCDD equivalent concentration present in a mixture, congener. This 2,3,7,8-TCDD-equivalent concentration provided that the concentrations of the relevant com of a particular congener would thus be equated with a pounds are known and that their biological activities hypothetical concentration of 2,3,7,8-TCDD capable of can be estimated. A biological end point of special con producing the same biological effect as the actual con cern for these compounds is carcinogenesis (1); unfor centration of that congener. The overall biological po tunately, only five dibenzodioxins and no dibenzofurans tency of the mixture could then be estimated by sum have been the subject of lifetime animal bioassays (10). ming the T C D D -equivalent concentrations of all Thus, short-term bioassay data, which are substantially congeners present. Such an approach is only applicable more numerous, must be utilized as the basis of the when the components of the mixture exhibit biological equivalence calculation. The end point used in the pres effects similar to those of 2,3,7,8-TCDD. ent study to estimate relative congener potency is the When first advanced (15), the TCDD-equivalent ap single-dose oral LDso in the guinea pig. This is advan proach was novel; more recently, conceptually similar tageous, since the LD50of the mixture in question (soot) approaches have been used in assessing other PCB- has been experimentally determined (24), thus allowing related incidents, e.g., the One Market Plaza incident direct verification of the calculation. However, since in San Francisco (16), and in assessing other sources of there is good correlation among various short-term as environmental exposure, e.g., emissions from resource says for these compounds (25 -2 9) any differences be recovery plants (17), and these have gained increasing tween results from particular short-term assays should acceptance as general solutions to the PCDD/PCDF have only minor effects on the calculation (18). mixture problem (18). However, it must be emphasized The issue of the relevance of short-term exposure that the only direct experimental support for this ap data to carcinogenic potency is less clear. However, the proach rests on work performed in connection with the very limited experimental data now available supports Binghamton incident. Here alone was sufficient contam the concept that short- and long-term effects are inated material (i.e., "soot") and appropriate laboratory roughly proportional. Thus, a mixture of two 2,3,7,8- resources available to allow in vivo comparisons of an substituted hexaCDDs exhibited carcinogenic^iot^ncy TCDD EQUIVALENCY OF COMPLEX MIXTURES 223 Table 2. Concentrations of chlorinated dibenzo-p-dioxins and dibenzofurans in the Binghamton soot sample used in animal toxicology experiments.* Compound Concentration, ppm Total diCDD Total triCDD Total tetraCDD 2,3,7,8-tetraCDD Total PeCDD l,2,3,7,8-PeCDDb Total HxCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD Total HpCDD 1.2,3,4,6,7,8-HpCDD Total OCDD Total diCDF Total triCDF Total tetraCDF 2,3,7,8-tetraCDF Total PeCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF Total HxCDF 1,2,3,4,7,8-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF Total HpCDF 1,2,3,4,6,7,8-HxCDF Total OCDF 0.5 0.5 1.5 0.5 1.4 0.5 1.2 0.2 0.2 0.2 1.4 0.7 1 5.9 27 120 17 160 22 12 64 26 1.2 7.7 1.5 26 10 13 *Unless otherwise indicated, concentrations listed are based on a recent redetermination of the PCDD/PCDF constituents in the soot using contemporary analytical methodology. This redetermination is considered more reliable than that previously reported (32), partic ularly in view of recent advances in the availability of labeled and unlabeled standards and chromatography. bEstimated from data of Silkworth et al. (32). estimated to be 0.04 times that of 2,3,7,8-TCDD (18); the ratio of corresponding guinea pig acute oral LD50 values is 0.03. Similarly, in two-year studies of unsub stituted dibenzodioxin (30) and 2,7-dichlorodibenzodioxin (31), doses up to 10,000 ppm in the diet elicited no carcinogenic response in either species. Unchlori nated and dichlorinated dibenzodioxins are essentially inactive relative to 2,3,7,8-TCDD in a battery of short term bioassays (10). Table 1 presents the currently available data on guinea pig acute oral LD50 values for dibenzodioxins and dibenzofurans. Use of this data set to predict the 2,3,7,8-TCDD equivalent concentration in a mixture (such as the Binghamton soot) is complicated by cer tain limitations, particularly the absence of any data on PCDFs other than 2,3,7,8-T C D F, 2,3,4,7,8PeCDF and 2,3,4,6,7,8-HxCDF. The following as sumptions will therefore be made about LD50 values not yet determined. (1) The ratio of the LD50 values of a particular PCDF congener and 2,3,7,8-TCDF will be the same as the ratio of the LD50 values of the correspondingly substituted PCDD congener and 2,3,7,8-TCDD. There is no direct experimental data to support this assumption. (2) The LD50 values of PCDFs and PCDDs lacking chlorines at any of the four lateral positions will be sufficiently high that their influence can be ignored for most environmental mixtures. This assumption is based on the guinea pig LD^ values of 2,8-diCDD, 2,3,7- triCDD, and 1,3,6,8-tetraC D D and 1,2,4,7,8- pentaCDD. All have LD50 values more than 450 times higher than 2,3,7,8-TCDD itself (Table 1). This simpli fication may be invalid for mixtures containing only very small percentages of 2,3,7,8-substituted isomers. (3) Introduction of a single additional chlorine sub stituent on a 2,3,7,8-substituted congener has essen tially no effect on the congener's guinea pig LD50. This assumption is based on comparison of the LDjo values of 2,3,7,8-TCDD versus 1,2,3,7,8-pentaCDD, and 2.3.7.8- TCDF versus 2,3,4,7,8-pentaCDF (Table 1). (4) Introduction of two additional chlorine substi tuents on a 2,3,7,8-chlorinated congener raises its LD50 by a factor of 30. This assumption is based on compar ison of the LDso values of 1,2,3,4,7,8-, 1,2,3,6,7,8-, and 1,2,3,7,8,9-hexaCDD versus 2,3,7,8-TCDD (Table 1). (5) The LD50 values of compounds with more than six chlorines will be sufficiently high that their influence can be ignored for most environmental mixtures. This assumption is based on comparison of the LDso values of 1,2,3,4,6,7,8-heptaCDD and 2,3,7,8-TCDD (Table 1). These assumptions require that attention be focussed only on 2,3,7,8-substituted PCDDs and PCDFs. The concentration of 2,3,7,8-TCDF in a bulk soot sample used in the animal toxicology experiments described later has recently been measured at 17.1 ppm (Table 2); since the data in Table 1 indicate that the LD50 value of 2,3,7,8-TCDF is about three times that of 2,3,7,8- TCDD, this is equivalent in terms of acute toxicity to a 2,3,7,8-TCDD concentration of about 6 ppm. Based on assumption (2), other tetraCDFs can be neglected in this calculation. The two pentaCDF isomers with 2,3,7,8 substitution together total 34 ppm. If, as is re quired by assumption (3) and as is consistent with ex perimental data for 2,3,4,7,8-pentaCDF, the toxicity of 2.3.7.8- substituted pentaCDFs is considered equal to that of 2,3,7,8-TCDF, this concentration corresponds to a 2,3,7,8-TCDD equivalent concentration of about 11 Hg/g. The hexaCDFs were measured at 64 (ig/g. How ever, based on assumption (2), only 2,3,7,8-substituted congeners (1,2,3,4,7,8-, 1,2,3,6,7,8-, and 2,3,4,6,7,8- hexaCDF) need be considered. Table 1 suggests that 2.3.4.6.7.8- hexaCDF is about 16 times less potent than 2.3.7.8- TCDF itself. Thus, the sum of observed 2,3,7,8- hexaCDF concentrations (36 |xg/g) corresponds to a 2.3.7.8- TCDD equivalent concentration of about 1 ppm (Table 3). Hepta and octaCDFs have insufficient con centrations and biological potencies to contribute to this calculation. The concentrations of 2,3,7,8-TCDD and 1.2.3.7.8- PeCDD have each been measured at 0.5 p.g/ g, corresponding to a "2,3,7,8-TCDD" equivalent con centration of 1 ppm. The concentration of hexaCDDs (1.2 jxg/g), when adjusted for biological potency by di viding by 30, makes a negligible contribution to the calculation. Similarly, hepta and octaCDDs can be ig nored. n. A A A r ' 224 EADON ET AL. Table 3. Calculation of 2,3,7,8-TCDD equivalents due to various chlorinated dibenzofurans, dibenzodioxins, and biphenylenes in Binghamton soot using chemical data.* c Concentration A Relative activity compound class vs. dibenzodioxins B Relative activity due to chlorine substitution 2,3,7,8-TCDD equivalents, ppm 2,3,7,8-TCDF/ 17.1 ppm 1,2,3,7,8- and 2,3,4,7,8PeCDF/33.8 ppm 1,2,3,4,7,8-, 2.3.4.6.7.81.2.3.6.7.8- , and 1,2,3,7,8,9hexaCDFs/36 ppm 2,3,7,8-TCDD/0.5 ppm 1,2,3.7,8-PeCDD/ 0.5 ppm 2,3,6,7-TCPB/l.l ppm 1,2,3,6,7-PeCPB/ 2.3 ppm 1/3 1/3 1/3 , 1 1 1 1 16 1 11 1/16 1 1 0.5 1 0.5 11 12 *2,3,7,8-TCDD calculated as the product C x A x B. Together, the dibenzodioxins and dibenzofurans ac count for about 19 (g/g "2,3,7,8-TCDD equivalents." Stalling's demonstration (12) that the soot contains a substantial concentration of PCBEs complicates the cal culation substantially. The only data available relevant to the biological potency of these derivatives are mea surements of the cytosolic receptor binding affinity and the cell keratinization activity of 2,3,6,7-tetraCBE (an alog of 2,3,7,8-TCDD) (33). These results suggest that 2,3,6,7-tetraCBE may have a biological potency similar to that of its structural analog, 2,3,7,8-TCDD. Unfor tunately, nothing is known about the rate of metabolic detoxification of these compounds. However, in the ab sence of more rigorous data, it will be assumed that biphenylenes exhibit the same biological potency as the corresponding dibenzodioxins, and three times the po tency of the corresponding dibenzofurans. Even today, quantitative determination of the PCBE concentration in the soot is complicated by the absence of any reliable labeled or unlabeled biphenylene stan dards. As a crude estimate of their concentration, Stal ling's observation that the PCBEs are present at ap proximately 1/15 the concentration of the PCDFs can be used. This factor, coupled with the foregoing as sumption that their biological potency is three times that of the dibenzofurans implies a contribution of about 3.4 ppm from the biphenylenes. Thus, the best estimate of the overall potency of the mixture of PCDDs, PCDFs and PCBEs is ca. 22 ppm (Table 3). Calculation of 2,3,7,8-TCDD Equivalents from Biological Data The best test of the various assumptions underlying the calculation of TCDD equivalents from chemical data is to perform an empirical determination of the mix ture's biological potency versus 2,3,7,8-TCDD itself. Since the estimate of biological potency was based on acute guinea pig oral LD50 data, the most direct point of comparison requires experimental measurement of this parameter for the mixture. Female guinea pigs were administered a single oral dose of a benzene ex tract of the soot equivalent to 4, 20, 100, 500, or 1000 mg of soot/kg (2L). The soot extract's LDso was deter mined to be 327 mg/kg, based on a 42-day observation period and calculated by a modification of the method of Bliss (3It). For purposes of comparison, female guinea pigs were similarly given single oral doses of 0.1, 0.5, 2.5, 12.5, or 20 |j.g/kg 2,3,7,8-TCDD (2U). The corre sponding calculated LDso value for 2,3,7,8-TCDD was 19 |xg/kg in an aqueous vehicle. Thus, by using the re lationship (1), 2.3.7.8- TCDD-equivaient concentration (|xg/g) = LDso for 2,3,7,8-TCDD (n-g/g) LD50 for soot (mg/kg) it can be calculated that the soot has a 2,3,7,8-TCDD equivalent concentration of (19 p.g/kg)/(327 x 103) p.g or 58 ppm. A more demanding test of the TCDD-equivalent con cept is to calculate the TCDD equivalent concentration of the soot based on a variety of exposure end points induced subchronically. Thus, Binghamton soot was in corporated into feed at concentrations of 0, 0.2,1.9, 9.3, 46.3, and 231.5 ppm and fed to guinea pigs for 90 days (34). In another study, 2,3,7,8-TCDD was incorporated into feed at concentrations of 0, 2, 10, 76, and 431 ppt and similarly fed to guinea pigs for purposes of com parison (35). End points chosen were those in which significant, dose-related differences from control values were observed after subchronic exposure in both stud ies. These endpoints included relative (to body) thymus weights (males), percent of initial body weight (males), triglyceride levels (males) serum ALT levels (females), and incidence of hepatocellular cytoplasmic inclusion bodies (females). For the continuous data parameters, linear regression analysis was performed to obtain the best straight-line fit for exposure (expressed as ppt 2.3.7.8- TCDD or log ppt 2,3,7,8-TCDD in the feed) ver sus response using data from the 2, 10, and 76 ppt dose groups in the present study. Correlation coefficients were obtained for each line. Response data were nor malized by expressing each value as a percentage of the corresponding control value. This was necessary to cor rect for small variations between the control values in each study. Response data for a particular dose level in the Bing hamton soot study were then compared to the standard curves generated above to obtain a hypothetical 2,3,7,8TCDD concentration in the feed which would have been expected to produce the same degree of response. The "2,3,7,8-TCDD equivalent" concentration of the soot was then calculated by using Equation (2). The partic ular dose levels of Binghamton soot chosen for this com parison were those in which the degree of response feJJ 1944o TCDD EQUIVALENCY OF COMPLEX MIXTURES 225 within the limits of the corresponding standard curve (1.9 ppm soot level was used). 2.3.7.8- TCDD-equivalent concentration (pg/g) _ hypothetical 2,3,7,8-TCDD concentration in feed (pg/g) concentration of Binghamton soot in feed (jj-g/g) A different method was employed for analysis of the quantal data obtained for the incidence of hepatocellular cytoplasmic inclusion bodies. In this case, use was made of the ED50 (dose level in feed expected to produce a 50% incidence) in order to facilitate comparison of pure 2.3.7.8- TCDD with the Binghamton soot. The ED50 fol lowing 90-day exposure to the soot was calculated using previously reported incidence data (39) by a modification of the method of Bliss (36). A corresponding ED50 for subchronic 2,3,7,8-TCDD exposure could not be calcu lated from results of the present study since a range of incidences was not obtained. For purposes of compar ison, 76 ppt was assumed to be the ED50 dose level for 2.3.7.8- TCDD, since 50% of female animals at this dose level exhibited inclusions at a severity grade of +1 or greater as previously defined (34,35). The 2,3,7,8TCDD equivalent concentration (ppm) of the soot was then calculated as: 2.3.7.8- TCDD-equivalent concentration (|xg/g) _ EDsqfor 2,3,7,8-TCDD (ppt) ED50 for soot (ppm) Table 4. Calculated 2,3,7,8-TCDD-equivalent concentrations of the Binghamton soot for various dose-related endpoints following subchronic exposure.* End point Sex Method0 2,3,7,8-TCDD equivalent concen f tration in soot, ppm Relative thymus weight (decrease) % of initial body weight (decrease) Serum triglycerides (increase) Serum ALT (decrease) Hepatocellular cytoplasmic inclusion bodies Mortality M Linear M Linear M Linear F Log F 'ED M LD50 0.951 0.994 0.998 0.960 ___ 19d 21d 5d 18* 10 (4-28)f 2 (1-3/ *Response data from the previously reported Binghamton soot (34), and 2,3,7,8-TCDD (35) subchronic studies 2,3,7,8-TCDD-equivalent concentrations for the soot. bMethod by which 2,3,7,8-TCDD dose-response standard curve was constructed: Linear = linear regression analysis of dose level (ppt) vs. response; LOG = linear regression analysis of log dose level (ppt) vs. response; EDM = calculation of ED, dose level by method of Carmines et al. (36); LD, = calculation of LD. for subchronic ex posure by method of Carmines et al. (36). ' Correlation coefficient from linear regression analysis. dCalculated using response data from 1.9 ppm Binghamton soot dose level. 'Calculated using response data from 3.9 ppm Binghamton soot dose level. f95% confidence limits. Finally, an effective LD50value for prolonged 2,3,7,8TCDD exposure was similarly derived by using total dose versus mortality data from a recovery experiment in the previous investigation (34). This was compared with corresponding calculations from the soot study to yield an equivalent concentration (ppm) by using Equa tion 1. Table 4 summarizes the results of calculations of the 2.3.7.8- TCDD equivalent concentration of Binghamton soot based on dose-response data from previous inves tigations (34,35). Excellent dose-response correlations (r > 0.950) were obtained for the continuous data end points (relative thymus weight, %of initial body weight, serum triglycerides, and serum ALT) after 2,3,7,8TCDD exposure. Log transformation of the 2,3,7,8TCDD feed concentrations was required for a good fit only in the case of serum ALT levels. Calculated "2,3,7,8-TCDD-equivalent" concentrations for the Binghamton soot were in good relative agreement for three of the four end points, with values ranging from 18 to 21 ppm (p.g 2,3,7,8-TCDD/g soot). Calculations based on serum triglyceride levels resulted in a some what lower value of 5 ppm. Data on the incidence of hepatocellular cytoplasmic inclusion bodies in female guinea pigs fed Binghamton soot were employed to determine a value of 7.6 ppm (2.7 to 21.6 ppm, 95% confidence range) as the ED50 dose level of soot for this lesion. Using a value of 76 ppt for the ED50 dose level following exposure to pure 2.3.7.8- TCDD, the "2,3,7,8-TCDD-equivalent" concen tration of the soot was determined to be 10 ppm (4 to 28 ppm, 95% confidence limits). A similar approach was employed for calculation of an effective LD50 for soot exposure, except that total soot consumption was used instead of soot concentration in the feed. This LD50dose was determined to be 383 mg soot/kg (186 to 790 mg/ kg, 95% confidence range). Comparison with the effec tive LD50 of 0.8 p.g 2,3,7,8-TCDD/kg (0.6 to 1.1 p,g/kg, 95% confidence limits) yielded a value of 2 ppm (1 to 3 ppm, 95% confidence limits) for the 2,3,7,8-TCDDequivalent concentration of the Binghamton soot. Comparison of Chemical and Biological Calculations The most direct test of the assumptions underlying the calculation of 2,3,7,8-TCDD equivalents from chem ical data and published acute oral LD50 values is com parison with the biological calculation based on the acute oral LD50 of the mixture and of 2,3,7,8-TCDD itself. The chemical data-based 2,3,7,8-TCDD equivalents es timate of 22 ppm is in reasonable agreement with the experimentally determined value of 58 ppm. From a practical perspective, however, the predic tions of magnitude of acute toxicity expressed by a mix ture is of less interest than prediction of the mixture's subchronic and chronic toxicity. The six subchronic end points listed in Table 4, although not necessarily rep resenting adverse effects of great clinical importance, nevertheless serve as sensitive indicators of exposure. 19446 -a- 226 EADON ET AL. The calculated equivalent concentration values varied over an approximately tenfold range (2 to 21 ppm), de pending upon the end point chosen. Such a range is not unexpected, since various soot components might be more or less effective than 2,3,7,8-TCDD in producing a specific end point. While the soot appeared to be very effective in causing alterations in thymus weight, body weight, and serum triglyceride levels, it was less ef fective at causing death. Taken together, these findings suggest that biologically based calculations of 2,3,7,8TCDD equivalent concentrations are best based on a range of biological end points. It is notable that for the Binghamton soot, use of the acute oral LD50 as the end point resulted in higher ap parent toxicity (i.e., the highest 2,3,7,8-TCDD equiv alent concentration) than the various subchronic end points. Thus, use of the acute LDso or use of the chemical data to perform a calculation may provide a conservative estimate of the acute or subchronic activity of the mixture. However, the observation that the 2,3,7,8-TCDD equivalent concentration varied by about an order of magnitude when calculated by using various subchronically induced end points is an indication of the uncertainty associated with the use of any single bio logical end point or of chemical analysis data alone to estimate a mixture's potency. Conclusions The acute oral LD50 of a soot sample containing a complex mixture of PCDDs, PCDFs, and PCBEs has been shown to correspond to that of a hypothetical ma terial containing only 2,3,7,8-TCDD at 58 ppm. The cor responding 2,3,7,8-TCDD equivalent concentration cal culated by using various dose-related subchronic end points ranged from 2 to 19 ppm. Conversely, using ex isting literature data to estimate the potency of indi vidual congeners and the results of chemical analyses on the soot, it can be "predicted" that the soot contains ca. 22 ppm 2,3,7,8-TCDD equivalents. In light of un certainties in the chemical analyses and toxicological determinations, as well as uncertainties in the sample congener toxicity data upon which this calculation re lies, the agreement between observed and predicted concentrations is good. However, the substantial vari ation in the results of the "equivalents" calculation de pending on endpoint chosen clearly shows that any de term ination based on a single end point or any calculation based on chemical concentrations has a sub stantial uncertainty associated with it. Nevertheless, these uncertainties will often be small in comparison to the many uncertainties and approximations inherent in a risk assessment for a single compound, such as 2,3,7,8TCDD. It appears likely that this approach, or a con ceptually similar one, will find increasing use in the evaluation of the hazard posed by complex mixtures. REFERENCES 1. Kimbrough, R. D., Falk, H., Stehr, P., and Fries, G. Health implications of 2,3,7,8-tetrachiorodibenzodioxin (TCDD) contam ination of residential soil. J. Toxicol. Environ. Health 14-47_a>> (1984). 2. Bowes, G. W., Mulvihill, M. J., Simmeit, B. R. T. Identification of chlorinated dibenzofurans in American polychlorinated diben zofurans. Nature 256: 306-307 (1975). 3. Tieman, T. 0 ., Taylor, M. L., Garrett, J. H., VanNess, G. F., Solch, J. G., Deis, D. A., and Wagel, D. J. Chlorodibenzodioxins' chlorodibenzofurans, and related compounds in effluents front combustion processes. Chemosphere 12: 595-606 (1983). 4. Bumb, R., Crummett, W., Cutie, S., Gledhill, J., Hummel, R., Kagel, R., Lamparski, L., Luoma, E., Miller, D., Nestrick, T.', Shadoff, L., Stehl, R., Woods, J. Trace chemistries of fire: a source of chlorinated dioxins. Science 210: 385-390 (1980). 5. Stalling, D. L., Smith, L. M., Petty, J. D., Hogan, J. W., John son, J. L., Rappe, C., and Buser, H. R. Human and Environ mental Risks of Chlorinated Dioxins and Related Compounds. Plenum Press, New York, 1983. 6. Rappe, C., Nygren, M., Hansson, M., Bergquist, P. A., Marklund, S., Lindstrom, G., Domellof, L., Hardell, L., and Olsson, M. Identification of 2,3,7,8-substituted polychlorinated dioxins (PCDDs) in environmental and human samples. Chemosphere, in press. 7. Buser, H. R. Formation of polychlorinated dibenzofurans (PCDFs) and dibenzo-p-dioxins (PCDDs) from the pyrolysis of chlorobenzenes. Chemosphere 6: 414-424 (1979). 8. Buser, H. R., Bosshardt, H. P., and Rappe, C. Formation of polychlorinated dibenzofurans (PCDFs) from the pyrolysis of PCBs. Chemosphere 7: 109-119 (1978). 9. United States Environmental Protection Agency. Health As sessment Document for Polychlorinated Dibenzo-p-dioxins. Ex ternal Review Draft, EPA 600/8-84-014A. 10. Kociba, R. J., and Cabey, O. Comparative toxicity and biological activity of chlorinated dibenzo-p-dioxins and furans relative to 2378-tetrachloro-dibenzo-p-dioxins (TCDD). Chemosphere, in press. 11. Smith, R. M., O'Keefe, P. W., Hilker, D. R., Jelus-Tyron, B. L., and Aldous, K. M. Analysis for 2,3,7,8-tetrachlorodibenzofuran and 2,3,7,8-tetrachlorodibenzo-p-dioxin in a soot sample from a transformer explosion in Binghamton, N.Y. Chemosphere, 12: 325-332 (1982). 12. Stalling, D. Chlorinated Dibenzofurans and Related Compounds in Soot Formed in a Transformer Fire in Binghamton, N.Y., Preliminary Report. Columbia National Fisheries Research Lab oratory, U.S. Fish and Wildlife Service, 1984. 13. Rappe, C., Marklund, S., Bergquist, P. A., and Hansson, M. Polychlorinated dibenzo-p-dioxins, dibenzofurans and other po lynuclear aromatics formed during incinerations and PCB fires. In: Chlorinated Dioxins and Dibenzofurans in the Total Environ ment, Vol. 1. Butterworth's Ann Arbor Science, Boston, 1983, pp. 99-124. 14. United States Environmental Protection Agency. Interim Eval uation of Health Risks Associated with Emissions of Tetrachlorinated Dioxins from Municipal Waste Resource Recovery Facil ities. November 1981. 15. Eadon, G., Aldous, K., Frenkel, G., Gierthy, J., Hilker, D., Ka minsky, L., O'Keefe, P., Silkworth, J. and Smith, R. Comparison of Chemical and Biological Data on Soot Samples from the Bing hamton State Office Building. New York State Department of Health Report, 1982. 16. Gravitz, N., Fan, A., and Neutra, R. R. Interim Guidelines for Acceptable Exposure Levels in Office Settings Contaminated with PCB and PCB Combustion Products. California Department of Health Services Report, 1983. 17. Assessment of Potential Public Health Impacts Associated with Predicted Emissions of Polychlorinated Dibenzodioxins and Poly chlorinated Dibenzofurans from the Brooklyn Navy Yard Re source Recovery Facility. Fred C. Hart Associates, Inc., August 17, 1984. 18. Bellen, J. S., and Barnes, D. B. Health hazard assessment for chlorinated dioxins and dibenzofurans. Presented at Symposium on Advances in Health Risk Assessment for Systemic Toxicants and Chemical Mixtures. Cincinnati, Ohio, 1984. 19. Silkworth, J. B., McMartin, D., DeCaprio, A. P., Rej, R., Kumar, 19447 TCDD EQUIVALENCY OF COMPLEX MIXTURES 227 S., Kaminsky, L. Acute toxicity in guinea pigs and rabbits of soot from a polychlorinated biphenyl-containing transformer fire. N. Y. State Dept, of Health Report, January 6, 1982. 20. Moore, J. A., McConnell, E. E., Dalgard, D. W., and Harris, M. W. Comparative toxicity of three halogenated dibenzofurans in guinea pigs, mice and rhesus monkeys. Ann. N. Y. Acad. Sci. 320: 151-163. 21. McConnell, E. E., Moore, J. A., Haseman, J. K., and Harris, M. W. The comparative toxicity of chlorinated dibenzo-p-dioxins in mice and guinea pigs. Toxicol. Appl. Pharmacol. 44: 335- 356 (1978). 22. Kawamura, K., Sato, R., and Kashima, M., Acute and chronic toxicity of 1,3,6,8-tetrachlorodibenzodioxin (1,3,6,8-TCDD). Oyo Yakuri 25: 703-711 (1983). 23. McKinney, J., and McConnell, E. E. Chlorinated dioxin and re lated compounds (0. Hutzinger, R. W. Frei, E. Merian, and F. Pocciara, Eds.), Pergamon Press, New York, 1981. ` 24. Silkworth, J., McMartin, D., DeCaprio, A., Rej, R., O'Keefe, P. and Kaminsky, L. Acute toxicity in guinea pigs and rabbits of soot from a polychlorinated biphenyl-containing transformer fire. Toxicol. Appl. Pharmacol. 65: 425-439 (1982). 25. Mason, G., Sawyer, T., Keys, B., Bandiera, S., Romkes, M., Piskorska-Pliszcynska, J., Zmudzka, B., and Safe, S. Polychlo rinated dibenzofurans (PCDFs): correlation between in vivo and in vitro structure-activity relationships. Toxicology 37: 1-12 (1985). 26. Bandiera, S., Sawyer, T., Romkes, M., Zmudzka, B., Safe, L., Mason, G., Keys, B., and Safe, S. Polychlorinated dibenzofurans (PCDFs): effects of structure on binding to the 2,3,7,8-TCDD cytosolic receptor protein, AHH induction and toxicity. Toxicol ogy 32: 131-144 (1984). 27. Bradlaw, J. L., and Ca3terline, J. L. Induction of enzyme activity in cell culture: a rapid screen for detection of planar polychlori nated organic compounds. J. Assoc. Off. Anal. Chem. 62: 904 (1979) . 28. Poland, A., and Knutson, J. C. 2,3,7,8-Tetrachlorodibenzodioxin and related halogenated aromatic hydrocarbons: examination of the mechanisms of toxicity. Ann. Rev. Pharmacol. Toxicol. 22: 517 (1982). 29. Poland, A., Greenlee, W. F., and Kende, A. S. Studies on the mechanism of toxicity of the chlorinated dibenzo-p-dioxin and re lated compounds. Ann. N.Y. Acad. Sci. 320: 214 (1979). 30. National Toxicology Program, NTP/NCI Report No. 122, NIH, Bethesda, MD, 1979. 31. National Toxicology Program, NTP/NCI Report No. 123, NIH, Bethesda, MD, 1979. 32. O'Keefe, P., Silkworth, J., Gierthy, J., Smith, R., DeCaprio, A., Turner, J., Eadon, G., Hilker, D., Aldous, K., Kaminsky, L., and Collins, D._ Chemical and biological investigations of a trans former accident at Binghamton, NY. Environ. Health Perspect. 60: 201-209 (1985). 33. Knutson, J. C., and Poland, A. Keratization of mouse teratoma cell line XB produced by 2,3,7,8-tetrachlorodibenzodioxin: an invitro model of toxicity. Cell 22: 27-36 (1980). 34. DeCaprio, A. P., McMartin, D. N., Silkworth, J. B., Rej, R., Pause, R., and Kaminsky, L. Toxicity in guinea pigs of soot from a polychlorinated biphenyl-containing transformer fire. Toxicol. Appl. Pharmacol. 6: 308-322 (1983). 35. DeCaprio, A. P., McMartin, D. N., O'Keefe, P. W., Rej, R., Silkworth, J. B. and Kaminsky, L. S. Subchronic oral toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the guinea pig: compari sons with a PCB-containing transformer fluid pyrolysate. Fundam. Appl. Toxicol. 6: 454-463 (1986). 36. Carmines, E. L., Carchman, R. A., and Borzelleca, J. F. A method for the evaluation of dose-effect data utilizing a pro grammable calculator. J. Environ. Pathol. Toxicol. 4: 23-30 (1980) . 19448 . TOXICOLOGY AND APPLIED PHARMACOLOGY 8 3 , 4 4 5 - 4 5 5 (1 9 8 6 ) Immunosuppression without Liver Induction by Subchronic Exposure to 2,7-Dichlorodibenzo-p-dioxin in Adult Female B6C3F1 Mice Michael P. Holsapple,*-1 J. Ann McCay,* and Donald W . Barnes! D epartm ent o f Pharm acology a n d Toxicology. M edical College o f Virginia. Virginia C om m onw ealth University. Richm ond. Virginia 23298: a n d t D epartm ent o f Pharmacology. School o f M edicine. East Carolina University. Greenville. North Carolina 27834 R eceived J u ly 29. 1985: accepted D ecem ber 16. 1985 Im m unosuppression w ithout Liver Induction by Subchronic Exposure to 2,7-DichIorodibenzop -d io x in in A d u lt F em ale B 6C 3FI M ice. Holsapple, M. P,, McCa y , J. a ., a n d Ba r n e s, D. W. (1986). Toxicol. Appi. Pharmacol. 83, 445-455. The overall objective o f this investigation was to begin to characterize the structure-activity relationship associated with dioxin-induced suppression o f hum oral im m unity. Subchronic exposure (14 days) to 2.3.7.8-tetrachlorodibenzo-p-dioxin (TCDD), the prototype o f the class, produced a suppression o f the antibody responses to both sheep erythrocytes, a T-dependent antigen, and dinitrophenyl-Ficoil, a T-independent antigen. Surprisingly, sim ilar results were observed with 2.7-dichlorodibenzo-p-dioxin (D C D D ), a dioxin congener lacking affinity for the A h receptor. In co n trast subchronic exposure to octachiorodibenzop-dioxin (OCDD). another dioxin congener w ithout affinity for the A h receptor, was devoid o f activity. Subchronic exposure to 2,3,7,8-TCDD, but not 2,7-D CD D , produced an induction of several liver param eters including: liver w eight am ount of m icrosom al protein, am ount o f cy tochrom e P-450, activity o f aminopyrine-iV-demethylase and activity o f aryl hydrocarbon hy droxylase. Subchronic exposure to 2,3.7.8-TCDD or 2,7-DCDD produced no m arked changes in thym us weighL Acute exposure to 2,3.7.8-TC D D also produced suppression o f the antibody response in th e absence o f effects o n th e thym us. 3 1986 Academic Pros. ine. 2 .3 .7 .8 - T etrach lorod ib en zo-p -d ioxin (T C D D ) c o m p e te n c e (H o lsa p p le et a i, 1984a). Sur represents the prototype and m ost potent prisingly, this was the first report o f im m u m em ber o f a fam ily o f low -m olecular-w eight nosuppression by any d ioxin congener other halogenated arom atic hydrocarbons. The than T C D D . H owever, it is im portant to em m echanism o f action by T C D D is at present phasize that other laboratories have tested unknow n, although m any o f the actions as n ond ioxin halogenated arom atic hydrocar sociated w ith exposure to T C D D are suspected bons. T hese results dem onstrated that 2.3.7,8- to be recep tor m ed ia ted . T h e Ah recep tor has d ib en zo fu ra n co u ld su p p ress th e a n tib o d y re been im p lica ted in a n u m b e r o f a c tio n s pro sp o n se (V e c c h i et a i, 1983) an d that A roclor duced by T C D D and related congeners 1254, a com m ercial polychlorinated biphenyl (G reenlee and N eal, 1984). Our interest in the (PCB) m ixture, or 3,3',4,4'-tetrachlorobi- chlorinated dibenzo-p-dioxins is in their effects phenyl, but not 2,2'.4,4',6.6'-hexachlorobi- on im m u n ocom p ten ce. W e have recently re phenyl, cou ld suppress the generation o f cy ported that su b c h r o n ic ex p o su re (1 4 d ays) to to to x ic T cells (Clark et a i. 1983). T h e ob jec 1.2.3.6.7.8- h exach lorodib en zo-p -d ioxin was tive o f the present investigation was to capable o f suppressing hum oral im m u n o- determ ine the effects o f 2,7-dichlorodibenzo- p-dioxin (D C D D ) on the hum oral antibody 1To whom reprint requests should be made. response and to com pare these effects with ; 0041-008X /86 S3.00 Copyright < 1986 by Academic Press. Inc. All rights ol reproduction in any form reserved. 19449 446 HOLSAPPLE. MC CAY. AND BARNES T C D D . P reviou s w ork b y G r ee n le e et al. (1 9 8 4 ) a n d b y P olan d et al. (1 9 7 6 ) h as sh o w n that 2 ,7 -D C D D was devoid o f affinity for the Ah recep tor a n d o f the a ctiv itie s attribu ted to binding to that receptor. In this investigation, w e also determ ined the effects o f exposure to octachlorodibenzo-p-dioxin (O C D D ) on the a n tib od y resp onse. P oland et al. (1 9 7 6 ) sh ow ed th at O C D D also lacked affinity for th e Ah re ceptor. T hese results are significant because the positive effects by 2,7-D C D D indicate that a chlorinated dibenzodioxin can produce ef fects (at least on im m u n ocom petence) which are n o t asso cia ted w ith th e Ah receptor. METHODS A nim als. Fem ale (C57BL/6 X C3H)F1 (B 6C 3FI) mice were purchased from Litton Bionetics. Inc. (Frederick. Md.). M ice arrived at 5-6 weeks o f age, were quarantined for at least I week, and were used for experim entation when they were 17-20 g in weight. U pon arrival, mice were random ized, weighed, and housed four m ice per cage. Individual mice within a cage were identified by earpunching. The anim al quarters were m aintained at 2124C and 40-60% relative hum idity. A 12-hr light/dark cycle was used. Chemicals. 2.3.7.8-TCD D was obtained through the N ational Toxicology Program and Bio-Rad Laboratories (Rockville C entre. N.Y.). 2.7-D CD D was a gift from Dr. W illiam F. G reenlee (CUT. Research Triangle Park. N.C.). O C D D was a gift from Dr. A nne N. T ucker (N IEH S. Re search T riangle Park). C o m oil was previously dem on strated to be an effective vehicle for the studies with 1,2,3.6,7,8-hexachlorodibenzo-p-dioxin (H C D D ) (Holsapple et al.. 1984a) and was th e vehicle for this investi gation. The route of exposure was by oral gavage. Range finding stu d ies d e term in ed th a t 14 daily exposures to 1 (ig/kg T C D D consistently produced a significant suppres sion o f the antibody response (data not shown), so a con centration o f 0.1 jig/m l T C D D was used in these studies. It is im portant to em phasize that the dose relatedness o f the suppression o f im m unocom petence by TCD D has been reported by oth er investigators (Vecchi et al., 1983; D ean and Lauer, 1984). O u r rationale for using a single con centration o f TC D D was based on the fact that TCDD was to serve as the prototype o f the class for com parative purposes to the other dioxin congeners. The concentrations o f 2.7-D C D D were 0.01. 0.1. and I jig/m l an d o f O CD D were 0.1 and I pg/m l. F o r the subchronic studies, mice w ere gavaged daily w ith 0.1 m l / 10 g body w t o f th e vehicle o r o n e o f th e d io x in congeners for 14 consecutive days unless otherwise stated. The various param eters were m easured 24 hr after the final exposure to the vehicle or dioxins, except for the acute studies with TCD D as de scribed below. The num ber o f anim als exposed to each treatm ent is indicated in the appropriate tables describing the results. B ody and organ eights and histopathology. Twentyfour hours after the final chem ical (or vehicle) treatm ent, mice were weighed and body weights recorded. Gross pathological exam inations were perform ed on all mice. The organs (brain, liver, spleen, lungs, thymus, and kid neys) were removed, trim m ed, and weighed. M aterial for histological analysis consisted o f organs removed at the tim e o f necropsy and consisted o f liver, spleen, thym us, kidney, a n d lung. These organs were fixed in 10% n e u tral buffered Form alin. T he tissues were sectioned at 6 and stained with hem atoxylin and eosin. The slides were stud ied by a pathologist who was unaw are o f the treatm ents. Hepatic microsomal parameters. Twenty-four hours af ter the final exposure to the vehicle. TCD D . o r DCDD, the anim als were killed and the livers were rem oved, weighed, rinsed, and hom ogenized a t 4 C in 9 vol o f 1.15% KC1 co n tain in g 0.1 M p o tassium phosphate buffer, pH 7.4. The homogenate was centrifuged at 9000g for 20 min, an d the supernatant fraction w as recentrifuged a t 100.000j for 1 hr in a Beckman M odel L5-50 ultracentrifuge (Beck m an Instrum ents Inc.. Fullerton. Calif.). The m icrosom al pellet was layered w ith 1.5% K.C1 an d frozen im m ediately at -7 0 C . W ithin 24 to 48 hr after freezing, the m icrosomes were resuspended in 0.05 m potassium phosphate buffer, pH 7.4 (0.5 g o f liver/m l), and sam ples were taken for study. Microsomal protein was assayed by the m ethod o f Low ry et al. (1951). T h e c o n te n t o f cytochrom e!s) P450 was determ ined by m easuring the carbon m onoxide difference spectrum in dithionite-reduced m icrosom es (O m ura and Sato, 1964). An extinction coefficient o f 9 1 cm "' mM"' was used to calculate /M 5 0 co n ten t Aminopyrine-iV-demethylase activity was determ ined by m ea suring formaldehyde production by the Nash reaction as described by Cochin and Axelrod (1959). Arvl hydrocarbon(benzo(a]pyrenej hydroxylase activity was evaluated spectrofluorom etrically as described by Yang et al. (1978) using 3-hydroxybenzo(a]pyrene as the standard. A ntibody response to a T-dependent antigen. On Day 11 o f subchronic exposure to the various treatm ents, mice were given injections o f 5 x 10a SR B C . ip. O n Day 15, the anim als were killed, and the spleens dissociated into single ceil suspensions by pressing the spleen between two frosted microscope slides. Spleen cells from individual an imals were assessed for the num ber o f anti-sheep red blood cells (SRBC) A FC-producing IgM in a m odified Jem e plaque assay (Bullock and Molter. 1972) as described pre viously (H olsapple et al.. 1984a). AFC (plaques) were counted a t 6 .5 x m agnification using a Bellco plaque viewer (Bellco Glass. Inc.. Vineland. N.J.). Spleen cell suspensions were counted with a Coulter C ounter (Model ZBI: Coulter Electronics. Inc.. Hialeah. Fla.) and AFC results were cal culated as A FC/10` recovered cells and AFC/spleen. IM M UNOSUPPRESSION BY 2,7-DCDD 447 It is im portant to em phasize that we have traditionally determ ined the effects o f xenobiotics on the IgG response as well as IgM antibody production an d on both Day 4 and Day 5 after im m unization. O ur decision to m easure only the D ay 4 IgM response is based on p revious results which showed that subchronic exposure to l.2.3,6.7.8H C D D suppressed all param eters, b u t th at the Day 4 IgM response was the most sensitive indicator (H olsapple et al.. 1984a). Antibody response to a T-independent antigen. This re sponse was determ ined as previously reported (Bide el al.. 1984). The thym ic-independent antigen, dinitrophenyl (D N PJ-Ficoll. was prepared using C N B r-Ficoil (400.000 M ,) and 2.4-dinitrophenyl-lysine (all from Sigma C hem icals) following the m ethod o f K laus et al. (1976). O n Day 11 o f subchronic exposure to eith er th e vehicle. T C D D . o r DCDD, mice received 20 ug o f D N P-Ficoll intrave nously in a volum e o f 0.2 ml. O n Day 15, the anim als were killed and the Day 4 IgM antibody response was measured basically as described above. The anti-D N P an tibody-form ing cell response was determ ined for each an imal in a m odified Jerae plaque assay using trinitrophenyl (TNP)-coupled SRBC as indicator cells. The SRBCs were densely coupled with trinitrophenyl following the m ethod o f Rittenberg and Pratt (1969). Polyclonal antibody response to lipopolvsaccharine. This response was determ ined as previously reported (Holsapple et al.. 1984b). A nim als were exposed to either th e vehicle. T C D D . o r D C D D daily for either 14 days (E xp erim en t 1) o r 5 days (Experim ent 2). Spleen cells w ere collected from individual anim als 24 hr after the final treatm ent and as sayed in q u ad ru p licate cultures (5 X IO6 cells/1.0 m l/culture well) in 24-well cluster plates (Falcon). Spleen cells were cultured with 200 jig/m l lipopolysaccharide (LPS) for 48 hr. It is im portant to em phasize th at these anim als receive no im m unological stim uli during the duration o f the chem ical exposure. Following culture, the cells were harvested using gentle resuspension w ith a Pasteur pipet. centrifuged, and resuspended to a suitable volum e for the determ ination o f antibody-form ing cells. T he antibody form ing cell response was determ ined using a modified Jem e plaque assay and TN P-coated sheep erythrocytes as target cells. Effects o f acute exposure to TC D D on the antibody re sponse to a T-dependent antigen an d thym ic weights. In light o f the traditional involvem ent o f the A h receptor in dioxin-induced thym ic atrophy, the lack o f effect on the thym us by subchronic exposure to T C D D was surprising. In order to rule ou t the possibility th a t the repeated ex posures (i.e., daily for 14 co nsecutive days) m ay have c o n tributed to the lack o f effect, a study was conducted to determ ine the effects o f a single exposure to T C D D . Mice were gavaged with either com oil (vehicle) o r 0.005,0.01. 0.05, 0.1, or 0.2 ug/m l T C D D (i.e., to give doses from 0.05 to 2.0 Mg/kg). Seven days later, all m ice w ere im munized w ith SRBCs as described above. Four days later, all mice were killed and the spleens were rem oved, weighed. and assayed for the num ber o f IgM antibody-producing cells as described above. In addition, the thym uses were removed and weighed. Statistics. For results o f all studies, a D unnett's t test was perform ed if a one-way analysis o f variance o f the m eans showed treatm ent effects. RESULTS Body and organ weights and histopathology. The results in Table 1 indicate that 14 daily exposures to 2.7-DCDD were devoid o f any effects on body weights or selected organ weights. For the sake o f brevity, we have only shown the effects on the thymus and the liver. Subchronic exposure to 2,3,7,8-TCDD, but not 2,7-DCDD, produced a significant in crease in liver weight. It is important to note that neither dioxin congener caused much of a change in thymus weight, although there was a significant reduction by TCDD, but only when reported as a percent o f body weight. The histopathology of the thymus, spleen, lungs, and kidneys o f mice receiving TCDD was normal. In agreement with the effects on liver size, the livers o f TCDD-treated mice showed focal areas o f hydropic degeneration. Hepatic microsomal parameters. As sug gested by the increase in liver weight shown in Table 1 and previous results described in the discussion, subchronic exposure to 2,3,7,8TCDD produced a marked induction o f a number o f microsomal parameters (Table 2). In contrast, subchronic exposure to 2.7DCDD was devoid of any changes. Antibody response-- T-dependent antigen. Two experiments were performed to deter mine the effects o f select dioxin congeners on the Day 4 IgM antibody response to a T-de pendent antigen. SRBC. In the first experi ment (Table 3, Section 1), subchronic exposure to 2,3,7.8-TCDD and to 2.7-D C DD produced a significant suppression at all concentrations tested. These effects occurred in the absence of any effects on spleen cell number and were therefore comparable when expressed as either IgM AFC/spleen or IgM AFC/106 recovered cells. The lack o f effect on spieen cell number 448 HOLSAPPLE. MC CAY. AND BARNES TABLE I Body and Select Organ Weights of B6C3FI Mice Exposed to 2.7-D C D D or 2.3,7.8-T C D D for 14 Days P a ram eter tV Vehicle (com oil) (9) 2,7-D C D D 0g/kg) 0.1 1.0 10.0 (5) (5) (9) 2 .3 .7 .8 -T C D D (dgAg) 1.0 (9) Body w eight (g) 17.5 * 0.3 18.4 * 0.4 Liver (mg) % Body weight 879.0 * 35.8 5.08 0.15 866.6 * 33.0 4.70 * 0.14 Thym us (mg) 65.2 * 2.7 % Body weight ' - 0 .3 8 * 0 .0 1 - 62.0 4.6 0 .3 4 * 0.02 .17.6 * 0.2 837.2 * 15.0 4 .9 8 * 0.08 63.6 + . 4.1 0 .3 6 * 0.03 18.6 * 0.3 944.3 * 26.9 5 .0 4 * 0.14 73.3 * 1,6 0 .3 9 * 0.01 18.6 * 0.3 1252.8 3 6 .8 * * 6.68 * 0.12** 54.6 4.6 0.29 * 0.02** Sole. B6C3FI female mice were adm inistered either vehicle (corn oil) o r 2,7-DCDD at one o f the designated con cen tratio n s. o r 2.3.7.8-T C D D daily for 14 days by gavage. T w en ty -fo u r hours after th e last treatm ent, m ice were necropsied. and the organs listed above removed, trim m ed o f fatty tissue, and weighed. Brain, spleen, lungs, and kidney were sim ilarly exam ined and there were no treatm ent-related effects. T he values represent the x SE derived from the num ber o f anim als indicated in parentheses. Values significantly different from vehicle control at p < 0.05 are indicated by an asterisk, while those significantly different at p < 0.01 are noted by a double asterisk, as determ ined by D unnett's t test. TABLE 2 Hepatic Microsomal Parameters in Female B 6C 3FI Mice Exposed to 2.7-D C D D and 2,3.7.8-T C D D for 14 Days 2.7-DCDD (ng/kg) 2.3.7.8-TCDD Parameter Vehicle O.i 1.0 10.0 1.0 jig/kg iV 9 5 5 9 9 Liver weight (mg) MicrosomaJ protein (mg/g liver) Cytochrome ?-450 (nmo!/mg protein) (nmol/g fiver) Glutathione imol/g liver Aminopyrine*<Vdemethyiase nmol/min/mg prot nmol/min/g liver Arylhydrocarbon hydroxylase nmol/min/mg prot nmol/min/g liver 874.4 * 40.3 16.0 0.2 0.932;i 0.03 15.0 0.6 8.27 0.26 9.09 0.33 145.8 6.42 0.54 i 0.02 8.62 0.35 866.6 * 33.0 14.3 * 0.6 0.934 0.02 13.3 * 0.6 8.59 * 0.38 9.28 * 0.54 131.7 * 4.82 0.63 i 0.03 8.95 * 0.32 873.2 * 15.0 15.2 * 0.3 0.865 * 0.02 13.2 t 0.6 8.30 * 0.20 7.30 0.47 110.9 6.16 0.61 0.03 9.21 0.46 945.1 * 26.8 15.4 s 0.6 0.903 r 0.04 13.9 0.7 8.20 0.32 8.41 0.56 130.0 10.5 0.52 = 0.02 8.04 z 0.34 1253.0 i 36.9** 21.7 * 0.5** 1263 = 0.09" 49.1 2.1" 9.69 i 0.28* 13.99 = 0.46" 303.3 * 9.56" 2.33 i 0.14" 50.15 a *>5* Sole. B6C3F1 female mice wereadministered DCDD or TCDD or com oil daily for 14 days. On Day 15. the livers were aseplically removed. The values represent the x * SEderived from the number ofanimals indicated at the top ofeach column, "p < 0.05 and mmp <0.01 as determined by Dunnett's r test. ..i IM M UNOSUPPRESSION BY 2.7-DCDD 449 table 3 Day 4 igM antibody Response to a T-Dependent Antigen. SRBC. by Female B 6C 3FI Mice Exposed to D C D D , T C D D . or O C D D for 14 Days Treatm ent N um ber o f spleen cells (X 1 0 ') IgM AFC/spleen (X 103) IgM A FC/10` recovered cells I. V ehicle (co m oil) 1.0 Mg/kg T C D D 0.1 fig/kg DCD D 1.0 fig/kg D C D D 10.0 jig/kg D CD D 11. V ehicle (co m oil) 1.0 jig/kg T C D D 1.0 /ig/kg O C D D 10.0 iig/kg O CD D 13.72 + 0.57 12.44 + 0.84 13.98 + 1.26 15.47 0 .6 7 15.38+ 0.96 16.01 + 0 .9 7 14.14 + 1.14 16.44 0.94 17.85 0 .9 9 343.8 + 41.0 69.2 22.3** 169.4 4 5 .4 " 157.9 + 3 4 .7 " 129.1 + 1 6 .5 " 205.7 29.2 34.9 7.7* 220.3 36.6 192.8 17.3 2510.1 282.3 533.1 1 5 7 .3 " 1 194.4 270.2** 1015.9 1 9 9 .3 " 843.8 + 9 9 .2 " 1269.4 132.7 278.7 5 4 .8 " 1349.4 2 1 6 .4 1082.4 82.8 Mote. Fem ale B6C3F1 m ice (five per tre a tm e n t group) were gavaged w ith vehicle (co m oil). T C D D , o r D C D D in c o m oil daily for 14 days in E xperim ent I an d w ith vehicle (c o m oil). T C D D , o r O C D D in c o m oil daily for 14 days in E x p erim en t II. In both ex perim ents, th e a n im a ls were im m u n iz e d w ith SR B C o n Day 11 o f exposure. O n D ay 15, the spleen cell IgM anti-SR BC A FC response was determ ined. V alues represent the x SE derived from five anim als per treatm ent group. Values significantly different from vehicle control at p < 0.01 are indicated by a double asterisk as d eterm ined by D unnett's l test. also indicates that neither d ioxin congener produced a decrease in the distribution o f lym phocytes to the spleen. T he possibility for alterations in the relative subpopulations o f im m u n ocom petent cells w ithin the spleen can n ot be estim ated by the present results. In the secon d experim ent (T able 3, S ection II), subchronic exposure to 2,3,7,8-T C D D pro duced an alm ost identical suppression o f the D ay 4 IgM response (i.e., com p are a 79% suppression in Section I w ith a 78% suppres sion in S ection II) even thou gh the v eh icle re sponses were different. It is im portant to em phasize that both vehicle responses are w ithin our historical controls as generated over the last 4 years, although the response in Section II is at th e low e n d o f th is range. In co n tra st with the p ositive effects by 2 ,3 ,7 ,8 -T C D D and 2,7-D C D D , subchronic exposure to O C D D was devoid o f activity. Antibody response-- T-independent antigen. As show n in Table 4, subchronic exposure to either 2,3,7,8-T C D D or 2 ,7-D C D D produced a significant suppression in the D ay 4 IgM an tibody response to a T -independent antigen, D N P -F ic o ll. In contrast w ith th e results in T able 3, there was a significant change in spleen cell num ber as the low dose o f 2,7D C D D produced a 24% increase. W e have no explanation for this effect, but it is im portant to em phasize that the average spleen weight o f this group o f m ice was 129.3 9.6, as com pared to 91.2 3.9 for the vehicle anim als. Polyclonal antibody response-- lipopoly saccharide. A s sh ow n in T a b le 5, the su p p res sion by subchronic exposure (14 days) to either T C D D or D C D D can be detected even when spleen cells are rem oved from treated anim als and im m unized in culture. Both the T C D D and the high dose o f D C D D produced a greater than 50% suppression o f the response o f LPS (m iddle colum n), which suggests functional im pairm en t o f the B lym p h ocyte. In contrast w ith the results show n in T ables 3 and 4, ex p osure to 1.0 Mg/kg D C D D for 14 d ays did not produce a significant suppression. In separate anim als, we attem pted a pre lim inary study with 5 days o f exposure to ei ther T C D D or the high d ose o f D C D D . T hese results are shown in the right-hand colu m n o f 194 53 I ........ f 450 HOLSAPPLE, MC CAY. AND BARNES TABLE 4 D a y 4 tgM A n t ib o d y Response to a T-In d e p e n o e n t a n t ig e n , D N P -F icoll, by Fem ale B6C 3F1 M ice Ex po sed t o D C D D o r T C D D for 14 Day s Treatm ent N um ber o f spleen cells (X 1 0 ') IgM AFC/spleen (X 101) IgM A FC /104 recovered cells Vehicle (com oil) 1.0 ug/kg T C D D 1.0 ug/kg D CD D 10.0 ug/kg D CD D 13.99 0.72 11.78 0 .5 7 17.41 0 .7 8 * 14.56 1.19 574.2 29.5 53.4 8.5** 516.6 4 7 .8 411.8 33.6" 4117.9 142.3 451.3 59.2" 3152.9 2 7 4 .7 " 2856.6 180.4" Note. Fem ale B6C3F1 m ice (six p er treatm ent group) were gavaged with vehicle (com oil), T C D D , o r D CD D in c o m oil daily for 14 days. O n D ay 11 o f exposure, ail m ice were im m unized w ith 20 D N P -F ic o ll. O n D ay 15. the spleen cell IgM anti-TN P A FC response was determ ined against T N P-haptenated SRBCs. Values represent the x SE derived from six anim als per treatm ent group. Values significantly different from vehicle control at p < 0.05 are indicated by a single asterisk, while those significantly different from vehicle control at p < 0.01 are indicated by a double asterisk, as determ ined by D unnett's t test. Table 5. T C D D produced a suppression after on ly 5 days o f exposure (62%) w hich was co m p a ra b le to that ob served after 14 d ays o f exp osu re (52%). In contrast, the suppression by D C D D was less after 5 days when com pared to the effect by 14 days o f exposure (i.e., TABLE 5 Polyclonal Antibody Response to Lipopolysac charide by Spleen Cells from Female B 6C 3FI Mice Exposed to D C D D or T C D D for 14 or 5 Days Treatm ent (#*g/kg) (gM A FC/104 recovered cells (14 days) IgM A FC /104 recovered cells (5 days) Vehicle (com oil) 840 19 542 21 23 and 56%, respectively) w hich suggested a possible tem poral relationship for the im m unotoxicity by D C D D . Immunosuppression-- acute exposure to TCDD. In light o f the m arginal effects that subchronic exposure to 2,3,7,8-T C D D pro d uced o n the th ym u s w eigh ts (T a b le 1), a stud y was conducted with acute exposure to T C D D . These results are sh ow n in T able 6. A s show n TABLE 6 Day 4 IgM Antibody Response to a T-Dependent Antigen and Thymus Weight of Female B6C 3FI Mice Exposed to a Single Injection of T C D D Treatm ent (ug/kg) IgM A FC /104 recovered cells Thymus weight 1.0 T C D D 1.0 D C D D 10.0 D C D D 402 l l " 737 80 372 18 203 28" ND 416 2 0 " Note. Fem ale B6C3FI mice (five per treatm ent group) were gavaged with vehicle (com oil), TCD D . or D CD D in c o m oil daily fo r eith e r 14 days (m iddle colum n) o r 5 days (right colum n). T w enty-four hours after the final ex posure, m ice were killed and spleen cell suspensions from individual m ice were aseptically prepared and cultured (5 X I04 cells/1.0 m l/culture well) in quadruplicate wells with 200 u g LPS in the presence o f 10% fetal ca lf serum . Cul tures were harvested after 48 hr and the num ber o f IgM antibody-form ing cells was determ ined. Values represent the m ean IgM A FC /104 recovered spleen cells SE. ND-- not done. Vehicle (com oil) 1495 304 51 3 0.05 TCD D 0.1 T C D D 0.5 T CD D 1.0 T C D D 2.0 TCD D 1558 = 230 1570 2 5 3 1104 47 826 32* 505 78* 60 5 49 5 57 5 53 7 54 2 Note. Female B6C3FI mice were gavaged with vehicle (co m oil) o r T C D D in corn oil o n D ay I. O n Day 8. the anim als were im m unized w ith SRB C. O n D ay 12. the spleen cell IgM anti-SRBC AFC response was determ ined and the spleen and thymuses were removed, trim m ed, and weighed. Values represent the ,v SE derived from five anim als per treatm ent group. Values significantly different from vehicle control at p < 0.05 are indicated by an as terisk. as determ ined by D u n n ett's t test. 13454 IMMUNOSUPPRESSION BY 2,7-DCDD 451 in the last colu m n , doses o f T C D D up to 2.0 Mg/kg had n o effect o n the th y m u s w eigh ts. This sam e concentration range did produce a dose-related suppression in the D ay 4 IgM an tibody response to a T -dependent antigen with significance noted at 1.0 and 2.0 ^ g/kg (45 and 66%, respectively). T h ese results in dicate that the marginal effects on the thym us show n in Table l are not an effect o f the repeated ex posures and suggest that the dioxin-induced suppression o f the T-dependent response can be com pletely separated from the dioxin-in duced thym ic atrophy. D ISC U S SIO N As recently reviewed by G reenlee and N eal (1984), 2.3,7,8-T C D D represents the proto type ligan d for th e Ah receptor. T h e Ah recep tor is suspected to be a protein w hich regulates the expression o f a gene battery which includes the structural genes for cytochrom e(s) P ,-450 (i.e., involved in the T C D D -ind uced liver in duction) and a second regulatory locus which controls a broader pleiotropic response (i.e., involved in action s w here T C D D is suspected as producing a change in cell proliferation and/ or differentiation). T his receptor has been im plicated in a num ber o f actions associated with exposure to T C D D and related congeners in cluding induction o f cytochrom e(s) i 'i-450 (Poland and G lover, 1974, 1975), thym ic atrophy (Poland and G lover, 1980), im pair m ent o f the ability o f thym ic epithelial cells to support the m aturation o f T lym phocyte precursors (G reen lee et al., 1985), su pp ression o f hum oral im m unity (i.e., generation o f an tibod y p ro d u c in g cells) (V e cc h i et al., 1983), and suppression o f cell-m ediated im m unity (i.e., generation o f cytotoxic T ceils) (Nagarkatti el al., 1984). T h e stu d ies w h ich have d em o n stra ted a p ossib le role b y th e Ah recep tor in the profile o f activity o f T C D D have capitalized on two im portant characteristics o f this locus. N eb ert et al. (1 9 7 2 ) reported that th e Ah locus is inherited as a sim ple au tosom al d o m inant trait. T h e genetic characteristics o f the locus have been particularly w ell worked out in m ice where the prototype responsive strains have been dem onstrated to be C 57B L /6 and C 3H , w hile the prototype nonresponsive strain is th e D B A /2 (P o la n d et al., 1976, 1979). Crosses (F I) betw een the various strains have an activity w hich is interm ediate to that o f the parents. Therefore, the B6C3F1 m ouse used in the present investigation w ould be expected to be Ah resp o n siv e, sin ce b o th p aren ts w ere responsive (C 57B L/6 and C3H ). The B6C3F1 m ouse was selected for these studies because it represents the m ouse approved for study by the N ational Toxicology Program and because we have had 5 years o f experience investigating the effects o f xenobiotics on its im m u n ocom petence. W hen one considers all the effects previously described for T C D D (this list is by no m eans intended to be all inclusive), the fol low ing have been dem onstrated to be pro d uced by low er con centration s o f T C D D in /1/i-responsive m ice com pared to /I/-nonre sponsive m ice: liver induction (Poland and G lover, 1974, 1975), thym ic atrophy (Poland and G lover, 1980), and suppression o f both h u m o ra l (V e c c h i et al., 1983) a n d c e ll-m e d ia ted im m u n ity (N agarkatti et al., 1984). In other words, the only parameter not to be show n to possess the genetic differential sen sitivity is the recent effects on th ym ic epithe liu m (G re en lee et al.. 1985). It is im p o r ta n t to e m p h a size th at in this report, G r e e n le e et al. only m easured the effects on prim ary cultures o f cells d erived from C 5 7 B L /6 m ic e , an Ahresponsive strain. T he second feature ofth e.-l/i receptor which has been used in the experim ental designs o f these studies is based on a clear stru ctu re-ac tivity relationship regarding the ability o f halogenated arom atic hydrocarbons (i.e., dibenzod ioxin s, dibenzofurans. azo[.v. y]benzenes and b ip h en yls) to bind to the Ah receptor and produce the associated effects. In general, stru ctures w h ich have affinity for th e Ah re ceptor m ust be planar or near planar and must be halogenated in the lateral ring positions. For the dibenzodioxins. halogen atom s must 452 HOLSAPPLE. MC CAY. AND BARNES occupy at least three (and for m axim al po tency, four) o f the lateral ring positions (2, 3, 7, and 8) and at least one ring position m ust be u nsu bstituted -- Le., octachlorodibenzo-p d ioxin (O C D D ) is inactive. O f particular im portance to the present investigation were the results w hich indicated that neither 2,7-D C D D nor the com pletely substituted O C D D could b in d to the Ah receptor. Param eters for w h ich these congeners have been dem onstrated to be devoid o f effect include liver induction (Poland and G lover, 1974, 1975), thym ic atrophy (Po land and G lover, 1980), and effects on thym ic e p ith e liu m (G reen lee et a i, 1985). N o tice a b ly absent from this list are param eters reflecting the dioxin-induced changes in im m u n ocom p tence. A s described in the introduction, V ecch i et al. (1 9 8 3 ) d em on strated that 2 ,3 ,7 ,8 dibenzofuran could suppress the antibody re sp o n se an d Clark et al. (1 9 8 3 ) d em o n stra ted that planar PCBs could suppress the genera tion o f cytotoxic T cells. Our recent report that subchronic exposure to 1,2,3,6,7,8-H C D D could suppress the antibody response (H olsap p le et al., 1984a) w as the first report to d e scribe im m unotoxicity by any chlorinated dibenzodioxin other than T C D D , the prototype o f the class. T he positive effect by 1,2,3,6,7,8H C D D supports the stru cture-activity rela tio n sh ip associated w ith th e Ah recep tor sin ce this congener possesses a m easurable affinity for th is lo c u s (P o la n d et a i, 1976) a n d is halogenated in the key lateral positions. T he positive activity presently dem onstrated for 2 ,7-D C D D in a num ber o f different im m unological m odels is the first m ajor evidence that an effect produced by the chlorinated dibenzodioxins need not be associated with the Ah receptor. T h a t w e w ere n o t p ro d u cin g the im m unosuppression by contam inants of T C D D (o r a n o th er Ah a ctiv e co n g en er) in o u r batch o f D C D D is indicated by the studies on liver induction. T C D D produced m arked in duction o f a variety o f liver parameters (Table 2) at the doses producing im m unosuppression (Tables 3 -5 ). In contrast, 2,7-D C D D was to tally d evoid o f effects on the liver at effective im m u n otoxic concentrations. W hile it is im portant to em phasize that the m ajority o f re ports on 2,7-D C D D have indicated no activity, th is is n o t th e first report for a p ositive effect by this congener. A s described by Kociba (1984), in conventional lifetim e 2-year carci nogenic studies sponsored by the N ational T oxicology Program (N T P ) 2,7-D C D D did not elicit any response in the rat, but there w as a suggestive effect in the form o f increased liver tum ors at the high dose in the m ouse. D ue to the paucity o f results for other dib en zod ioxin s, it is im possible to predict how m uch the structure-activity relationship (SA R ) for im m unosuppression m ight differ from the SA R reported for effects on the liver or thym us. W e have begun to screen the ac tivity o f a num ber o f chlorinated dibenzodioxins varying in the num ber and position o f the chlorine substituents. O ne particularly in terestin g co n g en er is 2 ,8 -D C D D . L u ster et al. (1984) have recently reported that this con gener was devoid o f activity w hen added di rectly to cultured spleen cells from B6C3F1 m ice. W e have recendy dem onstrated that both 2,3,7,8-T C D D and 2,7-D C D D were di rectly im m unosuppressive w hen added to cu l tured sp leen cells (H o isa p p le et al.. 1985). In addition to indicating that im m u n otox icity by the chlorinated dibenzo-dioxins need n o t be m ed iated through th e Ah receptor, th e present results also suggest that the effects on the antibody response can be com pletely dis sociated from the effects on the thym us. This interpretation is based o n the ability o fT C D D and D C D D to suppress the response to a thym ic independent antigen. D N P -F icolI (Table 4), as well as the response to the poly clonal activator, LPS (T able 5). T hese results are in agreem ent with previous results w hich indicated that in adult m ice. T C D D suppresses hum oral im m unity at doses which do not alter ce ll-m ed ia te d im m u n ity (M a n to v a n i et al.. 1980; V ecchi et al., 1980). T h e results in T a bles 3, 4, and 5 indicated that subchronic ex posure to T C D D could suppress antibody re sponses in the com p lete absence o f effects on th e th y m u s (T ab le 1). B ecau se o f the classical association between exposure to T C D D and IM M UNOSUPPRESSION BY 2,7-DCDD 453 thym ic atrophy, w e were som ew hat surprised by these results and q uestioned w hether the lack o f effect on the thym us m ay be a con se quence o f repeated exposures. Since m ost o f the earlier investigations have utilized single exposure to T C D D , it was im portant to d e term ine the effects o f acute exposure. As show n in Table 6, acu te exp osu re to T C D D at doses low er than 2 tg/kg was capable o f suppressing the antibody response in the ab sence o f effects on the thym us. W e have care fully considered the available literature and have sum m arized com parable effects in Table 7 from four previous reports plus the present results. T h e first ob servation is the co n siste n c y o f effect by T C D D -- these results were gen erated over the last 6 years by four different laboratories-- the suppression by exposure to ~ 1.0 Mg/kg T C D D has been 54, 78, 57, 40, and 45%, respectively. T h e second observation is that significant effects on the thym us were only reported for doses o f T C D D which were 5 -1 0 -fo ld higher-- 5 -1 0 /xg/kg. T o sum m arize, these results indicate differ en tial con trib u tion s by the Ah receptor in m e diating the effects o f exposure to chlorinated dibenzodioxins. T h e liver and im m unity are affected by doses w hich are devoid o f activity on the thym us. T h e lack o f effect o f 2,7-D C D D on thediver is inagreement with previous re su lts w h ich h ave im p lica ted the Ah receptor in this d ioxin-induced action. T he im m u n o suppression by 2 ,7 -D C D D indicates that this effect by the dioxins is not necessarily a con sequ en ce o f the Ah receptor and indicates the im portance for further study into the SAR o f dioxin-induced im m unotoxicity. These studies are currently underw ay. W hile it is tem ptin g to sp eculate that the activity by 2 ,7 -D C D D is either suggestive o f m ultiple dioxin receptors TABLE 7 Com parison o f t h e Effects o f a c u t e Ex po su r e to 2 ,3 .7 ,8 -T etra c h lo r o d ibe n zo d io x in (T C D D ) o n T h y m u s w e ig h t a n d T -D epe n d e n t a n t ib o d y Pr o d u c t io n in Fo u r Labo r a to r ies Dose of TCDD Atg/kg) T hym us weight (% o f control) IgM A FC/106 spleen cells (% o f control) Reference (m ouse strain) 1.2 6.0 30.0 1.2 6.0 30.0 0.1 1.0 10.0 0.2 1.0 5.0 10.0 0.1 0.5 1.0 2.0 ND ND ND 89% 56% 36%** 84% 94% 61%* 94% 100% 77%* 57%** 96% 117% 104% 106% 46%** 14%** 5%** 22%** 16%** 3%** 97% 43%** 19%** 35% 40% * 47%** ND 105% 74% 55%* 34%** Vecchi et ai. (1980) (C57BL/6) Vecchi et al. (1983) (C57BL/6) Dean and Lauer(l984) (B 6 C 3 F I) L uster et ai. (1984) (B6C3F1) Sum m arv from Table 6 (B6C3FI) p < 0.05: **p < 0 .0 1 . ND-- not determined. 18457 % > 454 HOLSAPPLE, M C CAY, AND BARNES or a nonreceptor effect, w e consider this in terpretation to be prem ature in the absence o f any binding studies w hich dem onstrate spe cific affinity for 2 ,7 -D C D D on on e o f the subpopulations o f im m u n ocom petent cells. ACKNOW LEDGM ENTS T his investigation was supported by contract NOI-ES1-5001 from th e N ational In stitu te o f E nviro n m en tal Health Sciences and by grant N O 2-RO I-ESO 2520 from the N ational Institutes o f H ealth. W e thank Dr. William G reenlee for the gift o f 2.7-D C D D an d for helpful dis cussion. D r. A nne T ucker for the gift o f O C D D . and Ms. Brenda Rexrode for preparing this m anuscript. REFERENCES Bick, P. H.. Tucker, a . N.. White. K. L., Jr., and Holsapple, M . P. (1984). Effects o f subchronic expo sure to diethylstilbestrol on hum oral im m une function in adult fem ale (C3B6)FI mice. Im m unopharm acology 7, 27-39. Bullock. W. W,, and Moller, E. (1972). S p o n tan eo u s B cell activation d u e to loss o f norm al m ouse serum suppressor. Eur. J. Im m unol. 2, 514-517. Clark, d . a .. Sweeney, G.. Safe, S.. Hancock, e .. Kilburn, D. G., and Gauldie, J. ( 1983). C ellular and genetic basis for suppression o f cytotoxic T cell gener ation by haloarom atic hydrocarbons. Immunopharm aco/ogy6, 143-153. Cochin, J., and Axelrod, J. (1959). B iochem ical an d pharm acological changes in the rat following chronic adm inistration o f m orphine, nalorphine, and norm orphine. J. Pharmacol. Exp. Ther. 125, 105-110. Dean, J. H.. and Lauer, L D. (1984). Im m unological effects following exposure to 2.3,7,8-tetrachlorodibenzop-dioxin: A review. In Public Health R isks o f the Dioxins (W . W. Lawrence, ed.) pp. 275-294. W illiam Kaufm ann. Inc.. Los Altos. CA. Greenlee. W. F., Dold, K. M., and Osborne, R . (1984). A proposed m odel for th e actio n s o f 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on epiderm al and thym ic epithelial target cells. In B anbury Report 18: Biologic M echanism s o f D ioxin A ction (A. Poland and R. K im brough, eds.), pp. 435-440. Cold Spring H arbor Lab oratory, Cold Spring H arbor, N.Y. Greenlee, W . F., Dold, K . M .. Irons, R. D.. and Os borne, R. (1985). Evidence for direct action o f 2,3.7,8tetrachlorodibenzo-p-dioxin (TC D D ) on thym ic epi thelium . Toxicol. Appl. Pharmacol. 79, 112-120. Greenlee, W. F., and Neal, R . a . (1984). T h e A h re c e p to r A biochemical and biological perspective. In The Receptors (P. M . C onn, ed.), Vol. 2. A cadem ic Press, New York. Holsapple, M. p .. McNerney, P. J., Barnes. D. w ,, and White. K. L . Jr. (1984a). Suppression o f hum oral antibody production by exposure to 1,2,3,6.7.8-hexachlorodibenzo-p-dioxin. J. Pharmacol. Exp. Ther. 231, 518-526. Holsapple. M. P.. McNerney, p . J.. Tucker, a . n .. and White. K. L . Jr. (1984b). Effects o f N -nitroso- dim ethylam ine on hum oral im m unity. J. Pharmacol. Exp. Ther. 229, 493-500. Holsapple. M. p ., McNerney, p . J.. Tucker, a . n .. and Luster. M. I. (1985). D irect suppression o f an ti body production by cultured m ouse spleen cells by 2J.7.8-T C D D . Toxicologist 5 .4 3 . Klaus, G . G . B,, Phillips, J. M ., Humphrey. J. M .. Dresser. D . w ,, and Cross, a . m . (1976). T h e im munological properties o f haptens coupled to thymicindependent carrier molecules. IV. The IgG response to dinitrophenylated-Ficoll. Eur. ]. Im m unol. 6 ,4 2 9 - 433. Kociba, R. (1984). E valuation o f the carcinogenic and m utagenic potential o f 2.3,7,8-TCDD and other chlo rinated dioxins. In B anbury Report 18: Biologic M ech anism s o fDioxin Action (A. Poland and R. Kimbrough, eds.), pp. 73-84. Cold Spring H arbor Laboratory. Cold Spring H arbor, N.Y. Lowry, O. H.. Rosebrough, N. J.. Farr, A. L.. and Randall, R. J. (1951). P rotein m easu rem en t with the Folin phenol reagent J. Biol. Chem . 193, 265-275. Luster, M. I., Tucker, a . N., Hong, L.. Boorman, G .. and Patterson, R . ( 1984). I n vivo a n d in vitro effects o f T C D D o n stem cell a n d B cell differentiation. In B anbury R eport 18: Biologic M e c h a n ism s o f D ioxin A c- tio n (A . Poland and R. K im brough, eds.). pp. 411-419. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Mantovani, A.. VECCHI, A.. LUINt. W,, SlRONt, M.. Canaiani. G. P., Spreafico, F.. and Garattini, S. (1980). Effect o f 2,3.7.8-tetrachlorodibenzo-p-dioxin on macrophage and natural killer cell-mediated cytotoxicity in mice. Biom edicine 32, 200-204. Nagarkatti. P. S.. Sweeney, G . P.. Gauldie. J., and Clark, D. A. (1984). Sensitivity to suppression o f cy totoxic T cell generation by 2.3.7,8-tetrachlorodibenzop-dioxin (T C D D ) is d e p e n d en t o n th e Ah genotype o f th e m urine host. Toxicol. Appl. Pharmacol. 72. 169- 176. Nebert, D. w .. Goujon. F. M.. a n d G ielen. J. E. (1972). Aryl hydrocarbon hydroxylase induction by polycyclic hydrocarbons. Simple autosom al d o m in an t trait in the m ouse. A'ature S e w Biol. 236. 107-110. Omura.T .. a n d Sato. R. (1964). T h e c arbon m onoxidebind in g pigm ent o f liver m icrosom es. I. Evidence for its hem oprotein nature. J. B io l Chem . 239. 2370-2378. Poland, A., and Glover, E. (1974). C o m p ariso n o f 2,3,7,8-tetrachlorodibenzo-p-dioxin. a potent inducer of aryl hydrocarbon hydroxylase with 3-m ethylcholanthrene. M ol. Pharmacol. 10. 349-359. 1S458 / IMMUNOSUPPRESSION BY 2,7-DCDD 455 Poland, A ., and Glover, E. (1975). G e n e tic expression o f aryl hydrocarbon hydroxylase by 2,3,7,8-tetrachlorodibenzo-p-dioxin: Evidence for a receptor m utation in genetically non-responsive mice. M ol. Pharmacol. 11, 389-398. Poland, a ., and Glover, E. (1980). 2,3,7,8-tetrachlorodibenzo-p-dioxin: Segregation o f toxicity with the Ah locus. Mol. Pharmacol. 17, 86-94. Poland, A ., Glover, E., and Kende, A . S. (1976). Ste reospecific high affinity binding o f 2,3,7,8-tetrachlorodiben zo -p -d io x in s an d related c o m p o u n d s. A n n . Af. Y. Acad. Sci. 320, 214-230. Poland, A., Greenlee, w . f ,, and Kende, a . S. (1979). Studies on the m echanism o f action o f the chlorinated dibenzo-p-dioxin$ a n d related co m p o u n d s. A n n . iV. Y. Acad. Sci. 320, 214-230. Rittenberg, M. B,, and Pratt, K. L. (1969). Antitrini- trophenyl (TNP) plaque assay: Prim ary response to BALB/c mice to soluble and particular im m unogen. Proc. Soc. Exp. Biol. M ed. 132, 575-581. VECCHI, A ., MANTOVANI, A., SlRONt, M ., LUINI, W ., Cairo, M ., and Garattini, S. (1980). Effect o f acu te exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin on hu m oral antibody production in mice. Chem .-Biol. In teract. 30, 337-342. VECCHI, a ., Sironi, M ., Canegrati, M . a ., Recchia, M ., and Garattinl S. (1983). Im m u n o su p p ressiv e ef fects o f 2,3,7,8-tetrachlorodibenzo-p-dioxin in strains o f m ice with different susceptibility to induction o f aryl hydrocarbon hydroxylase. Toxicol. Appl. Pharmacol. 68, 434-441. Yang, G S., Stricxhart, F. S., and Gcha, L P. (1978). Analysis o f the aryl hydrocarbon hydroxylase assay. Biochem . Pharmacol. 27, 2321-2326. J co CO Ann. Rev. Pharmacol. Toxicol. 1986. 26:371-99 L Copyright 1986 by Annual Reviews Inc. Ail ri reserved COMPARATIVE TOXICOLOGY AND MECHANISM OF ACTION OF POLYCHLORINATED DIBENZO-P-DIOXINS AND DIBENZOFURANS S. H , Safe Department of Physiology and Pharmacology, College of Veterinary Medicine, Texas A&M University, College Station, Texas 77843 INTRODUCTION Polychlorinated dibenzo-p-dioxins (PC D D s) and polychlorinated dibenzofurans (PCDFs) are members o f a chemical family (polyhalogenated aromatics) that also includes the polychlorinated biphenyls, naphthalenes, azobenzenes, and azoxybenzenes, and the polybrominated biphenyls. In con trast to the polychlorinated biphenyls and naphthalenes and (he polybrominated biphenyls, the PCDDs and PCDFs are not primary industrial products. PCDFs are found as by-products ( < 1 ppm) in commercial polychlorinated biphenyls and naphthalenes and are probably derived from dibenzofuran impurities in the industrial hydrocarbons that are subsequently chlorinated (1 -3 ). There is also circumstantial evidence that the effects o f heat or arcing may produce PCDFs from polychlorinated biphenyls during use (4). This is evidenced by the relatively high levels of PCDFs (ca 100 ppm) detected in the polychlorinated biphenyl-containing heat transfer fluids that were the toxic agents in the Yusho disasters in Japan and Taiwan (3, 5, 6). PCDDs and PCDFs are also found as impurities in chlorinated phenols and their derived products ( 3 ,7 , 8), and it is apparent that the combustion o f chlorinated aromatics and diverse types o f chemical, industrial, and municipal waste results in the formation and release o f these toxic chemicals into the environment (9-14). The potential for the 0362-1642/86/0415-0371 $07.00 371 372 SAFE fonnation o f PCDDs and PCDFs from nonindustrial sources, i.e. energyderived combustion and forest fires, led to the "trace chemistries o f fire'' hypothesis, which speculated that the origins o f PCDDs and PCDFs in the environment were nonanthropogenic (15, 16). However, analysis o f aquatic sediment cores from the Saginaw River and Bay and from Lake Huron does not support the trace chemistries o f fire hypothesis (17, 18). The PCDD and PCDF congener composition o f dated sediment cores demonstrates that the con centrations o f these compounds in sediments have greatly increased since the 1940s and that "this historical increase is similar to trends for the production, use, and disposal o f chlorinated organic compounds'' (17, 18). Human exposure to PCDDs and PCDFs has occurred via three major path ways: occupational, accidental, and environmental. Industrial workers en gaged in the manufacture or use o f polychlorinated biphenyls, chlorinated phenols, and their derived products are exposed to PCDDs and/or PCDFs in combination with their associated major commercial product (19-21). A cci dents in which PCDDs have been released into the workplace or into the environment (e. g. the Seveso accident in Italy) have also resulted in human exposure to mixtures o f the industrial chemicals and their PCDD/PCDF toxic contaminants (22-24). The Yusho poisoning in Japan and Taiwan involved the exposure o f several thousand individuals to PCBs and their PCDF contaminants ( 5 ,6 , 2 5 -27). The uptake o f environmental residues o f PCDDs and PCDFs into higher trophic levels o f the food chain is only now being investigated, and trace levels (parts per trillion) have been detected in fish, wildlife, and human tissues (19, 2 8 -32). In common with accidental and industrial exposures to PCDDs and PCDFs, exposure levels to these toxins represent only a small fraction of the total bioavailable lipophilic environm ental pollutants. 2 ,3 ,7 ,8 Tetrachlorodibenzo-p-dioxin (TCDD) is the major by-product formed from 2,4,5-trichlorophenol and its derived products; this highly toxic compound has been the focus o f most biologic and toxic studies on the PCDDs. However, all other human and environmental exposures to PCDDs and PCDFs involve a complex mixture o f isomers and congeners in combination with other chem icals. The scientific, regulatory, and media attention focused on PCDDs, PCDFs, and particularly 2,3,7,8-T C D D has continued unabated; moreover, with the recent identification o f trace levels o f these toxins in human tissue, domestic animals, the environment, and toxic chemical waste dumpsites, the scientific and societal concern about this class o f compounds will no doubt continue. It is CD apparent from the scientific literature that one member o f this class o f com pounds, namely 2,3,7,8-T C D D , ranks with benzo[aJpyrene as one o f the most O j thoroughly studied toxins. Unfortunately, the biologic and toxic effects o f the remaining 74 PCDD and 135 PCDF congeners have not been thoroughly PCDDs A N D PCDFs-- SARs 373 investigated, and the interactive effects of these compounds or theiractivities in combination with polychlorinated biphenyls and chlorinated phenols have also not been addressed. This article (a) briefly reviews the species-dependent toxic and biologic effects of PCDDs and PCDFs, (b) demonstrates the parallel modes of action of PCDDs and PCDFs and endogenous cellular hormones, (c) summa rizes the data that support the proposed receptor-mediated mechanism of action, including the structure-activity relationships (SARs) that have been developed for both PCDD and PCDF congeners, and (d) discusses the few interactive studies published. TOXIC AND BIOLOGIC EFFECTS OF PCDDs AND PCDFs Several review articles (33-45) have summarized the toxic and biologic effects elicited by PCDDs, PCDFs, and related toxic halogenated aryl hydrocarbons. The toxic effects resulting from exposure to this group of chemicals are dependent on a number of factors which include the dose of the toxin, and the age, strain, species, and sex of the animals used. The complete spectrum of toxicity is not usually observed in any single animal species; however, the limited data available indicate that the toxic PCDDs, PCDFs, and related compounds elicit the same qualitative pattern of responses within each species. The differences in species susceptibility to this group of chemicals are illus trated by the LD50 values for 2,3,7,8-TCDD, which vary over 5000-fold (33, .34) from the highly sensitive guinea pig to (he resistant hamster 1LDj0s (|xg/kg): guinea pig (0.6-2.0), rat (22-45), chicken (25-50), monkey (70), rabbit (115), dog (100-200), mouse (114-284), bullfrog (>1000), hamster (1157-5051)]. The quantitative differences in the toxicity of PCDF congeners have recently been demonstrated (47) for a series of ten congeners (see structure-activity section, below). The 2,3,4,7,8-pentachlorodibenzofuran (PeCDF) EDS0 val ues for thymic atrophy and body weight loss in the rat were 0.21 and 1.04 p.mol/kg; the 1,2,4,7,8-PeCDF isomer elicited the same toxic effects, but the ED50 values were 220 and 47 times higher, respectively (47). The toxic responses observed in several animal species by PCDDs and PCDFs include dermal toxicity, teratogenicity, reproductive problems, body weight loss, hepatotoxicity, gastric lesions, lymphoid involution, immunotoxicity, and carcinogenicity. The two most characteristic toxic effects observed in all laboratory animals are lymphoid involution and/or immunotoxicity and body weight loss. Chloracne and related dermal lesions are the most frequently noted signs of PCDD and PCDF toxicosis inhumans; dermal lesions arealso observed in rhesus monkeys, hairless mice, and rabbits that have been exposed to this group of toxins. In contrast, rats, most strains of mice, guinea pigs, and hamsters do not develop chloracne and related dermal toxic lesions after 374 SAFE Figure t Structure of the polychlorinated dibenzofurans and dibenzo-p-dioxins. exposure lo 2,3,7,8-TCDD. Poland & Knutson (38) have noted that many of the observed toxic lesions are either hyperplastic/metaplastic or hypoplastic, and primarily affect epithelial tissues. The mechanisms by which PCODs and PCDFs elicit this diverse group of species-dependant toxicities remain un explained although several hypotheses have been advanced and arediscussed in this review. PCDDs and PCDFs cause diverse biological responses in mammals and mammalian cells in culture including the highly characteristic induction of microsomal benzo[a]pyrene hydroxylase (aryl hydrocarbon hydroxylase, AHH) and several related cytochrome P-450-dependent monooxygenases (37, 38, 44, 47-53). In the rat, these activities are associated with the preferential induction of cytochrome P-450c, P-450d, and P-450a, with the former isozyme responsible for most of the induced monooxygenase enzyme activities (54). In the mouse, 2,3,7,8-TCDD induces cytochromes Pr 450 and Pj-450 (55-57), and the former isozyme exhibits antigenic and enzymatic similarities with the rat cytochrome P-450c (58). 2,3,7,8-TCDD induces two cytochrome P-450 isozymes (forms 4 and 6) in the rabbit, and their inducibility is highly tissuespecific (59, 60). 2,3,7,8-TCDD and related compounds also induce glu tathione S-transferases (61) and glucuronosyl transferase (62, 63) and several other enzymes including DT-diaphorase (64), ornithine decarboxylase (65), 8-aminolevulinic acid synthetase (48), epidermal transglutaminase (66), and hepatic DNA polymerase B (67). Detailed summaries of these and other biochemical effects of 2,3,7,8-TCDD have recently been reviewed (38, 42). The remarkably broad spectrum of biologic and toxic responses observed in animals exposed to 2,3,7,8-TCDD and related toxic halogenated aromatics has stimulated research on the mechanism or mechanisms of action of these chem icals. Unlike many toxins, the most active halogenated aryl hydrocarbons do not appear to require metabolic activation into presumed "toxic" intermediates that alkylate specific cellular acceptors (e. g. DNA, RNA, and protein) or initiate cellular lipoperoxidation. On the contrary, the most toxic halogenated aryl hydrocarbons are highly resistant to oxidative metabolic degradation and exhibit minimal metabolically mediated alkylation of cellular macromolecules (68-70); moreover limited data suggest that PCDD and PCDF metabolites are much less toxic than their parent hydrocarbons (71). Many of the effects of this PCDDs A N D PCDFs-- SARs 375 class of environmental toxins are comparable to those associated with modula tion of several hormone-mediated responses. For example, there are many similarities between animals that exhibit thyroid dysfunction and those treated with toxic halogenated aryl hydrocarbons. Daily injections of the active thyroid hormone, triiodothyronine (T3) to male mice treated with a lethal dose (200 p-g/kg) of 2,3,7,8-TCDD did increase theirmean surival times; however, all the animals in the 2,3,7,8-TCDD and 2,3,7,8-TCDD + T3-treated groups died (43). More recent studies (71, 72) have demonstrated that thyroid hormones may play a more important role in modulating the toxicity of 2,3,7,8-TCDD; radiothyroidectomy protected rats against 2,3,7,8-TCDD-mediatedT-cell immunotoxicity (as measured by the spleen anti-SRBC plaque-forming cell assay), mortality, and body weight loss (73). Like the glucocorticoids, PCDDs, PCDFs, and related toxic halogenated aryl hydrocarbons cause lymphoid involution (33, 34, 37, 38,40,'41,74,75), are teratogens in mice (76-80), and induce cytochrome P-450-dependent monooxygenases (37, 38, 44, 46-54). However the mechanisms of action of glucocorticoids and 2,3,7,8-TCDD are not directly linked since the latter compound is toxic to adrenalectomized rats, does not bind to the glucocorticoid receptor, and does not induce tyrosine aminotransferase (43) or the dexamethasone-induced cytochrome P-450 iso zyme (81). Other adrenal steroids resemble the toxic halogenated aryl hydro carbons since they also induce several hepatic drug-metabolizing enzymes, including monooxygenases and glucuronosyl transferases (81-84). It has been proposed that 2,3,7,8-TCDD and related toxic isostereomers, like the steroid hormones, elicit their responses via the initial noncovalent interac tion with a cytosolic receptor protein in target tissues (37, 38, 75). The synthesis of radiolabelled I3HJ-2,3,7,8-TCDD with high specific activity (52.5 Ci/mmol) resulted in the identification of a specific binding protein in hepatic cytosol of responsive C57BL/6J mice, whereas minimal binding activity was observed in nonresponsive DBA/2J hepatic cytosol (85). The role of this Ah receptor protein in the mechanism of action of toxic halogenated aryl hydrocar bon has been thoroughly investigated and satisfies most of the specific criteria (hat support a receptor mediated cellular process. These criteria include: (a) the existence of a finite number of binding or receptor sites and therefore saturable binding, (6) high affinity ligand binding that is commensurate with the usually low levels of circulating hormones, (c) stereoselective binding capacity for the receptor, (d) tissue or organ response specificity for the receptor ligand, and (e) a correlation between binding affinities, receptor occupancy, and the magni tude of the response. This review focuses on research (hat supports the role of the Ah receptor in the mechanism of action of PCDDs and PCDFs and high lights the detailed structure activity relationships (SARs) that have been de veloped for this group of environmental and industrial toxins. 376 SAFE PCDD AND PCDF ACTIVITIES: EVIDENCE THAT SUPPORTS THE ROLE OF THE CYTOSOLIC Ah RECEPTOR PROTEIN High Affinity Saturable Binding The saturable binding of [3H]-2,3,7,8-TCDD with hepatic and extrahepatic cytosolic receptorprotein fromseveral species has been demonstrated using the following receptor assay procedures: charcoal/dextran absorption, protamine sulfate precipitation, hydroxylapatite absorption, isoelectric focusing in polyacrylamide gels, gel permeation chromatography, sucrose density gradient centrifugation, and gel permeation high performance liquid chromatography (85-110). Scatchard plot analysis of [3H]-2,3,7,8-TCDD specific binding to hepatic cytosolic receptor protin gives dissociation consants (KD) that are dependent on a number of factors including the animal species and strain, the receptor binding assay used, and the age of the animal. The KD values for responsive C57BL/6 mice and rats vary from 0.27-3.0 nM (85, 95) and 0.13-1.2 nM (91, 95, 102), respectively, and the value for cynomolgus monkeys was approximately 3 nM (108). The concentration of hepatic cytosol ic receptor was also highly variable, but the upperlimit for most studies was less than 110 fmol/mg cytosolic protein (85, 91,95). One study demonstrated that hepatic receptor levels in the rat varied with age and that these levels were endocrine independent since hepatic receptor levels were virtually unaltered by orchiectomy, ovariectomy, adrenalectomy, or hyposectomy (106). This observation was consistent with the inactivity of several steroid hormones as competitive ligands for this receptor protein (85, 86, 89, 99, 102). Since certain polycyclic aromatic hydrocarbons resemble 2,3,7,8-TCDD in their mode of induction of AHH and related cytochrome P-450 isozymes in responsive strains of mice and mammalian cells in cultures, it is not surprising that many of these compounds competitively displace [3H]-2,3,7,8-TCDD from the receptor protein. Moreover, I3II]-3-methylcholanthrene, benzo{a]pyrene, and dibenzo(a,h]anthracene, three active AHH inducers, exhibit satur able binding with the rat hepatic receptor protein and the ligand-receptor complex sediments at 8-9 S under low ionic strength conditions using the sucrose density gradient technique (97, 105). The radioactive binding peaks were eliminated after competition with a 200-fold molar excess of 2,3,7,8TCDD. Similar results were also observed with responsive C57BL/6 mouse hepatic cytosol. However, [3H]-benzo[a]pyrene unexpectedly does not yield a radiolabelled ligand-receptor binding peak that is eliminated after competition with a 200-fold molar excess of unlabelled 2,3,7,8-TCDD. ^ Tissue/Organ, Strain, and Cell Culture Response Specificity The 2,3,7,8-TCDD receptor levels in several organs and tissues in SpragueDawley rats, C57BL/6J and DBA/2J mice have been reported (96, 103, 104, PCDDs A N D PCDFs-- SARs 377 106). The cytosolic Ah receptor concentrations (fmol/mg cytosolic protein) in the C57BL/6J mice and Sprague-Dawley rat organs and tissues were: liver, 32 1.5 and 39 1.9; lung, 23 6.4 and 47 4.3; kidney, 10 0.4 and 1.2 0.9; intestine, 8 2.3 and 15 1.7; thymus, 8 2.2 and 54 3.9 (The Ah receptor was not detectable in the adrenals, heart, brain, skeletal muscle, and testis). Although the Ah receptor was not detected in the cytosol of DBA/2J mice, 18 hr after administration of l3H]-2,3,7,8-TCDD to these animals levels of the receptor-ligand complex could be measured in nuclear protein extracts from liver (5.4 0.3 fmol/mg), lung (7.4 0.3 fmol/mg), and kidney (4.7 0.1 fmol/mg). The appearance of hepatic nuclear radiolabeled ligand-receptor protein complexes and the elimination of this radioactivity by preinjection with a large excess of unlabeled 2,3,7,8-TCDD has been reported by several groups (89, 91,96, 99, 102, 106); however, Mason & Okey (96) demonstrated that lung, liver, and kidney nuclearAh receptor levels were higher in the responsive C57BL/6J mice than in the nonresponsive DBA/2J strain. This observation is consistent with the fact that 2,3,7,8-TCDD and other toxic halogenated aryl hydrocarbons elicit biologic and toxic responses in both strains of mice but at different dose levels. Unfortunately there are insufficient data available to correlate tissue/organ receptor levels with the magnitude of specific responses in these target sites. The criteria for receptor response specificity are supported by numerous studies with genetically inbred responsive and nonresponsive strains of mice and with some mammalian cells in culture. For example, there is an excellent rank order correlation between the maximum AHH inducibility in several inbred strains of mice and F| hybrids and the number of Ah receptor molecules per liver cell (109). Nebert and co-workers have also shown alinear correlation (r = 0.99) between the amount of 2,3,7,8-TCDD-receptor complex appearing in hepatic nuclei of C57BL/6 and DBA/2 mice and thepercentage of maximally induced cytochrome P(-450 mRNA (104). Hudson and co-workers have demonstrated that for several human squamous cell carcinoma lines, the rela tive amount of receptor measured in each cell line correlated well with the 7-ethoxycoumarin 0-deethylase inducibility in these cells by 2,3,7,8-TCDD (HO). These data that support the receptor-mediated response specificity are in contrast to data in several other studies with animals and cell cultures. Hepatic 2,3,7,8-TCDD receptor levels in guinea pigs, rats, mice, hamsters, and nonhu man primates vary less than tenfold (10-100 fmol/mg cytosolic protein) and exhibit comparable KD values for [3H]-2,3,7,8-TCDD binding (93); these levels show no correlation between their maximal hepatic AHH inducibility or susceptibility to the toxic effects of 2,3,7,8-TCDD andrelatedhalogenated aryl hydrocarbons (93, 97, 108). For several mammalian cells in culture there is no correlation between receptor levels and their AHH inducibility (98, 99, l l l ll 3). Recent studies by Whitlock and co-workers indicate "that transcriptionof 378 SAFE the cytochrome Pr 450 gene is under both positive and negative control by at least two trans-acting regulatory factors" (113). The factors that control cytochrome Pr 450 in variant mouse heptoma cells may also play arole in some animal species and requires further investigation. It is apparent that response specificity to Ah receptor ligands is a highly complex process that depends not only on receptor levels but also on many other factors, an observation not unique to the Ah receptor protein (113, 114). Structure-Activity Relationships RECEPTOR BINDING AFFINITIES OF PCDDs AND PCDFs: SUBSTITUTION e ff e c t s Poland, Glover & Kende first reported the relative binding affinities of 23 halogenated dibenzo-p-dioxins and dibenzofurans using the dextran charcoal receptor assay and l3H|-2,3,7,8-TCDD as the competing radioligand (85). This study included 10 PCDD congeners and 7 PCDF congeners that differed only with respect to their degree of chlorination and substitution pattern. Table 1 summarizes results from a more recent study of the effects of structure on the receptor binding affinities of 14 PCDDs and 14 PCDFs using rat hepatic cytosol and the sucrose density gradient assay procedure (47, 51; G. Mason, J. Pikorska-Pilszczynska, B. Keys & S. Safe, unpublished results). 2.3.7.8- TCDD and 1,2,3,7,8-pentachlorodibenzo-p-dioxin were the most avid PCDD competitive binding ligands for displacement of [3H]-2,3,7,8-TCDD from the receptor protein, and their EC50 values were 1.0 x 10-8 and 7.9 x 10"8 M, respectively. Inspection of these data clearly demonstrated the im portance of the lateral Cl substituents in facilitating the interaction between the PCDD ligands and the cytosolic receptor protein. The relative receptor binding EC50 values for a series of telrachloro isomers were 2,3,7,8- > 2,3,6,7- > 1.3.7.8- > 1,2,3,4-, in the order of decreasing number of lateral substituents. 'I'lie fivefold difference in the receptor binding activities of the 2,3,6,7- and 1.3.7.8- TCDD isomers illustrates a more subtle structural feature that affects binding. The increased affinity of the former compound must be due to the receptor binding site preference for a vicinal 6,7- (or, 1,2) group over a meta 1,3-dichloro functionality. The data also illustrate that the degree of chlorina* lion of non-lateral sites is an important structural determinant for interaction * with the receptor protein. The 2,3,7,8-tetra-, 1,2,3,7,8-penta-, 1,2,3,4,7,8j^hexa-, and 1,2,3,4,6,7,8,9-octachlorodibenzo-p-dioxins all contain four laterjal Cl substituents; however, there is a marked decrease in their receptor binding tv,avidities with increasing Cl substitution at the nonlateral 1, 4, 6, and 9 (^positions. The stepwise addition of Cl groups at 1,4, 6, and 9 would result in Olseveral structural changes in the more highly chlorinated PCDDs including increased molecular size and volume, increased lipophilicity, a possible de crease in PCDD coplanarity associated with steric crowding, and decreased PCDDs A N D PCDFs-- SARs 379 9o 0 O 00 o o W Os 00 o Os SO h m O o O**4O o O o O Om O** O"* O^** O O** XXXXXXXXXXXXXX en Vi o es 00 Vi en en SO si~*4 00 Vi oo en oo en ei-Hs o00 V--i1 e-4s g wuX p > t-4 SsB 00 o O o o** Ow* O***o**Ho o O-i O o o O XXXXXXXXXXXXXX sfOS oo Os en o\ so V) es t-* es so Os r r-< s seon Vi sO 00 o OM &i o8 5 00 O CO1O O1o OO1O 0o10 rO>1 rO1* rO1* rO1 r O rO rO <o O-4 <o1o XXXXXXXXX XXXXX m *n -4 r- v M S' s r- o es o es o es eens res o S' en en U, o o o RCU o rs* en es r^> o S' rm en <N es rso S' en es o reens ren es 00 rs* en Os r* s en es oo s m eS r-> s en es r-> Os SO ren en es epHs e>o rSes' p4 00 SO" 4 es o 00 00 os r- <o V r* r- V QO O O O O O o O o O O O O O 0 XXX Os r_- '_-4 XXXXXXXXXX es m --i s es s_4 o es oo es es O f* om-l AA s own PnBr -- OO ooo i o pO1 OO1O O1o iO1^-**o*1** rO nO*1* r O V O"* g > *3 X X X X X X X X X X X X X fS V--4 f-4 Os -i r" SO o Vi es en m SO oo SO S' -4 <n o--4 AA H*5 o OOo o rO1 oi o1 oo O Oo <c O <n O n O V O cS* m'S XXXXXXXXXXXXXX o os sO es oo Os m es S' en O o es es r- r~ en o AA o oo r^' oo r- J) H(9 Q c-j h o r-* c r{ en r* 'O fi s es 4 en es o en r-; 'S es en4 es r* en es 00 es en es 4 es oQQn 380 SAFE +++ +++ +++ +++ Figure 2 The differential effects of chlorine substituents at different positions in the dibenzo-pdioxin and dibenzofuran rings on the relative receptor binding affinities of PCDD and PCDF congeners. aromatic ring electron density (due to the additional electronegative Cl groups). One or more of these changes may be related to the decrease in binding affinities of the more highly chlorinated 2,3,7,8-substituted PCDDs. Figure 2 summa rizes the differential effects of Cl substituents on the affinities of PCDDs for the cytosolic receptor protein and illustrates the importance of lateral chloro groups. It has also been suggested that the other critical structural factors that contribute to the high binding affinities of 2,3,7,8-TCDD include the planar ring structure and an ideal ligand area, 3 x 10 A (37, 38). The direct binding of radiolabelled 2,3,7,8-tetrachlorodibenzofuran (TCDF) and other PCDFs to the Ah receptor protein has not been demonstrated. However, the competitive binding affinities of three 2,3,7,8-substituted com pounds demonstrated their relatively high binding affinities for the receptor protein (85). The development of new procedures for the synthesis of PCDFs (115) has resulted in the preparation of over40 congeners that have been used to develop detailed SARs for this series of halogenated aryl hydrocarbons (47, 51). The dibenzofuran ring system possesses a single axis of symmetry (Figure 2); therefore there are four geometrically different positions on each aromatic ring, namely C-l (or C-9), C-2 (or C-8), C-3 (or C-7), and C-4 (or C-6). A complete SAR for PCDFs as ligands for the receptor protein must distinguish between the differential contributions of all four positions on the dibenzofuran ring. (See Table 1 for compounds selected for this study.) Inspection of these data confirms that the most active congeners, 2 ,3 ,4 ,7 ,8-pentachlorodibenzofuran (PeCDF, 1.5 x 1(T8M), 2,3,7,8-TCDF(4.1 x 1(T8M), 2.3.4.6.7.8- hexachlorodibenzofuran (HCDF, 4.7 x 10~8 M), and 1,2,3,7,8PeCDF (7.5 x 10~8M) were all fully substituted in their lateral 2, 3, 7, and 8 positions. Moreover, a comparison of the receptor binding ECS0values for the 2.3.4.7.8- , 1,2,4,7,8-, and 1,2,4,6,8-PeCDF isomers demonstrates the im portance of lateral chloro substituents since there is a decrease in receptor binding affinities with decreasing lateral substitution. Two pairs of PCDF isomers, namely 1,3,4,7,8- and 1,2,4,7,8-PeCDF, 2,3,4,7- and 2,3,4,8TCDF, differ only with respect to their substitution of C-2 (or C-8) and C-3 (or PCDDs A N D PCDFs-- SARs 381 C-7). In both cases the C-3 (or C-7) substituted compounds were 6.5-8 times more active than the corresponding C-2 (orC-8) isomers as competitive ligands for the rat hepatic cytosolic receptor protein. A comparison of the relative binding affinities of a series of C-l (or C-9) and C-4 (or C-6) isomer pairs illustrates the higher binding activities of the isomer that retains the C-4 (orC-6) substituent. For example the EC50 values for the 2,3,4,7-, 2,3,4,7,8-, 2,3,4,7,9-, and 2,3,4,6,7,8-substituted PCDFs were 2.5 X 10~8 M, 1.5 X 10_8M ,2 .0 x 10_7,and4.7x 10- 8 M whereas the values for the correspond ing C-l (or C-9) isomers (i.e. 1,2,3,7-TCDF, 1,2,3,7,8-, and 1,2,3,7,9PeCDF, and 1,2,3,6,7,8-HCDF) were 1.1 X 1(T7 M, 7.5 X 1(T8 M, 3.4 X 10-7 M, and 2.7 X 10~7 M, respectively. Figure 3 illustrates an overlay of 2,3,7,8-TCDD and 2,3,4,7,8-PeCDF, the two most active PCDD and PCDF ligands for the Ah receptor. The molecular areas and volumes of the dibenzofuran and dibenzo-p-dioxin ring systems are similar, but the spatial orientations of their substituents exhibit marked dif ferences. The C-3 (or C-7) substituents occupy a position between the lateral 2,3 (or 7,8) groups in 2,3,7,8-TCDD and clearly occupy the dominant lateral position in the dibenzofuran ring system. The spatial orientations of the C-4 (and C-6) and C-2 (and C-8) substituents are comparable and exhibit less overlap with the lateral positions of 2,3,7,8-TCDD; the C-l (or C-9) PCDFsubstituents exhibit the least overlap with the lateral positions of 2,3,7,8TCDD. These observations on the molecular orientations of the dibenzofuran Cl substituents are consistent with the observed SARs for PCDF receptor binding affinities and illustrate the stereospecific nature of the receptor proteinligand interactions. RECEPTOR BINDING AFFINITIES OF PCDDs AND PCDFS-. A QSAR ANALY SIS The receptor binding avidities of PCDDs and PCDFs summarized in Table 1 and in other studies (37, 38, 47, 51) not only demonstrate the im portance of Cl substitution patterns on ligand-receptor protein complex forma tion but also show that substituents are important structural determinants for these interactions. For example, the receptor binding ECSo values for 2,3,7trichlorodibenzo-p-dioxin is 7.1 X 10-8 M; replacement of the 7-C1 substituent with H gives 2,3-dichlorodibenzo-p-dioxin, which exhibits a greatly di minished receptor binding EC50 value (> 10-s M). It is clear that these substituent effects at this lateral C-7 position must be related to differences in their physicochemical characteristics which in turn influence ligand-receptor avidities. A series of substituted PCDD, PCDF, and polychlorinated biphenyl analogs have been synthesized (Figure4) as probes fordelineating the effects of substituent structure on ligand-receptor binding affinities (116-118). Each series of analogs contains a variable substituent group at a single lateral position, and it is apparent that substituent structure has a remarkable effect on 382 SAFE o -P C D D O -P C D F Figure 3 Overlay of 2,3,7,8-TCDD and 2,3,4,7,8-PeCDF structures. the receptor binding avidities of these compounds. For example the EC50value for 7-trifluoromethyl-2,3-dichlorodibenzo-p-dioxin (1.95 x 10-8 M) was 1000 times lower than the value for 7-amino-2,3-dichlorodibenzo-p-dioxin (2.88 x I0- 5 M). The effects of different substituents on the activity of aseries of analogs can be analyzed quantitatively by correlating the differences in a biological effect (e.g. receptor binding) with known substituent physicochem ical parameters (116-119), such as lipophilicity (it), electronegativity (ct), hydrogen bonding capacity (HB), and substituent width (AB5). Multiparameter linear regression analysis of the receptor binding data for sixteen 7-substituled2,3-dichlorodibenzo-p-dioxins gave the following equation (I): log (l/ECso) = I.24tt + 6.11 1. (n = 14, j = 0.29, r = 0.950), where it is the substituent lipophilicity, s is the standard deviation, and r is the correlation coefficient. The only substituents treated as outliers for the deriva tion of this equation were the bulky C6Hj and t-C4H9groups, which possess van der Waals volumes of 48.5 and 41.8 cm3/mol, respectively. This suggests that substituent molecular volumes are also important structural determinants for determining ligand affinities for the receptor protein binding site. Previous studies with the 4'-substituted-2,3,4,5-tetrachlorobiphenyls indicated that the maximum molecular volume for lateral substituents was < 35 cm3/mol (118). However it is apparent that if substituent molecular volume requirements are satisfied, the the receptor binding affinities of these analogs are directly related to the lipophilicity of the 7-substituents. The competitive receptor binding affinities of a series of thirteen 8substituted-2,3,4-trichlorodibenzofurans and ten 8-substituted-2,3-dichlorodibenzofurans (Figure 4) have also been determined, and the competi- PCDDs A N D PCDFs-- SARs 383 X X = i-C 4H9 , F, Br. I. -C3H7 , Cl, CH 3 i OCH3 i OH, H a X = C I, Br, CF3 , I. F, CH 3 i i-C 3H 7 , C2Hs , t C4H9, H, OCH3 , OH, C H jB r F igure 4 Slruclures of substituted PCDDS, PCDFs, and polychlorinated biphenyls used for QSAR studies. live displacement binding data results have been analyzed by multiparameter linear regression analysis to give Equations 2 and 3, respectively. log (1/ECjo) = 1.09ir + 5.77 2. log (1/ECso) = l.lOir + 5.19 3. For the 8-substituted-2,3,4-trichlorodibenzofurans, both the t-C4H9and i-C3H7 were outliers, whereas only the t-C4H9 substituent was not included in the derivation of Equation 3. Analysis of the collective data for 33 substituted polychlorinated dibenzofurans and dibenzo-p-dioxins (Figure 5) has demon strated the excellent linear correlation between the log (1/EC50) receptor bind ing data and lipophilicity (it ). Moreover the slopes and intercepts for Equations 1-3 were not significantly different. These data are consistent with a receptor protein that binds the PCDDs and PCDFs at a common binding site(s) on the protein; this site must accommodate the molecular area and volume encum bered by these ligands, and the QSAR results are consistent with a binding site that is highly hydrophobic. The effects of substituent structure on the rat hepatic cytosolic receptor binding affinities of 4-substituted-2,3,4,5-tetrachlorobiphenyls have also been 384 SAFE (-a CO tDs O QO f-u* <0 lO 'f (dQ-`2-X-8) 0fi036oT- *0 .60 PCDDs A N D PCDFs-- SARs 385 reported (118). Multiparameter linear regression analysis of the results of these analogs gave Equation 4, which suggests that substituent lipophilicity, electronegativity (a) and hydrogen bonding capacity (HB) are factors that influence receptor-ligand affinities: log (1/EC50) = 1.39 a + 1.3 lir + 1.12 HB + 4.20 4. The bulky t-C4H9 and CHj substituents were also treated as outliers for the derivation of Equation 4. These results demonstrate that substituent molecular volumes play a role in ligand-receptor interactions for the substituted PCBs, PCDDs, and PCDFs. However, the data analysis for the former group of analogs suggests that the critical 4'-substituent for the PCBs interacts with both polar and hydrophobic regions of the receptor binding site. The differences between Equations 1-3 and 'Equation 4 are somewhat paradoxical since molecular overlap of the four sets of substituted ligands does not indicate that there are major differences in their spatial orientation (Figure 4). Therefore it is likely that the differences observed for the substituted PCBs are due to the free rotation about the Ph-Ph bond and alimited population of the copianarconformers. Chlorinated biphenylenes that possess a fixed coplanar ring structure exhibit binding affinities comparable to those of 2,3,7,8-TCDD and related isostereomers; this emphasizes the importance of aplanarring system. It is also possible (hat the receptor binding sites for the substituted PCBs and PCDDs/ PCDF are not identical, a problem currently being investigated in my labora tory. ahh in d u ctio n a ctiv ities o f pcdds and pcdfs The in vivo and in vitro SARs for PCDDs and PCDFs sinducers of hepatic andextrahepatic AHH have been reported by several groups (37, 38, 47-49, 51-53). The most active PCDDs were substituted in tlieir 2,3,7, and 8 position; inspection of the data in Table 1 indicates that there were comparable SARs for PCDDs as ligands for the receptor protein and as AHH inducers; however, there is not a linear correlation between these two bioassays. SARs for several PCDF congeners as in vitro AHH inducers were comparable to those already discussed for receptor binding. Moreover, for the PCDFs summarized in Table 1 a comparison of in vitro ECjovalues for AHH induction in rat hepatoma H-4-11Ecells and in vivo EDsos for AHH induction in male Wistar rats showed a linear correlation between these two values. Like the PCDD congeners, however, there was not a strong correlation between AHH induction potencies and receptor binding avidities for the PCDF congeners. A comparison of the AHH and EROD induction potencies of the 7-substituted-2,3-dichlorodibenzo-p-dioxins with their rat hepatic cytosolic receptor binding avidities also showed that there was not a linear correlation between the two in vitro activities for this series of 386 SAFE analogs (116). Multiple parameter linear regression analysis of the AHH induction results for these compounds gave the following equation: log (1/EC50)auh = 1.60 tt - 0.33(ABS)2 + 5.85 5. Like the receptor binding avidities for these substituted PCDDs, their AHH induction potencies were dependent on substituent lipophilicity; however, a second parameter, STERIMOL (AB5) has also been included in the derivation of this equation. The STERIMOL parameter (119) is a measure of the max imum width of the substituents (compared to H) from the axis connecting the 7-substituent to the rest of the molecule and has previously been used in QSAR studies that involve the interaction of substituted organic ligands and macro molecules. The dependence of AHH induction activities by the substituted PCDDs on both ABSand it suggests that substituent-dependent effects such as conformational changes in the ligand-receptor complex occur after the initial receptor-ligand binding process. The AHH and EROD induction EC50values for the 8-substituted-2,3-di- and 2,3,4-trichlorodibenzofurans in rat hepatoma H-4-11Ecells were also subjected to multiple parameter linear regression analysis to give Equations 6 and 7, respectively (117). log (1/EC50)a,ih = 0.80-rr + 0.87AB5 - 0.35(ABS)2 + 4.63 6. log (1/EC50)a,ui = 0.76-n + 1.11AB5 + 2.23 a p + 6.78 7. Both Equations 5 and 6 showed that AHH induction potencies for the 8substituted-2,3-dichlorobenzofurans and 2,3-dichlorodibenzo-p-dioxins were dependent on substituent and AB3 parameters. The correlation for the more highly chlorinated set of analogs, the 8-substituted-2,3,4-trichlorodibenzofurans, also includes a Hammett substituent parameter (ap); presumably the requirement for a p must be due to the effects of the C-4 chlorine group, which constitutes the only structural difference between the two sets of substituted PCDF analogs. Equations 8-10 were developed from the AHH and EROD induction data for the 8-substituted-2,3-dichlorodibenzo-p-dioxins, -2,3-dichlorodibenzofurans, and -2,3,4-trichlorodibenzofurans, respectively. log (I/ECsoJerod = 0-99 log ( 1/EC5o)amm --0-07 8. log (1/EC50)EROd = 0.90 log (1/EC50)ahh + 0.83 log (1/EC50)Erod = 0.92 log (1/EC50)ahh + 0.28it + 0.27 9. 10. PCDDs AN D PCDFs-- SARs 387 For a total of twenty-five 7-substituted-2,3-dichlorodibenzo-p-dioxins and 8-substiluted-2,3-dichlorodibenzofurans and fifteen 4'-substituted-2,3,4,5tetrachlorobiphenyls there was a linear correlation between the EC50values for AHH and EROD induction, and the slopes for these equations were not significantly different from that of equation one. These data suggest that both ethoxyresorufin and benzo[a]pyrene are catalyzed by the same cytochrome P-450 isozyme(s). In contrast, Equation 10 required a it term to correlate the effects of the substituted 2,3,4-trichlorodibenzofurans; the rationale for these differences between the two sets of substituted PCDF analogs is unknown. OTHER BIOLOGIC EFFECTS OF PCDDs AND PCDFs: STRUCTURE-ACTIVITY r e l a t io n s h ip s PCDDs and PCDFs elicit a broad spectrum of speciesdependent biologic effects, but because pure standards have been unavailable few studies report qualitative or quantitative SARs for these compounds. The SARs for PCDDs as inducers of ALA synthetase were comparable to the effects of structure on their activities as AHH inducers (48). Knutson & Poland have used cultured XB cells derived from a mouse teratoma as an in vitro model for halogenated aryl hydrocarbon toxicity (120). 2,3,7,8-TCDD and related toxic isostereomers produce a dose-dependent keratinization response which in part resembles the in vivo dermal toxicity that develops in some animals after exposure to these toxins. The most active PCDD congeners in this in vitro assay possessed three or four lateral substituents, and the SARs were similar to those reported for theirreceptorbinding affinities. 2,3,7,8-TCDD causes comparable dermal toxicity in cultures of newborn foreskin keratinocytes (121). Several human squamous cell carcinoma (SCC) lines have been utilized as model systems for investigating the mechanism of action of toxic halogenated aryl hydrocarbons (110, 122). SCC cells possess variable Ah receptor levels, and the relative amount of receptor in several cell lines correlates with the maximal 7-ethoxycoumarin 0-deethylase inducibility in these cell lines. 2.3.7.8- TCDD causes down-regulation of the epidermal growth factor (EGF) receptor in the SCC-12F cell line, and this effect is dose- and structuredependent. Both 2,3,7,8-TCDD and 2,3,7,8-tetrabromodibenzofuran, which exhibit a high affinity for the Ah receptor, decrease EGF receptor binding, whereas 2,7-dichlorodibenzo-p-dioxin is inactive (123). Comparable results have been observed in keratinocyte strains derived from normal neonatal foreskin; the authors report that 2,3,7,8-TCDD acts (in part) through the Ah receptor in epidermal basal cells to enhance terminal differentiation (123). 2.3.7.8- TCDD also down-regulates EGF receptor activity in hepatic plasma membranes in several animal species and cultured mouse hepatoma cells (124, 125); however, structure-activity effects in the rat do not necessarily support the role of the Ah receptor in mediating this process (125). 388 SAFE TOXICOLOGY OF PCDDs AND PCDFs: STRUCTURE-ACTIVITY RELATIONSHIPS The dose-response acute toxicities of nine PCDD isomers and congeners in the guinea pig and responsive mouse have been reported (33, 34, 126). The relative LD50 values in both species were highly dependent on the number of lateral Cl substituents and the degree of substitution; their rank order of toxic potencies was similar to their in vitro receptor binding and AHH induction activities as discussed above. A comparative study (127) of the toxicity of 2,3,7,8-TCDF, 2,3,7,8-tetrabromodibenzofuran, and 2,3,4,7,8-pentachlorodibenzofuran in guinea pigs, mice, and rhesus monkeys confirms that these compounds elicit the characteristic broad spectrum of toxic effects observed for 2,3,7,8-TCDD and related isostereomers. Poland & Glover (75) reported the effects of several PCDD congeners and other toxic halogenated aryl hydrocarbons on genetically inbredstrains of mice. Although dose-response studies were carried out only with 2,3,7,8-TCDD, the relative toxicities (i. e. thymic atrophy) of these compounds correlated with their in vitro binding and induction activities. The dose-response toxicities (thymic atrophy and body weight loss) of the PCDFs listed inTable 1have been determined in the immature male Wistar rat (47). Inspection of the ED50 data for the toxic effects showed that the potencies of these congeners were struc ture-dependent and that the in vivo SARs for toxicity were identical to those observed for their in vitro AHH induction potencies. Figure 6 summarizes aplot of the --log EDS0values for thymic atrophy and body weight loss in immature male Wistar rats vs their in vitro AHH induction activities. The linear correla tion constant (r) and slope for the plots of the reciprocal log values for AHH induction vs body weight loss were 0.96 and 1.30 (slope), respectively, and values of 0.88 (r) and 1.16 (slope) were obtained for the comparable plot of the reciprocal log values for AHH induction vs thymic atrophy. The linear correla tion was observed only for those compounds that do not contain vicinal unsubstituted carbon atoms and are not significantly metabolized. Forexample, the toxicity of 1,2,3,7- or 2,3,4,8-TCDF in the rat was lower than predicted by the in vitro AHH (or EROD) induction data (not shown) owing to in vivo metabolism. Current research in my laboratory (S. H. Safe, unpublished results) indicates that for several PCDD isomers and congeners there is a linear correlation between --log EC50 (AHH induction) and --log ED50 (thymic atrophy and body weight loss in the rat). These results suggest that the rat hepatoma cell monooxygenase induction bioassay may serve as a short-term test system for predicting the toxicities of PCDDs, PCDFs, and related haloge nated aryl hydrocarbons. Poland and co-workers have investigated the effects of 2,3,7,8-TCDD, several PCDD congeners, and related halogenated aryl hydrocarbons inthe skin of inbred HRS/J hairless mice segregating for the hr locus (128, 129). The homozygous hrlhr hairless and heterozygous hr/+ hairedmice exhibit identical ~\ O* ./ / i(7> % 1.I 1 i23. k53 tn 1 l _l_____ I______ I______I-- OD N (O03 Bo|-) pu| HHV <0 IO ! -- 2A7.8-TCOD Figure 6 A plot of the - log EC values for in vitro AHH induction vs the - log'ED values for thymic atrophy (right) and body weight loss (left) in the male Wistar rat for several PCDF congeners and 2,3,7,8-TCDD. PCDDs A N D PCDFs-- SARs 389 8 aui I? I (vl 390 SAFE genetics except for one allele at the hr locus. Topical application of 2,3,7,8TCDD to the dorsal skin of hairless mice resulted in epidermal hyperplasia, sebaceous gland metaplsia, and hyperkeratosis, but these histological lesions are not observed in hrl+ haired mice. The development of a graded epidermal response by several PCDD congeners, 2,3,7,8-TCDF, and related toxic aryl hydrocarbons was structure-dependent and correlated with rankorder of recep tor binding affinities for these compounds (128). For example, the total dose (nmol/mouse) required to produce a2+ response was 0.36,1.76,1.2> 360 and > 360 for 2,3,7,8-TCDD, 1,2,3,6,7,8-HCDF, 2,3,7,8-TCDD, 1,3,6,8TCDD, and 2,7-DCDD, respectively. These results are consistent with the involvement of both theAh and hr loci in the development of murine epidermal lesions after exposure to PCDDs, PCDFs, and related toxins. 2,3,7,8-TCDD can also act as a potent tumor promoter in HRS/J mice, and the results of this study confirm the segregation of the activity with the A h and hr loci. A qualitative structure-activity study also suggests a possible role for the Ah receptor in mediating the tumor promotion activities of the toxic halogenated aryl hydrocarbons (130). The SARs for PCDDs and PCDFs clearly support an Ah receptor-mediated mechanism of action for these compounds; comparable studies have been reported for other classes of halogenated aryl hydrocarbons. It is assumed that the persistent effects elicited by these toxins are related to a sustained receptorligand occupancy of nuclear binding sites, but this has not yet been demon strated experimentally. The identity and role of any endogenous ligand(s) for the Ah receptor have not been determined; however, lumichrome, a riboflavin metabolite, does bind to the receptor (131). Genetic Evidence Pharmacogenetic studies with genetically inbred strains of mice typefied by (he Ah-responsive C57BL/6 and nonresponsive DBA/2 mice have provided strong evidence in support of the role of the Ah receptor in mediating the biologic and toxic effects of toxic halogenated aryl hydrocarbons. Nonresponsive DBA/2 mice contain relatively low levels of hepatic orextrahepatic cytosolic or nuclear Ah receptor (< 1fmol/mg cytosolic protein), whereas much higher levels of the receptor are detected in responsive strains of mice. The ED50 for 2,3,7,8TCDD mediated hepatic microsomal AHH induction in C57B1/6J mice was 1 * nmol/kg whereas this value is at least tenfold higher in DBA/2J mice. Ingenetic crosses and backcrosses between C57BL/6J and DBA/2J mice the trait or W* responsiveness to AHI1 induction is inherited in a simple autosomal mode (50, CO 132-134). The responsive backcross animals also had detectable hepatic recep t tor levels (85, 89,97,103,109). The segregation of the toxicity of PCDDs and "O pcDFs with the Ah locus has been determined primarily with 2,3,7,8-TCDD using both responsive and nonresponsive genetically inbred mice and their PCDDs A N D PCDFs-- SARs 391 crosses and backcrosses. The results illustrate that several toxic effects includ ing teratogenicity, porphyria and hepatotoxicity, immunotoxicity, and body weight loss segregate with the Ah locus (75-80, 135-139). Dermal toxic lesions appear tobe dependent on the interaction between the Ah andhrlocus as previously noted (128-130). It has also been suggested that additional genetic loci may also be involved in the hepatotoxic effects of 2,3,7,8-TCDD, however this observation requires further substantiation (140). PCDDs, PCDFs, AND RELATED COMPOUNDS-- INTERACTIVE EFFECTS Although the SARs and toxicology of PCDDs and PCDFs have been ex tensively studied the interactive effects of PCDD/PCDF mixtures and related compounds are not well understood. 2,3,7,8-TCDD can act as a tumor promot er for several initiators in the rodent liver (141), mouse skin (130), and C3H/10T1/2 cells (142), and as acocarcinogen causing 3-methylcholanthreneinitiated subcutaneous tumors in nonresponsive DBA/2 mice (143). Incontrast, 2,3,7,8-TCDD exhibits anticarcinogen activity in female CD-I mice (144). These effects are related to the agent's induction of drug-metabolizingenzymes that alter rates of metabolic activation of polynuclear aromatic hydrocarbon initiators. Several studies (51, 53, 120, 145, 146) report the application of in vitro bioassays as short-term tests for assessing the potential toxicity of PCDD/ PCDF mixtures; a comparison of the in vitro AHH induction activity of a reconstituted mixture of PCDFs identified in Yusho patients (147) and the toxicity of this mixture (Figure 6) suggest that the effects of the individual PCDFs in this mixture are additive (148). In contrast, the immunotoxicity and AHH induction activity of 2,3,7,8-TCDD in C57BL/6 mice were decreased by coadministering a nontoxic or noninducing dose of 2,3,7,8-TCDF (10 p.g/kg) (149). A rational explanation for the antagonistic effects of 2,3,7,8-TCDF is not apparent. Bimbaum and co-workers (77) have reported that treatment of pregnant mice with a combination of 2,3,7,8-TCDD (3 p.g/kg) and a nontoxic dose of 2,3,3',4,4',5-hexachlorobiphenyl (20 p,g/kg) resulted in a tenfold increase in incidence in cleft palate compared to those animals receiving only 2,3,7,8TCDD. A second PCB congener, 2,2',4,4',5,5'-hexachlorobiphenyl, at dose levels of 50 or 25 mg/kg in combination with 2,3,7,8-TCDD (3p.g/kg), did not effect the teratogenic potency of the latter compound. Although 2,3,3',4,4',5hexachlorbiphenyl is less toxic than 2,3,7,8-TCDD, this monortho coplanar PCB congener elicits several receptor-mediated biologic and toxic effects (74). It is conceivable that the interactive effects of 2,3,3',4,4',5-hexachlorobiphenyl and 2,3,7,8-TCDD may be additive if the dose-response curve for the former compound is steep and if the 20 mg/kg dose level is just 392 SAFE below lhe minimum observable teratogenic dose. Current research in my laboratory has demonstrated that administration of several compounds, includ ing several polychlorinated biphenyl congeners, increase hepatic 2,3,7,8TCDD receptor levels in rats and C57BL/6 mice. Pretreatment of rats and mice with these receptor modulators followed by administration of 2,3,7,8-TCDD results in markedly increased hepatic AHH and EROD induction activities. Both positive and negative modulators and antagonists of the hepatic and extrahepatic cytosolic receptor protein are currently being investigated as probes for delineating the mechanism of action of PCDDs and PCDFs and the role of the receptor protein in mediating these effects. These interactive studies will also be important for assessing the effects of polyhalogenated aromatic environmental pollutant mixtures and their potential human health impact. Acknowledgments The author gratefully acknowledges the financial assistance of the National Institutes of Health (EC-03554), the Environmental Protection Agency, and the Texas Agricultural Experiment Station. The collaboration of G. Mason, M. A. Dnomm, K. Homonko, L. Safe, B. Keys, S. Bandiera, T. Sawyer, M. Romkes, J. Piskorska-Pliszczynska, B. Zmudzka, andT. Fujita are gratefully acknowledged. Literature Cited 1. Bowes, G. W., Mulviliill, M. J., Simoneit, B. R. T., Burlingame, A. L., Risebrough, R. W. 1975. Identification of chlorinated dibenzofurans in Amer ican polychlorinated biphenyls. N ature 256:305-7 2. Vos, J. G., Koeman, J. H., Van der Maas, H. L., Ten Noeverde Brauw, M. C.. de Vos, R. H. 1970. Identification and toxicological evaluation of chlori nated dibenzofuran and chlorinated napthalene in two commercial polychlorin ated biphenyls. F o o d C osm et. Toxicol. 8:625-33. 3. Rappe, C., Buser, H. R. 1980. Chemical properties and analytical methods. In H alogenated Biphenyls, Terphenyls, Napthalenes, Dibenzodioxins and R e la ted Products, ed. R. D. Kimbrough, pp. 41-76. Amsterdam: Elsevier. 406 pp. 4. Morita, M., Nakagawa, J., Rappe, C. 1978. Polychlorinated dibenzofuran (PCDF) formation from PCB mixture by heat and oxygen. B u ll. E nviron. C ontam . T oxicol. 19:665-70 5. Kunila, N., Kashimoto, T., Miyata, H., Fukushima, S., Hori, S., Obana, H. 1984. Causal agents of Yusho. A m . J . In d . M ed. 5:45-58 6. Masuda, Y., Yoshimura, H. 1984. Poly chlorinated biphenyls and dibenzofurans in patients with Yusho and their toxicolo gical significance. A m . J . In d . M ed. 5:31-44 7. National Research Council of Canada. 1978a. P hen o xy H erbicides-- T h eir E f fe c ts on E nvironm ental Q uality, Ottawa NRCC Publication No. 16075, 440 pp8. Firestone, D. 1977. Determination of polychlorodibenzo-p-dioxins and polychlorodibenzofurans in commercial gela tins by gas liquid chromatography. J . A g ric. F ood. Client. 25:1274-80 9. Rappe, C., Marklund, S., Buser, H. R., Bosshardt, H. P. 1978. Formation of polychlorinated dibenzo-p-dioxins heat ing chlorophenates. C hem osphere 7: 269-81 10. Ballschmiter, K., Zoller, W., Scholz, C. H., Notlrodl, A. 1983. Occurrence and absence of polychlorodibenzofurans and polychlorodibenzodioxins in fly ash from municipal incinerators. C hem osphere 12: 585-94 O, r PCDDs AN D PCDFs-- SARs 393 11. Buser, H. R., Bosshardl, H. P., Rappe, C. 1978. Identiflcation of polychlori nated dibenzo-p-dioxin isomers found in fly ash. C hem osphere. 7:165-72 12. Buser, H. R. 1979. Formation of polych lorinated dibenzofurans (PCDFs) and di- benzo-p-dioxins (PCDDs) from the pyrolysis of chlorobenzenes. C hem o sphere 8:415-24 13. Olie, K., Vermeulen, P. L., Hutzinger, O. 1977. Chlorodibenzo-p-dioxins and chlorodibenzofurans are trace com ponents of fly ash and flue gas of some municipal incinerators in The Nether lands. C hem osphere 6:455-59. 14. Rappe, C., Marklund, S., Bergqvist, P. A., Hansson, M. 1983. Polychlorinated dioxins, dibenzofurans and other poly nuclear aromatics formed during in cineration and PCB Tires. In C hlorinated D ioxins and D ibenzofurans in the Total E nvironm ent, ed. G. Choudhary, L. H. Keith, C. Rappe, pp. 99-124. Boston: Bulterworlh 15. Bumb, R. R., Crummelt, W. B., Cutie, S. S ., Gledhill, J. R., Hummel, R. H., et al. 1980. Trace chemistries of fire: A source of chlorinated dioxins. Science 210:385-90 16. Crummelt, W. B., Townsend, D. I. 1984. The trace chemistries of Are: re view and update. C hem osphere 13:777-- 88 17. Czuczwa, J. M., McVeety, B. D., Hites, R. A. 1984. Environmental fate of com bustion-generated polychlorinated di oxins and furans. E n viro n . S ci. T echnol. 18:444-50 18. Czuczwa, 1. M., McVeety, B. D., Hites, R. A. 1984. Polychlorinated dibenzo-pdioxins and dibenzofurans in sediments from Siskiwit Lake, Isle Royale. Science 226:568-69 19. Rappe, C., Bergquist, P. A., Hansson, M., Kjeller, L-O., Lindstrom, G., et al. 1984. In B io lo g ica l M ech a n ism s o f D i o xin A ctio n , ed. A. Poland, R. D. Kim brough, pp. 17-25. Banbuiy Report 18, Cold Spring Harbor Laboratory. 500 pp. 20. Lalhrop, G. D,, Wolfe, W. D., Albanese, R. A., Moynahan, P. 1984. An epidemiologic investigation of health effects in Air Force personnel following exposure to herbicides. In M echanism s o f D ioxin A ction, ed. A. Poland, R. D. Kimbrough, pp. 471-74. Banbury Re" port 18, Cold Spring Harbor Laboratory. 500 pp. 21. Fingerhut, M. A., Halperin, W. E., Hon- char, P. A., Smith, A. B., Groth, D. H., Russell, W. O. 1984. An evaluation of reports of dioxin exposure and soft tissue sarcoma pathology in U.S. chemical workers. In B iological M echanism s o f D ioxin A ctio n , ed. A. Poland, R. D. Kimbrough, pp. 461-70. Banbury Re port 18, Cold Spring Harbor Laboratory. 500 pp. 22. Fara, G., Del Corono, G., Bonelti, F., Caramaschi, F., Dardanoni, L., et al. Chloracne after release of TCDD at Sevesco, Italy. In C hlorinated D ioxins a nd Related Compounds: Impact on the E nvironm ent, ed. O. Hutzinger, R. Frei, E. Merian, F. Pocchiari, pp. 545-59. Oxford: Pergamon 658 pp. 23. Kimbrough, R. D., Falk, H., Stehr, P., Fries, G. 1984. Health implications of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) contamination of residential soil. J . T oxicol. E nviron. H ealth 14:47-- 93 24. Zack, J., Gaffey, W. 1983. A mortality study of workers employed at the Mon santo Company plant in Nitro, West Vir ginia. In H um an a n d E nvironm ental Risks o f Chlorinated D ioxins and Related C om pounds, ed. R. Tucker, A. Young, A. Gray, pp. 575-91. New York: Ple num. 823 pp. 25. Masuda, V., Kuroki, H., Yamaryo, T., Haraguchi, K., Kuralsune, M., Hsu, ST. 1982. Comparison of causal agents in Taiwan and Fukuoka Japan polychlori nated biphenyl poisoning. C hem osphere 2:199-206 26. Hsu, S-T., Ma, C-L, Kwo-Hsiung, S., Wu, S-S., Hsu, N. H-M., Yeh, C-C. 1984. Discovery and epidemiology of PCB poisoning in Taiwan. A m er. J . Ind. M ed. 5:71-79 27. Kashimoto, T., Miyata, H., Shiriji, K., Tung, T., Hsu, S-T., etal. 1981. Role of polychlorinated dibenzofuran in Yusho (PCB Poisoning). A rch. E nviron. H ealth 36:321-26 28. Norstrom, R. J., Hallet, D. J., Simon, M., Mulvihill, M. J. 1982. Analysis of great lakes herring gull eggs for tetrachlorodibenzo-p-dioxins. In C hlorinated D i oxins and Related Compounds. Impact on the E nvironm ent, ed. O. Hutzinger, R. W. Frei, E. Merian, F. Pocchiari, pp. 173-81. Oxford: Pergamon 658 pp. 29. Van Den Berg, M., Olie, K., Hutzinger, O. 1985. Polychlorinated dibenzofurans (PCDFs): environmental occurrence, physical, chemical and biological proper ties. Toxicol. E nviron. C hem . 9:171--217 30. Rappe, C., Buser, H. R., Stalling, D. L., Smith, L. M., Dougherty, R. C. 1981. Identiflcation of polychlorinated di benzofurans in environmental samples. N a tu re 292:524-26 394 SAFE 31. Tieman.T. O., Taylor, M. L.,Garrett, J. H., Van Ness, G. F., Solch, J. G., et al. 1983. Sources and fate of polychlori nated dibenzo-p-dioxins, dibenzofurans and related compounds in human en vironments. E n viro n . H ealth P ersp. 59:145-158 32. Stalling, D. L., Smith, L. M., Petty, J. D. , Hogan, J. W., Johnson, J. L., et al. 1983. Residues of polychlorinated di benzo-p-dioxins and dibenzofurans in Laurentian Great Lakes fish. In H um an and Environmental Risks o f Chlorinated D io xin s a n d R e la te d C o m pounds, ed. R. E. Tucker, A. L. Young, A. P. Gray, pp. 221-40. New York: Plenum. 823 pp. 33. Kociba, R. J., Schweiz, B. A. 1982. Toxicity of 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD). D rug M etabolism R e view s 13:387-406 34. Kociba. R. J., Schwetz, B. A. 1982. A review of the toxicity of 2,3,7,8-letrachlorodibenzo-p-dioxin . (TCDD) with a comparison of the toxicity of other chlorinated dioxin isomers. A ssoc. F ood D ru g O fficia ls Q uart. B u ll. 46:168- 88 35. Lowrance, W. W., ed. 1984. P ublic H ea lth R isks o f th e D io xin s. New York: Rockefeller University Press. 389 pp. 36. National Research Council of Canada. 1981. P o ly ch lo rin a ted dibenzo-p-dio- xins: Criteria fo r Their Effects on M an a n d his E nvironm ent. NRCC Publication No. 18574. 251 pp. 37. Poland, A., Greenlee, W. F., Kende, A. S. 1979. Studies on the mechanism of action of the chlorinated dibenzo-pdioxins and relied compounds. A n n . N .Y . A ca d . S ci. 30:214-30 38. Poland, A., Knutson, J. C. 1982. 2,3,7,8-Telrachlorodibenzo-p-dioxin and related halogenalcd aromatic hydro carbons: examination of the mechanisms of toxicity. A n n . R ev. P ha rm a co l. T o x icol. 22:517-24 39. Reggiani.G. 1983. Toxicology of TCDD and related compounds: Observations in man. C hem ospltere 12:463-75 40. Poiger, II., Schlatter, C. 1983. Animal toxicology of chlorinated dibenzo-pdioxins. C hem ospltere 12:453-62 41. McConnell, E. E. 1980. Acute and chronic toxicity, carcinogenesis, repro duction leratogcnesis and mutagenesis in animals. In H a lo g en a ted B iphenyls, Ter- phenyls. Naphthalenes, Dibenzodioxins a n d R elated P roducts, ed. R. D. Kim brough, pp. 109-50. Amsterdam: Else vier. 406 pp. 42. Matsumura, F. 1983. Biochemical as pects of action mechanisms of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related chemicals in animals. P harm a c. Ther. 19:195-209 43. Neal, R. A., Beatty, P. W., Gasiewicz, T. A. 1979. Studies on the mechanisms of toxicity of 2,3,7,8-telrachlorodibenzo-p-dioxin (TCDD). A n n . N . Y. A ca d . Sci. 320:204-13 44. Poland, A. 1984. Reflections on the mechanism of action of halogenated aromatic hydrocarbons. In B iological M echanism s o f D ioxin A ction, ed. A. Po land, R. D. Kimbrough, pp. 109-17. Banbury Report 18, Cold Spring Harbor Laboratory. 450 pp. 45. McConnell, E. E. 1984. Clinicopathologic concepts of dibenzo-p-dioxin in toxication. In B iological M echanism s o f D ioxin A ction, ed. A. Poland, R. D. Kimbrough, pp. 27-37. Banbury Report 18, Cold Spring Harbor Laboratory. 450 PP46. Kociba, R. 1984. Evaluation of the carcinogenic and mutagenic potential of 2.3.7.8- TCDD and other chlorinated di oxins. In B iological M echanism s o f D i oxin A ction, ed. A. Poland, R. D. Kim brough, pp. 73-84. Banbury Report 18, Cold Spring Harbor Laboratory. 450 PP47. Mason, G., Sawyer, T., Keys, B., Ban- diera, S., Romkes, M., et al. 1985. Polychlorinated dibenzofurans (PCDFs): in vivo and in vitro quantitative structureactivity relationships (QSAR). Toxicol. 37:1-12 48. Poland, A., Glover, E. 1973. Chlori nated dibenzo-p-dioxins: Potent inducers of 8-aminolevulinic acid synthetase and aryl hydrocarbon hydroxylase II. A study of the structure activity relationship. M o l. P harm acol. 9:737-47 49. Yoshihara, S., Nagata, K., Yoshimura, H., Kuroki, H., Masuda, Y. 1981. In ductive effect on hepatic enzymes and acute toxicity of individual polychlori nated dibenzofuran congeners in rats. T o xico l. A p p l. P h arm acol. 59:580-88 50. Poland, A. P., Glover, E., Robinson, J. R., Nebert, D. W. 1974. Genetic expres sion of aryl hydrocarbon hydroxylase activity. Induction of monooxygeuase activities and cytochrome P-450 forma tion by 2,3,7,8-telrachlorodibenzo-pdioxin in mice generally "nonresponsi ve" to other aromatic hydrocarbons. J . Biol. Client. 249:5599-606 51. Bandiera, S., Sawyer, T., Romkes, M., Zmudzka, B., Safe, L., et al. 1984. Polychlorinated dibenzofurans (PCDFs): effects of structure on binding to the 2.3.7.8- TCDD cytosolic receptor pro tein, AHH induction and toxicity. Tox icol. 32:131-44 PCDDs A N D PCDFs-- SARs 395 52. Nagayama, J., Kuroki, H., Masuda, Y., Kuratsune, M. 1983. A comparative study of polychlorinated dibenzofurans, polychlorinated biphenyls and 2,3,7,8- tetrachlorodibenzo-p-dioxin on aryl hy drocarbon hydroxylase inducing potency in rats. A rch . Toxicol. 53:177-84 53. Bradlaw, J. A., Casterline, J. L. 1979. Induction of enzyme activity in cell cul ture: a rapid screen for detection of planar polychlorinated organic compounds. J . A sso c . O ff. A n a l. C hem . 62:904-16 54. Goldstein, J. A., Linko, P. 1984. Differ ential induction of 2,3,7,8-telrachlorodibenzo-p-dioxin-inducible forms of cy tochrome P-450 in exlrahepatic versus hepatic tissues. M o l. P harm acol. 25: 185-91 55. Tukey, R. H., Negishi, M., N.ebert, D. W. 1982. Quantitation of hepatic cyto chrome Pr 450 mRNA with the use of cloned DNA probe: effects of various P450 inducers m C57BL/6N am) DBA/2N mice. M o l. P harm acol. 22:779-86 56. Tukey, R. H., Nebert, D. W. 1984. Regulation of mouse cytochrome Pj-450 by the Ah receptor studies with P3-450 DNA clone. B iochem . 23:6003-8 57. Kimura, S., Gonzalez, F. J., Nebert, D. W. 1984. The murine Ah locus. Com parison of the complete cytochrome P,- 450 and Pj-450 DNA nucleotide and ami no acid sequences. J . B io l. C hem . 259:10705-13 58. Thomas, P. E., Reedy, J., Reik, L. M., Ryan, D. E., Koop, D. R., Levin, W. 1984. Use of monoclonal antibodies probes against rat hepatic cytochromes P-450c and P-450d to detected immunochcmically related isozymes in liv er microsomes from different species. A rc h . B io ch em . B io p h ys. 235:23953 59. Serabjit-Singh, C. J., Albro, P. W., Robertson, I. G. C., Philpol, R. M. 1983. Interaction between xcnobiotics that increase or decrease the levels of cytochrome P-450 isozymes in rabbit lung and liver. J . B iol. C hem . 258: 12827-34 60. Lient, H. H., Muller-Eberhard, U., Johnson, E. F. 1980. Differential induc tion of 2,3,7,8-telrachlorodibcnzo-pdioxin of multiple-forms of rabbit micro somal cytochrome P-450: evidence or tissue specificity. M ol. P harm acol. 18: 565-70 ' 61. Baars, A. J., Jansen, M., Breimer, D. D. 1978. The influence of phnobarbital, 3methylcholanlhrene and 2,3,7,8-telra- chlotodibenzo-p-dioxin on glutathione Stransferase activity of rat liver cytosol. B iochem . P harm acol. 27:2487-94 62. Owens, I. S. 1977. Genetic regulation of UDP-glucuronosyltransferase induction by polycyclic aromatic hydrocarbons in mice. J . B io l. C hem . 252:2827-33 . 63. Thunberg, T., Ahlborg, U. G., Wahl- strom, B. 1984. Comparison between the effects of 2,3,7,8-tetrachIorodibenzo-pdioxin and six other compounds on vita min A storage, the UDP-glucuronosyl transferase and the aryl hydrocarbon hy droxylase activity in the rat liver. Arch. T oxicol. 55:16-19 64. Beatty, P. W., Neal, R. A. 1976. Induc tion of DT-diaphorase by 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD). Biochem . Biophys. Res. Commun. 68:197-204 65. Nebert, D. W., Jensen, N., Peny, J., Oka, T. 1980. Associations between ornithine decarboxylase E.C.-4.I.1.I7 induction and the Ah locus in mice treated with polycyclic aromatic com pounds. J . B io l. C hem . 255:683642 66. Puhvel, S. M., Ertl, D. C., Lynberg, C. A. 1984. Increased epidermal transgluta minase activity following 2,3,7,8-tetra- chlorodibenzo-p-dioxin: In vivo and in vitro studies in mouse skin. Toxicol. A p p l. P harm acol. 73:42-47 67. Kurl, R. N., Lund, J., Poellinger, L., Gustafsson, J-A. 1982. Differential ef fects of 2,3,7,8-tetrachlorodibenzo-pdioxin on nuclear RNA polymerase activ ity in the rat liver and thymus. Biochem . P harm acol. 31:2459-62 68. Sawahata, T., Olson, J. R., Neal, R. A. 1982. Identification of metabolites of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) formed on incubation with iso lated rat hcpatocytes. Biochem . Biophys. R es. C om m un. 105:341-46 69. Poland, A., Glover, E. 1979. An es timate of the maximum in vivo covalent binding of 2,3,7,8-letrachlorodibenzo-p- dioxin to rat liver protein, ribosomal RNA and DNA. C ancer R es. 39:334144 70. Weber, H., Poiger, H., Schlatter, C. 1982. Fate of 2,3,7,8-telrachlorodiben- zo-p-dioxin metabolites from dogs and rats. X enobiotica 12:353-57 71. Weber, IL, Poiger, H., Schlatter, C. 1982. Acute oral toxicity of TCDD metabolites in male guinea pigs. Toxicol. L ett. 14:177-82 72. Rozman, K., Rozman, T., Greim, H. 1984. Effect of thyroidectomy and thy roxine on 2,3,7,8-tetrachlorodibenzo-p- dioxin (TCDD) induced toxicity. T ox icol. A p p l. P harm acol. 72:372-76 73. Pazdemik, T. L., Rozman, K. L. Effect of thyroidectomy and thyroxine on 396 SAFE 2.3.7.8- tetrachlorodibenzo-p-dioxin - in duced immunoloxicity. Life Sci. 36: 695-703 74. Safe, S. 1984. Polychlorinated biphenyls (PCBs) and polybrominated biphenyls (PBBs): biochemistry, toxicology and mechanism of action. C R C C rit. R ev. T oxicol. 13:319-95 75. Poland. A.. Glover, E. 1980. 2.3,7,8- Tetrachlorodibenzo-p-dioxin: Segrega tion of toxicity with the Ah locus. M ol. P harm acol. 17:86-94 76. Pratt, R. M., Dencker, L., Diewert, V. M. 1984. 2,3,7,8-Tetrachlorodibenzo-pdioxin-induced cleft palate in the mouse: evidence for alterations in palatal shelf fusion. T eralogenesis C a rcin . a n d M u tagen 4:427-36 77. Bimbaum, L. S., Weber, H., Hanis, M. W . Lamb, J.C ., McKinney, J.D . 1985. Toxic interaction of specific polychlori nated biphenyls and 2,3,7,8-tetrachlorodibenzo-p-dioxin: increased incidence of cleft palate in mice. Toxicol. A p p l. P h a r m acol. 77:292-302 78. Weber, H., Lamb, J. C.. Hanis, M. W., Moore, J. A. 1984. Teratogenicity of 2.3.7.8- tetrachlorodibenzofuran (TCDF) in mice. Toxicol. L ett. 20:183-88 79. Hassoun, E., d'Argy, R., Dencker, L. 1984. Teratogenicity of 2,3,7,8- tetrachlorodibenzofuran in the mouse. J. T oxicol. E n viro n . H ealth 14:337-51 80. Hassoun, E., d'Argy, R., Dencker, L., Lundin, L. G.. Borwell, P. 1984. Ter atogenicity of 2,3,7,8-tetrachlorodibenzofuran in BXD recombinant inbred strains. Toxicol. L ett. 23:37-42 81. Scheutz, E. G., Wrighton, S. A., Bar- wick, J. L., Guzelian, P. S. 1984. Induc tion of cytochrome P-450 by glucocorti coids in rat liver. J . B iol. C hem . 259: 1999-2006 82. l.ax, R. E., Baumann, P., Scliriefcrs, II. 1984. Changes in the activities of micro somal enzymes involved in hepatic steroid metabolism in the rat after ad ministration of androgenic, estrogenic and progestational anabolic and catatoxic steroids. B iochem . J . 33:1235-41 83. Rumbaugh, R. C., McCoy, Z., Lucier, G. W. 1984. Correlation of hepatic cyto solic androgen binding proteins with an drogen induction of hepatic microsomal elhylmorphine N-demcthylase in the rat. J . S te ro id B iochem . 21:243-54 84. Leakey, J. A., Althaus, Z. R., Bailey, J. R., Slikker, W. 1985. Dexamelhasone increased UDP-glucuronosyl transferase activity towards bilirubin, oestradiol and testosterone in foetal liver from rhesus monkey during late gestation. B iochem . J . 225:183-88 85. Poland, A. P., Glover, E., Kende, A. S. 1976. Stereospecific, high affinity bind ing of 2,3,7,8-tetrachlorodibenzo-pdioxin by hepatic cytosol. Evidence that the binding species is the receptor for the induction of aryl hydrocarbon hydroxy lase. J . B iol. C hem . 251:4936-46 86. Greenlee, W. F., Poland, A. P. 1979. Nuclear uptake of 2,3,7.8-tetrachlorodibenzo-p-dioxin in C57BL/6J and DBA/ 2J mice. J . B io l. C hem . 254:9814-21 87. Carlstedt-Duke, J., Elfstrom, G., Snochowski, M., Hogberg, B., Gustafs- son, i-A. 1978. Detection of the 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD) re ceptor in rat liver by isoelectric focusing in polyacrylamide gels T oxicol. L ett. 2:365-73 88. Hannah, R. R., Nebert.D. W., Eisen, II. J. 1981. Regulatory gene product of the A h complex. Comparison of 2,3,7,8tetrachlorodibenzo-p-dioxin and 3-methylcholanthrene binding to several moiet ies in mouse liver cylocol. J . B iol. C hem . 256:4584-90 89. Okey, A. B., Bondy, G. P., Mason, M. E., Kahl, G. F., Eisen. H. J., etal. 1979. Regulatory gene products of the A h re ceptor complex and evidence for its nu clear translocation. J . B io l. C hem . 254:11636-48 90. Gasiewicz, T. A., Neal, R. A. 1982. The examination and quantitation of tissue cytosolic receptors for 2,3,7,8-tetrachlorodibenzo-p-dioxin using hydroxyapa tite. A n a l. B io ch em ., 124:1-11 91. Gasiewicz, T. A., Rucci, G. 1984. Ex amination and rapid analysis of hepatic cytosolic receptors for 2,3,7,8-tetrachlorodibenzo-p-dioxin using gel permeation high performance liquid chromatogra phy. B iochem . B iophys. A c ta 798:3745 92. Carlstedt-Duke, J. M. B. 1979. Tissue distribution of the receptor for 2,3,7,8- tetrachlorodibenzo-p-dioxin in the rat. C ancer R es. 39:3172-76 93. Gasiewicz, T. A. Evidence for a homologous nature of Ah receptors among . mammalian species 1984. In M ech a n ism s o f D ioxin A ctio n , ed. A. Po land, R. D. Kimbrough, pp. 161-75. Banbury Report 18, Cold Spring Harbor Laboratory. 500 pp. 94. Kahl, G. F., Friederici, D. E., Bigelow, S. W., Okey, A. B., Nebeit, D. W. Ontogenetic expression of regulatory and structural gene products associated with the Ah locus. Comparison of rat, mouse, PCDDs A N D PCDFs-- SARs 397 rabbit and Sigm oden liispedies. D evel. B io c h em . B iophys. R es. C om m un. 118: P harm acol. Ther. 1:137-52 183-90 95. Okey, A. B., Vella, L. M. 1982. Binding 105. Okey. A. B., Dube, A. W., Vella, L. M. of 3-melhylcholanlhrene and 2,3,7,8- 1984. Binding of benzo[a]pyrene anddi- tetrachiorodibenzo-p-dioxin to a com benz[a,h]anthracene to (he Ah receptor in mon Ah receptor site in mouse and rat mouse and rat hepatic cytosols. Cancer hepatic cytosols. E ar. J . B iochem . R es. 44:1426-32 127:39-47 106. Carlstedt-Duke, J. M. B., Elfstrom, G., 96. Mason, M. E., Okey, A. B. 1982. Hogberg, B., Gustafsson, J. A. 1979. Cytosolic and nuclear binding of2,3,7,8- Ontogeny of the rat hepatic receptor for tetrachlorodibenzo-p-dioxin to the Ah re 2,3,7,8-tetrachlorodibenzo-p-dioxin and ceptor in extrahepalic tissues of rats and its endocrine independence. C ancer R es. mice. E a r. J . B iochem . 123:209-15 39:4653-56 97. Okey, A. B., Mason, M. E., Vella, L. 107. Poellinger, L., Lund, J., Gillner, M., M. 1983. The Ah receptor: species and Hansson, L. A., Gustaffson, J. A. 1984. tissue variation in binding of 2,3,7,8- Physiochemical characterization of spe telrachlorodibenzo-p-dioxin and carcino cific and nonspecific polyaromatic genic aromatic hydrocarbons. In E x hydrocarbon binders in rat and mouse trahepalic D rug M etabolism and Chem i liver cytosol. J . B iol. C hem . 258:13535- c a l C arcinogenesis, ed. J. Rydstrom, M. 42 Bengtsson, pp. 389-93. Amsterdam: 108. Okey, A. B., Vella, L. M., Iverson, F. Elsevier 1984. Ah receptor in primate liver: bind 98. Miller, A. G., Israel, D., Whitlock, J. P. ing of 2,3,7,8-tetrachlorodibenzo-p- 1983. Biochemical and genetic analysis of variant mouse hepatoma ceils de dioxin and carcinogenic polycyclic aromatic hydrocarbons. C an. J. Physiol. fective in the induction of benzo[a]- P harm acol. 62:1292-95 pyrene metabolizing enzyme activity. 109. Nebert, D. W., Eisen, H. J., Hankinson, J . B io l. C hem . 258:3523-27 O. 1984. The Ah receptor: binding 99. Okey, A. B., Bondy, G. P., Mason, M. specificity only for foreign chemicals. E., Nebert, D. W., Forster-Gibson, C. B io ch em . P harm acol. 33:917-24 J., et al. 1980. Temperature-dependent 110. Hudson, L. G.,Shaikh, R.,Toscano, W. cytosol-to-nucleus translocation of the A., Greenlee, W. F. 1983. Induction of Ah receptor for 2,3,7,8-letrachIorodi- 7-ethoxycoumarin 0-deethylase activity benzo-p-dioxin in continuous ceil culture in 'ultured human epithelial cells by lines. J . B io l. C hem . 255:11415-22 z,3,7,8 - tetrachlorodibenzo - p - dioxin 100. Carlstedt-Duke, J. M. B., llamemu, U. (TCDD): evidence for a TCDD receptor. B., Hogbert, B.,Gustafsson, J. A. 1981. ' B io ch em . B io p h ys. R e s. C om m un. IIS: Interaction of hepatic receptor protein for 611-17 2.3.7.8- tetrachlorodibenzo-p-dioxin 111. Jones, P. B. C., Miller, A. G., Israel, D. with DNA. B io ch em . B iophys. A cta L, Galeazzi, D. R., Whitlock, J. P. 672:131-41 1984. Biochemical and genetic analysis 101. Whitlock, J. P., Galeazzi, R. 1984. of variant mouse hepatoma cells which 2.3.7.8- Tetrachlorodibcnzo-p-dioxin re overtranscribe the cytochrome P|-450 ceptors in wild type and variant mouse gene in response to 2,3,7,8-tetrachloro- hepatoma cells; nuclear location and dibenzo-p-dioxin. J . B iol. C hem . 259: strength of nuclear binding. J . B iol. 12357-63 C hem . 259:980-85 112. Israel, D. L, Whitlock, J. P. 1984. Regu 102. Poellinger, L., Kurl, R. N., Lund, J., lation of cytochrome P|-450 gene Gillner, M., Carlstedt-Duke, J. M. B., et transcription by 2,3,7,8-tetrachlorodi- al. 1982. High affinity binding of benzo-p-dioxin in wild type and variant 2.3.7.8- telrachlorodibenzo-p-dioxin in mouse hepatoma cells. J . B iol. Chem . cell nuclei from rat liver. B iochem . B io 254:5400-2 p h ys. A cta 714:516-23 113. Jones, P. B. C., Galeazzi, D. R., Fisher, 103. Gasiewicz, T. A. and Rucci, G. 1984. J. M., Whitlock, J. P. 1985. The control Cytosolic receptor for 2,3,7,8-tetrachlo- of cytochrome P,-450 gene expression by rodibenzo-p-dioxin. M o l. Pharm acol. dioxin. Science 227:1499-1502 26:90-98 114. Ringold, G. M. 1985. Steroid hormone 104. Gasiewicz, T. A., Ness, W. C., Rucci, regulation of gene expression. A nn. Rev. G. 1984. Ontogeny of the cytosolic re P harm acol. Toxicol. 25:529-66 ceptor for 2,3,7,8-tetrachlorodibenzo-p- 115. Safe, S., Safe, L. 1984. Synthesis and dioxin in rat liver, lung and thymus. characterization of twenty-two poly- 398 S A F E chlorinated dibcnzofurans. J . A gric. pig, mouse and hamster. P roc. N atl. F o o d C hem . 32:68-72 A c a d . S ci. U .S .A . 81:7407-11 116. Denomme, M. A., Homonko, K., Fujita, 126. McConnell, E. E., Moore, J. A., Hase- T., Sawyer, T., Safe, S. 1985. The man, J. K., Harris, M. W. 1978. The effects of subslituents on the cytosolic comparative toxicity of chlorinated di- receptor binding avidities and AHH in benzo-p-dioxin in mice and guinea pigs. duction potencies of 7-substituted-2,3- Toxicol. A p p l. P harm acol. 44:335-56 dichlorodibenzo-p-dioxins-QSAR anal 127. Moore, J. A., McConnell, E. E., Dal- ysis. M o l. P harm acol. 27:656-61 gard, D. N., Harris, M. W. 1979. Com 117. Denomme, M. A., Homonko, K., Fujita, parative toxicity of three halogenated di- T., Sawyer, T., Safe, S. 1986. Sub benzofurans in guinea pigs, mice and stituted polychlorinated dibenzofuran re rhesus monkeys. A n n . N . Y. A ca d . Sci. ceptor binding affinities and AHH induc 30:151-63 tion potencies--a QSAR analysis. C hem . 128. Knutson, I. C., Poland, A. 1982. Re B io l. In teract. In press sponse of murine epidermis to 2,3,7,8- 118. Bandiera, S., Sawyer, T., Campbell, M. tetrachlorodibenzo-p-dioxin. Interaction A., Fujita, T., Safe, S. 1983. Competi of Ah and h r loci. C ell 30:225-234 tive binding to the cytosolic 2,3,7,8- 129. Poland, A., Knutson, J. C., Glover, E. TCDD receptor: effects of structure on 1984. Histologic changes produced by the affinities of substituted halogenated 2,3,7,8-tetrachlorodibenzo-p-dioxin in biphenyls--a QSAR approach. Biocliem . the skin of mice carrying mutations that P harm acol. 32:3803-13 affect integument. J . In vest. D erm . 119. Verloop, A. 1983. The STER1MOL 83:454-59 approach: further development of the 130. Poland, A., Paen, D., Glover. 1982. method and new applications. In P esti Tumour promotion by TCDD in skin of cide Chemistry, H um an Welfare and the HRS/J hairless mice. N ature 300:271-73 E nvironm ent, ed. J. Miyamoto, P. C. 131. Kurl, R. N. and Villee, C. A. 1985. A Kearney, pp. 334-44. Oxford: Pergamon metabolite of riboflavin binds to the 120. Knutson, I. C., Poland, A. 1980. 2.3.7.8 - tetrachlorodibenzo - p - dioxin Keralinizalion of mouse teratoma cell (TCDD) receptor. P harm acol. 30:241- line XB produced by 2,3,7,8- 44 tetrachlorodibenzo-p-dioxin: an in vitro 132. Poland, A., Glover, E. 1975. Genetic model of toxicity. C ell 22:27-36 expression of aryl hydrocarbon hydroxy 121. Milstone, L. M., LaVigne, J. F. 1984. lase by 2,3,7,8-tetrachlorodibenzo-p- 2.3.7.8- Telrachlorodibenzo-p-dioxin in dioxin: Evidence for a receptor mutation ducers hyperplasia in confluent cultures in genetically non-responsive mice. M ol. of keraloinocytes. J . Invest. D erm atol. P harm acol. 11:389-98 82:532-34 133. Nebert, D. W., Robinson, J. R., Niwa, 122. Hudson, L. G., Toscano, W. A., Green A., Kumaki, K., Poland, A. P. 1975. lee, W. F. 1985. Regulation of epidermal Genetic expression of aryl hydrocarbon growth factor binding in a human kera- hydroxylase activity in the mouse. J. tinocyte ceil line by 2,3,7,8-tetrachloro- C e ll. P hysiol. 85:393-14 dibenzo-p-dioxin. Toxicol. A p p l. P hrnia- 134. Niwa, A., Kumaki, K., Nebert, D. W. col. 77:251-59 1975. Induction of aryl hydrocarbon hy 123. Osborne, R., Greenlee, W. F. 1985. droxylase in various cell cultures by 2.3.7.8- Telrachlorodibcnzo-p-dioxin 2.3.7.8 - tetrachlorodibenzo - p - dioxin. (TCDD) enhances terminal differentia M o l. P harm acol. 11:399-08 tion of cultured human epidermal cells. 135. Jones, K. G., Sweeney, G. D. 1980. Toxicol. A p p l. P harm acol. 77:434-43 Dependence of the porphyrogenic effect 124. Karenlampi, S. O., Eisen, H. J., Hankin- of 2,3,7,8-tetrachlorodibenzo-p-dioxin son, O., Nebert, D. W. 1983. Effects of upon inheritance of aryl hydrocarbon cytochrome Pj-450 inducers on the cell- hydroxylase responsiveness. Toxicol. surface receptors for epidermal growth A p p l. P harm acol. 53:42-49 factor, phorbol 12,13-dibutyrate or in 136. Nagarkatti, P., Sweeney, G. D., Gaul- sulin of cultured mouse hepatoma cells. die, J., Clark, D. A. 1984. Sensitivity to J . B io l. C hem . 258:10378-83 suppression of cytotoxic T cell genera 125. Madhukar, B. V., Brewster, D. W., tion by 2,3,7,8-letrachlorodibenzo-p- Matsumura, F. 1984. Effects of in vivo dioxin (TCDD) is dependent on the Ah administered 2,3,7,8-letrachlorodiben- genotype of the murine host. Toxicol. zo-p-dioxin on receptor binding of A p p l. P harm acol. 72:169-76 j-21 epidermal growth factor in the hepatic 137. Clark, D. A., Sweeney, G. D., Safe, S., . plasma membrane of rat, guinea Hancock, E., Kilbum, D. G., Gauldie, J. ' J oi . PCDDs AN D PCDFs-- SARs 399 1983. Cellular and genetic basis for suppression of cytotoxic T cell genera tion by haloaromatic hydrocarbons. Im m u n o p h a rm a co l. 6:143--53 138. Vecchi, A., Sironi, M., Canegrati, M. A., Recchia, M.,Garatlini,S. 1983. Im munosuppressive effects of 2,3,7,8tetrachlorodibenzo-p-dioxin in strains of mice with different susceptibility to in duction of aryl hydrocarbon hydroxylase. T o xico l. A p p l. P h a rm a co l. 68:434--41 139. Vecchi, A., Mantovani, A., Sironi, M., Luini, W., Spreafico, F., Garattini, S. 1980. The effect of acute administration of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on humoral antibody production and cell-mediated activities in mice. A rc h . T oxicol. 4:163-65 140. Greig, J. B., Francis, i. E., Kay, S. J. E., Lovell, D. P.. Smith, A. G. 1984. In complete correlation of 2,3,7,8-letrachlorodibenzo-p-dioxin hepatotoxicity with Ah phenotype in mice. ToxicoI.A p p l. P h a rm a co l. 74:17-25 141. Pitot, H. C., Goldsworthy, T., Camp bell, H. A., Poland, A. 1980. Quantita tive evaluation of the promotion of 2,3,7,8-telrachlorodibenzo-p-dioxin of hepatocarcinogenesis from dielhylnitrosaniine. C ancer R es. 40:3616-20 142. Abemethy, D. J., Greenlee, W. F., Huband, J. C., Boreiko, C. J. 1985. 2,3,7,8 - Tetrachlorodibenzo - p - dioxin (TCDD) promotes the transformation of C3H/I0TI/2 cells. C arcin. 6.651-53 143. Kouri, R. E., Rude, T. H., Joglekar, R., Dansette, P. M., Jenna, D. M., et al. 1978. 2,3,7,8-Tetrachlorodlbenzo-p- dioxin as cocarcinogen causing 3melhylcholanthrene-iniliated subcuta neous tumors in mice genetically nonresponsive at Ah locus. C ancer R es. 38: 2777-83 144. DiGiovanni, J., Berry, D. L., Gleason, G. L., Kishore, G. S., Slaga, T. J. 1980. Time-dependent inhibition by 2,3,7,- 8-tetrachlorodibenzo-p-dioxin of skin tumorigenesis with polycyclic hydrocar bons. C ancer R es. 40:1580-87 145. Sawyer, T., Bandiera, S., Safe, S., Hutzinger, O., Olie, K. 1983. Bioanaly sis of polychlorinated dibenzofuran and dibenzo-p-dioxin mixtures in fly ash. C hem osphere 12:529-35 146. Gierthy, J. F., Crane,D., Frenkel, G. D. 1984. Application of an in vitro keratinization assay to extracts of soot from a fire in a polychlorinated biphenyl- containing transformer. F und. A ppl. T oxicol. 4:1036-41 147. Bandiera, S., Farrell, K., Mason, G., Kelley, M., Romkes, M., et al. 1984. Comparative toxicides of the polychlo rinated dibenzofuran (PCDF) and bi phenyl (PCB) mixtures which persist in Yusho victims. C hem osphere 13:507-12 148. Sawyer, T., Safe, S. 1985. In vitro AHH induction by polychlorinated biphenyl and dibenzofuran mixtures: additive effects. C hem osphere 14:79-84 149. Rizzardini, M., Romano, M., Tursi, F., Salmona, M., Vecchi, A., et al. 1983. Toxicological evaluation of urban waste emissions. C hem osphere 12:559-64 pound, it does not vaporize and thus will only be found in the atmosphere when carried on dust particles. Dioxin has a very low solubility in water and, when released in rivers or lakes, eventually settles to the bot tom and accumulates in the sediment. Dioxin binds tightly to soil. This tendency keeps dioxin from migrating through soil, making contamination of ground water highly unlikely.1Soil particles protect the compound from the ultraviolet rays of the sun, which would otherwise cause the dioxin to break down rap idly.2In most cases of environmental dioxin contamina tion, the vast majority of the chemical has remained on or near the soil surface. The half-life, or time required for the breakdown of half the dioxin residue present, has been estimated at one to three years.' However, if the dioxin does manage to settle deep beneath the surface, its half-life may exceed ten years because of the additional protection against the sun's ultraviolet rays. In the sediment of a river or lake, dioxin undergoes a slow yet significant degree of microbiological degrada tion. The half-life in this environment has also been estimated to be one to three years.1 If d io x in is generated in tiny a m o u n ts, w h y is there so m uch concern about a health threat? Animal experiments have shown dioxin to be extremely toxic to some animals, even in small amounts. The compound has caused a number of adverse effects in laboratory animals including severe blood chemistry changes, liver damage, skin disorders, lung lesions, loss of weight and death. Animal species vary greatly in their susceptibility to dioxin poisoning. Guinea pigs are the most sensitive of the species tested. Indeed, dioxin's reputation as being extremely toxic is based primarily on tests done on these highly susceptible animals. Hamsters, in contrast, are approximately 5,000 times more resistant than guinea pigs to a single dose of dioxin. The variability of dioxin sensitivity is presented in Table 1, which lists the dose necessary to kill half of the test animals (called LDMvalue) fora number of species. Dioxin is also an animal carcinogen, causing many different tumor types at a number of sites including the liver and thyroid. The specific roles which dioxin plays in cancer development, however, are still being explored. Depending on the test species, dose and other experimental conditions, dioxin may act as a cocarcinogen with another cancer-causing agent or even, surprisingly enough, an anti-carcinogen, meaning that the presence of dioxin inhibits some carcinogens from 4 TABLE 1 D IO X IN LDVALUES Animal LD,, (/xg/kg body weight) Guinea Pig 1 Rat-male 22 Rat-female 45 Monkey <70 Mouse 114 Rabbit 115 Dog >300 Bullfrog >500 Hamster 5,000 Source: Poland and Knutson, A n n u a l Review o fP h a r- m aco io gy & Toxicology, 1982. inducing cancer in laboratory animals.3Evidence that it is an actual in itia to r of cancer remains weak. Repro duction studies (three generations) with dioxin have shown that it is not mutagenic in mammals.* Maternal exposure to relatively high levels of dioxin during critical periods of fetal development causes birth defects and fetal death in many animal species. While the defects observed vary from species to species, the high frequency of adverse effects leaves no doubt that dioxin can cause serious reproductive problems, at least at doses above some minimum level, in the female test animal. There is no scientific evidence that male sperm cells are affected by dioxin. In w h at kinds of situations are people exposed to dioxin? Human exposure to dioxin can be separated into four distinct groups: 1) Accidents at chemical manufacturing plants have exposed workers to large single doses (and in the case of Seveso, Italy, local residents as well); 2) Occupational exposure to herbicides has exposed applicators over long periods, including U.S. Air Force personnel who were involved in Agent Orange spray ing, and others who work with dioxin-containing products; 3) Contamination of the general environment may expose local residents to low doses over potentially lengthy periods, primarily from improper disposal of contaminated industrial waste; 4) It is believed that very, very low levels of dioxin con- 5 .J.1 J SL> 3t Exhaust systems of vehicles driven by diesel engines have been shown to contain trace amounts of dioxin, suggesting that dioxin is formed during the combustion of diesel fuel. tamination are present everywhere, at least in the developed countries, due to the use of herbicides and solvents and the burning of wastes. Dioxin has been detected in pristine lakes, in freshwater fish and in nor mal human tissues. W h a t are the k n o w n health effects in hum ans of dioxin exposure? Only two health effects have so far been repeatedly observed following dioxin exposure; and both are reversible. The first isa skin disorder called chloracne and the second isa general "not feeling well" syndrome including sleeplessness, headache, nausea and irritabil ity. Other effects have been reported in single studies at borderline levels of significance and have thus far failed to be confirmed in any second study. These included: soft-tissue sarcoma (cancer) in Swedish herbicide appli cators, increased incidence of stomach ulcers, some birth defects among the offspring of Vietnam veterans and some small changes in certain metabolic measure ments. Conclusions are difficult to draw from the single study effects because any random sample from an unaf fected population is likely to display s o m e statistically improbable results if many different properties are mea sured and tested for significance. Despite the frightening image which the word "dioxin" has come to represent, there are no known human 6 deaths from exposure to this substance. Some contami nation episodes are described below: Monsanto Workers An accident at Monsanto Chemical's plant in Nitro, West Virginia on March 8, 1949 exposed more than 200 workers to dioxin. Fumes containing dioxin were emit ted when a relief valve opened as a result of an explo sion. One hundred twenty-two workers developed chloracne.within a few weeks. Other immediate but transient effects were also reported, including head ache, dizziness, nausea and general weakness. It is impossible to know what symptoms were specifically caused by dioxin, as many of the observed symptoms can also result from exposure to the other chemicals contained in the fumes. A 1980 epidemiological study' of 121 of the 122 chloracne victims, 30 years after the incident, permits some judgment about the long-term effects of dioxin. Fewer of these workers had died from cancer than would be expected in a control (unexposed) population. Only 32 of the workers had died from all causes, where 46 deaths would be expected in a comparable sample of that age and size. The authors noted, "It is important that no apparent excess in total mortality or in deaths from malignant neoplasms or diseases of the circulatory system were observed in a group of workers with a high peak exposure to tetrachlorodibenzodioxin who were followed over a period of nearly 30 years." Unfortu nately, we do not have data on the dose of dioxin to which the workers were exposed. But we do know that it was sufficient to cause chloracne and other immedi- Guinea pig. Dioxin's reputation as being extremely toxic is based pri marily on tests done on guinea pigs. Hamsters, in contrast, are approximately 5,000 times more resistant than guinea pigs to a single dose of dioxin. 7 ACSH: J. Jones Following the explosion of a chem ical reaction cham ber at an IC M E S A chemical plant in Seveso, Italy, a section of the com m unity was dusted with an estimated one to four pounds of dioxin. There were no deaths am ong humans, but plants, birds, rabbits and chick ens in the area died soon after the accident. ate symptoms. Although the total number of people exposed was not large, the results are reassuring for other exposed populations, especially those whose exposure was less than the amount needed to cause chloracne, since no long-term effects have yet been documented among the exposed Monsanto workers who did have chloracne. Dow Workers In 1964, 61 Dow Chemical employees were exposed to dioxin levels potentially reaching 10,000 parts per mil lion (ppm), or ten micrograms dioxin per kilogram body weight. After a plant mishap in Michigan, 49 of the exposed employees developed chloracne. As of 1980 (15 years later), no excess in total deaths had occurred among these workers.4 Seveso, Italy One of the most publicized chemical accidents in recent years was the July 1976 accident in Seveso, Italy. Following the explosion of a chemical reaction cham ber at the ICMESA chemical plant, a section of the Seveso community was dusted with an estimated one to four pounds of dioxin. In this case, the dioxin was produced during the manufacture of trichlorophenol, a material which was being used to make hexachlorophene. Some 37,000 people were potentially exposed to the dioxin and several hundred did develop symptoms of acute poisoning. Children and adults complained of nausea, nervous symptoms and chloracne, which was 8 severe in some cases. The relative contribution of other chemicals released in the explosion towards producing these symptoms is unknown. There were no deaths among humans caused by acute poisoning. Plants, birds, rabbits and chickens in the area died soon after the accident. The animal mortality was propor tional to the dioxin levels in the soil and grass, which were as high as 5.5 ppm in the most heavily contami nated, 180- acre zone.4This finding illustrates a point concerning the relative hazard posed by dioxin in soil. Animals are, for the most part, at a much greater risk, as they are likely to ingest significant and repeated doses of the compound during their normal feeding activities, unlike humans who are unlikely to ingest dirt or plants tainted with dioxin residue. In the days following the accident, a number of public health measures were instituted by the Italian authori ties. Entry into the area of highest contamination was prohibited and 736 inhabitants were evacuated two to three weeks after the accident. Young children, preg nant women and the elderly who resided in the sur rounding region that had less, but still significant, con tamination were also evacuated. Those remaining were forbidden from consuming fruits, vegetables, meat and dairy products from the area, and domestic animals and livestock were destroyed. An extensive health surveillance system was put into effect to record any continuing and long-term effects of dioxin exposure on the population. Medical examina tions and laboratory tests were performed, pregnant women were closely monitored to record rates of mis carriage and birth defects, and a cancer registry was created to track new cases of cancer. Chromosomal studies of humans at Seveso who were exposed to high concentrations of dioxin show no evi dence of chromosome abnormalities due to dioxin exposure.7 A thorough analysis of the health data from the Seveso population shows that affected residents developed chloracne and minor, reversible nerve damage. Clinical studies revealed some impaired liver function, the long term effects of which are as yet unknown. No other organs or body functions appear to have been affected, nor was reproduction adversely affected. No cases of cancer related to the incident have been observed but study will continue for many years so that long-term effects can be monitored. D o e s dioxin cause can ce r in h u m a n s? Experiments on laboratory animals have shown that dioxin can cause cancer at very low doses, and this of 9 course suggests that dioxin could cause cancer in humans. On the other hand, studies of people who have been exposed to dioxin have not shown a rela tionship between dioxin and cancer and this offers us considerable assurance that traces in the environment do not constitute a cancer hazard. However, the actual number of exposed persons is too small to rule out the possibility of such a hazard. A chromosomal study done at a plant where dioxin-contaminated 2,4,5-T was man ufactured showed no chromosomal abnormalities.4 Much of the research on dioxin and human carcinogen esis has centered around the compound's alleged ability to cause a rare form of cancer called soft tissue sar coma (STS). STS is a group of tumors that occur in the muscles, tendons and other connective tissues, fat tis sues, blood vessels and nerves. Researchers in Sweden concluded that development of STS was more likely to occur among workers occupa tionally exposed to dioxin-contaminated phenoxy acids in agricultural chemicals. Between 1970 and 1976 there were 19 cases of STS among such workers compared with 11 expected cases. Another Swedish study found that railroad workers who handled dioxin-contami nated herbicides suffered an elevated rate of cancer mortality, with six cases observed compared with three cases expected in the sample of 207 men.' Since no specific tumor type predominated among the six cases,' it is highly questionable whether dioxin was, in fact, related to the cancer mortality. The dioxin/STS hypothesis was tested by researchers in Finland and New Zealand. The results there differed from those in Sweden. In neither case was any associa tion seen between occupational dioxin exposure and STS, despite the fact that these workers used the same herbicides during the same time period as did the work ers in Sweden. Dioxin content in commercial 2,4,5-T used in Finland ranged from 0.1 to 0.9 ppm during the 1960s.' The daily dose of dioxin absorbed by an appli cator has been calculated to be some 500 times less than the highest level failing to produce a carcinogenic effect (called the no-effect level) in humans based on extrapolation from experimental rodent data.' Soft tissue sarcomas are relatively rare, thus few pathologists have been able to achieve a high degree of consistency in their diagnosis. Most experts believe that if dioxin was acting as a carcinogen it would cause an increase of one type of soft tissue sarcoma. In the Swedish stud ies there was no documented evidence either to inten sity, frequency or time of duration of exposure. The results can be confused by exposure to other chemi cals, such as amitrole; which-iaAknovyn carcinogen. > In short, while the Swedish conclusions suggest a link 10 W hen dioxin contamination occurs, pets and livestock are more likely to be in danger than hum ans are, because the animals may ingest tainted soil and plants during their normal feeding activities, while hum ans are unlikely to d o so. between dioxin and STS, there has been no confirma tory evidence to support it, despite an effort by researchers to specifically monitor STS rates in compa rable sample populations. Studies on workers exposed to dioxin during occupa tional accidents (between 1949 and 1968) failed to establish a dioxin/cancer association. Monsanto, BASF, Boehringer, Philips, Dow, Coalite and Spolana workers all suffered some degree of acute dioxin toxicity symp toms. But of the seven groups, only the BASF group showed any excess cancer mortality. There, the three cases of stomach cancer seen in one age bracket meant that the rate for that cancer was nine times higher than normal. The Monsanto and Dow studies did reveal a few cases of STS, although the cancer rate for all tumor types was not elevated. While the doses to these work ers are unknown, the dioxin levels in waste residues were very high, ranging from 140 to 2,400 ppm.' Does dioxin cause birth defects? Laboratory studies show that dioxin causes birth defects when female test animals are exposed to the com pound. However, as with cancer, no solid evidence exists linking dioxin with birth defects in humans. Claims have been made to the contrary, but they have been based largely on medically unvalidated and statis tically unrepresentative anecdotal reports lacking the sound data base necessary to establish a cause-andeffect relationship. 11 1947 9 E. O. (ones Typical of these are the Alsea studies released by the Environmental Protection Agency (EPA) in 1979. EPA scientists concluded that there was a relationship between the use of a dioxin-containing herbicide, 2,4,5-T, and an increase in miscarriages among women living in a sprayed region in Alsea, Oregon. The Alsea II report was understood by many in the lay community as proving that dioxin in 2,4,5-T caused reproductive abnormalities. However, after the report was made public, many scientists challenged the EPA's conclusions. More than 18 reviews of this study have uncovered numerous flaws in the study design, statisti cal analysis, and interpretation of the results. Criticism of the study was extensive, consistent and practically unanimous. For example, EPAconcluded that a sea sonal peak in miscarriages followed 2,4,5-T spraying. Close examination of the data by other scientists showed that this peak did not exist. A number of epidemiological studies have been done to determine what, if any, relationship exists between paternal exposure to dioxin and birth defects. Children fathered by Australian veterans of the Vietnam War who served during the period when dioxin-contami nated defoliants such as Agent Orange were sprayed have not experienced elevated numbers of birth defects. Mothers of these children have not had unusu ally high numbers of miscarriages. Reproductive out come of the offspring of male applicators of 2,4,5-T with an average dioxin content of 0.1 ppm in New Zealand has been normal.1Similarly, all children born to men affected by the Spolana accident in Czechoslovakia were free of birth defects; miscarriage incidence among their wives was below expected levels.1 Two major epidemiological studies were completed in 1984 on Vietnam veterans exposed to dioxin in the period 1963 to 1971 .'-'"The first, the so-called "Ranch Hand Study," because of the code name for the defolia tion operations, was conducted by the Air Force and compared the 15 to 20 year health histories of 1,045 herbicide applicators with 19,000 other veterans believed to have had no exposure. The study stressed investigation of chloracne and any incidence of soft tissue sarcoma. In all cases, no adverse health effects were found. Some borderline incidence of increased infant deaths and minor birth defects was observed and noted for follow-up in future studies. The investigators concluded, "current evidence is insufficient to support a causal relationship between herbicide exposure and adverse health." The second study was conducted by the federal Centers for Disease Control (CDC) on the family health histories of 7,000 children born with birth defects in the Atlanta 12 Dioxin has a very low solubility in water, and when released in rivers or lakes, eventually settles to the bottom and accumulates in the sedi ment. area. The parents with Vietnam experience (about 10 percent of the total sample) composed the contrast group for comparison with the others as the control group. Reassuringly, no significant correlation of birth defects or changes in fertility was found to be associ ated with the Vietnam experience. Weak indications were observed at the level of statistical significance for two specific birth defects: spina bifida (a spinal defor mation) and congenital tumors. Again, these results were deemed inconclusive, and will be watched in the future. Dow Chemical has a large herbicide manufacturing plant in Midland, Michigan. The Michigan Department of Public Health looked at the number of birth defects among Midland County residents and found no excess of any kind. Researchers in Hungary studied the relationship between 2,4,5-T use and birth defects from 1966 to 1977, a period which coincided with a 30-fold increase in use of the herbicide in that country.1No increase in the national frequency of birth defects was detected. Reproductive patterns were monitored closely in Seveso in the years following the 1976 dioxin exposure. Had the dioxin contamination induced miscarriage, rates would have been highest during the remainder of 1976, the period when women already pregnant in July (the time of the accident) would still be carrying their unborn offspring. Yet the frequency of miscarriage did not deviate from expected levels during these six months. A small decrease in fertility (defined as the 13 19460 ACSH: ). Jones number of new pregnancies) was observed during the months following the accident, possibly suggesting that there may have been an adverse effect from the dioxin or other chemicals released into the environment. More likely though, this reflected a conscious decision by many parents not to have children for fear of dioxin exposure. In summary, the epidemiological evidence collected to date does not implicate dioxin as a cause of birth defects or miscarriages in humans. A m ong exposed populations which have been studied, dioxin exposures were not found to adversely affect reproduction. These findings should be a source of comfort to residents of places like Times Beach and Newark, where exposure levels have been substantially lower than in the various industrial accidents. ; > W hat happened at Times Beach, Missouri? The dioxin contamination of Times Beach, Missouri was perhaps the most widely publicized episode of chemi cal contamination in the U.S. since the Love Canal story of 1978. Times Beach, a small town of approximately 2,000 residents located 25 miles southwest of St. Louis, drew national attention in December, 1982 after flood waters swept through the area. Concern was raised that contaminated soil which had been present along the roadside had been dispersed widely throughout the town. The incident resulted in a federal buyout and relocation of the Times Beach residents. How did the dioxin get there? In 1971, a private waste hauler sprayed dioxin-contami nated waste oil on unpaved roads as a dust retardant. In addition to the roads, two horse arenas were sprayed with the oil. A number of horses and small animals died and two children playing in one of the arenas devel oped chloracne." Investigations were conducted by state officials and the federal government's Centers for Disease Control (CDC). By 1974, C D C scientists concluded that dioxin poisoning was the cause of the horse and other animal deaths. M onitoring of soil samples revealed dioxin con centrations of 31 to 33 ppm in the horse arenas. Dioxin levels in Times Beach soil taken after the 1982 flood ranged from non-detectable levels to 0.3 ppm .'2The highest level measured in town then was 100 times less than that found in the horse arenas some ten years earlier and approximately 15 times less than the highest level measured in the soil at Seveso, Italy after the 1976 accident. During the autumn of 1982, EPA tested residues at the chemical plant which was the source of the dioxin 14 Surveys of Great Lakes fish show that the average levels of dioxin residues in various species range from 40 ppt in Lake O ntario eel and smelt to less than 10 ppt in most species in Lake Erie. Com parisons show that consumers of fish with 25 ppt of dioxin residue w ould be exposed to less than 1/70th the no-effect level and less than 1/7000th of the carcinogenic level in test anim als. sprayed on roads in 1971. Results showed dioxin present in concentrations of up to 0.3 ppm. After the December floods, officials feared that the dioxin along the roads might have dispersed, increasing the chances that the population would be exposed. The news hit the front pages across the country. M an y Times Beach residents began to fear for their health. The EPA had never dealt with a comparable dioxin con tamination situation. There was no tolerance level for dioxin to use as a criterion upon which to base risk assessment and strategy. The C D C was called in and after extensive epidemiological and statistical model ing, a limit or action level of one ppb (one part per billion) dioxin in soil was announced. C D C calculated, on the basis of hypothetical, yet currently accepted, models of risk estimation, that a 70-year exposure to soil concentrations of one ppb dioxin might cause one excess case of cancer per million people. A major con sideration was the possibility that children would be most heavily exposed by ingestion of dirt while play ing." N ew evidence suggests that laboratory animals exhibit symptoms of dioxin poisoning after Ingestion of dioxin-contaminated soil.14With Times Beach soil con taminated at levels reaching 0.3 ppm in some samples, C D C recommended evacuation. Times Beach was placed on top of the national priority list of hazardous waste sites. A state of emergency was declared and 15 19481 federal funds in excess of $30 million and a $3 million state contribution were allocated to buy the houses and resettle the residents. D o the residents of Times Beach exhibit any symptoms of dioxin poisoning? A pilot epidemiological study of Times Beach residents and of controls conducted b yth e C D C , Missouri Divi sion of Health, St. Louis University School of Medicine and St. Joseph's Hospital in Kirkwood, Missouri was completed in October 1983. The project consisted of three phases-- a questionnaire survey to obtain informa tion on medical histories and risk of exposure to dioxin, a clinical screening for chloracne, and extensive exami nations com paring persons with high potential of dioxin exposure to persons with little or no risk of dioxin exposure. The results of the study were encouraging. In a letter to all participants, the Missouri Division of Health con cluded that "w e have not been able to identify mean ingful ill-health effects related to potential exposure to d io x in .... The results show that the high risk group of participants had the same rate of health problems as the low risk group ... every participant had a skin examina tion, but no case of chloracne was fo u n d .. . . To sum marize, this pilot investigation did not turn up evidence suggesting that dioxin has caused any serious health problems in the 104 persons who were studied."15 Extensive clinical studies of the Times Beach residents reported in 198616were similar to the earlier work in that no conclusive evidence of dam aging health effects was found. The study compared 154 exposed persons with an unexposed control group of 155 people having similar lifestyles and medical histories. In an extensive battery of over fifty tests, the differences found between the two groups were some subclinical changes in liver function and some depression of the cellular immune function in the exposed group. In neither case, after whatever exposure took place between 1971 and 1983, had there been any evidence of increased liver disease or general illness in the exposed group as of 1986. Researchers will follow up on the effects noted in future years. Has the area been cleaned up? Not yet. Authorities in Missouri have yet to determine how best to dispose of the wastes. An interim disposal site at some Missouri location will be selected while the search for a permanent facility continues. 16 Is the average American being exposed to dioxin today and, if so, at what levels? Large scale surveys to answer these questions have never been done but are in the planning stage by the EPA at the present time. The C D C recently announced plans for a $57 million study of dioxins, "its most com plex, ever." With the advent of instruments capable of detecting extremely low concentrations (parts per tril lion) of specific chemicals, it comes as no surprise that evidence is accumulating that dioxins, like many other contaminants, are present in extremely low quantities nearly everywhere in the developed world. When compared with large-scale and indisputable public health haz ards such as cigarette smoking, drug abuse and drunk driving, the damage to society and individuals caused by dioxin exposure is neg lig ib le . In spite of the widespread, low level contamination, based on the facts which have been accumulated to date it is highly unlikely that the health of Americans is being affected adversely by dioxin. The symptom-caus ing exposures of the past have exclusively consisted of herbicide-related accidents and occupational expo sures. The EPA severely restricted use of 2,4,5-T in 1979 (inappropriately, in the opinion of many scientists) and is m oving towards complete elimination of the use of the herbicide in this country (presently, it is permitted 17 ACSH: J. (ones 19482 on rice paddies and rangelands). The 2,4,5-T that is currently used has an average dioxin concentration of only 0.01 ppm, a one hundred-fold reduction com pared to the 1 ppm level often present in the 2,4,5-T manufactured twenty years ago. Indeed, some 2,4,5-T made 20 years ago contained 25 ppm or more.'7 The Food and Drug Administration (FDA) has set guide lines for dioxin concentration in edible fish at 25 parts per trillion (ppt). This regulatory decision was arrived at by extrapolating data from rodent laboratory experi ments. A simplified explanation of this complex proce dure follows. Surveys of Great Lakes fish determined that the average levels of dioxin residues in various species offish ranged from 40 ppt in Lake Ontario eel and smelt to less than 10 ppt in most species in Lake Erie.18By com paring the amount of dioxin ingested by fish eaters with the no-effect level of dioxin in rodents, it was shown that consumers of fish with 25 ppt of dioxin residue would be exposed to less than 1/70th the no-effect level and less than 1/7000th of the carcinogenic level in test ani mals.18These guidelines are believed to provide a more than adequate margin of safety. H ow effective are epidemiological studies and how much of this type of investigation is enough? It is frustrating to the concerned public to find so many large epidemiological studies ending with "inconclu sive results" on the main questions and then reporting weak statistical signals suggesting that some new and previously unsuspected health effect may be present. Unfortunately, it is in the nature of these studies that the two dilemmas exist. They must be expected from future epidemiological studies as well. Suppose that in fact there is no health effect from a particular exposure. The epidemiological study, or indeed, even a perfectly controlled experiment, will never be able to "prove" that fact, for to do so would be to prove a negative proposition. The best that scien tists can hope to do is investigate more and more cases to strengthen their conviction that "n o evidence can be found" that the effect in question exists. At some point we must decide when the investigation has gone far enough. The C D C study of Vietnamrelated birth defects in Atlanta cost $2.8 million; the Air Force Ranch Hand Study cost $11 million; the future C D C study was planned to cost $57 million. O n the other hand, as was pointed out earlier, it is likely that somef/i/ngunusual will show up in every one of these studies through the workings of chance alone. It is 18 like a bridge hand which, while it is unlikely to hold any specific combination of cards, is quite likely to contain some unusual combination. Again, scientists must make a judgment about the likely significance of the newly discovered effect and respon sible lay people must decide whether to allocate more scarce resources to further investigation. Epidemiological studies of dioxin have certainly reached a high degree of sophistication and are receiv ing the benefit of continuing public concern. For exam ple, the Veterans Administration periodically reviews ail literature on dioxin with a team of independent medical specialists as mandated by Public Law 96-151, passed in 1979. The 1985 review covered 250 documents pub lished in 1984, two-thirds of which were primary sources.19Five major epidemiological studies were reviewed for all of the suspected effects of dioxin expo sure. N o significant new results were found and the overall conclusion was that present information "w as not sufficiently advanced for definitive conclusions on human health effects of low-level exposure." The great value of epidemiological studies is that where a strong cause-and-effect relationship exists, it will almost certainly show up conclusively in a large sam ple. This will be true even when the cause (high dioxin exposure) and thus the effect (chloracne) are quite rare in the general population. O n ly such strong relation ships can be conclusively established. It can be very useful, however, to know through epidemiological studies that a strong relationship probably does not exist if the studies keep com ing in with inconclusive results. W e may never know whether or not weak effects exist. It might be argued that at som e point we have enough information to be able to say, "The effect, if any, from this cause is lost in the background of the effect from all other causes." Then we would set prudent limits on future contamination by the target substance and turn our scientific attention, which is a limited resource, to other concerns. In the judgment of many scientists, this time has come for dioxin. Dioxin Contamination: A Perspective The lack of any lingering effects am ong the heavily exposed citizens (including the young and elderly) of Seveso, Italy is comforting for every civilian population exposed to environmental dioxin, since the doses at Seveso were higher than in any other civilian exposure situation, except the Missouri horse arenas. Knowledge about any possible long-term effects based on the Seveso experience is limited because a relatively short time (only ten years) has passed since the accident. 19483 However, the M onsanto workers who were exposed to enough dioxin to cause chloracne and other acute symptoms in 1949 have been studied for more than thirty years. Epidemiological studies of this group have yet to demonstrate any long-term effect of dioxin on the people who were exposed. At the Scientific Dispute Resolution Conference in Arlington, Virginia" it was concluded that "phenoxy herbicides containing T C D D have not been shown to be carcinogenic in humans in retrospective epidemio logic studies to d ate .. . . Analysis of the available data leads this group to the conclusion that no adverse effects on human reproduction have yet been dem on strated after exposure to 2,4,5-T or T C D D ." These data came from the United States, Sweden, N ew Zealand, Australia, Vietnam and Italy. Possibilities that dioxin may yet cause long-term health problems do remain. The low exposure over many years in a Newark neighborhood adjacent to a chemi cal manufacturing plant may affect health differently than the large exposure at the M onsanto plant. C on clu sions drawn from study of workers may not be able to provide information about the general population, which includes people of all age groups with the com plete scope of human illnesses, in addition to healthy working men. Nevertheless, thus far we have no evi dence suggesting that either short-term or long-term health effects will be seen in populations exposed to low levels of environmental dioxin such as in Times Beach and Newark. For further reading, the 1985 book, Dioxins in the Environment, by Michael A. Kamrin, published by Harper & Row, is recommended. In perspective, it should be recalled that the evidence suggesting dioxin causes cancer or birth defects is based on animal experiments. Genuine conflict exists in the scientific community about the validity of direct extrapolation of laboratory data to estimates of human risk. Beyond that, in interpreting the laboratory data on dioxin, one should be aware of the fact that a number of naturally occurring chemicals have also been shown to be potent animal carcinogens and teratogens. For example, aflatoxin (present in peanut butter and grainbased foods) is a potent animal carcinogen. However, there is no evidence that the tolerated trace levels in peanut butter and other foods pose any hazard to human health. Further, while there are some 600 to 800 agents known to be teratogenic in laboratory animals, only about 25 or 30 are known to be responsible for human malformations. It is interesting that the first com pounds which showed high teratogenic activity were such essential materials as the vitamins A, D, and E, when administered in excessive quantities. 20 Public health policy must incorporate both scientific and nonscientific considerations. Medical, ecologic, economic, and legislative input, combined with com mon sense, is necessary for the framing of future public policy on hazardous waste issues such as dioxin. N o one likes to learn that chemicals are polluting the environment. As a health-conscious society, we strive to minimize health risks whenever possible. That potential health damage resulting from dioxin exposure has received so much media attention and caused enor mous public concern is indeed ironic. W hen compared with large-scale and indisputable public health hazards such as cigarette smoking, drug abuse and drunk driv ing, the damage to society and individuals caused by dioxin exposure is negligible. Based on laboratory ani mal studies, however, some suspicion of adverse health effects persists. Fortunately, that potential has not been realized in humans, nor is there good reason to expect it to do so in the years ahead. References 1. Reggiani, G., Toxicology of 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD): short review of its formation, occurrence, toxicology, and kinetics, discussing human health effects, safety measures, and disposal. R e g u la to r y T o x ic o lo g y a n d P h a r m a c o lo g y 1:212, 1981. 2. Crosby, G.M., and A.S. W ong, Environmental degradation of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). S c ie n c e 195:133 7, 1977. 3. Cohen, G.M., W.M. Bracken, R.R Iyer, D.L. Berry, J.K. Selkirk, and T.J. Slaga, Anticarcinogenic effects of 2,3,7,8tetrachlorodibenzo-p-dioxin on benzo(a)pyrene and 7,12dim ethylbenz(a)anthracene tum or initiation and its relation ship to DNA binding. C a n c e r R e s e a r c h 39:4027, 1979. 4. Murray, F.J., et al., Three-generation reproduction study of rats given 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the diet. T o x ic o lo g y a n d A p p lie d P h a r m a c o lo g y 50:241, 1979. 5. Zack, J.A., and R.R. Suskind, The mortality experience of workers exposed to tetrachlorodibenzodioxin in a trichlorophenol process accident. Jo u rn a l o f O c c u p a tio n a l M e d ic in e 22:11, 1980. 6. The Dow Chemical Company, D io x in , A g e n t O r a n g e a n d H u m a n H e a lth , April 1984. pp. 13-15. 7. Tuchamann-Dupleisis, H., Pollution of the environm ent and offspring apropos of the accident at Seveso. M e d e c in e e t H y g ie n c e 36:1758, 1978. 8. Kilian, D.J., et al.. Cytogenetic studies of personnel who m anufacture 2,4,5-T. Presented at the New York Academy of Sciences Workshop on Occupational Monitoring and Genetic Hazards, March 28-29, 1975. 9. Lathrop, G. D., et al. A n E p id e m io lo g ic I n v e s tig a tio n o f H ea lth E ffects in A ir F orce P e rso n n e l F o llo w in g E xp o su re to H e r b ic id e s : B a s e lin e M o r b i d i t y S t u d y R e s u lts , USAF School of A erospace Medicine, Brooks Air Force Base, TX, 1984 (19 Chapters). 21 11. Gj A Ro As 10. Erickson, J. D., et al. V ie tn a m V e te r a n s ' R is k s fo r F a th e r in g B a b ie s w ith B irth D e fe c ts , Centers for Disease Control, Atlanta, GA, 1984 (370 Pages). 11. Carter, C.D., R.D. Kimbrough, J.A. Liddle, R.E. Cline, M.M. Zack, and W.F. Barthel, Tetrachlorodibenzodioxin: an accidental poisoning episode in horse arenas. S c i e n c e 188:738, 1975. 12. Interim Report of the Missouri Dioxin Task Force, June 1, 1983. 13. Centers for Disease Control, Health-risk estim ates for 2,3,7,8-tetrachlorodibenzodioxin in soil. M o r b id ity a n d M o r ta lity W e e k l y R e p o r t 33(3):27, 1984. 14. McConnell, E.E., G.W. Lucier, R.C. Rumbaugh, P.W. Albro, D.J. Harvan, J.R. Hass, and M.W. Harris, Dioxin in soil: bioavailability after ingestion by rats and guinea pigs. S c ie n c e 223:1077, 1984. 15. Donnell, H.D., Jr. (Co-director, Missouri Dioxin Study), letter to participants, O ctober 16,1983. 16. Hoffman, R. E., et al. Health effects of long-term expo sure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. J o u rn a l o f th e A m e r ic a n M e d ic a l A s s o c ia tio n 255:2081, 1986. 17. Poland, A.R, D. Smith, G. Metter, and R Possick, A health survey of w orkers in a 2,4-D and 2,4,5-T plant. A r c h iv e s o f E n v ir o n m e n ta l H e a lth 22:316, 1971. 18. Cordle, F., The use of epidem iology in the regulation of dioxins in the food supply. R e g u la to r y T o x ic o lo g y a n d P h a r m a c o lo g y 1:385, 1981. 19. Veterans Administration, R e v i e w o f L ite r a tu r e o n H e r b i cides, Including P h en o xy H erbicides a n d A sso cia ted D ioxins, Volume V, Clem ent Associates, Arlington, VA, May, 1985. 20. Scientific Dispute Resolution C onference on 2,4,5-T sponsored by the American Farm Bureau Federation (225 TouhyAve., Park Ridge, IL 60068). ORDER FORM No. of CopiM Title Price ____ Cancer in the U.S.: Is There an Epidemic? .. _____ ____ Food Additives and Hyperactivity............. ....... ____ The Health Effects of Caffeine................. ....... ____ Wood As Home Fuel: A Source of Air Pollution........................ ....... ____ Fast Food and the American D i e t .............' _____ ____ "Vitamin B-15":Anatomy of a Health Fraud . ____ ____ Irradiated Foods .................................. ....... ____ Formaldehyde..................................... ....... ____ America's Health: A Century of Progress But a Time of Despair............. ........ ____ Breast or Bottle?.................................. ....... ____ Health and Safety Aspects of Video Display Terminals ........................ ....... ____ Antibiotics in Animal Feed: A Threat to Human Health?..................... ....... ____ Smoking or Health: It's Your C h o ic e .......... ........ ____ Of Mice and Men: The Benefits and Limitations of Animal Cancer Tests . . . . ____ ____ Pesticides in Your Home and G a rd e n ........ ....... ____ Ethylene Dibromide (EDB) ..................... ....... ____ Searching for a Way Out-- Smoking Cessation Techniques .............. ....... ____ Low-Calorie Sweeteners Aspartame-Saccharin-Cyclamate............. ....... ____ Dioxin in the Environment: Its Effect on Human H e a lth ..................... ....... ____ Reye Syndrome................................... ....... ____ PCBs: Is the Cure Worth the C o s t ? ........... ....... ____ Diet and Cancer................................... ....... ____ Obstetric Anesthesia........................... . ....... ____ Premenstrual Syndrom e........................ ....... ____ Does Nature Know Best? Natural Carcinogens in American Food ___ ____ ____ Automobile Occupant Restraint Systems ... ____ ____ Hay Fever .......................................... ....... ____ Answers About A I D S ............................. ....... ____ Osteoporosis...................................... ....... ____ Microwave Ovens ................................ ....... ____ Malignant Melanoma............................. ....... ____ Sugars and Your Health.......................... ....... Total ____ U se this form for ordering and we will bill you. Prices: 1 - 9 copies $2.00 per copy 10- 49 copies $1.75 per copy 50- 99 copies $1.50 per copy 100-199 copies $1.25 per copy NAME___ ADDRESS 22 23 I L D< American Council on Science and Health Yes, I will join ACSH! Annual Membership Rates -- Please check category 8ENEFACT0R $25,000 PATRON $15,000 0 SPONSOR $10,000 SUPPO RTING MEMBER $5,000 INSTITUTION $1,000 Q INDIVIDUAL $35 (Outside U.S. $50*) SEN IO R CITIZEN STUDENT (Outside U.S. $25*) LIBRARY (Outside U.S. $25*) A C S H News & Views Subscription only (Outside U.S. $16') $15 $15 $15 $10 'A ll overseas subscriptions must be prepaid, U.S. currency. I wish to contribute $________________________ in addition to membership dues. AH contributions are tax-deductible as provided by law. Checks should be made payable to American Council on Science and Health and mailed to 47 Maple Street, Summit, N.J. 07901-2580 PLEASE PRINT Name______________________________________ Address____________________________________ City________________________________________ State__________________ Zip__________________ Telephone__________________________________ Affiliation___________________________________ A C SH members receive: A C S H News & Views, a bimonthly publication with scientific articles on health and safety; book reviews; guest editorials; and updates on current issues being researched by ACSH. Copies of latest A C S H reports, summaries, annual reports, and other materials. Information on national and local conferences and seminars. T3 3* a 3 'S c IS 3 os Q. 3oj oo Send membership information to: Name________________________ Address______________________ City_________________________ State__________________ Zip___ Telephone____________________ Affiliation_____________________ 24 American Council on Science and Health 1995 8roadway New York, NY 10023 Telephone: 212 362-7044 47 Maple Street Summit, NJ 07901 Telephone: 201 277-0024 Elizabeth M. Whelan, Sc.D., M.P.H. Executive Director Edward G. Hammers, Sc.D. A ssociate Director Lynne P. Mlddelveen, M.S. A ssistant Director Board of Directors Stephen S. Sternberg, M.D. C hairm an ol th e Board, ACSH Member. Sloan-Kettering Institute lor Cancer Research Attending Pathologist Memorial Hospital, New York, New York Norman E. Bortaug, Ph.O. Distinguished Professor of International Agriculture Departm ent of Soil and Crop Sciences Department of Plant Sciences Texas A&M University College Station, Texas F.J. Francis, Ph.0. Professor Department of Food Science and Nutrition University of M assachusetts Amherst, M assachusetts Alfred E. Harper, Ph.0. professor of Nutritional Sciences Professor of Biochemistry University of Wisconsin Madison, Wisconsin Joseph F. Murphy, Esq. LeBoeuf, Lamb. Letby & M acflae New Yark, New York Robert . Olson, M.D., Ph.0. Professor of M edicine and Professor of Pharm acological S ciences S tate University of New York Stony Brook, New York Fredrick J. Stare, M.O., Ph.D. Professor of Nutrition, E m eritus Harvard School of Public Health Boston, M assachusetts Elizabeth M. Whelan, Sc.0., M.P.H. Executive Oirector American Council on Science and Health New York. New Ybrk Board of Scientific arid Policy Advisors Alwynelle S. Ahl, Ph.O. Michigan S tate University Roslyn 8. Alfln-Slater, Ph.D. University of California, Los Angeles G. Harvey Anderson, Ph.0. University of Toronto Donald Austin, M.D., M.P.H. Calif. Dept, of Health S ervices Lewis A. Bamess, M.0. University of South Florida Cotlege of Medicine Stephen Barrett, M.D. Lehigh Valley Com mittee Against Health Fraud. Inc. Michael T. Belongla, Ph.D. St. Louis, MO Blaine L. Blad, Ph.0. University of N ebraska David Bodansky, Ph.D. University of Washington Hlnrlch L. Bohn, Ph.O. University of Arizona John C. Bolton, M.D. San Francisco. CA Norman E. Borlaug, Ph.D. Texas A&M University Joseph F. Borzelleca, Ph.D. Medical C ollege of Virginia George A. Bray, M.0. University ot Southern California Medical C enter Edward E. Burns, Ph.D. Texas A&M University Francis F. Busta, Ph.O. University of Rorida Ogbourne Butler, Ph.D., Ret. Texas A&M University William G. Cahan, M.D. Memorial Sloan-Kettering Cancer Center Elwood F. Caldwell, Ph.O. University of M innesota John P. Caftan, M.D. Ubertyville. IL Barbara N. Campaigns, Ph.O. University ot Cincinnati College of Medicine Ernest E. Campaigns, Ph.D. Indiana University 25 Zerte L. Carpenter, Ph.D. Texas A&M University System C. Jelleff Carr, Ph.O. Columbia, MO Robert G. Cassens, Ph.O. University of W isconsin James J. Cerda, M.D. University of Florida Bobbe L. Christensen, Ph.O. University of Texas F.M. Clydesdale, Ph.O. University of M assachusetts Bernard L. Cohen, D.Sc. University of Pittsburgh Neville Colman, M.O., Ph.O. Mount Sinai School of Medicine Julius M. Coon, M.D.VPh.O. Thom as Jefferson University T.J. Cunha, Ph.O. California State Polytechnic University Bernard O. Davis, M.O. Harvard Medical School Thomas R. DeGregori, Ph.O. University of Houston Robert M. Devlin, Ph.D. University of M assachusetts John Dlebold The Oiebokj Group. Inc. Janet B. Oougfass, M.S., R.N. University of Lowell, MA Theron Downes, Ph.O. Michigan S tate University Henry Oymaza, Ph.O., R.O. University of R hode Island Merrtl Eisen bud, Sc.O. New tortt University Medical C enter James E. Enstrom, Ph.D. University of California, LosA ngelea Myron E. Essex, D.V.M., Ph.O. Harvard School of Public Health Daniel F. Parkas, Ph.O. University of Oelaware J.S. Felton, M.O. University of California, Irvine Owen R. Fennema, Ph.D. University of W isconsin Lloyd Jackson Filer, Jr., M.D., Ph.O. University of Iowa OeanC. Fletcher, Ph.O. W ashington S tale University Ralph W. Fogleman, D.V.M. Upper Black Eddy, PA J. 0. Fox, Ph.D. University of Kentucky F.J. Francis, Ph.O. University of M assach u setts Glenn Froning, Ph.D. University of N ebraska. Lincoln LaNelle E. Geddee, Ph.O., R.N. Purdue University William Paul Glezen, M.O. Baylor College of Medicine Roger E. Gold, Ph.D. University of N ebraska, Lincoln Leonard J. Goldwater, M.O. Chapel Hill, NO Vincent P. Gotz, M.S., R.Ph. University of Florida 26 Saxon Graham, Ph.O. SUNY at Buffalo Jam es Ian Gray, Ph.D. Michigan S tate University Jesse F. Gregory III, Ph.O. University of Florida Helen A. Guthrie, Ph.O. Pennsylvania S tate University Allred E. Harper, Ph.D. University of Wisconsin William R. Havender, Ph.O. Berkeley, California WSyland J. Hayes, Jr., M.D., Ph.O. Vanderbilt University School of Medicine Virgil Hays, Ph.O. University of Kentucky William E. Hazeltlns, Ph.O. Butte County, California Mosquito Abatem ent District Norman Heidslbaugh, V.M.O., M.P.H., Ph.O. Texas A&M University L. M. Henderson, Ph.D. Sandy, UT Victor Herbert, M.O., J.O. Bronx VA Medical C enter S late University of New York 0. Bonta Hlscoe, M.O. Health Central Lansing, Ml Helen B. Hlscoe, Ph.O. Michigan S tate University John Holbrook, M.D. University of Utah Robert M. Holllngworth, Ph.O. Purdue University C. Stuart Houston, M.D. University Hospital; S a sk a tc h e w a n Lucisn R. Jacobs, M.O. University of California William T. Jarvis, Ph.O. Loma Linda University Norge Jerome, Ph.O. University of K ansas Michael Kamrln,Ph.O. Michigan S tate University Edward Harold Kass, M.O., Ph.D. Harvard Medical School Philip Keeney, Ph.O. Pennsylvania State University G.A. Keyworth, II, Ph.D. Chairm an. Keyworth Co. W ashington, O.C. John Klnsella, Ph.O. Cornell University Kathryn M. Kolasa, Ph.O., R.O. East Carolina University Oavld Kritchevsky, Ph.O. The Wlstar Institute, Philadelphia Manfred Kroger, Ph.O. Pennsylvania S ta te University Leonard T. Kurland, M.O., Or.P.H. Mayo Clinic, R ochester Carolyn Lackey, Ph.O., R.D. North Carolina S tate University J. Claybum LaForce, Ph.D. University of California, Los Angeles Lawrence E. Lamb, M.O. Com m unications Inc., TX Bernard J. Uska, Ph.O. P urdue University Jam es A. Lowell, Ph.D. Pima Community College Daryl Lund, Ph.D. University of Wisconsin Howard 0. Maccabee, Ph.O., M.D. Radiation Oncology C enter W alnut Creek. CA Robert MacVIcar, Ph.O. O regon S tate University Roger P. Malckel, Ph.O. Purdue University Henry G. Manne, J.S.O. Emory University Kart Maramorosch, Ph.D. R utgers University Charles 0. May, M.O. Grantham , NH John McKetta, P.E., Ph.O. University of Texas at Austin W. W. Melvin, M.O., Sc.O., M.P.H. Colorado S tate University Robert Menzer, Ph.O. University of Maryland Thom as Milby, M.O., M.P.H. Lafayette, CA Joseph M. Miller, M.O., M.P.H. New Hampton, NH W.J. Miller, Ph.O. University of Georgia J.A. Milner, Ph.O. University of Illinois Oade W. Moeller, Ph.D. Harvard School of Public Health A. Alan Moghlssl, Ph.O. Institute of Regulatory Science, VA Grace P. Monaco, J.O. White, Fine &Verville Eric W. Mood, LL.D., M.P.H. Yale University John P. Morgan, M.O. City College of New York Robert J. Morin, M.O. UCLA Medical Center Stephen J. Moss, D.D.S., M.S. New York University Oental Center Joseph F. Murphy, Esq. LeBoeuf, Lamb, Leiby & M acRae Philip Nelson, Ph.D. Purdue University John Neuberger, Or. P.H. University of K ansas Richard Oksas, M.P.H., Pharm. 0. South Bay Free Clinic, CA J.E. Oldfield, Ph.O. Oregon S tate University Robert E. Olson, M.O., Ph.O. SUNY at Stony Brook Jane Orient, M.O. Tucson, AZ Robert Oseasohn, M.O. University of Texas Grace L. Ostenso, Ph.O. Committee on Science and Technology U-S. House of Representatives M. Alice Ottobonl, Ph.O. Berkeley, CA Rose Marie Pangbom, M.S. University of California, Oavis Michael Pariza, Ph.O. University of Wisconsin Albert Pearson, Ph.D. Michigan S ta te University T.W. Perry, Ph.O. Purdue University Timothy Dukes Phillips, Ph.O. Texas A&M University Mary Frances Plcclano, Ph.O. University of Illinois John J. Powers, Ph.O. University of Georgia William D. Powrle, Ph.O. University of British Columbia John Renner, M.D. St. Mary's Hospital K ansas City, MO Rita Ricardo-Campbell, Ph.O. Hoover Institution, Stanford University David B. Roll, Ph.D. University of Utah Dale R. Rom sos, Ph.O. Michigan S ta te University David Peter Rose, M.D., Ph.D. American Health Foundation Sheldon Rovin, D.D.S., M.S. University of Pennsylvania Edward Runge, Ph.O. Texas A&M University Paul 0. Saltman, Ph.O. University of California. San Oiego Herbert P. Sarett, Ph.O. S araso ta. FL Lowell 0. Satterlee, Ph.O. University of N ebraska Frank 0. Schaumberg, Ph.O. O regon S la te University Harry Schwartz, Ph.O. Columbia University Leroy L. Schwartz, M.O. T he Princeton Institute for Health Policy B.S. Schwelgert, Ph.O. University of California, Oavis David P. Sheridan, M.O., M.S. Medical College of Wisconsin Sidney Shindell, M.O., LL.B. Medical College of Wisconsin Sarah Short, Ph.O., Ed.D., R.O. Syracuse University A.J. Sledler, Ph.O. University of Illinois John Sfllfker, Ph.O. Silliker Laboratories, CA Julian L. Simon, Ph.O. University of Maryland S. Fred Singer, Ph.O. University of Virginia Roy F. Spalding, Ph.D. University of Nebraska Robert R. Spitzer, Ph.D.* Milwaukee School of Engineering Ronald T. Stanko, M.O. University of Pittsburgh School of Medicine Fredrick J. Stare, M.D., Ph.O. Harvard School of Public Health 27 r* /? o ' j ;; u James H. Steele, D.V.M., M.P.H. University of Texaa Judith S. Stem, Sc.O. University of California. Davis Stephen S. Sternberg, M.D. Memorial SloaxvKettering Cancer Center Elizabeth F. Stier, Ph.D. Rutgers University Martha Barnes Stone, Ph.O. K ansas S tate University Slta R. Tatinl, Ph.D. University of M innesota Steve L. Taylor, Ph.D. University of W isconsin Murray M. Tuckerman, Ph.O. Temple University John W. TUrner, M.D. Baystate Medical Center Springfield. MA Varro E. Tyler, Ph.D. Purdue University Robert P. Upchurch, Ph.D. University of Arizona Stanley E. Wallen, Ph.D. Harris Laboratories. Inc. Lincoln, NE W.F. Wardowskl, Ph.O. University of Florida Miles Weinbarger, M.D. University of Iowa Esther M. Wander, M.D. Montefrore Medical C enter Bronx. NY Philip L. White, Sc.O. American Medical Association C. K. Whitehalr, D.V.M., Ph.D. Michigan S ta te University Carol Whitlock, Ph.D., R.D. Rochester Inst, of Technology Christopher Wilkinson, Ph.D. Cornell University Warren Wlnkelsteln Jr., M.D., M.P.H. University of California, Berkeley Virgil Wodlcka, Ph.O. Fullerton, CA George T. Woods, D.V.M. University of Illinois James Harvey Young, Ph.D. Emory University John A. Zapp, Jr., Ph.D. Kennett Sq u are, PA Panayiotis Michael Zavos, Ph.D. University of Kentucky 28 / American Council on Science and Health 47 Maple Street Summit, NJ 07901 Nonprofit Org. U.S. Postage Paid Summit, N.J. Permit No. 84 CD VJ CO P art V. Carcinogenicity LONG-TERM TOXICOLOGIC STUDIES OF 2,3.7,8-TETRACHLORODIBENZO-p-DIOXIN (TCDD) IN LABORATORY ANIMALS R. J. Kociba, D. G. Keyes, J. E. Beyer, R. M. Carreon, and P. J. Gehring T oxicology R esearch Laboratory H ealth and E nvironm ental Research D ow C hem ical. U .S.A . M id la n d , M ichigan 48640 V- Introduction The compound 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is a highly toxic impurity that may be formed as an unwanted contaminant under certain conditions during the production of 2,4,5-trichlorophenoI. TCDD has been considered one of the causes of chloracne, which has been associated historically with the industrial production of 2,4,5-trichlorophenol and other products made from 2,4,5-trichlorophenol. Over the past decade, a number of toxicologic studies with TCDD were conducted to assess the potential for acute toxicity and teratogenesis. Results of these earlier studies have been summarized in a previous publication by Kociba e t a l? In that publication, we also reported the results of a subchronic study in which rats were given 1.0, 0.1, 0.01, 0.001 or 0 Mg TCDD/kg 5 days/week for 13 weeks. Doses of 1.0 Mg TCDD/kg/day caused multiple toxicologic effects, with significant morphologic changes in the liver, thymus and reproductive organs. A dose level of 0.1 Mg TCDD/kg/day caused lesser degrees of toxicity, and rats given 0.01 or 0.001 Mg TCDD/kg/5 days/week for 13 weeks had no alterations considered of any toxicologic significance. Since that time, studies have been in progress to assess the potential for chronic toxicity associated with long-term exposure to TCDD. Allen et a l? reported on a subchronic study in which monkeys consumed a diet containing 500 ppt of TCDD for up to 9 months. It was calculated that these monkeys ingested 2-3 MgTCDD/kg over the course of the 9-month study. Clinically, these monkeys had changes similar to those described by McConnell e t a l? as well as some hematologic depression and hemorrhages in various tissues. Hypertrophy, hyperplasia and/or metaplasia were noted in the epithelium of the bile ducts, salivary glands, bronchi, pancreatic ducts, sebaceous glands, skin, gastric lining and urinary tract of these monkeys given diets containing 500 ppt of TCDD. In 1969, Innes et a l? reported on a long-term carcinogenic study conducted in mice given 2,4,5-trichlorophenoxyacetic acid contaminated with a level of TCDD sufficient to supply approximately 0.27 MgTCDD/kg/day. Neither of the two strains of mice had an increase in tumors as a result of the treatment. DiGiovanni e t a l? reported on a study in which TCDD was reported to be only a weak tumor initiator in the two-stage system of mouse skin carcinogenesis with 7,12-dimethylbenz(a)anthracene (DMBA). Van Miller and Allen6 issued a preliminary report on a study of small groups of male rats fed diets containing TCDD for 65 weeks. All 10 rats of each group receiving 1.0, 0.5 or 0.05 ppm TCDD in the diet died within 4 weeks, with acute toxic effects. 397 0077-8923/79/0320-0397 S0I.75/0O 1979, NYAS 398 Annals New York Academy of Sciences Groups of male rats on diets containing 5000 or 1000 ppt of TCDD had increased mortality, decreased weight gain and liver toxicity. In a more recent publication by this same group of researchers, Van Miller e t a t.7 reported lung and/or liver tumors in those rats given 1000 or 5000 ppt of TCDD. Neoplasms of various organs, but not of lung or liver, were reported in rats given diets containing 500, 50 or 5 ppt of TCDD. Van Miller e t a t. reported a zero incidence of tumors in the group of rats given 1ppt of TCDD and in a group of 50 control rats. In view of the need for an evaluation of the chronic toxicity and potential for carcinogenicity of TCDD, the study summarized herein was conducted. For the sake of brevity, only the more salient data are included in this presentation. The full report can be found in a recent paper by Kociba e t a t} Materials and Methods The TCDD sample was prepared by the Dow Chemical Company, Midland, Michigan, USA, and had a purity exceeding 99%. On the basis of results from a previous 13-week long toxicity study1in rats dosage levels of 0.1, 0.01 and 0.001 fig of TCDD/kg/day were given in the diets. The test diets were adjusted at intervals to provide these dose levels of TCDD as required by body weight and food comsiimption determinations. Groups of 100 rats (50 male, 50 female) of the Sprague-Dawley strain were maintained on diets supplying these dose levels for two years. The control group consisted of 85 males and 86 females. Samples of the prepared diets were analyzed periodically to ascertain if the dietary levels of TCDD were being maintained as scheduled.. Parameters monitored during the course of the study included body weights, food consumption, hematologic parameters including total erythrocyte count (RBC). total and differential leukocyte counts (WBC), thrombocyte and reticulocyte counts, packed cell volumes (PCV) and hemoglobin concentration, routine urinalyses, and urinary excretion of creatinine, coproporphyrin, uroporphyrin and delta-amino levulinic acid (delta-ALA). Serum samples were collected preterminally or terminally for determination of urea nitrogen (BUN), glutamic pyruvic transaminase (SGPT), bilirubin (total, direct and indirect), cholesterol, triglycerides, alkaline phosphatase (AP), gamma-glutamyl transferase (GGT), total protein, albumin and globulin. All rats killed after two years, as well as those dying or culled during the study, were subjected to a complete necropsy examination by a veterinary pathologist. Representative tissues from all organ systems of the body of each rat were preserved in formalin fixative, with subsequent preparation of paraffin-embedded sections for light microscopy examination. Additional sections of livers collected at terminal necropsy from three female rats per group were processed for transmission electron microscopy. The weights of the liver, kidney, brain, heart, thymus, spleen and testes or ovaries/ uterus were recorded at the time of terminal necropsy. Liver and fat samples collected from three females per group at terminal necropsy were subsequently analyzed for TCDD content, using gas chromatography lowresolution spectrometry. The various data were statistically analyzed according to the procedures listed in the previously cited publication by Kociba e t a t which includes the full details of this research. R esults Analyses of the test diets indicated that rats given 0.1, 0.01 and 0.001 fig TCDD/kg/day were ingesting an average of 2,193, 208 and 22 ppt of TCDD, respectively, in the diet for two years. i:. ppt of TCDD had increased more recent publication by i lung and/or liver tumors in of various organs, but not of ; 500, 50 or 5 ppt of TCDD. e group of rats given 1 ppt of c toxicity and potential for vas conducted. For the sake presentation. The full report mical Company, Midland, '.he basis of results from a of 0.1,0.01 and 0.001 Mg of ;re adjusted at intervals to ight and food comsumption the Sprague-Dawley strain 'O years. The control group j to ascertain if the dietary7 .-* icluded body weights, food. irocyte count (RBC), total and reticulocyte counts, n, routine urinalyses, and rphyrin and delta-amino ireterminally or terminally ric transaminase (SGPT), des, alkaline phosphatase ilbumin and globulin, r culled during the study, a veterinary pathologist each rat were preserved in nbedded sections for light :ted at terminal necropsy >sionelectron microscopy, s, spleen and testes or psy. oup at terminal necropsy as chromatography low- p the procedures listed in des the full details of this 3.1, 0.01 and 0.001 ftg and 22 ppt of TCDD, j The rate of mortality was increased in female rats given 0.1 Mg TCDD/kg/day. The body weights of male and female rats given 0.1 Mg TCDD/kg/day and also the females given 0.01 Mg TCDD/kg/day were decreased. However, these parameters were not affected at the lower dose levels. , Hematologic data indicated some effects at the high dose level of 0.1 Mg TCDD/kg/day: this included slight decreases in PCV and Hgb of males and decreases in PCV, total RBC and WBC counts and Hgb values of females; reticulocyte counts appeared to be slightly increased. No effect was noted on thrombocyte counts and WBC differentials at all the dose levels studied. No hematologic changes were noted j in rats given 0.01 or 0.001 Mg TCDD/kg/day. Changes in urinary parameters were j limited only to females given 0.1 or 0.01 Mg TCDD/kg/day; these groups had i increased urinary excretion of coproporphyrin and uroporphyrin. Urinary excretion of ! delta-ALA was increased only in females given 0.1 Mg TCDD/kg/day. Table 1 Major G ross Necropsy Findings in Rats attributed to Ingestion of Diets Containing TCDD for Two Years Dose Level of TCDD 0g/kg/day) 0.1 0.01 Emaciation (M, Fe) Icterus (Fe) _* -- Liver toxicity (M, Fe) Liver toxicity (M, Fe) Liver nodules (Fe) Liver nodules (Fe) Thymic atrophy (Fe) -- Splenic atrophy (M, Fe) Anemic paleness of tissue and focal -- hemorrhages (Fe) -- Increased incidence of mesenteric periarteritis (M, Fe) Mesenteric periarteritis (Fe) Pulmonary congestion, edema. effusion and mass formation (Fe) Decrease in incidence of pituitary enlargement, endometrial hyper plasia, uterine polyps, and sub- cutaneous masses (Fe) Decrease in severity of chronic -- renal disease (M) Increased incidence of keratinized proliferative lesions near pharynx (M , Fe) -- 0.001 -- -- -- -- -- -- -- -- -- *(--) Indicates observation was not noted at that dose level, or incidence was com parable to control data. Tabulation excludes voluminous data considered unrelated to treatment with TCDD. ri ( Of all the clinical chemical parameters evaluated, the only treatment-related ( changes were noted in females given 0.1 Mg TCDD/kg/day, which had increases in SGPT, AP, and GGT activities. :'t For the sake of brevity, treatment-related gross and microscopic observations on i i ; tissues have been summarized in Tables 1 and 2. Gross necropsy examination indicated that the target organs of chronic TCDD toxicity in rats were primarily the liver, lymphoid organs, vascular and respiratory systems (Table 1). Light microscopy examination revealed the liver to be the organ most consistently affected, and rats 94 82 I 400 . Annals New York Academy of Sciences Table 2 Major Histopathologic Findings in Rats Attributed to Ingestion of Diets Containing TCDD for Two Years___________ Dose Level o f T C D D ^jg /k g /d ay ) 0.1 0.01 0.001 Liver Hepatic degenerative inflamatory, ne crotic and proliferative changes. Cytomegaly, distortion of lobular pattern and atropy of cords, cyto plasmic vacuolation, fatty change, altered tinctorial properties, he patic necrosis and inflammation, multinucleated hepatocytes, foci or areas of hepatocellular alteration (swollen hepatocytes), pigment ag gregates, bile duct hyperplasia, periportal inflammation, fibrosis Hepatocellular carcinoma (Fe) Hepatocellular hyperplastic (neoplas tic) nodules (Fe) Livers had much lesser degree of change. Hepatocellular nodules (Fe) No changes. Livers of females had statistical increase in foci or area of hepatocellular al teration (swollen hep atocytes), and livers of males had statistical decrease in area of hepatocellular altera tion (swollen hepato cytes). R ep ro d u c tiv e Decreased incidence of uterine hyperplasia, cyst formation and polyp formation (Fe) Questionable decreased incidence of uterine polyps (Fe) M am m ary Tissue Decreased incidence of mammary tumors --. E n d o crin e Pituitary--decreased incidence of pituitary hemangiectasis and ade noma formation (Fe) Adrenal--decreased incidence of medullary hyperplasia (M, Fe) and pheochromocytoma (M) Increased incidence of cortical necrosis and hemorrhage (Fe), corti cal adenoma and hematocyst (M) Pancreas--decreased incidence of acinar adenoma (M) Thyroid--occurrence of various isolated follicular changes of ques tionable significance (M) Parathyroid--decreased incidence of parathyroid hyperplasia, secondary to chronic renal disease (M) Lym phoid Isolated cases of thymic/splenic atrophy (Fe) -- -- -- -- -- -- -- R espiratory Focal alveolar hyperplasia, pigment aggregates, accumulation of alveolar macrophages and cholesterol clefts, pulmonary edema, focal in terstitial inflammation and fibrosis Focal alveolar plasia (Fe) hyper enees TED TO INGESTION Y ears 0.001 ;-------- ------------ No changes. Livers of females had statistical increase in foci or area of hepatocellular alteration (swollen hepatocytes), and livers of males had statistical decrease in area of hepatocellular altera- . .tion (swollen hepatocytes). Kociba et al.: Long-Term Toxicologie Studies of TCDD 401 Table 2 (Continued) Dose Level of TCDD (Mg/kg/day) 0.1 0.01 0.001 Keratinizing squamous metaplasia (Fe) . Keratinizing squamous cell carci noma (Fe) -- -- C a rd io va sc u la r Isolated cases o f hemorrhace in C N S. -- (Fe) Increased incidence of mesenteric and Increased incidence o f thoracic periarteritis, with cases of periarteritis (M) thrombosis or hematoma forma- tion(M , Fe) Increased incidence of myocardial degenerative changes (Fe) -- t. -- -- *(--) Indicates observation was not noted at that dose level, o r incidence was comparable to control data. Tabulation excludes voluminous data considered unrelated to treatment with TCDD. given 0.1 or 0.01 Mg TCDD/kg/day had a dose-dependent incidence of hepatic degenerative, inflammatory and necrotic changes listed in T able 2. Female rats given 0.1 MgTCDD/kg/day also had a statistical increase in lesions described morpholog ically as hepatocellular carcinomas (no metastasis) and hepatocellular nodules; females given 0.01 MgTCDD/kg/day had an increase in hepatocellular nodules. I Histopathologic examination of livers from rats given 0.001 Mg TCDD/kg/day revealed none of the treatment-related effects that were noted at the higher dose rr levels. On a statistical basis, the incidence rate of a slight hepatocellular alteration (focal swollen hepatocytes) was increased in females and decreased in males of the group given 0.001 MgTCDD/kg/day; these statistical variations were not considered ? indicative of any toxicologic significance. Ultrastructural examination of hepatocytes from females of the>0.1 Mg TCDD/kg/day revealed the most consistent change to be in the rough endoplasmic reticulum (RER) which showed proliferation, distortion and fragmentation. Smooth endoplasmic reticulum (SER) and mitochondrial structures were within control z' limits. Other .changes in hepatocytes from this group given 0.1 Mg TCDD/kg/day >& included focal cytoplasmic vacuolization, increased lysosomal activity with residual body formation and an occasional multinudeated hepatocyte. In hepatocytes from if females given 0.01 Mg TCDD/kg/day, the most notable changes included a lesser v degree of proliferation of the RER and some proliferation of the. SER. This was accompanied by a slight increase in the number of hepatocytes with lipid droplet accumulation. The hepatocytes from females given 0.001 Mg TCDD/kg/day were within the limits of variation seen in the hepatocytes of the control group. There was no general increase in lipid droplet content of the hepatocytes, although an occasional hepatocyte had some lipid droplet content. As predicted fromprevious studies, gross and histopathologic examination of some ; rats given .1 MgTCDD/kg/day had isolated cases of thymic and/or splenic atrophy. . Lower dose levels did not produce this effect. ^ The respiratory system of both male and female rats was affected by ingestion of '% 0.1 MgTCDD/kg/day, with focal alveolar hyperplasia, pigment aggregates, alveolar v macrophages, cholesterol clefts, edema, focal interstitial inflammation, fibrosis, 402 Annals New York Academy of Sciences keratinizing squamous metaplasia, and keratinizing squamous cell carcinomas. At 0.01 p g TCDD/kg/day, these effects were not noted, except for focal alveolar hyperplasia in female rats. At the dose level of 0.001 Mg TCDD/kg/day, there were no treatment-related changes in the respiratory tract. Examination of the reproductive organs of females given 0.1 (and possibly those given 0.01 Mg TCDD/kg/day) revealed a decreased incidence of the uterine changes typically encountered as spontaneous lesions in this strain of rat. The high dose group of females also had decreased incidences of both mammary and pituitary tumors. Adrenal changes noted only at the high dose level of 0.1 Mg TCDD/kg/day included a Table 3 Increased Incidence of Tumor and Tumor-like Lesions Attributed to Lsoestion by Rats of Diets Containing TCD D for Two Years Dose Level of TCDD (Mg/kg/day) 0 0 1 0.01 _______________________________________________________________ _________________ _________________ 0.001 H ard P a late/N asal Turbinates Stratified squamous cell carcinoma N um ber o f M ales with L esio n /N u m b er in Croup 0/85 4/50 0/50 0/50 Tongue Stratified squamous cell carcinoma 0/85 3/50*. 1/50 1/50 Lung Squamous cell carcinoma 0/85 1/50 0/50 0/50 H ard P a late/N asal Turbinates Stratified squamous cell carcinoma N um ber o f Females with L esio n /N u m b er in Group 0/86 4/49* 1/50 0/50 Tongue Stratified squamous cell carcinoma Lung Squamous cell carcinoma Liver Hepatocellular carcinoma Hepatocellular nodule(s) 1/86 2/49 0/86 7/49* 1/86 11/49* 8/86 23/49* 0/50 0/50 2/50 18/50* 0/50 0/50 0/50 3/50 Statistically increased above control data, p < 0.05. Table excludes additional data on incidence of tumors considered unrelated to treatment with TCDD. decreased incidence of medullary hyperplasia and pheochromocyioma formation, and an increased incidence of cortical necrosis, hemorrhage, hematocyst and adenoma formation. The incidence of pancreatic acinar adenoma was decreased in males given 0.1 p g TCDD/kg/day. This group of males also had a decrease in the incidence rate of secondary parathyroid hyperplasia; this was due to a decrease in the severity of chronic renal disease noted in males given this high dose level of 0.1 p g TCDD/kg/day. Examination of the cardiovascular system revealed several effects assumed to be due to ingestion of 0.1 p g TCDD/kg/day. These effects included isolated cases of hemorrhage in different organs of the body, and an increase in the incidence of periarteritis and myocardial degenerative changes, both of which occur spontaneously ,, - ,-?5 . r-r-.;.Jrxv.. fri X X' .; :X' ^X .''XvX.;\ .^ r'XX:i >^XV'-X,X';'X res is cell carcinomas. At ;pt for focal alveolar 'kg/day, there were no i.l (and possibly those of the uterine changes t. The highdose group and pituitary tumors. DD/kg/day includeda Lesions D P for Two Y eaVrHsS >D ( M g / k g / d a y ) 0.01 0.001 M ales iber in Group 0/50 0/50 1/50 1/50 J/50 0/50 F em a les ib er in Group 1/50 0/50 0/50 0/50 0/50 0/50 2/50 18/50 0/50 3/50 unrelated to treatment ytoma formation, and itocyst and adenoma in males given 0.1 Mg the incidence rate of se in the severity of ise level of 0.1 Mg effects assumed to be ded isolated cases of : in the incidence of l occur spontaneously Kociba et al.: Long-Term Toxicologic Studies of TCDD 403 Table 4 Decreased Incidence of Tumors Attributed to Ingestion by Rats of Diets Containing TCDD for Two Years Pancreas Acinar adenoma Pheochromocytoma Dose Level of TCDD (Mg/kg/day) 0 0.1 N u m b e r o f tW ales hith L e s io n s / T o ta l N u m b e r in G ro u p 14/85 28/85 2/50 , 4/50* N u m b er o f Fem ales hiih L esion /T o ta l N um ber in Group U teru s Benign tumor 30/86 8/49 I M a m m a ry Gland Benign tumor 73/86 24/49 P itu ita ry Adenoma 43/86 12/49 Statistically decreased from control data, p < 0.05. Data at lower dose levels comparable to controls, except for possible decrease in incidence of benign uterine tumors in females given 0.01 Mg TCD D /kg/day. in this strain of rat. None of these observations were noted at the two lower dose levels, except -for. an increase in incidence of periarteritis in male rats given 0.01 Mg TCDD/kg/day. Other treatment-related effects included an increase above the background incidence of squamous cell carcinomas of the tongue and hard palate/nasal turbinate region of rats given 0.1 Mg TCDD/kg/day. Historically, squamous cell carcinomas have occurred in these organs of this strain of rat at a spontaneous incidence rate of 1-3%. Only the high dose level of 0.1 Mg TCDD/kg/day increased the incidence of this type of neoplasm. All the data on incidence rates of neoplasms that were increased or decreased by treatment with the high dose level of 0.1 Mg TCDD/kg/day are summarized in T ables 3 and 4. Review of organ weights recorded at the terminal necropsy indicated the following effects considered related to treatment (a) an increase in absolute and/or relative liver weight of rats given 0.1 or 0:01 MgTCDD/kg/day and (b) a decrease in weight of the thymus of females given 0.1 Mg TCDD/kg/day. There were no organ weight changes at the dose level of 0.001 Mg TCDD/kg/day. T able 3 lists the-TCDD content of liver and fat samples collected from female rats at terminal necropsy after two years of ingesting diets containing TCDD. -V. Wfic-' Table 5 TCDD. Content in Rat Livers and Fat Samples Collected AFTER Two Years Ingestion of Diets Containing TCDD Dose Level of TCDD (Mg/kg/day) 0.1 0.01 0.001 TCDD Content (ppt) Fat Liver 8100 24000 1700 5100 540 540 19 4 9 i 404 Annals New York Academy of Sciences' Discussion The results of this study of rats ingesting TCDD for a lifetime serve as a basis for assessing the long-term chronic toxicity of TCDD. Continuous ingestion of a high dose level of 0.1 n g TCDD/kg/day (approximately 2200 ppt in diet)*predictably caused multiple toxicologic effects. Liver toxicity was the most consistent observation, and this was accompanied by morphologic changes of the lymphoid, respiratory, and vascular tissues of the body. The incidence of hepatocellular carcinomas of the liver and squamous cell carcinomas of the lung, hard palate/nasal turbinates or tongue was increased at this dose level. Conversely, the incidence of tumors of the pituitary, uterus, mammary gland, pancreas and adrenal medulla was decreased at this high dose level of treatment. Similarly, the incidence of other spontaneous lesions such as chronic renal disease was also decreased at this high dose level. Lifetime ingestion by rats of 0.01 jig TCDD/kg/day (approximately 210 ppt in diet) caused a lesser degree of toxicity, primarily of the liver. However, there was no increase in the incidence of neoplasia at this dose level. Rats ingesting 0.001 /zg TCDD/kg/day for two years had no adverse effects in spite of the fact the liver and fat each contained 540 ppt of TCDD at termination of the study. Thus, these data indicate a good dose response for the long-term toxicity of TCDD. Whereas higher doses were predictably toxic and did alter the incidence of tumors, an intermediate dose level caused lesser toxicity but no neoplasia. A lower dose was tolerated for a lifetime with no adverse effects noted in any of the parameters evaluated. R eferences 1. Kociba, R. J., P. A. K eeler , C. N. Park & P. J. Gehring. 1976. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD): Results of a 13-week oral toxicity study in rats. Toxicol. AppL Pharmacol. 35: 553-574. 2. Allen, J. R,, D. A. Barsotti, J. P. Van Miller. L. J. abrahamson, & J . J . Lalich. 1977. Morphological change in monkeys consuming a diet containing five hundred parts per trillion of 2.3,7,8-tetrachloro-dibenzo-p-dioxin. Food Cosmet Toxicol. 15:401-410. 3. McConnell, E. E., J . A. Moore & D. W. Dalgard. 1978. Toxicity of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) in rhesus monkeys (Macaca mulatta) following a single oral dose. Toxicol. Appl. Pharmacol. 43: 175-187. 4. Innes, J. R. M., B. M. U lland, M. G. Valerio. L. Petrucelli, L. Fishbein, E. R- Hart, A. J. Pallotta. R. R. Bates, H. L. Falk, J. J. Gart, M. Klein, I. Mitchell & J. Peters. 1969. Bioassay of pesticides and industrial chemicals for tumorigenicity in mice: A preliminary note. J. Nat. Cancer Inst. 42: 1101-1114. 5. DiGiovanni, A. Viaje, D. L. Berry, T. J. Slaga & M. R. J uchau. 1977. Tumor initiating ability of 2,3,7,8-tetrachlorodibenzo-/vdioxin (TCDD) and Arochlor 1254 in the twostage system of mouse skin carcinogenesis. Bull. Environ. Contam. Toxicol. 18: 552557. 6. Van Miller, J. P. & J. R. Allen. 1977. Chronic toxicity of 2,3,7,8-tetrachlorodibenzo/j-dioxin in rats. Fed. Proc. 36:396 (Abstract 573).- 7. Van Miller, J. P,, J. J. Lalich & J. R. Allen. 1977. Increased incidence of neoplasms in rats exposed to low levels of 2,3,7,8-tetrachlorodibenzo-p-dioxin. Chemosphere 9: 537544. 8. Kociba, R. J., D. G. Keyes, J. E. Beyer, R. M. Carreon, C. E. Wade, D. A. Dittenber, R. P. Kalnins, L. E. Frauson, C. N. Park, S. D. Barnard, R. A. Hummel & C. G. Humiston. Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) in rats. Toxicol. Appl. Pharmacol. In press, 1978. CLASSIFICATION OF SCFT TISSUE TUMORS AND TUMOR-L IKE LESIONS Site TUMORS OF FIBROUS TISSUE Non-Neopl asti c L es i ons F as ci i ti s Kel oi d El astof i broma Fibromatosi s TUMORS OF TENDOSYNOVIAL TISSUE Tendosynovi al cyst Tendosynovi ti s Synovial chondromatosis TUMORS OF ADIPOSE TISSUE Fat necrosis Lipogr anul ana Lipodystrophi a Piezogeni c papul es Adiposis dolorosa Steatopygi a TUMORS CF MUSCLE TUMORS OF VESSELS Atrophy Dystrophy Polymyosi ti s Rhabdomyolysi s Proliferative myositis Fibromatosi s Myositis ossificans Pyogeni c granul ana Angi obi asti c lynphoid hyper pi as i a Hereditary hemorrhagic te l angi ectasi a Arten' ovenous f i stul a Beni gi Neopl asti c L es i ons Beni gi f i brobl asti c fibrous histiocytoma Benign h i s t i o c y t i c fibrous histiocytoma Benigp pleomorphic fibrous histiocytoma Fibroma F i broma Lipoma Villonodular synovitis Benigi giant cell tunor Wei 1 - d i f f erenti ated 1i poma Myxoi d 1i poma Fibrobl asti c 1 ipoma Pleomorphi c 1ipoma Angiol ipoma Angiomyjl ipoma Myelol ipoma Lipobl astoma Hibernoma Leiomyoma Angianyol ipoma Rhabdomyoma Malignant Neoplastic L es i ons Malignant fib ro b la stic fibrous histiocytoma Mali giant h istio cytic fibrous histiocytoma Malignant pleomorphic fibrous histiocytoma Desmoi d tunor Fibroblastic fibrosarcoma Pleomorphic fibrosarcoma Malignant giant cell tunor Biphasic tendosynovi al sarcoma Monophasic sp in d le c e ll type tendosynovi si sarcoma Epi thel ioi d sarcoma Clear cell sarcoma Chordoid sarcoma W e ll-d iffe re n ti ated l i p o s ar coma Myxoi d 1iposarcoma Lipobl asti c 1iposarcoma Fibroblastic 1iposarcoma Pleomorphic 1iposarcoma Lei cmyosarcoma Leiomyobl astoma Embryonal rhabdomyosarcoma Pleomorphic rhabdomyosarconu Rhabdomyobl astoma C a p i l l a r y hemangioma Cavernous hemangioma Venous hemangioma Papi 11 ary endothel i al hyper pi asi a Hypertrophic hemangioma Angiomatosi s Angi of ibroma Angi ol i poma Angiamyol ipoma Hemangi obi astoma Glanus tunor Lymph angi oma Lymph angi amyoma Hen angi operi cytoma Hen angi os ar coma K a p o si' s sarcoma Lymph angi osar coma J1L Qo 'pj.- O 9'j CLASSIFICATION OF SCFT TISSUE TUMORS AND TUMOR-LIKE LESIONS ( c o n t in u e d ) TUMORS OF PERIP FERAL NERVE Non-Neopl asti c L es i ons Degenerati on Hypertrophy Trainati c neuroma EXTRAS KE LETAL BCNE TUMORS MISCELLANEOUS SCFT TISSUE TUMORS Chondroi d met api asi a Osteoi d met api asi a Myositis ossificans Benign Neoplastic L es i ons Neurof i broma Neurof i bromatosi s Angi oneuroma Plexiform neuroma Pacinian neurofibroma Glandular schwannoma Pleomorphic schwannoma Cysti c schwannoma S tori form schwannoma E p it h e lio id schwannoma Chondroma Osteochondroma Osteoma Benign granular cell tunor Benigi mesenchynoma Mai ignant Neopl asti c Lesi ons Malignant collagenous schwannoma Malignant non-coll age no us schwannoma Primitive neuroectrodermal timor Osteogenic sarcoma Chondros arcoma Ewi ng1s sarcoma Malignant granular cell tun or Alveolar s o f t part sarcoma Malignant lynphoma Granulocytic sarcoma Plasmacytoma P o s t ir r a d ia t io n sarcoma Malignant mesenchynoma U n sifferen ti ated s o f t tissue sarcoma From: Hajdu, S . I . : Pathology of S o f t Tissue Tunors, Lea & Febiger, P h ila d e lp ia , 1979 "i d0 . 9 GROWTH PATTERN 1. Arranged pattern 2. Spreading pattern 3. Lacy pattern 4. E p it h e lio id pattern 5. Alveolar pattern 6. Disarranged pattern CELL MORPHOLOGY 1. Slender sp in d le c e l l s 2. Plunp sp in d le c e l l s 3. Granular e p it h e lio id c e l l s 4. Clear e p it h e lio id c e l l s 5. Isomorphic giant c e l l s 6. Pleomorphic giant c e l l s APPEARANCE OF STROMA 1. F i b r i l l a r stroma 2. Sclerosed stroma 3. Myxoi d stroma 4. Vascular stroma 5. Inflamed stroma 6. Necroti c stroma 7. Chondrified stroma 8. O ssifie d stroma 9. C a lc i f ie d stroma PRODUCT OF CELLS 1. Collagen 2. Glycogen 3. Polysaccharides 4. Fat 5. Melanin 6. Secretory granul es 7. C r y s t a ls 8. Fine structure 9. Tissue antigens From: Hajdu, S . I . : D if f e r e n t ia l D ia gn osis of S o ft T issu e and Bone Tunors, Lea & Febiger, Ph iladelp hia, In Press 19500 REFERENCES Hajdu, S . I . , Bean, M.A., Fogh, J, Hajdu, E.O. and R i c c i , A.: Papanicolaou smear of cultured hunan tunor c e l l s . A c t a C y t o l . 18:327332, 1974. Hajdu, S . I . and Hajdu, E.O.: Cytopathol ogy of Sarcomas and Other None p i t h e li a l Malignant Tunors. W.B. Saunders, Ph iladelphia, 1976. R u s s e ll, W.O., Cohen, J . , Enzinger, F . , Hajdu, S . I . et'_al_: A c li n i c a l and pathological staging system f o r s o f t tissu e sarcomas. Cancer 40:1561-1570, 1977. Hajdu, S . I. : Pathology of Soft Tissue Tunors. Philadelphia, 1979. Lea & Febiger, Levine, E . , Lee, K .R ., Neff, J .R., Makland, N.F., Robinson, R.G. and Preston, D.F.: Comparison of computed tomography and other imaging m odalities in the evaluation of musculoskeletal tunors. Radiology 131:431-437, 1979. McDonald, A.D. and McDonald, J.C.: Malignant mesothelioma in North America. Cancer 46:1650-1656, 1980. Hajdu, S . I .: S o ft tissu e sarcomas. C l a s s i f i c a t i o n and natural h ist o ry . Ca. 31:271-280, 1981. Hajdu, S . I . , Lemos , L .B., K o za la/ich , H., Helson, L. and Beattie, E.J.: Growth pattern and d iffe r e n tia tio n of hunan s o f t tissu e sarcomas in nude mice. Cancer 47:90-98, 1981. Fine, G ., Hajdu, S . I ., Morton, D.L ., Eil ber, F.R ., Sui t, H.D. and Weiss, S.W.: So ft Tissue Sarcomas. C l a s s i f i c a t i o n and Treatment (A Symposium). S.C. Sommers and P.P. Rosen (e d it o r s ) . Pathology Annual 17:155-195, 1982. Weiss, S.W. and Enzinger, F.M.: Epidermoid hemangioendothelioma. A vascular tunor often mistaken fo r a carcinana. Cancer 50:970 - 981, 1982. Denk, H., Krepler, R ., A r tlie b , U. et_ _al_: Proteins of intermediate filam ents. An inmunohistochemical and biochemical approach to the c l a s s i f i c a t i o n of s o f t tissu e tunors. Am. J. Pathol. 110:193 - 208, 1983. Hajdu, S . I . : D if f e r e n t ia l D iagnosis of S o f t T issu e and Bone Tunors. Lea & Febiger, Philadelphia, In Press. Shiu, M.H., Turnbull, A.D., N o r i, D., H i l a r i s , B. and Hajdu, S . I . : Control of l o c a l l y advanced extremity s o f t tis s u e sarcomas by fu n c tio n sv in g resection and brachytherapy. Cancer, In P re ss. 19501 Cheirosphere,Vol.1 2 ,N o.4 /5 ,nn 429-446,19e3 Printed in Treat B ritain 0045-6535/93/050429-18*03.00/0 1983 Perrair.on P r e s s L td . Reprinted by permission of Pergamon Press Ltd. STATUS OF ANALYTICAL SYSTEMS FOR THE DETERMINATION OF PCDDs AND PCDFs Warren B. Crummett Analytical Laboratories 574 B uilding Dow Chemical U.S.A. Midland, MI 48640 During the past decade the development of ana lytical methodology for the determination of chlorinated dibenzo-p-dioxins (CDDs) in many and varied m aterials has been e x tra o rd in a rily comprehensive in scope, s e le c t iv it y , and s e n s it iv it y . The state o f the a rt i s now such that ana lytical methodology protocols can be custom-made to generate data having any desired degree of ce rta in ty at concentration le v e ls down to sub parts per t r i l l i o n . The rig o r required to achieve assurance o f the q u a lity o f data varie s with the many factors which govern the separation and measurement processes. Although these fa cto rs often do not need to be optimized, they must be understood, controlled, and monitored in order to under stand the ce rta in ty o f the r e s u lt s . ' >12 This has been emphasized by a National Research Council of Canada1*2 panel who wrote a comprehensive review o f PCDD a n a ly sis with emphasis on lim ita tio n s. The choice o f methodology depends on the matrix to be analyzed, the s e n s it iv it y , s p e c ific it y and the degree of ce rta in ty required in the data. Thus, a n a lytica l systems su ita b le fo r the determinations of the CDDs in commercial products do not need to be as so phisticate d as those applied to the environmental m a te ria ls.15 Various a n a ly tica l systems have been p re viou sly c la s s if ie d 6; 11* ) 15 according to the degree o f re so lu tio n achieved in each o f three stages: (1) sample preparation, (2) sample introduction into the mass spectrometer, and (3) mass spectrometry. Sample preparation c o n sists of (1) i n i t i a l sample treatment, (2) liq u id - liq u id p a rtitio n , (3) column chromatography, and (4) high pressure liq u id chromatography. These are used in numerous combinations to iso la te the PCDDs from most o f the interferences. References to the early work have been given in a previous re p o rt.15 Recent pu blications which are repre sentative o f sta te -o f-th e -a rt technology and present d e ta ils o f the techniques are given in the follow ing table: 430 TABLE 1: Analytical System - Stage I: Extraction-Cleanup Techniques INITIAL SAMPLE TREATMENT Dissolution Digestion In Base In Acid Direct Extraction LIQUID-LIQUID PARTITION Against Cone. H 2S 0 4 Against Aqueous Base RECENT REFERENCES C O C H R A N E et al. (1981) JENSEN et al. (1981) LANGHORST 4 SHADOFT (1980) ZABIK 4 ZABIK (1980) C O C H R A N E et al. ( 1 9 B U H A R L E S S et al. (1960) J E N S E N et al. (1981) P f E I E E E R et al. (1980) COLUMN CHROMATOGRAPHY Alumina Silica Gel Flonsil Ion Exchange Resin Charcoal Gel Permeation Reagent Modified Silica COCHRANE eteal. (1981) C O C H R A N E et a l . (1981) RYAN & PILON (1980) B LASER et al. (1978) HUCKINS (1978) D O U G H E R T Y et al. (1980) LAMPARSKI 4 NESTRICK (1980) HIGH PERFORMANCE LIQUID CHROMATOGRAPHY Normal Phase Reversed Phase *R e v e r s e d Normal Phase LAMPARSKI 4 NESTRICK (1980) LANGHORST 4 SHADOFT (1980) RYAN 4 PILON (1980) LAMPARSKI 4 NESTRICK (1980) LANGHORST 4 SHADOFT (1980) These sample preparation techniques have been used in many and varied combinations to su c c e ssfu lly separate PCDDs and PCDFs from a wide va rie ty of matrices including plant and animal tissu e s and products, chemical products, sediment, s o il, combustion products, and waste m aterials. A great deal is known about the re la tiv e effectiveness for doing th is assuring that measurable well-defined sig n a ls are obtained. The power of high performance liq u id chromatography to separate isomers has been demonstrated by the separation of a ll 22 isomers of T<,CDD*7 and the 10 isomers of HSCDD33 using reversed phase and normal phase HPLC consecutively. Although co lla b o ra tive studies using these techniques have not yet been done, several la b o ra to rie s50' 57i 39 have reported success using these and sim ila r modified approaches. Very few d ire ct comparisons of the e ffic ie n c ie s o f the various extraction-cleanup procedures have yet been done. One o f the most elu cid a ting has been that o f the U.S. Food and Drug A d m inistration5 who examined s ix extracts o f f is h samples (3 spiked and 3 unspiked) prepared by each o f s ix d iffe re n t labora tories using procedures in use at that time in th e ir respec tive la b o ra to rie s. Under the GC/MS conditions used by the FDA the procedures o f Lamparski et a l . 35 and Huckins et a l . 29 were the most effective. SAMPLE INTRODUCTION Introduction o f the sample into the mass spectrometer is accomplished by (1) d ire c t probe, (2) packed column gas chromatography, (3) c a p illa r y column gas chromatography, or (4) a second mass spectrometer. The various types o f these techniques are summarized in the follow ing tab! e s . 1950 431 TABLE 2: Analytical System - Stage I I : TECHNIQUE Direct Probe Pached Column GC Mass Spectrometer Sample Introduction SELECTED RECENT REFERENCES iB A U G H M A N MESELSON (1973) ov-i V-101 OV-3 OV-7 OV-17 O V - 105 OV-2IO CV-225 OV-17/Poly S-179 FAHELLI et 1. (I960) PHILLIPSON & PUMA (1980) LAMPARSKI et >1. (1978) E RX et al. (1979) LI8 ERTI et 1. (1981) LAMPAKSKI et al. (1978) H ASS et al. (1978) C H A E et al. (1977) CHESS & GROSS (1980) LAMPAKSKI & NESTRICK (1981) SE-30 X-60 Carbowax 20M SP-2100 Dexil 300 C O C H R A N E et al. (1981) H I E U R E et al. (1977) E I C E M A N et al. (1980) PKILLIPSON & PUMA (1980) V A N NES S et al. (1980) Caoillarv Column GC OV-17 WCOT OV-61 WCOT 0V-10 WCOT OV-17/Poly S-179 S i l a r IOC SE-30 WCOT SE-54 SP-2100 Methyl Silicone WCOT LIBERTI et al. (1981) BUSER (1975) CAVALLARO et al. (1982) N E S T R 1 C X et al. (1980) BUSER 6 RAPPE (1980) H A R L E S S et al. (1980) C O C H R A N E et al. (1981) M 1 T C H U M et al. (1980) N O R S T R O M et al. (1982) As can be e a sily seen from Table 2, there are many ways whereby samples can be introduced into the mass spectrometer. Although no d ire ct comparison o f these various methods has been made, many in ve stig a to rs become strong advocates o f the la te st technology which presently is c a p illa r y column gas chromatography. The main advantage o f these columns is th e ir c a p a b ility to separate isomers. For example, Buser and Rappe7 demonstrated that 2,3,7,8-TCDD could be separated from a ll i t s 21 isomers by use o f a S i la r IOC c a p illa ry column using OV-17 and 0V-101 c a p illa r ie s fo r confirm ation. Harless et a l . JI*>25 emphasize "the total enhancement of the high reso lu tio n mass spectrometry detection techniques" produced by the use o f c a p illa ry columns. MEASUREMENT BY MASS SPECTROMETRY Mass spectrometry in the electron impact mode was f i r s t reported for the id e n t ific a tio n of PCDDs by Ress (1970). Since then i t has had extensive development. As shown in the table below, at le a st a dozen models operating at low re so lu tio n (M/AM <3000) have been used to both id e n tify and determine PCDOs after separation by Stages I and I I . 43? TABLE 3: Stage I I I : Mass Spectrometry (Low Resolution Electron Impact) M/AM <2000 MS MODEL LKB-9000 LKB-900QS LKB-209I FINNICA* 3000 FINNICA* 3200 FINNIGAN 4023 KEVP-5984A -5992A -5992 E X T R A - UC. K R A T O S 30 VARIA* - 311A t c d d LIMIT OF DETECTION PX 6 5-10 2-10 50 20 250 50 20 40-80 1-2 20 40-60 5-10 1-5 100C 5 10 M/AM 600 400 400 -- ... 400 UNIT UNIT UNIT UNIT UNIT UNIT UNIT UNIT UNIT 1000 1000 REFERENCE CRUMMETT & STEHL (1973) KUMMEL (1977) SHAD0F7 & KUMMEL (1978) AOAMOLI et 1. (1978) DI D O M E N I C O et a l . (1979) FANELLI et a l . (1980) ADAMOLI et a l . (1976) DI D O M E N I C O et a l . (1979) C AVALLARO et a l . (1980) TOSINE et a l . (1981) DI D O M E N I C O et a l . (1979) LAMPARSXI & NESTRICK (1980) NORSTROM et a l . (1961) REES et a l . (1981) TIERNAN et a l . (1980) LANGHORST & SKADOFF (1980) RYAN & PILON (1980) Very l i t t l e work has been done at medium re so lu tio n (M/aM = 2500 to 9000) as shown below. Although th is approach would be expected to have some advantage in the s e n s it iv it y over that o f the same instrument operated at higher reso lu tio n. TABLE 4: Stage I I I : Mass Spectrometry - Medium Resolution Electron Impact MS MODEL OEATOS-30 KRATOS-50 T.CDD LIMIT OF DETECTION PK 5-10 50 M/M 3000 2000 REFERENCE KUMMEL (1977) CHESS & CROSS (1980) Al though high reso lu tio n mass spectrometry (M/AM >9000) was used in the early work o f Baughman and Mese!son (1973), i t has only recently been emphasized by some mass spectrosco p ists as being important. This is p rim a rily due to the work o f Harless et a l. (1980) The experience with th is approach is summarized below: TABLE 5: Stage I I I : Mass Spectrometry - High Resolution Electron Impact MS MODEL TCDD LIMIT OF DETECTION PR M/AM REFERENCE V A R I A N 3 1 1A 5-10 9000 H A R L E S S et .1. (1930) AEI MS-9 KRATOS MS-30 5 10000 BAUGHMAN & MESELSON (1973) 100 12000 TIERNAN et a l . (1980) KR A T O S M S - 30 KRATOS MS-50 2-10 1-10 9000 10000 SHADOFF 3 HUMMEL (1978) SMITH et a l . (1981) Chemical ion iza tion mass spectrometry has been under in v e stig a tio n as a p o ssib le detector and monitor fo r PCDDs. The qua ntita tive re su lts o f t h is work are summarized in the follow ing table: j 19505 433 TABLE 6: Stage I I I : Mass Spectrometry - Chemical Io n iza tio n MS MODEL FINNIGAN 3300 T,CDD LIMIT OF DETECTION _______ El________ SO-SOO M/OH UNIT REFERENCE HASS I al- (197) EJCTRA-NUC. UNIT NITCHUM et e l . (1980) CAIRNS et e l . (1980) More re ce n tly55 (MS/MS) has been used as a detector. Using a Finnigan t r ip le stage quadrupole mass spectrometer the lim it of detection for 2,7-DCDD and 1,3,7-T 3CDD was found to be about 10 picograms. It is evident from the tables that the lim it of detection for TtCDD is very much the same on high re so lu tio n instruments as on low reso lu tio n ones. The advantage of high re so lu tio n lie s in it s a b ilit y to separate interferences in the mass spectrometer. C rite ria for the confirm ation o f the homologue and s p e c ific isomer i s essential to any a n a ly sis. Harless et a l. (1980) have proposed the follow ing: HARLESS CRITERIA FOR HOMOLOGUE AND SPECIFIC ISOMER CONFIRMATION 1. Correct HRGC-HRMS retention time for 2,3,7,8-Ti,CDD 2. Correct HRGC-HRMS m ultiple ion response for 37C1 u-TuCDD and TwCDD masses (simultaneous response for elemental composition of m/z 320, m/z 322, m/z 328). 3. Correct chlorine isotope ra tio for the molecular ions (m/z 320 and m/z 322). 4. Correct responses for the co -in jectio n o f sample fo rt ifie d with 37C1u-Ti.CDD and T..CDD standard. 5. Response of the m/z 320 and m/z 322 must be greater than 2.5 times the noise le v e l. SUPPLEMENTAL CRITERIA PROPOSED BY HARLESS (A) C0C1 lo s s in d ica tiv e o f T,,CDD structure, and (B) HRGC-HRMS peak matching a n a ly sis o f m/z 320 and m/z 322 in real time to confirm the TuCDD elemental composition. These c r it e r ia are useful and are agreeable to almost a ll a na lytical s c ie n t is t s . However, i t is not necessary to specify c a p illa r y gas chromatography (HRGC) in either c rite rio n 1 or 2 since i t i s not needed i f high re so lu tio n liq u id chromatography (HPLC) i s used. A lso , the lim it o f detection in c r it e r ia 5 would be better defined as 3 times the noise le v e l. The supplemental c r it e r ia are nice to have but u su a lly cannot be obtained at concentration levels near the lim it o f detection where such information is most needed. l!hen HPLC i s used in a manner to separate a ll 22 TCDD isomers, the follow ing c r it e r ia pertains: HESTRICK AND LAM.PARSKI CRITERIA FOR SPECIFIC ISOMER CONFIRMATION 1. Correct reverse-phase HPLC retention time fo r TCDD isomers 2. Correct normal phase HPLC retention time for TCDD isomers 3. Correct GC (packed or c a p illa r y ) retention time for TCDD isomers 4. M u ltip le ion detection at m/z 320, 322, and 324 with correct C l-iso to p e ra tio 5. Response o f m/z 320 and m/z 322 greater than 3 times the noise level 434 THE ANALYTICAL PROCESS A ll three stages o f the a n a lytica l process may be thought o f as co n sistin g o f three categories o f re so lvin g power: low (L), medium (M), and high (H). The categories o f sample preparation are described as follows: TABLE 7: Stage I - Sample Preparation RESOLVING POWER ________________ DESCRIPTION________________ L CHEMICAL TREATMENT AND/OR EXTRACTION NO CHROMATOGRAPHY M L COLUMN CHROMATOGRAPHY H M HIGH PRESSURE LIQUID CHROMATOGRAPHY The sample introduction into the mass spectrometer may be described as follow s: TABLE 8: Stage I I - MS Sample Introduction RESOLVING POWER ___________________D E S C R I P T I O N ________________ L DIRECT PROBE (NO GAS CHROMATOGRAPHY) H PACXED COLUMN GC H CAPILLARY COLUMN GC Likewise, the operation of the mass spectrometer may be assigned rankings as follow s: TABLE 9: Stage I I I - Mass Spectrometry (MS) RESOLVING POWER L M H ___________________ D E S C R I P T I O N ___________________ LOW RESOLUTION (M/AM <2500) MEDIUM RESOLUTION (M/AM 2500 to 9000) HIGH RESOLUTION (H/AM > 9000) O bviously, these can be put together in a ll p o ssib le combinations to creat various methods. Thus, methods could be developed which could be described by each o f these acronyms. TABLE 10: Possible Analytical Methods For The Determination of PCDDs and PCDFs LLL MLL HLL LLM MLM HLM LLH MLH* HLH LML MML* HML* LMM MMM KKM LMH MMH* KMH LHL MHL HHL* LKM MHM HHM LHH MHH* HHH* At le a st seven o f these {marked with a s te ris k s ) have been used recently to determine dio xin s in environmental samples. Some o f the studies are summarized in Table 11 along with the lim it s o f detection and estimated a b ilit y to resolve TCDD isomers. r r X1 U i 435 TABLE 11: Performance o f PCDO A n a lytica l Methods on Environmental Samples ANALYTICAL SYSTEM LIMIT OF DETECTION.opt MLH 0.2-3 MMH 0 .1-6 MMH 3-6 MMH 1-6 HML 0-5-1 MMH 2-4 MHH 3-5 MML MML MML MML HML HHL KHL MHH KHH 1 2-10 10 3-5 4-20 0 .6 - 2 1-2 4-13 1-5 NUMBER OF T 4CDD ISOMERS SEPARATED SAMPLE MATRIX 0 HUMAN MILK 4-12 HUMAN MILK 4-12 HUMAN MILK 15-22 HUMAN MILK 15-22 HUMAN MILK 4-12 HUMAN ADIPOSE TISSUE REFERENCE M I S E L S O N & O 'K E E F E ( 1 9 7 7 ) GROSS (1980) TIERNAN (1981) SHADOFF el al. (1977) L O OMIS et 1. (1980) * GROSS (1980) 15-22 4-12 4-12 4-12 4-12 HUMAN ADIPOSE TISSUE BOVINE MILK H A H L E el *1. (1977) B O VINE TIS S U E S J ENSEN el al. (1981) FISH FUWJ H A R A el al. (1975) CULL EGGS NORSTROM et al. (1982) 15-22 22 22 15-22 22 FISH FlYASH FISH FISH FISH LAUPARSXI et al. (1979) NESTRICK 8 LAMPARSKI (1982) T O SINE et *1. (1981) HARLESS 8 LEVIS (1982) S M I T H et 1. (1981) Of course for most o f these studies the s p e c ific it y o f the method for 2,3,7,8-TCDD is the most important consideration and many o f these methods are sp e c ific fo r t h is isomer. Even though each step in the methodology ju st described is well understood, re su lts generated by it s use need to be c a re fu lly interpreted and documented i f they are to be compared with other sim ila r data. This is made cle a r by the ACS g u id e lin e s1 and reaffirmed at the IUPAC Conference in H e ls in k i12. A Canadian panel112 applied the ACS guidelines to rate publications and found only two that came close to meeting the gu id elines. However, the references which are cited in the follow ing tables a ll contain essentia l elements which permit the data to be compared even though co lla b o ra tive studies have not always been done. To be included in the tables the data must meet the follow ing c r it e r ia : TABLE 12: C rite ria For Intercom parability 1. Isomer s p e c if ic it y defined and sim ila r 2.. Su ffic ie n t recovery studies reported 3. Interferences removed or th e ir role explained 4. Appropriate a n a lytica l standards used 5. Detection lim it s defined and reported 6. Sampling protocol documented ; I t is of in te re st to look at the re su lts o f some o f these methodologies in more detail e sp e c ia lly as they relate to a p a rtic u la r matrix. Table 13 summarizes data taken from the i determination o f 2,3,7,8-TCDD in fis h . P o sitiv e re su lts were reported fo r fis h from bodies o f water close to in d u stria l s it e s but not found in those accummulating water from areas j heavily sprayed with 2 ,4 ,5-T. The p o sitiv e re s u lts , however, have not been system atically i > 135C8 436 traced to a p a rtic u la r source causing Harless (1982) to conclude that "the source or sources o f the contamination i s unknown". Data taken by d iffe re n t labora tories on f is h from the same body o f water appear to be in reasonable agreement although the methodology is considerably different. Data taken on various species o f f is h shown in Table 14 are generally higher on bottom feeders. This would be predicted and gives further confidence in the r e l ia b i li t y of the data. TABLE 13: Search For 2,3,7,8-TCDD In Fish LABORATORY Dow ANALYTICAL SYSTEM MHL SAMPLE SOURCE Texas Reservoir YEAR SAMPLED 1976 NUMBER OF SAMPLES 20 NUMBER APPARENT POSITIVE RESULTS 0 APPARENT TCDD CONC. RANGE ND5-7) REFERENCE SHADOFF et al. (1977) Arkansas Rice Pond 1976 20 1 ND(6)-6(5) Dow HML Michigan Tittabawassee River Saginaw Bay 1976 10 2 NESTRICK & LAMPARSE1 (1979) 9 ND(20)-170(120) 0 ND(10) MDNR-EPA MHH Michigan Tittabawassee River Saginaw River Saginaw River Mouth Saginaw Bay Grand River Lake Michigan 1978 18 5 3 2 7 1 OSWALD (1978) 13 N D ( 2 ) - 6 9 5 6 N D ( 11)-105 HARLESS (1982) 3 11(6)-153 1 *D(3)-(2) 5 N D ( 2 ) - 6 1 (8) 0 ND(7) Dow HML Michigan Tittabawassee River Saginaw River Saginaw River Mouth Saginaw Bay Flint River Cass River Shiawassee River 1979 13 12 12 18 6 5 6 NESTRICK & LAMPARSE1 (1979) 7 ND(3)-500 11 N D ( 3 ) - 3 1 0 12 5 ( 0 - 3 2 ( 2 ) 16 N D ( 4 ) - 2 9 ( 5 ) 5 ND(3)-11(2) 3 N D ( 5 )-106 1 ND(5)-9(9) NYS-DOH HHH Lake Ontario Lake Huron 1978-1980 1978-1980 10 7 10 6 ( 0 - 1 6 2 6 ND(2)-29 S M ITH et al. ( 1 9 8 0 HOE KHL Niagara River Lake Ontario Lake Erie Lake Huron Lake Superior 1980 1980 1980 1980 1980 23 52 3 18 6 0 ND(1 0 ) NATIONAL RESEARCH COUNCIL CANADA (1981) 16 H D ( 10)-19(103 0 ND(10) 0 ND(10) 0 ND(10) NYS-DOH HHH Detroit River Grand River C a n a g a g i q u e Cr. Lake Ontario Lake Erie Lake Superior Lake Michigan 1980 1980 1980 1978-1980 1979 1979 1979 3 2 6 66 17 1 5 0 ND(10) 0 ND(10) 0 ND(10) 45 ND(3)-I07 NRCC (1981) 5 ND(l)-2 1 1 (0 0 ND(2)-ND(5) 437 TABLE 14: Relationship of 2,3,7,8-TCDD Concentration To Fish Species SPECIES Channel Catfish Ca rp Bullhead Crappte Yellow Perch Lake Trout YEAR SAHPLEO 1976 1978 1979 1976 1978 1979 1976 1979 1976 1979 1978 1979 1980 LABORATORY Oow MDNR-EPA Dow Dow MDNR-EPA Dow Dow Dow Dow Dow MDNR-EPA Dow NYS-DOK NUMBER SAMPLES . 2 8 17 1 14 1 3 14 3 16 6 12 3 NO. APPA R E N T POSITIVE RESULTS 2 8 17 0 10 0 3 9 3 10 3 9 3 APPARENT TCDD CONC. RANGE. PPT 90-170 28<13)-69S 9(3)-500 ND(80) 20(7)-153 ND(5) 20(20)-140(40) 1(0-14(7) 40(20)-90(30) ND(3)-I18(14) ND(5)-20 N D ( 4 ) - 10(5) 41-72 MEAN APPARENT CONC. PPT 130 160 30 (av8 55 -- 70 9 80 10 13 5 42 Although not a ll the data generated by the EPA Dioxin Implementation Plan have been published, some of i t i s shown in Table 15. These data can e a s ily be m isinterpreted and require special explanation. The Dow data were taken on the fat cut from steaks. With a lim it of detection of 3 times noise only one sample gave a p o sitiv e sig n a l. The samples analyzed by EPA-Dow were a complete set of peritoneal beef fat samples taken in phase I o f the Dioxin Implementation Plan while those analyzed by EPA-WSU were a part o f the same set. Other a n a ly tica l colla bora tors were EPA and the U n ive rsity o f Utah. From the table i t appears that EPA-Dow found 9 p o sitiv e re su lts on 88 samples. However, most of these same samples were shown to be negative by other in ve stiga to rs with lower detection lim it s . Only the M issouri sample showing a con centration o f 60 parts per t r i l l i o n at a detection lim it o f 25 was proven p o sitiv e (EPA-WSU found 54(25) for example). S im ila r comments apply to the data generated by other co lla b o rators. So the table i s a good example o f fa lse p o sitiv e data re su ltin g from both the uncertainty of the methodology and interferences. TABLE 15: Search For 2,3,7,8-TCDD In Beef Fat LABORATORY ANALYTICAL SYSTEM SAMPLE SOURCE YEAR SAMPLED NUMBER OF SAMPLES NUMBER OF APPARENT POSITIVE RESULTS Dow MML/MMH Texas Oklahoma Missouri 1976 19 2 3 3 0 0 EPA-Dow (DIP) MML/HHH Texas Oklahoma Kansas Missouri 1976 20 5 (controls) 19 S (controls) 15 10 (controls) 14 0 0 2 0 1 1 5 EPA-WSU (DIP) MMH Texas/ 1976 Oklahoma/ Missouri/Kansas 28 8 HANGE OF APPARENT POSITIVE 2 , 3 , 7 ,8 - T C D D CONC. (PPT) REFERENCE 3(3)-6(4) ND(5) ND(5) CRUMMETT (1979) ND(6)-ND(23) NDO)-ND(IB) 9(5)-30(20) ND(5)-ND(20) 7(6) 6(4) 10(5)-60(25) ll(10)-54(25) TIERNAN (1981) Unknown Unknown 39 4 10(7)-33(8) 19510 438 TABLE 16: Search For 2,3,7,8-TCDD In Beef Liver LABORATORY EPA-Oov ANALYTICAL SYSTEM SAMPLE SOURCE MML Texas Kansas O k 1ahorna EPA-VSC (DIPI WMH Missouri Texas Kansas O k Iahorna Missouri YEAR SAMPLED NUMBER OF SAMPLES NUMBER OF APPARENT POSITIVE RESULTS 1976 20 5 (controls) 0 0 1976 10 ! (control) 0 0 1976 5 2 (controls) 0 0 1976 6 0 1976 37 3 APPARENT TCDD CONC. RANGE ND(2)*ND(100) ND(3)-ND(7) ND(2 J-ND6 ) ND(5) ND(2)*ND(25) ND(2)-ND(5) ND(2)-ND(IS) U(S)-40(8) REFERENCE CRUHMETT (1979) T1ERNAN (1981) TABLE 17: Search For 2,3,7,8-TCDD In Bovine M ilk LABORATORI' ANALYTICAL ,_SYSTEM. Dow M M I. SAMPLE SOURCE YEAR SAMPLED Oklahoma Arkansas Missouri Michigan 1976 (controls) N UM8 ER OF SAMPLES 10 7 8 2 NUMBER OF APPARENT POSITIVE RESULTS APPARENT TCDD CONC. RANGE 2 ND(l)-Kl) 2 N D ( I )1(1) 2 ND(l)-Kl) 1 ND(l)-Kl) REFERENCE M A H L E et al. (1977) Table 16 shows some of the re su lts on the EPA DIP beef liv e r sampling. Many o f the samples are from the same c a ttle as the beef fat previously discussed. The 3 apparent p o sitiv e re su lts obtained by WSU were not detected by other co lla borators at lower detection lim it s. Table 17 shows the small amount o f work that has been reported on bovine m ilk. The authors used 2.5 times noise as a detection lim it . At 3 times noise a ll re su lts would have been negative at a detection lim it o f 1 part per t r i l l i o n . This c le a rly illu s t r a t e s the importance o f having w ell-defined and accepted c r it e r ia fo r the determination of chemicals in environ mental samples. This point is also illu s t ra te d in Table 18. The p o sitiv e re su lts reported by Meselson would be negative at 3 times noise. The p o sitiv e re su lts being reported by Loomis are also rig h t at the detection lim it. One explanation is that the lim it o f detection has now been lowered by improved a n a ly tica l methodology to a level corresponding to that of the background it s e lf . TABLE 18: Search For 2,3,7,8-TCDD In Human M ilk LABORATORY Dow Harvard Harvard Dow EPA/UN EPA/WSU ANALYTICAL SYSTEM SAMPLE SOURCE MMH Texas MLH Texas Hi ssouri Oregon Massachusetts MLH Texas Kansas HML Washington MMH Wash. /Ore./ California Calif./Alaska MMH W a s h ./ O r e g o n Calif./Alaska YEAR SAMPLED 1974 1976 1976 1976 1976 1978 1978 1978 1978 1978 1978 NUMBER OF SAMPLES 6 6 6 7 6 9 6 46 72 31 7 APPARENT POSITIVE RESULTS CONCENTRATION RANGE OF SAMPLES. PPT REFERENCE 1)0 N D( -ND(6 ) SHADOFF et a l . (1977) J3 0 .6 ( 0 .6 ) - .S ( 1 .5) M E S E L S O N & O ' K E E F E ( 1 9 7 7 ) 0 NIMO. 6)-ND(3) 1 0 . 8 ( 0 . 7 ) - l .4(1.5) 0 N D ( 0 .6 ) - N 0 ( 1.6) 2 0.4(0.2)-0.6(0-3) CR U M H E T T et al. (1979) 1 I (0.6)- N D ( 1.1) 5 ND(0.5)-2(1) L O O M I S et al. (1980) )0 ND(0.2 -ND(6 ) GROSS (1980) 0 N D ( 0 . 1)-ND(3) 0 ND(3)-ND(6) TIERNAN (1981) 439 TABLE 19: Search For Dioxins Other Than 2 ,3,7,8- TCDD In Human M ilk DIOXIN ANALYTICAL LABORATORY SYSTEM SAMPLE SOURCE YEAR SAMPLED NUMBER OF SAMPLES APPARENT POSITIVE RESULTS CONCENTRATION RANGE. PPT REFERENCE ISO - TCDD* Dow HML Washington 1978 46 0 ND(0.S)-ND(0.7) LOOMIS et a l . (1980) HCDD 1 , 2 , 3 , 4 , 6 , 7 .9 - H 7 C D D 1 ,2 , 3 , 4 , 6 , 7 , fi-H7 C D D Dow Dow Dow KML HML HML Washington Washington Washington 1978 1978 1978 46 46 46 27 N D ( l )-4(3) ( )3 N D ( 1 ) - 2 1 )33 W O f 1 -- 7 ( 2 3 LOOMIS et a l . (1980) aL O O M I S e t 1 ( 1 9 8 0 ) LOOM I S et a . (1980) OCDD Dow HML Washington 1978 46 38 K D ( 6 ) - 4 I ( 6 ) LOOM I S et a . (1980) TABLE 20: Search For 2,3,7,8-TCDD In Adipose From Vietnam Veterans LABORATORY UNL UNL-EPA EPA ANAL. SYSTEM MNH HHH MHH SAMPLE SOURCE Heavily Exposed USAF Officers Other Exposed Controls Heavily Exposed USAF Officers Other Exposed Controls Heavily Exposed USAF Officers Other Exposed Controls NUMBER OF SAMPLES 3 2 5 5 1 1 3 2 3 2 5 5 NUMBER OF APPARENT POSITIVE RESULTS 2 2 3 3 1 1 2 2 3 2 2 4 RANGE OF APPARENT 2 , 3 . 7 ,8 - T C D D CONC. PPT HD(3)-99 4(3)-5<3) ND(2)-7(3) ND(4)-20 36 3 3-5 20-173 10-24 N D ( 5 )-18 ND(3)-46 REFERENCE GROSS (1980) TABLE 21: Search For 2 ,3 ,7 ,8-TCDD In Human Adipose From Rice Growing States LABORATORY UNL ANALYTICAL SYSTEM SAMPLE SOURCE YEAR SAMPLED MMH Mtss/Tenn/ Arkansas 1977 NUMBER OF SAMPLES 41 2i NUMBER OF APPARENT POSITIVE RESULTS 0 0 0 RANGE OF APPARENT 2.3,7.8-TCDD CONC. (PPT) ND ( 1)-ND(12) N D ( 17)-ND(19)* ND(I6 O)** REFERENCE Grots (1980) *Sample size too small ^'Recovery o n l y a b o u t 1 The Loomis study also included other selected polychlorinated dibenzodioxins - total isomers o f 2,3,7,8-tetrachlorodibenzo-p-dioxin, total hexachlorodibenzo-p-dioxins, the two hepta isomers, and octachlorodibenzo-p-dioxin. The re su lts in Table 19 are p o sitiv e in a large number o f cases but are generally close to the lim it o f detection - only two being greater than ten times noise. Table 20 on the determination of 2,3,7,8-TCDD in Vietnam veterans shows the type o f agreement co lla b o ra tive studies w ill produce using the most recently developed methodology. Table 21 suggests that persons liv in g in rice-grow ing areas are not as lik e ly to have detect able le v e ls of TCDD in th eir fat as those used fo r controls in the Vietnam veterans study. A summary o f these data is attempted in Table 22 in which the ACS gu id elines o f 3 times noise and 10 times noise for lim it o f detection and lim it o f quantitation is applied. Only 1 beef fa t sample gives a quantitative signal and no human or bovine m ilk. However, 84 o f 233 samples o f f is h contain quantitative sig n a ls for TCDD. 1951 440 TABLE 22: Summary of 2,3,7,8-TCDD Data From Biological Samples MATRIX TOTAL SAMPLES ANALYZED Beef fat 179 Beef Liver 51 Bovine Milk 27 Fish 233 Human Milk 193 Human Adipose (non-milicary) 49 NUMBER OF APPARENT POSITIVE RESULTS 24 0 7 146 12 4 NUMBER AT CONCENTRATION LEVEL >3 X NOISE >10 X NOISE 17 1 00 00 134 84 60 43 Chlorinated Dioxins In Chemical Products The determination of chlorinated dioxins in chemical products uses the same analytical systems of LC/GC/MS methodology as used for environmental samples. However, the detection limit is higher, usually in the part per billion range. Also, it has usually already been established that the dioxin being measured is present and it's merely a matter of deciding whether or not the level is above or below a regulatory limit or a level of concern. The proper development of such methods requires the cooperation of industrial and governmental scientists. An excellent example of collaboration between Agriculture Canada and Dow Chemical 1s illustrated by the data shown in Tables 23, 24, and 25. Although the methodology was considerably different, the results agree to a remarkable degree. This kind of cooperative effort must be supported and encouraged if reasonable regulations are to result. TABLE 23: Dioxins In 2,4-D Butoxyethyl Ester SAMPLE NUMBER 1 2 3 t. 5 29 30 31 APPARENT DCDO. PPB AGRICULTURE CANADA DOW 21 54 2l ND 1 21 ND 3 ND 1 ND 1 APPARENT T ^CDD. PPB AGRICULTURE CANADA DOW 01 44 21 ND 1 11 ND I ND 1 ND 1 4 41 TABLE 24: Dioxins In Formula 40 (DEA) SAM PLE NUM BER 6 7 8 9 37 APPARENT DCCD. A G R IC U L TU R E CANADA PPB DOW ND 4 21 I ND 1 ND S8 A PP A R E N T T -iC D D , PP B a g r ic u l t u r e CANADA DOW ND 3 21 ND ND ND ND 67 TABLE 25-: Dioxins In DMA 6 Unsequestered SAM PLE N L T lB E R 11 12 19 20 21 22 23 25 26 A G R IC U L TU R E CANADA ND ND 6 20 1 1 2 1 2 DOW 1 1 10 30 1 1 2 1 3 A G R IC U L TU R E CANADA ND ND 5 24 -ND ND ND 2 DOW 1 ND 8 38 -- ND 1 ND 2 This report has been concerned primarily with the use of LC/GC/MS systems for the quantitative determination of the chlorinated dibenzo-p-dioxins in environmental and biological samples and products. Other analytical methods which are at an early stage of development largely for screening purposes include radio immunoassay (McKinney et al. 1982), aryl hydrocarbon hydroxylase induction assay (Bradlaw and Casterline, 1979), cytosil-receptor assay (Hutzinger et al. 1981), and kertatization of mouse teratoma cell line XB (Knutson and Poland, 1980). SUMMARY AND CONCLUSIONS Many techniques are available to dissolve or extract samples containing chlorinated dioxins, to remove the bulk of the interferences, to separate homologues and isomers, and to measure isomers either specifically or as part of a homologue. These techniques can be combined in an almost infinite number of ways to yield procedures which appear to produce reliable data. Because these procedures are tedious and require considerable'time to carry out, true measures of the certainty of results are often neglected. Thus, validation and collaborative studies are seldom done to the satisfaction of the scientific community. However, the quality assurance programs now placed on much of the work are so rigorous that the certainty of the data is assured. In fact no other analysis known to me is subjected to such a complete and comprehensive control. Recent methods thus serve as models for other investi gations in trace analysis. Analytical scientists can now write protocols which are based on good estimates of the sensitivity and specificity of the total analytical scheme. This approach is much preferred over the use of "best available analytical methodology" as often specified by regulators. 19514 442 Of course more validation and collaborative studies are needed if regulation is indicated. However, these should not require that all participants use identical procedures but that the methodologies which are used are equivalent and are subjected to suitable quality control. Both intra- and inter-laboratory studies are required. However, regulations should be based on inter-laboratory studies. Moreover, any regulatory level should include allowance for the certainty of the method. This generally requires that regulation be attempted at levels no lower than the limit of quantitation. REFERENCES 1. ACS Committee on Environmental Improvement (1980). Data Acquisition and Data Quality Evaluation In Environmental Chemistry. Anal. Chem., 52, 2242. 2. Baughman, R. and M. Meselson (1973). An Analytical Method For Detecting TCDD (Dioxin): .Levels of TCDD In Samples From Vietnam. Environ. Health Persp., 5^ 27. 3. Blaser, W. W., R. A. Bredeweg, L. A. Shadoff, and R. H. Stehl (1975). Determination of Chlorinated Dibenzo-p-dioxins In Pentachlorophenol By Gas Chromatography/Mass Spectrometry. Anal. Chem., 4 8 , 984. 4. Bradlaw, J. A. and 0. L. Casterline (1979). Induction of Enzyme Activity in Cell Culture: A Rapid Screen For Detection of Planar Polychlorinated Organic Compounds. J. Assoc. Off. Anal. Chem.. 62, 904. 5. Bramley, W. C., J. A. G. Roach, J. A. Sphon, P. A. Dreifuss, D. Andrzejewski, R. A. Neiman, and D. Firestone (1981). Low-Resolution Multiple Ion Detection Gas Chromatographic - Mass Spectrometric Comparison of Six Extraction Cleanup Methods For Determining 2,3,7,8Tetrachlorodibenzo-p-dioxin in Fish. J. Aqri. Food Chem., 29, 1040. 6. Buser, H. R. (1975). Analysis of Polychlorinated Dibenzo-p-dioxins and Dibenzofurans In Chlorinated Phenols By Mass Fragmentography. J. Chromatoqr., 107, 295. 7. Buser, H. R. and C. Rappe (1980). High-Resolution Gas Chromatography of The 22 Tetrachlorodibenzo-p-dioxin Isomers. Anal Chem., 52, 2257. 8. Cavallaro, A., G. Bandi, G. Invernizzi, L. Luciani, E. Mongini, and G. Gorni (1982). Negative Ion Chemical Ionization MS As A Structure Tool In The Determination of Small Amounts of PCDD and PCDF. Chlorinated Dioxins and Related Compounds Impact on The Environment. 0. Hutzinger, R. W. Frei, E. Merian, and F. Pocchiari, ed. Perqamon Press, Oxford, pp. 55-65. 9. Chae, K., L. K. Cho, and J. D. McKinney (1977). J. Agri. Food Chem., 25, 1207. 10. Chess, E. K. and M. L. Gross (1980). Determination of Tetrachlorodioxins By Mass Spectrometric Metastable Decomposition Monitoring. Anal. Chem., 52, 2057. 11. Cochrane, W. P., J. Singh, W. Miles, and B. Wakeford (1981). Determination of Chlorinated Dibenzo-p-dioxin Contaminants in 2,4-0 Products By Gas Chromatography/Mass Spectrometric Techniques. J. Chromatoqr.. 217, 289. 12. Crummett, W. B. and J. K. Taylor (1982). Guidelines For Data Acquisition and Data Quality Evaluation In Environmental Chemistry. IUPAC Collaborative Interlaboratory Studies in Chemical Analysis. Edited by H. Egan and T. S. West, Perqamon Press, Oxford and New York. 13. Crummett, W. B. (1979). Fundamental Problems Related to Validation of Analytical Data Elaborated on The Example of TCD0. Toxic. Environ. Chem. Rev., 3, 61. 443 14. Crummett, W. B., R. H. Stehl, R. A. Hummel, and L. A. Shadoff (1979). A Search For Chlorinated Dioxins in The Environment. CIPAC Proceedings Symposium Papers. W. R. Bontoyan, ed., CIPAC Proceedings Symposium Series 1. CIPAC Publication, Hertfordshire, England. 15. Crummett, W. B. (1981). Analytical Methodology For The Determination of PCODs and PCDFs in Products and Environmental Samples - An Overview and Critique. Second International Symposium on Chlorinated Dioxins and Related Compounds, Arlington, VA. 16. Crummett, W. B., R. H. Stehl, R. A. Hummel, and L. A. Shadoff (1979). A Search For Chlorinated Dioxins in The Environment. CIPAC Proceedings Symposium Papers, W. R. Bontoyan, ed., Collaborative International Pesticides Analytical Council Publications, Cambridge, England. 17. Cutie, S. S. (1981). Recovery Efficiency of 2,3,7,8-Tetrachlorodibenzo-p-dioxin From Active Carbon and Other Particulates, Anal. Chim. Acta., 123, 25. 18. Dougherty. R. C. (1980). Environ, Health Persp.. 36, 103. 19. Eiceman, G. A., A. C. Viau, and F. W. Karasek (1980). Ultrasonic Extraction of Poly chlorinated Dibenzo-p-dioxins and Other Organic Compounds From Fly Ash From Municipal Incinerators. Anal Chem., 52 1492. 20. Erk, S., M. L. Taylor, and T. 0. Tiernan (1979). Determination of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Residues on Metal Surfaces. Chemosphere. 8, 7. 21. Fanelli, R., M. P. Bertoni, M. Bonfani, M. G. Castelli, C. Chiabrando, G. P. Martelli, M. A. Noe, A. Noseda, S. Garrattini, C. Binaghi, V. Marazza, F. Pezza, D. Pozzoli, and G. Cicognetti (1980). Bull. Environ. Contam. Toxicol., 24, 460. 22. Fukahara, K., M. Takeda, M. Uchiyama, and H. Tanabe (1975). Analytical Method For 2,3,7,8-Tetrachlorodibenzo-p-dioxins in Sea Foods. Eisei Kagaku, 21, 318. 23. Gross, M. L. (1980). Direct Testimony of Dr. Michael L. Gross. Before the Administra tion, United States Environmental Protection Agency. EPA Exhibit No. 223. FIFRA Docket Nos. 415 et a l . 24. Harless, R. L., E. 0. Oswald, M. K. Wilkinson, A. E. Dupuy, Jr., D. 0. McDaniel, and Han Tai (1980). Sample Preparation and Gas Chromatography/Mass Spectrometry Determina tion of 2,3,7,8-Tetrachlorodibenzo-p-dioxin. Anal. Chem., 52, 1239. 25. Harless, R. L. and R. G. Lewis (1982). Quantitative Determination of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Residues By Gas Chromatography/Mass Spectrometry. Chlorinated Dioxins and Related Compounds. Impact on The Environment. 0. Hutzinger, R. W. Frei, E. Merian, and F. Pocchiari, ed. Pergamon Press, Oxford, pp. 25-35. 26. Harless, R. L., E. 0. Oswald, R. G. Lewis, A. E. Dupuy, Jr., D. D. HcDaniel, and Han Tai (1982). Determination of 2,3,7,8-Tetrachlorodibenzo-p-dioxin in Fresh Water Fish. Chemosphere, 11, 193. 27. Hass, J. R., M. D. Friesen, D. J. Harvan, and C. E. Parker (1978). Determination of Polychlorinated Dibenzo-p-dioxins in Biological Samples by Negative Chemical Ionization Mass Spectrometry. Anal. Chem., 50, 1474. 28. Huckins, J. N., D. L. Stalling, and W. A. Smith (1978). Foam Charcoal Chromatography For Analysis of Polychlorinated Dibenzodioxins in Herbicide Orange. J. Assoc. Off. Anal. Chem., 61, 32. h | 19516 444 29. Hutzinger, 0., K. 01ie, J. W. A. Lustenhouwer, A. B. Okey, S. Bandiera, and S. Safe (1981). Polychlorinated Dibenzo-p-dioxins and Dibenzofurans: A Bioanalytical Approach. Chemosphere, 10, 19. 30. Jensen, 0. J., R. A. Humnel, N. H. Mahle, C. W. Kocher, and H. S. Higgins (1981). A Residue Study on Beef Cattle Consuming 2,3,7,8-Tetrachlorodibenzo-p-dioxin. J. Aqric. Food Chem., 29, 265. 31. Knutson, J. C. and A. Poland (1980). Keratization of Mouse Teratoma Cell XB Produced by 2.3.7.8- Tetrachlorodibenzo-p-dioxin: An In-Vitro Model of Toxicity. Cel 1 , 2 2 , 27. 32. Lamparski, L. L. and T. J. Nestrick (1980). Determination of Tetra-, Hexa-, Hepta-, and Octachlorodibenzo-p-dioxin Isomers In Particulate Samples at Parts Per Trillion Levels. Anal. Chem., 52, 2045. 33. Lamparski, L. L. and T. J. Nestrick (1981). Synthesis and Identification of The 10 Hexachlorodibenzo-dioxin Isomers by High Performance Liquid and Packed Column Gas Chromatography. Chemosphere, 10, 3. 34. Lamparski, L. L., N. H. Mahle, and L. A. Shadoff (1978). Determination of Pentachlorophenol, Hexachlorodibenzo-p-dioxin, and 0ctachlorodibenzo-p-diox1n in Bovine Milk. J. Aqri. Food Chem., 26, 1113. 35. Lamparski, L. L., T. J. Nestrick, and R. H. Stehl (1979). Determination of Part-PerTrillion Concentrations of 2,3,7,8-Tetrachlorodibenzo-p-dioxin in Fish. Anal. Chem., 51, 1453. 36. Langhorst, M. L. and L. A. Shadoff (1980). Determination of Parts-Per-Trillion Concen trations of Tetra-, Hexa-, Hepta-, and Octa-Chlorodibenzo-p-dioxins in Human Milk Samples. Anal. Chem., 52, 2037. 37. Liberti, A., D. Brocco, A. Cecinato, and M. Possanzini (1981). Determination of Poly chlorinated Dibenzo-p-dioxins and Dibenzofurans in The Emission From Urban Incinerators. Mikrochim. Acta Wein, 1, 271. 38. Loomis, T. A., L. A. Shadoff, and M. L. Langhorst (1980). Chlorinated Dioxins in Human Milk and Their Toxicological Significance. Submitted for publication, Fund. and Ap pl. Toxicol . 39. Mahle, N. H. and L. A. Shadoff (1982). The Mass Spectrometry of Chlorinated Dibenzo-pdioxins. Biomed. Mass Spectrom., , 45. 40. Mahle, N. H., H. S. Higgins, and M. E. Getzendaner (1977). Search For The Presence of 2.3.7.8- Tetrachlorodibenzo-p-dioxin in Bovine Milk. Bull. Environ, Contam, Tox., 18, 123. 41. McKinney, J., P. Albro, M. Luster, B. Corbett, J. Schroeder, and L. Lawson (1982). Development and Reliability of a Radioimmunoassay For 2,3,7,8-Tetrachlorodibenzo-p-dioxin. Chlorinated Dioxins and Related Compounds Impact on The Environment, 0. Hutzinger, R. W. Frei, E. Merian, and F. Pocchiari, ed., Perqamon Press, Oxford, England. 42. McLeod, H. (Chairman), Panel of Associate Committee on Scientific Criteria For Environ mental Quality (1981). Polychlorinated Dibenzo-p-dioxins: Limitations to The Current Analytical Techniques. National Research Council of Canada, NRCC No. 18576. 43. Meselson, M. and P. W. O'Keefe (1977). Human Milk Monitoring: Preliminary Results For Twenty-One Samples: 12/15/76. Letter to Congressman Jim Weaver. 445 44. Mieure, J. P., 0. Hicks, R. G. Kaley, and P. R. Michael (1977). Determination of Trace Amounts of Chlorodibenzo-p-dioxins and Chlorodibenzofurans in Technical Grade Pentachlorophenol. J. Chromatogr. Sci., 15, 275. 45. Mitchum, R. K., G. F. Moler, and W. A._ Korfmacher (1980). Combined Capillary Gas Chromatography/Atmospheric Pressure Negative Chemical Ionization/Mass Spectrometry For The Determination of 2 ,3 ,7 ,8 -Tetrachlorodibenzo-p-dioxin in Tissue. Anal. Chem., 52, 2278. 46. National Research Council Canada (1981). Polychlorinated Dibenzo-p-dioxins: Criteria For Their Effects on Man and His Environment. NRCC No. 18574. Publications, NRCC/CNRC, Ottawa, Canada K1A 0R6. 47. Nestrick, T. J . , L. L. Lamparski, and R. H. Stehl (1979). Synthesis and Identification of the 22 Tetrachlorodibenzo-p-dioxin Isomers by High Performance Liquid Chromatography and Gas Chromatography. Anal. Chem., 51, 2273. 48. Nestrick, T. J., L. L. Lamparski, and D. I. Townsend (1980). Identification of Tetrachlorodibenzo-p-dioxin Isomers at The 1-ng Level by Photolytic Degradation and Pattern Recognition Techniques. Anal Chem., 5 2 , 1862. 49. Nestrick, T. J., L. L. Lamparski (1982). Isomer-Specific Determination of Chlorinated Dioxins for Assessment of Formation and Potential Environmental Emission From Wood Combustion. Anal. Chem., in press. 50. Nieman, R., W. Brumley, D. Firestone, and J. Sphan (1981). Multi-Dimensional High Resolution Chromatographic Technique Applied to PPT Level Measurements of 2,3,7,8-TCDD in Fish. International Symposium on Chlorinated Dioxins and Related Compounds. Arlington, Virginia, October 25-29. 51. Norstrom, R. J . , D. J. Hallet, M. Simon, and M. J. Mulvihill (1982). Analysis of Great Lakes Herring Gull Eggs For The Tetrachlorodibenzo-p-dioxins. Chlorinated Dioxins and Related Compounds. Impact on The Environment. 0. Hutzlnger, R. W. Frei, E. Merian, and F. Pocchiari, ed., Perqamon Press, Oxford, pp. 173-181 52. Oswald, E. 0. (1978). Summary of Results For Analyses of Samples of Fish From MichiganEPA Region V - for 2,3,7,8-Tetrachlorodibenzo-paradioxin (TCDD). 20 December letter to Mr. Karl . Bremmer, Toxic Substances Coordinator, U.S. EPA, Chicago, Illinois 60604. 53. Pfeiffer, C. D., T. J. Nestrick, and C. W. Kocher (1980). Anal. Chem., 52, 2328. 54. Phillipson, D. W. and B. J. Puma (1980). Identification of Chlorinated Methoxyblphenyls as Contaminants in Fish and as Potential Interferences in The Determination of Chlorin ated Dibenzo-p-dioxins. Anal. Chem., 52, 2328. 55. Ryan, J., J. C. Pilon (1980). High Performance Liquid Chromatography in The Analysis of Chlorinated Dibenzodloxins and Dibenzofurans in Chicken Liver and Wood Shaving Samples. J. Chromatogr.. 197, 171. 56. Shadoff, L. A., R. A. Hummel, L. L. Lamparski, and J. H. Davidson (1977). A Search for 2,3,/,8-Tetrachlorodibenzo-p-dioxin (TCDD) in an Environment Exposed Annually to 2,4,5Trichlorophenoxy-acetlc Acid Ester (2,4,5-T) Herbicides. Bull. Envir. Contam. Tox., 18, 478. 57. Smith, R. M., D. R. Hilker, P. W. O'Keefe, K. M. Aldons, C. M. Meyer, S. N. Kumar, and B. M. Jelus-Tyror (1981). Determination of Tetrachlorodibenzo-p-dioxins and Tetrachlorodibenzofurans in Environmental Samples by High Performance Liquid Chromatography, Capillary Gas Chromatography and High Resolution Mass Spectrometry. International Symposium on Chlorinated Dioxins and Related Compounds. Arlington, Virginia, Oct. 25-29. 446 58. Tiernan, T. 0., M. L. Taylor, J. G. Solch, G. F. Van Ness, and J. H. Garrett (1981). Quantitation of Parts-Per-Trll1ion Levels of 2,3,7,8-Tetrachlorod1benzo-p-dioxin 1n EPAFurnished Extracts of Biological Materials. Final Report. Cooperative Agreement No. CR8068646-01. Josephine Huang, EPA Project Officer, Office of Research and Development, Washington, D.C. 20460. 59. Tosine, H., 0. Smillie, and G. A. V. Rees (1981). Comparative Monitoring and Analytical Methodology For Polychlorinated Dibenzo-p-dioxins in Fish. International Symposium on Chlorinated Dioxins and Related Compounds. Arlington, Virginia, October 25-29, 1981. 60. Tou, J. C., D. Zakett, and V. Caldecourt (1982). Applications of MS/MS to Chemical Problems. Tandem Mass Spectrometry (MS/MS), F. W. McLafferty, editor. To be published by John Wiley and Sons, New York. 61. Van Ness, G. F., J. G. Solch, M. L. Taylor, and T. 0. Tiernan (1980). Tetrachlorodibenzop-dioxins in Chemical Wastes, Aqueous Effluents, and Soils. Chemosphere, 9, 553. 62. Zabik, M. E. and M. J. Zabik (1980). Bull. Environ. Contam. Toxicol., 24, 344. WEDNESDAY, 19 OCTOBER 8:00 [REGISTRATION AND COFFEE] WELCOME AND INTRODUCTION 9:00 WELCOME Joshua Lederberg, PhD (President, The Rockefeller University) 9:15 INTRODUCTION TO THE SYMPOSIUM William Lowrance, PhD (Senior Fellow and Director, Life Sciences and Public Policy Program, The Rockefeller University) 9:30 SPECTRUM OF METABOLIC EFFECTS FOUND IN ANIMAL EXPERIMENTS i Robert Neal, PhD (President, Chemical Industry Institute of Toxicology) 10:00 GENERAL DISCUSSION CHEMICAL ANALYSIS AND EXPOSURE ESTIMATION J 10:15 POSSIBILITIES, LIMITATIONS, AND PITFALLS OF CHEMICAL ANALYSIS T h o m a s O. T i e r n a n , PhD ( P r o f e s s o r of C h e m i s t r y and Director, Brehm Laboratory, Wright State University) y1 0 : 4 5 U.S. A N D C A N A D I A N A N A L Y S E S OF H U M A N T I S S U E S ' Alvin L. Y o u n g , PhD, Lt. Col. U S A F ( S p e c i a l A s s i s t a n t for Environmental Sciences, Agent Orange Projects Office, U.S. Veterans Administration) '11:00 SWEDISH ANALYSES OF HUMAN TISSUES Christopher Rappe, PhD (Professor and Chairman, Department of Organic Chemistry, University of Umea) 19520 11:15 PANEL DISCUSSION OF ANALYTIC AND EXPOSURE-ESTIMATION ISSUES Charles Nauman, PhD, MPH (Environmental Scientist, Office of Health and Environmental Assessment, U.S. Environmental Protection Agency) M i c h a e l L. T a y l o r , P h D ( A s s o c i a t e P r o f e s s o r of Pharmacology/Toxicology, Wright State University) Helle Tosine, PhD (Advanced Instrumentation Supervisor, Pesticides Section, Laboratory Services Branch, Canadian Ministry of the Environment) * M a r y S. Wo l f f , Ph D ( A s s i s t a n t P r o f e s s o r of C h e m i s t r y , M t . Sinai School of Medicine) 11:40 GENERAL DISCUSSION 12:00 [LUNCH BREAK] CARCINOGENICITY AND CANCER RISK ASSESSMENT 1:30 SUMMARY AND CRITIQUE OF RODENT CARCINOGENICITY ASSAYS Richard J. Kociba, PhD (Senior A s s o c i a t e S c i entist, Toxicology Research Laboratory, Dow Chemical U.S.A.) 2:10 RISK ASSESSMENT PROJECTIONS BASED ON RODENT CARCINOGENICITY ASSAYS 'John V a n Ryzin, PhD (Professor of Bios t a t i s t i c s , School of Public Health, Columbia University) [reporting c a l c u l a t i o n s c o n d u c t e d j o i n t l y with David G. Hoel and C h r i s t o p h e r J. P o r t i e r o f t h e U.S. N a t i o n a l I n s t i t u t e of Environmental Health Sciences] 2:30 COMMENTS ON RISK ASSESSMENT MODELS N a n c y K. Kim, P h D ( D i r e c t o r , B u r e a u o f T o x i c S u b s t a n c e s Assessment, New York State Department of Health) 2:45 A N A P P R O A C H TO RIS K A S S E S S M E N T FOR DIOXINS IN SOIL R e n a t e D. K i m b r o u g h , M D ( M e d i c a l Offi c e r , C e n t e r for Environmental Health, U.S. Centers for Disease Control) 3:15 [COFFEE BREAK] 3:1S DIOXINS AS CARCINOGENIC PROMOTERS I. B e r n a r d W e i n s t e i n , M D ( P r o f e s s o r o f M e d i c i n e a n d Director, Division of Environmental Sciences, Columbia University College of Physicians and Surgeons) 3:40 DIOXINS AS ANTICARCINOGENS John Di Giovanni, PhD (Assistant Professor of Biochemistry, Systems Cancer Center, University of Texas) 3:50 GENERAL DISCUSSION HUMAN SOFT-TISSUE SARCOMA ISSUE 4:HQ PATHOLOGICAL DEFINITION OF SOFT-TISSUE SARCOMAS S t e v e n I. H a j d u , M D ( A t t e n d i n g P a t h o l o g i s t a n d C h i e f , Pathology Service, Memorial Sloan-Kettering Cancer Center) ^4:20 REINTERPRETATION OF RECENT U.S. OCCUPATIONAL SOFT-TISSUE SARCOMA CASES / / '4:35 Marilyn Fingerhut, PhD (Epidemiologist, Division of Surveillance, Hazard Evaluations, and Field Studies, U . S . N a t i o n a l -IrLSti t u t e f o r O o c u n a t i a n a l . S a f e t v a n d -flpflJJ-h) DIOXINS AND SOFT-TISSUE SARCOMAS R a l p h R. Cook, MD, M P H ( D i r e c t o r of E p i d e m i o l o g y , D o w Chemical U.S.A.) 4:55 GENERAL DISCUSSION 5:30- [COCKTAIL RECEPTION FOR ALL PARTICIPANTS AND OBSERVERS] 7:00 THURSDAY, 20 OCTOBER 8:00 [COFFEE] IMMUNOTOXICITY 9:00 IM M U N O L O G I C A L O B S E R V A T I O N S IN ANIMALS J a c k H. Dean, P h D (Head, D e p a r t m e n t of C e l l B i o l o g y , Chemical Industry Institute of Toxicology)' 9:15 POSSIBLE HUMAN IMMUNOTOXIC EFFECTS A l b e r t E. M u n s o n , Ph D ( A s s o c i a t e P r o f e s s o r of Pharmacology and Toxicology, Medical College of Virginia, Virginia Commonwealth University) 9:30 RECENT CLINICAL IMMUNOLOGICAL OBSERVATIONS J. G e o r g e B e k e s i , M D ( P r o f e s s o r o f M e d i c i n e and Neoplastic Disease; and Director, Environmental Immunobiology Laboratory, M t . Sinai School of Medicine) 9:45 GENERAL DISCUSSION REPRODUCTIVE TOXICITY 10:00 OVERVIEW OF REPRODUCTIVE EFFECTS Donald Mattison, MD (Medical Officer, Pregnancy Research Branch, U.S. National Institute for Child Health and Human Development) 10:15 TERATOGENICITY AND FETOTOXICITY W i l b u r P. M c N u l t y , M D (Head, D i v i s i o n o f P r i m a t e Medicine, Oregon Regional Primate Research Center) 10:30 M O N I T O R I N G F O R R E P R O D U C T I V E EFFECTS IN P O P U L A T I O N S AT RISK Zena A. Stein, MB, BCh ( P r o f e s s o r of E p i d e m i o l o g y , S chool of Public Health, Columbia University; and Director, Epidemiology of Developmental Brain Disorders, New York State Psychiatric Institute) 10:45 GENERAL DISCUSSION BROAD METABOLIC TOXICITY 11:00 ALTERATION OF CELL MEMBRANE FUNCTION, CAUSING WIDESPREAD METABOLIC CHANGES Fumio Matsumura, PhD (Director, Pesticide Research Center, Michigan State University) 11:20 THYROTOXIC EFFECTS Carl Potter, PhD (Postdoctoral Trainee, Environmental Toxicology Center, School of Pharmacy, University of Wisconsin-- Madison) 11:30 H O R M O N A L C H A N G E S A N D " W A S T I N G " O B S E R V E D IN M O D E R A T E L Y - H I G H DOSE ANIMAL EXPERIMENTS Richard Peterson, PhD (Associate Professor of Pharmacology and Toxicology, School of Pharmacy, University of Wisconsin-- Madison) 1 1 :45 SOME O B S E R V A T I O N S ON Y U S H O D I S E A S E Masanori Kuratsune, MD (Professor of Public Health, Kyushu University) 12:00 GENERAL DISCUSSION [LUNCH BREAK] 2:00 PANEL TO COMMENT ON ASSESSMENT ISSUES [tentative] P a u l F. D e i s l e r , J r . , P h D ( V i c e P r e s i d e n t f o r H e a l t h , Safety, and Environment, Shell Chemical Company) P e r r y J. G e h r i n g , P h D (Vice P r e s i d e n t for A g r i c u l t u r a l Products R&D; and Director of Health and Environmental Science U.S.A., Dow Chemical Company) iM i c h a e l G o u g h , P h D ( S e n i o r A n a l y s t , O f f i c e o f T e c h n o l o g y ' Assessment, U.S. Congress) R o b e r t Harris, PhD (Co-Director, Hazardous Waste and Research Division, Center for Energy and Environmental Studies, Princeton University) Ellen Silbergeld, PhD (Chief Scientist, Environmental Defense Fund) & SUMMARIES OF CURRENT ACTIVITIES 2:45 THE NIOSH OCCUPATIONAL DIOXIN REGISTRY Marilyn Fingerhut, PhD (Epidemiologist, Division of Surveillance, Hazard Evaluations, and Field Studies, U.S. National Institute for Occupational Safety and Health) 3:00 PROPOSED FOOD AND DRUG ADMINISTRATION APPROACH TO TOLERANCE-SETTING FOR DIOXINS IN FOODS Robert Scheuplein, PhD (Associate Director for Toxicological Sciences, U.S. Food and Drug Administration) 3:15 CURRENT ENVIRONMENTAL PROTECTION AGENCY ACTIVITIES D o n a l d G. B a r n e s , P h D ( S c i e n c e A d v i s o r to the A s s i s t a n t Administrator, Office of Pesticides and Toxic Substances, U.S. Environmental Protection Agency) 3:45 PANEL TO COMMENT ON BIOCHEMICAL AND CLINICAL ISSUES [t e n t a t i v e ] Attallah Kappas, MD (Professor and Head, L a b o r a t o r y of Metabolism-- Pharmacology; and Physician-in-Chief, The Rockefeller University) R o b e r t W. Miller, MD (Chief, C l i n i c a l E p i d e m i o l o g y Branch, U.S. National Cancer Institute) Ruth Lilis, MD (Associate Professor of Occupational and Environmental Medicine, Mt. Sinai School of Medicine) J a m e s P. W h i t l o c k , J r., M D ( A s s o c i a t e P r o f e s s o r of Pharmacology, Stanford University School of Medicine) 4:15 PANEL TO COMMENT ON RESEARCH STRATEGY AND AGENDA [tentative] C h a r l e s M. B e n b r o o k , P h D (Staff D i r e c t o r , S u b c o m m i t t e e Department Operations, Research, and Foreign Agriculture, of the Committee on Agriculture, U.S. House of Representatives) on A n t h o n y Cerami, PhD (Professor and Head, Laboratory of Medical Biochemistry, The Rockefeller University) Glenn Paulson, PhD (Vice President for Science, National Audubon Society) Christoffer Rappe, PhD (Professor and Chairman, Department of Organic Chemistry, University of Umea) PARTICIPANTS William C. Agosta, PhD Professor of Organic Chemistry The Rockefeller University 1230 York Avenue New York, NY 10021 Karim Ahmed, PhD Director Natural Resources Defense Council 122 East 42nd Street New York, NY 10168 Ronald Altman, HD Assistant Commissioner Division of Epidemiology and Disease Control New Jersey Department of Health John Fitch Plaza CN360, Room 70S Trenton, NJ 0862S Karl E. Anderson, HD Associate Professor of Hetabolism-Pharmacology The Rockefeller University 1230 York Avenue New York, NY 10021 Susan G. Austin, ScD Corporate Director of Epidemiology Union Carbide Corporation Section P-2586 Old Ridgebury Road Danbury, CT 06817 Donald G. Barnes, PhD Science Advisor to the Assistant Administrator Office of Pesticides and Toxic Substances (TS-78) u.S. Environmental Protection Agency 40) H Street, SW Washington, DC 20460 J. George Bekesi, HD Professor of Hedicine and Neoplastic Disease, and Director, Environmental Immunological Laboratory Ht. Sinai School of Hedicine 10 East 102nd Street, Room 336 New York, NY 10029 Charles H. Benbrook, PhD Staff Director Subcommittee on Department Operations Research, and Foreign Agriculture Committee on Agriculture U.S. House of Representatives Longworth House Office Building, Room 1301 Washington, DC 20515 Brendan Birmingham, PhD Standards Coordinator Hazardous Contaminants and Standards Branch Ministry of the Environment province of Ontario 135 St. Clair Avenue West Toronto, Ontario N4V 1P5 CANADA Etcyl H. Blair, PhD Vice President Director, Health and Environmental Sciences The Dow Chemical Company 2020 Dow Center Midland, Michigan 48640 John Blodgett Head, Environmental Protection Section Environmental, Energy, and Natural Resources Division Congressional Research Service Library of Congress Washington, DC 20540 Jon Blyth Program Officer Charles Stewart Mott Foundation 1200 Mott Foundation Building Flint, MI 48502 Angela Boggs, MS Research Assistant Department of Health Policy and Management Harvard School of Public Health 677 Huntington Avenue Boston, MA 02115 CD cn i' j cn Thomas A. Burke Director Office of Science and Research New Jersey Department of Environmental Protection 190 Nest State Street Trenton, NJ 08625 Robert Cardeil, PhD Professor of Anatomy and Cell Biology (Hail Location 521) College of Medicine University of Cincinnati Cincinnati, OH 45267 D. Martin Carter, MD Professor and Head Laboratory of Investigative Dermatology The Rockefeller University 1230 York Avenue New York, NY 10021 Anthony Cerami, PhD Professor and Head Laboratory of Medical Biochemistry The Rockefeller University 1230 York Avenue New York, NY 10021 Joel Cohen, PhD, DrPH Professor of Populations The Rockefeller University 1230 York Avenue New York, NY 10021 Jonathan Cole, PhD Professor of Sociology, and Director of the Center for the Social Sciences Columbia University New York, NY 10027 Ralph R. Cook, MD, MPH Director of Epidemiology U.S. Medical, Health and Environmental Sciences 1803 Building Dow Chemical U.S.A. Midland, MI 48640 Susan Daum, MD Occupational Physician 16 East 96th Street New York, NY 10028 Devra Lee Davis, PhD Staff Director Board on Toxicology National Research Council 2101 Constitution Avenue, NW Washington, DC 20418 Jack H. Dean, PhD Head, Department of Cell Biology Chemical Industry Institute of Toxicology P.O. Box 12137 Research Triangle Park, NC 27709 Paul F. Deisler, PhO Vice President for Health, Safety, and Environment Shell Chemical Company One Shell Plaza P.O. Box 2463 Houston, TX 77001 William Dennlgan, MD Assistant Professor of Dermatology University of Missouri School of Medicine Columbia, MO 65212 John Di Giovanni, PhD Assistant Professor of Biochemistry Systems Cancer Center Science Park University of Texas Smithville, TX 78957 George Eadon, PhD Assistant Professor Division of Environmental Sciences Center for Laboratories and Research New York State Department of Health Albany, NY 12201 David L. Eaton, PhD Assistant Professor Department of Environmental Health, SC-34 School of Public Health and Community Medicine University of Washington Seattle, WA 98195 Marilyn Fingerhut, PhD Epidemiologist Division of Surveillance, Hazard Evaluations and Field Studies U, S. National Institute for Occupational Safety and Health 4676 Columbia Parkway Cincinnati, OH 45226 A. Betty Fischmann, MBBS, MRCPEd Chief of Dermatology Washington Veterans Administration Medical Center 50 Irving Street, NW Washington, DC 20422 Marvin Friedman, PhD Professor of Biological Sciences Hunter College 695 Park Avenue New York, NY 10021 Perry J. Gehring, PhD Vice President for Agricultural Products R&D; and Director of Health & Environmental Science U.S.A. The Dow Chemical Company 2030 Dow Center Midland, MI 48640 John Gierthy', PhD Research Scientist Center for Laboratories and Research New York State Department of Health Albany, NY 12202 Michael Gochfeld, MD Acting Chairman Environmental and Community Medicine Department University of Medicine and Dentistry of New Jersey-- Rutgers Piscataway, NJ 08854 Michael Gough, PhD Senior Analyst Health Program Office of Technology Assessment U.S. Congress Washington, DC 20510 Donald Grant, PhD Toxicologist Toxicological Evaluation Division Health and Welfare Canada HPB Building Tunney's Pasture Ottawa, Ontario K1A 012 CANADA Joseph H. Graziano, PhD Associate Professor of Parmacology Division of Pediatric Hematology- Oncology College of Physicians 6 Surgeons of Columbia University 630 West 168th Street New York, NY 10032 Laura Green, PhD Research Associate Department of Helath Policy and Management Harvard School of Public Health 677 Huntington Avenue Boston, MA 02115 CD cn to <1 William F. Greenlee, PhD Scientist, Department of Cell Biology Chemical Industry Institute of Tox icology P.0. Box 12137 Research Triangle Park, NC 27709 Steven I. Hajdu, HD Attending Pathologist and Chief, Cytology Service Memorial Sloan-Kettering Cancer Center 1275 York Avenue New York, NY 10021 William Halperin, MD, HPH Chief, Industry Wide Studies Branch U.S. National Institute for Occupational Safety and Health 4676 Columbia Parkway Cincinnati, OH 45226 Douglas Harding Ontario Ministry of Labor 400 University Avenue Toronto, Ontario M7A 1T7 CANADA Robert Harris, PhD Co-Director Hazardous Waste and Research Division Center for Energy and Environmental Studies School of Engineering/Applied Science Princeton University Princeton, NJ 08544 Maureen Hatch, PhD, MPH Assistant Professor of Epidemiology Columbia University College of Physicians and Surgeons 600 West 168th Street New York, NY 10032 Shelly Hearne, BS Research Assistant Natural Resources Defense Council 122 East 42nd Street New York, NY 10168 C.A. Heijden Head, Laboratory of Carcinogenicity and Mutagenicity National Institute of Public Health P.O. Box 1 3720BA Bilthoven THE NETHERLANDS John Hesse, MD Acting Chief Center for Environmental Health Sciences Michigan Department of Public Health P.O. Box 30035 Lansing, MI 48909 Joseph Highland, PhD Co-Director Hazardous Waste and Research Division Center for Energy and Environmental Studies School of Engineering/Applied Science Princeton University Princeton, NJ 08544 Donald Hillman, MD Corporate Medical Director Diamond Shamrock Corporation 717 N. Harwood Dallas, TX 75201 /Laurie Hudson Doctoral Candidate Department of Toxicology Harvard School of .Public Health 665 Huntington Avenue Boston, MA 02115 Peter Kahn, PhD Associate Professor of Biochemistry Department of Biochemistry 328 Litman Hall Rutgers University New Brunswick, NJ 08903 Emil Thomas Kaiser, PhD Professor and Head Laboratory of Bioorganic Chemistry and Biochemistry The Rockefeller University 1230 York Avenue New York, NY 10021 Lawrence Kaminsky, PhD Chief, Biochemical and Genetic Toxicology Laboratory Center for Laboratories and Research New York State Department of Health Albany, NY 12202 Han Kang, DrPH Chief Research Section Agent Orange Projects Office Department of Medicine and Surgery U.S. Veterans Administration Washington, DC 20420 Attallah Kappas, MD ' Physician-in-Chief, and Head Laboratory of Metabolism- Pharmacology The Rockefeller University 1230 York Avenue New York, NY 10021 Nathan Karch, PhD President Karch 4 Associates 7713 14th Street, NW Washington, DC 20012 Nancy K, Kim, PhD Director, Bureau of Toxic Substances Assessment New York State Department of Health Empire State Plaza Tower Building, 14th Floor Albany, NY 12237 Renate D. Kimbrough, MD Medical Officer Center for Environmental Health .U.S. Centers for Disease Control 1600 Clifton Road, NE Atlanta, GA 30333 Florence K. Kinoshita, MD Senior Toxicologist Medical Department Hercules, Inc. Hercules Plaza Wilmington, DE 19894 Judith Klotz, DrPH Postdoctoral Fellow Memorial Sloan-Kettering Cancer Center; and Science Associate Natural Resources Defense Council 122 East 42nd Street New York, NY 10168 Richard J. Kociba, PhD Senior Associate Scientist Toxicology Research Laboratory 1803 Building Dow Chemical U.S.A. Midland, MI 48640 Masanori Kuratsune, MD Professor of Public Health Faculty of Medicine Kyushu University 1-1, 3 Chme, Maidashi Higashi-ku, Fukuoka-shi 812 JAPAN Robert W. Leader, DVM Acting Director Center for Environmental Tox icology East Fee Hall, Room A-136 Michigan State University East Lansing, Michigan 48824 Joshua Lederberg, PhD Pres ident The Rockefeller University 1230 York Avenue New York, NY 10021 CO ai V>' uo Hilton C. Lewinsohn, MD Assistant Corporate Medical Director Union Carbide Corporation Section P-2590 Old Ridgebury Road Danbury, CT 06817 Steven C. Lewis, PhD Toxicology Associate Medicine and Environmental Health Department Exxon Corporation P.O. Box 235 East Millstone, NJ 08873 Ruth Lilis, MD Associate Professor of Occupational and Environmental Medicine Environmental Sciences Laboratory Mt. Sinai School of Medicine 10 East 102 Street New York, NY 10029 William Lower, PhD Group Leader Environmental Trace Substances Research Center University of Missouri Rural Route 3 Columbia, MO 65201 William W. Lowrance, PhD Senior Fellow and Director Life Sciences and Public Policy Program The Rockefeller University 1230 York Avenue New York, NY 10021 Paul Lundy, MS Toxic Substances Associate Toxic Substances Program National Audubon Society 950 Third Avenue New York, NY 10022 Karl Mackerer, PhD Manager of Biochemical Toxicology Mobil Oil Corporation P.O. Box 1029 Princeton, NJ 08540 Bonnie M. Marlnelli Senior Attorney-Advisor Board of Veterans Appeals U.S. Veterans Administration Washington, D.C. 20420 Fumio Matsumura, PhD Director Pesticide Research Center Michigan State University East Lansing, MI 48824 Donald R. Mattison, MD Medical Officer Pregnancy Research Branch National Institute for Child Health and Human Development U.S. National Institutes of Health Building 10, ACRF 86313 Bethesda, MD 20205 Donald J. McGraw, MD, MPH Corporate Medical Director Koppers Company, Inc. Pittsburgh, PA 15219 Wilbur P. McNulty, MD Head Division of Primate Medicine Oregon Regional Primate Research Center 505 NW 185th Avenue Beaverton, OR 97006 Samuel Milham, Jr., MD Head, Epidemiology Section Population Study Unit' Washington State Department of Social and Health Services 1212 Southeast Quince Mail Stop ET13 Olympia, WA 98405 Robert W. Hiller, HD Chief, Clinical Epidemiology Branch National Cancer Institute A-S21 Landow Building National Institutes of Health Bethesda, MD 20205 Sonya Mirsky, MS Librarian The Rockefeller University 1230 York Avenue New York, NY 10021 Marion Moses, MD 500 W University Parkway Baltimore, MD 21210 Warren Muir, PhD Vice President Idea Tech Associates, Inc. One Skyline Place, Suite 1313 5205 Leesburg Pike Falls Church, VA 22041 Devdas Mukerjee, PhD Senior Environmental Health Scientist Environmental Criteria and Assessment Office Environmental Research Center u.S. Environmental Protection Agency 26 W. St. Clair Street Cincinnati, OH 45268 Albert E. Munson, PhD Associate Professor of Pharmacology and Toxicology School of Basic Sciences Medical College of Virginia Virginia Commonwealth University Box 110, MCV Station Richmond, VA 23298 F. Jay Murray, PhD Director of Environmental Affairs Syntex (U.S.A) Inc. 3401 Hillview Avenue Palo Alto, CA 94304 Charles Nauman, PhD, MPH Environmental Scientist Office of Health and Environmental Assessment U.S. Environmental Protection Agency 401 M Street, SW Washington, DC 20460 Robert A. Neal, PhD President Chemical Industry Institute of Toxicology P.O. Box 12137 Research Triangle Park, NC 27709 Herbert Northrop, MD Director Department of Occupational Medicine Stauffer Chemical Company Westport, CT 06881 Kathleen O'Halloran, MPH Policy Analyst New York City Department of Environmental Protection 2448 Municipal Building New York, NY 10007 Patrick O'Keefe, PhD Director, Dioxin Analytical Laboratory Center for Laboratories and Research New York State Department of Health Albany, NY 12202 William Parkin, DVM, DrPH State Epidemiologist Division of Epidemiology and Disease Control New Jersey Department of Health John Fitch Plaza CN360 Trenton, NJ 08625 Glenn Paulson, PhD Vice President for Science National Audubon Society 950 Third Avenue New York, NY 10022 co Ol t CO Dennis J. Paustenbach, PhD Product Health Risk Specialist Stauffer Chemical Company Westport, CT 06880 Frederica P. Perera, DrPH Senior Staff Scientist Natural Resources Defense Council 122 East 42nd Street, 45th Floor New York, NY 10168 Richard Peterson, PhD Associate Professor of Pharmacology and Toxicology School of Pharmacy University of Wisconsin-- Hadison 425 N. Charter Street Hadison, WI 53706 Carl Potter, PhD Postdoctoral Trainee Environmental Toxicology Center School of Pharmacy University of Wisconsin-- Hadison 425 N. Charter Street Hadison, Wl 53706 Christoffer Rappe, PhD Professor and Chairman Department of Organic Chemistry University of Umea S-901 87 Umea SWEDEN Harilyn Reeves Vice President Natural Resources Coordinator League of Women Voters of the U.S. 1730 H Street, NW Washington, DC 20036 Arleen B. Rifkind, HD Professor of Pharmacology Cornell University Hedical College E-319 New York, NY 10021 Richard A. Rlkfind, HD Director, Sloan-Kettering Division Graduate School of Hedical Sciences Hemorial Sloan-Kettering Cancer Center 1275 York Avenue New York, NY 10021 Thomas Robinson, PhD Director, Environmental Affairs Vulcan Chemicals P.O. Box 7689 Birmingham, AL 35253 t ( Kenneth Roseman, HD Director Occupational and Environmental Health Services New Jersey Department of Health John Fitch Plaza CN360 Trenton, NJ 08625 B. Sanjsder, HD National Institute of Public Health P.O. Box 1 3720BA Bilthoven THE NETHERLANDS Shigeru Sassa, DHedSci Associate Professor of Hetabolism-Pharmacology The Rockefeller University 1230 York Avenue New York, NY 10021 Arnold Schechter, HD, HPH Professor of Preventive Hedicine Clinical Campus Upstate Hedical Center State University of New York Binghamton, NY 13901 Robert Scheuplein, PhD Associate Director for Toxicological Sciences U.S. Food and Drug Administration HFF-101 200 C Street, SW Washington, DC 20204 SoX Schreiber Court Appointed Special Master, Agent Orange Cases U.S. District Court, Eastern District of New York One Pennsylvania Plaza (Suite 4915) New York, NY 10119 Carl Schulz, PhD Associate Scientist Clement Associates 1515 Wilson Blvd. Arlington, VA 22209 Barclay H. Shepard, MD Director Agent Orange Projects Office {10A7) U.S. Veterans Administration 810 Vermont Avenue, NW Washington, DC 20420 Ellen Silbergeld, PhD Chief Scientist Environmental Defense Fund 1525 18th Street, NW Washington, DC 20071 Burton H. Singer, PhD Professor of Mathematical Statist ics Center for Social Sciences Columbia University 420 west 118th Street New York, NY 10027 Kenneth Skipka Camp, Dresser t McKee 250 Broadway Suite 2601 New York, NY 10007 Andy Smith National Ministries American Baptist Churches, USA Box 851 Valley Forge, PA 09482-0857 Jerry F. Stara, DVM Director Environmental Criteria and Assessment Office Environmental Research Center Environmental Protection Agency 26 West St. Clair Cincinnati, OH 45268 Zena A. Stein, MB, BCh Professor of Public Health (Epidemiology), Columbia * University; and Director Epidemiology of Developmental Brain Disorders New York State Psychiatric Institute 722 West 168th Street New York, NY 10032 Steven Stellman, PhD Assistant Vice President for Epidemiology American Cancer Society 4 West 35th Street New York, NY 10001 Acos Szakolcai Air Resources Branch Ministry of the Environment 880 Bay Street Toronto, Ontario M5S 1Z8 CANADA Michael L. Taylor, PhD Associate Professor of Pharmacology/Tox icology Brehm Laboratory Wright State University Dayton, OH 45435 David E. Tenzer Professional Staff Media Resource Service Scientists' Institute for Public Information 355 Lexington Avenue New York, NY 10017 C cn GO O Stephen R. Thomas, PhD Assistant Professor of Political Science and Environmental Policy Department of Health Policy and Management Harvard School of Public Health 677 Huntington Avenue Boston, MA 02115 Thomas O. Tiernan, PhD Professor of Chemistry and Director Brehm Laboratory Wright State University *Dayton, OH 45435 Michael Torrusio Special Legislative Assistant on Energy and Environment Office of U.S. Congressman Guy V. Molinari Fort Wadsworth, Building 203 Staten Island, NY 10305 William A. Toscano, PhD Assistant Professor of Toxicology Harvard School of Public Health 665 Huntington Avenue Boston, MA 02115 Helle Tosine, PhD Advanced Instrumentation Supervisor Pesticides Section Laboratory Services Branch Ministry of the Environment P.O. Box 213 Rexdale, Ontario M9W 5L1 CANADA Thomas Umbreit, PhD Research Specialist Environmental and Community Medicine Department University of Medicine and Dentistry of New Jersey-Rutgers' Piscataway, NJ 08854 Joanna Underwood Executive Director Inform 381 Park Avenue South New York, NY 10016 John Van Ryzln, PhD Professor of Biostatistics School of Public Health Columbia University 600 West 168th Street New York, NY 10032 J.J. Vostal, MD Senior Medical Research Advisor General Motors Research Laboratory Warren, MI 48090 I. Bernard Weinstein, MD Director, Division of Environmental Sciences Columbia University College of Physicians and Surgeons 701 West 168th Street New York, NY 10032 James p. Whitlock, Jr., MD Associate Professor of Pharmacology School of Medicine Stanford University Stanford, CA 94305 pohn Wilkinson, PhD Technical Director Wood Treating Products Reichhold Chemicals Inc. 2340 Taylor Way Tacoma, Washington 98401 Mary S. Wolff, PhD Assistant Professor of Chemistry Environmental Sciences Laboratory Department of Community Medicine Mt. Sinai School of Medicine One Gustave L. Levy Place New York, NY 10029 James S. Woods, PhD Senior Research Scientist Batelle Human Affairs Research Centers 4000 NE 41st Street P.O. Box C-5395 Seattle, WA 98105 Alvin L. Young, Lt. Col., USAP, PhD Special Assistant for Environmental Sciences Agent Orange Projects Office Veterans Administration 810 Vermont Avenue, NW Washington, DC 20420 tM S4. 13 si- iS v1. 05 05 ftn o a s St- ft05 1 3 3 ftf t si- ^ . 1 3 05 S 4 tX, is uf t . ftf t f t s 4. IS 05 f t s S i 05 f t f t 05 s4. *3 t c t? 3 IS ea r* .S sfti K sfftti4 ftsi- f t f t f t . IS K O 05 f t si s4. g?Cf?c*i i si- ? 3 sf 05 asa+*- ffaooisttsJ s+ a Cri ,, 3 0 ^ 0 $a us+ 05 faat 4 sa ?0e5* 3 aianas- a is Pancho Villa's (Mexican) Garden of Eating (salads) 62 Manana Mix (Mexican) Wishbone (sandwiches) Misono (Japanese) F r i d a y 's (burgers and salads) 63 Chitose (Japanese) Hunan Lucky Deli 64 Mr. K's Noodle Parlor The Rockefeller University o< MZCW 66th Street 65 F l a n a g a n 's (Burgers and NY Irish) Wok on First (Chinese) 66 Peppermint Park (sandwiches and icecream) Tanj ore (Indian) w Rascal's * (burgers I and salads) 69 RESTAURANT GUIDE / Symposium Announcement and Call for Abstracts DIOXINS IN THE ENVIRONMENT December 6 and 7,1983 Kellogg Center for Continuing Education Michigan State University East Lansing, Michigan 48824 SEP 1 1983 Sponsored by Center for Environmental Toxicology, Michigan State University in cooperation with Cranbrook Institute of Science and Limno-Tech, Inc., Ann Arbor Continuing concern with dioxins in the United States, especially the Great Lakes region, has led to heightened public awareness and increased effort on the part of government, academic and industrial scientists to understand and deal with possible health and environmental problems due to these chemicals. The purpose of this symposium is to bring together scientists from these various sectors to discuss what we know at present, what we need to know for the future and what is presently being done to fill in these knowledge gaps. Symposium topics include analytical methodology, monitoring, environmental source, distribution and fate, human and animal health effects, mechanisms of action and public policy issues. Experts from U.S., Canada and Europe will deliver comprehensive talks which will place the most recent research results into an overall framework. Interested parties from State, Provincial and Federal agencies, academia and industry are invited to attend and participate. Special invitation is extended to individuals concerned with environmental quality in the Great Lakes region. Space will be provided for poster presentations. Poster abstracts on dioxins and related chemicals (single space, approximately 472x772) should be submitted to the Center for Environmental Toxicology by November 1, 1983. The abstracts will be reproduced and distributed at the meeting. This symposium will be followed by two days of workshops designed to summarize the state-of-knowledge of the symposium topics, explore relevant issues, and make recommendations for future research and policy decisions. The proceedings of the symposium and the results of the workshops will be published at a later date. 19532 Speakers include: D. Crosby, Department of Environmental Toxicology University of California - Davis W. Crummett, Analytical Laboratories Dow Chemical Company 3. Doull, Department of Pharmacology and Toxicology University of Kansas Medical Center M. Gross, Chemistry Department University of Nebraska R. Hites, Chemistry Department Indiana University O. Hutzinger, Laboratory of Environmental and Toxicological Chemistry University of Amsterdam, Netherlands W. Lowrance, Life Sciences and Public Policy Institute The Rockefeller University T. Mill, Department of Physical Organic Chemistry Stanford Research Institute E. E. McConnell, National Institute of Environmental Health Sciences National Toxicology Program A. B. Morrison, Health Protection Branch Health and Welfare, Canada E. Smuckler, Pathology Department University of California - San Francisco D. Stalling, Columbia National Fisheries Research Laboratory U.5. Fish and Wildlife Service R. Suskind, Institute of Environmental Health University of Cincinnati A. Young, Office of Environmental Medicine Veterans Administration The registration fee for this meeting will be $50. Meals and housing will be available in Kellogg Center at reasonable rates. Please use the attached form to obtain more detailed information and registration materials. Send further information on this meeting to: N am e_________________________________________ A ffiliation_________________________________ Address Title Information on Registration & Housing Joan Martin Alam Lifelong Education Programs 23 Kellogg Center Michigan State University East Lansing, MI 48824 (517) 353-7822 Scientific Information Center for Environmental Toxicology C-231 Holden Hall (517) 353-6469 f~The S c i e n c e o f t h e T o t a l E n v i r o n m e n t , 31 (1983) 203--218^ Elsevier Science Publishers B.V., Amsterdam --Printed in The Netherlands 203 CARCINOGENICITY AND TOXICITY OF 2,4-DICHLOROPHENOXYACETIC ACID MELVIN DWAINE^EUBER 11014 Swansfield Road, Columbia, M D 2 1 0 4 4 (U.S.A.) (Received February 9th, 1983; accepted May 4th, 1983) ABSTRACT 2.4- Dichlorophenoxyacetic acid (2,4-D) is carcinogenic in male and female rats and probably also in mice. Male and female rats ingesting 2,4-D developed increased incidences of malignant neoplasms. Lymphosarcomas were increased in rats of both sexes, and neoplasms of the mammary gland in female rats. Male rats also had carcinomas of the endocrine organs. 2,4-D isooctyl ester was carcinogenic for the lymphoreticular system in female mice. 2,4-D and 2,4-dichIorophenol also were promoters of neoplasms of the skin in mice. Male mice given 2,4-D isopropyl ester developed an increased incidence of neoplasms of the lung. 2.4- D also is mutagenic and teratogenic in animals and causes poisoning in animals and human beings. INTRODUCTION The auxinlike, plant growth-promoting chemical, 2,4-dichlorophenoxyacetic acid (2,4-D), has been used for may years as a plant herbicide [1]. The acid, its salts, and its esters are used to kill dicotyledonous weeds and certain other plant species. The chemical is readily absorbed from the digestive tract of animals and excreted by the kidney. Acute and subacute toxicity tests have shown that 2,4-D can cause poisoning in animals. 2,4-D is widely used as a herbicide to control certain types of vegetation. It is sprayed from airplanes on many millions of acres to kill shrub and broad leaved plant life on specialized farms and range and pasture lands used for grazing. 2,4-D is used to kill vegetation adjacent to railroads, highways and other roads, pipe lines, power lines, and in and around lakes, ponds and irrigation ditches. It is used in aerial spraying operations of forest lands to kill broad-leaved vegetation and encourage soft wooded trees such as spruce and pine. It is present in preparations sold at retail for home gardeners and is widely used on home lawns and grass covered areas used for recreational purposes. 2,4-D is also used on cereal crops such as rice, on sugar cane and to control ripening o f bananas and citrus fruits, to delay preharvest dropping of some fruits, and for the control of rot when lemons are stored. 2,4-D 0048-9 6 9 7 /8 3 /SO3.00 1983 Elsevier Science Publishers B.V. o)Ht 4^' i` l u U was extensively sprayed together with 2,4,5-trichiorophenoxyacetic acid on millions of acres as a defoliant in Vietnam. With the widespread use of this herbicide, possible toxic effects are important. 2,4-D finds its way to humans through direct contact either occupationally or by spraying humans purposely or accidentally. Drift after spraying can contaminate gardens, plant nurseries, water and other areas. Residues may inadvertently be encountered in food, vegetables, meat and water. The production and use of 2,4-D can only be estimated because the data often is considered as trade secret by the manufacturers. The latest data available are for the year 1975. An estimated 27 million kg were used in the U.S.A. The Federal Republic of Germany and the U.K. are the major pro ducing countries of western Europe where annual production was estimated to be 3--30 million kg. In eastern Europe it was estimated to be less than 10 million kg. In Japan production amounted to 511 thousand kg. This review includes, to the best of our knowledge, every study on the carcinogenicity of 2,4-D, or its esters, and a metabolite, 2,4-dichlorophenol in animals. The raw data and histological sections were reviewed for the FDA 2,4-D Rat Study and FDA 2,4-D Dog Study [1--3] and the results are based on this examination. The Innes et al. Mouse Studies for 2,4-D, or its esters [4 ], the Archipov and Kozlova 2,4-D Mouse and Rat Studies [5 ], and the Boutwell and Bosch 2,4-Dichlorophenol Mouse Study [6] are also included. Statistical tests of significance were p (probability) values obtained with Fisher's exact test and tests for positive linear trend and departure from linear trend. Osbome-Mendel rats, three weeks of age, ingested either 0, 5, 25, 125, 625, or 1250 ppm of 2,4-D in the diet for 104 weeks [1, 3 ]. There were 25 males and 25 females at each dose. The rats were individually caged and were weighed weekly. Hematologic studies were done periodically. Survivors were killed after 104 weeks, and organ weights were taken for heart, liver, spleen, kidneys, and testes. Organs were also weighed from moribund rats killed after 52 weeks; however, no organ weights were in cluded from rats that died. 1953-8 205 TABLE 1 N U M B E R OF M A L E RATS INGESTING 2,4-D WITH CARCINOMAS A N D SARCOMAS (FDA RAT STUDY) Dose (ppm) Carcinomas P Sarcomas P Botha P 0 5 25 125 625 1250 5--1250 0/25 (0%) 2/25 (8%) 6/25 (24%) 6/25 (24%) 5/24(21% ) 3/23(13% ) 22/122 (18%) 0.011 0.011 0.022 0.012 0.057 1/25 (4%) 2/25 (8%) 4/25 (16%) 6/25 (24%) 4/24 (17%) 7/23 (30%) 23/122 (19%) 0.049 0.018 0.052 0.021b 1/25 (4%) 3/25 (12%) 9/25(36% ) 9/25 (36%) 6/24 (25%) 9/23(39% ) 36/122 (30%) 0.0053 0.0053 0.043 0.0033 0.0038 0.039b 0.034 1 Some rats had both carcinoma and sarcoma and are counted only once. b Test for positive trend. 0 Departure from linear trend. Detailed histologic examination was made of tissues from six male rats and six female rats from the 1250 ppm 2,4-D groups and the control groups. Liver, kidney, spleen, and ovary or testes, as well as tumors and other gross lesions, were sectioned histologically from the rats on all other doses of 2,4,-D. At the end of 104 weeks the following rats were alive: 12 males, 6 females (Oppm); 8 males, 4 females (5 ppm); 9 males, 9 females (25 ppm); 9 males, 8 females (125 ppm); 10 males, 9 females (625 ppm); and 5 males, 11 females (1250 ppm). Body weights and the organ--body weight ratios for liver, kidney, heart and testes were reported as showing no differences among the various groups. These results have little meaning, however, since the rats that died with the most chance of being abnormal were not included. The sp leenbody weight ratio was slightly elevated in rats given 625 ppm or 125 ppm 2,4-D (p < 0 .0 5 ). Terminally, the red blood cells of rats given 5 ppm, 625 ppm, or 1250 ppm 2,4-D showed macrocytosis, slight polychromasia, and slight to moderate hypochromasia. These changes were not present, or were of a minor degree in control rats. M a lig n a n t neo p la sm s at all sites in male rats There were increased incidences of malignant neoplasms in male rats given all doses of 2,4-D (Table 1). Nine of 25 rats (36%) given 25 or 125 ppm {p = 0.0053) and 9 of 23 (39%) given 1250 ppm of 2,4-D {p = 0.0033) developed carcinomas and sarcomas. The incidences o f these neoplasms were dose related (p = 0.039). For some unexplained reason, the rats re ceiving 625 ppm, 6 of 24 (25%), had fewer malignant neoplasms than male TABLE 2 N U M B E R OF M A L E RATS INGESTING 2,4-D WITH LY MP HO SA RC OM AS A N D NEUROSARCOMAS (FDA RAT STUDY) Dose (ppm) Lymphosarcoma P Subcutaneous neurosarcomas 0 5 25 125 625 1250 5--1250 0/25 (0%) 2/25 (8%) 4/25 (16%) 5/25 (20%) 3/24 (13%) 6/23 (26%) 20/122 (16%) 0.055 0.025 0.0082 0.018 0.034a 0/25 (0%) 0/25 (0%) 0/25 (0%) 1/25 (4%) 1/24 (4%) 1/23 (4%) 3/122 (2%) Test for positive trend. rats in other groups (p = 0.043). Thirty-six of 122 (30%) 2,4-D-treated male rats had carcinomas and sarcomas (p = 0.0038). Sa rc o m a s in m ale rats The sarcomas in 2,4-D-treated male rats were mainly lymphosarcomas, however, there were a small number of subcutaneous neurosarcomas (Table 2). Sarcomas were present in 4 of 25 rats (16%) ingesting 25 ppm, 6 of 25 rats (24%) ingesting 125 ppm {p = 0.049), and 7 of 23 (30%) receiving 1 ,2 5 0 ppm of 2,4-D (p = 0.018) (Table 1). Twenty-three of 122 (19%) male rats given 2,4-D had sarcomas (p = 0.052). C a rcin o m a s in m ale rats Carcinomas in male, rats treated with 2,4-D were seen in the endocrine system. Six of 25 male rats (24%) ingesting 25 or 125 ppm (p = 0.011) and 5 of 24 rats (21%) given 625 ppm (p = 0.022) o f 2,4-D developed carcino mas (Table 1). Twenty-two of 122 (18%) male rats given 2,4-D had car cinomas (p = 0.012). M a lig n a n t n eop la sm s at all sites in fem ale rats Eight of 20 (40%) female rats given 5 ppm (p = 0.019), 11 of 22 (50%) given 25 ppm (p = 0.058), 13 of 23 (57%) on 125 ppm (p = 0.022), 18 of 24 (75%) receiving 625 ppm (p = 0.00047), and 17 of 25 (68%) female rats ingesting 1,250 ppm (p = 0.0022) of 2,4-D developed malignant neo plasms at all sites (Table 3). Altogether, malignant neoplasms were ob served in 67 of 114 female rats (59%) ingesting 2,4-D (p = 0.0018). L e s io n s o f the ly m p h o re ticu la r system in fem ale rats The highest incidence of lymphosarcomas was 12 of 24 (50%) female rats given 625 ppm of 2,4-D (p = 0.00007) (Table 4). Female rats in the other TABLE 3 NUMBER OF FEM ALE RATS INGESTING 2,4-D WITH CARCINOMAS AND SA R COMAS (FDA RAT STUDY) Dose Carcinomas P (ppm) Sarcomas P Botha P 0 5 25 125 625 1250 5--1250 4/22 (18%) 8/20 (40%) 7/22 (32%) 10/23 (43%) 0.065 14/24 (58%) 0.0059 13/25 (52%) 0.017 52/114(46% ) 0.013 0.017b 1/22 (5%) 5/20 (25%) 8/22 (36%) 0.011 8/23 (35%) 0.013 12/24 (50%) 0.00007 10/25 (40%) 0.0044 43/114 (36%) 0.0017 0.024 0.089 5/22 (23%) 8/20 (40%) 11/22 (50%) 0.058 13/23 (57%) 0.022 18/24 (75%) 0.00047 17/25 (68%) 0.0022 67/114 (59%) 0.0018 0.0022b 0.090c a Some rats had both carcinomas and sarcomas and are counted only once. b Test for positive trend. 0 Departure from linear trend. 2,4-D treatment groups had 24--27% lymphosarcomas (p = 0.011--0.018). Thirty-one percent of all females (p = 0.00073) developed lymphosarcomas. The findings are even more significant if hyperplasia of the lymphoreticular system is considered along with lymphosarcomas, i.e., 39% (p = 0.00005) of female rats ingesting 2,4-D. Female rats also had sarcomas of the uterus. Ca rcinom a s at all sites in fem ale rats Carcinomas at all sites were seen in 43% (10 of 23) (p = 0.065) of female rats on 125 ppm, 58% (14 of 24) (p = 0.0059) given 625 ppm, 52% (13 of 25) (p = 0.017) receiving 1250 ppm and 46% (52/114) (p = 0.013) of female rats given 2,4-D (Table 3). Carcinomas were found in the repro ductive system, particularly the mammary gland, and occasionally in endo crine organs. N e o p la sm s o f the m a m m a ry gla n d in fem ale rats The incidence of mammary gland tumors was higher in treated female rats, particularly on gross examination, which is generally quite reliable. The number of rats with such tumors was less in all but one group after histologic examination (Table 5). On gross examination, 18 of 21 female rats (86%) ingesting 25 ppm (p = 0.0064), 20 of 24 rats (83%) ingesting 625 ppm (p = 0.0080) and 16 of 23 rats (70%) given 1250 ppm of 2,4-D had neoplasms of the mammary gland, compared to 10 of 22 (45%) control female rats. The findings are quite different for the numbers oi rats with histological sections examined by pathologists at FDA and sections available for this examination. 208 TABLE 4 NUMBER OF FEMALE RATS INGESTING 2,4-D WITH LESIONS OF THE LYMIMIORETICULAR SYSTEM (FDA RAT STUDY) Dose (ppm) Hyperplasia 0 5 25 125 625 1250 5-1250 0/22 (0%) 0/22 (0%) 2/22 (9%) 2/23 (9%) 4/24 (17%) 2/25 (8%) 10/114(9% ) P 0.065 Lymphosarcoma 0/22 (0%) 5/20 (25%) 6/22 (27%) 6/23 (26% 12/24 (50%) 6/25 (24%) 35/114 (31%) P 0.018 0.011 0.012 0.00007 0.016 0.00073 Hyperplasia and lymphosarcoma P 0/22(0% ) 5/20 (25%) 8/22(36% ) 8/23 (35%) 16/24 (67%) 8/25 (32%) 45/114 (39%) 0.018 0.0018 0.0023 < 0.00001 0.0034 0.00005 0.030" 0.0004b a Test for positive trend. b Departure from linear trend. N TABLE 5 NUMBER OF FEMALE RATS INGESTING 2,4-D WITH NEOPLASMS OF THE MAMMARY GLAND AND HISTOLOGIC EXAM INATION" (FDA RAT STUDY) Dose (ppm) 0 5 25 125 625 1250 5-1250 Grossb 10/22(45% ) 8/20 (40%) 18/21 (86%) 8/23(35% ) 20/24 (83%) 16/23(70% ) 70/111(63% ) P 0.0064 0.0080 0.091 0.0233* 0.0008' Histologic* 10/22(45% ) 7/20(35% ) 11/21 (52%) 10/23(43% ) 14/24(58% ) 15/23(65% ) 57/11 1 (52%) P 0.0334* IIistologicd 8/22(36% ) 7/20 (35%) 11/21 (50%) 9/23 (39%) 15/24 (63%) 11/23 (45%) 53/1 11 (47%) P 0.070 " Corrected for survival time, i.e., time of appearance of first neoplasms o f the mammary gland. b Gross necropsy at FDA. c Histological examination by FDA. d Histological examination by Rcuber. e Test for positive trend. ' Departure from trend. 209 r.< u-V' 210 Com m ents It is worth noting that histological examination of tissues from this study was inadequate. Microscopic neoplasms would have been overlooked at the 5 to 625 ppm doses of 2,4-D (only gross neoplasms were sectioned histologi cally) and also even at the highest dose because only six rats of each sex were examined in detail. Also histological sections were not available for 19 mammary gland neoplasms described at the time of necropsy, and no explanation was given for this discrepancy. Despite the shortcomings of this study, it must be considered as an acceptable study. The largest numbers of rats alive were in the 25 to 1250 ppm doses; and the smallest number of rats alive was in the lowest dose, 5 ppm. These results suggest that rats in at least some o f the 25 to 1250 ppm groups were not ingesting their diets because of toxicity. The differences in the incidences of neoplasms, therefore, would not be great. Tumors were analyzed, on the basis of 25 rats per group, and the following conclusions were made by the authors: "There is a statistically significant (p < 0.05) linear relationship between the proportion of female rats with tumors and the level of the log dose, but not the arithmetic dose, indicating that there is a tendency for the proportion of female rats with tumors to increase with the log dosage. There is also a statistically significant (p < 0.05) linear relationship between the proportion of male rats with malignant tumors and the level of arithmetic and log dose. A comparison was made of the control group with each treatment group. Statistically significant ( p < 0 .0 5 ) differences were found only between the control group and the 1250 ppm dose level with respect to male rats with malignant tumors." The authors also believed that the raw data and the pathologic interpre tation did not support the statistical analyses and that "a carcinogenic effect of 2,4-D has not been shown". They also stated that "additional support for this interpretation has been given by the long-term study in m ice" . Sum m ary Male and female rats ingesting 2,4-D developed increased incidences of malignant neoplasms. Lymphosarcomas were increased in rats of both sexes, and neoplasms of the mammary gland in female rats. C o n c lu sio n s 2,4-D is carcinogenic for male and female rats. FDA 2,4-D DOG STUDY Beagle dogs, 6 to 8 months old (3 males and 3 females per group) ingested 0, 10, 50, 100, or 500 ppm 2,4-D in the diet for 104 weeks [1, 2 ]. Organ weights were taken for brain, heart, liver, kidneys, spleen, thyroid, adrenals, and testes. Tissues from all dogs were studied grossly and microscopically. i ii on'J ~fj\t 9 A/ Many of the dogs ingesting 2,4-D lost weight. There were scattered lesions such as atrophy of the testes and prostate, interstitial nephritis, hemangioma of the adrenal, atrophic or cystic pituitary, atrophy of the thyroid, and hypoplasia of the bone marrow. Most of the lesions were seen in the endocrine organs. Control dogs generally did not have lesions. Com m ents Previous experience at FDA has shown that long-term chronic dog studies should be carried out for six years or longer in order for neoplasms to develop. Since they occurred predominantly in 2,4-D treated dogs, lesions may have progressed to neoplasms had the dogs been treated for a longer period of time. Sum m ary A two-year feeding study in dogs cannot be considered a carcinogenicity study. INNES et al. 2,4-D, OR ITS ESTERS, ORAL MOUSE STUDIES 1. 2 ,4 -D The maximum tolerated dose of 2,4-D was given to two hybrid strains of mice, (C57BL/6 x C3H/Anf)F! designated as "strain A " and (C57BL/6 x AKR)F[ designated as "strain B" mice [4, 7 ], There were 18 treated mice and 18 untreated controls of each strain and each sex. 46.4 mg/kg was given in 0.5% gelatin daily by stomach tube beginning at 7 days of age. After the mice were weaned at 28 days o f age, 149 ppm of 2,4-D was mixed directly in the diet and provided ad libitum. "Strain B" male and female mice were also. given 100 mg/kg, followed by 323 ppm. Treatment was continued approximately 18 months. Postmortem included thorough external examination and internal examination of the neck glands and the thoracic and abdominal cavities, with histologic examination of major organs and of all grossly visible lesions. Thyroid glands were not examined. There was no increase in neoplasms in the 2,4-D-treated mice. 2. 2 ,4 -D iso p ro p y l ester "Strain A " and "strain B" male and female mice were given 46.6 mg/kg of 2,4-dichlorophenoxyacetic acid, isopropyl ester in 0.5% gelatin daily by stomach tube. At 28 days of age the mice received 111 ppm in the diet for approximately 18 months. There were neoplasms in the lungs of 4 of 18 "strain A " male mice (22%) ingesting 2,4-D isopropyl ester compared to 2 of 17 matched controls (12%) and 5 of 79 pooled male control mice (6%) (p = 0.0584) (Table 6). In this study, "strain A " male mice receiving 2,4-D isopropyl ester de veloped an increased incidence o f neoplasms of the lung. 212 TABLE 6 NUMBER OF MALE AND FEMALE MICE INGESTING 2,4-D ISOPROPYL ESTER WITH NEOPLASMS OF THE LUNG (INNES et al. [4, 7 ], MOUSE STUDY) Strain "A" Matched "A " Pooled "A" "B" Matched "B " Pooled "B" Dose (ppm) 0 0 111 0 0 111 Male 2/17 (12%) 5/79 (6%) 4/18 (22%) 2/18 (11%) 9/90 (10%) 2/18 (11%) P 0.0584 Female 1/18 (6%) 3/87 (3%) 1/18 (6%) 0/17 (0%) 3/82(4% ) 0/17 (0%) 3. 2 ,4 -D b u ty l ester "Strain A" .and "strain B " male and female mice received 46.4m g/kg of 2,4-dichlorophenoxyacetic acid, butyl ester in 0.5% gelatin daily by stomach tube. At 28 days of age the mice received 149 ppm in the diet for approximately 18 months. Three of 18 "strain A" female mice (17%) and 4 of 87 pooled controls (5%) had reticulum cell sarcomas (p = 0.0955) (Table 7). In this study, there was an increased incidence of reticulum cell sarcoma in 2,4-D butyl ester-treated "strain A " female mice. 4. 2 ,4 -D iso o cty l ester "Strain A" and "strain B" male and female mice received 46.4m g/kg of 2,4-dichlorophenoxyacetic acid, isooctyl ester in 0.5% gelatin by stomach tube. After 28 days they ingested 130 ppm in the diet for approximately 18 months. Neoplasms in treated mice were found mainly in the liver and lung. One "strain A" treated male mouse had an "angioma" o f the liver and another an "angioma" o f the spleen. In this study, neoplasms of the liver were slightly increased in 2,4-D TABLE 7 NUMBERS OF MALE AND FEMALE MICE INGESTING 2,4-D BUTYL ESTER WITH RETICULUM CELL SARCOMAS (INNES et al. [4, 7 ], MOUSE STUDY) Strain "A " Matched "A " Pooled "A" "B " Matched "B " Pooled "B " Dose (ppm) 0 0 149 0 0 149 Male 1/17 (6%) 5/79 (6%) 0/18 (0%) 1/18 (6%) 1/90 (1%) 2/18 (11%) P 0.0714 Female 2/18 (11%) 4/87 (5%) 3/18 (17%) 2/17 (12%) 3/82 (4%) 0/16 (0%) P 0.0955 19544 isooctyl ester-treated "strain A" male mice and tumors of the lung in "strain A" female mice. Two male mice had rare tumors; "angiomas" of the liver and spleen. INNES et al. 2 ,4 -D, OR ITS ESTERS, SUBCUTANEOUS MOUSE STUDY A single subcutaneous injection of 2,4-D, or its esters, was given in the nape of the neck to two hybrid strains of mice, (C57BL/6 x C3H/Anf)Fj designated as "strain A" and (C57BL/6 x AKR)F[ designated as "strain B" at approximately the 28th day of age [7 ]. There were 18 treated mice and 18 untreated controls of each strain and each sex. They were killed after approximately 18 months. Mice received the following doses: 215m g/kg or 464 mg/kg of 2,4-D in DMSO; 100 mg/kg of 2,4-D isopropyl ester, 25.5 mg/kg of 2,4-D butyl ester, or 21.5 mg/kg of 2,4-D isooctyl ester in corn oil. Necropsy included thorough external examination and internal exam ination of the neck glands and the thoracic and abdominal cavities. Histo logical examination was done on all major organs and of all grossly visible lesions. R e t ic u lu m cell sarcom as in "strain B " fem ale m ice Five of 17 "strain B" female mice (29%) given 2,4-D isooctyl ester had reticulum cell sarcomas, compared to 5 of 157 control female mice (3%) [p = 0.0009) (Table 8). Sum m ary 2,4-D isooctyl ester was carcinogenic for the lymphoreticular system (reticulum cell sarcomas) in "strain B" female mice. ARCHIPOV AND KOZLOVA MOUSE AND RAT STUDIES Three groups, 100 each, of CBAx C57/BL hybrid mice were used [5]. Two groups of mice received one drop of a 0.5% solution of 3-methylcholanthrene TABLE 8 NUMBER OF "STRAIN B" MALE AND FEMALE MICE GIVEN A SINGLE SUB CUTANEOUS INJECTION OF 2,4-D ISOOCTYL ESTER WITH RETICULUM CELL SARCOMAS (INNES et al. [ 4 , 7 ] , MOUSE STUDY) Dose (mg/kg) 0 1 Males 0/161 (0%) 0/18 (0%) Females 5/157 (3%) 5/17 (29%) P 0.0009 in benzene on the skin for 3 weeks. Later the skin of one group o f mice was painted with a 10% solution o f the amine salt of 2,4-D in acetone. The skin of the third group of mice was painted with a 10% solution of the herbicide. Mice were observed for 20 months. Neoplasm s o f the skin Papillomas developed on the skin of 17.7% of the mice treated with 3methylcholanthrene followed by 2,4-D, and none for the mice receiving 3-methylcholanthrene only or 2,4-D only. Com m ents Groups of rats also ingested l/1 0 th the LDS0 of the amine salt of 2,4-D [5 ]. The dose was not given. Sum m ary 2,4-D is a promoter of neoplasms of the skin in mice. BOUTWELL AND BOSCH 2,4-DICHLOROPHENOL MOUSE SKIN STUDY Female Sutter mice, 2--3 months of age, were used [6 ]. Mice were painted on the skin with a single initiating dose of 0.3% dimethylbenzanthracene in acetone followed by 20% 2,4-dichlorophenol in benzene, benzene only, or no treatment. 2,4-Dichlorophenol was applied twice weekly. One group of mice was observed for 15 weeks and a second for 24 weeks. Neoplasm s o f the skin Mice treated with both chemicals developed an increased incidence of neoplasms of the skin. Thirteen of 27 treated mice (48%) (p = 0.00607) had papillomas and 3 o f 27 treated mice (11%) had carcinomas o f the skin, compared to 1 of 15 untreated mice (7%) with papillomas and 0 of 15 untreated mice (0%) after 15 weeks. Papillomas of the skin were observed in 12 of 16 treated mice (75%) (p = 0.00004) and carcinomas of trie skin in 10 of 16 treated mice (62%) {p < 0.00001); whereas, papil lomas were seen in 3 of 27 mice (11%) and carcinomas in 0 of 27 mice (0%) receiving benzene only after 24 weeks. Conclusion 2,4-Dichlorophenol promoted the incidence of neoplasms of the skin in mice when administered after a single initiating dose o f dimethylbenz anthracene. DISCUSSION 2,4-D was found to be carcinogenic for rats in the original publication by 215 Hansen et al. [1 ]. It was also concluded that "2,4-dichlorophenoxyacetic acid is carcinogenic in rats" in a preliminary review of the raw data, but not the histological sections, requested by Senator Edward M. Kennedy's Subcommittee on Administrative Practice and Procedure of the Committee on the Judiciary, United States Senate [8 ]. The examination and diagnoses of the histological sections, which was done for this review further strength ened the findings and conclusion that 2,4-D is carcinogenic in rats. Experience has shown that it is necessary to review the raw data and histological sections of old chronic toxicity and carcinogenicity studies in animals [9, 10, l l ] . The data from studies often is not adequately analyzed and the conclusions are suspect. It also is not unusual for results of such studies, which are statistically significant, to be ignored because they are described as not biologically significant. The pathology report for the FDA 2.4- D study in rats was completed in 1964; however, the results were not published until 1971 [1, 3 ]. 2.4- D is mutagenic and teratogenic as well as carcinogenic [12]. 2,4-D and its derivatives are teratogenic in mice, rats and hamsters. 2,4-D induces skeletal malformations, cleft palats, eye malformations, subcutaneous edema and hemorrhages in the snout, abdominal cavity, liver and soft tissues in rats when administered during early pregnancy [7, 13--17]. 2,4-D is embryotoxic [18]. 2,4-D penetrates the placenta and the fetuses [1 9 ]. 2.4- D caused point mutations in animal cells without liver activation, damages DNA in a manner similar to ionizing radiation and stimulates mitosis [20--22]. Endothel-treated fruit flies developed an increase of recessive lethal mutations [2 3 ]. 2.4- Dichlorophenol is a major product of the breakdown of 2,4-D by microorganisms [2 4 ]. The compound is also teratogenic in rats and carcino genic in mice. [6 ,1 4 ]. 2.4- D is rapidly absorbed from the human gastrointestinal tract and distributed widely throughout the body in animals and in human beings [25--2 8 ]. It is fat soluble and rapidly absorbed through the skin and lungs [2 9 ]. Acute effects of 2,4-D poisoning in humans include headache, weakness, dizziness, nausea and vomiting, sore throat, irritation of nasal mucosa, impared sense of taste and smell, substemal chest pain and loss of consciousness [3 0 ]. Peripheral neuropathy, which begins with tingling and numbness in the extremeties, has been reported in humans hours or days after exposure to 2,4-D on thejkin. The symptoms in some people increased through several weeks until pain, paresthesia, and paralysis were severe. Disability was protracted and recovery was incomplete even after a lapse of years [31, 3 2 ]. These symptoms resemble those of multiple sclerosis, and this has been confirmed by the finding o f plaques of acute demyelination in all parts of the brain o f a man who died after ingestion of 2,4-D [3 3 ]. The formation and occurrence of tumors in man and other mammals, such as rats and mice, is quite similar [1 1 ]. In tests undertaken to date, it has been demonstrated that virtually every chemical which has been found to be carcinogenic in man is also carcinogenic in one or more mammalian test animals. It has been shown that if these compounds will produce tumors in one species, they will very likely produce tumors in more than one; thus adding weight to any finding of carcinogenicity in tests using any mammalian species. Sufficient documentation is available on qualitative extrapolation of animal data that one must conclude that a finding of carcinogenicity in one mammalian species should be deemed to have relevance for other mammalian species -- including man. This threat may not be manifested for up to 30 or 40 years, given the long latent period of many known human carcinogens. Given the state of carcinogenicity testing and knowledge, there can be only one safe tolerance for a carcinogen -- an absolute zero tolerance -- and the only way a substance shown to be carcinogenic in test animals will not be a threat to human health is simply to prevent that substance from entering the environment. To the extent that it does enter man's environment, it will constitute a very real threat to his health. Recent reports have described an increase of malignant neoplasms in human beings exposed to phenoxyacetic acids. Humans exposed to phenoxyacetic acids or chlorophenols showed an increased risk for the development of soft-tissue sarcomas [34, 3 5 ]. Results also have indicated that humans in contact with phenoxy acids may also develop malignant lymphomas [36]. ACKNOWLEDGEMENT Statistical analyses were done by C. A. Riggs, Information Systems Department, Frederick Cancer Research Center. English translations were kindly supplied by Ruth M. Shearer, Issaquah Health Research Institute, Issaquah, Washington. W. H. Hansen, R. T. Haberman and 0 . G. Fitzhugh, Chronic toxicity o f 2,4dichlorophenoxyacetic acid in rats and dogs, Toxicol. Appl. Pharm., 20 (1971) 122-192. T. Haberman, FDA Memorandum entitled: Pathological Changes in Dogs Fed 2,4'.Dichlorophenoxy Acid for Two Years, A. J. Lehman dated November 5, 1963. T. Haberman, FDA Memorandum entitled: Pathological Changes in Rats Fed 2,4Dichlorophenoxy Acid for Two Years, A. J. Lehman dated March 20, 1964. J. R. M. Innes, B. M. Ulland, M. G. Valerio, L. Petrucelli, L. Fishbein, E. R. Hart, A. J. Pallota, R. R. Bates, H. L. Falk, J. J. Gart, G. M. Klein, I. Mitchell and <1. Peters, Bioassay of pesticides and industrial chemicals for tumorigenicity in mice: A preliminary note, J. Natl. Cancer Inst., 42 (1969) 1101--1114. G. N. Archipov and I. N. Kozlova, Study of the carcinogenicity properties o f the herbicide amine salt of 2,4-D, Voprosy Pitaniya, 5 (1974) 83--84 (English trans lation from Russian). 217 6 R. K. Boutwell and D. K. Bosch, The tumor-promoting action o f phenol and related compounds for mouse skin, Cancer Res., 19 (1959) 413--424. 7 Bionetics Research Labs, Inc.: Evaluation of carcinogenic, teratogenic, and mutagenic activities of selected pesticides and industrial chemicals, Vol. 1 Carcinogenicity, National Technical Information Service, U.S. Department o f Commerce, 1968. 8 M. D. Reuber, Preliminary review o f some oncogenicity studies for 2,4-dichlorophenoxyacetic acid, Appendix of the Environmental Protection Agency and the Regulation of Pesticides, U.S. Government printing office, Washington, DC, 1977, pp. 471--476. 9 M. D. Reuber, Review o f toxicity test results submitted in support o f pesticide tolerance petitions, Sci. Total Environ., 9 (1977) 135--148. 10 M. D. Reuber, Review of additional toxicity test results submitted in support of pesticide tolerance petitions. Appendix of the Environmental Protection Agency and the Regulation of Pesticides, U.S. Government printing office, Washington, DC, 1977, pp. 460--470. 11 M. D. Reuber, Carcinogenicity testing of chemicals with particular reference to organochlorine pesticides, Sci. Total Environ., 10 (1978) 105--115. 12 R. W. Shearer, Personal communication (1979). 13 B. A. Schwetz, G. L. Sparschu and P. J. Gehring, The effect of 2,4-dichloro- phenoxyacetic acid (2,4-D) and esters of 2,4-D on rat embryonal, foetal and neonatal growth and development, Food. Cosmet. Toxicol., 9 (1971) 801--817. 14 T. K. Konstantinova, L. P. Ephimenko and T. A. Antonenko, The embryotropic effect of the dissociation products of herbicides based on 2,4-D, Gigienal. Sanitariya, 11 (1976) 102--105 (English translation from Polish). 15 K. S. Khera and W. P. McKinley, Pre- and postnatal studies on 2,4,5-trichiorophenoxyacetic acid, 2,4-dichlorophenoxyacetic acid and their derivatives in rats, Toxicol. Appl. Pharmacol., 22 (1972) 14--28. 16 T. F. X. Collins and C. H. Williams, Teratogenic studies with 2,4,5-T and 2,4-D in the hamster, Environ. Contam. Toxicol., 6 (1971) 559--567. 17 D. L. Courtney, Prenatal effects of herbicides: Evaluation by the prenatal develop ment index, Arch. Environ. Contam. Toxicol., 6 (1977) 33--46. 18 Z. A. Alekaashina, S. Y. Buslovich and V. M. Kolosovskaya, Embryotoxic action of the diethylamine salt of 2,4-D, Gigiena I. Sanitariya, 2 (1973) 100--101 (English translation from Polish). 19 L. M. Fedorova and R. S. Belova, 2,4-D detected in reproductive organs and fetuses of treated pregnant rats, ways and dynamics of its removal, (English translation from Russian). 20 S. V. Basrur, R. A. Fletcher and P. K. Basrur, In vitro effects o f 2,4-dichlorophenoxy acetic acid (2,4-D) on bovine cells, Can. J. Comp. Med., 40 (1976) 410--415. 21 R. W. Hart, S. Hays, D. Brash, F. B. Daniel, M. T. Davis and N. J. Lewis, In vitro assessment and mechanism o f action of environmental pollutants, Ann. N. Y. Acad. Sci., 298 (1977) 1 4 1 -1 5 8 . 22 F. E. Ahmed, R. W. Hart and N. J. Lewis, Pesticide-induced DNA damage and its repair in cultured human cells, Mutat. Res., 42 (1977) 161--174. 23 J. C. Hadder, Endothal induced mutations in drosophila melanogaster, Trans. 111. State Acad. Sci., 63 (1970) 157--159. 24 D. F. Paris and D. L. Lewis, Chemical and microbial degradation o f ten selected pesticides in aquatic systems, Residue Rev., 45 (1973) 107. 25 J. D. Kohli, R. N. Khanna, B. N. Gupta, M. M. Dhar, J. S. Tandon and K. P. Sircar, Absorption and excretion of 2,4-dichlorophenoxyacetic acid in man, Xenobiotica, 4(1974) 97-100. 26 S. Khana and S. C. Fang, Metabolism of Cl4-labeled 2,4-dichlorophenoxyacetic acid in rats, J. Agr. Food Chem., 14 (1966) 500--503. 27 K. Erne, Distribution and elimination of chlorinated phenoxyacetic acid in animals, Acta Vet. Scand., 7 (1966) 240--256. 4 ^Vl IM U \ ^ *** 5 <"* 1; i > A 4J 218 28 M. W. Sauerhoff, W. H. Braun, G. E. Blau and P. J. Gehring, The fate o f 2,4-dichloro- phenoxyacetic acid (2,4-D) following oral administration to man, Toxicology, 8 (1977) 3--11. 29 J. A. Burton, T. H. Gardiner and L. S. Schanker, Absorption of herbicides from the rat lung. Arch. Environ. Health, 29 (1974) 31--33. 30 A. D. Radionov, A. N. Chumachemko and I. I. Kirilenko, The toxic properties of the herbicide 2,4-D, Hyg. Sanit., 32 (1967) 116--118 (English translation from Russian). 31 N. P. Goldstein, P. H. Jones and J. R. Brown, Peripheral neuropathy after exposure to an ester of dichlorophenoxyacetic acid, J. Amer. Med. Assoc., 171 (1959) 1306-- 1309. 32 M. C. Berkley and K. R. Magee, Neuropathy following exposure to a dimethylamine salt of 2,4-D, Arch. Int. Med., I l l (1963) 351--352. 33 A. W. Dudley, Jr. and N. T. Thapar, Fatal human ingestion of 2,4-D, a common herbicide, Arch. Pathol., 94 (1972) 270--275. 34 L. Hardell and A. Sandstrom, Case-control study: Soft tissue sarcomas and exposure to phenoxyacetic acids or chlorophenols, Br. J. Cancer, 39 (1979) 711--717. 35 M. Eriksson, L. Hardell, N. O. Berg, T. Moller and O. Axelson, Case control study on malignant mesenchymal tumors of the soft-tissue and exposure to chemical sub stances, Lakartidningen, 76 (1979) 3872--3875. 36 L. Hardell, M. Eriksson and P. Lenner, Malignant lymphoma and exposure to chemi cal substances especially organic solvents, chlorophenols and phenoxy acids, Lakar tidningen, 77 (1980) 208--210. 13550 V. e: E. BI DC Di So (R. AE ope she in : dec dat: of r estir resu INT F 19c Ana con rad: tail! cou: R haza hum coni * Su: Actic **Pr> Engir 0048 4 Sarcomas in Male Rats The sarcomas in 2,4-D treated male rats were mainly lymphosarcomas, however, there were a small number of subcutaneous neurosarcomas (Table 2). Sarcomas were present in 4 of 25 rats (16%) ingesting 25 ppm (p = 0.055), 6 of 25 rats (24%) ingesting 125 ppm (p = 0.011), and 7 of 23 (30%) receiving 1250 ppm of 2,4-D (p = 0.0033). Twenty-three of 122 (19%) male rats given 2.4- D had sarcomas (p = 0.0092). Carcinomas in Male Rats Carcinomas in male rats treated with 2,4-D were seen in the endocrine system. Six of 25 male rats ingesting 25 or 125 ppm(p = 0.011)and 5 of 24 rats (21%) given 625 ppm (p = 0.022) of 2,4-D developed carcinomas (Table 1). Malignant Neoplasms at All Sites in Female Rats Eight of 20 (40%) female rats given 5 ppm (p = 0.019), 11 of 22 (50%) given 25 ppm (p = 0.058), 13 of 23 (57%) on 125 ppm (p = 0.022), 18 of 24 (75%) receiving 625 ppm (p = 0.00047), and 17 of 25 (68%) femalerats ingesting 1250 ppm of 2,4-D developed malignant neoplasms at all sites (Table 3). Altogether, malignant neoplasms were observed in 67 of 114 female rats (59%) ingesting 2,4-D (p = 0.0018). Lesions of the Lymphoreticular System in Female Rats The highest incidence of lymphosarcomas was 12 of 24 (50%) female rats given 625 ppm of 2,4-D (p = 0.00007) (Table 4). Female rats in the other 2.4- D treatment groups had 24 to 27% lymphosarcomas (p = 0.011 to p = 0.018). Thirty-one percent of all females (p = 0.00073) developed lesions of the lymphoreticular system. The findings are even more significant if hyperplasia is considered along with lymphosarcomas, i.e., 39% (p = 0.00005) of female rats ingesting 2,4-D. Female rats also had sarcomas of the uterus (see Table 3). Ji Lob O D^ 5 Carcinomas at All Sites in Female Rats Carcinomas at all sites were seen in 43% (10 of 23) (p = 0.065) of female rats on 125 ppm, 58% (14 of 24) (p = 0.0059) given 625 ppm and 52% (13 of 25) (p = 0.017) receiving 1250 ppm 2,4-D (Table 3). Carcinomas were found in the reproductive system, particularly the mammary gland, and occasionally in endocrine organs. Neoplasms of the Mammary Gland in Female Rats The incidence of mammary gland tumors were higher in treated female rats, particularly on gross examination, which is generally quite reliable. The number of rats with such tumors was less in all but one group after histologic examination (Table 5). On gross examination, 18 of 21 female mice (85%) ingesting 25 ppm (P = 0.0064) 20 of 24 rats (83%) ingesting 625 ppm (p = 0.0080) and 16 of 23 rats (70%) given 1250 ppm of 2,4-D had neoplasms of the mammary gland, compared to 10 of 22 (45%) control female rats. The findings are quite different for the numbers of histological sections examined by pathologists at FDA and sections available for my examination. Comments It is worth noting that histological examination of tissues from this study was inadequate. Microscopic neoplasms would have been overlooked at the 5 to 625 ppm doses of 2,4-D (only gross neoplasms were sectioned histologically) and also even at the highest dose because only six rats of each sex were examined in detail. Also histological sections were not available for 19 marrenary gland neoplasms described at the time of necropsy, and no exaplanation was given for this discrepancy. Despite the shortcomings of this study, it must be considered as an acceptable study. t o*-' o Jl O vj O ^ The largest numbers of rats alive were in the 25 to 1250 ppm; and the smallest number of rats alive was in the lowest dose, 5 ppm. These results suggest that rats in at least some of the 25 to 1250 ppm groups were not ingesting their diets because of toxicity. The differences in the incidences of neoplasms, therefore, would not be great. Tumors were analyzed, on the basis of 25 rats per group, and the following conclusions were made by the authors: "There is a statistically significant (p = <0.05) linear relationship between the proportion of female rats with tumors and the level of the log dose, but not the arithmetic dose, indicating that there is a tendency for the proportion of female rats with tumors to increase with the log dosage. There is also a statistically significant (p = <0.05) linear relationship between the proportion of male rats with malignant tumors and the level of arithmetic and log dose___ A comparison was made of the control group with each treatment group. Statistically significant (p = <0.05) differences were found only between the control group and the 1250 ppm dose level with respect to male rats with malignant tumors." The authors also believed that the raw data and the pathologic interpretation did not support the statistical analyses and that "a carcinogenic effect of 2,4-D has not been shown." They also stated that "additional support for this interpretation has been given by the long term study in mice (i.e., the Innes, et al. 2,4-D Mouse Study). Summary Male and female rats ingesting 2,4-D developed increased incidences of malignant neoplasms. Lymphosarcomas were increased in rats of both sexes, and neoplasms of the mammary gland in female rats. Conclusions 2,4-D is carcinogenic for male and female rats. FDA 2,4-D DOG STUDY Beagle dogs, 6 to 8 months old (3 males and 3 females per group) ingested 0, 10, 50, 100, or 500 ppm 2,4-D in the diet for 104 weeks (1-3). Organ weights were taken for brain, heart, liver, kidneys, spleen, thyroid, adrenals, and testes. Tissues from all dogs were studied grossly and microscopically. Many of the dogs ingesting 2,4-D lost weight. There were scattered lesions such as atrophy of the testes and prostate, interstitial nephritis, hemangioma of the adrenal, atrophic or cystic pituitary, atrophy of the thyroid, and hypoplasia of the bone marrow. Most of the lesions were seen in endocrine organs. Control dogs generally did not have lesions. Comments Previous experience at FDA has shown that long-term chronic dog studies should be carried out for six years or longer in order for neoplasms to develop. Since they occurred predominantly in 2,4-D treated dogs, lesions may have progressed to neoplasms had the dogs been treated for a longer period of time. Summary A two-year feeding study in dogs cannot be considered a carcinogenicity study. INNES, ET AL. 2,4-D, OR ITS ESTERS, ORAL MOUSE STUDIES T. 2,4-D The maximum tolerated dose of 2,4-D was given to two hybrid strains of mice, (C57BL/6 x C3H/Anf)F-| designated as "strain A" and (C57BL/6 x AKR)F-| 8 designated as "strain B" mice (4-5). There were 18 treated mice and 18 untreated controls of each strain and each sex. 46.4 mg/kg was given in 0.52 gelatin daily by stomach tube beginning at 7 days of age. After the mice were weaned at 28 days of age, 149 ppm of 2,4-D was mixed directly in the diet and provided ad libitum. "Strain B" male and female mice were also given 100 mg/kg, followed by 323 ppm. Treatment was continued approximately 18 months. Postmortem included thorough external examination and internal examination of the neck glands and the thoracic and abdominal cavities, with histologic examination of major organs and of all grossly visible lesions. Thyroid glands were not examined. There was no increase in neoplasms in the 2,4-D-treated mice. 2. 2,4-D Isopropyl Ester "Strain A" and "strain B" male and female mice were given 46.6 mg/kg of 2,4-dichlorophenoxyacetic acid, isopropyl ester in 0.52 gelatin daily by stomach tube. At 28 days of age the mice received 111 ppm in the diet for approximately 18 months. There were neoplasms in the lung of 4 of 18 strain A male mice (222) ingesting 2,4-D isopropyl ester compared to 2 of 17 matched controls (122) and 5 of 79 pooled male control mice (62) (p = 0.0584) (Table 6). In this study, strain A male mice receiving 2,4-D isopropyl ester developed an increased incidence of neoplasms of the lung. 3. 2,4-D Butyl Ester "Strain A" and "strain B" male and female mice received 46.4 mg/kg of 2,4-dichlorophenoxyacetic acid, butyl ester in 0.52 gelatin daily by stomach tube. At 28 days of age the mice received 149 ppm in the diet for approximately 18 months. V? JL CO- 9 Three of 18 strain A female mice (17%) and 4 of 87 pooled controls ( 5%) had reticulum cell sarcomas (p = 0.0955) (Table 7). In this study, there was an increased incidence of reticulum cell sarcoma in 2,4-D butyl ester-treated strain A female mice. 4. 2,4-D Isooctyl Ester "Strain A" and "strain B" male and female mice received 46.4 mg/kg of 2,4-dichlorophenoxyacetic acid, isooctyl ester in 0.53 gelatin by stomach tube. After 28 days they ingested 130 ppm in the diet for approximately 18 months. Neoplasms in treated mice were found mainly in the liver and lung. One strain A treated male mouse had an "angioma" of the liver and another an "angioma" of the spleen. In this study, neoplasms of the liver were slightly increased in 2,4-D isooctyl ester-treated strain A male mice and tumors of the lung in strain A female mice. Two male mice had rare tumors, "angiomas" of the liver and spleen. INNES, ET AL. 2,4-D, OR ITS ESTERS, SUBCUTANEOUS MOUSE STUDY A single subcutaneous injection of 2,4-D, or its esters, was given in the nape of the neck to two hybrid strains of mice, (C57BL/6 x C3H/Anf)F-j designated as "strain A" and (C57BL/6 x AKR)F-j designated as "strain B" at approximately the 28th day of age (4-5). There were 18 treated mice and 18 untreated controls of each strain and each sex. They were killed after approximately 18 months. Mice received the following doses: 215 mg/kg or 464 mg/kg of 2,4-D in DMSO; 100 mg/kg of 2,4-D isopropyl ester, 25.5 mg/kg of 2,4-D butyl ester or 21.5 mg/kg of 2,4-D isooctyl ester in corn oil. J> O K C J-' i j O 10 Necropsy included thorough external examination and internal examination of the neck glands and the thoracic and abdominal cavities. Histological examination was done on all major organs and of all grossly visible lesions. Reticulum Cell Sarcomas in "Strain B" Female Mice Five of 17 "strain B" female mice (3%) given 2,4-0 isooctyl ester had reticulum cell sarcomas, compared to 5 of 157 control female mice (29%) (p = 0.0009) (Table 8). Reticulum cell sarcomas were seen in 2 of 18 male mice (11%) and 0 of 161 control mice (p = 0.0096). Summary 2,4-D isooctyl ester was carcinogenic for the lymphoreticular system (reticulum cell sarcomas) in "strain B" female and male mice. COMMENTS Virtually every chemical which has been found to be carcinogenic in humans is also known to be carcinogenic in one or more other mammalian test animals (Reuber, 1978). It has been demonstrated that if a compound will produce neoplasms in any one species, it will likely produce them in another. We strongly feel that a finding of carcinogenicity in one mammalian species is therefore relevant to all mammals, including humans, even though direct evidence of danger to humans may not be available. iq 1 REFERENCES 1. Hansen, W.H., Quaife, M.L., Haberman, R.T., and Fitzhugh, O.G.: Chronic toxicity of 2,4-dichlorophenoxyacetic acid in rats and dogs. Toxicol. Appl. Pharm. 20:122-192, 1971. 2. FDA Memorandum; "Pathological Changes in Rats Fed 2,4-Dichlorophenoxy Acid for Two Years" from Robert T. Haberman to A. J. Lehman dated March 20, 1964. 3. FDA Memorandum: "Pathological Changes in Dogs Fed 2,4-Dichlorophenoxy Acid for Two Years" from Robert T. Haberman to A. J. Lehman dated November 5, 1963. 4. Innes, J.R.M., Ulland, B.M., Valerio, M.G., Petrucelli, L., Fishbein, L., Hart, E.R., Pallota, A.J., Bates, R.R., Falk, H.L., Gart, J.J., Klein, G.M., Mitchell, I., and Peters, J.: Bioassay of pesticides and industrial chemicals for tumorigenicity in mice: A preliminary note. J. Nat'l. Cancer Inst. 42:1101-1114, 1969. 5. Bionetics Research Labs, Inc.: Evaluation of carcinogenic,, teratogenic, and mutagenic activities of selected pesticides and industrial chemicals. Vol. 1 Carcinogenicity. National Technical Information Service, U.S. Department of Commerce, 1968. 1955 TABLE 1 . NUMBER OF MALE RATS INGESTING 2,4-D WITH CARCINOMAS AND SARCOMAS (FDA Rat Study) Dose (ppm) Carcinomas Sarcomas Both^ 0 0/25 (025) 1/25 (4%) 1/25 (4%) 5 2/25 (8%) 2/25 (8%) 3/25 (12%) 25 6/25 (24%) p = 0.011 4/25 (16%) p = 0.055 9/25 (36%) p = 0.0053 125 6/25 (24%) p = 0.011 6/25 (24%) p = 0.011 9/25 (36%) p = 0.0053 625 5/24 (21%) p = 0.022 4/24 (17%) p = 0.050 6/24 (25%) p = 0.043 1250 3/23 (13%) 7/23 (30%) p = 0.0033 9/23 (39%) p = 0.0033 5-1250 22/122 (18%) p = 0.012 23/122 (19%)p = 0.0092 36/122 (30%)p = 0.0038 CO CO oo oo 0.013-/ 0.057-/ a/ Some rats had both carcinoma and sarcoma and are counted only once, b/ Test for positive trend c/ Departure from linear trend I TABLE 2. NUMBER OF MALE RATS INGESTING 2,4-D WITH LYMPHOSARCOMAS AND NEUROSARCOMAS (FDA Rat Study) Dose (ppm) 0 5 25 125 625 1250 5-1250 Lymphosarcoma 0/25 (0%) 2/25 (8%) 4/25 (16%) p=0.055 5/25 (20%) p=0.025 3/24 (13%) 6/23 (26%) p=0.0082 20/122 (16%) p=0.018 0.034^ Subcutaneous Neurosaromas 0/25 (0%) 0/25 (0%) 0/25 (0%) 1/25 (4%) 1/24 (4%) 1/23 (4%) 3/122 (2%) Total 0/25 (0%) 2/25 (8%) 4/25 (16%) p=0.055 6/25 (24%) p=0.011 4/24 (17%) p=0.050 7/23 (30%) p=0.0033 23/122 (19%) p=0.0092 0.013^ a/ Test for positive trend T A B L E 3. N U M B E R O F F E M A L E R A T S I N G E S T I N G 2 , 4 - D W I T H C A R C I N O M A S A N D S A R C O M A S (FDA Rat Study) Dose (ppm) Carcinomas Sarcomas Both^/ 0 4/22 (18%) 1/22 (5%) 5/22 (23%) 5 8/20 (40%) 4/20 (20%) 8/20 (40%) p=0.019 25 7/22 (32%) 8/22 (36%) p-0.011 11/22 (50%) p=0.058 125 10/23 (43%) p=0.065 8/23 (35%) p=0.013 13/23 (57%) p=0.022 625 14/24 (58%) p=0.0059 12/24 (50%) p=0.00007 18/24 (75%) p=0.00047 1250 13/25 (52%) p=0.017 10/25 (40%) p=0.0044 17/25 (68%) p=0.0022 5-1250 52/114 (46%) p=0.013 41/114 (36%) p=0.0017 67/114 (59%) p=0.0018 0.017--^ 0.024^ 0.0022--^ 0.089^ 0.090^/ a/ Some rats had both carcinomas and sarcomas and are counted only once, b/ Test for positive trend c/ Departure from linear trend M CD CJi CD i T A B L E 4. N U M B E R O F F E M A L E R A T S I N G E S T I N G 2 , 4 - D W I T H L E S I O N S O F T H E L Y M P H O R E T I C U L A R S Y S T E M (FDA Rat Study) Dose (ppm) 0 5 25 125 625 1250 5-1250 Hyperplasia 0/22 (0%) 0/20 (0%) 2/22 (9%) 2/23 (9%) 4/24 (17%) p=0.065 2/25 (8%) p=0.28 10/114 (9%) p=0.16 Lymphosarcoma 0/22 (0%) 5/20 (25%) p=0.018 6/22 (27%) p=0.011 6/23 (26%) p=0.012 12/24 (50%) p=0.00007 6/25 (24%) p=0.016 35/114 (31%) p=0.00073 Hyperplasia and Lymphosarcoma 0/22 (0%) 5/20 (25%) p=0.018 8/22 (36%) p=0.0018 8/23 (35%) p=0.0023 16/24 (67%) p=<0.00001 8/25 (32%) p=0.0034 45/114 (39%) p=0.00005 0.039^ 0.0004-/ a/ Test for positive trend b/ Departure from linear trend CD CJ1 to T A B L E 5. N U M B E R O F F E M A L E R A T S I N G E S T I N G 2 , 4 - D W I T H N E O P L A S M S O F T H E M A M M A R Y G L A N D O N G R O S S A N D H I S T O L O G I C E X A M I N A T I O N ^ (FDA Rat Study) Dose (ppm) Grossi Histologic^/ Histologic^ 0 5 25 125 625 1250 625-1250 10/22 (45%) 8/20 (4035) 18/21 (8635) p = 0.0064 8/23 (3535) 20/24 (8335) p = 0.0080 16/23 (70%) p = 0.091 70/111( %) 0.0233^/ 0.0008^/ 10/22 (45%) 7/20 (35%) 11/21 (52%) 10/23 (43%) 14/24 (58%) 15/23 (65%) 57/111( %) 0.0334^/ 8/22 (36%) 7/20 (35%) 11/21 (50%) 9/23 (39%) 15/24 (63%) 11/23 (45%) 53/111( %) a/Corrected for survival time, i.e., time of appearance of first neoplasm of the mammary gland. b/Gross necropsy at FDA. c/Histological examination by FDA. d/Histological examination by Reuber. e/Trend for dose response curve. f/Departure from trend. I TABLE 6. NUMBERS OF MALE AND FEMALE MICE INGESTING 2,4-D ISOPROPYL ESTER WITH TUMORS OF THE LUNG Strain "A" Matched "A" Pooled "A" "B" Matched "B" Pooled "B" Dose (ppm) 0 0 111 0 0 111 Male 2/17 (12%) 5/79 ( 6%) 4/18 (22%) p = 0.0584 2/18 (11%) 9/90 (10%) 2/18 (11%) Female 1/18 (6%) 3/87 (3%) 1/18 (6%) 0/17 (0%) 3/82 (4%) 0/17 (0%) CO c/: c I TABLE 7 NUMBERS OF MALE AND FEMALE MICE INGESTING 2,4-D BUTYL ESTER WITH RETICULUM CELL SARCOMAS Strain "A" Matched "A" Pooled "A" "B" Matched "B" Pooled "B " Dose (ppm) 0 0 149 0 0 149 Male 1/17 ( 6%) 5/79 ( 635) 0/18 ( 0%) 1/18 ( 6%) 1/90 ( 1%) 2/18 (11%) p = 0.0714 Female 2/18 (11%) 4/87 ( 5%) 3/18 (17%) p = 0.0955 2/17 (12%) 3/82 ( 4%) 0/16 ( 0%) O y^ u O; o TABLE 8. NUMBER OF "STRAIN B" MALE AND FEMALE MICE GIVEN A SINGLE SUBCUTANEOUS INJECTION OF 2,4-D ISOOCTYL ESTER WITH RETICULUM CELL SARCOMAS Dose (mg/kg) 0 1 Males 0/161 ( 0) 2/18 (11%) p = 0.0096 Females 5/157 ( 3%) 5/17 (29%) p = 0.0009 t tD crt C '- air samples Although we apply dispersion staining primarily to bulk sam ple it m ay also be applied to air samples. When air samples are exam ined it is necessary to separate the sample dust from the filter. A convenient method o f eliminating the filter is low temperature ashing. Portions of filters, cut in the shape of sectors, may be low temperature ashed directly on microscope slides. During ashing the filter sector must adhere to the slide. Adherence can be accom plished by macing the filter sector on a slide that has a drop o f the 1.580 m ounting liquid applied to its surface. The mounting liquid should be spread over an area slightly arger than the filter sector. references 1. Crossmon, G.: Determination of Free Silica by Dispersion Staining Microscopical Methods. Am. Ind. Hyg. Assoc. Quart. 72:117-120 (1951). 2. Crossmon, G.: Dispersion Staining Microscopy as Applied to Industrial Hygiene. Am. Ind. Hyg. Assoc. Quart. 18:341 344(1957). 3. Brown, K.M. and W.C. McCrone: Dispersion Staining. The Microscope. 73:311-322 (1963). 4. Walter C. McCrone Associates, Inc., 2820 South Michigan Avenue, Chicago, IL60616. 5. McCrone, W.C. and J.G. Delly: The Particle Atlas. 2nd Ed., Vol. 1, pp. 97-114. Ann Arbor Science, Ann Arbor (1973). 6. R.P. Cargille Laboratories, Inc., 55 Commerce Read, Cedar Grove, NJ 07009. Mo excess in cancer deaths found in largest group ever studied for long-term effects of dioxin exposure . . . A standardized mortality analysis has been completed of a qioup of 121 workers believed to have been exposed toTCDD, one of a family of toxic impurities commonly called `"dioxin," in an industrial accident 30 years ago at Monsanto Company's Nitro, West Virginia plant. The study found no apparent excess deaths from cancer or cardiovascular disease. The total deaths observed (32) were also less than expected when compared to the national average. The medical study was co-authored by Judiih A. Zack, a Monsanto epidemiologist, and Raymond R. Suskind, M.D., Director of the Institute of Environmental Health at the University of Cincinnati Medical Center. A report of their findings will be published in the Journal of Occupational Medicine. TCDQ is sometimes formed as a trace contaminant during me production of certain industrial chemicals. One of these is the herbicide, 2,4,5-T, currently the subject of widespread controversy centering on its use in the defoliant "Agent Orange" in Vietnam and in agricultural applications in the United States and abroad. The 121 Monsanto workers involved in this study represent i : ' of the employees who were exposed in the accident and vho subsequently developed "chloracne," a skin disorder commonly associated with dioxin exposure. All 121 people 'ere traced through plant medical records, workmen's compensation files and. where applicable, official death certificates. Thus, a complete follow-up was achieved. According to the authors, the acute or short-term health effects of dioxin exposure are described in the literature but cntil now little has been known regarding the chronic or long:erm effects. They stressed that this analysis is important in that "no apparent excess of total mortality or deaths from malignant neoplasms or disease of the circulatory system was observed in a group of workers with a high peak exposure to TCDD (dioxin) followed over a period of nearly 30 years. The 1949 Nitro plant accident is the earliest recorded incident of dioxin exposure involving a population of this size. Since then, over 200 dioxin-related industrial accidents have occurred around the world, including a widely reported 1976 accident in Seveso, Italy. Accordingly, the authors stated that the Nitro group represents the best opportunity thus far, in terms of people exposed and duration of latency period since exposure, to study the long-term effects of dioxin or mortality. Despite the relatively small number of deaths involved in the study, the authors said that "by augmenting these data with the results of comparable mortality studies, the long-term effects of TCDD (dioxin) may be more definitely evaluated." This study will become a part of a larger analysis of not only mortality data but also of health information gathered last June in examinations of over 400 present and former Nitro plant employees by Dr. Suskind and his medical team from the University of Cincinnati's Institute of Environmental Health. Included in this larger study are both workers exposed during the 1949 accident and those not exposed at that time but who worked in the 2,4,5-T operation between 1948 when the unit started up and 1969 when Monsanto ceased production of the product. A control group of employees who worked in other areas of the plant during that time frame is also being studied for comparative purposes. This extensive medical investigation is being directed by Dr. Suskind. His findings are expected to be published later this year or early next year. t'inan industrial Hygiene Association JOURNAL (4l> U i. 73 105ST- ~ - 2- ABSTRACT The toxicological significance of exposures to synthetic chemicals is examined in the context of exposures to-naturally-occurring chemicals. We calculate that 99.99% (by weight) of the pesticides in the American diet are chemicals that plants produce to defend themselves. Only 52 natural pesticides have been tested in high-dose animal cancer tests, and about half (27) are rodent carcinogens; these 27 are shown to be present in many common foods. We conclude that natural and synthetic chemicals are equally likely to be positive in animal cancer tests. We also conclude that at the low doses of most human exposures the comparative hazards of synthetic pesticide residues are insignificant. KkJ -3 - Toxicological examination of synthetic chemicals such as pesticides and industrial pollutants, without similar examination of the chemicals in the natural world to use for comparison, has generated an imbalance in both data and perception about potential hazards to humans (1-6). In this and two accompanying papers (7,8), we try to redress this imbalance and discuss in detail one major group of natural chemicals in our diet -- nature's pesticides. About half of all chemicals (whether natural or synthetic) tested chronically in animal cancer tests at the maximum tolerated dose (MTD) are carcinogens (7,9-14). The MTD of the test chemical is a near-toxic dose that can cause chronic mitogenesis, often as a result of cell killing (7). We have argued that mitogenesis increases mutagenesis, and therefore that a high percentage of all chemicals might be expected to be carcinogenic when tested chronically at the MTD (7). A high proportion of both natural and synthetic test chemicals is positive for carcinogenicity. Natural chemicals constitute the vast bulk of chemicals in the human diet, and therefore should be used as a reference for evaluating possible carcinogenic hazards from synthetic chemicals. In recent years, we have compared the possible hazards of various rodent carcinogens, using the HERP ratio: human exposure/rodent potency (1,6,). It should be emphasized that as the understanding of carcinogenesis mechanisms improves, these comparisons can be refined but they cannot provide a direct estimate of human hazard. This paper does not extend the HERP comparisons (1) because our purpose is different and space does not allow a proper analysis. 19569 i 4- - NATURE'S PESTICIDES: MUTAGENICITY AND CARCINOGENICITY "Plants are not just food for animals. . . The world is not green. It is colored lectin, tannin, cyanide, caffeine; aflatoxin, and canavanine" [Janzen (16)]. Dietary Pesticides are 99.99% All Natural. Nature's pesticides are one important subset of natural chemicals. Plants produce toxins to protect themselves against fungi, insects, and animal predators (5,16-23). Tens of thousands of these natural pesticides have been discovered, and every species of plant analyzed contains its own set of perhaps a few dozen toxins. When plants are stressed or damaged, such as during a pest attack, they may greatly increase their natural pesticide levels, occasionally to levels that can be acutely toxic to humans. We estimate that Americans eat about 1.5 g of natural pesticides per person per day, which is about 10,000 times more than they eat of manmade pesticide residues (see below). As referenced in this paper (see 16-21 and Legends to Table 1, 2) there is a very large literature on natural toxins in plants and their role in plant defenses. The human intake of these toxins varies markedly with diet and would be higher in vegetarians. Our estimate of 1.5 g of natural pesticides per person per day is based on the content of toxins in the major plant foods (e.g. 13 g roasted coffee per person per day contains about 765 mg of chlorogenic acid, neochlorogenic acid, caffeic acid, and caffeine; see (22,23) and Table 2). Phenolics from other plants are estimated to contribute another several hundred mg of toxins. Flavonoids and glucosinolates account for several hundred mg; potato and tomato toxins may 18570 - 5- contribute another hundred, and saponins from legumes another hundred. Grains such as white flour and white rice contribute veiy little, but whole wheat, brown rice, com (maize) may contribute several hundred mg more. The percentage of a plant's weight that is toxin varies, but a few percent of dry weight is a reasonable estimate: e.g. 1.5% of alfalfa sprouts is canavanine and 4% of coffee beans is phenolics. However, the percentage in some plant cultivars is lower, e.g. potatoes and tomatoes. Concentrations of natural pesticides in plants are usually measured in pans per thousand or million (16-23) rather than parts per billion (ppb), the usual concentration of synthetic pesticide residues or of water pollutants (1,24). We estimate that humans ingest roughly 5,000 to 10,000 different natural pesticides and their breakdown products (16-23). For example, Table 1 shows 49 natural pesticides (and metabolites) that are ingested when eating cabbage and indicates how few have been tested for carcinogenicity or clastogenicity. Lima beans contain a completely different array of 23 natural toxins that, in stressed plants, range in concentration from 0.2 to 33 parts per thousand fresh weight; none appears to have been tested yet for carcinogenicity or teratogenicity (19). Many Leguminosae contain canavanine, a toxic arginine analog that, after being eaten by animals, is incorporated into protein in place of arginine. Feeding alfalfa sprouts (1.5 % canavanine dry weight) or canavanine to monkeys causes a lupus erythematosus-like syndrome (44). Lupus in man is characterized by a defect in the immune system that is associated with autoimmunity, antinuclear antibodies, chromosome breaks, and various types of pathology. The toxicity of non-food plants is well known; plants are among the most commonly i.O 'ij i ingested poisonous substances for children under five. Surprisingly few plant toxins have been tested for carcinogenicity (10- 13,45). Among 1052 chemicals tested in a t le a s t o n e sp e c ie s in chronic cancer tests, only 52 are naturally occurring plant pesticides (10-13). Among these, about half (27/52) are carcinogenic.^ Even though only a tiny proportion of plant toxins in our diet has been tested so far, the 27 natural pesticides that are rodent carcinogens are present in the following foods: anise, apples, apricots, bananas, basil, broccoli, Brussels sprouts, cabbage, cantaloupe, caraway, carrots, cauliflower, celery, cherries, cinnamon, cloves, cocoa, coffee, collard greens, comfrey herb tea, currants, dill, eggplant, endive, fennel, grapes, grapefruit juice, guava, honey, honeydew melon, horseradish, kale, lentils, lettuce, mangoes, mushrooms, mustard, nutmeg, orange juice, parsley, parsnips, peaches, pears, peas, black pepper, pineapples, plums, potatoes, radishes, raspberries, rosemary, sesame seeds, tarragon, tea, tomatoes, and turnips. Thus, it is probable that almost every fruit and vegetable in the supermarket contains natural plant pesticides that are rodent carcinogens. The levels of these 27 rodent carcinogens in the above plants are commonly thousands of times higher than the levels of manmade pesticides. Table 2 shows a variety of natural pesticides that are rodent carcinogens occurring in the ppm range in plant foods. The catechol-type phenolics such as tannins, and caffeic acid and its esters (chlorogenic and neochlorogenic acids), are more widespread in plant species than other natural pesticides (e.g., Table 1 and 2). It may be that these phenolics have an antimicrobial role analogous to the respiratory burst of -7- oxygen radicals from mammalian phagocytic cells. The phenolics oxidize when a plant is wounded, yielding a burst of mutagenic oxygen radicals, e.g. the browning when an apple is cut. Caution is necessary in interpreting'the implications of the occurrence in the diet of natural pesticides that are rodent carcinogens. It is not argued here that these dietary exposures are necessarily of much relevance to human cancer. Indeed, a diet rich in fruit and vegetables is associated with lower cancer rates (86,87). This may be because ancarcinogenic vitamins and antioxidants come from plants (86,87). What is important in our analysis is that exposures to natural rodent carcinogens may cast doubt on the relevance of far lower levels of exposures to synthetic rodent carcinogens. Residues of Manmade Pesticides. A National Research Council report has discussed the regulation of synthetic pesticides that are rodent carcinogens, but ignored natural pesticides (88). The U.S. Food and Drug Administration (FDA) has assayed food for 200 chemicals including the manmade pesticide residues thought to be of greatest importance and the residues of some industrial chemicals such as polychlorinated biphenyls (PCBs) (24). FDA found residues for 105 of these chemicals: the U.S. intake of the sum of these 105 chemicals averages about 0.09 mg per person per day, which we compare to 1.5 g of natural pesticides (i.e. 99.99% natural).^ Other analyses of synthetic pesticide residues are similar (90). About half (0.04 mg) of this daily intake of synthetic pesticides is composed of 4 chemicals (24) that were not carcinogenic in rodent tests: ethylhexyl diphenyl phosphate, chlorpropham, malathion, and dicloran (10,89). Thus, the intake of rodent carcinogens from synthetic ibo 3A o p* !; r\ i - 8- residues is only about 0.05 mg a day (averaging about 0.06 ppm in plant food) even if one assumes that all the other residues are carcinogenic in rodents (which is unlikely). Cooking food. The cooking of food is also a major dietary source of potential rodent carcinogens. Cooking produces about 2 g per person per day of mostly untested burnt material that contains many rodent carcinogens, e.g. polycyclic hydrocarbons (81,91) heterocyclic amines (92,93), furfural (22,23), nitrosamines and nitroaromatics (1,94), as well as a plethora of mutagens (9195). Thus, the number and amounts of carcinogenic (or total) manmade pesticide residues appear to be minimal compared to the background of naturally-occurring chemicals in the diet. Roasted coffee, for example, is known to contain 247 volatile chemicals (22); 11 have been tested chronically and 8 are rodent carcinogens (10-13); caffeic acid, a non-volatile rodent carcinogen, is also present (Table 2). A typical cup of coffee contains at least 9 mg (40 ppm) of rodent carcinogens (mostly caffeic acid, catechol, furfural, and hydrogen peroxide)(Table 2). The evidence on coffee and human health has been recently reviewed, and the evidence to date is not sufficient to show that coffee is a risk factor for cancer in humans (81,86). The same caution about the implications for humans of rodent carcinogens in the diet that were discussed above for nature's pesticides apply to coffee and the products of cooked food. Clastogenicity and Mutagenicity Studies. Results from in v itro studies also indicate that the natural world should not be ignored and that positive results are commonly observed in high-dose protocols. For example, Ishidate et 4Jl Q K *y -9- a l (26) reviewed experiments on the ciastogenicity (ability to break chromosomes) of 951 chemicals in m a m m a lia n cell cultures. Of these 951 chemicals, we identified 72 as natural plant pesticides: 35 (48%) were positive for ciastogenicity in at least one test. This is s im ila r to the results for the remaining chemicals, of which 467/879 (53%) were positive in at least one test. Of particular interest are the levels at which some of the carcinogenic plant toxins in Table 2 were clastogenic (26): a) Allyl isothiocyanate was clastogenic at a concentration of 0.0005 ppm, which is about 200,000 times less than the concentration of sinigrin, its glucosinolate, in cabbage. Allyl isothiocyanate was among the most potent chemicals in the compendium (26), and is also effective at unusually low levels in transforming (96) and mutating animal cells (30). (See also the discussion of cancer tests in Table 1.) b) Safrole was clastogenic at a concentration of about 100 ppm, which is 30 times less than the concentration in nutmeg, and roughly equal to the concentration in black pepper. The rodent carcinogens safrole and estragole, and a number of other related dietary natural pesticides that have not been tested in animal cancer tests, have been shown to produce DNA adducts in mice (97). c) Caffeic acid was clastogenic at a concentration of 260 and 500 ppm, which is less than its concentration in roasted coffee beans and close to its concentration in apples, lettuce, endive, and potato skin. Chloiogenic acid, a precursor of caffeic acid, was clastogenic at a concentration of 150 ppm, which is 100 times less than its concentration in roasted coffee beans and similar to its concentration in apples, pears, plums, peaches, cherries and apricots. Chlorogenic acid and its metabolite caffeic acid are also mutagens (Table 1). 10 The genotoxic activity of coffee to mammalian cells has been demonstrated (98). The carcinogenicity and mutagenicity of many plant pesticides have been recently reviewed (45): 5- and 8-methoxypsoralen are light activated mutagens (17); benzyl acetate and ethyl acrylate mutate mouse lymphoma cells (30). Plant phenolics such as caffeic acid, chlorogenic acid, and tannins (esters of gallic acid) have been reviewed for their mutagenicity and anti-mutagenicity, clastogenicity, and carcinogenicity (99). -11 - ACKNOWLEDGMENT We arc indebted to R. Peto, N. B. Manley, T. H. Slone, C. Wehr, R. Beier, L. W. Wattenberg, R. Hall, T. Jukes, G. R. Fenwick, J. Caldwell, J. Duke, C. VanEtten, D. Freedman, R. Prokopy, and N. Ito. This work was supported by National Cancer Institute Outstanding Investigator Grant CA39910, by National Institute of Environmental Health Sciences Center Grant ES01896; Contract No. DE-AC03-76SF00098: Director, Office of Energy Research, Office of Health and Environmental Research, Division of the U.S. Department of Energy. We dedicate this paper to the memory of William Havender. - 12 - REFERENCES 1. Ames, B. N., Magaw,R.& Gold, L. S. (1987) S cien ce 2 3 6 , 271-280. 2. Ames, B. N. & Gold,L. S. (1987) S c ien ce 238, 1634. 3. Ames, B. N., Magaw,R.& Gold, L. (1987) S cien ce 237, 1283-1284. 4. Ames, B. N., Magaw,R. & Gold, L. S. (1987) S cien ce 237, 235. 5. Ames, B. N. (1983) S cien ce 221, 1256-1264. 6. Ames, B. N. & Gold, L. S. (1989) S cien ce 244, 755-757. 7. Ames, B. N. & Gold. L. S. P ro c . N a tl. A c a d . S ci. USA, this issue. 8. Ames, B.N., Profet, M. & Gold, L.S. P ro c . N a tl. A ca d . S ci. U SA , this issue. 9. Gold, L. S., Bernstein, L., Magaw, R. & Slone, T. H. (1989) E n viron . H ealth P ersp ect. 81, 211-219. 10. Gold, L. S., Sawyer, C. B., Magaw, R., Backman, G. M., de Veciana, M., Levinson, R., Hooper, N. K., Havender, W. R., Bernstein, L., Peto, R., Pike, M. C. & Ames, B. N. (1984) E n viron . H ea lth P e rsp e c t. 58, 9-319. 11. Gold, L. S., de Veciana, M., Backman, G. M., Magaw, R., Lopipero, P., Smith, M., Blumenthal, M., Levinson, R., Bernstein, L. & Ames, B. N. (1986) E n viron . H ea lth P e r s p e c t. 67, 161-200. 12. Gold, L. S., Slone, T. H., Backman, G. M., Magaw, R., Da Costa, M., Lopipero, P., Blumenthal, M. & Ames, B. N. (1987) E n viron . H ea lth P e rsp e c t. 74, 237-329. 13. Gold, L. S., Slone, T. H., Backman, G. M., Eisenberg, S., Da Costa, M., Wong, M., Manley, N. B., Rohrbach, L. & Ames, B. N. (1990) E n viron . H ea lth P e r s p e c t., 84, 215-285. - 13 - 14. Gold, L. S., Slone, T. H. & Bernstein, L. (1989) E n viron . H ea lth P e rsp e c t. 79, 259-272. 15. Haseman, J.K. (1985) F und. A p p l. T o x ic o l. 5, 66-78. 16. Janzen, D. H. (1977) A n n. M is s o u r fR o t. G a rd . 64, 706-736, 1977. 17. Beier, R. C. (1990) in R e v ie w s o f E n v iro n m en ta l C o n ta m in a tio n a n d T o x ic o lo g y , ed. Ware, G. W., (Springer-Verlag, New York), pp. 47-137. 18. Rosenthal, G. A. & Janzen, D. H., eds. (1979) H e r b iv o re s : T h eir In te ra c tio n w ith S e c o n d a ry P la n t M e ta b o lite s, (Academic Press, New York). 19. Green, M. B. & Hedin, P. A., eds. (1986) in N a tu ra l R e sista n c e o f P la n ts to P e s ts : R o le s o f A lle lo c h e m ic a ls, ACS Symposium 296, (American Chemical Society, Washington, D.C.). 20. VanEtten, H. D., Matthews, D. E. & Matthews, P. S. (1989) Annu. R ev. P h y to p a th o l. 27, 143-165, 1989. 21. Watson, D. H., ed. (1987) N a tu ra l T o x ic a n ts in F o o d . (VCM Veriagsgesellschaft, Weinheim, Germany). 22. Maarse, H. & Visscher, C. A., eds. (1983-1988) V o la tile c o m p o u n d s in F o o d s, (CIVO-TNO Zeist, The Netherlands), Vols. 1-7. 23. Stofberg, J. & Grundschober, F. (1987) P e rfu m e r & F la v o r is t 12, 27- . 24. Gunderson, E. L. (1988) J . A sso c . Off. A n a l. C h em . 71, 1200-1209. 25. Stich, H. F., Rosin, M. P,, Wu, C. H. & Powrie, W. D. (1981) M u ta t. R es. 90, 201-212. 26. Ishidate, Jr., M., Hamois, M. C. & Sofuni, T. (1988) M u tat. R es. 195, 151-213. 1 CW Q 9O - 14 - 27. Ariza, R. R., Dorado, G., Barbancho, M. & Pueyo, C. (1988) M utat. R es. 201, 89-96. 28. Fung, V. A., Cameron, T. P., Hughes, T. J., Kirby, P. E. & Dunkel, V. C. (1988) M u tat. R es. 204, 219-228. 29. Hanham, A. F., Dunn, B. P. & Srich, H. F. (1983) M u tat. R es. 116, 333339. 30. McGregor, D. B., Brown, A., Cattanach, P., Edwards, I., McBride, D., Riach, C. & Caspary, W. J. (1988) E n viron . M ol. M u tagen . 12, 85-154, 1988. 31. National Toxicology Program (1982) C a rc in o g e n e sis B io a ssa y o f A lly l Isoth iocyan ate (CAS N o. 57-0 6 -7 ) in F 344/N R ats an d B 6 C 3 F j M ice (G a v a g e S tu d y), Technical Report 234, NIH Pub. No. 83-1790, National Toxicology Program, NIH, (Research Triangle Park, NC). 32. Huff, J. E., Eustis, S. L. & Haseman, J. K. (1989) C a n c e r a n d M e ta sta sis R e v ie w s 8, 1-21. 33. Morse, M. A., Wang, C.-X., Amin, S. G., Hecht, S. S. & Chung, F.-L. (1988) C a rc in o g e n e sis 9, 1891-1895. 34. National Toxicology Program (1989) D r a ft T e c h n ic a l R e p o rt: T o x ic o lo g y a n d C arcin ogen esis Studies o f d -C arvon e in B 6 C 3 F j M ice a n d T oxicology S tu d ie s in F3 4 4 /N R a ts, Technical Report 381, NIH Pub. No. 90-2836, National Toxicology Program, NIH, Research Triangle Park, NC. 35. Wakabayashi, K., Suzuki, M., Sugimura, T. & Nagao, M. (1989) P ro ceed in g s o f the 48th Annual M eeting o f the Japanese C ancer A s so c ia tio n , October 1989, Nagoya, Japan, Abstract No. 284. ^ O r ' P C\ - 15 - 36. Ito, N. & Hirose, M. (1987) Jpn . J. C a n ce r R e s. (G a n n ) 78, 1011-1026. 37. Hirose, M., Fukushima, S., Shirai, T., Hasegawa, R., Kato, T., Tanaka, H., Asakawa, E. & Ito, N. (1990) Jpn . J. C a n c e r R e s. 81, 207-212. 38. VanEtten, C. H. & Tookey, H. L. (i979) in H e rb iv o re s: T h e ir In teraction w ith S e c o n d a ry P la n t M e ta b o lite s, eds. Rosenthal, G. A. & Janzen, D. H., (Academic Press, New York), Ch. 13, pp. 471-500. 39. Fenwick, G. R., Heaney, R. K. & Mullin, W. J. (1983) in C R C C rit. R ev. F o o d S ci. N u tr. 18, 123-201. 40. McDanell, R., McLean, A.E.M., Hanley, A.B. & Fenwick, G.R. (1988) F d . C h em . T ox. 26, 59-70. 41. Nishie, K. & Daxenbichler, M. E. (1980) F d. C o sm e t. T o x ico l. 18, 159172. 42. Nishie, K. & Daxenbichler, M. E. (1982) F d. C h em . T oxic. 20, 279-280. 43. Villasenor, I. M., Lim-Sylianco, C. Y., Dayrit, F. (1989) M u ta t. R es. 224, 209-212. 44. Malinow, M. R., Bardana, Jr., E. J., Pirofsky, B., Craig, S. & McLaughlin, P. (1982) S c ie n c e 216, 415-417. 45. Hirono, I. (ed.) (1987) N a tu r a lly O c c u rrin g C a rc in o g e n s o f P la n t O rig in : Toxicology, P athology and Biochem istry, B ioactive M olecules, (Kodansha/Elsevier Science Publishers B.V., Tokyo/Amsterdam), Vol. 2. 46. International Agency for Research on Cancer (1986) IA R C M o n o g ra p h s on the E valuation o f C arcinogenic R isks to H um ans: Som e N aturally O ccurring an d Synthetic F ood Com ponents, F urocoum arins and U ltr a v io le t R a d ia tio n , vol. 40. (International Agency for Research on - 16 - Cancer, Lyon, France) 47. National Toxicology Program (1989) T o x ic o lo g y a n d C a rc in o g e n e sis Studies o f 8-M ethoxypsoralen (C A S N o. 2 9 8 -8 1 -7 ) in F 344/N R ats (G a v a g e S tu d ies), Technical Report 359, NIH Pub. No. 89-2814. (National Toxicology Program, Research Triangle Park, NC). 48. McManus, B. M., Toth, B. & Patil, K. D. (1987) L ab. In vest. 57, 78-85. 49. Toth, B. (1986) A n tica n cer R es. 6, 917-920. 50. National Toxicology Program (1988) T o x ic o lo g y a n d C a rc in o g e n e sis Studies o f d-U m on en e (CAS N o. 5 9 8 9 -2 7 -5 ) in F 344/N R ats an d B 6 C 3 F / Alic e , G a v a g e S tu dies, Technical Report 347, Peer Review Draft, April 1988. (National Toxicology Program, Research Triangle Park, NC). 51. Miller, E. C., Swanson, A. B., Phillips, D. H., Fletcher, T. L., Liem, A. & Miller, J. A. (1983) C a n ce r R es. 43, 1124-1134. 52. National Toxicology Program (1986) T o x ic o lo g y a n d C a rc in o g e n e sis Studies o f B enzyl A cetate (CAS N o. 140-11-4) in F344IN R ats and B 6 C 3 F j M ic e (G a v a g e S tu d ies), Technical Report 250, NIH Pub. No. 862506, (National Toxicology Program, Research Triangle Park, NC). 53. National Toxicology Program (1989) T o x ic o lo g y a n d C a rc in o g e n e sis Studies o f a-M ethylbenzyl A lcoh ol in F 344/N R a ts an d B 6C 3F M ice (G a v a g e S tu d ies), Technical Report No. 369, Galley Draft. (National Toxicology Program, Research Triangle Park, NC). 54. Hirose, M., Masuda, A., Imaida, K., Kagawa, M., Tsuda, H. & Ito, N. (1987) J p n . J. C a n c e r R es. (G a n n ) 78, 317-321. 19582 - 17 - 55. Beier, R. C , Ivie, G. W., Oertli, E. H., & Holt, D. L. (1983) F d . C hem . T oxicol. 21, 163-165. 56. Chaudhary, S. K., Ceska, O., Ttu, C , Warrington, P. J., Ashwood-Smith, M. J. & Poulton, G. A. (1986) P la n ta M e d ic a 6, 462-464. 57. Ivie, G. W., Holt, D. L. & Ivey, M. C. (1981) S cien ce 213, 909-910. 58. Berkley, S. F., Hightower, A. W., Beier, R. C., Fleming, D. W., Brokopp, C. D., Ivie, G. W. & Broome, C. V. (1986) A nn. Intern. M e d ., 105, 351355. 59. Seligman, P. J., Mathias, C. G. T., O'Malley, M. A., Beier, R. G , Fehrs, L. J., Serrill, W. S. & Halperin, W. E. (1987) A rch . D e rm a to l. 123, 1478-1482. 60. Carlson, D. G., Daxenbichler, M. E., VanEtten, C. H., Kwolek, W. F. & Williams, P. H. (1987) J. A m e r. S oc. H o rt. S ei. 112, 173-178. 61. Schreier, P., Drawert, F. & Heindze, I. (1979) C h em . M ik ro b io l. T ech n ol. L e b e n sm . 6, 78-83. 62. Engel, K. H. & Tressl, R. (1983) J. A g ric . F o o d C h em . 31, 796-801. 63. Hasselstrom, T., Hewitt, E. J., Konigsbacher, K. S. & Ritter, J. J. (1957) A g ric. F o o d C hem . 5, 53-55. 64. Hecker, E. (1981) J. C a n c e r R es. C lin . O n co l. 99, 103-124. 65. Miura, Y., Ogawa, K. & Tabata, M. (1987) P la n ta M e d ic a 53, 95-96. 66. Archer, A. W. (1988) J . C h ro m a to g r. 438, 117-121. 67. Concon, J. M., Swerczek, T. W. & Newburg, D. S. (1979) N u tr. C a n ce r 1 (Spring), 22. - 18 - 68. Ohta, H., Kinjo, S. & Osajima, Y. (1987) J. C h ro m a to g r. 409, 409-412. 69. Wootton, M., Edwards, R. A., Faraji-Haremi, R. & Williams, P. J. (1978) J. A p ic u ltu ra l R es. 17, 167-172. 70. Luo, S. J., Gue, W. F. & Fu, H. J. (1988) D ev. F o o d S ei. 17, 191-199. 71. Karawya, M. S., Hashim, F. M. & Hifnawy, M. S. (1974) J. A g ric. F o o d C h em . 22, 520-522. 72. Risch, B. & Herrmann, K. (1988) Z. L eben sm . U n ters. F o rsc h . 186, 225- 263. 73. Schmidtlein, H. & Herrmann, K. (1975) Z. L eb en sm . U n ters. F o rsch . 159, 255-263. 74. Mller, B. & Herrmann, K. (1983) P h yto ch em istry 22, 477-481. 75. Mosel, H. D. & Herrmann, K. (1974) Z. L eben sm . U n ters. F o rsch . 154, 6- 11. 76. Schfers, F. I. & Herrmann, K., (1982) Z. L eben sm . U n ters. F o rsch . 175, 117-121. 77. Winter, M., Brandi, W. & Herrmann, K. (1987) Z. L eb en sm . U n ters. F o rsch . 184, 11-16. 78. Herrmann, K. (1978) Z. L eb en sm . U n ters. F orsch . 167, 262-273. 79. Sthr, H. & Herrmann, K., (1975) Z. L eb en sm . U n ters. F o rsc h . 159, 219- 224. 80. Schuster, B,, Winter, M. & Herrmann, K. (1986) Z. N a tu rfo rsch . 41c, 511-520. 81. Clarke, R. J. & Macrae, R., eds. (1988) C o ffee, vols. 1-3, (Elsevier Appl. Sei.: New York), Vols. 1-3. _i. t-* U1 - 19 - 82. Baltes, W. (1979) in 8 th In te rn a tio n a l S cien tific C o llo q u iu m on C o ffee, Abidjan, Nov. 1977, pp. 85-96. 83. Tressl, R., Bahn, D., Kppler, H. & Jensen, A. (1978) Z. L eb en sm . U n ters. F o rsch . 167, 111-114. 84. Rahn, W. and Knig, W. A. (1978) J. H igh R e so lu tio n C h ro m a to g r. & C h ro m a to g r. C om m on . 1002, 69-71. 85. Fukuda, Y., Nagata, M., Osawa, T. & Namiki, M. (1986) A g ric . B io l. C h em . 50, 857-862. 86. National Research Council (1989) D ie t a n d H ealth , Im p lic a tio n s f o r R ed u cin g C h ron ic D is e a s e R isk, (National Academy Press, Washington, D.C.). 87. National Research Council (1982) D ie t, N u tritio n , a n d C a n ce r, (National Academy Press, Washington, D.C.). 88. National Research Council, Board on Agriculture (1987) R eg u la tin g P e s tic id e s in F o o d , (National Academy Press, Washington, D.C.). 89. Treon, J., Dutra, F. & Cleveland, F. (1953) A rch . Ind. H yg . O ccu p . M ed . 8, 170-184. 90. Nigg, H.N., Beier, R.C., Carter, O., Chaisson, G , Franklin, C , Lavy, L., Lewis, R., Lombardo, P., McCarthy, J.F., Maddy, K., Moses, M., Norris, D., Peck, C., Skinner, K. & Tardiff, R.G. (1990) in P e s tic id e s '.H ea lth E ffects, eds. Baker, S., Wilkinson, C , (Princeton Scientific Publishing, Princeton, NJ). 91. Furihata, C. & Matsushima, T. (1986) A nn. R ev. N u tr. 6, 67-94. Gy O^ Op 11 - 20 - 92. Sugimura, T. (1988) M m a t. R es. 205, 33-39. 93. Takayama, S., Nakatsuru, Y. & Sato, S. (1987) Jpn. J. C a n c e r R es. (G an n ) 78, 1068-1072. 94. Beije, B. & Mller, L. (1988) M u ta t. R es. 196, 177-209. 95. E n viron . H ea lth P e r s p e c t. (1986) Dietary Mutagens (Special Issue), 67. 96. Kasamaki, A., Yasuhara, T. & Urasawa, S. (1987) J. T o x ico l. S c ien ce s 12, 383-396. 97. Randerath, K., Randerath, E., Agrawal, H. P., Gupta, R. C , Schurdak, M. E. & Reddy, V. (1985) E n v. H ea lth P e rsp e c t. 62, 57-65. 98. Tucker, J. D,, Taylor, R. T., Christensen, M. L., Strout, C. L. & Hanna, M. L. (1989) M u ta g en esis 4, 343-348. 99. Stich, H. F. & Powrie, W. D. (1982) in C a rcin o g en s a n d M u ta g en s in the E n viro n m en t, V ol. I: F o o d P ro d u c ts, ed. Stich, H. (CRC Press, Boca Raton, FL), ch. 11, pp. 135-145. I95CS - 21 - Table 1. C abbage: 49 N atural P esticides and M etabolites Glucosinolates: 2-propenyl glucosinolate (sinigrin) , 3- methyl-thio-propyl glucosinolate, 3-methyl-sulfinylpropyl glucosinolate, 3-butenyl glucosinolate, 2hydroxy-3-butenyl glucosinolate, 4-methyl-thio-butyl glucosinolate, 4-methyl-sulfinyl-butyl glucosinolate, 4- methylsulfonyl-butyl glucosinolate, benzyl glucosinolate, 2-phenyl-ethyl glucosinolate, propyl glucosinolate, butyl glucosinolate Indole Glucosinolates And Related Indoles: 3-indolylmethyl glucosinolate (glucobrassicin), l-methoxy-3- * indolylmethyl (neoglucobrassicin), indole-3-carbinol , indole-3-acetonitrile, 3,3'-diindolylmethane * Isothiocyanates And Goitrin: allyl isothiocyanate , 3- methyl-thio-propyl isothiocyanate, 3-methylsulfinyl-propyl isothiocyanate, 3-butenyl isothiocyanate, 5-vinyloxazolidine-2-thione (goitrin), 4- methylthiobutyl isothiocyanate, 4-methylsulfinylbutyl isothiocyanate, 4-methylsulfonyl butyl isothiocyanate, 4-pentenyl isothiocyanate, benzyl isothiocyanate, phenylethyl isothiocyanate Cyanides: l-cyano-2,3-epithiopropane, l-cyano-3,4epithiobutane, l-cyano-3,4-epithiopentane, th r e o -1cyano-2-hydroxy-3,4-epithiobutane, e ryth ro - 1-cyano-2- - 22 - hydroxy-3,4-epithiobutane, 2-phenylpropionitrile, allyl * cyanide , l-cyano-2-hydroxy-3-butene, l-cyano-3methylsulfinylpropane, l-cyano-4-methylsulfinylbutane * Terpenes: menthol, neomenthol, isomenthol, carvone Phenols: 2-methoxyphenol, 3-caffoylquinic acid ** (chlorogenic acid) , 4-caffoylquinic acid , 5-caffoyl * quinic acid (neochlorogenic acid) , 4-p-coumaroylquinic acid, 5-p-coumaroylquinic acid, 5-feruloylquinic acid -23 - Legend to Table 1 * Discussed below; all others untested. C la s to g e n ic ity : Chlorogenic acid (25) and allyl isothiocyanate are positive (26). Chlorogenic acid and its metabolite caffeic acid are also mutagens (27-29), as is allyl isothiocyanate (30). C a rc in o g e n ic ity : Allyl isothiocyanate induced papillomas of the bladder in male rats (a neoplasm that is unusually rare in control rats) and was classified by NTP as carcinogenic. There was no evidence of carcinogenicity in mice: however, NTP indicated "the mice probably did not receive the MTD" (31,32). Sinigrin (the glucosinolate, i.e., thioglycoside of allyl isothiocyanate) is co-carcinogenic for the rat pancreas (33). Carvone is negative in mice (34). Indole acetonitrile has been shown to form a carcinogen, nitroso indole acetonitrile, in the presence of nitrite (35). Caffeic acid is a carcinogen (36,37) and clastogen (25) and is a metabolite of its esters 3-, 4-, and 5-caffoylquinic acid (chlorogenic and neochlorogenic acid). M e ta b o lite s: Sinigrin gives rise to allyl isothiocyanate on eating raw cabbage (e.g., coleslaw); in cooked cabbage it also is metabolized to allyl cyanide, which is untested. Indole carbinol forms dimers and trimers on ingestion, which mimic dioxin (TCDD) (8). O ccu rren ce (18,21,38,40). T o x ic o lo g y : The mitogenic effects of goitrin (which is goitrogenic) and various organic cyanides from cabbage suggest that they may be potential carcinogens (41,42). Aromatic cyanides related to those from cabbage have been shown to be mutagens and are metabolized to hydrogen cyanide and potentially mutagenic aldehydes (43). 19589 Table 2. Natural Pesticide Carcinogens in Food Plant Food Concentrauon Rodent Carcinogen (ppm) Parsley 5- and 8-methoxypsoralen 14 Parsnip, cooked tt 32 Celery 0.8 Celery, new culrivar " 6.2 Celery, stressed ti 25 Mushrooms p-hydrazinobenzoate 11 ft glutamyl-p-hydrazinobenzoate 42 Cabbage sinigrin (allyl isothiocyanate) 35-590 Collard greens H 250-788 Cauliflower I 12-66 Brussels sprouts It 110-1,560 Mustard (brown) " 16,000-72,000 Horseradish M 4,500 Orange juice limonene 31 Mango 40 Pepper, black M 8,000 Basil estragole 3,800 Fennel M 3,000 Nutmeg safrole 3,000 Mace N 10,000 Pepper, black n 100 Pineapple ethyl acrylate 0.07 Sesame seeds (heated oil) sesamol 75 Cocoa a-methylbenzyl alcohol 1.3 Basil benzyl acetate 82 Jasmine tea tf 230 Honey ti 15 Coffee (roasted beans) catechol 100 Apple, Carrot, caffeic acid 50-200 Celery, Cherry, Eggplant, Endive, Grapes, Lettuce, Pear, Plum, Potato Absinthe, Anise, Basil, > 1,000 Caraway, Dill, Marjoram, Rosemary, Sage, Savory Tarragon, Thyme Coffee (roasted beans) Apricot, " 1,800 ** chlorogenic acid 50-500 Cherry, Peach, Plum (caffeic acid) Coffee (roasted beans) 21,600 ** Apple, neochlorogenic acid 50-500 Apricot, Broccoli, (caffeic acid) Brussels sprouts, Cabbage, Cherry, Kale, Peach, Pear, Plum Coffee (roasted beans) 11,600 - 26 - Legend to Table 2 Sinigrin is a co-carcinogen (33) and is metabolized to the rodent carcinogen allyl isothiocyanate, although,no adequate test has been done on sinigrin itself. The proportion converted to allyl isothiocyanate or to allyl ** cyanide depends on food preparation (38-40). Chlorogenic and neochlorogenic acid are metabolized to the carcinogens caffeic acid and catechol (a metabolite of quinic acid), but have not been tested for carcinogenicity themselves. The clastogenicity and mutagenicity of the above compounds are referenced in Table 1. C a rcin o g en refs are in (10-13) and the following: 5-methoxypsoralen (light-activated), 8-methoxypsoralen (47), p-hydrazinobenzoate and glutamyl-phydrazinobenzoate (48,49), allyl isothiocyanate (31,32), D-limonene (50), estragle and safrole (45,51), ethyl acrylate, benzyl acetate (52), amethylbenzyl alcohol (53), caffeic acid (37), sesamol (37), catechol (37). C o n cen tra tio n refs.: 5-, 8-methoxypsoralen (17,55-59), p-hydrazinobenzoates (in commercial mushrooms) (48,49), sinigrin (38-40,60), D-limonene (61-63), estragle and safrole (64-67), ethyl acrylate (68), benzyl acetate (69-71), amethylbenzyl alcohol (23), caffeic acid, chlorogenic acid, neochlorogenic acid (72-80) (in coffee (81,82)), catechol (83,84), sesamol (85). M u ta g e n ic ity a n d cla sto g en icity refs.: see text - 27 FOOTNOTES 1 References to, and analyses of, individual cancer tests are in the Carcinogenic Potency Database papers (10-13). Our analyses are based on this database, which repons only results of chronic, long-term bioassays that are adequate to detect a carcinogenic effect or lack of effect More than 4000 experiments met the inclusion criteria of the database, but thousands of others did not, e.g. tests that lack a control group, that are too short or include too few animals to detect an effect, that use routes of administration not likely to result in whole body exposure (like skin painting or subcutaneous administration), co-carcinogenesis studies, and bioassays of particulate or fibrous matters. One third of the chemicals in the database have been tested by the National Cancer Institute/National Toxicology Program, using standard protocols that include testing in two species at the MTD (15). About half of the chemicals in the database, however, have been tested in only one species. Positivity rates and prediction between species have been analyzed (9). We classify the results of an experiment as either positive or negative on the basis of the author's opinion in the published paper, and classify a chemical as positive if it has been evaluated as positive by the author of at least one experiment. We use the author's opinion to determine positivity because it often takes into account more information than statistical significance alone, such as historical control rates for particular sites, survival and latency, and/or dose response. Generally, this designation by author's opinion corresponds well with the results of statistical reanalyses of the significance of the doseresponse effect (9). i ; rr '7 ' <ilJ i j 0 ! ' 5 28 X + The list of 52 natural plant pesticides includes 7 toxins from edible mushrooms because mushrooms are commonly considered a plant food. Fungal toxins are not included in this list, but are given below. P la n t p e s tic id e s . C a rc in o g e n s: acetaldehyde methylformylhydrazone, allyl isothiocyanate, arecoline.//C7, benzaldehyde, benzyl acetate, caffeic acid, catechol, clivorine, cycasin and methylazoxymethanol acetate mixture, estragle, ethyl acrylate, glutamyl p-hydrazinobenzoic acid, p-hydrazinobenzoic acid, lasiocarpine, N methyl-iV-formylhydrazine, D-limonene, a-methylbenzyl alcohol, methylhydrazine, 8-methoxypsoralen, monocrotaline, pentanal methylformalhydrazine, petasitenine, reserpine, safrole, senkirkine, sesamol, symphytine. (Cycasin as well as its metabolite methylazoxymethanol are positive in numerous tests (45) that do not meet the inclusion criteria of the database.) 5- and 8-methoxypsoralen are isomeric, light activated mutagens (17,45,46). 8-methoxypsoralen is positive in an NTP Gavage study (without light) and is in the database (47). 5-methoxypsoralen has only been tested in a skin painting study (with light) and is positive (46); it is not in our database because it is a skin painting test N o n -c a rc in o g e n s: atropine, benzyl alcohol, biphenyl, caffeine, d-caivone, deserpidine, disodium glycyirhizinate, emetine .2 H C l, ephedrine sulphate, eucalyptol, eugenol, b e ta -N -[y-L (+ )glutamyl]-4-hydroxymethylphenylhydrazine, isosafrole, kaempferol, DLmenthol, nicotine, norharman, pilocarpine, piperidine, rotenone, rutin sulfate, sodium benzoate, vinblastine. U n certa in : tra/is-anethole, quercetin. Among fungal toxins tested for carcinogenicity, 11/16 were positive. F u n g a l toxins: C a rcin o g en s: aflatoxin, 5-azacytidine, azaserine, citrinin, - 29 griseofulvin, luteoskyrin, mitomycin-C, ochratoxin A, sterigmatocystin, streptozotocin, zearalenone. N o n -c a rc in o g e n s: erythromycin stearate, fusarenon-X, oxytetracyclineJ/CV, patulin, penicillin VK. ^ Figures here are based on males aged 25-30 in 1982-1984. Cancer test results are in (10-13). The negative test on 2-ethylhexyl diphenyl phosphate is in (89). The latest FDA figures on actual exposures do not include every known manmade pesticide, and diets vary. Nevertheless, 0.05 mg of possiblycarcinogenic pesticide residues consumed in a day seems to be a reasonable rough estimate. JH tj o u U; 'f? Ranking Possible Carcinogenic Hazards Bruce N. Ames,* Renae Magaw, Lois Swirsky Gold doses to effects on humans exposed to low doses is routinely This review discusses reasons why animal cancer tests attempted by regulatory agendes when formulating polidcs at cannot be used to predict absolute human risks. Such tempting to prevent future canccr. There is little sound scientific tests, however, may be used to indicate that some chemi basis for this type o f extrapolation, in part due to our lack o f cals might be of greater concern than others. Possible knowledge about mechanisms o f cancer induction, and it is viewed hazards to humans from a variety of rodent carcinogens with great unease by many epidemiologists and toxicologists (5, 9 - are ranked by an index that relates the potency of each 11). Nevertheless, to be prudent in regulatory policy, and in the carcinogen in rodents to the exposure in humans. This absence o f good human data (almost always the case), some reliance ranking suggests that carcinogenic hazards from current on animal cancer tests is unavoidable. The best use o f them should levels of pesticide residues or water pollution are likely to be made even though few, if any, o f the main avoidable causes o f be of m inim al concern relative to the background levels of human cancer have typically been the types o f man-made chemicals natural substances, though one cannot say whether these natural exposures arc likely to be of major or minor importance. that are being tested in animals (10). Human cancer may, in part, involve agents such as hepatitis B virus, which causes chronic infiammation; changes in hormonal status; defidencies in normal / - protective factors (such as selenium or p-carorcnc) against endoge nous carcinogens (22); lack o f other anticarcinogens (such as dietary fiber or calaum) (4)\ or dietary imbalances such as excess consump Epid em io l o g ists estimate th a t at least 70% o f hum an cancer would, in principle, be preventable if the main risk and anririsk factors could be identified (2). This is because tion o f fat (3, 4, 12) or salt (13). i There is a need for more balance in animal canccr testing to y \ emphasize the foregoing factors and natural chemicals as well as the incidence o f specific types o f cancer differs markedly in differsyenntthetic chemicals (12). There is increasing evidence that our parts o f the world where people have different life-styles. For normal diet contains many rodent carcinogens, all perfeedy natural example, colon and breast cancer, which are among the major types or traditional (for example, from the cooking o f food) (22), and that \T' VS of cancer in the United States, are quite rare among Japanese in no human diet can be entirely free o f mutagens or agents that can be Japan, but not among Japanese-Americans. Epidemiologists are carcinogenic in rodent systems. We need to identify the important providing important dues about the specific causes o f human causes o f human cancer among the vast number o f minimal risks. &cancer, despite inherent methodological difficulties. They have This requires knowledge o f both the amounts o f a substance to identified tobacco as an avoidable cause o f about 30% o f all U.S. which humans are exposed and its carcinogenic potency. cancer deaths and o f an even larger number o f deaths from other Animal canccr tests can be analyzed quantitatively to give an causes (2, 2). Less specifically, dietary factors, or their absence, have estimate o f the relative carcinogenic potendcs o f the chemicals been suggested in many studies to contribute to a substantial tested. We have previously published our Carcinogenic Potency proportion o f cancer deaths, though the intertwined risk and Database, which showed that rodent carcinogens vary in potency by antirisk factors are being identified only slowly (1, 3, 4). High fat more than 10 millionfold (14). intake may be a major contributor to colon cancer, though the This artide attempts to achieve some perspective on the plethora evidence is not as definitive as that for the role o f saturated fat in o f possible hazards to humans from exposure to known rodent heart disease or o f tobacco in lung cancer. Alcoholic beverage carcinogens by establishing a scale o f the possible hazards for the consumption, particularly by smokers, has been estimated to con amounts o f various common carcinogens to which humans might be tribute to about 3% o f U .S. cancer deaths (I) and to an even larger chronically exposed. We view the value o f our calculations not as number o f deaths from other causes. Progress in prevention has providing a basis for absolute human risk assessment, but as a guide been made for some occupational factors, such as asbestos, to which to priority setting. One problem with this type o f analysis is that few workers used to be heavily exposed, with delayed effects that still o f the many natural chemicals we are exposed to in very large contribute to about 2% o f U .S. cancer deaths (1 ,5). Prevention may amounts (relative to synthetic chemicals) have been tested in animals also become possible for hormone-related cancers such as breast for cardnogenidty. Thus, our knowledge o f the background levels cancer (2, (5), or virus-related cancers such as liver cancer (hepatitis o f human exposure to animal carcinogens is fragmentary, biased in ) and cancer o f the cervix (papilloma virus HPV16) (1, 7). favor o f synthetic chemicals, and limited by our lack o f knowledge o f Animal bioassays and in vitro studies arc also providing dues as to human exposures. nich carcinogens and mutagens might be contributing to human canccr- However, the evaluation o f carcinogenidty in rodents is ^pensive and the extrapolation to humans is difficult (8-11). We use the term "possible hazard" for estimates based on rodent pCcr tcsts and "risk" for those based on human cancer data (10). B. N. Ames is associated with the Department of Biochemistry, University of California, Berkeley, CA94720. R- Magaw and L. SwirskyGold ate associatedwith the Biology and Medicine Division, Lawrence Berkeley Laboratory, Berkeley, CA 94720. *trapolation from the results o f rodent cancer tests done at high *To whom reprint requests should be sent. A;'S2 L 1987 Sc i e n c e v. Z3 4> ARTICLES 271 19586 ioM R anking o f P ossible Carcinogenic Hazards Pesticide residues. Intake o f man-made pesticide residues from food Tf: in the United States, including residues o f industrial chemicals such Since carcinogens differ enormously in potency, a comparison o f as polychlorinated biphenyls (PCBs), averages about 150 jig/day. possible hazards from various carcinogens ingested by humans must Most (105 jig) o f this intake is composed o f three chemicals take this into account. The measure o f potency that we have (erhylhexyl diphenyl phosphate, malathion, and chlorpropham) developed, the T D jo, is the daily dose rate (in milligrams per shown to be noncartinogenic in tests in rodents (16). A carcinogen- kilogram) to halve the percent o f tumor-free animals by the end o f a ic pesticide residue in food o f possible concern is DDE, the principa] standard lifetime (14). Since the TD jo (analogous to the LDjo) is a metabolite (>90% ) o f DDT (16). The average U.S. daily intake of dose rate, the lower the T D jovalue the more potent the carcinogen. DDE from DDT (HERP = 0.0003%) is equivalent to the HERP To calculate our index o f possible hazard we express each human o f the chloroform in one glass o f tap water and thus appears to be exposure (daily lifetime dose in milligrams per kilogram) as a insignificant compared to the background o f natural carcinogens in percentage o f the rodent TD j0 dose (in milligrams per kilogram) for our diet (Table 1). Even daily consumption o f 100 times the average each carcinogen. We call this percentage HERP [Human Exposure intake o f DDE/DDT or PCBs would produce a possible hazard that ? dose/Rodcnt Potency dose]. The TDjo values arc taken from our is small compared to other common exposures shown in Table 1. 4 ongoing Carcinogenic Potency Database (currendy 3500 experi tlature's pesticides. We are ingesting in our diet at least 10,000 ' ments on 975 chemicals), which reports the TD S0 values estimated times more by weight o f natural pesticides than o f man-made j from experiments in animals (14). Human exposures have been pesticide residues (12). These are natural "toxic chemicals" that have estimated from the literature as indicated. As rodent data are all an enormous variety o f chemical structures, appear to be present in calculated on the basis o f lifetime exposure at the indicated daily all plants, and serve to protect plants against fungi, insects, and : \f 9 i -' dose rate (14), the human exposure data are similarly expressed as animal predators (12). Though only a few arc present in each plant " lifelong daily dose rates even though the human exposure is likely to species, they commonly make up 5 to 10% o f the plant's dry weight ' : |f e be less than daily for a lifetime. (12). There has been relatively little interest in the toxicology or j. It would be a mistake to use our HERP index as a direct estimate carcinogenicity o f these compounds until quite recently, although o f human hazard. First, at low dose rates human susceptibility may they are by for the main source o f "toxic chemicals" ingested by differ systematically from rodent susceptibility. Second, the general humans. Only a few dozen o f the thousands present in the human shape o f the dose-response relationship is not known. A linear dose diet have been tested in animal bioassavs, and only some o f these ~ response has been the dominant assumption in regulating carcino tests arc adequate for estimating potency in rodents (14). A sizable gens for many years, but this may not be correct. If the dose proportion o f those that have been tested are carcinogens, and many ? responses are not linear but are actually quadratic or hockey-stick others have been shown to be mutagens (12), so it is probable that ; shaped or show a threshold, then the actual hazard at low dose rates many more will be found to be carcinogens if tested. Those shown might be much less than the HERP values would suggest. An in Table 1 are: estragle (HERP = 0.1% for a daily 1 g o f dried additional difficulty is that it may be necessary to deal with basil), safrolc (HERP = 0.2% for a daily natural root beer), sym- 7 carcinogens that differ in their mechanisms o f action and thus in phyrinc (a pyrrolizidine alkaloid, 0.03% for a daily cup. o f comfrey their dose-response relationship. We have therefore put an asterisk tea), comfrcy tablets sold in health food stores (6.2% for a daily a next to HERP values for carcinogens that do not appear to be active dose), hydrazines in mushrooms (0.1% for one daily raw mush- * through a genotoxic (DNA damaging or mutagenic) mechanism room), and ailyl isothiocyanate (0.07% for a daily 5 g o f brown 7 (15) so that comparisons can be made within the genotoxic or mustard). nongenotoxic classes, Plants commonly produce very much larger amounts o f their [- Table 1 presents our HERP calculations o f possible cancer natural toxins when damaged by insects or fungi (12). For example, hazards in order to compare them within several categories so that, psoralens, light-activated carcinogens in celery, increase 100-fold [ for example, pollutants o f possible concern can be compared to when the plants are damaged by mold and, in feet, can cause an5 natural carcinogens in the diet. A convenient reference point is the occupational disease in celery-pickers and in produce-checkers at possible hazard from the carcinogen chloroform in a liter o f average supermarkets (12, 17). -fj (U.S.) chlorinated tap water, which is close to a HERP o f 0.001%. Molds synthesize a wide variety o f toxins, apparently as antibiotics j Chloroform is a by-product o f water chlorination, which protects us in the microbiological struggle for survival: over 300 mycotoxins j from pathogenic viruses and bacteria. have been described (18). They arc common pollutants o f human | Contaminated, water. The possible hazards from carcinogens in food, particularly in the tropics. A considerable percentage of those% contaminated w ell water [for example, Santa Clara ("Silicon") . tested have been shown to be mutagens and carcinogens: some, such - Valley, California, or W obum, Massachusetts] should be compared as afiatoxin and sterigmatocystin, are among the most potent known| to the possible hazard o f ordinary tap water (Table 1). O f 35 wells rodent carcinogens. The potency o f afiatoxin in different species^ shut down in Santa Clara Valley because o f their supposed carcino varies widely; thus, a bias may exist as the HERP uses the most* genic hazard, only two have HERP values greater than ordinary tap sensitive species. The afiatoxin content o f U.S. peanut butter^ water. Well water is not usually chlorinated and typically lacks the averages 2 ppb, which corresponds to a HERP o f 0.03% for the: chloroform present in chlorinated tap water. Water from the most peanut butter in an average sandwich (Table 1). The Food and Dmf polluted wcU (HERP = 0.004% per liter for trichloroethylene), as Administration (FDA) allows ten times this level (HERP = 0.3%) indicated in Table 1, has a HERP value orders o f magnitude less than for the carcinogens in an equal volume o f cola, beer, or wine. Its HERP value is also much lower than that o f many o f the common natural foods that arc listed in Table 1, such as the average peanut butter sandwich. Caveats for any comparisons are given and certain foods can often exceed the allowable limit (18). AflJ` toxin contaminates wheat, com (perhaps the main source o f dietati afiatoxin in the United States), and nuts, as well as a wide variety " j stored carbohydrate foodstuffs. A carcinogenic, though less potent! metabolite o f afiatoxin is found in milk from cows that eat mol<ty| I* [? [' 1. r below. Since the consumption o f tap water is only about 1 or 2 liters grain. T per day, the animal evidence provides no good reason to expect that chlorination o f water or current levels o f man-made pollution o f water pose a significant carcinogenic hazard. There is epidemiologic evidence that afiatoxin is a human caran0-i gen. High intake in the tropics is associated with a high rate of cancer, at least among those chronically infected with the hepad05 ui Ah 272 1 ^ 1 17 a 1 9 5 9 cf NCE' vou virus 20)- Considering the potency o f those mold toxins that have been tested and the widespread contamination o f food with molds, they may represent the m ost significant carcinogenic pollu tion o f the food supply in developing countries. Such pollution is much less severe in industrialized countries, due to refrigeration and modem techniques o f agriculture and storage, including use o f synthetic pesticides and fumigants. Preparation offoods and beverages can also produce carcinogens. Alcohol has been shown to be a human carcinogen in numerous epidemiologic studies (1, 21). Both alcohol and acetaldehyde, its a d it it )r :h iy n se le iy at vn ce rn- :tV reino- : .flivtf ribs )L. T a b le 1 Ranking possible carcinogenic hazards. Potencyi f cardnogenr. A number in parentheses indicates aT D jo value not used in HERB calculation because t is the less sensitive species; (--) 3 negative in cancer tese (+) = positive for carcinogenicity in tcst(s) not suitable for calculating a TD30; (?) = is not adequately tested for carcinogenicity. TD30values shown are averages calculated by takingthe harmonic mean oftheT D jo's ofthe positive tests in that species from the Carcinogenic Potency Database. Results are similar ifthe lowest TDjovalue (most potent) is used instead. For each test die target site with the lowesr TD50 value has been used. The averageTDjo has beencalculated separately for rats and mice, and the more sensitive species is used for calculating the pos sible hazard. The database, with references to the source of the cancer tests, is complete for tests published through 1984 and for the National Toxicology Program bioassays through June 1986 (14). We have not indicated the route of exposure or targetsites or other particularsofeach test, although these are re ported in the database. Daily human exposure. We have tried to use average or reasonable daily intakes to facilitate comparisons. In several cases, such as contaminated well water or factory exposure to EDB, this is difficult to determine, and we give che value for the worst found and indicate pertinent information in the References and Notes. The calculations assume adaily dose for a lifetime; where drugs arc normally taken for only ashort period we have bracketed the HERP value. For inhaladon exposures we assume an inhalation o f9,600 liters per 8 hours for the workplace and 10,800 liters per 14 hours for indoorair at home. Possiblehazard-. The amount of rodent carcinogen indicated undercarcinogen dose is divided by 70 kg to give amilligram per kilogram of human exposure, and this human dose is given as the percentage of the T D jo dose in the rodent (in milligrams per kilogram) to calculate the Afuman fxposure/Rodent Potency index (HERP). Possible hazard: K IR P(%) " ' ~ Daily human exposure Carcinogen dose per 70-kg person 0.001* 0.004* 0.0004* 0.0002* 0.0003* 0.008* 0.6 0.004 2.1 0.0002* 0.0003* 0.0004 0.003 0.006 0.003 0.03 0.03 0.06 0.U7 0.1 0.1 0.2 0.008 2.8* 4.7* 6.2 1.3 0.0002 0.06* [0.3] 15.6] [14] 16* 17* 5.8 140 Environmental pollution Tap water, 1 liter Well water, 1 liter contaminated (worst well in Silicon Valley) Well water, 1 liter contaminated, Woburn Swimming pool, 1 hour (for child) Conventional home air (14 hour/day) Mobile home air (14 hour/day) Chloroform, 83 pg (U.S. average) Trichloroethylene, 2800 pg Trichloroethylene, 267 pg Chloroform, 12 pg Tctrachloroethylene, 21 pg Chloroform, 250 pg (average pool) Formaldehyde, 598 pg Benzene, 155 pg Formaldehyde, 22 mg Pesticide and other residua PCBs: daily dietary intake DDE/DDT: daily dietary intake EDB: daily dietary intake (from grains and grain products) PCBs, 0.2 pg (U.S. average) DDE, 2.2 pg (U.S. average) Ethylene dibromide, 0.42 pg (U.S. average) Naturalpesticide and dietary toxins Bacon, cooked (100 g) Sake (250 ml) Comfrcy herb tea, 1 cup Peanut butter (32 g; one sandwich) Dried squid, broiled in gas oven (54 g) Brown mustard (5 g) Basil (1 g of dried leaf) Mushroom, one raw (15 g) (Agaricus bisporus) Natural root beer (12 ounces; 354 ml) (now banned) Beer, before 1979 (12 ounces; 354 ml) Beer (12 ounces; 354 ml) Wine (250 ml) Comfrcy-pepsin tablets (nine daily) Comfrcy-pepsin ablets (nine daily) Dimethylnitrosamine, 0.3 pg Diethylnitrosamine, 0.1 pg Urethane, 43 pg Symphytine, 38 pg (750 pg of pyrrolizidine alkaloids) Aflatoxin, 64 ng (U.S. average, 2 ppb) Dimethylnitrosamine, 7.9 pg Allyl isothiocyanare, 4.6 mg Estragole, 3.8 mg Mixture of hydrazines, and so forth Safrolc, 6.6 mg Dimethvlnitrosaminc, 1 pg Ethvl alcohol, 18 ml Ethyl alcohol, 30 ml Comfrcy root, 2700 mg Symphytine, 1.8 mg Pood additive AF-2: daily dietary intake before banning Diet Cola (12 ounces; 354 ml) AF-2 (fiirylforamidc), 4.8 pg Saccharin, 95 mg Drugs Phenacctin pill (average dose) Metronidazole (therapeutic dose) Isoniazid pill (prophylactic dose) Phnobarbital, one sleeping pill Qofibrate (average daily dose) Phenacetin, 300 mg Metronidazole, 2000 mg Isoniazid, 300 mg Phnobarbital, 60 mg Clofibrate, 2000 mg Occupational exposure Formaldehyde: Workers' average daily intake EDB: Workers' daily intake (high exposure) Formaldehyde, 6.1 mg Ethylene dibromide, 150 mg ensks indicate HERP t e n carcinogens thought to be nongenoraxic. 17aprh. V&-J Potency of carcinogen: TDjo (mg/kg) Rats Mice (119) (-) (-) (119) 101 (119) 1.5 (157) 1.5 1.7 (" ) 1.5 90 941 941 90 (126) 90 (44) 53 (44) (9.6) 13 (5.1) Refer- 96 97 98 99 100 28 101 16 102 (0.2) 0.02 (41) 1.9 0.003 (0.2) 96 (?) (?) (436) (0.2) 9110 9110 626 1.9 29 2143 1246 (542) (150) (+) 169 1.5 1.5 0.2 (+) 22 (?) (+) 0.2 (-) 52 20,300 56 0.2 (?) (?) (?) (?) (131) (-) (2137) 506 30 5.5 (?) (44) (5.1) 40 24 103 18 37 47 48 104 105 38 23 23 103 44 106 51 107 108 50 52 109 55 I jfincLES 273 %J o O V,' major metabolite, are carcinogens in rats (22, 23). The carcinogenic potency o f ethyl alcohol in rats is remarkably low (23), and it is among the weakest carcinogens in our database. However, human intake o f alcohol is very high (about 18 g per beer), so that the possible hazards shown in Table 1 for beer and wine are large (HERP = 2.8% for a daily beer). The possible hazard o f alcohol is enormous relative to that from the intake o f synthetic chemical residues. If alcohol (20), trichloroethylene, DDT, and other pre sumptive nongenotoxic carcinogens are active at high doses because they are tumor promoters, the risk from low doses may be minimal. Other carcinogens are present in beverages and prepared foods. Urethane (ethyl carbamate), a particularly well-studied rodent car cinogen, is formed from ethyl alcohol and carbamyl phosphate during a variety o f fermentations and is present in Japanese sake (HERP = 0.003% ), many types o f wine and beer, and in smaller amounts in yogurt and bread (24). Another fermentation product, the dicarbonvl aldehyde methvlglyoxal, is a potent mutagen and was isolated as the main mutagen in coffee (about 250 |xg in one cup). It was rccendy shown to be a carcinogen, chough not in a test suitable for calculating a TD 50(25). Methvlglyoxal is also present in a variety o f other foods, such as tomato puree (25, 26). Diacctvl (2,3butanedione), a closely related dicarbonyl compound, is a fermentadon product in wine and a number o f other foods and is responsible for the aroma o f butter. Diacetyl is a mutagen (27) but has not been tested for carcinogenicity. Formaldehyde, another natural carcinogenic and mucagenic alde hyde, is also present in many common foods (22,26-28). Formalde hyde gas caused cancer only in the nasal turbinates o f che nosebreathing rodents and even though formaldehyde is genotoxic, foe dose response was nonlinear (28, 29). Hexamethylenetetramine, which decomposes eo formaldehyde in foe scomach, was negative in feeding studies (30). The effects o f oral versus inhalation exposure for formaldehyde remain to be evaluated more thoroughly. As formaldehyde is almost ubiquitous in foods, one can visualize various formaldehyde-rich scenarios. Daily consumption o f shrimp (HERP = 0.09% per 100 g) (31), a sandwich (HERP o f two slices o f bread = 0.4%) (22), a cola (HERP = 2.7%) (32), and a beer (HERP = 0.2%) (32) in various combinations could provide as much formaldehyde as living in some mobile homes (HERP = 2.1%; Table 1). Formaldehyde is also generated in animals metabolically, for example, from mefooxy compounds that humans ingest in considerable amounts from plants. The level o f formaldehyde reported in normal human blood is strikingly high (about 100 \iM or 3000 ppb) (33) suggesting that detoxification mechanisms are important. The cooking o ffood generates a variety o f mutagens and carcino gens. Nine heterocyclic amines, isolated on the basis- o f their mutagenicity from proteins or amino adds that were heated in ways that occur in cooking, have now been tested; all have been shown to be potent carcinogens in rodents (34). Many others are still being isolated and characterized (34). An approximate HERP o f 0.02% has been calculated by Sugimura et ai. for the daily intake o f these nine carcinogens (34). Three mutagenic nitropyrcncs present in diesel exhaust have now been shown to be carcinogens (35), but the intake o f these carcinogenic nitropyrcncs has been estimated to be much higher from grilled chicken than from air pollution (34, 36). The total amount o f browned and burnt material eaten in a typical day is at least several hundred times more than that inhaled from severe air pollution (12). Gas flames generate NO 2, which can form both foe carcinogenic nitropyrcncs (35, 36) and che potently carcinogenic nitrosamines in food cooked in gas ovens, such as fish or squid (HERP = 0.06%; Table 1) (37). We suspect that food cooked in gas ovens may be a major source o f dietary nitrosamines and nitropyrcncs, though it is not clear how significant a risk these pose. Nitrosamines wen ubiquitous in beer and ale (HERP = 0.008% ) and were formei- from NO2 in foe gas flame-heated air used to dry foe malt However, foe industry has switched to indirect heating, which resulted in markedly lower levels (< 1 ppb) o f dimefoylnitrosammc (38). The dimefoyinitrosamine found in human urine is thought t<. be formed in part from NO2 inhaled from kitchen air (39). Cooker bacon contains several nitrosamines (HERP = 0.009% ) (40): Oxidation offats and vegetable oils occurs during cooking and also spontaneously if antioxidant levels are low. The result is the formation o f peroxides, epoxides, and aldehydes, all o f which appear to be rodent carcinogens (8, 12, 27). Fatty acid hydroperoxide- (present in oxidized oils) and cholesterol epoxide have been showi to be rodent carcinogens (though not in tests suitable for calculador a TD50). Dried eggs contain about 25 ppm o f cholesterol epoxide (' sizable amount), a result o f foe oxidation o f cholesterol by foe NO in foe drying air that is warmed by gas flames (12). Normal oxidation reactions in fruit (such as browning in a cm apple) also involve production o f peroxides. Hydrogen peroxide is 3 mutagenic rodent carcinogen that is generated by oxidation 01 natural phenolic compounds foac arc quite widespread in edible plants. A cup o f coffee contains about 750 p.g o f hydrogen peroxide (25); however, since hydrogen peroxide is a very weak carcinoger (similar in potency to alcohol), the HERP for drinking a daily cup o f coffee would be very low [comparable to D DE/DDT, PCBs, or ethylene dibromide (EDB) dietary intakes]. Hydrogen peroxide L< also generated in our normal metabolism; human blood contain: about 5 puVi hydrogen peroxide and 0.3 \iM o f foe cholesterol ester o f fatty acid hydroperoxide (41). Endogenous oxidants such a< hydrogen peroxide may make a major contribution to cancer and aging (42). Caloric intake, which could be considered the most striking rodent carcinogen evpr discovered, is discussed remarkably little in relation to human cancer. It has been known for about 40 years that increasing the food intake in rats and mice by about 20% above optimal causes a remarkable decrease in longevity and a striking increase in endocrine and mammary tumors (43). In humans, obesity (associated with high caloric intake) leads to increased levels o f circulating estrogens, a significant cause o f endometrial and gall bladder cancer. The effects o f moderate obesity on other types of human cancer are less clear (1). Food additives are currently screened for carcinogenicity before use if they are synthetic compounds. AF-2 (HERP = 0.0002% ), a food preservative, was banned in Japan (44). Saccharin (HERP = 0.06%) is currently used in the United States (the dose- response in rats, however, is clearly sublinear) (45). The possible hazard o f diethylstilbestrol residues in meat from created farm animals seems miniscule relative to endogenous estrogenic hor mones and plant estrogens (46). Some natural carcinogens are also widely used as additives, such as allyl isochiocyanatc (47), estragle (48), and alcohol (23). a A ir pollution. A person inhales about 20,000 liters o f air in a day: thus, even modest contamination o f foe atmosphere can result it inhalation o f appreciable doses o f a pollutant. This can be seer, in the possible hazard in mobile homes from formaldehyde (HERP = 2.1%) or in conventional homes from formaldehyde (HERP = 0.6%) or benzene (HERP = 0.004% ; Table 1). Indoor air pollution is, in general, worse than outdoor air pollution, pard} because o f cigarette smoke. The most important indoor air pollutant may be radon gas. Radon is a natural radioactive gas that is presen* in the soil, gets trapped in houses, and gives rise to radioactive decay products that arc known to be carcinogenic for humans (49). It d** been estimated that in 1 million homes in foe United States foe o f exposure to products o f radon decay may be higher than tha* Z74- t C S7 r Q SCIENCE, VOL- a* JL ^ - r s were : c. d. which samine ight to y ] looked ') * ad also 1 is the appear roxides shown "ulating >xide (a ic NO, n a cut .tide is a icion of i edible icroxidc cinogcn ally cup CBs, or oxide is :ontains rol ester such as cer and .it relation an that o above striking lumans, ;d levels and gall types of :fore use ;1 32%), a \ \ accharin he dose- *[ possible cd form seen. VOL. teceived by today's uranium miners. Two particularly contaminated houses were found that had a risk estimated to be equivalent to jeceiving about 1200 chest x-rays a day (49). Approximately 10% o f die lung cancer in the United States has been tentatively attributed to radon pollution in houses (49). Many o f these cancers might be preventable since the most hazardous houses can be identified and modified to minimize radon contamination. General outdoor air pollution appears to be a small risk relative to the pollution inhaled by a smoker: one must breathe Los Angeles smog for a year to inhale the same amount o f burnt material that a smoker (two packs) inhales in a day (12), though air pollution is inhaled starting from birth. It is difficult to determine cancer risk from outdoor air pollution since epidemiologists must accurately control for smoking and radon. Some common drugs shown in Table 1 give fairly high HERP percentages, primarily because the dose ingested is high. However, since most medicinal drugs are used for only short periods while the HERP index is a daily dose rate for a lifetime, the possible hazard would usually be markedly less. We emphasize this in Table I by bracketing the numbers for these shorter exposures. Phenobarbital (HERP = 16%) was investigated thoroughly in humans who had rake;', it for decades, and there was no convincing evidence that it caused cancer (50). There is evidence o f increased renal cancer in long-term human ingestion o f phenacetin, an analgesic (51). Acet aminophen, a metabolite o f phenacetin, is one o f the most widely used over-the-counter pain killers. Clofibrate (HERP = 17%) is used as a hypolipidemic agent and is thought to be carcinogenic in rodents because it induces hydrogen peroxide production through peroxisome proliferation (52). Occupational exposures can be remarkably high, particularly for volatile carcinogens, because about 10,000 liters o f air are inhaled in a working day. For formaldehyde, the exposure to an average worker (HERP = 5.8%) is higher than most dietary intakes. For a number o f volatile industrial carcinogens, the ratio o f the permitted exposure limit [U.S. Occupational Safety and Health Administraon (OSHA)] in milligrams per kilogram to the TD50 has been calculated; several arc close to the TD 50 in rodents and about twothirds have permitted HERP values >1% (53). The possible hazard estimated for the actual exposure levels o f the most heavily exposed EDB workers is remarkably high, HERP = 140% (Table 1). Though the dose may have been somewhat overestimated (54), it was still comparable to che dose causing cancer in half the rodents. An epidemiologic study o f these heavily exposed EDB workers who inhaled EDB for over a decade did not show any increase in cancer, thougn because o f the limited duration o f exposure and the relatively small numbers o f people monitored the study would not have detected a small effect (54, 55). OSHA still permits exposures above the TD5o level. California, however, lowered the permitted level over 100-fold in 1981. In contrast with these heavy workplace ^posures, the Environmental Protection Agency (EPA) has banned ^e use o f EDB for fumigation because o f the residue levels found in Sfam (HERP = 0.0004% ). Uncertainties in R elying on Anim al Cancer Tests for H um an Prediction Species variation. Though we list a possible hazard if a chemical is a ^fcinogcn in a rat but not in a mouse (or vice versa), this lack o f ^Steemcnt raises the possibility that the risk to humans is nonexis- 22a ^ chemicals in our database tested in both rats and mice, . Wcre carcinogens in at least one test, but 96 o f these were 'ttve in the mouse and negative in the rat or vice versa (56). This 0rdance occurs despite the fact that rats and mice are very closely 17 ,q87 related and have short life-spans. Qualitative extrapolation o f cancer risks from rats or mice to humans, a very dissimilar long-lived species, is unlikely to be as reliable. Conversely, important human carcinogens may not be detected in standard tests in rodents; this was true for a long time for both tobacco smoke and alcohol, the two largest identified causes o f neoplastic death in the United States. For many o f the chemicals considered rodenc carcinogens, there may be negative as well as positive tests. It is difficult to deal with negative results satisfactorily for several reasons, including the fact that some chemicals are tested only once or twice, while others are tested many times. The HERP index ignores negative tests. Where there is species variation in potency, use o f the more sensitive species, as is generally done and as is done here, could introduce a tendency to overestimate possible hazards; however, for most chemicals that are positive in both species, the potency is similar in rats and mice (57). The HERP may provide a rough correlate o f human hazard from chemical exposure; however, for a given chemical, to the extent that the potcnqrin humans differs from the potency in rodents, the relative hazard would be different. Quantitative uncertainties. Quantitative extrapolation from ro dents to humans, particularly at low doses, is guesswork that we have no way o f validating (1,5 ,1 0 ,1 1 ,5 8 ). It is guesswork because o f lack o f knowledge in at least six major areas: (i) the basic mechanisms o f carcinogenicity; (ii) the relation o f cancer, aging, and life-span (1,10,42,59); (iii) the riming and order o f the steps in the carcinogenic process that are being accelerated; (iv) species differ ences in metabolism and pharmacokinetics; (v) species differences in anticarcinogens and other defenses (1, 60); and (vi) human hetero geneity--for example, pigmentation affects susceptibility to skin cancer from ultraviolet light. These sources o f uncertainty are so numerous, and so substantial, that only empirical data will resolve them, and little o f this is available. Uncertainties due to mechanism in multistage carcinogenesis. Several steps (stages) arc involved in chemical carcinogenesis, and the doseresponse curve for a carcinogen might depend on the particular stagc(s) it accelerates (58), with multiplicative effects if several stages are affected. This multiplicative effect is consistent with the observa tion in human cancer that synergistic effects are common. The three steps o f carcinogenesis that have been analyzed in most detail arc initiation (mutation), promotion, and progression, and we discuss these as an aid to understanding aspects o f the dose-response relation. Mutation (or DNA damage) as one stage o f the carcinogenic process is supported by various lines of evidence: association o f active forms o f carcinogens with mutagens (61), the changes in DNA sequence o f oncogenes (62), generic predisposition to cancer in human diseases such as retinoblastoma (63) or DNA-repair deficiency diseases such as xeroderma pigmentosum (64). The idea that gcnotoxic carcinogens might show a linear dose-response might be plausible if only the mutation step o f carcinogenesis was acceler ated and if the induction o f repair and defense enzymes were not significant factors (65). Promotion, another step in carcinogenesis, appears to involve cell proliferation, or perhaps particular types o f cell proliferation (<5<5), and dose-response relations with apparent thresholds, as indicated by various lines o f evidence: (i) The work o f Trosko et al. (67) on promotion o f carcinogenesis due to interference wich cell-cell com munication, causing cell proliferation, (ii) Rajewsk/s and other work indicating initiation by some carcinogenic agents appears to require proliferating target cells (68). (iii) The work o f Farber et al. (69) on liver carcinogenesis supports the idea that cell proliferation (caused by partial hepatectomy or cell killing) can be an important aspect o f hepatocarcinogenesis. They have also shown for several chemicals that hepatic cell killing shows a toxic threshold wich dose, (iv) Work on carcinogenesis in the pancreas, bladder and stomach 19600 articles 275 N. 0), and other tissues (58) is also consistent with results on the liver (71, 72) though the effect o f cell proliferation might be different in tissues that normally proliferate, (v) The work o f Mirsalis a d . (71) suggests that a variety o f nongenotoxic agents are hepatocardnogens in the B6C3F1 mouse (commonly used in cancer tests) because o f their toxicity. Other studies on chloroform and trichloroethylene also support this interpretation (72, 73). Cell proliferation resulting from the cell killing in the mouse liver shows a threshold with dose (72). Also relevant is the extraordinarily high spontaneous rates o f liver tumors (21% carcinomas, 10% adenomas) in the male B6C3F1 mouse (74). These spontaneous tumors have a mutant ms oncogene, and thus the livers in these mice appear to be highly initiated (mutated) to start with (75). (vi) Oncogenes: As Weinberg (62) has pointed out, "Oncogene-bearing cells surrounded by normal neigh bors do not grow into a large mass if they carry only a single oncogene. But if the normal neighbors are removed . . . by killing them with a cytotoxic drug. . . then a single oncogene often suffices." (vii) Cell killing, as well as mutation, appears to be an important aspect o f radiation carcinogenesis (76). Promotion has also been linked to the production o f oxygen radicals, such as from phagocytic cells (77). Since chronic cell killing would usually involve inflammatory reactions caused by neutrophils, one would commonly expect chemicals tested at the maximally tolerated dose (MTD) to be promoters because o f the chronic inflammation. Progression, another step in carcinogenesis, leading to selection for invasiveness and mctastascs, is not well understood but can be accelerated'by oxygen radicals (78). Chronic cell toxicity caused by dosing at the MTD in rodent cancer bioassays thus not only could cause inflammation and cell proliferation, but also should be somewhat mutagenic and clastogenic to neighboring cells because o f the release o f oxygen radicals from phagocytosis (12, 79, 80). The respiratory burst from phago cytic neutrophils releases the same oxidative mutagens produced by radiation. (77, 79). Thus, animal cancer tests done at the MTD o f a chemical might commonly stimulate all three steps in carcinogenesis and be positive because the chemical caused chronic cell killing and inflammation with som e mutagenesis. Some o f the considerable human evidence for chronic inflammation contributing to carcino genesis and also some evidence for and against a general effect o f inflammation and cytotoxicity in rodent carcinogenesis have been discussed (81). Another set o f observations may also bear on the question o f toxicity and extrapolation. Wilson, Crouch, and Zeise (82) have pointed out that among carcinogens one can predict the potency in high-dose animal cancer experiments from the toxicity (the LD50) o f the chemical, though one cannot predict whether the substance is a carcinogen. We have shown that carcinogenic potency values are bounded by the MTD (57). The evidence from our database suggests that the relationship between TD 50 and MTD has a biological as well as a statistical basis (57). We postulate that a just sublethal level o f a carcinogen causes cell death, which allows neighboring cells to proliferate, and also causes oxygen radical production from phagocytosis and thus chronic inflammation, both important aspects o f the carcinogenic process (57). The generality o f this relationship and its basis needs further study. If most animal cancer tests done at the MTD are partially ensuring cell killing and consequent cell proliferation and phago cytic oxygen radical damage as steps in the carcinogenic process, one might predict that the dose-response curves would generally be nonlinear. For those experiments in our database for which life table data (14) were available, a detailed analysis (83) shows that the doseresponse relationships arc more often consistent with a quadratic (or cubic) model than with a linear model. Experimentally, it is very difficult to discriminate between the various extrapolation models at low doses (11, 58). However, evidence to support the idea that a nonlinear dose-response relation ship is the norm is accumulating for many nongenotoxic and some gcnotoxic carcinogens. Dose-response curves for saccharin (45), butylatcd hydroxyanisolc [BHA (54)], and a variety o f other nongenotoxic carcinogens appear to be nonlinear (85). Formalde hyde, a gcnotoxic carcinogen, also has a nonlinear dose response (28, 29). The data for both bladder and liver tumors in the large- scale study on acetylaminofluorene, a gcnotoxic chemical, could fit a hockey stick-shaped curve, though a linear model, with a decreased effect at lower dose rates when the total dose is kept constant (86), has not been ruled out. Carcinogens effective at both mutating and killing cells (which includes most mutagens) could be "complete" carcinogens and therefore possibly more worrisome at doses far below the MTD than carcinogens acting mainly by causing cell killing or prolifera tion (15). Thus, all carcinogens are not likely to be directly comparable, and a dose o f 1/100 the TD50 (HERP = 1%) might be much more o f a carcinogenic hazard for the gcnotoxic carcinogens dimethylnitrosamine or aflatoxin than for the apparently nongeno toxic carcinogens trichloroethylene, PCBs, or alcohol (HERP values marked with asterisks in Table 1). Short-term tests for mutagenicity (61, 87) can have a role to play, not only in understanding mechanisms, but also in getting a more realistic view o f the background levels o f potential gcnotoxic carcinogens in the world. Knowledge o f mechanism o f action and comparative metabolism in rodents and humans might help when estimating the relative importance o f various low-dose exposures. Human cancer, except in some occupational or medicinal drug exposures, is not from high (just subtoxic) exposures to a single chemical but is rather from several risk factors often combined with alack o f antirisk factors (60); for example, aflatoxin (a potent mutagen) combined with an agent causing cell proliferation, such as hepatitis B virus (19). High salt [a possible risk factor in stomach cancer (23)] and high fat [a possible risk factor in colon cancer (4)] both appear to be effective in causing cell killing and cell proliferation. Risk from carcinogenesis is not linear with time. For example, among regular cigarette smokers the excess annual lung cancer " incidence is approximately proportional to the fourth power o f the duration o f smoking (88). Thus, if human exposures in Tabic 1 are much shorter than die lifetime exposure, the possible hazard may be markedly less than linearly proportional. ? A key question about animal cancer tests and regulatory policy is - the percentage o f tested chemicals that will prove to be carcinogens - (89). Among the 392 chemicals in our database that were tested in . both rats and mice, 58% are positive in at least one species (14). For . the 64 "natural" substances in the group, the proportion o f positive results is similar (45%) to the proportion o f positive results in the ; synthetic group (60%). One explanation offered for the high proportion o f positive results is that more suspicious chemicals are| being tested (for example, relatives o f known carcinogens), but w e | do not know if the percentage o f positives would be low among less j suspicious chemicals. If toxicity is important in carcinogenicity, as we have argued, then at the MTD a high percentage o f all chemicals t might be classified as "carcinogens." The Background o f N atural Carcinogens j The object o f this article is not to do risk assessment on naturally occurring carcinogens or to worry people unduly about an occasional raw mushroom or beer, but to put the possible hazard o f in**1' made carcinogens in proper perspective and to point out that've 276 S C IE N O E . VOL. some ( ), other laldc- - 1 ponsc largeid fit a rcased 1 (86), which ts and MTD fliferaiirectly ightbc nogens ngeno1values icnicity anding o f the world. >lism in relative 5 i su.gie ;d with utagen) jaritis B 3)] and ar to be xample, cancer x o f the )lc 1 are J i may be policy is :inogens rested in 14). For 'positive ts in the he high lic&ls arc , but wc long less nicity, ** * naturally occasion* 1o f man* t that * e lade the knowledge to do low-dose "risk assessment." Wc also are For example, the aflatoxin in the average peanut butter sandwich, or almost completely ignorant o f the carcinogenic potential o f the a raw mushroom, arc 75 and 200 times, respectively, the possible enormous background o f natural chemicals in the world. For hazard o f EDB. Before banning EDB, a useful substance with rather example, cholinesterase inhibitors arc a common class o f pesticides, low residue levels, it might be reasonable to consider whether the both man-made and natural. Solaninc and chaconine (the main hazards o f the alternatives, such as food irradiation, or the conse alkaloids in potatoes) are cholinesterase inhibitors and were intro quences o f banning, such as increased mold contamination o f grain, duced generally into the human diet about 400 years ago with the pose less risk to society. Also, there is a disparity between OSHA dissemination o f the potato from the Andes. They can be detected in not regulating worker exposures at a HERP o f 140%, while the the blood o f almost all people (12,90). Total alkaloids are present at EPA bans the substance at a HERP o f 0.0004% . In addition, the i level o f 15,000 |ig per 200-g potato with not a large safety factor FDA allows a possible hazard up to a HERP o f 0.3% for peanut (about sixfold) from the toxic level for humans (91). Neither butter (20 ppb), and there is no warning about buying comffcy pills. alkaloid has been tested for carcinogenicity. By contrast, malathion, Because o f the large background o f low-level carcinogenic and die main synthetic organophosphate cholinesterase inhibitor in our other (93) hazards, and the high costs o f regulation, priority setting diet (17 jtg/day) (16), is not a carcinogen in rodents. is a critical first step. It is important not to divert society's attention The idea that nature is benign and that evolution has allowed us away from the few really serious hazards, such as tobacco or cocope perfectly with the toxic chemicals in the natural world is not saturated fat (for heart disease), by the pursuit o f hundreds o f minor compelling for several reasons: (i) there is no reason to think that or nonexistent hazards. Our knowledge is also more certain about natural selection should eliminate the hazard o f carcinogenicity o f a the enormous toll o f tobacco-- about 350,000 deaths per year (1,2). plant toxin that causes cancer in old age past the reproductive age, There arc many trade-offs to be made in all technologies. Trichlo though there could be selection for resistance to the acute effects o f roethylene and tctrachlorocthylene (perchlorocthylenc) replaced particular carcinogens. For example, aflatoxin, a mold toxin that - hazardous flammable solvents. Modem synthetic pesticides dis prca'imably arose early in evolution, causes cancer in trout, rats, placed lead arsenate, which was a major pesticide before the modem mice, and monkeys, and probably people, chough the species are not chemical era. Lead and arsenic are both natural carcinogens. There is equally sensitive. Many o f the common metal salts are carcinogens also a choice to be made between using synthetic pesticides and (such as lead, cadmium, beryllium, nickel, chromium, selenium, and raising the level o f plants' natural toxins by breeding. It is not clear arsenic) despite their presence during all o f evolution, (ii) Given the that the latter approach, even where feasible, is preferable. For enormous variety o f plant toxins, most o f our defenses may be example, plant breeders produced an insect-resistant potato, which general defenses against acute effects, such as shedding the surface has to be withdrawn from the market because o f its acute toxicity to lining o f cells o f our digestive and respiratory systems every day; humans due to a high level o f the natural plant toxins solaninc and protecting these surfaces with a mudn layer; having detoxifying chaconine (12). enzymes that are often inducible, such as cytochrome P-450, This analysis on the levels o f synthetic pollutants in drinking conjugating enzymes, and glutathione transferases; and having water and o f synthetic pesticide residues in foods suggests that this DN.:_repair enzymes, which would be useful against a wide variety pollution is likely to be a minimal carcinogenic hazard relative to the ofingested toxic chemicals, both natural and synthetic Some human background o f natural carcinogens. This result is consistent with the cancer may be caused by interfering with these normal protective epidemiologic evidence (1). Obviously prudence is desirable with systems, (iii) The human diet has changed drastically in the last few regard to pollution, but wc do need to work out some balance ' thousand years, and most o f us are eating plants (such as coffee, between chcmophobia with its high costs to the national wealth, - potatoes, tomatoes, and kiwi font) that our ancestors did not. (iv) and sensible management o f industrial chemicals (94). Normal metabolism produces radiomimctic mutagens and carcino Human life expectancy continues to lengthen in industrial coun gens, such as hydrogen peroxide and other reactive forms o f oxygen. tries, and the longest life expectancy in the world is in Japan, an Though we have defenses against these agents, they still may be extremely crowded and industrialized country. U.S. cancer death major contributors to aging and cancer. A wide variety o f external rates, except for lung cancer due to tobacco and melanoma due to agents may disturb this balance between damage and defense (12, ultraviolet light, arc not on the whole increasing and have mosdy 42). been steady for 50 years. N ew progress in cancer research, molecular biology, epidemiology, and biochemical epidemiology (95) will probably continue to increase the understanding necessary for Implications for D ecision-M aking lengthening life-span and decreasing cancer death rates.*123456 For all o f these considerations, our scale is not a scale o f risks to humans but is only a way o f setting priorities for concern, which should also take into account the numbers o f people exposed. It should be emphasized that it is a linear scale and thus may 'Overestimate low potential hazards if, as wc argue above, linearity is not the normal case, or if nongenotoxic carcinogens are not o f very much concern at doses much below the toxic dose. Thus, it is not scientifically credible to use the results from rodent done at the MTD to directly estimate human risks at low doses. ">ram ple, an EPA "risk assessment" (92) based on a succession o f worst case assumptions (several o f which are unique to EDB) included that EDB residues in grain (HERP = 0.0004% ) could ^use 3 cases o f cancer in 1000 people (about 1% o f all U.S. cancer). ** ^nsequcnce was the banning o f the main fumigant in the ^ n try . It would be more reasonable to compare the possible "^rd o f EDB residues to that o f other common possible hazards. REFERENCES AND NOTES 1. R. Doll and R. Peto, The Causa o fCancer (Oxford Univ. Press, Oxford, England, 1981). 2. Smoking and H ealth: A Report o f the Surgeon General, Department of Health, Education and Welfare Publication No. (PHS) 79-50066 (Office ofthe Assistant Secretary for Health, Washington, DC, 1979). 3. G. J. Hopkins and K. K. Carroll,/. Environ. Pathol. Toxicol. Oncol. S, 279 (198S); J. V. Joossens, M. J. Hill, J. Geboers, Eds., D iet and H um an Carcinogenesis (Elsevier, Amsterdam, 1985); L Knudsen.Ed., Genetic TccdtologyafthtD iet (Liss, New York, 1986); Committee on Diet, Nutrition and Cancer, Assembly of Life Sciences, National Research Council, D iet, N utrition and Cancer (National Academy Press, Washington, DC 1982). 4. R. P. Bird, R. Schneider, D. Stamp, W. R. Bruce, Carcinogenesis 7,1657 (1986); H. L. Netmurk ee ai,, in Large Board Cancer, vol. 3 in Cancer Research Monographs, A. J. Mastromarino and M. G. Brattain. Eds. (Praeger, New York, 1985), pp. 102-130; E. A. Jacobson, H. L. Newmark, E. Bright-See, G. McKcown-Evssen, W. R. Bruce, Nucr. Rep. Ins. 30, 1049 (1984); M. Busct, M. Lapkin, S. Winawer, S. Swaroop, E. Friedman, Cancer Res. 46, 5426 (1986). 5. D. G. Hocl, R. A. Merrill. F. P. Pcrera, Eds., Banbury Report 19. Risk Q uantitation and Regulator, Polio (Cold Spring Laboratory, Cold Spring Har bor, NY, 1985). ' <i p - , o r ' 6. B. E. Henderson et al.. Cancer Res. 42, 3232 (1982). JL cl? 'O U VOL. I 17 ^ R l L 1987 ARTICLES 277 7. R. Peto and H. zur Hausen, Eds., Banbury Report 21. Viral Etiology ofCervical Cancer (Cold Spring Harbor Laboraoocy, Cold Spring Habor, NY, 1986); F.*$. Ych et aL, Cancer Res. 45, 872 (1985). 8. International Agency for ResearchoaCatKcr,IARCMonographson theEvaluation ofthe CarcinogenicRisk ofChemicalstoHumans(International Agencyfor Research on Cancer, Lyon, France, 1985), voL 39. 9. D. A. Freedman andH. Trasd,FromMousetoMan: TheQuantitativeAssessmentof Cancer Risks (Tech. Rep. No. 79, Department of Statistics, University of California, Berkeley, 198/). 10. R*VexQ,\nAssessmentofRjskfhmLo^Ltvd Exposure to Radiation and Chemicals, A. D. Woodhcad, C. J. SheUabargcr, V. Pond, A. Hollaender, Eds. (Plenum, New York and London, 1985), pp. 3-16. 11. S. W. Samuels and R. H. Adamson,/. Nad. Cancer Inst. 74, 945 (1985); E. J. Calabrese, Drug Metah. Rev. 15, 505 (1984). 12. B. N. Ames, Science 221, 1256 (1983); Aid. 224, 668, 757 (1984). 13. H. Ohgaki etal., Gann 75,1053 (1984); S. S. Mirvish,/. Nad. CancerInst. 71, 630 (1983); J. V. Joosscns and J. Geboezs, inFrontiers m Gastrointestinal Cancer, B. Levin and R. H. Riddell, Eds. (Elsevier, Amsterdam, 1984), pp. 167--183; T. Hirayama,/p./. Clin. Oncol. 14,159 (1984); C Furihata ctal.,Biochem. Biophys. Res. Common. 121, 1027 (1984). 14. R. Peto, M. C Pike, L. Bernstein, L. S. Gold, B. N. Ames, Environ. Health Perspea. 58,1 (1984); L. S. Gold ref., Aid., p. 9; L. S. Gold et al.. Aid. 67,161 (1986); L. S. Gold et al., ibid., in pres. 15. G. M. Williams and J. H. Wcisburger, in Gaamr andDoulPsToxicology. The Basic Science( fPoisons, C. D. Klaassen, M. O. Amdur, J. Doull, Eds. (Macmillan, New York, ccL3, 1986), chap. 5, pp. 99-172; B. E. Butterworth andT. J. Slaga, Eds., Banbury Report 25. Non-Gcnotaac Mechanisms in Carcinogenesis (Cola Spring Harbor Laboratory, Cold Spring Harbor, NY, 1987). 16. The FDA has estimated the average U.S. dietary intake of 70 pesticides, herbicides, and industrial chemicals for 1981/1982 [M. J. Gartrcli, J. C. Craun, D. S. Podrebarac, E. L. Gunderson,J.Assoc. OffAnaL Chem. 69, 146 (1986)]. The negative test on 2-ethylhexyl diphenyl phosphate is in J. Treon, F. Dutra, F. Cleveland, Arch. Ind Hvg. Ocatp.Med 8, 170 (1953). 17. R. C Beier et al., Food'Chem. Taxied. 21, 163 (1983). 18. L. Sroioff, M. Castcgnaro, P. Scoa, L K. O'Neill, H. Bartsch, Eds., Some Mycotaodns, voL 5 inEnvironmentalCarcinogens. SelectedMethodsofAnalysis (IARC ' Scientific PubL No. 44, International Agency for Research on Cancer, Lyon, France, 1982); H. Mori etaL, CancerRes. 44,2918 (1984); R. ROschenchalcr, E. E. Creppy, G. Dirheimer,/. Taxicoi.-TaxmRev. 3, 53 (1984); W. F. O. Marasas, N. P. J. Kriek, J. E. Fincham, S. J. van Rcnsburg, Ins. J. Cancer 34, 383 (1984); Environmental Health Criteria 11: Myootoxats (World Health Organization, Geneva, Switzerland, 1979), pp. 21-^85; W. F. Busby ct al., in Chemical Carcinogens, C. E. Scarle, Ed (ACS Monograph 182, American Chemical Society, Washington, DC, ed. 2,1984), voL 2, pp. 944-1136. 19. S. J. Van Rcnsburgrr/.,Br./. Cancer51,713 (1985); 5. N. Taman etoL,Lancet 1985-1, 1357 (1985); H. Austin et aL, CancerRes. 46, 962 (1986). 20. A. Takada, J. Nci, S. Takase, Y. Manuria, Hepatology 6, 65 (1986). 21. J. M. Elwood et al., Int. /. Cancer 34, 603 (1984). 22. Aldehydes and ketones are largely responsible for the aroma and flavor of bread [Y. Y. Linko, J. A. Johnson, B. 5. Miller, Cereal Chemistry 39, 468 (1962)]. In freshly baked bread, formaldehyde (370 m per two slices of bread) accounts for 2.5% of the total carbonyl compounds (kL Lorenz and J. Maga, J. Agric. Pood Chem. 20, 211 (1972)]. Acetaloehyde, which is present in bread at about twice die level of formaldehyde, is a carcinogen in rats [R. A. Woutcrsen, L. M. Appdman, V. J. Feron, C A. Vandohe^den, Toxicology 31, 123 (1984)] and a DNA cross-linking agent in human crib [B. Lambert, Y. Chen, S.-M. He, M. Sten, Mutat. Res. 146, 301 (1985)]. ^23. Ethyl alcohol contents of wine and beer were assumed to be 12% and 5%, respectively. The TD50 calculation is based on M. J. Radike, K. L. Stemmer, E. Bingham, Environ. Health Perspea. 41, 59 (1981). Rats exposed to 5% ethyl alcohol in drinking water for 30 months had increased incidences of endocrine ^ and liver tumors. 24. C. S. Ou#\,J.Agric. FoodChem. 24,323 (1976). Urethane is alsocarcinogenic in hamsters and rhesus monkeys. 25. Y. Fujita, K. Wakabayashi, M. Nagao, T. Sugunur% Mutat. Res. 144, 227 (1985); M. Nagao, Y. Fujita, T. Sugimura, in IARC Workshop, in press. 26. M. Petro-Turza and I. Szarfoldi-Szalma,A m Aftmcnnma 11, 75 (1982). 27. L. J. Mamctt a al., Mutat. Res. 148, 25 (1985). 28. Formaldehyde in air samples taken from all the mobile homes examined ranged from50 to 660 ppb (mean, 167ppb) (T. H. Connor, J. C.Theiss, H. A. Hanna, D. K. Monteith, T. S. Mamey, Taxied Lea. 25,33 (1985)]. The important role ofcell toxicity and cell proliferation informaldehyde carcinogenesis is diynwtj in T. B. Starr and J. E. Gibson [Amu. Rev. Pharmacol. Toxicol. 25, 745 (1985)]. 29. J. A. Swenberg a al., Caranogenehs 4, 945 (1983). 30. G. Della Porta, M. L Colnaghi, G. Parmiani, FoodCasmce. Toxicol. 6,707 (1968). 31. Formaldehyde develops postmortem in marine fish and crustaceans, probably through the metabolism of trimethyiamine oxide. The average level found in shrimp from four U.S. markets was 94 mgrikg (T. Radford ana D. E. Daisis, f. Agric. Food Chem. 30, 600 (1982)]. Formaldehyde is found in-remarkably high concentrations (300 ppm, HERP 3 29%per 100g) in Japaneseshrimp that have been bleached with a sulfite solution [A. Xoshida and M. Imaida,/. FoodHygienic Soe. Japan 21, 288 (1980)]. 32. J. F. Lawrence and J. R. Ivcngar, Int. J. Esmrm. Anal. Chem. 15, 47 (1983). 33. H. d'A Heck a al.. Am. ind. Hyg. Assoc./. 46, 1 (1985). 34. T. Sugimura et al., in Genetic Taaedagy ofthe Diet, I. Knudsen, Ed. (Liss, New York, 1986), pp. 85-107; T. Sugimara, Sdena 233, 312 (1986). 35. H. Ohgaki et Cancer Lea. 2S, 239 (1985). 36. T. Kinouchi, H. Tsutsui, Y. Ohnishi, Mutat. Res. 171, 105 (1986). 37. T. KawabacaetaL, in N-NitnxoCompounds:Analysis, Formation and Occurrence, E. A. Walker, L Griciute, M. Castcgnaro, M. Borzsonyi, Eds. (IARC Scientific Pub!. No. 31, International Agencv for Research oh Cancer, Lyon, France, 1980), pp. 481--490; T. Maki, Y. Tannin, Y. Shimamura, and Y. Naot [BulL Environ. Contam. Toxicol. 25, 257 (1980)] have surveyed Japanese food for nitrosamines. 38. T. Fazio, D. C Havcry, J. W. Howard, in N-Nitraso Compounds: Analysis, Formation and Occurrence, E. A. Walker, L. Griciute, M. Castcgnaro, m! 7 Borzsonyi, Eds. (IARC Scientific Pubi. No. 31, International Agency f0r - Research on Cancer, Lyon, France, 1980), pp. 419-435; R. Prcussmann and G. * Eisenbrand, in Chemical Carcinogenesis, C. . Searle, Ed. (ACS Monograph 182* * American Chemical Society, Washington, DC ed. 2,1984), voi 2, pp. 829-868* 3. D. C. Havery, J. H. Hotchkiss, T. Fazio,/. Food So. 46, 501 (1981). '* 39. W. A. Garland a al., Cancer Res. 46, 5392 (1986). ` 40. E. A Walker, L Griciute, M. Castcgnaro, M. Borzsonyi, Eds., N-Nirrc - j Compounds: Analysis, Formation and Occurrence (IARC Scientific Pubi No. 31 ^ International Agency for Research on Cancer, Lyon, France, 1980), pp. 457- 2 463; B. Spicgelhalder, G. Eisenbrand, R. Prcussmann, Oncology 37,211 (1980); 2 ' R_ A. Seaman and S. R. Tannenbaum, Eds,, N-Nitroso Compounds (ACS Symposium Scries No. 174, American Chemical Society, Washington, DC, & 1981), pp. 165-180. Nitrosamines are formed in cured meats through reactions of seconaary amines with nitrites added during the manufacturing process. One % survey of bacon commercially available in Canada identifiedN-nicrosodimcthvU- mine (DMN), N-nitrosodicmylaminc (DEN), andN-nicrosopyrrodinc (NPYR) in most samples tested, with average levels of 3.4,1.0, and 9.3 ppb, respectively. ~ The cooked'ouc fat from the bacon samples contained DMN and NPYR at 3 average levels of 6.4 and 21.9 ppb, respectively [N. P. Sen, S. Seaman, W. F. ^ Miles, /. Agric. Food Chem. 27, 1354 (1979); R. A. Scanian, Cancer Res. 43, 2435s (1983)]. The average levelsofNPYRin cooked bacon havedecreased since 1971 because of reduced levels of nitrite and increased levelsof ascorbate used in bacon curing mixtures [D. C Havery, T. Fazio, J. W. Howard,.Assoc. Off.Anal. Chem. 61, 1379 (1978)]. 41. Y. Yamamoto a al.,AnaL Biochem. 160, 7 (1987). 42. B. N. Ames and R. L Saul, in Theories of Carcinogenesis, O. H. Jversen, Ed. (Hemisphere, New York, in press); R. Carnean, E. Schwiers, R. L. Saul, B. N. Ames, Proc. Nad. Acad. So. USA.. 81, 5633 (1984). 43. B. P. Yu, E. J. Masoro, I. Murata, H. A. Bertrand, F. T. Lvnd,/. Gerontol. 37,130 (1982) ; F. J. C Roe, Pro. Nutr. Soc. 40, 57 (1981); Nature (London) 303, 657 - (1983) ; M. J. Tucker, Int. /. Gwurr23, 803 (1979). 44. Y. Tazima, Environ. Health Perspea. 29,183 (1979); M. Kinebuchi, T. Kawachi, N. Masukura, T. Sugimura, Food Cosmee. Taxied. 17, 339 (1979). 45. F. W. Cariborg, FoodChem. Taxied. 23, 499 (1985). 46. T. H. Jukes, Am. Stat. 36, 273 (1982); /. Am. Med Assoc. 229, 1920(1974). f 47. Alivi isochiocyanate (ATTQ is the major flavor ingredient, and natural pesticide, ^ of brown mustard and also occurs nanirelhr in varving concentrations in cabbage, kale, broccoli, cauliflower, and horseradish [Y. M. Ioannou, L. T. Burka, H. B. v" Matthews, Taxied. Appi. Pharmacol. 75, 173 (1984)]. It is present in the plant's volatile oil as the giucoside sinigrin. (The primary flavor ingredient of yellow mustard isp-hydroxybcnzyl isothiocyanacc.) The ATTCyield frombrown mustard ? is approximately 0.9% by weight, assuming all of the sinigrin is converted to ^ ATTC [A. Y. Leung, Encyclopedia ofCommon Natural Ingredients Used in Food, ^ Drugs and Cosmetics (Wiley, New York, 1980), pp. 233-241]. Synthetic ATTCis ^ used in nonalcoholic beverages, candy, baked goods, meats, condiments, and s svrups at average levels ranging from 0.02 to 88 ppm [T. E. Furia and B. Nicolo, Eds., FenarolPsHandbookofFlavorIngredients, (C&C Press, Cleveland, OH, 2 ed, 1975), voL 1, p. 19]. ^ 48. Estragle, one of numerous safroie-like compounds in plants, is present in the z voladle oilsofmanyedible plants, including basil, tarragon, bay, anise, and fennel ? aswell as in pine oUand turpentine [A. Y. Leung, EncyclopediaofCommonNatural g Ingredients Used in Food, Drugs and Cosmetia (Wiley, New York, 1980)]. Dried % basil has a voladle oil content of about 1.5 to 3.0%, which contains (on average) S 25% estragle [H. B. Heath, Source Book ofFlavors (AVI, Westport, CT, 1981), % pp. 222-223]. Estragle is used commercially in spice, anise, licorice, and fruir Savors. It is added to beverages, candy, baked goods, chewing gums, ice creams, and condiments at average levels ranging from 2 to 150 ppm [NAS/NRC Food ^ Protection Committee, Food and Nutrition Board, Chemicals Used in Food ^ Processing(NAS/NRC Pubi No. 1274, Nadonai Academy of Sciences, Washing* ron, DC, 1965), p. 114]. 49. The esdmadon of risk is from human data on uranium miners and estimates of t- intake. E. P. Radford, Environ. HealthPernea. 62,281 (1985); A V. Nero ct aL, Science 234, 992 (1986); A V. Nero, Tecbnd. Rev. 89, 28 (1986); R- Hanley, f ; TheNew Tor Times, 10 March 1986, p. 17. % 50. The average daily adult dose of phnobarbital for sleep induction is 100 to 320 9 mg (HERP = 26 to 83%), though its use is declining [AMA Division of Drugs f AMA Drug Evaluations (American Medial Assodadon, Chicago, IL, ed. 5, ^ 1983), pp. 201-202]. The TD data in the table is for phnobarbital which, so ^ 6ir, has been shown to be carcinogenic only in mice; the sodium salt of ^ phnobarbital is carcinogenic in both rats and nuce. Human studies on phenobar* Dirai and cancer are reviewed in A E. M. McLean, H. E. Driver, D. Lowe, L ^ Sutherland, Taxied. Lea. 31 (suppl), 200 (1986). ,^ 51. Phenacedn use has gradually decreased following reports of urinary bladder and 2 kidney tumors in heavyusen (J. M. Piper, J. Tonasda, G. M. Matanoski,N. Engl- J. Med 313, 292 (1985)]. Phenaccnn also induces urinary bladder and kidney g tumors in rats and mice. J 52. The human dose of dofibrate is 2 g per day for many years [R. J. Havel and J- * Kane, Annu. Rev. Med 33, 417 (1982)]. The role of dofibrate as a peroxisome m proliforaror is reviewed in J. K. Reddy and N. D. Lalwani [CRC Crit. Rf9* j | Taxied. 12,1 (1983)]. An epidemiologic study is in World Health OrgamzaDon M Report, Lanca 1984-H, 600 (1984). ^ 53. L. S. Gold, G. Backman, N. K. Hooper, R. Peto, Lawrence Berkeley Laborase* S Repore23161 (1987); N. K. Hooper and LS. Gold, mMonisoringofOasipatjotid - Genotaxicana, M. Sorsa and H. Norppa, Eds. (Lisa, New York, 1986), pp- a ll"" M 228; K. Hooper and L. S. Gold, in CancerPrevention: Strategies m the Workplace, 2 C Becker, Eo. (Hemisphere, Washington, DC 1985), pp. 1--11. 54. California Department of Health Services, EDB Criteria Document (1985)- ^ 55. M. G. Ott, H. C J. C Ramsey, C Schamweber, R. R. Langner, Br./. Ind Med 37, 163 (19*' N. Park, M. G. Ott, P. J. Gehring, Taxied.Appi. Pharmacd. 47. 4 411 (1978). This has been disputed (54). The carcinogen dose reported in table assumes a time-weighted average air concentration of 3 ppm and an 8*hottf 2 workday 5 days per week for 50 weeks per year for life. ^ 278 P 1 SCIENCE, VOL. -i 1O * ca R. Maeaw, Li S. Gold, L. Bernstein, T.'H. Slone, B. N. Ames, in preparation, s? L Bernstein, L. S. Gold, B. N. Ames, M. C. Pike, D. G. HoeL Fumum. Appl. 9`' l'aiaL S, 79 (198S); L. Bernstein, L. S. Gold, B. N. Ames, M. G Pike, D. G. Hod, Risk Anal. 5, 263 (1985). 5g D. B. Qayson, Taxied. Pathol 13, 119 (1985); O. B. Claysoa. Mutt. Res., in g n 0 . S. E. Parish, R. G. Gray, in Age-Related Foam in Carcinogenesis, A. Likhachev, V. Anisimov, R- Montcsano, Eds. (IARC Scientific PubI No. 58, International Agency for Research on Cancer, Lyon, France, 1985), pp. 43-53. 50. D. M. ShankcL, P. Hartman, T. Kada, A. Hollacnder, Eds.,Antanutagenesis and ' Anticarnnogcncsir. Mechanisms (Plnum, New York, 1986). 51 B. N. Ames and J. McCann, Cancer Bar. 41, 4192 (1981). R- A. Weinberg, Same 230, 770 (1985). 53. A. G. Knudson, Jr., CancerSot. 45, 1437 (1985). 51' }. B. Geaver, in Gena and Conter, J. M. Bishop, J. D. Rowlev, M. Greaves, Eds. ' (Liss, New York, 1984), pp. 117-135. 55. A. D. Woodhead, G ). Shdlabatgcr, V. Pond, A. Hollacnder, Eds,,Assessmentof Risk from lasv-Lcvcl Exponat to Radiation and Chemieais: A Crixieai Overview Plenum, New York, 1985). 55. J. Caims, Nature (London) 255, 197 (1975); G G Karris and T. Sun, Carcinogenesis5,697 (1984); A. M. Edwards and G M. Lucas, Biecbem. Biophys. Res. Commun. 131, 103 (1985); H. Tsuda et al., CaneerRxs. 39, 4491 (1979); W. H. Haese and E. Bisecting,J. Pharmacol. Exp. Ther. 197, 703 (1976). 57. J. E. Trosko and G G Chang, inMethodsfor Estimating Risk ofChemicalInjury: Human and Non-Human Biota and Ecosystems, V. B. Vouk, G G Buder, D. G. HoeL D. B. Peakail, Eds. (Wiley, New York, 1985), pp. 181-200; J. E. Trosko and G G Chang, inAssessment ofRisk from Low-Level Exposure to Radiation assd Chemicals:A Critical Overview, A. O. Woodhcad, G ). Shcllabarger, V. Pond, A. Hollacnder, Eds. (Plenum, New York, 1985), pp. 261-284; H. Yamasaki, Toxicol. Pathol. 14, 363 (1986). 68. M. F. Rajewsky, in Age-Related Factors in Carcinogenesis, A. Likhachev, V. Anisimov, R. Montcsano, Eds. (IARC Scientific PubL No. 58, International Agency for Research on Cancer, Lyon, France, 1985), pp. 215-224; V. Kinsel, G. Furscenberger, H. Lochrke, F. Vlarks, Carcinogenesis 7, 779 (1986). 69. E. Frber, CancerRes. 44,5463 (1984); E. Frber, S. Parker, M. Gruensrein, ibid. 36, 3879 (1976). 70. A. Denda, S. Inui, M. Sunagawa, S. Takahashi, Y. Konishi, Gann 69, 633 (1978); R. Hasegawa and S. M. Cohen, Cancer Lett. 30, 261 (1986); R. Hasegawa. S. M. Cohen, M. St. John, M. Cano, L. B. Ellwein, Carcinogenesis 7, 633 (1986); B. L Ghanavem, R. R. Maronpot, H. B. Matthews, Taxicdpgy6,189 (1986). 71. J. G Mirsalis et al.. Carcinogenesis 6, 1521 (1985); J. G Mirsalis etai.. Environ. Mutag. 8 (suppL 6), 55 (1986); J. MirsalisetaL, Abstract for Fourth Intemation- al Conference on Environmental Mutagens, held 24-28 June in Stockholm, Sweden (1985). 72. W. T. Stott, R. H. Reitz, A- M. Schumann, P. G. Wannabe, FoodCosmet. Tceciccl. 19, 567 (1981). 73. D. H. Moore. L. F. Chasscaud, S. K. Majccd, D. E. Prentice, F. J. G Roc, ibid. 20, 951 (1982). 74. '. K. Haseman, J. Huff; G A. Boorman, TacicoL Pathol. 12, 126 (1984); R. E. Tarone, K. G Chu, J. M. Ward, f. NatL CancerInst. 66, 1175 (1981). 75. S. H. Reynolds, S. J. Stowen, R. R. Maronpot, M. W. Anderson, S. A. Aaronson,"Proc. Nad Acad S. UBA. 83, 33 (1986); T. R. Fox and P. G. Watanabe, Science 2^8, 596 (1985). 76. T. D. Jones, Health Phys. 4, 533 (1984); J. B. Litde, A. R_ Kennedy, R. B. McGandy, Radias. Res. 103, 293 (1985). 77. T. W. Knsler and B. G. Taffe,Ad. FreeRadical Biol. Med 2,347 (1986); P. A. Cerutd, in UCLA Symposium on Molecular and Biology Growth Factors, Tumor Promoters and CancerGenes, in press; P. A. Cerutti, in Biochemical sendMolecular Epidemiology ofCanter, voL 40 of UCLA Symposium on Molecular and Cellular Biology, G Harris, Ed. (Liss, New York, 1986), p. 167; in Theories ofCarcino genesis, O. H. Iversen, Ed. (Hemisphere, New York, in press); H. G Bimboim, Carcinogenesis 7, 1511 (1986); K. Frenkel and K. Chrzan, ibid. 8, 455 (1987). 78. J. Rotstein, J. O. O'Connell, T. Slaga, Proc.Assoc. CancerRes. 27,143 (1986); J. S. O'Connell, A J. P. Klein-Szanto, J. DiGiovanni, J. W. Frics, T. J. Slaga, Cancer, lies. 46, 2863 (1986); J. S. O'Connell, J. B. Rotstein, T. J. Slaga, in Banbury Report 25. Non-Genotoxic Mechanisms in Cccrcinpgenesis, B. E. Butterwotth and T. J. Slaga, Eds. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1987). 79. M. A. Trush, J. L. Seed, T. W. Kcnsler, Proc. Nad. Acad S. USA.. 82, 5194 (1985); A I. Tauber and B. M. Babior, Adv. Free-Radical Biol. Med 1, 265 (1985); G. J. Chellman, J. S. Bus, P. K. Working, Proc. Nad. Acad S U SA. 83, 8087 (1986). 80. I. . Schraufsntter et aL, Proc. Nad Acad S. U SA. 83, 4908 (1986); M. O. Biadicv, in Basic andAppliedMutagenesis, A Muhammcd and R. C. von Boistd, Eds. (Plenum, New York, 1985), pp. 99-109. 81. L. Diamond, T. G. O'Brien, W. M. Baird, Adv. Cancer Res. 32, 1 (1980); D. Schmahl, /. Cancer Res. Clin. Oncol. 109, 260 (1985); O. H. Iversen and E. G. Astrup, CancerInvest. 2,51 (1984); A Hagiwara and J. M. Ward, Fundam.Appl Taxied. 7,376 (1986); J. M. Ward, in CarcmogcnesisandMutagenesesTesting, J. F. 82. Douglas, Ed. (Humana, Gifton, NJ, 1984), pp. L. Zeise, R. Wilson, E. Crouch, Risk Analysis 4, 97-100. 187 (1984); L. Zeise, E. A C. Crouch, R_ Wilson, ibid 5,265 (1985); L. Zeise, E. A. G Crouch, R. Wilson,/. 83. 84. Am. College Toxicol 5, 137 (1986). D. Hoel, personal communication. Ito, S. Fukushima, A Hagiwara, M. Shibata, T. Ogiso, }. Nad Cancer Inst. 85. 70, 343 (1983). F-W. Cariborg, FoodChem. Toxic. 20,219 (1982); FoodCosmet. Taxied. 19,255 (1981). 86. G. Brown and D. G. Hoch Fundam. Appl. Taxied. 3. 470 (1983); N. A 87. Lrtdefield and D. W. Gavlor,/. TaeoL Environ. Health J. Ashbv, Mutagenesu L, 3 (1986). IS, 545 (1985). 2" Cancer Res. 38, 3573 (1978); _____ and R. Peto, /. Epidemid. 'immunity Health 32, 303 (1978). '7 Ap r il 1987 89. J. E. Huff, E. E. McConnell, J. K. Haseman, Environ.Mutagenesis 7 ,427 (1985); H. S. Roscnknnz, ibid., p. 428. 90. M. H. Harvey, B. A Morris, M. McMillan. V. Marks, Human Taeicol 4, 503 (1985). 91. S. J. Jadhav, R. P. Sharma, D. K-Salunkhe, CR.CCrit.Rer. Taxied. 9 ,21 (1981). 92. Environmental Protection Agency, PositionDocument4 (Special Pesticide Review Division, Environmental Protection Agency, Arlington, VA 1983). 93. R. Wilson and E. Crouch, RiskIBenejit Analysis (Ballinger, Cambridge, MA, 1982); W. F. Allman Science 85 6, 30 (1985).' 94. P. Huber, Regulation, 33 (March/April 1984); G Whipple, ibid. 9, 37 (1985). 95. B. A Bridges, B. E. Butterwotth, I. B. Weinstein, Eds., Banbury Report 13. Indicators of Genoteocic Exposure. (Cold Spring Harbor Laboratory, Cola Spring Harbor, NY, 1982); P. E. Enterline. Ed., Filth Annual Symposium on Environ mental Epidemiology, Environ. Health Pcrsptct. 62,239 (1985). 96. Anational surveyof U.S. drinking water supplies identified the concentrations of about 20 organic compounds. The mean total trihalomcthane concentration was 117 iig/lircr, with the major component, chloroform, present at a mean concen tration of 83 |ig/liter (83 ppb). Rawwater that is rdariveiv free oforganic matter results in drinking water pdativclv free of ofrhalomcthanes after chlorination. Thesestudies arcreviewed in S. J. Williamson, TheScienceofthe TotalEnvironment 18. 187 (1981). 97. Public and private drinking water wells in Santa Clara Valley, California, have been found to be contaminated with a variety of halogenatect hydrocarbons in small amounts. Among 19 public water systemwells, the most commonly found contaminants were 1,1,1-tridiloroethane (TCA), and 1,12-trichloro-1,2,2-tnfiuoroethane (Freon-113). TCAwas found in 15 wells generally at concentrations of less than 30 ppb, though one well contained up to 3800 ppb, and Freon-113 was found in sixwells at concentrations up to 12 ppb. Neither chemical has been adequately tested for carcinogenicity in long-term bioassays. In addition to these compounds, threewells alsocontained catrinogenic compounds at lowconcentra tions. Water from public supply wells mav be mixed with treated surface water before delivery, thus the concentrations of these compounds that people actually receive may be somewhat reduced. Thirty-five private drinking water supply wells were examined; the major contaminant was the carcinogen trichloroethylene (TCE), at levels up to 2800 ppb. TCA and Freon-113 were also found in some wells, at maximum levels of 24 ppb and 40 ppb, respectively. Though fewer people drink from private water wells, the contaminant concentrations may be higher because the water is not mixed with water from other sources [California Department of Health Services, California Regional Water Quality Control Board 2, Santa Clara County Public Health Department, Santa Clara Valley Water District, U.S. Environmental Protection Agency, Ground Water and Drinking Water in the Santa Clara Valley: A White Paper (1984), table 8]. ' Trichloroethylene may not be a carcinogen in humans at low doses [R. D. Kimbrough, F. L. Mitchell, V. N. Houk, /. Taxied. Environ. Health 15, 369 (1985)]. 98. Contaminateddrinkingwaterin the area ofWoburn, Massachusetts, was found to contain 267 ppb trichloroethylene, 21 ppb tetrachloroethyiene, 12 ppb chloro form, 22 ppb trichlorotrifiuoroethane, and 28 ppb 12-traw-dichloro<nhyiene [S. W. fjgakns, B. J. Wessen, M. Zclen, J. Am. Stat. Assoc. 81, 583 (1986)]. 99. The amount of chloroform absorbed by a 6-ycar-old child in a chlorinated freshwater swimming pool has been estimated (J. A Beech, Med. Hypotheses 6, 303 (1980)1. Table 1 refers to the chloroform in an average pool (134 p-g/lirer) and for a 37-kg child. Three other trihalomethanes were identified in these freshwater pools: bromofotm, bromodichloromcthane and chlorodibromomethane. U. Laid, J. Vondusze, B. Gabel B. Scadid, W. Thiemann [Water Res. 15, 803 (1981)] have estimated absorption in covered swimming pools. 100. J. McCann, L. Horn, J. Girman, A V. Nero, inShort-Term Bioassttysin theAnalysis ofComplexEnvironmentalMixtures, V. S. Sandhu, D. M. De Marini M. J. Mass, M. M. Moore, J. L. Mumfbrd, Eds. (Plenum, New York, in press). This estimate (Table 1) for formaldehyde in conventional homes, excludes foam-insulated houses and mobile homes. The figure is a mean of the median or mean of the reported samples in each paper. For benzene, the figure is a mean of all reported median or mean samples. The level of benzene in Los Angeles outdoor air is similar (U.S. EPAOffice of Air Quality Planning and Standards, EPA 450/4-86012, 1986). 101. The average adultdaily PCB intake from food estimatedby the FDA in fiscal years 198171982 was 0.2 irg/dav (76). Many slightlydifferent PCB mixtures have been studied in long-term animal cancer bioassays; the calculation ofTD was from a test of Arodor 1260 which was more potent than other PCBs (74). 102. The average consumption of EDB residues in grains has been estimated bv the EPA for adults as 0.006 yg kg' 1day' 1and for children as 0.013 izg kg' 1day' 1 [U.S. EPA Office of Pesdciae Programs, Ethylene Dibromide (EDB) Scientific Support and Decision Documentfor Grain and Grain Milling Fumigation Uses (8 February 1984)]. 103. The leaves and roots of Russian comftcv are widely sold in health food stores and arc consumed as a medicinal herb or salad plant or are brewed as a tea. Comftcv leaf has been shown to contain 0.01 to 0.15%, by weight, total pvnolizidine alkaloids, with an avenge level of 0.05% for intermediate size leaves [C. G J. Culvenor, J. A Edgar, J. L. Frahn, L. W. Smith, Aust. J. Chem. 33, 1105 (1980)]. The main pyrrolizidine alkaloids present in comftcy leaves are echimidinc and 7-acetyilvcopsamine, neither of which has been tested for carcinogenic ity. Almost all tested 1,2-unsaiuntcd pyrrolizidine alkaloids have been shown to be genotoxic and carcinogenic [H. Mon et al. Cancer Res. 45, 3125 (198S)]. Symphydne accounts for 3% ofme total alkaloidin the leaves and has been shown to be carcinogenic [G G J. Culvenor et al, Experientia 36, 377 (1980)]. We assume thac 1.5 g of intermediate size leaves are used per cup of comftcy tea (Table 1). The primary alkaloids in comftcy root are symphvdne (0.67 g per kilogramof root) and echimidine (0.5 g per kilogramofroot) [T. FuruvaandM. Hitachi, Phytochemistry 10, 2217 (1971)]. Comfrcy-pepsin tablets (300 mg of root per tablet) have a recommended dose of one to three tablets three times per day. Comftcy roots and leaves both induce liver tumors in rats [I. Hirono, H. Mori. M. Haga,/. Nad CancerInst. 61, 865 (1978)1, and the TD value is based on these results. Those pyrrolizidine alkaloids tested nave been found to be at least 1 S 6 C 4 ARTICLES 279  aspotent ascarcinogenssuch as symphytinc. Ifthe other pyrrolizidinc alkaloids in comfrevwere as potent carcinogens as symphydne, the possible hazard ofa daily cup of tea would be HEAP * 0.6% and that of a daily nine tablets would be HERJ? - 7.3%. 104. Aganau bitponuis the most commonly eaten mushroom in the UnitedStateswith an trimarrrf annual consumption of 340 million kilograms in 1984-85. Mush rooms contain various hydrazine compounds, some ofwhich have been shown to cause turnon in mice. Raw mushrooms fed over a lifetime to male and female miceinduced bone, forestomach, liver, and lung turnon [B. Toth and I. Erickson, Canaria. 46, 4007 (1986)]. The 15-g raw mushroom is given as wet weight. The TDjo value based on the above report is expressed as dry weight of mushroomsso as to be comparable to other values for TDn inTable 1; 90%ofa mushroom is assumed to be water. A second mushroom, Gyivmitiu tteulentn, has been similarlystudied and found to contain a mixture of carcinogenic hydrazines [B. Toth, J. Environ. So. Health C2, 51 (1984)]. These mushrooms arc eaten in considerable quantities in several countries, chough less frequently in the United States. 105. Safioieis the maincomponent (up to 90%) ofoilofsassafras, formerlyusedas the main flavor ingredient m root beer [J. B. Wilson, /. Auoc. Off. Anal Chem. 42, 696 (1959); A. Y. Leung, Encyclopedia of Common Natural Ingredient! Used m Food,Drug!andCotmetia (Wiley, New York, 1980)]. In 1960, safrole and safrole. containing sassafras oils were banned from use in foods in the United Sates Bjgitt. 25, 12412 (I960)]. Safrole is also naturally present in the oils of sweet basil, cinnamon leaf, nutmeg, and pepper. 106. Diet cola available in a local market contains 7.9 mg ofsodiumsaccharin per fluid ounce. 107. Metronidazole is considered to be the drug of choice for trichomona! and CardntreUn infections (AMA Division of Drugs,AMA Drug Evaluation! (Ameri can Medical Association, Chicago, IL, ed. 5, 1983), pp. 1717 and 1802], 108. Isoniazid is used both prophylacdcally and as a treatment for active tuberculosis. The adult prophylactic dose (300 mg daily) is continued for 1 year (AMA Division of Drugs, AMA D nu Evaluation! (American Medical Association. Chicago, IL, ed. 5, 1983), pp. 1766-1777], 109. D. M. Sicgal, V. H. Frankos, M. A. Schncidcrman, Keg. Toxintl Pharmaod. 3 355 (1983). 110. Supported by NCI Outstanding InvestigatorGrant CA39910 to B.N.A., NIEHS Center Giant ES01896, and NIEH5/DOE InteragencyAgreement 222-Y01-ES10066. We are indebted to numerous colleagues for criticisms, particularly W. Havendcr, R. Peto, ]. Caims, J. Miller, E. MiUer, D. B. Clayson, J. McCann, and F. J. C Roe. Perception o f Risk P a u l S l o v ic Studies o f risk perception exam in e d ie jud gm ents people m ake w h en th ey are ask ed to characterize and evaluate hazardous activities an d tech n o lo g ies. T h is research aim s to aid risk analysis an d p o licy -m a k in g b y (i) p rovid in g a basis for u n d erstan d in g an d an ticip atin g p u b lic responses to hazards a n d (ii) im p ro v in g th e com m u n ication o f risk in form ation am on g lay p eop le, technical experts, and decision-m akers. T h is w o rk assu m es th at th ose w h o pro m ote and regulate h ealth an d safety n eed to understand h o w p eop le th in k a b o u t an d resp o n d to risk. W ith o u t such u nd erstan ding, w ell-in ten d ed p olicies m ay be inef fective. experience with hazards tends to come from the news media, which rather thoroughly document mishaps and threats occurring throughout the world. The dominant perception for most Ameri cans (and one that contrasts sharply with the views of professional riskassessors) is that they face more risk today than in the past and that future risks will be even greater than today's (2). Similar views appear to be hdd by drizens of many ocher industrialized nations. These perceptions and the opposition to technology that accompa nies them have puzzled and frustrated industrialists and regulators and have led numerous observers to argue that the American public's apparent pursuit of a "zero-risk sodety" threatens the nation's political and economic stability. Wildavsky (3, p. 32) commented as follows on this state of affairs. How extraordinary! The richest, longest lived, best protected, most resourceful civilization, with the highest degree of insight into its own technology, is on its way to becoming the most frightened. T he ability to sense and avoid harm ful environmental conditions is necessary for the survival of all living organisms. Survival is also aided by an ability to codify and Is it our environment or ourselves that have changed? Would people like us have had this sort of concern in the past?. .. Today, there arc risks from numerous small dams for exceeding those from nudear reactors. Why is the onefeared andnot the other? Is itjust that we arcused to the old or aresome lcam from past experience. Humans have an additional capaboilfituys looking differendy at essentially the same sorts of experience? that allows them to alter their environment as well as respond to it. This capacity both creates and reduces risk. During the past decade, a small number of researchers has been In recent decades, the profound development of chemical and attemptingto answersuch questions byexaminingthe opinions thac nudear technologies has been accompanied by the potential to cause people express when they are asked, in avariety ofways, to evaluate catastrophic andlong-lasting damage to the earth and the life forms hazardous activities, substances, and technologies. This researchhas that inhabit it. The mechanisms underlying these complex technolo attempted to develop techniques for assessing the complex and gies are unfamiliar and incomprehensible to most drizens. Their subtle opinions that people have about risk. With these techniques, most harmful consequences are rare and often delayed, hence researchers havesought to discoverwhat people meanwhentheysay difficult to assess by statistical analysis and not well suited to that something is (or is not) "risky," and to determine what factors management by trial-and-error learning. The elusive and hard to underlie those perceptions. The basic assumption underlying these manage qualities oftoday's hazards have forcedthe creationofanew efforts is thatthosewho promote andregulatehealthandsafety need intellectual discipline called risk assessment, designed to aid in to understand the ways in which people think about and respond to identifying, characterizing, and quantifying risk (1). Whereas technologically sophisticated analysts employ riskassess 19605 ment to evaluate hazards, the majority of drizens rely on intuitive riskjudgments, typically called "riskperceptions." For these people. The author is president of Decision Research, 1201 Oak Street, Eugene, OR 9/-Wk and professor of psychology at the University of Oregon. 280 S C IE N C E , V O L. I- Red; rrirr~rr Cause Cancer 2 0 THIS WORID, OCTOBER S. 1986 o&JCf t* w tn G iniC H kC hiii BYBRUCES. AMfS ?hecarcinogensand pesticideresidues currently being found InCalifornia '.! water supplies,such asinSiliconVal- I ley,are present in tiny amounts that, except inrarecases, aretrivial relativetothe back groundlevel ofcarcinogens innature. Therefore, 1amconvinced thatsuch water pollution isirrelevant asacauseof human cancer. The maincurrent fallacy inour ap proachtosuch pollution consistsin believ ingthat carcinogens--cancer-causingsubstances--arc rare and that theyare mostly man-made chemicals. Quitethe contraryis thecase. Myestimate isthat more than 99.99 percent of thecarcinogensCalifornians in gest are from natural sources(that is,sub stances normally present infood)or tradi tionalsources flhat is, cigarets, alcohol a n d .' chemicalsformed by cooking food). J- . . . . . . -W ... :.* The taneerjtausing h a za rd a/ a fish highly. confam ipafed withjPDT/DDE or PCB-- th a t is, UJO tim jifsth e efi/erage le v e l--is Every meal hasmany carcinogens, and ' when one compares the level of carcinogens incontaminatedwater or pesticide residues y .'t, , . . i . "- * , -.Vjt# ' \ .... com parable to the'hazarft of the average peanut bpfjet'jahdw ich. infood with the level of natural carcinogens alsopresent inthe diet, it isclear that water pollutionor pesticide residuesrepresent a trivial exposure by comparison! Water pollution and pesticideresidues ` are almost always present in the ppb(parts per billion) range. One part per billion(Le^ one person in all of China)isan exlraordl- - narilysmall amount. Bycomparison, the carcinogens in afewcommondrinks are ` listedbelow. Every commondrinkcontains'' carcinogens. - : v-.i i.ii.; teston rats must take into account the po tencyof the carcinogen inratsaswellaslhe human dose. Somecolleagues and Iare just completingastudy in which wecompare possible hazards for humansdue totypical dailyintakeof carcinogens, after adjusting forthe potencyof eachcarcinogen fromthe animal data.Thisadjustment isnecessary becausethe potencyof carcinogens varies ' DDEin the average U.S. dailyintake is equivalent tothat of the chloroformin a glassof tap water and is insignificantcom pared with natural carcinogens in ourdiet. Evenanoccasional fish highlycontaminat ed with DDT/DDEor PCB--that is, 100 times the average level--wouldcontribute a possiblehazard that iscomparableto the average peanut butter sandwichandissmall Coffeecontains the knownnatural carcin-, morethan a millionfold. For example, afla- compared with other very common minimal ogens hydrogen peroxide and methylglyox-: toxin, a moldcarcinogen present insmall risks,such asa glassof beer. al. each at about 4000ppb.' amounts in peanut butter(2ppb isthe U.S. average)andin com productssuch as torti Tapwater.asaconsequenceofchlorina- llas, requiresabouta million timessmaller lion,containsthe carcinogenchloroformat j- dosetocause thesameincidence of cancer ' anational average of 83ppb.-- sciif :-a ' i+'i.-iif. Coladrinks contain the carcinogenfor-., maldebyde at 7900ppb, though this level;/., in test animalsastrichloroethylene, which wasthe maincontaminant inSilicon Valley wells. Consider the followingcomparisons. Nature's pesticides: Weare ingesting nat ural pesticides inour diet inamounts at least 10,000times greater than manmade pesti cide residues. Natural pesticidesare natural We must ignore the notmuch higher than that inhuman blood, which averages about 3000ppbin formalde-' hydefrom normal metabolism. Contaminated water The levelof carcino gens in.contaminated well water(for exam ple, trichloroethylene in SiliconValleyor trivia i t we wish to deal with the im portant Beer contains nitrosamines, formalde-! . Woburn, Mass.)only rarely involvesa possi hyde(700ppb)and alcohol(50millionppb, or. blehazard greater than that of ordinary 5percent), aUknowncarcinogens.Alcohol consumptionisa knowncause ofhuman' !' chlorinatedtap water. Of 35private wells' t shut downinSiliconValley becauseof their cancer (3percent of UB. cancer), and.ethyl. i . supposedcarcinogenic hazard in an EPA alcohol isa carcinogen in rats. ' study, onlytwowereof greaterpossiblehaz ard than ordinary tap water(well water usu Milkconuins ahigh percentageoffat, allylacks the chloroformpresent inchlori andhigbfat consumption hasbeenImplicit- ,- nated tap water),and the most polluted well ed inhuman breast and coloncancer and ro 0900ppbtrichloroethylene)isstill at least dentcancer uhough milkisan Important::' 1000timeslessof a possiblehazard than an sourceofcalcium, whichmaybe important Vv equal volumeofcola,beer or wine. This is as an anti-carcinogen). . becausetrichloroethyleneisan extremely causes o f taneer. toxic chemicals present in all plants, usually making up5to 10percent of a plant's weight. Theyhavean enormousvariety of chemicalstructures, thoughonlya feware present Ineach plantspecies.Their function isto protect against fungi, insectsand ani mal predators. Thus, a major aspect of the evolutionof plants ischemical warfare. There has been relatively little research on the toxicologyor carcinogenicityof these Fruit juices may bavevariousamountsof ' carcinogenic moIdtoxins, dependingon- - ' howmany moldyfruitswere processed.' ' weakcarcinogen. It iscomparable tosaccha rin andonly 10timesmore potent than alco hol which Ispresentat about50million ppb ina beer. Giventhat the consumptionoftap compounds until quiterecently, and thus . very fewofthe large number present in the human diet have been testedfor theirabili tyto cause cancerinanimals. Calculating a possiblehazardto humans frominformation obtainedfroma cancer ; Bax, N. Ames is dtairman.of the Department o f waterisonlyabout oneor two liters per day, itseen)s unlikelythatman-made waterpol lutionin the trace amounts typicallyseenis causingmorethan a minimal hazard. Afair percentage ofthosefewnatural pesticidematerialsfrom plantsthat have been tested have turned out to becarcino gens in ratsor mice.Theyincludeestragole Biochemistry at the University of California at Berke Pesticide residues:Manmade pesticide (inbasil),safroie(inherbs), symphitine(In ley and was formerly on the Boardo f Directors o f the .residues present in our foodamount to National Cancer Institute. He is a member'ol th e-. NationalAcademy o f Sciences oadrtcefresiJhe ptes- . i tigiovs General Motors Cancer Research Foundation - ' Prise in 1983 as well as the highest 'award for ' abaullOOppban theaverage; mostofthese residuesare composedof noncarcinogenic compounds. Themanmade carcinogenic res environmental achievement, the Tyler Prize, if\, 1P8S- idue ofmost interest in foodis likely to be He does no consulting for chemical, drug or food -DDTandits metabolite(or breakdown prod industries, or for law firms. - - --- uct), DDE.Thepossible hazard of the DDT/ comfrey tea), psoralens(inparsley and cel ery), hydrazines(inmushrooms),and allyl is othiocyanate(inmustard).The passiblecar cinogenic hazardsof nature's pesticides completelyovershadowthe tracesof man made pesticide residues found in the daily diet. Plants alsocontain anti-carcinogens and valuable nutrients, so 1believethat even these possiblehazardsseemtoosmall to wor ry about, particularly given myskepticism about extrapolating risks from rodents. Allcalculationsof human risk based on ratandmousecancertests.bothfromnatu- _ ral and manmadecarcinogens,are hypo thetical.Thus, they should betaken with a great doseof skepticism, unlike known hu man carcinogenssuch assmoking(which causes400,000deaths fromcancer, heart dis easeand other illnessesa year in the United Stalestor alcoholic beverages(100.000deaths fromcancer andother diseaseseach yean. There are many newreasons for being . skeptical of uncritically extrapolating ani mal data --obtained byfeeding enormous dosesof carcinogens--to human risks from lowdosesof acarcinogen,but these cannot bediscussed here.Thiscontention is reinforced bythe studies ofcancer epidemi ologistswhoare makingconsiderable prog ressin understanding thesmoking, dietary, hormonal, viral and occupational contribu tions to human cancer, but who are finding remarkably littlesolidevidence for any sig nificant contribution frompollution. Even in the well-publicized casesofLoveCanal, Contra CostaCounty,Santa Clara County and Woburn. Mass., they found noconvinc ingevidence that pollution wasthe cause of the hazard. Sincewe nowknow that carcinogens are common, not rare--more than half of the chemicalstested inrats or mice have beenjudged carcinogens--wemust of ne cessityignorethe trivia if wewish todeal with the important causesofcancer. We might possiblyeliminate everytrace of man made carcinogen fromour water or food supply, but it wouldcost an enormous amountof California'swealth, beof minimal relevance tocausesof human cancer and distract health workers from real more im portant risks. Thus, onecancither chaseafter parts per billionof every manmadecarcinogen that turns upor havesomesensible regula tionsabout pollution. , } REVIEW Perspectives on Comparing Risks of Environmental Carcinogens1 Frederica Perera,* Paolo Boffetta23 In 1987, investigators (Ames et al.) concluded that the risks of man-made industrial carcinogens and pesticides (outside of the workplace) are trivial compared with the risks of nat urally occurring carcinogens found mostly in the diet. They used a ranking system based on human exposure and rodent potency (HERP) data to arrive at this conclusion. As a re sult, they recommend that regulatory agencies, such as the Environmental Protection Agency and the Food and Drug Administration, base their priorities in this area on their HERP system. We analyzed the assumptions and data set upon which the HERPs were based, concluding that such a simplified approach to set public health policy is inap propriate given the underlying uncertainties. However, we note that when comparisons are consistently based on esti mates of average daily exposure to common carcinogens, the HERP scores of many man-made pollutants are compara ble to those of naturally occurring carcinogens in the diet. [J Natl Cancer Inst 1988;80:1282-1293] Background T h e m ajority (an estim ated 60% -90% ) o f hum an can cer is considered to be attributable to environm ental factors, broadly defined to include cigarette sm oking, industrial pol lutants, radiation, diet, and perhaps other life-style factors and viruses (7 ). T hus, in theory m ost cancer is preventable through the identification and control o f causative factors, in cluding exposure to carcinogens. For decades, policym akers concerned w ith the assessm ent and regulation o f environ m ental carcinogens have searched for a system atic w ay to set priorities am on g the m any candidates. T his paper criti cally evalu ates the m ost recent proposal for such a ranking schem e (2). Identification o f sp ecific etiologic factors and estim ation o f their relative im portance constitute a form idable task. F ew can cers are attributable to sin gle factors or exposures; rather, com p lex interactions b etw een environm ental and host fac tors are generally involved (5 -5 ). M oreover, because o f the lim itations o f ep id em iology (6), on ly rarely are hum an data available that directly link an environm ental agent to hu m an cancer. For exam p le, ep id em iological studies strongly suggest, although they do not con clusively establish, an as sociation b etw een organ ic ch em ical carcinogens in drinking w ater (such as chloroform ) and cancers at several sites, in cluding the rectum , colon, and bladder (7-11). C ertain d i etary and nutritional factors (such as dietary fat and fiber) have been im p licated in can cer o f the breast, colon , rectum , and stom ach (12,13), but here too a direct cau sal a sso cia tio n h as not b een estab lish ed for sp ecific dietary constituents. In addition to active cigarette sm oking and a significant num b er o f p ollu tan ts in the w ork p lace, established to b e hum an lu ng carcinogens (14), there is grow ing evidence that pas siv e sm ok in g (15) and pollutants in the am bient air (16,17) contribute sign ifican tly to lu ng can cer m ortality. H ow ever, becau se o f cost and feasibility constraints and the difficulty in id en tifyin g an appropriate study population, the vast m a jority o f anim al carcin ogen s, both naturally occurring and m an-m ade, have neither been the subject o f ep id em iological in vestigation nor are they lik ely to b e (18). Thus, for practi cal purposes, as a m atter o f long-standing policy, regulatory agen cies accep t the use o f p ositive anim al data as predictive o f ca rcin o g en ic hazard in h um an b ein g s (19). T h e alterna tive, aw aitin g p o sitiv e ev id e n c e o f carcin ogen icity in hum ans, 'Received May 26, 1988; revised July 1, 1988; accepted July 8, 1988. division of Environmental Sciences, Columbia University School of Public Health, New York, NY. 3We gratefully acknowledge the valuable contributions of Dis. Ian Nisbet and Karim Ahmed. We also thank Drs. L B. Weinstein, Marvin Scbneiderman, Dale Hattis, David Rail, Norton Nelson, Philip Landrigan, Devra Davis, Lauren Zeise, Irva Hertz-Picciotto, William Pease, and Paolo Vineis for helpful discussions during the preparation of this manuscript and Drs. Michael Waters and Frank Stack for providing results from the U.S. Envi ronmental Protection Agency Gene-Tox Data Base. We are grateful to Jan Roby for excellent preparation of the manuscript *Correspondence to: Dr. Frederica Perera, Division of EnvironmentalSciences, Columbia University School of Public Health, 60 Haven Ave., B-109, New York, NY 10032. 2 Journal of the National Cancer Institute traditionally has been rejected as morally and socially unac ceptable. A major limitation of epidemiology (and risk assess ment) is that reliable data on human exposure to specific chemicals are frequently lacking. Therefore, by necessity, most epidemiological studies have relied on crude or in direct measures of exposure. A significant number of car cinogens have been detected in drinking water, ambient air, and the food supply; however, reliable monitoring data ex ist for only a small fraction of these chemicals. For ex ample, while dozens of pesticides and industrial chemical carcinogens have been measured routinely in surface wa ter, groundwater, and drinking water, they represent only a small percentage of chemical pollutants present (10,20-23). Over 700 organic chemicals have been found to be present in the U.S. drinking water supply, including 40 known or suspected carcinogens (24). Numerous carcinogenic air pol lutants (trace metals, polycyclic aromatic hydrocarbons, and volatile organic chemicals) have been detected in ambient air, again, there are little or no reliable monitoring data on the majority of airborne carcinogens (25). Similarly, many carcinogenic pesticide residues have been identified in the food supply, but reliable exposure data are lacking for most (26). Testifying to the pervasiveness of environmen tal contamination are studies showing significant concen trations of synthetic organic chemicals in the blood, urine, and/or adipose tissue of the U.S. population. These include l,l,l-trichloro-2,2-bis(p-chlorophenyl)ethane (DDT), dield- rin, heptachlor epoxide, polychlorinated biphenyls (PCBs), and dioxin (27,28). Again, data are far from comprehensive; however, they do show a decline in the concentrations of DDT and PCBs as a result of regulation. Despite their limitations, available exposure and epi demiologic data have served as the basis for a number of widely varying estimates of the proportion of human cancer in the U.S. population that can be attributed to life-style, occupational exposures, or other environmental pollution. These exercises have generated significant debate, as much over the underlying assumptions as the data used to gener ate them (17,29-35). Unfortunately, various such estimates (ranging, for example, from 4% to >20% for occupational exposures) have been cited as a basis for setting priorities for public health protection. This approach ignores both the underlying uncertainties, the relative preventability of var ious risk factors (36), and the disproportionate impact on some segments of the population. For example, once rec ognized, most chemical pollutant hazards can be reduced or eliminated by practical means. Moreover, the involuntary nature of these exposures necessitates control at the source, in contrast to exposures related to life-style (e.g., diet and smoking), which can be addressed more effectively through public education regarding personal behavioral choices. An other inherent problem with the approach of estimates is that it obscures the much higher risks to certain subpopulations. For example, if the contribution of occupational carcinogens to all cancer deaths in the United States were as low as 3%, for male industrial workers as a group, workplace carcino gens would account for at least 25% of all identified causes of cancer (33). __ Another tool that has been used increasingly by regula tory agencies to set priorities and even to determine accept able levels of exposure to individual environmental contam inants has been quantitative risk assessment Here, also, the lack of good information on human exposure as well as the usual paucity of epidemiological data are compounded by uncertainties regarding the proper way to extrapolate from high to low dose and from experimental animals to hu mans (37). To offset these uncertainties, the four major U.S. regulatory agencies, including the Environmental Protection Agency (EPA), the Occupational Safety and Health Admin istration (OSHA), the Consumer Product Safety Commis sion, and the Food and Drug Administration (FDA), have traditionally preferred conservative models that incorporate an assumption of low-dose linearity, regardless of the pre sumed mechanism of action of the chemical carcinogen (19). However, in certain instances, these conservative models may underestimate cancer risk. For example, the widely ac cepted linearized multistage model (38), considered to be one of the most conservative of the biologically plausible risk-assessment models, works on the assumption that the ex posed population is of uniform susceptibility and that interac tions do not occur between chemical exposures and other risk factors. Yet significant intraindividual variability has been demonstrated for human metabolism and binding of drugs and carcinogens (3 9-4 5) as well as for repair of DNA dam age (46). Moreover, a number of epidemiological studies have demonstrated synergism between chemical exposures and host factors, such as cigarette smoking and air pollutants in the workplace and urban air (47,48). To further compli cate the situation, although nonlinear (both superlinear and sublinear) dose-response relationships have been observed experimentally and epidemiologically, the available data do not allow low-dose linearity to be ruled out in any of these cases (49). Given these uncertainties, it is reassuring that, in a number of cases, risks observed in humans have been con sistent with those calculated from high-dose animal experi ments with the use of models that incorporated linearity at low dose. These include benzene, ethylene dibromide (EDB), gasoline, asbestos, and ethylene oxide (50-56). Therefore, there is general agreement that the use of quantitative risk assessment, performed with appropriate and consistent as sumptions and models, affords the possibility of comparing risks for the purpose of setting priorities among selected can didates for regulation. However, most scientists do not view quantitative risk assessment as capable of providing precise estimates of human risk from individual chemicals; general sources of chemical exposures are considered even less likely candidates for risk estimation by this method. Human Exposure/Rodent Potency (HERP) Index Most recently, researchers at the University of Califor nia at Berkeley and Lawrence Berkeley Laboratory have suggested still another approach to priority setting (2). They have calculated a possible hazard index for selected carcinogens by expressing the human exposure (in milligrams/kilogram) as a percentage,^i^he ^gdent TD50 dose Vol. 80, No. 16, October 19, 1988 REVIEW 1283 (also in milligrams/kilogram).4 They have compared the re sultant HERP indices for four pollutants found in drinking er and indoor air, three man-made pesticides and other dues, 10 natural pesticides and dietary toxins, two food additives, five drugs, and two occupational exposures (see table 1). The authors conclude that man-made environmen tal pollutants, such as pesticide residues and contaminants in drinking water, are "likely to be of minimal carcinogenic hazard" relative to the background of natural carcinogens (found largely in the diet). They recommend that regulatory agencies that traditionally have emphasized control of expo sures to man-made or industrial carcinogens (in addition to those in the occupational setting) revise their priorities. The authors acknowledge several major limitations of the HERP system, such as the possibility of interspecies (rodent and human) variation in susceptibility to carcinogens and quantitative uncertainties regarding the general shape of the dose-response curve, including the possibility of synergistic effects and thresholds for nongenotoxic carcinogens, such as promoters (see discussion below). They caution that it would be a mistake to use the HERP index as a direct estimate of human hazard, but they conclude that the scale provides "a way of setting priorities for concern." Although this is an innovative approach, it suffers from several inherent flaws. First, as we will show in table 2, the results are influenced strongly by the selection of chemicals and whether one classifies them as "man-made" or "natural." The rationale for selection of the individual compounds in table 1 was not provided by the authors, but presumably it dictated by the nature and availability of both exposure ana rodent potency data. As mentioned, the rodent potency data base is not comprehensive. For example, it omits a number of carcinogenic pesticides including alachlor, which is of current concern as a food contaminant (26) and has been found in water supply wells at significant concentrations (61). Certainly, the four selected drinking water and air Table 1. Possible carcinogenic hazards, as ranked by Ames et al. (2)* Possible hazard (HERP %) Carcinogenic exposure 0.0002 0.0003 0.0004 0.0002 0.06 0.003 0.006 0.003 0.03 0.03 0.06 0.07 0.1 0.1 0.2 0.008 2.8 4.7 6.2 1.3 0.6 0.004 2.1 0.001 0.004 0.0002 0.0003 0.008 0.3 5.6 14 16 17 5.8 140 Man-made chemicals in foods and beverages PCBs.t U.S. average daily dietary intake DDE/DDT. average daily dietary intake EDB, average daily dietary intake from grains/grain products Furylfuramide in 2-fluorenamine, daily dietary intake before banning Saccharin t in 12-oz diet cola Natural carcinogens in foods and beverages DMN In 100 g of cooked bacon DEN in 100 g of cooked bacon Urethane in 250 mL of sake Symphytine in 1 cup of comfrey herb tea Aflatoxin in 1 peanut butter sandwich DEN in dried squid, broiled in gas oven Allyl isothiocyanate in 5 g of brown mustard Estragle in 1 g of dried basil leaf Hydrazines in 1 raw mushroom Safrole in natural root beer, before ban DMN in 12-oz beer, before 1979 Ethyl alcoholt in 12-oz beer Ethyl alcoholt in 250 mL of wine Comfrey root in comfrey-pepsin tablets, 9 daily Symphytine in comfrey-pepsin tablets, 9 daily Indoor air pollutants Formaldehyde in conventional home air, 14 hr/day Benzene in conventional home air, 14 hr/day Formaldehyde in mobile home air, 14 hr/day Water pollutants Chloroform t in tap water. 1 L U.S. average Tetrachloroethylenet in well water, 1 L, highly contaminated Chloroform t in well water, I L. contaminated Tetrachloroethylenet in well water. 1 L. contaminated Chloroform t in average swimming pool, 1 hr Drugs Phenacetin, average dose Metronidazole, therapeutic dose Isoniazid, prophylactic dose Phenobarbital, 1 sleeping pill Clofibrate.f average daily dose Occupational exposure Formaldehyde, worker's average daily exposure EDB. worker's daily intake, high exposure JHere the TD30 is the average daily dose rate to halve the percent of tumor-free animals by the end of a standard lifetime (57). The average TD,,, is calculated by taking the harmonic mean of the TD30s of the posi tive tests in the most sensitive species. From each test, the target site with the lowest TDS0 value was used. In general, the harmonic mean and the lowest TD30 differ by a factor of " 2 (53). The source of TD30 values is the Carcinogenic Potency Data Base (CPDB) (57-60). The data base is a compilation of results from >3,500 experiments on 975 chemicals. It in cludes results from the Carcinogenesis Bioassay Program of the National Cancer Institute/National Toxicology Program (through May of 1986) as well as studies published in the literature (through December of 1984). The data base is restricted to tests that meet very stringent methodologic crite ria. Thus certain human carcinogens (such as asbestos and tobacco smoke) are excluded; seven chemicals regarded by the International Agency for Re search on Cancer (IARC) as having sufficient evidence of carcinogenicity in ar s (cadmium chloride, cadmium sulfate, epichlorohydrin, glycidaldehy sosafroie, mestranole, and 2-nitropropane) are recorded in the CPDB as having only negative tests. The CPDB is a useful tool, but its limitations should be kept in mind. *DMN = iV-nitrosodimethylamine. and DEN = iV-nitrosodiethylamine. t Carcinogens characterized by Ames et al. as nongenotoxic and likely to have thresholds. pollutants and the pesticides listed cannot represent the large number of industrial chemicals and pesticides that have been detected frequently in the U.S. drinking water, air, and food supply and that also have evidence of carcinogenicity in humans and/or laboratory animals (62,63). Moreover, although we are aware that there are many potential dietary hazards, the majority of which also are not well characterized (2,64), the 10 natural dietary carcinogens in table 1 include a number of exotic foods to which the U.S. general population has limited exposure (sake, comfrey Journal of the National Cancer Institute herb tea, dried squid, brown mustard, and comfrey-pepsin dioxin by sizeable segments of industrialized populations is tablets). Therefore, comparisons between drinking 1 L of 1 pg/kg (28), which corresponds to a HERP of 0.004. water containing average concentrations of chloroform and To avoid the problems of inconsistent exposure indices, we eating a daily serving of dried squid ignore the fact that the have adopted in table 2 the standard approach of uniformly average American adult ingests an estimated 2 L or more of providing average daily dose to the U.S. adult. We recognize water a day (65)s and rarely, if ever, eats dried squid. both the uncertainties in available exposure data (143) and An additional problem is that several "natural pesticides the fact that the average estimates mask wide interindividual and dietary toxins" in table 1 are misclassified in that they variation in exposure depending on geographical, cultural, can result from harvesting, manufacturing, or cooking pro economic, social, and host factors. For example, a child's cesses and therefore cannot be considered to be strictly exposure to pollutants in drinking water is proportionally natural. For example, aflatoxin in nuts and grains is par greater than exposure of the adult, because children ingest an tially attributable to improper harvesting and storage pro estimated 1 L of water per 10 kg of body weight compared cedures, whereas, as the authors acknowledge, nitrosamines with 2 L or more per 70 kg of body weight for the adult are formed in cured meats through the reaction of secondary (65). Children may also consume more of a contaminated amines with nitrites added as preservatives. Carcinogenic food than adults. In the case of the pesticide daminozide, nitropyrenes and nitrosamines occur in browned or burned the daily dose to the U.S. child (1-6 yr) from consumption meats as a result of cooking with gas flames that generate of apples, apple juice, and peanut butter is from fivefold to N 02 (2). 15-fold higher than the daily dose to the U.S. adult (70,71.) Moreover, a number of natural substances or food addi Thus, there are obvious drawbacks to using each of the tives in table 1 have been banned (safrole in natural root possible exposure indices (average, worst case, general pop beer and AF-2, a Japanese food additive never used in the ulation, or sensitive subpopulation). However, it is imperative United States) ((57), so that there is no current exposure to the in making comparisons that the same measure of exposure U.S. population. Several of the environmental pollutants have be used consistently. This is demonstrated by table 3 in which been regulated (chloroform, PCBs, and EDB) or even banned we compared HERPs from tables 1 and 2 for the same com (DDT), so that postregulatory exposures (and HERPs) are pound. predictably low, testifying to the effectiveness of regulation. As mentioned above, postregulatory exposures (and A second major limitation of the approach of Ames et al. HERPs) for environmental carcinogens such as DDT/DDE derives from the fact that, as can be seen in table 1, varying (l,l-dichloro-2,2'-bis(p-chlorophenyl)ethylene) are low. exposure indices were used. For waterborne and airborne Therefore, in a number of cases we have included preregu- contaminants, daily exposure was calculated; for pesticides latory and postregulatory values for purposes of comparison. and other residues in food, daily average dietary intake was Unfortunately, any listing of chemicals such as in tables provided; for "natural" carcinogens in food, one serving was 1-3 cannot convey the reality of cumulative exposures to assumed to occur daily; for food additives and drugs, several different carcinogens in the same medium. Also, it does different measures were used. not reflect the possibility of interactions among them or the To illustrate the effect of chemical selection and of as need to consider exposures to the same chemical via several sumptions regarding levels of exposure, we have constructed different media. For example, the individual has exposure to table 2; it includes all of the chemicals/exposures in ta synthetic volatile organic compounds in the drinking water ble 1, except for those dietary constituents not widely con from ingestion, from dermal absorption while bathing or sumed in the United States and those that have been banned showering, and from inhalation of the volatilized compound and have no current U.S. exposure. We have also omitted (144). Humans may be exposed concurrently to the same drugs because exposure is generally of short duration; drugs carcinogenic substance via a number of different sources and are a special case because they are prescribed when ben media. For example, in considering the risk of EDB in grain, efits are thought to outweigh risks to the individual. Fi the New York Department of Health reasonably chose to nally, we have included in table 2 several chemicals or sum the potential risks of the pesticide in food, ambient air sources of exposure that are encountered commonly by (from use of unleaded gas), and drinking water (145). the U.S. population and for which rodent potency (58-60) Finally, an important distinction not conveyed by either and exposure data are available. Unfortunately, in several table 1 or table 2 is that between voluntary and involuntary cases environmental chemicals of concern were in the ro exposure. As discussed in the introduction, individuals are dent potency data base, but we could not find reliable ex capable of voluntarily reducing exposure to substances in diet posure data for specific media. This was true for dioxin or and cigarette smoke that have been identified as carcinogenic 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). However, ac hazards. By contrast, individuals cannot feasibly control their cording to the EPA, a crude estimate of total daily intake of exposure to air, water, and workplace pollution. In tables 2 and 3, we have attempted to demonstrate the susceptibility of the HERP system (or any such simplified approach) to the effects of selection of both chemicals and exposure estimates. As is clear from table 3, the differences 3In fact, the results of a recent water consumption survey show that for between tables 1 and 2 are largely because of these two 5% of adults 20-64 yr old, the average daily consumption of tap water is factors. In contrast to that of Ames et al. (2), our approach, 2.71 L/day, whereas the average daily total water intake is 3.79 L (66\ incorporating a representative set of exposures to the U.S. Vol. 80, No. 16, October 19, 1988 19610 REVIEW 1285 Possible hazard (HERP %) 0.02 0.002 0.03 0.003 0.0003 0.002 0.001 0.004 0.0004 0.01 0.0002 0.003 0.003 < 0.0001 1.6 0.4 1.3 0.01 0.001 0.002 0.03 0.009 0.0005 0.0004 0.0002 0.00003 0.004 1.8 0.4 0.002 0.0001 0.003 0.001 0.001 0.005 0.0002 0.01 0.6 2.1 0.02 0.001 0.0001 0.003 0.01 0.007 0.03 0.0003 0.0002 0.00002 0.0002 86 Table 2. Ranking possible carcinogenic hazards with the use of the methodology of Ames et aL (2)* Average daily carcinogen dose (70-kg adult) Potency of carcinogen TDjo (mg/kg) Man-made chemicals in foods/beverages Daminozide in treated apples and apple juice (1987) Daminozide in peanuts and peanut butter (1987) DBCP in treated carrots (preregulatory, 1976) ODT. DOO, and ODE in food (preregulatory. 1968-1969) DDT, DDD. and DDE in food (postreguiatory, (1980-1982) Dieldrin in food (preregulatory, 1968-1969) Dieldrin in food (postreguiatory, 1980-1982) EDB in treated apples (preregulatory) EDB in grain products (preregulatory, 1983) PCBs in food (preregulatory, 1971) PCBs in food (postreguiatory, 1980-1982) Sodium saccharin in diet soda (1977-1978) 20 Mg 1.9 Mg 5.1 Mg 29.0 Mg 2.3 Mg 1.5 Mg LI Mg 4.1 Mg 0.42 Mg 15 Mg 0.2 Mg 4.9 ng Natural carcinogens in foods and beverages Aflatoxins in peanuts and peanut butter (1977) Estragle in basil Ethyl alcohol in beer (1981) Ethyl alcohol in wine (1981) Ethyl alcohol in hard liquor (1981) Hydrazines in mushrooms (1977) DMN in cured meat and bacon (1980) DEN in cured meat and bacon (1980) 5.8 ng <3.8 Mg 10.2 g 2-7 g 8.1 g 0.16 g 0.12 Mg 0.034 Mg Ambient air pollutants Benzene (Los Angeles, preregulatory, 1968) Benzene (Los Angeles, postreguiatory, 1984) Carbon tetrachloride (U.S. urban and surburban areas, 1973-1974) Carbon tetrachloride (U.S. urban areas, 1980) DDT (U.S. rural areas, preregulatory, 1972) DDT (U.S. rural areas, postreguiatory, 1974) EDB (U.S. urban areas, 1980-1981) Formaldehyde (Los Angeles. 1966) Formaldehyde (Los Angeles, 1979) PCBs (U.S. suburban areas, preregulatory, 1975) PCBs (U.S. urban areas, postreguiatory, 1979) Tetrachloroethylene (Bayonne, NJ, 1973) Tetrachloroethylene (Bayonne, NJ, 1983) Toxaphene (U.S. rural areas, 1972) 1.0 mg 0.32 mg 48 Mg 42 Mg 2.0 Mg 0-24 Mg 4.3 Mg 1.9 mg 370 Mg 2 Mg 150 ng 220 Mg 92 Mg 5.2 Mg Indoor air pollutants Benzene (personal average. New Jersey, 1981) Carbon tetrachloride (personal average, New Jersey, 1981) Chlordane (average in treated homes, 1976-1982) Formaldehyde in conventional homes (average of all reported U.S. data) Formaldehyde in mobile homes (US. average, 1984) Heptachlor (average in treated homes. 1982) Tetrachloroethylene (personal average. New Jersey, 1981) 173 Mg 16.2 Mg 20.5 Mg 600 Mg 2.2 mg 13.9 Mg 80 Mg Water pollutants Chlordane (Kansas City drinking water. preregulatory, 1965-1967) Chloroform (average U.S. drinking water, 1976) DBCP (California, postreguiatory, 1984) EDB (Florida, groundwater, 1983) Heptachlor (South Carolina rural drinking water, preregulatory, 1977) PCBs (US. surface water, preregulatory. 1971-1974) Tetrachloroethylene (New Jersey water supplies. 1985) TCE (US. water supplies, 1985) Vinylidine chloride (New Jersey water supplies. 1985) 0.14 Mg 170 Mg 2.0 Mg 7-8 Mg 24 Mg 0.4 Mg 12 Mg 14 Mg 4 Mg 1.2 1.2 0.24 13 13 1.1 1.1 1.5 IS 1.7 1.7 2,100 0.0026 52 9,100 9,100 9,100 20,000 0.16 0.021 53 53 140 140 13 13 IS IS IS 1.7 1.7 100 100 5.8 ... 53 140 2.4 IS IS 1.2 100 2.4 90 0.24 IS 1.2 1.7 too 940 24 Commentt ( 1) ( 1) (2) (3) (3) (4) (4) (5) (5) (6 ) (6 ) (7) (8) (9) ( 10) ( 10) ( 10) (ID ( 12) ( 12) (13) (13) (14) (14) (15) (15) (16) (17) (17) (18) (18) (19) (19) (2 0 ) (2 1 ) (2 2 ) (23) (24) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) Journal of the National Cancer Institute .................. . 1 9 6 1 1 Possible hazard (HERP %) Carcinogenic exposure T ab le 2. C ontinued Average daily carcinogen dose (70-kg adult) Potency of carcinogen TDjo (mg/kg) C om m ent! 32.3 0.06 105.0 3.0 6.2 0.2 Occupational exposures Benzene (rubber industry, preregulatory, 1942) Benzene (rubber industry, postregulatory, 1980s) Formaldehyde (resin and paper manufacture, 1961) Formaldehyde (resin and plastic manufacture, 1980s) TCE (small factories, preregulatory, 1940s) TCE (postregulatory, 1980s) 1-2 g 2.4 mg 110 mg 3.2 mg 4-1 g 0.1 g 53 53 1.5 1.5 940 940 (36) (36) (37) (37) (38) (38) *The selection of chemicals and the estimates of exposure differ somewhat from those in Ames et al,, as described in the text. To calculate average daily dose over an individual lifetime, we assumed: a) food consumption according to nationwide surveys: b) water consumption: 2 L/day, c) ambient air. inhalation of 20,000 L/day, d) indoor a ir inhalation of 10,800 L/14-hr day, e) workplace a ir inhalation of 9,600 L/day, 5 days/wk, 50 wk/yr, 40/70 yr (i.e,, 3,768 L/day over an average lifetime) (68). For carcinogens listed as ambient and indoor air pollutants, the respective HERPs cannot be considered additive, since the 20,000 L/day may include both types of exposure. We also calculated exposure for a 70-kg male adult, although a 60-kg adult is more reasonable (69). When only a range of values was reported in the literature, their geometric mean was used as the average exposure. The HERP is derived by dividing the daily carcinogen dose by 70 kg to provide a milligram-per-kilogram value, which is then given as the percentage of the TD;o dose in the rodent (also in mg/kg). t See appendix for comments. Table 3. Comparison of possible carcinogenic hazards (HERPs) as estimated by Ames et ai. and with the use of average exposure estimates**-! Carcinogenic exposure Ames et al. estimate Average Worst-case exposure exposure Our estimate, average exposure Man-made chemicals in foods and beverages DDE/DDT in food Preregulatory Postregulatory EDB in grains PCBs in food Preregulatory Postregulatory Sodium saccharin in diet sodas -- 0.0003 0.0004 -- 0.0002 -- -- -- -- -- -- 0.06 0.003 0.0002 0.0004 0.01 0.0002 0.003 N atural carcinogens in foods and beverages Aflatoxins in peanuts and peanut butter DMN in cured meat and bacon DEN in cured meat and bacon Estragle in basil Ethyl alcohol in beer Ethyl alcohol in wine Hydrazines in mushrooms -- _ -- -- -- -- -- 0.03 0.003 0.006 0.1 2.8 4.7 0.1 0.003 0.001 0.002 <0.0001 1.6 0.4 0.01 Ambient air pollutants Formaldehyde in conventional home air Formaldehyde in mobile home air 0.6 2.1 -- 0.6 _ 2.1 Water pollutants Chloroform in water Tetrachloroethylene in water TCE in water 0.001 -- -- -- 0.0003 0.004 0.003 0.0002! 0.000021 Occupational exposures Formaldehyde in workplace air 5.8 -- 3.0 *DMN = /V-nitrosodimethylamine, DEN = N -nitrosodiethylamine, and TCE = trichloroethylene. t See tables 1 and 2 for details on daily carcinogenic dose and TD;o+Worst-case assumption: HERP % = 0.007. Worst-case assumption: HERP % = 0.1. population, shows that the selected man-made or industrial pollutants generally are comparable in terms of HERP scale to naturally occurring carcinogens in the diet. Because of the limitations in the HERP approach, however, we stress that regulatory agencies would be unwise to base public health policies principally on comparisons such as these. Finally, Ames et al. (2) asserted that nine of the 26 car cinogens listed in table 1 "are thought to be nongenotoxic" and are therefore likely to have nonlinear dose-response curves or a decreased risk at lower dose. This subject has been discussed frequently (146-149). The general consen sus on the pan of regulatory agencies and expert groups has been that such policy distinctions are premature be cause they are supported inadequately by scientific data (.19,63,147,150,151). There are few, if any, clear-cut promoters and initia tors; rather, there is evidence that under different condi tions the same carcinogen can operate as a complete car cinogen, an initiator, or a promoter (147). For example, TCDD has demonstrated the ability to act both as a com plete carcinogen and a promoter (152-155). It is just as difficult to distinguish between genotoxic and nongenotoxic agents because in most cases short-term tests for genetic toxicity have generated a mixture of positive and negative results. This phenomenon has been observed with a vari ety of chemicals regarded up to this time as model "epi genetic, late stage" carcinogens: asbestos, the phorbol es ter 12-O-tetradecanoylphorbol-13-acetate, diethylstilbestrol, and DDT. These compounds have induced a variety of ge netic effects, either indirectly or directly, in experimental sys tems or in humans (149). The nine so-called nongenotoxic carcinogens in table 1 illustrate the difficulty of making such categorical distinc tions. Although in most instances the majority of short-term test results have been negative, each of the compounds (with the exception of clofibrate) has tested positive in at least one assay for each of several different genetic toxicity end points Voi. 80, No. 16, October 19, 1988 REVIEW 1287 ( 156).6 For eight of the nine chemicals, although the evi tinue to reduce involuntary exposures to carcinogens. The dence of genetic toxic effects is generally limited, it cannot dramatic decrease in estimated cancer risk following regula be dismissed. Therefore, it is not possible to conclude defini tion of a number of industrial chemicals illustrates this point. tively that these are nongenotoxic carcinogens that do not Controlling exposures to carcinogens such as those listed in act at some stage and under some conditions, either directly table 2 has important side benefits in that many carcinogens or indirectly, by damaging the genetic material. Viewed in a are mutagenic, teratogenic, reproductive, or neurological tox larger context, the proposed distinction among carcinogens icants (36). on the basis of presumed mechanism or stage at which they act is belied by the observation that control of late-stage carcinogens that may not be genotoxic may lead to the most References rapid reduction in risk, as has been seen with postmenopausal 1. Higginson J, Muir CS. The role of epidemiology in elucidating the estrogen therapy and cigarette smoking (158). In summary, there is no question that both occupational and food carcinogens are real concerns. However, while the hazards of the workplace are relatively well characterized (68), there is clearly a need for more research on dietary carcinogens. At the same time, there is evidence from epi demiologic, experimental, and monitoring sources that the cumulative risks of environmental pollution are important. Despite its limitations, the HERP analysis for a selection of importance of environmental factors in human cancer. Cancer Detect Prev 1976;1:79-105. 2. Ames BN, MaGaw R, Gold LS. Ranking possible carcinogenic hazards. Science 1987;236:271-280. 3. Hiatt HH, Watson JD, Winsten JA, eds. Origins of human cancer. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1977. 4. Mulvthill JJ. Genetic repertory of human neoplasia. In: Mulvihill. JJ. Miller RW, Fraumeni JF Jr. eds. Genetics of human cancer. New York: Raven Press. 1977:137-143. 5. Miller DG. On the nature of susceptibility to cancer. Cancer 1980:46:1307-1318. 6. D a y NE. Statistical considerations. In: Wald NJ, Doll R, eds. Inter exposures prevalent in the U.S. environment tends to support this conclusion. Although it is not possible to estimate the magnitude of these risks with certainty, it is prudent to con- pretation of negative epidemiologic evidence of carcinogenicity. Lyon, France: IARC, 1985:13-27. 7. Wilkins JR, Reiches NA, Kruse CW. Organic chemical contaminants in drinking water and cancer. Am J Epidemiol 1979;110:420-448. 8. National Research Council. Drinking water and health, vol 3. Wash ington, DC: Natl Acad Press, 1980. 9. Gottueb MS, Carr JK, Clarkson JR. Drinking water and can cer in Louisiana: a retrospective mortality study. Am J Epidemiol 1982:116:652-667. 10. Crump KS, Guess HA. Drinking water and cancer review of rece; ' 4 As summarized by the IARC, chloroform has been positive in only one of epidemiological findings and assessment of risks. Annu Rev Public Health 1982;3:339-357. many assays for gene mutation in bacteria and has been largely negative in other systems used. However, it has tested positive in lower eukaryotic sys tems (inducing differential toxic effects in DNA repair-deficient strains, gene conversion, and/or recombination and reverse mutation). Trichloroethylene has been positive in a number of assays. It has induced genetic toxic ef fects in bacteria (mutation), in yeast (gene conversion or recombination and mutation), in Tradescantia (mutation), in rodent ceils (transformation), in hu man cells in vitro [unscheduled DNA synthesis (UDS) and sister chromatid exchanges (SCHs)l, in animals in vitro (DNA damage), and in animal cells 11. Cantor KP, Hoover R, Hartge P, et al. Bladder cancer, drinking water source, and tap water consumption: a case-control study. JNCI 1987;79:1269-1279. 12. Maclure KM, MacMahon B. An epidemiologic perspective of envi ronmental carcinogenesis. Epidemiol Rev 1980:2:19-48. 13. Cohen L. Diet and cancer. Sci Am 1987;257:42-48. 14. Decoufl P. Occupation. In: Schottenfeld D, Fraumeni JF Jr, eds. Cancer epidemiology and prevention. Philadelphia: Saunders. 1982:318-335. 15. National Research Council. Environmental tobacco smoke: measuring in vivo (mutation and micronuclei). Tetrachloroethyiene, PCBs, and DDT have been largely negative in assays for genetic toxic effects; however, for tetrachloroethyiene there have been positive results in yeast (recombination or gene conversion), in Traaexaima (mutation), and in animal cells (trans formation and DNA damage). PCBs have induced DNA damage in animal cells in vitro and UDS in rat primary hepatocytes. DDT has been positive in insect systems (dominant lethal mutation and aneuploidy), animal cells in vitro (chromosome aberrations), and animal cells in vivo (chromosome aberrations and dominant lethal mutation). Ethyl alcohol has been studied exposures and assessing health effects. Washington, DC: Natl Acad Press, 1986. 16. Blot WJ. Clues to environmental determinants of cancer from its geographic patterns: In: Breslow NE, Whittemore AS, eds. Energy and health. Proceedings of a SIMS conference. Philadelphia: SIAM, 1979:151-167. 17. Nisbet ICT, Schneiderman MA, Karch NJ, et aL Review and eval uation of evidence for cancer associated with air pollution. Research Triangle Park: EPA Office of Air Quality Planning and Standards, 1984 [publication No. (EPA) 450/5-83-006R], extensively; a significant number of positive results were found. Evidence of genetic toxic effects includes gene conversion or recombination, muta tion, and aneuploidy in lower eukaryotic systems; SCE, micronuclei, and chromosome aberrations in plant systems; SCE, chromosome aberrations, and aneuploidy in animal cells in vitro; SCE, chromosome aberrations, mi cronuclei, dominant lethal mutations, and aneuploidy in animal cells in vivo; and SCE and chromosome aberrations in human cells in vivo. Alcoholism is associated with increased incidence of chromosome aberrations (757). For phenobarbitol, the preponderance of results has been negative. How 18. Karstadt M, Bobal R, Selkoff U. A survey of availability of epi demiologic data on humans exposed to animal carcinogens. In: Peto R, Schneiderman M, eds. Quantification of occupational cancer. Banbury report No. 9. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1981:223-242. 19. Office of Technology AssessmenL Identifying and regulating carcino gens. Washington, DC: US Govt Print Off, 1987. 20. Council on Environmental Quality. Contamination of groundwater by toxic organic chemicals. Washington, DC: US Govt Print Off,1981. 21. Office of Technology AssessmenL Protecting the nation's groundwa ever, there is evidence for the induction of gene mutation in bacteria and aneuploidy in yeast; gene mutation, SCE, and chromosome aberrations in rodent cells in vitro; cell transformation in rodent cells; as well as gene mu tation and chromosome aberrations in human ceils in vitro. Clofibrate has only been tested in two assays, both of which were negative. Finally, sodium saccharin has evidence of genetic toxic effects in lower eukaryotic systems (gene conversion or recombination and mutation), in insects (mutation), in animal and human cells in vitro (SCE and chromosome aberrations), and in animals in vivo (SCE and mutation). ter from contamination. Washington, DC: US Govt Print Off, 1984 [publication No. (OTA)-0-233], 22. National Research CounciL Drinking water and health, vol 1-6. Wash ington, DC: Natl Acad Press, 1977-1987. 23. Wallace LA, Peluzzari ED, Hartwell TD, et aL The TEAM study: personal exposures to toxic substances in air, drinking water, and breath of 400 residents of New Jersey, North Carolina and North Dakota. Environ Res 1987;43:290-307. 24. Harris RH, Page T, Reiches NA. Carcinogenic hazards of organic chemicals in drinking water. In: Hiatt HH, Watson JD, Winsten JA, Journal of the National Cancer Institute _ ... . . . . i s e i 3 eds. Book A. Incidence of cancer in humans. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1977:309-330. 25. Hunt WF Jr, Faoro RB. Curran TC, et al. Estimated cancer incidence rates for selected toxic air pollutants using ambient air pollution data. Research Triangle Park: EPA Office of Air and Radiation, 1984. 26. National Academy of Sciences. Regulating pesticides in food: the Delaney paradox. Washington, DC: Natl Acad Press, 1987. 27. Murphy RS, Kutz FW, Strassman SC. Selected pesticide residues or metabolites in blood and urine specimens from a general population survey. Environ Health Perspect 1983:48:81-86. 28. Environmental Protection Agency. A cancer risk-specific dose estimate for 2,3,7,8-TCDD (draft). Washington DC: EPA. 1987. 29. Cole P. Cancer and occupation: status and needs of epidemiologic research. Cancer 1977;39:1788-1791. 30. Bridbord K. Decoufle P, Fraumeni JF Jr, et al. Estimates of the fraction of cancer in the United States related to occupational fac tors. National Cancer Inst, NIEHS and NIOSH, September 15, 1978 (reproduced in ref. 18, pp. 701-726). 31. Cederlof R, Doll R, Fowler B. et al. Air pollution and cancer risk assessment methodology and epidemiological evidence. Environ Health Perspect 1978:22:1-12. 32. Doll R, Peto R, eds. The causes of cancer quantitative estimates of avoidable risk of cancer in the United States today. London: Oxford Univ Press, 1981. 33. Nicholson WJ. Quantitative estimates of cancer in the workplace. Am J Ind Med 1984:5:341-342. 34. Pike MC, Gordon RJ, Henderson BE, et al. Air pollution. In: Frau meni JF Jr, ed. Persons at high risk of cancer. An approach to cancer etiology and control. New York: Academic Press, 1975:225-238. 35. Davts DL, Lihenfeld AD. Gittelsohn A, et al. Increasing trends in some cancers in older Americans: fact or artifacL Toxicol Ind Health 1986:2:127-144. 36. Tomatis L. Relation between mutagenesis, carcinogenesis, and teratogenesis: experience from the IARC monographs programme. Pre sented at the fourth international conference on environmental muta gens. Stockholm, June 24-28, 1985. 37. Perera F. Quantitative risk assessment and cost benefit analysis for carcinogens at EPA: a critique. J Public Health Policy 1987;8:202-221. 38. Environmental Protection Agency. Guidelines for carcinogen risk assessmem. Federal Register, September 24, 1986;51:33992-34003. 39. Cartwright RA, Rogers HJ, Barham-Hall D, et al. Role of /V-acetyltransferase phenotypes in bladder carcinogenesis: a pharmacogenetic epidemiologic approach to bladder cancer. Lancet 1982:2:842845. 40. Harris CC, Trump BF, Grafstrom R, et al. Differences in metabolism of chemical carcinogens in cultured human epithelial tissues and cells. J Cell Biochem 1982;18:285-294. 41. Ritchie JC, Idle JR. Population studies of polymorphism in drug ox idation and its relevance to carcinogenesis. In: Bartsch H, Armstrong, B, eds. Host factors in human carcinogenesis (IARC Sci Publ No. 39). Lyon. France: IARC, 1982:381-394. 42. Omenn GS. Advances in genetics and immunology: the importance of basic research to prevention of occupational diseases. Arch Environ Health 1984;39:173-182. 43. Marquis JK, Siek GC. Sensitive populations and risk assessment in environmental policy-making. In: Saxena J, ed. Hazard assessment of chemicals, vol 6. Washington, DC: Hemisphere, 1988. 44. Perera F. The significance of DNA and protein adducts in human biomonitoring studies. Mutat Res 1988;205:255-269. 45. Hattis D, Erdreich L, DiMauro T. Human variability in parameters that are potentially related to susceptibility to carcinogenesis. I. Pre liminary observations. Cambridge, MA: M.I.T. Center for Technology, Policy and Industrial Development. 1986 (report No. CTPID 86-4). 46. Setlow RB. Variations in DNA repair among humans. In: Harris CC, Autrup H, eds. Human carcinogenesis. New York: Academic Press, 1983:231-254. 47. Seukoff U. Lung cancer and mesothelioma during prospective surveil lance of 1249 asbestos insulation workers, 1963-1974. Ann NY Acad Sci 1976;271:448-456. 48. Archer VE. Enhancement of lung cancer by cigarette smoking in uranium and other miners. Carcinog Compr Surv 1985:8:23-37. 49. Zeise L, Wilson R, Crouch EAC. Dose-response relationships for carcinogens: a review. Environ Health Perspect 1987;73:259-308. 50. Crouch E, Wilson R. Interspecies comparison of carcinogenic po:.' tency. J Toxicol Environ Health 1979;5:1095-1118. 51. California Department of Health Services. Health effects of benzene, report to the scientific review panel, part B. Sacramento, CA: CDHS, 1984. 52. Rowe JN, Springer JA. Asbestos lung cancer risks: comparison of animal and human extrapolation. Risk Anal 1986:6:171-180. 53. Enterline PE. A method for estimating lifetime cancer risks from limited epidemiologic data. Risk Anal 1987;7:91-96. 54. Hertz-Picciotto I, Neutra RR, Collins JF. Ethylene oxide and leukemia. JAMA 1987:257:2290. 55. Hertz-Picciotto, I, Gravttz N, Neutra RR. How do cancer risks predicted from animal bioassays compare with the epidemiologic evi dence? The case of ethylene dibromide. Risk Anal. In press. 56. California Department of Health Services. Controlling risks from low dose exposure to proposition 65 carcinogens and reproductive toxins. DHS response to the proposed science advisory panel caveaL Sacra mento, CA: CDHS, December 23. 1987. 57. Peto R, Pike MC, Bernstein L, et al. The TD30: a proposed general convention for the numerical description of the carcinogenic potency of chemicals in chronic-exposure animal experiments. Environ Health Perspect 1984:58:1-8. 58. Gold LS, Slone TH, Backman GM, et a l Second chronological sup plement to the carcinogenic potency data base: standardized results of animal bioassays published through December 1984. Environ Health Perspect 1987:74:237-329. 59. Gold LS, Sawyer CB, Magaw R, et al. A carcinogenic potency data base of the standardized results of animal bioassays. Environ Health Perspect 1984;58:9-13. 60. Gold LS, de Veciana M, Backman GM, et al. Chronological sup plement to the carcinogenic potency data base: standardized results of animal bioassays published through December 1982. Environ Health Perspect 1986:67:161-200. 61. Hallberg GR, Libra RD, Long ICR, et al. Pesticides, groundwater and rural drinking water quality in Iowa. In: Pesticides and groundwater a health concern for the MidwesL Navarre, MN: Public Freshwater Foundation, 1987:83-104. 62. National Toxicology Program. Fourth annual report on carcinogens: summary 1985. Research Triangle Park, NC: NTP, 1985 [publication No. (NTP)85-002]. 63. International Agency for Research on Cancer. Overall evaluations of carcinogenicity: an updating of IARC monographs, vol 1-42. IARC Monogr Eval Carcinog Risk Chem Hum (suppl 7). Lyon, France: IARC. 1988. 64. Sugimura T, Sato S. Mutagens-carcinogens in foods. Cancer Res 1983;43(suppl):24l5S-2421S. 65. National Research Council. Drinking water and health, vol 6. Wash ington, DC: Natl Acad Press, 1986. 66. Ershow AG, Cantor KP. Population-based estimate of water intake. Fed Proc 1986;45:706. 67. International Agency for Research on Cancer. Some food additives, feed additives, and naturally occurring substances. IARC Monogr Eval Carcinog Risk Chem Hum 1983;31:47-61. 6 8 . Gold LS, Backman GM, Hooper NK, et aL Ranking the potential carcinogenic hazards to workers from exposures to chemicals that are tumorigenic in rodents. Environ Health Perspect 1987;76:211-219. 69. Environmental Protection Agency. Captan special review position doc ument 2/3. Washington, DC: EPA, Office of Pesticides and Toxic Sub stances, 1985. 70. Environmental Protection Agency. Daminozide special review. Phase HI market basket survey, 1987 [Uniroyal's submissions dated February 13 and 20, 1987; and April 13, 1987 (memo from L Cheng to W. Waldrop, May 18, 1987)]. 71. Environmental Protection Agency. Tolerance assessment system. An nualized chronic consumption data based upon the United States De partment of Agriculture 1977 nationwide food consumption survey. Washington, DC: EPA. Office of Pesticide Programs. 1987. 72. Newsome WH, Iverson F, Panopio LG, et al. Residues of dibromochloropropane in root crops grown in fumigated soiL J Agr Food Chem 1977;25:684-685. 73. Duggan RE, Corneuussen PE. Dietary intake of pesticide chemi cals in the United States (HD, June 1968-April 1970. Pest Monit J 1972;5:331-341. 74. Environmental Protection Agency. DDT: a review of scientific and economic aspects of the decision to ban its use as a pesticide. Wash ington, DC: EPA, 1975 [publication No. (EPA) 540/1-75-022]. 75. GaRTELL MJ, C raun JD, PodrebaraC DS, et al. Pesticides, selected elements and other chemicals in adult total diet samples, October 1980-March 1982. J Assoc Off Anal Chem 1986;69:146-161. 76. Dumas T. Inorganic and organic bromide residues in foodstuffs fumi gated with methyl bromide and ethylene dibromide at low tempera tures. J Agr Food Chem 1983:21:433-436. 77. Environmental Protection Agency. Ethylene dibromide (EDB). Scien- Vol. 80, No. 16, October 19, 1988 19614 REVIEW 1289 tie support and decision document for grain and grain milling fumi gation uses. Washington, DC: EPA, 1984. 78. Jeunek CF, C orneuussen PE. Levels of PCBs in the U.S. food sup ply. In: Proceedings of (he national conference on polychlorinated biphenyls, Chicago, 1975. Washington, DC: EPA, 1976:163-165 [publication No. (EPA)560/6-75-004|. 79. International Agency for Research on Cancer. Some non-nutritive sweetening agents. IARC Monogr Evai Carcinog Risk Chem Hum 1980:22:111-170. 80. Pao EM, Fleming KH, Guenther PM. Foods commonly eaten by individuals: amount per day and per eating occasion. Washington, DC: US Dept of Agriculture, 1982. 81. National Research Council. Evaluation of cyclamates and carcino genicity. Washington, DC: Natl Acad Press, 1985. 82. Busby WF, Wogan GN. Aflatoxins. In: Searle CE, ed. Chemical car cinogens. Am Chem Soc Monogr 1984;182:945-1136. 83. Stoloff L. Aflatoxin control: past and present J Assoc Off Anal Chem 1980:6:1067-1073. 84. Leung AY. Encyclopedia of common natural ingredients. New York: Wiley, 1980. 85. T uyns AJ. Alcohol. In: Schottenfeld D. Fraumeni JF Jr, eds. Cancer epidemiology and prevention. Philadelphia: Saunders, 1982:293-303. 86. Distilled Spirits Council of the United States. Public revenues from alcohol beverages, Washington, DC: DISCUS, 1981. San Francisco; Wine Institute economic research report CA: Wine Institute, 1983. 87. Toth B, Erickson J. Cancer induction in mice by feeding of the un cooked cultivated mushroom of commerce Agaricus bisporus. Cancer Res 1986;46:4007-4011. 88. Toth B. Mushroom hydrazines: occurrence, metabolism, carcinogene sis. and environmental implications. In: Miller EC, ed. Naturally occur ring carcinogens, mutagens, and modulators of carcinogenesis. Tokyo: Japan Sci Soc 1979:57-65. 89. Preussmann R. Occurrence and exposure to N-nitroso compounds and precursors. In: O'Neill IK. Von Borstel RC, Miller CT, et al, eds. n-Nitro compounds: occurrence, biological effects and relevance to human cancer. Lyon, France: [ARC, 1984:3-15. 90. C hoi BCK. N-nitroso compounds and human cancer. Am J Epidemiol 1985;121:737-743. 91. Sen NP, Seaman S, Miles WF. Volatile nitrosamines in various cured meat products: effects of cooking and recent trends. J Agr Food Chem 1979:27:1354-1357. 92. Scanlan RA. Formation and occurrence of nitrosamines in food. Can cer Res 1983;43:2435S-2440S. 93. International Agency for Research on Cancer. Some industrial chem icals and dyestuffs. [ARC Monogr Eval Carcinog Risk Chem Hum 1982;29:93-148; 345-389. 94. Lonneman WA, Bellar A, Altshuller AP. Aromatic hydrocarbons in the atmosphere of the Los Angeles basin. Environ Sci Technol 1968:2:1017-1020. 95. National Academy of Science. Benzene in air. Washington, DC: Natl Acad Press. 1980. 96. Hunt WF Jr, Faoro RB, Freas W. Report on the interim data base for state and local air toxic volatile organic chemical measurements. Washington. DC: EPA, 1986 [publication No. (EPA) 450/4 86 012]. 97. International Agency for Research on Cancer. Some halogenated hydrocarbons. [ARC Monogr Eval Carcinog Risk Chem Hum 1979;20:371-399. 98. Lillian D, Singh HB, Appleby A, et aL Atmospheric fates of halo genated compounds. Environ Sci Technol, 1975;9:1042-1048. 99. Lovelock JE, Maggs RJ, Wade RJ. Halogenated hydrocarbons in and over the Atlantic. Nature 1973;241:194-196. 100. HanST PL, Spiller LL, Watts DM, et al. Infrared measurement of fluorocarbons, carbon tetrachloride, carbonyl sulde, and other atmo spheric trace gases. J Air Pollut Control Assoc 1975;25:1120-1226. 101. Singh HB, Fowler DP, Peyton TO. Atmospheric carbon tetrachloride: another man-made pollutant Science 1976;192:1231-1234. 102. Arthur RD, Cain JD, Barrentine BF. Atmospheric levels of pesticides in the Mississippi Delta. Bull Environ Contam Toxicol 1976:15:129-134. 103. Stanley CW, Barney JE, Helton MR, et al. Measurement of atmo spheric levels of pesticides. Environ Sci Technol 1971;5:430-435. 104. Singh HB, Salas LJ, Smith AJ, et al. Measurements of some potentially hazardous organic chemicals in urban environments. Atmos Environ 1981;15:601-612. 105. Singh HB, Salas LJ, Stiles RE. Distribution of selected gaseous or ganic mutagens and suspect carcinogens in ambient air. Environ Sci Technol 1982:16:872-880. 106. Patterson RM, Bornstein MI, Garshick E. Assessment of formalde hyde as a potential air pollution problem, vol 8. Bedford, MA: EPA, 1976. 107. Kitchens RD, Casner RE, Edwards GS, et aL Investigation of se lected potential environmental contaminants: formaldehyde. Washing ton, DC: EPA [publication No. (EPA)560/2-76-009]. 108. Versar Inc. Human exposure to formaldehyde. Draft report (contract No. 68-01-5791 for Office of Pesticides and Toxic Substances, Envi ronmental Protection Agency, Springfield, VA), 1980. 109. Cleveland WS, Graedel TE. Kleiner B. Urban formaldehyde: ob served correlation with source emissions and photochemistry. Atmos Environ 1977;11:357-360. 110. Kutz FW, Yang HSC. A note on polychlorinated biphenyls in air. In: Proceedings of the national conference on polychlorinated biphenyls, Chicago, 1975. Washington, DC: EPA, 1976:182 [publication No. (EPA)560/6-75-004], 111. Eisenreich SJ, Looney BB. Thornton JD. Airborne organic con taminants in the Great Lakes ecosystem. Environ Sci Technol 1981;15:30-38. 112. Eisenreich SJ, Looney BB, Hollod GJ. PCBs in the Lake Superior atmosphere 1978-80. In: Mackay D, Paterson S, Eisenreich SJ, et al, eds. Physical behavior of PCBs in the Great Lakes. Ann Arbor, MI: Ann Arbor Sci, 1983:115-125. 113. International Agency for Research on Cancer. Some halogenated hydrocarbons. IARC Monogr Eval Carcinog Risk Chem Hum 1979;20:491-514. . 114. Singh HB. Phosgene in the ambient air. Nature 1976;264:428-429. 115. Livingston JM, Jones CR. Living area contamination by chlordane used for termite treatmenL Bull Environ Contam Toxicol 1981^27:406411. 116. Wright CG, Leidy RB. Chlordane and heptachlor in the ambi ent air of houses treated for termites. Bull Environ Contam Toxicol 1985;28:617-623. 117. Fenske RA. Sternbach T. Indoor air levels of chlordane in residences in New Jersey. Bull Environ Contam Toxicol 1987;39:903-910. 118. Louis JB, Kisselbach KC Jr. Indoor air levels of chlordane and hep tachlor following termiticide applications. Bull Environ Contam Tox icol 1987;39:911-918. 119. Connor TH, Theiss JC. Hanna HA, et aL Genotoxicity of organic chemicals frequently found in the air of mobile homes. Toxicol Lett 1985;25:33-40. 120. Schafer ML, Peeler JT, Gardner WS, et aL Pesticides in drinking water waters from the Mississippi and Missouri rivers. Environ Sci Technol 1969;12:1261-1269. 121. Williamson SJ. Epidemiological studies on cancer and organic com pounds in U.S. drinking water. Sci Total Environ 1981;18:187-203. 122. Williams DT. Formation of trihalomethanes in drinking water. In: Fishbein L, O'Neill IK, eds. Environmental carcinogens: selected meth ods of analysis, vol 7 (IARC publications No. 68). Lyon, France: IARC, 1985:69-88. 123. Environmental Protection Agency. Tolerance assessment system. An nualized chronic consumption data based upon the United States De partment of Agriculture 1977 nationwide food consumption survey. Washington, DC: EPA. Office of Pesticide Programs, 1987. 124. Cohen DB, Bowes GW. Water quality and pesticides: a California risk assessment program, vol L Sacramento, CA: State Water Resources Control Board (Toxic Substances Control Program), 1984. 125. Environmental Protection Agency. Ethylene dibromide (EDB) position document 4. Washington, DC: EPA, Sept 27, 1983. 126. Sandhu SS, Warren WJ, Nelson P. Pesticidal residue in rural potable water. J Am Water Works Assoc 1978;70:41-45. 127. Dennis DS. Polychlorinated biphenyls in the surface waters and bot tom sediments of the major drainage basins of the United States. In: Proceedings of the national conference on polychlorinated biphenyls, Chicago, 1976. Washington, DC: EPA, 1976:183-194 [publication No. (EPA)560/6-75-004], 128. New Jersey Department of Environmental Protection. Results of testing for hazardous contaminants in public water supplies under Assembly Bill A-280. Final report. Trenton: NJDEP, 1987. 129. International Agency for Research on Cancer. Some halogenated hydrocarbons. IARC Monogr Eval Carcinog Risk Chem Hum 197920:545-572. 130. American Water Works Association. Materials for research workshop on volatile organic chemicals. Denver AWWA, 1972. 131. International Agency for Research on Cancer. Some monomers, plas tics and synthetic elastomers, and acrolein. IARC Monogr Eval Car cinog Risk Chem Hum 1979;19:439-459. 132. California Department of Health Services. Organic chemical contam- 290 Joumal of the National Cancer Institute ination of large public water systems in California. Sacramento, CA: CDHS, 1986. 133. Environmental Protection Agency. Ambient water quality criteria for benzene. Washington, DC: EPA, 1980 [publication No. (EPA) 440/580/018], 134. Runion HE, Scott LM. Benzene exposure in the United States 19781983: an overview. Am J Ind Med 1985;7:385-393. 135. National Institute for Occupational Safety and Health. Criteria for a recommended standard. Occupational exposure to formaldehyde. Washington. DC: US Govt Print Off, 1976 [DHEW publication No. (NIOSH) 77-125]. 136. National Academy of Sciences. Formaldehyde: an assessment of its health effects. Prepared for the Consumer Products Safety Commis sion. Washington. DC: Natl Acad Sci Press. 1980. 137. Siegel DM, F rankos VH, Schneiderman MA. Formaldehyde risk as sessment for occupationally exposed workers. Regul Toxicol Pharma col 183:3:355-371. 138. Bernstein RS, Stayner LT, Elliott U , et al. Inhalation exposure to formaldehyde: an overview of its toxicology, epidemiology, monitoring and control. Am Ind Hyg Assoc J 1984;45:778-785. 139. Environmental Protection Agency. Technical document Formalde hyde. Washington, DC: EPA, Office of Pesticides and Toxic Sub stances. November 16, 1981. 140. Funn FB. Industrial exposures to chlorinated hydrocarbons. Am J Med 1946:1:388-394. 141. Kleinfeld M, Tabershaw IR. Trichloroethylene toxicity. AMA Arch Ind Hyg 1954;10:134-141. 142. Kimbrough RD, Mitchell FL, Houk VN. Trichloroethylene: an up date. J Toxicol Environ Health 1985:15:369-383. 143. Silbergeld E. Risk assessment Science 1987(237:1399. 144. Brown HS, Bishop DR, Rowan CA. The role of skin absorption as a route of exposure for volatile organic compounds (VOCs) in drinking water. Am J Public Health 1984;74:479-484. 145. New York State Department of Health. A risk assessment for ethylene dibromide. Bureau of Toxic Substance Assessment NYSDH, Feb 21, 1984. 146. Williams GM, Weisburger JH. Carcinogen risk assessment Science 1983:221:6. 147. International Agency for Research on Cancer. Approaches to classi fying chemical carcinogens according to mechanism of action. Lyon, France: IARC, 1983 (technical report No. 53/001). 148. Weinstein IB. Letter to the editor. Science 1983:219:794-796. 149. Perera F. The genotoxic/epigenetic distinction: relevance to cancer policy. Environ Res 1984;34:175-191. 150. Office of Science and Technology Policy. Chemical carcinogens: a review of ihe science and its associated principles. Federal Register, March 14, 1985;50:10372-10442. 151. California Department of Health Services. Guidelines for chemical carcinogen risk assessments and their scientific rationale. Sacramento, CA: CDHS, 1985. 152. Kociba RJ, Keyes DG. Beyer JE. Results of a two-year chronic toxicity and oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats. Toxicol Appl Pharmacol 1978:46:279-303. 153. National Toxicology Program. Carcinogenesis bioassay of 2,3,7,8tetrachlorodibenzo-p-dioxin in Osborne Mendel rats and B6C3F, mice (gavage study). Research Triangle Park, NC: NTP, 1982 (technical report No. 209). 154. Pitot HC, Goldsworthy T. C ampbell HA. et at. Quantitative evalua tion of the promotion by 2,3,7,8-ietrachlorodibenzo-p-dioxin of hepatocarcinogenesis from diethylnitrosamine. Cancer Res 1980:40:36163620. 155. Poland A, Palen D, Glover E. Tumor production by TCDD in skin of HRS/J hairless mice. Nature 1982:300:271. 156. International Agency for Research on Cancer. Genetic and related effects: an update of selected IARC monographs, vol 1-42. IARC Monogr Eval Carcinog Risk Chem Hum (suppl 6). In press. 157. Waters MD, Stack HF, Brady AL, et al. Use of computerized data listings and activity profiles of genetic and related effects in the review of 195 compounds. Mutat Res 1988(205:295-312. 158. Day NE, Brown CC. Multistage models and primary prevention of cancer. JNCI 1980;64:977-989. Vol. 80, No. 16, October 19, 1988 REVIEW 1291 APPENDIX: TABLE 2 COMMENTS (1) Daminozide is a plant growth retardant In 1986, the EPA adopted interim measures to reduce its use, and a ban has been proposed. The estimated residues are 0.86 ppm in peanuts, 0.47 ppm in peanut butter, 1.02 ppm in apples, and 0.38 ppm in apple juice (70). The estimated average adult daily consumption of these foods is: apples, 18.1 g; apple juice, 4.4 g; and peanuts and peanut butter, 2.9 g (7/). (2) l,2,-Dibromo-3-chloropropane (DBCP), a soil fumigant for the control of nematodes, was widely used on a variety of vegetable, fruit, and nut crops until the ban in 1979. Concentrations were highest in root crops. The DBCP concentration in peeled raw carrots 9 wk after treatment at seeding was 0.607 mg/kg; the concentration of DBCP was 60% lower after 5 min of boiling in water (72). The average daily consumption of carrots is 8.4 g/day (71). (3) The dose calculation for DDT, l,l-dichloro-2J2'-bis(p-chlorophenyl)ethane (DDD). and DDE is based on FDA's Total Diet Studv. which yielded estimates of daily intakes of 29 Mg/day in 1968-1969 (73). DDD and DDE are the major metabolites of DDT and are therefore present mainly in meat, fish, and dairy products (74). The TD50S of the three chemicals are similar. DDT was phased out from use on many food crops in the late 1960s and was banned in 1972. According to an FDA study on pesticide intake from the diet in 1980-1982, the average daily intake of DDT was <40 ng/day; for DDE it was 2.2 Mg/day (75). The higher levels of DDE reflect the fact that long after the DDT ban, significant residues of DDT were available for metabolic conversion to DDE by animals used for food consumption. (4) Dieldrin is a cyclodiene pesticide that was banned in 1974, as was aldrin. another pesticide that was oxidized in the environment to yield residues of dieldrin. The estimates of average daily intake of dieldrin are derived from Duggan and Comeliussen (73) and Gartell et al. (75). (5) EDB residue (0.23 ppm) was estimated 12 days after fumigation, the average interval between spraying and consumption (76). Daily apple consump tion is estimated at 18 g/day for a 70-kg adult (71). Citrus fruits, plums, peaches, and cherries were also fumigated with EDB prior to its being banned for this use in 1984. In 1983, EPA estimated the average adult consumption of EDB in grain products at 0.006 ug/kg/day (77). The EDB residues in these foods are lower than in fruits, reflecting the relatively long interval between fumigation and consumption of grain and grain products. Fumigation of grains with EDB was banned in 1984. (6) The dose calculation is based on FDA's Total Diet Study, which yielded estimates of average daily PCB intake of 15.0 Mg/day in 1971, falling to 8.7 Mg/day in 1975 ( 78). Although PCBs were used in pesticide manufacture, the main source of dietary exposure was the widespread environmental contami nation from industrial uses. Fish, cheese, and eggs were the foods most significantly contaminated with PCBs. A number of regulatory and other actions were taken in 1971-1972 to reduce food contamination, and PCBs were banned from use in nonclosed systems in 1979. Although manufacture was discontinued in 1977, residues persist in the environment. The 1980-1982 estimate of daily intake was 0.196 Mg (75). (7) Consumption of saccharin increased substantially after cyclamates were banned in 1970. In recent years, other artificial sweeteners have become more popular. In 1976. an estimated 45% of all saccharin consumed in the United States was in soft drinks (79). A U.S. Department of Agriculture nutritional survey in 1977-1978 reported an average intake of artificially sweetened soft drinks at 14.7 g/day for males aged 19 and over (80). The average concentration of sodium saccharin is 33.5 m g/100 g of diet soda (79). A survey during 1977-1978 indicated an average daily intake of saccharin from all sources of 9.24 mg for males aged 18 and over (81). (8) Aflatoxins are substances produced by molds that contaminate nuts, cereals, and other foods. Although they are naturally occurring, their concentra tion is inversely proportional to the quality of food storage techniques. Thus, they are, in part, man-made carcinogens, and exposures can be appreciably controlled by improvements in commercial processes. The average consumption of peanuts and peanut butter is 2.9 g/day (71), and the U.S. average concentration of aflatoxins in peanuts and peanut butter is assumed to be 2 ppb (8Z83). (9) The average concentration of estragole in dried basil is 3.8 mg/g (84). The EPA reports "zero" average daily consumption of basil (71). However, since EPA reports no daily intake <1 mg, we are assuming basil intake to be < 1 mg. Other herbs (anise, fennel, bay, and marjoram) are considered to be the most important sources of estragole. They are also reported to have zero average daily consumption in the United States (71). (10) Ethyl alcohol is the best known human dietary carcinogen. It has been associated with cancers of the oral cavity, larynx, esophagus, and liver (85). Health and societal effects of alcohol consumption other than carcinogenesis are more important in terms of mortality, morbidity, and costs. Valid estimates of alcoholic beverage consumption are difficult to obtain. Data from household surveys, estimated consumption based on tax revenues, and estimates by producers do not cover all aspects of drinking patterns. The dose calculation is based on estimates by the Distilled Spirits Council of the United States and by the Economic Department of the Wine Institute (86): 116.56 L/yr of beer (3.2% alcohol by weight), 10.36 L/yr of wine (9.6% alcohol by weight), and 9.26 L/yr of hard liquor (32% alcohol by weight). (11) The TD50 is derived from data on dry Agaricus bisporus (Z87). The average daily consumption of mushrooms has been estimated as 1.6 g (71). The HERP is calculated assuming 90% of mushroom weight to be water and that all eaten mushrooms have the carcinogenic potency of A. bisporus. the most commonly eaten mushroom in the United States (88). No data are available concerning possible transformation of hydrazines, the principal carcinogens, during the cooking process. (12) N-nitrosamines constitute an important group of carcinogens formed by the reaction of amines with a nitrosating agent, usually nitrous acid. Human exposure to nitrosamines has a complex pattern with four major aspects: a) endogenous formation of precursors (from saliva and gastric juice) and endogenous formation of N-nitrosamines, b) endogenous formation after uptake of precursors (mainly nitrite and nitrate in foods), c) uptake of Nnitrosamines present in food, and d) uptake from other sources (tobacco smoke, drugs, cosmetics, and various occupations) (89). It has been estimated that 94% of exposure to /V-nitrosodimethylamine (DMN) is from exogenous sources (90). N-nitrosamines are present in cooked bacon, mainly from the frying process, whereas they occur in cured meat from added nitrate (89,91). The DMN content of beer, formerly high, has been significantly reduced through technological changes. Levels of N-nitrosamines in cured meat are also decreasing over time. The estimated daily intake is based on average concentrations in both cured meat and bacon of 3.4 Mg/kg (DMN) and 1 Mg/kg [A/-nitrosodiethylamine (DEN)] (91) and a total consumption of 34 g/day (bacon and cured meat) (92). Daily intake of DEN from meat in the Federal Republic of Germany in 1976 was estimated to be 0.1 Mg/day (89). (13) Benzene is mainly used as an intermediate in the synthesis of numerous chemicals, as a solvent, and as a gasoline component The major source of ambient air concentration is from gasoline, both from refueling and motor operation (93). The average concentration of benzene in Los Angeles ambient air in 1968 was 0.05 mg/m3 (94), corresponding to a daily intake of 1.0 mg. In 1980, the average intake from the ambient air of a number of U-S. cities was estimated to be 0.6 mg/day (95). The 1984 Los Angeles estimate is 16 Mg/m3 (96). The average 1979-1982 concentration from five urban areas was 12.3 Mg/m3 (25). (14) Carbon tetrachloride is used as a solvent and in the production of frons. It was used as a grain fumigant until banned for that use in 1985. The major sources of ambient air contamination are solvent emissions; the use of carbon tetrachloride is declining because of the concern about its toxic effects (97). The mean of eight measurements in urban and suburban areas in the Northeastern United States in the early 1970s was 2.4 Mg/m] (98). Other measurements during the same period ranged between 0.45 and 9.1 Mg/m3 (99-101). A decade later, the average concentration in five urban areas (2.1 Mg/m3) was comparable (25). (15) The estimates derive from a 1972-1974 ambient air survey in the Mississippi Delta area where DDT was widely used on cotton crops until its ban in January of 1973. The average concentrations were 0.1 Mg/m3 in 1972 and 0.012 Mg/m3 in 1974 (102). A previous estimate from nine urban and agricultural areas in the United States had an upper estimate of 1.56 Mg/m3, corresponding to a HERP% of 0.003 (103). (16) The average ambient air concentration of EDB in 10 U.S. cities was 213 ng/m~ (104,105), largely from its use as a scavenger in leaded gas. Regulations to reduce the use of leaded gas have been in effect since 1975. Journal of the National Cancer Institute (17) Formaldehyde is an important industrial chemical used as an intermediate in the production of resins for the construction, antomotive, and appliance industries. It is also used in the manufacture of fungicides and bactericides, various textiles, paper, and a number of consumer products, such as cosmetics, air fresheners, and drugs (93). Formaldehyde is a component of tobacco smoke. There is no major environmental (other than occupational) regulation of formaldehyde on the basis of its carcinogenicity. Data from 1966 are derived from a survey that showed a range of daily average concentrations in Los Angeles ambient air between 60 and 148 /ig/m3 (geometric mean, 94 ig/m3) (06). Other estimates in the same period were similar (107). The 1979 estimate (18.2 /ag/m3) is considerably lower (108). Other investigators have also reported a decline in formaldehyde levels over tme (109). (18) The average ambient air concentration of PCBs in suburban areas of Florida, Mississippi, and Colorado in 1975 was 100 ng/m3 (110). The most significant reduction in PCB air concentrations followed the production ban in 1979. The average urban air concentration in 1979 was estimated to be 5-10 Mg/ni3 (111); the HERP is based on the midpoint A 1979 study in Minneapolis reported an average ambient air concentration of 7.1 yig/m3 (112). (19) Tetrachloroethylene is used largely in the textile (processing and finishing) and dry cleaning industries, in vapor degreasing and cold cleaning of metals, and as an intermediate in the chemical industry. In 1972, it constituted >62% of the total dry cleaning solvent usage (113). Both exposure estimates refer to Bayonne, NJ, ambient air. The average daily ambient concentrations were 11.1 yig/m3 in 1973 (98) and 4.6 /ig/m3 in 1983 (23). Studies of other urban areas reported similar average daily concentrations: Los Angeles 1976, 4.6 ig/m3 (114); Baltimore 1978-1983, 4.0 jig/m3; and Philadelphia 1979-1982, 5.9 Mg/m3 (25). (20) Use of toxaphene has increased in the last decade, partly because of the ban on DDT. It is mainly used for the control of cotton insect pests. The average concentration measured in 1972 in the Mississippi Delta area was 0.26 Mg/m3 (102). (21) The dose calculation for benzene is based on personal air samplers carried by " 600 New Jersey residents in the fall of 1981. These samplers would have integrated outdoor and indoor exposure, but in all cases outdoor concentrations were lower than personal average concentrations; hence, the personal samples would have underestimated indoor exposure. Tobacco smoke was thought to be the main source of indoor concentrations of benzene (23). (22) The dose calculation for carbon tetrachloride is based on personal air sampling in New Jersey (23). See comment 21. (23) Chiordane and heptachlor are chlorinated cyclodiene insecticides, structurally similar to dieldrin. Chlordane was widely applied to the soil below and around the foundations of buildings to control subterranean termites until this use was terminated by agreement with the EPA in 1987. The dose calculation is based on a survey of military housing (115). Data for nonmilitary housing in the same period suggested higher average doses of 28-48 Mg/day (116,117). Chiordane is a complex mixture. The methods of measurement used in the cited studies would not have included the more volatile fractions and hence would have underestimated total exposures. Local data for 1983-1985 suggest a substantial reduction in exposure levels, perhaps resulting from improved application techniques introduced in the 1980s (118). (24) The release of formaldehyde from urea-formaldehyde foam insulation in mobile homes as well as in conventional homes containing particle board and plywood are the most important sources of indoor exposure to formaldehyde, but it has widespread use in construction materials, wood products, and furniture, textiles, and paper. Sources of indoor contamination for benzene include tobacco smoke and solvents (93). The estimate of average daily uptake of formaldehyde in the air in conventional homes is based on the means of all reported median or mean concentrations, as estimated by Ames et al. (2). The mean concentration of formaldehyde in the air of mobile homes (0.205 mgj/m3) is derived from Connor et aL (119). (25) Heptachlor is a component of technical chiordane (5.8%) and is often applied with it in a 17 mixture. The dose calculation is based on measure ments in homes treated with either this mixture or with technical chiordane (116). Local data for 1983-1985 suggest a substantial reduction in exposure levels, perhaps resulting from improved application techniques introduced in the 1980s (118). (26) The dose calculation for tetrachloroethylene is based on personal air sampling in New Jersey (see comment 21). Dry-cleaned clothes were thought to be a source of indoor concentrations of tetrachloroethylene (23). (27) Chiordane was used on many commercial fruit, vegetable, and grain crops, as well as in private gardening. Chiordane was detected in 21 out of 50 samples of Kansas City drinking water between 1965 and 1967 (120). Because only concentrations higher than 0.1 jig/L were listed separately in this study, a level of 0.05 ugli. was assumed for the remaining positive samples; the average (0.07 ^ig/L) is used in the table. The maximum chiordane concentration reported in the same study was 8 #ig/L (HERP%, 0.01). (28) The average drinking water concentration of chloroform estimated in a 1976 nationwide study of 113 public water systems was 84 *ig/L (121). It is mainly formed through the reaction of chlorine with organic chemicals in water (122). The risk posed by chloroform must be compared with the benefits of chlorination, and its benefits should be compared with those of other water-disinfection technologies. (29) DBCP concentrations in positive groundwater samples from five states in the 1980s ranged between 0.02 and 20 Mg/L (123). The upper value corresponds to a HERP% of 0.2. A concentration of 137 Mg/L (HERP% 1.6) has been reported for a single polluted well in Arizona (10). Monitoring of wells in both small domestic and large water systems in California showed that " 30% were contaminated with DBCP. The majority had concentrations >1 Mg/L; hence this value is used in the table as an average estimate (124). (30) An EPA groundwater EDB contamination study in four states found > 100 contaminated wells. Eighty-six of them were in Florida; the concentra tion ranged between 1.0 and 15.0 Mg/L. The HERP is based on the geometric mean (3.9 Mg/L). The highest concentration in this study was 100 Mg/L, corresponding to a HERP% of 0.2, and was found in Georgia (125). (31) Before being banned in 1983, heptachlor was mainly used to treat corn crops and seeds. A 1977 study of drinking water in two U.S. rural counties showed mean levels of heptachlor in positive samples of 15 and 9 Mg/L, respectively (126). The HERP is based on the average value of 12 (32) The estimated surface water concentration of PCBs in 17 major U.S. drainage basins during 1971-1974 was 0.1-3.0 Mg/L (127). The HERP is based on the average of positive results for 1974: 0.2 fig/L. (33) The median concentration of tetrachloroethylene in NJ public water supplies that tested positive from July to December of 1985 was 6 Mg/L: maximum concentration was 46 Mg/L (128) (HERP%, 0.001). Concentrations as high as 1500 Mg/L (HERP%, 0.04) have been detected in polluted wells ( 1 ) . (34) Trichloroethylene (TCE) is used in the vapor degreasing of lubricated metal parts, as a solvent in the textile, adhesive, and lubricants industry, as well as in some consumer cleaning products. In 1977, FDA prohibited its use as an extraction solvent in the food (coffee and spices) industry (129). The major source of environmental contamination by TCE is the metal industry (129). The median contamination of TCE in New Jersey public water supplies that tested positive from July to December 1985 was 7 Mg/L (128). The same study showed a maximum concentration of 190 /ig/L (HERP%, 0.0006). Levels as high as 35,000 Mg/L (HERP%, 0.1) have also been found in contaminated private drinking water supply wells (130). (35) In the New Jersey study described in comment 33, the median concentration of vinylidine chloride was 2 Mg/L, and the maximum was 9 Mg/L (HERP%, 0.001). Vinylidene chloride is the basis for the production of copolymers, which are used in food packaging, coatings for paper, fibers, tubes, and pipes (131). Vinyl chloride, whose polymers are also used in the production of plastics, is a known human and animal carcinogen and is strictly regulated both in the workplace and in the environment It has been found in polluted wells at concentrations as high as 20 Mg/L (132) and 10 Mg/L (128). Inasmuch as the TD50 is 8.0 mg/kg, the respective HERP% values are 0.007 and 0.004. (36) Concentrations of 100 ppm of benzene or more were common in the rubber, printing, leather, and shoe industries until the 1970s (93,133). The 1942 exposure corresponded to 100 ppm. In 1980, a permissible exposure limit (PEL) of 100 ppm was issued by OSHA; today the limit is 1 ppm or 3.2 mg/m3. The estimated average postregulatory benzene concentration in the rubber industry is 0 7 0 ppm (134). (37) The estimated workplace air concentration (28.3 mg/m3) is the midpoint of values obtained in a 1961 U.S. study on resin manufacture and paper production (135,136). In 1980, OSHA set the PEL at 3.7 mg/m3 (HERP%, 13.3) on the basis of noncarcinogenic effects. In 1987, it was lowered to 1-2 mg/m3 (HERP%, 4.3). Estimates of the concentration of formaldehyde (0.86 mg/m3) in the resin and plastic industry during the 1980s indicates a 33-fold decrease (137-139). -(38) The HERP is based on an estimated workplace concentration of 200 ppm, which was not uncommon before regulations were imposed (140). Concentrations as high as 8,000 ppm, corresponding to a HERP% of 256, have also been measured (141). Today, the OSHA PEL is 100 ppm or 538 mg/m3 (HERP% 3.2), and the HERP is based on an estimated workplace concentration of 5 ppm (142). Vol. 80, No. 16, October 19, 1988 1 n,pf 1 jL o -4- REVIEW 129:- Science is only orderly after the fact; in process it is chaotic and fiercely controversial. standards of scientific validity without significant improvements in the state of the science. Thus, a tension was established, which exists to this day, between the need to act and the availability of knowledge. Even at the outset as the environmentalist idea moved, in Winston Churchill's phrase, "from the wonderful cloudland of aspiration to the ugly scaffolding of attempt and achievement," it confronted a problem that would plague EPA henceforth: the basic disjunction between the reality of science and the assumption, inherent in the mission of all governmental protective agencies, that it is always possible, and therefore obligatory, to provide swift and sure regulation of all substances and situations that threaten public health and the environment. This disjunction is hardly ever acknowledged by the general public, so completely has science been accepted as the handmaiden of its material desires. The relationships among basic science, applied science, and improve ments in daily life are usually regarded as simple, being much like those among growing trees, cutting lumber, and building houses. This concept feeds the notion that when we want something from science, we can order it, as we order lumber to build a house. Ifthere is not enough lumber, we can grow and cut more trees. It follows that there is a way to "manage" this orderly process so as to make it more efficient, or more suitable to our current needs. Even though a scientific explanation may appear to be a model of rational order, we should not infer from that order that the genesis of the explanation was itself orderly. Science is only orderly after the fact; in process, and especially at the advancing edge of some field, it is chaotic and fiercely controversial. Thus, the expectation built into environmental law, that sci ence can provide definitive answers to the kinds of questions that policymak ers are obliged to ask under the terms of that law, will be disappointed to the degree that such answers derive from the forward edge of research. We can direct our resources in various directions with the expectation that talented researchers will work on questions that bear on important policy issues, but we cannot order answers as efficiently as the public would like. As Louis Pasteur said. "There is no such thing as applied science, there are only applications of science." Even in those scientific establishments specifically funded to explore policy questions for federal protective agencies, we find scientists behaving rather more like scientists in "basic" research than their budget justifications would suggest. Nor can we order a consensus in the areas of greatest interest to environmental policy: pollutant exposure and effects. Policymakers, includ ing me, have often deplored the tendency of scientific panels to engage in interminable debate rather than reach the agreement that was clearly indi cated on the invitation. O f course scientists will disagree on issues involving the advancing edge of research: that is what they do for a living. And even if we could somehow get a group of scientists to endorse a consensus position, it would be, in the first place, only tentative and subject to revision with the arrival of new discoveries; and in the second place, it may be entirely wrong. In science, the majority does not rule, as the history of science amply demonstrates. Everybody but Semmelweiss was wrong on childbed fever. Everybody but Wegener was wrong on continental drift Scientists outside the ISSUES IN SCIENCE A N D TECH N O LO G Y 19613 R ISK . SC IE N C E . A N D D E M O C R A C Y co n sen su s, w h o m a y n o t b e averse to v o icin g th eir o b je ctio n s in p u b lic, m ay be cranks or th ey m a y b e right. P u b lic officials ca n n o t m a k e this d istin ction : on ly the slow m ills o f scien ce can grind o u t th e truth. P u b lic officials, th o u g h , d o n o t h ave th a t k in d o f tim e . F rom its earliest d a y s E P A w a s o f t e n c o m p e l l e d t o act under conditions o f substantial scientific uncertainty. T h e fu ll im p l ic a t io n s o f th is p r o b le m w e r e p a r tia lly m a s k e d a t th e b egin n in g b ecau se th e k in d o f p o llu tio n w e w ere tryin g to con trol w as so blatan t. A lth o u g h scien tists w ere o ften in th e forefron t o f th e early struggles again st p ollu tan ts, m o st p eo p le d id n ot n eed a scien tific p an el to tell th em that air is n o t su p p o sed to b e b ro w n , th at strea m s are n o t su p p o sed to ign ite an d stink, that beach es are n o t su p p o sed to be covered w ith raw sew age. W h en I left E P A in 1973. th e m e a n s for e n d in g th e w o rst fo rm s o f p o llu tio n w ere fairly w ell in h an d . If em issio n co n tro ls w ere p u t o n cars, if th e req u isite sew age trea tm en t p lan ts w ere b u ilt, a n d if in d u strial firm s in stalled readily availab le con trol tech n ologies, p o llu tio n w ou ld be essen tially elim in a ted as a n ation al p rob lem . W h ile I h ad so m e reservation s a b o u t th e scien tific basis o f so m e o f the criteria w e had esta b lish ed for health p ro tectio n an d ab o u t th e ab ility o f in d ustry to m eet co n g ressio n a l d ead lin es, an d w h ile w e w ere p rob ably n ot p ro tectin g as efficien tly as w e m ig h t h ave, all in ail it d id n o t seem to m atter m u ch . I w as certain that C on gress w ou ld ad ju st the m ore o b v io u s in firm ities in o u r e n v ir o n m e n ta l la w s o n c e it rea lized th e so c ia l a n d e c o n o m ic d isto r tio n in h eren t in p u rsu in g zero-risk e n v ir o n m en ta l goals. W e h ad m a d e a start an d had d em o n stra ted th at th e federal g o v ern m en t co u ld m o v e effectively against m ajor p ollu ters in su ch a w ay that th e gen eral p u b lic co u ld literally see th ro u g h its w in d o w s. IV T h e events o f th e past d ozen or so years have d em onstrated that m y o p in io n w as in o n e se n se q u ite correct a n d in a n o th er far to o sa n g u in e. M an y o f the grosser sorts o f p o llu tio n are in d eed u n d er con trol. B ut the level o f con troversy ab ou t en v iro n m en ta l p rotection has n ot d im in ish ed . O n the ev id en ce o f m y m o re recen t ex p erien ce as a d m in istrator o f EPA, as in d icated by w h at regulatory issu es w ere o f greatest con cern , w h at the various con gres sio n a l c o m m itte e s w ere in terested in h earin g ab ou t, press coverage, a n d so on. e n v ir o n m e n ta l co n tro v e rsy is n o w largely fo cu sed o n th e ca r cin o g e n ic risk to h u m an health from to x ic ch em icals, an d o n th e rem oval o f ever sm aller in cr em en ts o f c o n v e n tio n a l p o llu tio n from th e air a n d w ater. " R isk " is th e k e y c o n c e p t h ere. It w a s h a rd ly m e n t io n e d in th e ea r ly y ea rs o f E PA , a n d it d o es n o t h a v e an im p ortan t p la ce in th e C lea n A ir or C lean W a ter A cts p assed in th a t p eriod . O f th e ev en ts th at co n trib u ted to th is ch an ge, th e m o st im p ortan t w ere th e focu s o f p u b lic a tten tion o n P C B s an d asb estos (tw o su b sta n ces th at are u b iq u ito u s in th e A m erica n e n v ir o n m e n t a n d th at are ca p a b le o f c a u sin g ca n cer) an d th e rea liza tio n th at ex p o su re to a very large n u m b e r o f u n fa m ilia r a n d largely u n tested c h e m ic a ls is u n iversal. T h e d iscovery by can cer ep id em io lo g ists that can cer rates vary w ith en v iro n m en t su g g ested th a t p o llu tio n m ig h t p lay a role in c a u sin g th is d isea se. A n d finally, th e ca n cer risk w a s p u sh ed to th e forefront by th e em e rg en ce o f a b a n d o n ed S P R IN G 1985 dumps o f toxic chemicals as a consuming public issue. As a direct result of this shift in attention, the relation o f EPA to its science base was altered; the problem o f uncertainty was moved from the periphery to the center. This shift occurred because the risks of effects from typical environmen tal exposures to toxic substances-- unlike the touchable, visible, and malodor ous pollution that stimulated the initial environmental revolution-- are largely constructs or projections based on scientific findings. We would know nothing at all about chronic risk attributable to most toxic substances if scientists had not detected and evaluated them. Our response to such risks, therefore, must be based on a set o f scientific findings. Science, however, is hardly ever unambiguous or unanimous, especially when the data on which definitive science must be founded scarcely exist. The toxic effects on health of many of the chemicals EPA considers for regulation fall into this class. "Risk assessment" is the device that government agencies such as EPA have adopted to deal with this quandary. It is the attempt to quantify the degree of hazard that might result from human activities-- for example, the risks to human health and the environment from industrial chemicals. Essentially, it is a kind of pretense; to avoid the paralysis o f protective action that would result from waiting for "definitive" data, we assume that we have greater knowledge than scientists actually possess and make decisions based on those assumptions. Of course, not all risk assessment is on the controversial outer edge of science. We have been looking at the phenomenon of toxic risk from environmental levels o f chemicals for a number of years, and as evidence has accumulated for certain chemicals, controversy has diminished and consen sus among scientists has become easier to obtain. For other substances-- and these are the ones that naturally figure most prominently in public debate-- the data remain ambiguous. In such cases, risk assessment is something o f an intellectual orphan. Scientists are uncomfortable with it when the method must use scientific information in a way that is outside the normal constraints of science. They are encroaching on political judgments and they know it. As Alvin Weinberg has written: Attempts to deal with social problems through the procedures of science hang on the answers to questions that can be asked of science and yet which cannot be answered by science. I propose the term trans-scientific for these questions.. . . Scientists have no monopoly on wisdom where this kind of trans-science is involved; they shall have to accommodate the will o f the public and its representatives. However, the representatives of the public, in this instance policy officials in protective agencies, have their problems with risk assessment as well. The very act of quantifying risk tends to reify dreaded outcomes in the public mind and may make it more difficult to gain public acceptance for policy decisions or push those decisions in unwise directions. It is hard to describe, 26 ISSU ES IN S C IE N C E A N D T E C H N O L O G Y R ISK . SC IEN CE. A N D D E M O C R A C Y say, o n e ca n cer case in 7 0 years a m o n g a p o p u la tio n o f a m illio n as an " a ccep tab le risk" w h en su ch a d escrip tio n m a y to o ea sily su m m o n up for an y in d ivid u al th e im age o f so m e clo se relative o n h is d eath b ed . A lso, the use o f risk a ssessm en t as a p o licy b asis in ev ita b ly p ro v o k es en d less a rg u m en ts a b o u t the valid ity o f the estim a tes, w h ich can serio u sly d isru p t th e regulatory tim etab les su ch officials m u st liv e by. D e sp ite this u n ea sin ess, th ere ap p ears to b e n o su b stitu te for risk a sse ssm en t, in th at so m e so rt o f risk fin d in g is w h a t tells us th at th ere is a n y b a sis for regu latory a ctio n in th e first p lace. T h e a ltern a tiv e to n o t p erfo rm in g risk a s s e s s m e n t is to a d o p t a p o lic y o f e ith e r r e d u c in g a ll potentially t o x ic e m issio n s to th e greatest d eg ree tec h n o lo g y a llo w s ( o f w h ich m o re later) o r b an n in g all su b stan ces for w h ich there is a n y e v id e n c e o f h arm fu l effect, a p o licy th at n o tec h n o lo g ic a l so c iety c o u ld lo n g su rv iv e. B e y o n d that, risk a ssessm en t is an irrep laceab le to o l for settin g p riorities a m o n g th e ten s o f th ou san d s o f su b stan ces that co u ld be su bjects o f con trol action s-- su bstan ces that vary en o rm o u sly in th eir apparent p o ten tia l for cau sin g d isease. In m y view , therefore, w e m u st u se a n d im p r o v e risk a ssessm en t w ith full recogn i tio n o f its cu rren t sh o rtco m in g s. T h is a cco m m o d a tio n w ou ld be m u ch easier from a p ub lic p olicy view p o in t w ere it p ossib le to esta b lish for all p o llu ta n ts th e en v ir o n m en ta l levels that p resen t zero risk. T h is is p rev en ted , h o w ever, by an im p o r ta n t lim ita tio n o f th e current tech n iq u e; th e difficulty o f estab lish in g d efin itive n o-effect lev els for ex p o su re to m o st c a r cin o g e n s. C o n se q u en tly , w h e n e v er th ere is an y ex p o su re to su ch su b sta n c e s, th ere is a ca lcu la b le risk o f d isease. T h e en v ir o n m en ta list eth o s, w h ic h is reflected in m a n y o f o u r e n v ir o n m e n ta l law s, an d w h ic h requ ires th a t zero -risk lev els o f p o llu ta n t ex p o su re b e esta b lish ed , is thus sh o w n to be an im p o ssib le goal for an in d u strial society, as lon g as w e re tain th e n o -th resh o ld m o d e l for ca rcin o g en esis. v In my view, we must use and improve risk assessment with full recognition o f its current shortcomings. T h is situ ation has g iv en rise to tw o co n flictin g v iew p o in ts o n p rotection . T h e first, u su a lly p roffered b y th e reg u la ted c o m m u n ity , a rgu es th a t regu la tion ou gh t n ot to b e b ased o n a set o f u n p rovab le a ssu m p tion s, but on ly on co n n ectio n s b etw een p o llu ta n ts an d h ealth effects that can b e d em on strated u n d e r th e c a n o n s o f s c ie n c e in th e stric t s e n s e . It p o in ts o u t th a t for th e v a st m ajority o f ch e m ic a l sp ec ies, w e h a v e n o e v id e n c e at all th at su ggests effects on h u m a n health from ex p o su res at en v iro n m en ta l levels. B ecau se m an y im p o rta n t risk a sse ssm en ts are b ased o n a ssu m p tio n s th at are scien tifically u n testa b le, th e m e th o d is to o su scep tib le to m a n ip u la tio n for p o litica l en d s an d , th e regu lated c o m m u n ity c o n te n d s, it h as b een so m a n ip u la te d b y en viron m en talists. T h e secon d view p oin t, w h ich has b een ad op ted by so m e en viron m en tal ists, co u n te r s th at w a itin g for firm e v id e n c e o f h u m a n h ea lth effects a m o u n ts to u s in g th e n a t io n 's p e o p le a s g u in e a p ig s, a n d th a t is m o r a lly u n a c c e p t a b le . It p ro p oses th at far from o v er estim a tin g th e risks fro m to x ic su b sta n ces, c o n v e n tio n a l risk a ssessm en ts u n d erestim a te th e m , for th ere m a y b e effects from ch em ica ls in co m b in a tio n that are greater th an w o u ld b e ex p ected from the SPRING 1985 27 su m effects o f all c h e m ic a ls a ctin g in d ep en d en tly . W h ile a p p ro v in g o f risk assessm en t as a priority-settin g tool, th is view p oin t rejects the id ea that w e can u se risk a sse ssm en t to d istin g u ish b etw een " sig n ifica n t" an d " in sign ifican t" risks. A n y id en tifia b le risk o u g h t to b e e lim in a te d u p to th e ca p acity o f availab le tech n o lo g y to d o so. It is im p o s s ib le to e v a lu a te th e m e r its o f th e se p o sitio n s w ith o u t first d raw in g a d istin c tio n b etw een the a ssessm en t o f risk an d th e p rocess o f d e c id in g w h a t to d o a b o u t it, w h ic h is " risk m a n a g e m e n t." T h e a rg u m en ts in th e form sk etch ed here are really d irected at b oth th ese p rocesses, a c o m m o n co n fu sio n th at h as lo n g sto o d in th e w ay o f sen sib le p olicym ak in g. R isk a sse ssm e n t is an ex ercise th at c o m b in e s availab le d a ta o n a su b s t a n c e 's p o t e n c y in c a u s i n g a d v e r s e h e a lt h e f f e c t s w i t h in f o r m a t i o n a b o u t lik ely h u m a n e x p o su r e , a n d th r o u g h th e u se o f p la u sib le a ssu m p tio n s , it gen era tes a n e stim a te o f h u m a n h ea lth risk. R isk m a n a g e m e n t is th e p ro cess by w h ich a p ro tectiv e ag en cy d ecid e s w h at a ctio n to tak e in th e face o f su ch es tim ates. Id eally th e a ctio n is b ased o n su ch factors as th e goals o f p u b lic h ealth an d en v ir o n m en ta l p ro tectio n , relevan t legisla tio n , legal p reced en t, an d a p p li c a t io n o f s o c ia l, e c o n o m i c , a n d p o lit ic a l v a lu e s . Risk Assessment in the Federal Government, a N a tio n a l R e se a r c h C o u n c il ( N R C ) d o c u m e n t , r e c o m m en d s th at regu latory agen cies estab lish a strict d istin ctio n b etw een th e tw o processes, to allay an y co n fu sio n b etw een th em . In m y view C on gress sh ou ld d o th e sa m e in all sta tu tes se ek in g to d eal w ith risk. R etu rn in g n o w to the o p p o sin g view p oin ts w e see that b oth reflect th e fear th at risk a sse ssm en t m a y b e im b u ed w ith v a lu es rep u gn an t to o n e or m o re o f th e p arties in v o lv ed . T h a t is. so m e p eo p le in th e regu lated c o m m u n ity b elie v e th a t th e stru ctu re o f risk a ssessm en t in h eren tly exaggerates risk, w h ile m a n y e n v ir o n m e n ta lists b elie v e that it w ill n o t cap ture all th e risk th at m ay a ctu ally ex ist. A s w e h ave seen , th is d isa g reem en t is n o t resolvab le in th e sh ort run th rou gh recou rse to scien ce. R isk a ssessm en t is n ecessarily d e p en d en t o n ch o ic es m a d e a m o n g a h o st o f assu m p tio n s, an d th ese ch o ices w ill in evitab ly be affected by the valu es o f the ch oosers, w h eth er they be scien tists, civil servants, or p olitician s. T h e N R C report suggests that th is p rob lem can be su bstan tially alleviated b y th e esta b lish m en t o f form al p u b lic rules g u id in g th e n ecessary in feren ces a n d a ssu m p tio n s. T h ese rules sh o u ld be based o n th e best availab le in fo rm a tion con cern in g th e u nd erlyin g scien tific m ech an ism s. A d op tion o f su ch gu id elin es red u ces th e p ossib ility that an EPA ad m in istrator m ay m an ip u late th e fin d in gs o f so m e risk a sse ssm en t so as to a v o id m a k in g th e d ifficu lt, an d p erhaps p olitica lly u n p opu lar, ch o ices in v o lv ed in a risk -m an agem en t d eci sio n . B oth in d u stry an d en v iro n m en ta lists fear th is m a n ip u la tio n -- from differen t b ran d s o f ad m in istrator, n eed less to say. A lth ou gh w e ca n n o t rem o v e v a lu es fro m risk a ssessm en t, w e can a n d sh o u ld k eep th o se v a lu es from sh iftin g arbitrarily w ith th e p olitica l w in d s. T h e ex p licit an d op en co d ification suggested b y the N R C w ill also en su re that th e a ssu m p tio n s u sed in risk a ssessm en t w ill at least b e u n iform a m o n g all agen cies th at a d o p t th em , w ill b e p lausible scien tifically, a n d w ill reflect a predictable a n d relatively co n sta n t p olicy a m id this co m p lex a n d ch aotic h ybrid d iscip lin e . It a lso offers th e p ossib ility th at o n e d ay all th e p ro tectiv e 28 ISSU ES IN S C IE N C E A N D T E C H N O L O G Y i RISK. SCIENCE. AND DEMOCRACY I I a g en cies o f g o v e r n m e n t w ill sp eak w ith o n e v o ic e w h e n th ey ad d ress risks, so th at estim a tes o f risk w ill b e co m p a ra b le a m o n g a g en cies a n d th e p u b lic at last w ill b e a b le to m a k e a fair co m p a riso n o f th e in d iv id u a l risk -m a n a g em en t d ecision s o f separate agen cies. T h e rem a in in g p o in ts o f b oth p o sitio n s are really a b o u t risk m a n a g em en t an d o n th is issu e b o th are flaw ed . A t its ex tr em e, th e first p o sitio n -- that regulation sh ou ld be b ased solely o n scien tifically p rovab le co n n ectio n s b etw een p ollu tan ts a n d h ealth effects-- w o u ld a llow th e release o f u n lim ited q u a n tities o f su b stan ces that cau se can cer in an im als, o n the assu m p tio n that there w ill be n o a n a lo g o u s effect on p eo p le a n d that there m u st be th resh old s for carcin ogen esis. I ex p ect that m ost A m erica n s w ou ld reject that assu m p tion as im p ru d en t, given ou r current k n ow led ge ab ou t carcin ogen esis (for ex a m p le. th e sim ilarity o f can cer-cau sin g g en es across sp ecies). A t so m e level w e h ave to regard th e p o ssib ility th at w e are c o n tro llin g so m ew h a t in ex cess o f th e tru e risk as a k in d o f in su ran ce, w ith th e c o st o f co n tro l as its p rem iu m . T h e ef fort to reduce ap p reh en sion , even so-called u nreason ab le ap p reh en sion , ab ou t th e fu tu re resu lts o f cu rren t p ractices is a v a lid so c ia l fu n ctio n . R isk -m a n a g e m en t agen cies su ch as EPA co u ld be chartered to d o p recisely that. If so, w e h a d b etter m a k e clea r w h a t w e are d o in g , a n d esta b lish ru les for d o in g it. T h e w eak n ess o f th e se c o n d v ie w p o in t, th at an y id en tifiab le risk o u g h t to be e lim in a ted up to th e cap acity o f ava ila b le tec h n o lo g y to d o so. lies in th e co n ce p t o f a best availab le tech n o lo g y th at m u st in variab ly be ap p lied w here risk is d isco v ered . " B est" a n d " a v a ila b le" are term s as in fin itely d eb atab le as th e a ssu m p tio n s o f risk a sse ssm en t. T h ere is a lw a y s a te c h n o lo g y c o n c e iv a b le th at is an im p r o v em en t o n a p rev io u s o n e. a n d as th e last in crem en ts o f p o llu tio n are r e m o v e d , th e c o st o f e a c h su c c e ssiv e fix g o e s u p very steep ly. B ecau se, accord in g to the n o-th resh old a ssu m p tio n , even m in u te q u an tities o f carcin ogen s can be projected ou t to cau se cases o f d isease, argu m en ts ab ou t te c h n o lo g y red u ce in th e en d to a rg u m en ts a b o u t risk an d cost: tec h n o lo g y A a llo w s a resid u al risk o f 10'5 an d c o s ts $1 m illio n : te c h n o lo g y B a llo w s a r e sid u a l risk o f 10'6 a n d c o s ts $ 1 0 m illio n , a n d s o o n a d in fin itu m . It is s p e c io u s to p reten d th a t c o sts d o n o t m atter, b e c a u se it is a lw a y s p o ssib le to sh o w th at at a certain level o f rem oval, co sts in fact d o m atter: tec h n o lo g y Z al l o w s r e s id u a l r is k s o f 1 0 '15 a n d c o s t s $ 1 t r illio n . O n c e th is is a d m itte d , as it a lm o st a lw a y s is w h en w e c o m e d o w n to d e b a tin g a ctu a l reg u la tio n s, th e p o sitio n is red u ced to a rg u m en ts a b o u t affordability. T h is to o is treach erou s grou n d . F irm s v a n ' in th eir ab ility to pay. a n d w h a t is afford ab le for o n e m a y b a n k ru p t an oth er. If req u irem en ts are a d ju sted so as n o t to crip p le the p o o rest firm s, th e p o licy a m o u n ts to an e n v ir o n m en ta l su b sid y to th e less efficien t players in o u r ec o n o m y . In th e en d . d iscu ssio n s a b o u t availab le tech n o lo g y an d affordab ility d ev elo p in to d iscu s s io n s o f c o r p o r a t e m a n a g e m e n t : " I f t h e y c a n 't a ffo r d p o llu t io n c o n t r o l, h o w c o m e th eir ex ecu tiv es got su ch a big b on u s?" T h is m ay b e a fascin atin g to p ic to so m e , b u t it d istracts from the m a in issue. M y p o in t is th at in co n fro n tin g a n y risk th ere is n o w a y to esca p e th e q u e s tio n " Is c o n tr o llin g it w o rth it?" W e m u st ask th is q u e s tio n n o t o n ly in term s o f th e rela tio n sh ip o f th e risk red u ced a n d th e co st to th e e c o n o m y b u t a lso as it a p p lies to th e reso u rces o f th e a g en cy in v o lv e d . P o licy a tte n tio n is th e S P R IN G 1VSS In confronting any risk there is no way to escape the question "Is controlling it worth it?" 1i. Qu> f l 9 + - 2SL That vision o f the essential unity o f ,nature has been lost and with it the capacity to reach social consensus on environmental policy. 30 m o st p reciou s co m m o d ity in g o v ern m en t, an d a regulation that m argin ally protects o n ly 20 p eop le m a y take up as m u ch atten tion as a regulation that su rely p rotects a m illion . " Is it w o rth it?" T h a t th is q u e s tio n m u st b e a sk ed a n d a sk ed ca r e fu lly is a tok en o f h o w th e m ain force o f th e en viron m en tal idea has b een m od ified by th e recen t fo cu s o n to x ic risk to h u m a n h ealth. In truth this q u estio n sh o u ld alw ays h ave b een ask ed , but b eca u se th e early goals o f en v iro n m en ta lism w ere so o b v io u s ly g o o d , th e r e q u ir e m e n t to ask , " Is it w o rth it?" w a s n o t firm ly b u ilt in to all o u r en v ir o n m e n ta l law s. W h o w o u ld d are to q u estio n th e w orth o f sa v in g L a k e Erie? E n v ir o n m e n ta lism at its in c e p tio n w a s a gran d v isio n , o n e th a t n early all A m e rica n s w illin g ly sh ared. S o m e h o w th at v isio n o f th e essen tial u n ity o f n atu re a n d o f th e n eed for b rin gin g in du strial so ciety in to h a rm o n y w ith it h as b een lo st a m o n g th e parts p er b illio n , a n d w ith it w e h a v e lost th e ca p a city to reach so cia l co n sen su s o n en v iro n m en ta l p olicy. VI >f j | f ? W h y h as th is h ap p en ed ? I b eliev e it is b ecau se en v ir o n m en ta lism , lik e m a n y a n o th er so cia l m o v e m e n t, is su fferin g from th e ex cesses o f its o w n y o u th fu l vigor. In th e early leg isla tio n p ro m ises w ere m a d e th at c o u ld n o t reason ab ly b e kept; ex p ec ta tio n s w ere raised that w ere b o u n d to b e d isa p p oin ted . A case in p oin t is th e lan gu age o f S ectio n 112 o f th e C lean A ir A ct, w h ich requ ires th at EPA estab lish " an am p le m argin o f safety" in th e co n tro l o f h a za rd o u s air p o llu ta n ts. W e can a ch iev e th is m argin o n ly w h e n it is p ossib le to d eterm in e a level at w hich there are no ap paren t effects o n h u m an s, as w e m ay b e ab le to w ith su ch su b stan ces as carbon m o n o x id e an d n itro g en o x id es. It th en b e c o m e s p o ssib le to roll that lev el b a ck to o n e p resu m ed to p ro tect th e m o st se n sitiv e in d iv id u a ls in th e p o p u la tio n , w h ic h is, o f co u rse, th e p o in t o f th e " a m p le m arg in " language. M o st o f th e air p ollu tan ts n ow u nder con sid eration , how ever, are su spected carcin ogen s. B e c a u se , a s a lr ea d y n o te d h ere, w e h a v e a ssu m e d that it is im p o s s ib le to estab lish n o-effect levels for ca rcin o g en s, w e ca n n o t estab lish th e m argin o f safety d em a n d ed by th e law w ith o u t b a n n in g th ese su b stan ces or th e p rocesses th at p ro d u c e th e m , a n d su ch a c tio n is n early alw a y s u n w ise for so c ia l a n d eco n o m ic reasons. T h e C lea n A ir A ct p ro m ised ab solu te p rotection from airb orn e ca rcin o gen s. but w e ca n n o t k eep th at p rom ise. T h e A m erican p eo p le w ere also p ro m ised u n d er th e c lea n air a n d w ater law s that th e air w o u ld b e p u re a n d w ater p o llu tio n w o u ld b e elim in a te d by a sp ecific date; w e h ave n o t k ep t th o se p ro m ises either. F ew th in gs are m o re corru p tin g to a free so ciety su ch as ou rs th an gran d p ro m ises u n fu lfilled . S en a to r D a n iel P atrick M o y n ih a n (D -N .Y .) has w ritten o f this; T h e m alaise o f o v erp ro m isin g d erives alm ost w h olly, in m y ex p erien ce, from th e failu re o f ex ecu tiv es an d legislators to u n d er s t a n d w h a t is risked w h e n p r o m i s e s a r e m a d e -------W h e n t h i n g s d o n 't w o r k o u t a s p r o m is e d it is a ll t o o e a s y t o s u s p e c t t h a t s o m e o n e intended th e y s h o u ld n o t. ISSUES IN SCIENCE AND TECHNOLOGY r is k , s c ie n c e , a n d d e m o c r a c y Many o f the promises made or implied by federal law as part of the triumph of environmentalism have indeed had this sad effect. Since the beginning o f EPA. there has been insufficient appreciation of the difficulties involved in the successive tasks the agency has been assigned by Congress. As a result many people on Capitol Hill assume that if unrealistic promises have not been kept, some people are not doing their jobs-- and that the way to ensure that they do is to write ever more stringent prescriptions into environ mental law, with appropriate deadlines. Thus more promises are added to unrealistic promises, and the cycle of disappointment and growing mistrust is accelerated. The foregoing should in no way be taken as implying that the control of toxic substances is unimportant or that EPA could not be better managed. My point is that EPA will not do an effective job in either case unless we realize that the debate about risk in its present form leads us ever deeper into a blind alley. We must revisit that crucial stage of problem solving in government at which one defines what kind of problem it is. We will emerge from the blind alley and reforge a practical consensus on the environment only when we are able to redefine the problem of environmental protection as being "the management of risk." When we adopt this definition, we specifically will abandon the impossi ble goal of perfect security and accept the responsibility for making difficult and painful choices among competing goods. We can then move forward from the present impasse and begin to look at a more productive set of questions. VII What then is risk management? As management in general can be defined as the distribution of current resources to shape some desirable future state, risk management in its broadest sense means adjusting our environmen tal policies to obtain the array of social goods-- environmental, health-related, social, economic, and psychological-- that forms our vision o f how we want the world to be. In practical terms, risk management means giving the protective agencies flexibility comparable to that which managers have traditionally exercised in other spheres. This suggestion, of course, runs counter to the inclination to further restrict the discretion accorded those agencies. Administrations do vary in their priorities and competence, so how can we ensure that this flexibility will not be misused? Obviously, flexibility cannot be taken to mean carte blanche for the agencies. Flexibility in risk management must be bounded by evidentiary rules of the type described in the NRC report, as well as by rules to assess the adequacy and competence of information. Most important, flexibility should be limited by broad public acceptance, including acceptance by individuals subject to risk, and also by more sensible and appropriate congressional oversight. It may be argued that the administrator of EPA already has sufficient flexibility. Did we not, for example, attempt to get around the impossible standard in Section 112 o f the Clean Air Act by adopting a best-available- SPRIN G 1985 31 technology approach that is not mentioned in the act? The answer is that this is the wrong kind of flexibility. Agency staffs can always finagle their way toward some workable solution out of almost any legislative language. This does not mean that we should not strive for clear and realistic language in our laws. Once the law is written. Congress has the responsibility to provide oversight to ensure that the reasonable goals of the law are carried out (an activity that Congress now performs very poorly). This is not the same as telling the EPA administrator what to do on specific issues. I believe that the tendency of Congress to give such direction will diminish over the long run if the protective agencies are able to redefine their missions and to accept the responsibility for showing what they are doing as clearly as possible; that is, what they are doing in terms of risk management. This approach requires FIGURE 2: DISTRIBUTION OF HAZARDOUS WASTE SITES INTHE UNITED STATES Key: Number of sites in states 0-2 3-10 11-25 over 25 32 much more flexibility to address specific pollution problems than many of the most important laws now permit. Two characteristics of such problems make it imperative that we begin to address them in this way. The first of these has to do with the distribution of toxic substances: we typically find large variations in risk in different parts of the country or even in different parts of the same community. The second characteristic is that toxic risk engenders extraordinary fear, which may paralyze rational public policy and lead to unproductive or even perverse actions on the part of government. Let us deal with each of these characteris tics in turn. The distribution of pollutants is obviously not uniform across the country. We find abandoned hazardous waste dumps only in particular places, and we find the same sort of localization in connection with industrialprocess releases of toxic substances. Receiving waters, whether on the surface ISSUES IN SCIEN CE A N D TECH N O LO G Y T RISK. SCIENCE. AND DEMOCRACY ) or in th e grou n d , d iffer in th eir v u ln era b ility to p o llu tio n , as d o air sh ed s. Further, p u b lic ex p o su re, a n d h en ce risk, varies greatly d ep en d in g o n the lo ca tio n o f p o llu tio n so u rc es in relation to p o p u la tio n cen ters. F inally, d ifferen t p o p u la tio n s m a y d iffer g reatly in th eir o p in io n s a b o u t w h at risks th ey are w illin g to en d u re in return for w h at b en efits. C urrent law m a k es it hard to reco g n ize su ch d istin ctio n s. F or exa m p le, the C lean A ir A ct requ ires EPA to im p o se ex p en siv e a u to m o b ile in sp ectio n an d m a in ten a n ce p rogram s in c o m m u n itie s w h ere certain p o llu ta n t criteria have b een exceed ed tw ice a year. EPA m u st d o this ev en th ou gh th e violation s m ay b e a co n seq u en ce o f th e p la cem en t o f th e air-quality m o n ito rin g d evices a n d m a y n o t reflect th e g e n e r a l q u a lity o f th e air. e v e n th o u g h th e re is n o d is cern ib le h ealth effect, an d ev en th ou gh th e p eo p le in th e co m m u n ity stron gly o p p o se th e action . T h e law d o es n ot a llo w th e federal g o v ern m en t to d istin gu ish b etw een (for ex a m p le) L os A n g eles an d S p ok an e, W ash in gton , in this regard-- a restriction th at d efies c o m m o n sen se. In th e sa m e w ay, w e ca n n o t d istin gu ish b etw een a p lant d isch argin g p ollu tan ts in to a highly stressed river in C o n n ecticu t an d o n e d isch argin g in to A lask an w aters that bear n o o th er p o llu ta n t b u rd en . In o th er w ord s, the law d o es n o t p erm it u s to act sensibly. V III In m y view , so u n d p u b lic p o licy w o u ld give E PA th e flexib ility to co n fro n t an d d eal w ith risks in th e local co n tex t. T h is flexib ility w o u ld en tail, first, b a la n c in g risks a g a in st th e lo c a l e c o n o m ic im p a c ts o f c o n tr o llin g th em ; seco n d , en su rin g th at o u r n a tio n a l p rogram s th at a ttem p t to d ea l w ith local risks o p era te a cco rd in g to r isk -m a n a g e m e n t p rin cip les; a n d th ird , in v o lv in g th e local p u b lic in a m ea n in g fu l w a y in th e d ecisio n m a k in g p rocess. A typ ical ex a m p le o f th e lo c a liz a tio n o f risk is th at in w h ich an in d u strial p lan t im p o ses so m e lo ca l risks d esp ite th e in stallation o f a d v a n ced p o llu tio n co n tro ls. L o ca l e c o n o m ic in terests th en co n fro n t local h ealth in terests in red u cin g risk. A p a rad igm o f su ch ca ses w as th e situ a tio n in T a co m a . W ash in gton , w here u n til recen tly a cop p er sm elter p rocessed arsen ic-rich ore a n d re le a sed q u a n titie s o f th is c a r c in o g e n in to th e a m b ie n t air. E v e n after th e p lan t w as h eavily co n tro lled , it ap p eared im p o ssib le to e lim in a te th e ca rcin o g en ic risk from th is release; th a t is. e lim in a tin g th e risk m ea n t e lim in a tin g th e plant. A s EPA ad m in istrator, I b eliev ed EPA had a resp on sib ility to ex p la in to the p eo p le w h o w o u ld b e m o st d irectly affected by th e d ecisio n w h at w e k n ew ab ou t th e risks from th e sm elter. T rue p u b lic in v o lv e m e n t m ea n t forcing th e p u b lic to co n fro n t th e trade offs in v o lv ed in th is risk -m a n a g em en t d ecisio n . W h en w e b egan to ex a m in e th e T a c o m a situ a tio n in d eta il, w e d isco v ered th at m o st o f th e lo ca l p eo p le w ere w illin g to v iew th e p ro b lem in term s o f th e risk -m a n a g em en t ch o ice, alth o u g h th eir p o sitio n s o n th e arsen ic risk d ep en d e d (u n d ersta n d a b ly ) o n w here th ey lived in relation to th e d istrib u tion pattern o f th e em issio n s, an d on w h eth er th ey had a p erson al e c o n o m ic stake in th e p lan t rem a in in g op en . A s th e p u b lic d isc u ssio n s c o n tin u e d a n d w e refin ed o u r d a ta o n b o th th e risk a n d th e m e a n s o f re d u cin g it, th e citiz e n s o f T a c o m a b egan to c o m e u p w ith In my view, sound public policy would give the EPA the flexibility to confront and deal with risks in the local context. S P R IN G 1985 33 19628 h elp fu l id eas a b o u t h o w w e co u ld m in im ize the actu al im p act o f th e arsen ic em issio n s a n d k eep th e sm elter op eratin g. A lt h o u g h t h e p la n t 's o w n e r s e v e n t u a lly d e c id e d t o c lo s e it (fo r r e a s o n s n o t d irectly c o n n e c te d to th e p o llu tio n issu e), w e learn ed so m eth in g valu ab le from th e ex p erien ce. D esp ite in itia l fears, it is p o ssib le for p eo p le su b ject to to x ic risk to th in k r a tio n a lly a b o u t it. It is p o ssib le for th e m to c o n fr o n t th e hard truth th at so lu tio n s to su ch p rob lem s n ecessarily in volve an u n even d istrib u tion o f risks a n d b en efits. T h is ration al th in k in g in v o lv es p lu n gin g in to th e u n certa in ties in v o lv ed in th e an alyses w e u se to d efin e th e issues. H o w sure are w e ab ou t exposure? H o w sure are w e ab out the effects o f the su bstan ce? H o w certain are th e ec o n o m ic im pacts? S u ch ra tio n a l th in k in g req u ires a k in d o f d em o cr a tic citizen sh ip th a t is w illin g to d ig d eep er th an th e glib h ead lin es a n d th e u su al in v ita tio n s to p an ic. It a lso req u ires m u ch m o re from th e regulatory a g en cy p rovid in g th e facts: T h e agen cy m u st b e w illin g to ex p lain an d ab le to co m m u n ica te, a n d m o st o f all, it m u st a d m it th e u n certa in ties b u ried in its ca lcu la tio n s. O n ly th e n can th e ap p rop riate b a la n cin g d ecisio n s take p lace. O u r en v iro n m en ta l law s sh o u ld sp ecifica lly p erm it E PA to m a n a g e a lo c a liz ed risk in th is m an n er. R isk m a n a g em en t th u s p rovid es us w ith th e flexible ap p roach n eed ed to d eal w ith th e w ay to x ic su b sta n ces are actu ally d istrib uted . H ow ever, it m u st also en ab le u s to d eal w ith th e seco n d characteristic o f the current situ ation w ith resp ect to to x ic su bstan ces: p u b lic fear an d th e m istrust often associated w ith it. H o w ca n risk m a n a g e m e n t, a logical a n d ration al p rocess, c o p e w ith em o tio n a lly b ased p u b lic reactions? W h ile w e d o n ot yet h ave a perfect so lu tio n to th is p rob lem , w e d o k n ow en o u g h , I thin k, to id en tify so m e n o n so lu tio n s. R elia n c e o n exp erts is n ot a so lu tio n . P sych o lo g ist P au l S lo v ic w rites o f this: S in ce ev en w ell-in fo rm ed layp eop le h ave d ifficulty ju d g in g risks a cc u r a tely , it is te m p tin g to c o n c lu d e th a t th e p u b lic s h o u ld b e re m o v ed from th e risk a ssessm en t p rocess. S u ch a ctio n w o u ld se e m to b e m isg u id e d o n several c o u n ts. First, w e h ave n o assu ran ce that experts' ju d g m en ts are im m u n e to biases o n ce they a r e f o r c e d t o g o b e y o n d h a r d d a t a ____ S e c o n d , in m a n y i f n o t m o st cases, effectiv e h azard m a n a g e m e n t requires th e co o p era tio n o f a large b o d y o f lay p eo p le. T h ese p eo p le m u st agree to d o w ith o u t so m e th in gs an d accep t su b stitu tes for others; th ey m u st v o te sen sib ly o n b allot m easu res and for legislators w h o w ill serve th em as su rrogate h azard m anagers: th ey m u st o b ey safety rules a n d u se th e legal sy stem resp onsibly. E ven if th e exp erts w ere m u c h b etter ju d g es o f risk th an la y p eo p le, g ivin g exp erts an ex clu siv e fran ch ise for hazard m an a g em en t w ou ld m ea n su b stitu tin g sh ort-term efficien cy for the lon g-term effort n eed ed to create an in fo rm ed citizenry. B u ryin g q u a n tita tiv e estim a tes o f risk w ith in so m e p roced u ral regulatory fram ew ork , an abstract scorin g system perhaps, is also a n o n so lu tio n . In ISSUES IN SCIENCE A N D TECH N O LO G Y RISK , s c i e n c e , a n d d e m o c r a c y debates surrounding any environmental decision, however, the subject o f risk is bound to arise, and then the agency will have to explain not only the risk it self but also why it was trying to "hide" it. The regulatory agencies must accept the fact that the public is part o f the regulatory process. Given the atmosphere of mistrust that has characterized governmental relations with the public since Vietnam and Watergate, it is vital that the agencies be forthcom ing with all the information involved in their decisions. Furthermore, as we have seen, regulatory approaches not based on quantified risk must eventually confront the economic realities and become risk management in all but name. It is thus impossible for modem environ mental policymakers to avoid dealing directly with named, quantified risks and the public fear these engender. IX I believe that we can mitigate this fear only by adopting as policy the full disclosure of the risks involved in regulatory decisions. This belief is not universally shared within the environmentalist or the regulated communities or even at the protective agencies. It is not pleasant to tell people that some re sidual risk is nearly always associated with practical levels of pollution control. It becomes less and less pleasant as one gets closer to the most affected locality. In a wav. the great national debates on environmental policy are resolvable only in high school auditoriums across the country and in terms of what public officials are able to say to worried citizens. The questions that these audiences naturally ask (Can we drink the water? Will my child get cancer?) can hardly ever be answered unequivocally. The uncertainties involved in risk assessment must be explained, and people must be brought to understand that "safety" is a social construct, the definition of which is. or ought to be. a part of each citizen's duty in our society. Our skill at developing such constructs for new apparent dangers is not yet what it should be. Communication and public education must become a much more important part of the work o f EPA and the other protective agencies, on a level with their scientific, engineering, policymaking, and legal functions. Risk management is not merely a set of techniques for arriving at the correct answers. It must include communication to the public about how we arrive at environmental protection decisions. The values and assumptions that underlie all such decisions must be made manifest. Transparency is the object o f the whole process, and public trust is the ultimate goal. But it is not easy, for we must confront one of the inevitable dilemmas of leadership in a democracy, that of having to choose between telling people what they want to hear and telling them what they ought to know. A good deal of our environmental legislation has bent toward the former, which has made things more difficult for the officials who have to stand up on the stages of those high school auditoriums. If people have been told in statutes that risk can be made to vanish through governmental actipn, they will be satisfied with nothing less, and they will not treat kindly the messenger who has to tell them it cannot be done. If risk management is to be used as a means o f coming to decisions at spe- I believe that we can mitigate this fear only by adopting as policy the fa ll disclosure o f the risks involved in regulatory decisions. S P R IN G 1985 13630 ria l lo c a les, th e p rin cip le m u st b e sp ecifica lly esta b lish ed in en v ir o n m e n ta l law . O u r sta tu tes sh o u ld reflect th e reality th a t e n v ir o n m e n ta l p ro tectio n in an im p erfect w orld m ea n s th e a ssessm en t o f risks a n d th e ab ility to m an age th em as p articu lar situ a tio n s w arrant. N o te th a t th is sh o u ld g o b ey o n d h ealth risk in th e strict sen se. W e sh o u ld learn to lo o k at all th e im p a c ts o f p o llu tio n from the p ersp ective o f b alan cin g so m e d efin ab le im p ro v em en t against ou r alw ays finite resources, w h eth er con trol exp en d itu res or govern m en tal atten tion . T h is is a lw a y s a hard ca se to m a k e, esp e cia lly to C on gress, w h ere the p rizes se e m to g o for n e w p ro g ra m s a n d e v e r m o r e strin g en t o n es. It so m etim es seem s that th o se w h o w rite o u r p rotective statu tes w o u ld rather h a v e EPA p reten d to b e d o in g a th o u sa n d tasks th a n h a v e it select th e h u n d red m o st im p o rta n t task s an d d o th em w ell. I a m n o t sayin g th at ou r p resen t law s h ave n ot w ork ed . T h ere is n o q u e stio n th at w e h a v e m o u n te d a su ccessfu l n a tio n a l effort again st th e m o re gross sorts o f p o llu tio n . T h o se law s, h ow ever, w ill b eco m e in creasin gly in effective if th ey force us to d eal w ith to x ic risk, or ev e n w ith th e final rem a in in g in cr em en ts o f c o n v e n tio n a l p ollu tan ts, in th e w ron g w ay. O u r law s n eed less p io u s h o p e a n d m o re d eterm in a tio n to strik e at th e sou rces o f actu al h arm . T h e m a n a g em en t o f E PA w o u ld surely b en efit from su ch legislative ch an ges. F or ex a m p le, in th e air p rogram w e w o u ld a llo c a te resou rces n o t to th e in evitab ly slow p rocess o f listin g an d con trollin g a h and fu l o f th e im m e n se n u m b er o f ch em ica l sp ecies that co u ld cau se trou b le b u t rather to th ose actu al situ a tio n s in w h ic h so m e u n a c cep ta b le risks ca n b e fo u n d . W e w o u ld w o rk w ith in d u stry an d th e affected c o m m u n itie s to d ev elo p accep tab le levels, w h ich w o u ld natu rally vary w ith co n d itio n s sp ecific to d ifferent areas. W h ere ap propriate, w e w o u ld h a v e th e flexib ility to estab lish a m b ien t, so u rcesp ecific. or tech n o lo g y -b a sed stan d ard s tailored to th e p articu larities o f each risk situ a tio n . In th e w a ter p rogram w e w o u ld p ay m o re a tte n tio n to th e im p a c t o f 36 ISSU ES IN S C IE N C E A N D T E C H N O L O G Y 19631 r is k , s c ie n c e , a n d d e m o c r a c y pollutants on aquatic ecosystems, and invest time and attention where it promises to do the most good. We would question whether it pays to go from 95 to 99 percent removal of toxic substances and conventional pollutants at industrial sources where we know that the bulk of the pollution load on the water comes from urban and agricultural nonpoint runoff. By now it is clear that we are not going to eliminate all pollutant discharges into our waterways by the date specified in the Clean Water Act. Should we therefore wring our hands and cry "failure"? No. we should decide what kind o f water quality we can live with, in which bodies of water, and what we are really prepared to pay to get it. and then we should put that kind of flexibility into law. In groundwater protection we should start relating our concern for this resource to its presumptive use. Groundwater is a resource like any other. Obviously we cannot allow our groundwater to be carelessly fouled, but let us cultivate a sense of proportion as well as a sense of outrage. There is a great dif ference between the careless and casual destruction of a supply of drinking water and the demand that a vast pool of groundwater lying beneath an industrial metropolis be scrubbed as clean as it was when a dozen farmers lived above it. We accept risks from chlorinated drinking water in most places in the United States right now. risk levels that many communities have declared to be unacceptable when presented by way of groundwater. Most chlorinated drinking waters contain chloroform, a product o f the reaction between chlorine and the humic acids occurring naturally in many water sources. We have established an acceptable level for chloroform in drinking water, based on a balancing of the carcinogenic risk of this substance against the risk of pathogenic diseases that could result from insufficient chlorination. We ought to eventually establish the same kind o f acceptable levels for the potential carcinogens that may show up in groundwater. The hazardous waste problem boils down to specifying where we want the waste to go. In the last analysis, after we have reduced production and have recycled and treated as much as we can of our waste, we have three choices as to where to put it: the land, the air. or the water. Congress has never analyzed the total risk associated with a certain hazardous waste policy and the total cost of eliminating it and then used the analysis as the basis for legislation. However, that is what we will increasingly have to do if our national environmental policies are to retain any contact at all with the dayto-day realities they are supposed to influence. In the control o f pesticides, we have relied heavily on the banning of chemicals that entail unacceptable risks. We must continue to do so. of course, but we must go beyond the single-substance ban as our major weapon. The industrial laboratory will always outpace the regulatory agency in providing substitutes for banned chemicals, and some of those substitutes in field use may prove as troublesome as the ones they replace. We must start looking at the long-term risks posed by whole systems of agriculture-- land management, tillage practices, and herbicide and pesticide applications-- and not just at the health risks associated with chemical residues on food or applicator contact. What is the impact o f all this chemical loading over the years on the The hazardous waste problem boils down to specifying where we want the waste to go: the land, the air, or the water. S P R IN G 1985 37 ecological systems in which human culture is embedded? After decades of socalled pesticide control, we have not even begun to ask this question. Indeed, it is odd how little time is spent at the upper levels of EPA thinking about such things and how much time is spent worrying about tiny increases in the risk of a single human disease. We need a return to the vision of environmentalism embodied in the documents that created EPA. X Some years ago, when pollution control seemed a far more overwhelm ing task than it does now, Ren Dubos said that the way to cope with such massive problems was to "think globally and act locally." It is still good advice, and may serve as a general prescription for successful management of technological risk in a democratic society as big and diverse as ours. Global thinking in the present case means dealing explicitly with the central questions of risk management: how to reconcile technological systems with social values; how to develop the consensus about potentially dangerous technol ogies that is necessary for continued growth; and how to establish and maintain trust in our protective institutions. We do not yet know how to deal very well with any of these questions, although I have tried to suggest some promising directions to follow. This is why local action--a diversity of local actions--is necessary as well. I suspect that the most efficient wav for our society to learn how to cope with risk is to enable hundreds or thousands oflocally based risk-management endeavors to take place. Local risk-taking preferences could then be ex pressed, under broad limits set by higher levels of government. This will inevitably change public perceptions of risk, for in such perceptions familiar ity breeds, not contempt, but the ability to discriminate between trivial and important risks. Fear is, after all. what has tended to paralyze public policy on these issues, and all our research shows that familiar risks engender less fear than unfamiliar ones. As our people begin to assess and manage risks at the lo cal level, they will be preparing themselves to cope as citizens of a democratic society moving into a future dominated by barely imaginable technologies and fraught with unfamiliar risks. 38 ISSUES IN S C IE N C E A N D T E C H N O L O G Y 19633 VOLUME I NUMBER 3 SPRING 1985 IN SCIENCE AND TECHNOLOGY NATIONAL ACADEMY OF SCIENCES. NATIONAL ACADEMY OF ENGINEERING. INSTITUTE OF MEDICINE 3 FORUM 19 Risk. Science, and Democracy William D. Ruckelshaus 39 GENETIC ENGINEERING: BALANCING RISK A N D REWARD 40 Regulating Biotechnology Thomas O. McGarity 57 Ecological Consequences: Reducing the Uncertainties Martin Alexander 69 Our Investment: What is at Stake? Ralph W. F. Hardy and David J. Glass HIGH TECHNOLOGY AND DEFENSE: HOW LARGE A ROLE? 83 Weapons. Platforms, and the New Armed Services Seymour J. Deitchman 100 Smart Weapons: But When? Richard L. Garwin 108 A Doubtful Revolution William S. Lind 113 China Joins the World: Prospects and Implications Michel C. Oksenberg 128 Antitrust Policy and Technological Innovation: A Response Daniel C. Schwartz and J. Michael Cooper 136 NEW PUBLICATIONS * V INSCIENCE AND TECHNOLOGY FRANK. PRESS, president National Academy ofSciences' RO B ER T M. W H ITE , president National Academy of Engineering F R E D E R IC K C. ROBBINS, president Institute of Medicine PH IL IP M. S M IT H , executive officer ISSUES IN SCIENCE AND TECHNOLOGY is published to inform public opinion and to raise the quality o f private and public decisionmaking by providing a forum for discussion and debate. Accordingly, the pages o/'ISSUES IN SCIENCE AND TECHNOLOGY are open to all responsible points o f view, and the material published here reflects'only the views o f the authors, not a consensus o f opinion or the opinion o f any institution. ALLEN L. H A M M O N D editor PHILIPS. COOK m a n a g i n g editor LESLIE ROBERTS senior editor M A R J O R I E B. D A HLIN assistant to the editors NORMAN METZGER contributing editor KEL L& C H A D IC K. INC. design V IR G IN IA B. M AR T IN publisher MARY KOPPAL circulation director MARY MULLINIX lulhllment m a n a g e r DAWN M. EICHENLAUB production m a n a g e r ADVISORY BOARD SIM O N RA M O . chairman T H O R N T O N F. BR A D SH A W J. FRED BUCY WILLIAM H. DANFORTH JO H N M. DEUTCH ROBERT C. FORNEY PA U L E. GRAY DONALD KENNEDY s.Ro b e r t McN a m a r a SHERM AN M. M ELLINKOFF FRANKLIN D. M U R PH Y PETER G. PETERSON JO H N S. R E ED IRVING S. SH A PIRO STEVEN W EINBERG M ARINA v.N. W HITM AN Published quarterly by the N ational A cadem y o f Sciences. 2101 C o nstitution A venue. W ashington. D .C . 2 0 4 1 8 . C o p y rig h t 19 85 by th e N a tio n a l A c a d e m y o f S c ie n c e s . P rin te d in th e U .S .A . IS S N 0 7 4 8 5492. E ditorial C orrespondence: By m ail to the address above, o r call (202) 334-3305. U nsolicited m anu scripts will be co n sid ered b u t can be retu rn e d only if acco m p an ied by a self-addressed, sta m p !: envelope. R eader Services: Inquiries o r co m m u n icatio n s concerning subscriptions, single-copy sales, perm is sions. and reprints should be sent to R eader Services. ISSU ES IN SC IEN C E A N D T E C H N O L O G Y , 2101 C onstitution A venue. W ashington. D .C. 20418. S ubscription prices: U.S. $24.00; foreign $32100. Second-class postage paid a t W ashington. D .C.. an d ad d itio n al m ailing offices. A d v e rtisin g In q u irie s : T e le p h o n e V irg in ia B. M a rtin a t (2 0 2 ) 3 3 4 -3 3 1 8 . P ostm aster: Send changes o f address to ISSUES IN SC IEN C E A N D T E C H N O L O G Y , 2101 C onstitu tion A venue. W ashington. D .C. 20418.' ISSUES IN SCIEN CE A N D T E C H N O L O G Y