Document kaQ3kMvkJBjmkoOpVxYkeEJZE

POLYCHLORINATED BIPHENYL COMPOUNOS (Slide l) I. INTRODUCTION (Slide 2) A. PRODUCTION o Polychlorinated biphenyl (PCB) compounds were first synthesized by Griefs In 1857. o PCBs were produced In the United States in 1929 by the Swann Chemical Company bought later In 193S by the Monsanto Company. o Although as many as six companies have had Ireglstered trade marks for commercial brands of PCBs, probably most PCB mixtures have been produced in this country by the Monsanto Company. o It has been estimated that between 1929 and 1970 some 990 million pounds of PCBs were sold in North America alone. o PCBs were produced and/or imported by probably all industri alized nations In the 1950s and 1960s. MOMS 019396 (Slide 3) B. PROPERTIES 0 PCBs range In appearance from a clear, thin oil to a light yel low, thick liquid with boiling points of approximately 300400"C. Haying boiling points 3 to 4 times that of water, very little PC8s evaporate into the air at normal temperatures. o PC8s have certain physical properties which make them very use ful in industry. These properties include thermal stability; resistance to oxidation by acids, bases, and other chemicals; and electrical insulation. (Slide 4) C. USES o Because of the above properties PCBs have been used extensively in the electrical Industry as an Insulator and coolant. They have also been used as sealants for wood or cement surfaces; as cutting oils and hydraulic fluids, as vapor suppressants for Insecticides to retard their evaporation and prolong their ac tion In soil; and In Inks, carbonless duplicating paper, adhe sives, Imeerslon oil for microscopes, and plasticizers. (Slides S. 6. 7) 0. OCCURRENCE o The average air concentration in American cities In 1975 (Miami, Fort Collins, and Jackson) was 0.1 u'g/m3; this level Is the currently allowable maximum concentration for PCBs In workroom air. o According to a 1959 report, PCB levels In workroom air In United States plants ranged from 0.2-10.5'mg/m^. This Is 2,000-100,000 times greater than today's allowable standard. MONS 019397 o It has been calculated that 2.2 million pounds of PCBs fall on the United States yearly in rain and particulate matter. We are constantly recycling PC8s through our environment. o A national study in 1971-74 showed that PC9s in unfiltered water samples ranged from 0.1-3.0 ug/L (ppb). o In 1976, it was found that the average PCS concentrations in seawater was 13 ng/L (ppt). o PCBs were found in the fat of arctic animals (porpoises, seals, foxes, sheep, and polar bears) at concentrations ranging from 0.3-21.0 ug/g (ppm). o A national survey In the United States during the years 1973-74 revealed that 40* of the population had PCS levels of 1 ug/g (ppm) or greater in fat tissue. o A 1973 Finnish study revealed that 11 healthy workers employed In a capacitor factory had PCS blood concentrations of 0.07-1.9 ug/g (ppm) which was 50-100 times that of control groups. o A Japanese survey reported the average PCS Intake as 4-50 ug/person/day. This is 50 times the allowable Intake via air for a worker in the United States. (Slides 8, 9) E. CONCERNS FOR PCS USE o Initial concern about the health' effects of PCS mixtures began in the late '30s and early '40s when production workers coming in direct contact and without proper ventilation developed chloracne and some signs of liver injury. MONS 019398 0 In 1949, Or. R.E. Kelly, describing a human patch test study using the technique developed by the Medical Olrectcr of The United States Public Health Service (USPHS), reported: "If the procedure recommended by the JSPHS is sufficient to distinguish between chemical compounds that are innocuous when applied to the skin and other compounds that are primary irritants or sen sitizers, then it would seem that Arodor (trademark, Monsanto) has no primary irritant effect nor is it a sensitizer of high potency." o In 1968, approximately 1,500 Japanese ingested large amounts of PCBs which had inadvertently leaked into a brand of cooking oil. o Also in the mid '60s, a Swedish researcher investigating OOT residues in the environment found PCBs in fish. The extensive presence of PCBs in the environment was confirmed by 1970. Although PC8s are not as inherently toxic as OOT, they raise similar concerns. PCBs are fat soluble and resist degradation In animals and the environment. Therefore, even though the overall impact of PCBs on the environment was not known, its presence in the environment raised concerns about potential adverse health and environmental effects because its chemical stability would make it hard to remove once a problem was dis covered. MOMS 019399 (Slides 10, 11, 12, 13) II. TOXICITY STUDIES IN ANIMALS A. SYSTEMIC TOXICITY o Toxicology Is the study of adverse effects. At high enough doses every chemical in our environment is toxic, be it a chemical used in industry, as a drug, or even our vitamins and hormone supplements. This aspect was pointed out 400 years ago by PARACELSUS (1493-1541) who stated: "All substances are poisons, there is none which is not a poison. The right dose differentiates a poison and a remedy.'' The interesting aspect of PCBs that is often overlooked is the high doses needed to cause a toxic effect. o The dose lethal to 50* of the animals (LDjq) range for rats from 4,000-14,000 mg/kg. The LDjg for common table salt is 4,000 mg/kg. o The major effects of toxic concentrations of PC8s in the liver include hypertrophy (increased cell size), marked fatty infil tration, and centrilobular necrosis (cell death and degeneration). o Exposure to the skin ranges from reddening, increased pigmenta tion, hair loss, chloracne, blisters, desquamation (peeling or shedding) and skin lesions. o Other effects observed include irritation to the stomach, ul cers, anemia (less than normal number of red blood cells), decreased number of white blood cells, depression of the immune response (inability to fight infections), and prolongation of the menstrual cycle. MONS 019400 (Slide 14) B. EM8RY0T0XICITY, BIRTH DEFECTS AMO REPRODUCTIVE EFFECTS o 100-500 mg/kg/day of the Aroclors in rats (equivalent to l oz. liquid/day in humans) caused decreased mating performance, lit ter size, and survival of offspring. Ho birth defects were seen and the no-effect level was 50 mg/kg. 0 In monkeys, PCBs have caused Irregular and prolonged menstrual cycles, resorptions, and infertility. o PCBs were fetotoxic in rabbits, but caused no birth defects. 0 'Karnofsky's Law"- any chemical administered at a high enough dose at the proper stage of development to the right species will cause disturbances In embryonic development. o For some perspective on Karnofsky's Law it may be noted that the list of known animal teratogens (agents causing birth defects) includes: vitamin A, folic acid, insulin, steroids, androgens, estrogens, adrenalin, stress (heat, cold), and vitamin and hormonal deficiencies. C. MUTAGENICITY o PCBs were tested in the Ames assay; although 4-chlorobiphenyl was mutagenic, the more chlorinated biphenyls showed little activity. o Aroclors did not produce chromosomal aberrations in rat bone marrow or sperm cells. They were also negative in the domi nant lethal mutation test in rats. o No chromosomal aberrations were seen in cultured hiaean white blood cells. MOWS 019401 o Currently there are probably more data against caffeine as a possible mutagen than there are against PCBs. 0. CARCINOGENICITY o The carcinogenicity data In animals concerning PCSs are con troversial. Reoorts have been published concluding that PCSs induce neoplasms at high doses in rats and mice. However, studies by Monsanto and the 1977 National Cancer Institute study were negative. Also Or. Pour of the Eppley Institute for Research in Cancer reviewed both the negative and the positive studies and concluded that there was no occurrence of liver cancer in either study. MOMS Q19402 (Slide 15) III. HUMAN PCB EXPOSURE: A. *YUSHO" At least 1291 persons ingested large mounts of PC8s (1-2 grams) that had leaked Into a brand of cooking oil. The oil contained 800-3,000 mg/kg (ppm) PCBs and required months of exposure before the onset of symptoms. o The most common clinical features In the early phases were dermatological In nature. These symptoms included ocular dis charge and swelling of the upper eyelids, extra skin growth around hair follicles, chloracne, and increased pigmentation. o Other clinical symptoms Included peripheral nerve disturb ances, liver enlargement without liver injury, irregular men struation in women, transient growth retardation in children, elevated serin triglyceride (fatty acids) levels, and bronchitis. o Only mild birth defects were attributed to PCBs; some children were born with skin problems similar to those of the adults. The other effects included premature eruption of teeth and less than average birth size. (Note: Smoking and drinking also correlates with small birth size.) B. OCCUPATIONAL EXPOSURES o For the most part occupational exposure has largely caused skin symptoms only. o The half life (time to eliminate 1/2 of the amount in the body) has been estimated at 90 days. o Currently there is no epidemiologic evidence in studies of workers exposed to PCBs for years concluding that PCBs cause cancer in humans. HONS 019403 C. FOA REGULATIONS The following are the allowable limits of PCBs as regulated by the Food and Crug Administration. 2.5 ppn in dairy products (on at fat basis-whole mtU-3.25X fat) 5.0 ppm in poultry (on a X fat basis) 0.5 ppm in eggs 2.0 ppm in fish and shellfish 0.2 ppm in infant food 10.0 ppm in paper packaging materials e Therefore, under federal regulations we can Ingest and accumu late more PCBs through our food than in the workplace. 0. PERSPECTIVES ON HUMAN EXPOSURE (Slide 16) e Under current regulations, the allowable human exposure levels to PC8s are much lower in the workplace than in everyday sur roundings. For example, total PC8 intake via Ingestion of food may be greater than permissible employee exposure limits. (Slide 17) e PC8s are less toxic than many safe over-the-counter medica tions. (Slide 18) e Vitamin A has been found to be toxic at the same level at which the highest concentration of PCBs have been found to be non toxic in human blood. MOMS 019404 (Slide 19) 8oth aldrfn and DOT are more toxic than and just as persistent as PCBs, but have higher allowable exposures Tor the workplace. The allowable levels for vinyl chloride, a chemical proven to cause cancer in humans, are also higher. HONS 019405 GLOSSARY Abbreviation ppm ppb ppt mg/kg ug/kg 9 g "9 TLV PEL acute chronic teratogen Part per million Part per billion Part per trillion Milligram compound per kilogram body weight Nlcrogram compound per kilogram body weight 1/1,000 of a kilogram 1/1,000 of a gram or 1/1,000,000 of a kilogram 1/1,000,000 of a gram or 1/1,000,000,000 of a kilogram Threshold Limit Value, the highest workplace air concentration as suggested by the American Conference of Governmental Industrial Hygien ists Permissible Exposure Limit, the standard for workroom air concentrations as regulated by OSHA (Occupational Safety and Health Administration) High level short term exposure Low level long term exposure Compound causing birth defects HONS 019406 Glossary (Cont.) Abbreviation mutagen embryotoxicity resorption Compound causing damage to the genetic material Toxic effects to embryo Indicates death of embryo HONS 019407 SLIDE I HONS 019408 PRODUCTION First Synthesized In IIC7 by Griefs e First Manufactured in U.S. in 1929 a Over 1 Billion Pounds Hava Bean Sold in the U.S. Alone a Produced or Imported by All Industrial Nations SLIDE 2 HONS 0 1 9 4 0 9 USEFUL PROPERTIES Thermal Stability Chamlcal Stability Good Heat Capacity a Excellent Dielectric (Electrical Insulation) SLIDE 3 HONS 019410 END USES OF AROCLORS BY TYPE End um 101# 1221 1212 1242 12S1 1250 1260 1262 1261 Current Capacitors Transformers '* X XX X Former Hast Transfer Hydraulics/Lubricants X Hydraulic Fluids Vacuum Pumps **Gas-Transmission Turbinas XXX XX XX XX Plasticisers Rubbers a Synthetic Resins a Carbonless Paper XXX X X X X X X XX X a Miscellaneous a Adhesives e Wax Extenders e Dedusting Agents e Inks a Culling Oils e Pesticide Extenders e Sealants and Caulking Compounds X XXX X XX XX X X X X X SLIDE HONS 0 1 9 4 1 1 OCCURRENCE (.2.-10.S ig/mJ In Air at Plants-1939 Averaga PCB Concantrations In Saawater Wara 13 ng/L Avaraga Japanesa Intaka Calculatad to ba a-50 ug/dsy 2.2 Million Pounds of PCBs Pall on tho U.S. Yaarly PCB RESIOUES IN ANIMALS Year 1971 1977 1973 1974 1970 Spades ppai (Fat) Fish Cows Birds Ducks Falcons and Hawks 3C0 s.a 15.3 3.1 2-UI.OM SLIDE i PCS RESIDUES IN THE GENERAL POPULATION Yaw 1*7* 197* 1772 1975 1772-74 Country Finland Cermany U.S.A. Japan Japan ppm (Fat) 1.77 1.3 1-2.0* 2.6 S.O * 40% of U.S. Population > 1*0 ppm 26% of U.S. Population 1.0-2.0 ppm 14% of U.S. Population > 2.0 ppm ppb (Blood) 3-21 - 0-29 3.2 4.9 HONS 0 1 9 * 1 * SLIDE 7 PCB RESIDUES IN OCCUPATIONALLY EXPOSED Yur 1973 I72 Country Finland Japan Occupation Laboratory Capacitor Plant Capacitor Plant ppb (Blood) 36-63 75-1900 100-650 SLIDE PCB RESIDUES IN HUMAN MILK Yaw 1*70 1*72 1973 1974 Country Coralany Sweden U.S.A. Japan FDA Regulation Cow Milk Whole Milk (ppb) 103 25 100 30 Milk Fat (ppm) 3.S - 1.2 Limit 2.5 HONS 0 1 9 4 1 6 SLIDE t TOXICOLOGY The Study of: (1) Chemical or Physical Agents Which Interact With (2) Any Living System in Which a Response Is Produced (31 Tha Response Must be Considered Harmful or Undesirable SLIDE 10 TOXICITY RATING CHART Rating Probabla Oral Lathal Oosa for Humans* Animal LC^ Expected Human Dose 1. Nontoxic 7. Waakly Toxic 3. Modarataly Toxic 0. Toxic S. Extreiaely Toxic C. Supartoxic IS,000 mg/kg 5,000-15,000 mg/kg $00-5,000 mg/kg 50-500- mg/kg 5-50 mg/kg > 5 mg/kg > 1 Quart 1 Pint-1 Quart 1 Ounce-1 Pint 1 Teaspoon-1 Ounce 7 Drops-1 Teaspoon Less Than 7 Drops * Avaraga Adult 70 kg MQNS 0X 9418 SLIDE II CROSSMATCH THE FOLLOWING CHEMICALS WITH THEIR TOXICITIES Aganti Alcohol Botulinua toxin DOT Dioxin (TCDDI Iron Tablott (FiSO|) Nlcotlno PCBs Strychnin# Table Salt (NaCI) Ltya.g/kgJ 14.000 10.000 4.000 1.S00 100 2 1 0.001 0.00001 6Tr6TM oT V* SLIDE 12 ACTUAL TOXICITY ANO RANKING Agants PCS* Alcohol Table salt Iron , DOT Strychnin* Nicotine TCOD Botuiinus toxin L^o 14,000 10.000 4,000 1.S00 100 2 1 0.001 0.00001 Expected Human Dote 1 Quart 1 Pinl-I Quart 1 Pint 1 Ounce-1 Pint 1 Teaspoon-1 Ounce 4 Drops 1 Drop Less Than 1 Drop Less Than 1 Drop 07V6T0 SNOW SLIDE I) ANIMAL. TOXICITY STUDIES Systemic Liver Skin Gel. Blood Hormonal Raproductlvo/Fotal Roforptions Litter Slio Survival Teratogen Karnofsky*. Law Mutagen Cancer Arne* Test Dominant Lathal Chromoaomal Studies Mice Rats T7V6T0 SNOW SLIDE U SYMPTOMS OF TOXICITY IN HUMANS o Chleracne, Pigmentation o Ocular Discharge o Liver Enlargement o Irregular Menstruation e Growth Retardation, In Children o Elevated Trlglycerides SLIDE IS TTt6T0 SNOM POSSIBLE EXPOSURES TO PCBa OSHA 0.1 ug/3 1 ug FOA Fish 2 pp 200-000 ug *MOWS O l9 23 SLIDE M COMPARISON OF PCS TOXICITY IN HUMANS TO COMMON COMPOUNDS Agants Dom (graaii) Predicted Huean Response PCBs Phanylaphrlna (NwiyiMphrln*. Isophrin Congasplrln, Dristan) Dxtroiithorphan (Cough Mediclni`DM Robitussln-DM) D iphenhydramine (Benadryl) Chlorpheniramine (Chlor-Trlmeton, Dristan) 2.0 1.0 0.5-1.0 0.4-1.0 1.0-2.0 "Yuaho" Oeath Death Death Death HONS 0 1 9 4 2 4 SLIDE 17 Agont Vltaain A PCB. PCB* Blood Lavals 3-20 ppm 1.1 ppa > 1.0 ppa Huaian Response Toxic Nona Non* HONS 0 1 9 4 2 5 SLIDE II Compound TLV WQC Chronic Study Cancer Aldrin DDT Nitro*amine Vinyl chloride PCB 15# ug/* 1.100 uglm3 - 10,000 ug/*1 0*1 ug/iR3 0.70 ng/l O.H ng/l 1.0 ng/l 20,000 ng/l 0.79 ng/l 10 ppm 1--10 ppm 10 ppm SO ppm 100-1000 pp y* yes yet yes yts/no HONS 0 1 9 4 2 6 SLIDE It 1 REFERENCES 1. IARC Monographs on the Evaluation of Carcinogenic RUE of Chewleels to Nan, 'Polychlorinated Biphenyls," Volute 13, pp. 4l-l03, 1978, 2. Occupational Exposure to Polychlorinated Biphenyl; (PCBs), OHEV (NIOSH), Publication 5i. 7TJ25, 1477/ *--^--L 3. Aablent Water Oual1t Criteria for Polychlorinated Biphenyls. US -------- EPA PM1-II779S, 1980. ------- ---------------------E--i-i* 4. Polychlorinated Biphenyls: A Report on Uses. Environmental and Health Effects and Disposal. Public Relations Department. Monsanto Co. 5. Handbook of Poisoning. R.H. Orelsback, Lange Books, 9th edition. 6. ToaIcology: The Basic Science of Poisons. 2nd edition, Editors Ooull, Klaassen. and Mur, MacMillan Publishing Co., New York. New York, 1980. 7. PC8 Poisoning and Pollution. Editor K. Hlguchl, Academic Press, New York, New York, 1?76. MONS 019427 General Comments These documents are tom*what unobjectively vritten with unscientific dltcutsiont tuch as: On page 3-23 th* author states: 'However, the overzealous ust of matheaatleal models and the assumptions made by the EPA without comparing these results to reality are easily demonstrated.* . The report by th* New England Gas Association begins 'The discovery of small amounts of PCBs, th* us* of which has been banned by th* EPA, in natural gas transmission lines is neither unexpected nor cause for public alarm.' Later in discussing acute PCB toxicity th* report states: 'Short term effects of moderate doses in. animals, are minimal. The toxicity is comparable to that expected from such substances considered relatively safe such as table salt, alcohol or caffeine.' In discussing chronic effects th* report states: '...PCBs produce minimal changes in new borns. In fact they are no more toxic to th* fetus than most chemicals, and less toxic with regard to birth defects than some common vitamins and hormones.* Summary of the Health Effects of PCBs The 'Summary of th* Health Effects of PCBs' tends to misguide the reader as to what the actual articles present. Th* overall tons of th* summary is a refutation of the technical presentation or an exaggeration of th* 'negative* arguments presented in th* six review articles. On pp. 1, 2-1, and 2-2 of th* Summary, the authors cite and discuss Nelsburgsr's and William's concept of promoters and Initiators, and epigenetic and gsnotoxic mechanisms. However, they do net point out that th* concept and the classification of carcinogens by Welsburgsr and Willalm are not shared by many- On pp. 1, 2, and 1-5 the authors discuss the large number of negative carcinogenicity studies for PCBs and present the inadequacies of th* positive studies. They do not mention the inadequacies of the negative studies. For example, the reference papers show that only a few of the negative studies are of sufficient duration to adequately characterize them as negative studies. Most of th* studies quoted in the six articles are of one ysar or less duration (e.q., see article by Crump, p. 24). For the positive studies, they criticize that the length of exposure was too short for an adequate study design (p. 2-3). This is not a relevant concern for a positive study. HONS 019428 1 Th* authors also stated that the positive studies should be discounted Because of inadequate design. If this is the case, the negative studies should also be discounted as well, it la far more difficult to prove a negative result than a positive one. Often, observations cited in the review papers are neglected or omitted in the CMA Summary, for example, p, 3, paragraph 2, sentence 3 of Summary> "The preponderance of studies has not identified a clinical disease associated with exposure to PCS, nor has it provided persuasive evidence of health impairment* competed to Paper No. 1, p. 22, paragraph 1, last sentence: *...a single study which suggests that diastolic blood pressure may be related to blood levels of PCBs.* This study suggests some cardiovascular toxicity. In addition, dermatologic reactions is known to be associated with PCBs poisoning. The Summary contains too much emphasis on the possible carcinogenicity of PCBs and does not mention much on the possible chronic health effects. The six articles have more details ofthe chronic effects. One of the controversial issues in this document is that the CHA contractor feels that PCBs are tumor promoters and not initiators. Since promoters may have threshold end reversible effects, the authors feel that low level exposures do not represent a haiard for cancer. The evidence for this assertion is presented on page 2-7 through 2-S and again on pages 3-17 through 3-19. These arguments can be summarized as follows. 1) PCBs increase the frequency of liver tumors in rodents: a tumor type which occurs spontaneously in these animals. 2) PCB Induced tumors are not metastatic. 3) PCB tumors are not transplantable. 4) Promoters are not mutagens. Many of the mutagenicity test on PCBs were negative. - 5) Continued presence of PCBs is required to sustain liver neoplasia and removal may cause tumor regression. There are several problems with this line of reasoning. First the mechanisms of action of Initiators and promoters are becoming less clear. The dogma that Initiators causa mutations and promotors have "SDigenetic* effects in cancer expression is no longer totally accepted. Cairns has recently proposed that initiators may not act through the induction of mutations by way of ttih damage as is usually thought. Further, the tern ssinanitic is not very useful; for example, what if a chemical caused a heritable change in chromatin structure by altering the action of DIM msthylasss. This affects the D'.M and therefore Is 2 HONS 019429 genetic," and yet ntither the sequence of bases nor the number of chromosomes is affaeted. Because of such difflcultlts It is bttttr, perhaps, not to stsrsotyps chemicals. Evtn If ws accept the contractors reasoning, it Is important to note that not all mutagenicity tests on PCBs were negative. In the Monsanto document entitled *A review and evaluation of carcinogenicity In mice and rats and mutagenicity studies with PCB' several positive mutagenicity tssts were cited along with the negetlvt results: Aroelor 1221 and 4-chloroblpbenyl were positive in tha Ames test. TCB and the 3, 4 epoxide of TCB induced single strand breaks in DMA of 0-929 cells. Kanechlor 300 was found to Induce chromosomal aberrations in vitro while Kanechlor 500 was positive in vivo. Continuous exposure of C3H 10T Vi cells to Aroelor 12S4 caused cell transformation to type 111 foci. These cells formed a sarcoma when Implanted into irradiated mice. Aroelor 1250 and 2,4,5,2.* ,4', 5' hexaehloroblphenyl also caused morphological transconation of C3H 10T Vi cells. 4-Chlorobiphenyl caused an Increase in unscheduled SNA synthesis in CHO cells while Aroelor 1254 induced chromosome damage in ring doves fed a 10 ppm diet. Clearly, based on these data one can argue that PCBs may have initiator activity. Finally, in calculating the risk estimate for Increased human cancer cases caused by exposure to PCBs, the contractor made (but did not state) an enormous assumption. It was assumed that because hepatomas were induced in rodents, this was the only tumor type for which humans were at risk. Although the correlation between carcinogenicity in animals and nan is good, the issue of site concordance is not clear. Therefore, humans may be at risk for other types Of tumors as well. Since the incidence of hepatomas in the human population is low, the risk estimate produces a very low number of PCB induced cancer. However, if these chemicals were also associated with lung carcinoma, for exas^le, risk estimates would be much higher. The latter part of the Summary consists mostly of a complete refutation of the risk assessment done by Crump and Hastaman, the last document on the 'Assessment of Carcinogenic Risks from PCBs in Food,* with very little documentation of the reasons for not accepting Dr. Crump's assessment. In his assessment, certain models are used to derive virtually safe doses of PCBs at specified risks. Be Includes the probit model, the one-hit model, and the multi-stage model. He also uses the one-hit and multi-hit models to derive estimates of lift time extra risks to humans for certain subpopulations at risk. Although the 'Summary* is interesting reading, in most cases it is not of sufficient scientific validity to warrant a response. In any case, the Summary conclusions for risk assessment are not supported by the technical review documents. In fact, unlike the other section such as epidemiology, the risk assessment section 3 HONS 019430 doss not even attempt to summarize the material In the risk assessment technical document. There are several specific problems with ths Summery. 1) Ssvsral places In the Summary (pages 4-5 and 1-3) the authors conclude that there is little reproductive risk from PCBs in animals. The Crump and Hastsrman document contains the data from the reproductive studies in monkeys which showed effects at levels as low as 2.5 ppm. The PCS data is unusual in showing reproductive effects at levels lower than those at which carcinogenic effects are seen. The fact that this is not even mentioned in the summary document, even though one of their own technical review documents Include the data, casts some suspicion in the summary documant. 2) Aside from the biological arguatents regarding mechanism of cancer which are used to refute Crump's -risk assessment, there Is another major argument proposed. It is that if the risks are as large as Crump's estimates indicate, these risks would have been detected by now in high risk populations. Ths review of the epidemiology studies show the highest number of cohort deaths studied for sufficient latency periods is less than 100. It is doubtful that relative risks of less than 15-20 times would be detected with this size population (we are .presently doing the calculation). 3) Using the EPA Hater Criteria documents, the authors in the summary ssetion calculated the expected Increase over background cancer for certain subpopulatlont. There is a mistake somewhere in the conversion figure given in Table 3-5 eguating 75 ng/1 to SO og/day. The 80 ug/day does not produce the expected increases over background shown in Table 3-6. A conversion to 80 ng/day rather than 80 ug/day would produce the estimates shown in Table 3-6 and on pages 4-10 the figure is given as 80 ng/day. 4) On pages 4-6, the authors use the CAG documents which show that if PCBs are not carcinogenic, the allowable daily Intake would be 210 ug of PCBs per day. This is calculated using reproductive data in minks (which was similar to reproductive data in monkeys) and a safety factor of 100. This allowablr dally Intake of 210 ug/day is not disputed by the authors, yet ths exposures shown in the table on page 4-11 for capacitor workers is 4 times higher than this allowable daily intaka. Therefore, even without considering carcinogenicity, the allowable dally Intake of 210 ug/day, which is used by the authors to question the much lower safe levels from cancer data, is already 4 times lower than average exposure of capacitor workers. All the arguments used in the Sunmary for negating certain models used for extrapolation of cancer risks by mechanistic arguments would have no relevance to reproductive data. 4 HONS 019431 S) Th authors of the summary document makes no attempt to estimate any risks according to their proposed nodal of cancar. This nay ba bacauaa tha probit nodal, also usad by Crunp, gives tha lowest risks of alnost any nodal availabla at low lavals. Tha safa lavals shown on page 34 of Crump's raport using this modal ara balov tha exposure of avaraga capacitor workars, avan for risks as high as 10"s. Tha Eoldamlology of PCBs Tha Epidaniology of PCBs by Gaffay was raviavad in graat datall. Our eonnants ara diractad to answar tha quastlon of whathar his conclusions and gsnarallsations ara justiflsd basad on our eursary rsvisw of tha cltad studies. Ha chosa to begin with tha review of tha carcinogenicity section since this is of nost concern. Carcinogenicityi In general, Gaffay's conclusions concerning the strength of epidemiologic evidence necessary to support a causal association between PCB axposura and carcinogenicity ara reasonable in light of tha criteria established in Richard Doll's paper. However, several of the studies reviewed by Gaffay (Sartassl st al, lack and Musch) suggest an overall increase in risk from death due to cancer and all studies reviewed by Gaffay do in fact reveal excesses for specific sites; although most of these excesses were not statistically significant or site concordant with one another. It should ba stated at the outset, that the presence of consistency strongly supports causality, but its absence does not rule it out. Gaffey deliberately omitted-tha preliminary mortality : findings of the Yusho incident from this section because this study was not an epidemiologic analysis directed at cancar. Uraba at al reported 11 deaths from malignant neoplasms among IS confirmed deaths of Yusho victims. Tha authors state that 'this rata I11/3S) is substantially higher than the 21.lt that Is the mortality rate from neoplasms in the same prefecture" for that year (1977) but the authors also state that 'it would ba prematura to conclude that this high mortality bears association with PC* poisoning.* Nevertheless, thraa of thasa deaths wars dua - to Hear cancar and two from lung cancar, which suggests some concordance with reports from previous studies (lack and Musch, Brown and Jones). Several of the studies reviewed by Gaffay (Bahn et al, lack and Musch) ara baaed on so few deaths, especially among those workars with tan or more years of exposure or IS years since first axposura, that it would ba prematura to conclude that PCBs ara definitely non-earcinogenic. For instance, in tha one study (Drown and Jonas) with tha greatest number of parson-years observed, there wars 143 deaths and it is not dear how many of those deaths wars among workers with five or more years of axposura or IS years latency since first axposura to PCBs. 5 HONS 019432 Furthermore. th power of detecting a significant elevated risk hu not bean reported for any of theee studies and these calculations should be performed as part of our in-depth review. In addition, it should be noted that most of ths studies reviewed by Gaffey art mortality studies, and not incidence studies, and are therefore based on death certificate information which is known to have certain limitations for detecting cancer risks due to chemical exposurs. The principal limitation being that analyses based on deaths alone may exclude those workers who left employment from a study plant and later developed cancer, but have not expired and are therefore not counted among the cancer deaths. With the exception of the Yusho study (Drabs et al), no reliable data on past Industrial exposure were reported. Two studies (Brown and Jones; Bertazxl et al) -reported relatively low airborne levels of PCBs as estimates of current exposure. Only Bahn et al reported that their study population were 'believed to have been heavily exposed* to PCBs. In the absence of high body burden levels or clinical Indications of ehloracne, we would have to assume that occupational exposure to PCBs was relatively low, and,therefore, would not be expected to produce a measurable differential effect on mortality. in summary, based on the preliminary results reported in the literature, Gaffey's conclusions with respect to cancer risk seem reasonable, but he unjustifiably dismisses the significance of suggestive evidence. Bertaxsi et al, for instance, recognizing the limitations of their analysis based on a few number of deaths (27), state that 'the results so far obtained are strongly suggestive of an excess of mortality from cancer among male workers and an excess of mortality from all causes and from cancer among female workers.* Several of the studies reviewed by Gaffey (Brown and Jones, Bertazzl et al, Zack and Musch) determined the vital status of their cohorts as of 1977 and 197S. it is possible that formal requests to these investigators to follow-up their investigations through 19B1 may resolve ambiguities that are due to insufficient numbers of deaths. Pinslly, Gaffey reports on the private communication of preliminary results from two Investigators (Rousch and Sinclair) whose results should be secured as soon as possible so that we can evaluate their relsvanca to Gaffey's conclusions. It is our opinion that further rigorous evaluation by more qualified ; experts is inherently desirable and will add to our understanding of the significance and deficiency of these studies, but we doubt that there is sufficient epidemiological evidence at this time to support a causal association between PCS exposure and cancer. Additional studies among more highly exposed workers with sufficient numbers of deaths should be supported in order to resolve reasonable concerns about the potential carcinogenicity of PCBs. 6 HONS 019*33 Accidental Heavy Exposure! Gaffey rtvisvi tv* incidents in this section (Meigs et ai, Yusho reports), but Meigs is more appropriately reviewed in the next section. The numerous studies which have been published in conjunction with the Tusho incident represent a sizeable fraction of the total health effect literature concerning KBs. Gaffey, and the authors of the General Electric health effect report, attempt to minimize the relevance of these findings for two main reasonsi (1) samples of the Kamenl Rice Oil were found to be contaminated with dibenzofurans, and (2) victims of this incident Ingested large doses of KBs which are irrelevant to assessing the health effects from lower occupational exposures found in the United States. Proper evaluation and Interpretation of these objections are crucial to the assessment of health effects related to KB exposure. Yusho disease is a Japanese term which refers specifically to acute and sub-acute poisoning due to ingestion of Kamenl Rice Oil which was contaminated with PCBs. There seems to be a general consensus that the average amount of Kanechlor ingested was ' estimated to be about 2 gm while the minimum amount ingested by a patient was about 0.5 gm. As of 1977, there were 1,SS victims of Yusho, who exhibited a variety of acute' symptoms ranging from severe ehloracne and hyperpigmentation to less specific effects such as vomiting and diarrhea. The most notable finding from this incident seems to be that patients seen three and six years after first coming down with Yusho disease still exhibited symptoms of Yusho, biochemical abnormalities and persistent blood levels of PCBs. Unless these effects can all reasonably be attributable to the dlbenzofuran contamination, then I would say that the severity, incidence and persistence of these effects all warrant serious concern for FCB exposure. Both the specific toxic effects (as determined from animal and human studies) of KDFs and KBs must be compared, as well as the quantitative evidence supporting the extent of PCDF contamination. tilth respect to the preliminary mortality reports, Gaffey justifiably calls attention to the problem that the populationat-risk, or denominator of the relative risk rates, cannot be determined, and therefore presents a problem tor determining the actual 'attack rate' as well as the relevance of elevated cancer deaths among Yusho victims. This problem is mitigated somewhat by the finding that among a group of 14( known users of oil, BO consumed leas that 720 ml, and 88 percent of these users were affected. Among those who used more than 720 ml, 100 percent were affected. This would suggest that Yusho victims, as defined by various clinical criteria, do in fact represent the total number of people exposed to contaminated oil. This, of course, is conjecture and is subject to much criticism. But until this problim is resolved, epidemiologic analyses utilyzlng the cohort approach may be inappropriate and unacceptable. HONS 01943* 7 Gaffey only briefly reviews th* reproductive effects mn among children of Yusho victims and he neglects entirely tc rsport on nor* recent findings. Th* EPA Hater Quality CritSr: document reports that 'thirteen women, 11 with Yusho and 3 without, but married to men with Yusho, delivered 10 live aid stillborn infanta beween February IS and December 31, lets.* Nine of the 10 had hyperpigmentation, and S had grayish-dark pigmentation of the gingiva and nails. Other findings of redi growth, prematurely erupted teeth and increased eye discharge were also reported. Nine more infants with hyperplgmentatlon were born to Yusho mothers between I960 and 1972 (Yoshiaura 1974). The Hater Quality Criteria document reports on several studies which found elevated levels of PCBs in breast milk of Yusho women. Again, the significance of these finding* depend upon our interpretation of the relevance of PCDP contamination as well as on the potential for lactating mothers with high be burden levels to deliver PCBs to nursed Infants. Environmental Levels and Body Burdens; Gatfey's conclosio seem reasonable and justified. Re concludes that high exposur to PCBs results in higher body burdens, all other things being egual. There ware a few inconsistencies which can bs adequate explained by more rigorous interpretation of the studies which will have to be addressed in a mors in-depth review. Specifically, what is the significance of high burden levels? there a threshold level for adverse health effects? Mhat is t significance of higher chlorinated compounds and what is the potential for future bloaceumulation under current exposure situations? Epidemiologic Studies of PCBs and Healtht Dermatologic effects: Gaffey justifiably concludes that 'the data suggest strongly that when PCB blood levels exceed about 150-200 ppb, ehloracn* can occur,* but most studies hsve shown that the occurence of ehloracn* is not further associated with blood pc levels.* Gaffey suggests that this apparent inconsistency may a function of personal idiosyncratic factors or that not all cases of high blood PCB levels are associated with ehloracn* because they do not have sufficient skin contact for ehloracn* dsvelop. In any event, dermatologic reactions have been consistently associated with PCB exposure and these reactions < occur under occupational exposure conditions as low as 0.1 mg/r Liver function: It is difficult to make any lnt*rpr*tat,lor of these findings without reviewing each of th* nine studied, general, there does not seem to be consistent findings of abnormal liver function or biochemical indicators of sneyrne activity; nevertheless, several studies did report abnormaliti* in enzyme levels which may have relevance beyond that suggested by Gaffey. Fat metabolism: Gaffey's conclusions as stated are consistent with the reported finding*. HONS 019*35 8 `Host studies, including on* non-oeeupatlonal study have associated lncr*M*d trlpycerides with PCB exposure. Th* data on chol*ttol act not consistent; an inertam3, a decrease and no chang* wtr* found (on* study *ach). HDL chol*stcrol either dtcrtastd or was unchanged (on* study each). Even if PCB exposure has son* effect on fat metabolism, it appears to be without any apparent clinicsl significance.* ' Symptoms. Illness and Other Conditions; Xll of the studies reviewed in this section should be more thoroughly evaluated. Gaffey quotes Smith it al. as saying `that no studies to date have shown that occupational exposure to PCBs is associated with any adverse health outcome, to be distinguished from demonstrable subclinlcal biochemical alterations.* It is possible that these biochemical abnormalities have not yet revealed any clinical manifestations, and it is not clear whether the subclinlcal alterations present serious health effect concerns. In general, most of these studies are cross-sectional, and therefore make no effort to ascertain th* true populatlon-at-- risk. In other words, the investigators did not identify their Study populations from historical employment records - thereby ensuring that all workers had some minimal- duration of exposure and furthermore, these studies mads no effort to identify workers who may have left employmment for occupational health reasons. In most cases, th* study population represented those workers currently employed and healthy. 9 HONS 019436 AdvtOt R*productive Outcow Dlicustion (Ntaltft EfEtctt o! Ei*cttic*l <5rd Kli) I. General problems with document A. It 1* impossible to discuss KB rtlatad problems In iaolation. Various amounts of PCDP art always found with PCB. B. Many isomers of KB art found In any mixture of PCBs, aoat of which ara extremely toxic. C. The fata of KB in nature baa id* bioaccuaulatlon la not clear, ror exaaple, photooxidation of PCB to PCDP do** occur (Hutslnaer 1974), but the probability of suph occurrence In nature la not known. D. Heating of KBa aa would occur '1th KB ua* In heat exchange rs results In lncreaaad converaion to PCOF. it la not clear how they addreaa thla laau* and lta inherent rlak. Row hot do capacitor* and 'closed* systems get In their life tlae. What converaion occur*? t. Given the Halt* of our atteapta to allainat* the reaalnlng PCS* in the environment, it* tranaport through the food chain, uncertain chealcal tranaforaatlona, and worriaoa* peralatence In the anlaal world a* the type of chlorination change* and the degree increaaea. It appear* axtraaaly unwla* to change any poaltion without auch further work at the health and baalc chealcal level. II. Peflclenclaa in conalderatlon of reproductive effect* A. Increaaed eatrogen effect aay be aeen via two different aechaniaaa. 1. Eatrogen like activity which would add to or compete with endogenoua eatrogen* (uterotrophle activity in rata, Gellett 1971). 2. lncreaaad metabolism of endogenous estrogens through PCB effect on hepatie HfO ayatea. The** types of effects could easily account for aenatrual irregularities seen as well as iapair the ability to aalntaln the fetus. B. Anlaal studies Ringer 1972, Linder 1974, Villsneuna 1971 show pronounced fertility problems depending upon KB type and anlaal species at less than 50 ppm PCB. 10 HONS 019437 C. Effects on humans 1. Yusho cooking oil ineidant showing low birth weight/ developmental delays ia of concarn. Raproductlva affacta are aaan in rhaaus aonkava at 2.5-5.0 ppm by purlfiad PCBa ad in tha sink, a sanaitiva animal apaclaa. Thus, although PCDF ia admittedly dangaroua PCB la not *obvioualy* aafa. 2. Raproductlva affacta ara balng investigated in a group of woman who work with capacitor grade PCBa in an Edison facility. Incraasad rata of low birth weight Infants is ona outcoaa under investigation by NIOSH. This type of study as wall as ona addrassing Incidence of infanta with developmental delays are needed. D. Central health vs PCB 1. Civan tha problems of small numbers and our presently inadaquats ability to addrass raproductlva haalth questions of menstrual lrregularitlas, dacraasad fertility, developmental delays and minimal brain dysfunction, animal studies must be given perhaps additional weight. 2. Each PCB mixture contains as many as 70 different compounds and has unique characteristics which reprsssnt a separata risk independent of its later chamieal fate. Chemical transformations durlnq use and after disposal rapressnt yat another risk needing evaluation recognising the uniqueness of each environmental situation. 11 HOhS 019438