Document KJ10ZLrZz402egQkjwMEyM41x

7^/? - ~Vh1^4 C^/ Occupational Canc&itB-RSfEIVED the Nervous System auG 121992 Ellen Kessler, M.D., M.P.H.' UB. uKutk Uth. uHLM. and Paul W. Brandt-Rauf, Sc.D., M.D., Dr.P.H.f The occurrence of nervous systems cancers has in creased markedly over the past 40 years, particularly brain tumors among white males over the age of 55, which has raised the possibility of occupational contri butions to etiology In fact, there has accumulated enough evidence to suggest that a link may exist be tween the incidence of brain tumors and certain occu pations and chemical exposures. The first inkling of this came from animal studies in the early 1940s that dem onstrated the induction of brain tumors in rats by direct intracranial implantation of known carcinogens. Subse quent studies in rats demonstrated that gliomas could be induced by inhalation or ingestion of acrylonitrile, whereas neuroblastomas developed with vinyl chlo ride inhalation. These and similar compounds have been used extensively in the plastics and rubber indus tries. Later epidemiologic studies, particularly in rub ber workers, suggested that humans with certain indus trial exposures may indeed be at increased risk for developing brain tumors. These investigations prompted research in other industries and there are now studies raising concern of a potential excess risk for brain tumors in a number of work-related environments in cluding rubber manufacturing, polyvinyl chloride pro duction, oil refining and petrochemical production, and nuclear fuels and weapons fabrication. Unfor tunately, in many cases the evidence is conflicting and inconclusive, and further study is necessary, but epi demiologic analysis has been complicated by lack of complete exposure data, prolonged latency periods, and difficulty in obtaining and accounting for different histologic cell types.12 RUBBER MANUFACTURING Mancusos studies on synthetic rubber workers arose following his observation that the 1947 death certifi cates among white males between the ages of 25 and 64 in Summit County, Ohio (the location ofmost of the U.S tire manufacturing at the time) revealed an excess pro portional mortality for brain cancer.3 Mancuso went on to determine that men employed in tire assembly and tire curing had a particularly elevated risk for develop ing brain/CNS malignancies, although a lower than ex pected risk existed for the entire plant populations.3 Although potentially toxic chemicals including coal tar, carbon disulfide, benzene, and other solvents had been used in the tire manufacturing process, none of these compounds could be directly linked to CNS carcino genesis. Nevertheless, subsequent studies by Monson et al. supported Mancusos original observations.4 3 However, later investigations, including a large coop erative study that used a cohort of active and retired male workers between 1963 and 1973 from 48 different plants, foiled to support the earlier findings.6 Other cohort studies after 1964 were also unrevealing.7 8 It has been argued that the earlier studies were representa tive of exposures initially found in the rubber tire manufacturing process that have since been removed. 'Medical Director, Corporate Healthcare Services Inc, Springfield. Virginia 'Director. Occupational Medicine, Columbia University, New York, New York Reprint requests. Dr Brandt-Rauf. Occupational Medicine, Columbia University, 60 Haven Ave, B-l, New York. NY 10032 19B7 by Thieme Medical Publishers. Inc, 381 Park Avenue South, New York, NY 10016. All tights reserved. WARNING: TV'* mt*ri1 may b* eev#J OLI 3258 by e pvrio!-* ' - SEMINARS i.N OCCUPATIONAL MEDICINE Volume 2, Number 4, December 1987 POLYVINYL CHLORIDE PRODUCTION When it was first reported that workers exposed to high levels of vinyl chloride were at increased risk for developing angiosarcoma of the liver, an otherwise rare tumor, NIOSH and otner study groups intensified their investigations in order to identify other possible haz ards from exposure. Waxweiler et al. reported that among a cohort of workers exposed to vinyl chloride for greater than 15 years, a significant risk for the devel opment of brain cancer existed with three observed and 0.6 expected.9 Overall, a total of ten brain cancers were identified among vinyl chloride workers on the basis of death certificates, pathology reports, and/or tissue evaluation, and nine of the ten brain tumors were histologically identified as glioblastoma multiforme; no histologic diagnosis was available on the tenth.9 Al though glioblastoma multiforme is a brain tumor that occurs in a relatively high proportion of adult cases, this percentage is considerably higher than would be ex pected. Furthermore, a proportional mortality ratio of 4.2 was calculated from a study by Monson on deceased workers from two plants with vinyl chloride monomer exposure.10 At about the same time, Maltoni et al. were able to show that brain tumors could be induced in rats when exposed to vinyl chloride by inhalation at levels resembling the human exposure data.1112 In Britain, Fox and Collier observed two brain cancer deaths as compared to 5.7 expected for a cohort ofvinyl chloride exposed workers.13 However, of the cohort members that were considered to have a high expo sure, one brain cancer death was observed versus 0.4 expected. Although the number ofbrain tumor cases in all these studies is relatively small, the results are gener ally regarded to be indicative of an association between vinyl chloride exposure and the development of brain malignancy, in particular glioblastoma multiforme. OIL REFINING AND PETROCHEMICAL PRODUCTION There has been considerable effort by numerous study groups to determine if oil refining and pe trochemical production could be linked to brain can cer. Although the potential carcinogenicity of certain petroleum products is well established, the cancer risk among refinery workers is less clear-cut In the United States, the focus came about when representatives of Union Carbide and the Oil, Chemical and Atomic Workers Union together identified 18 cases of brain cancer among deceased and living former employees of the Union Carbide plant in Texas City, Texas.14 Epi demiologists from OSHA"NIOSH joined the effort with a cohort mortality study on workers employed at Union Carbide and consequently found a significantly in creased SMR of 2.06 for white male hourly workers at the same Union Carbide plant.13 In addition, Rewp al. from NIOSH found that residents of Brazoria Countv the site of Dow Chemical, had a twofold risk of dyjn ' from a brain tumor it, at any time, they had worked * the Dow petrochemical plant.16 However, Ralph Cook, the director of epidemiology at Dow, and his groUri failed to find an association between a specific job category or exposure with an elevated risk for develop, ing brain tumors among employees at Dow1' Subsequently, Thomas et al. from the National Can cer Institute calculated a proportional mortality ratio for brain tumors of 2.28 in deceased active and retired members of the Oil, Chemical and Atomic Workers Union among hourly workers employed at Mobil, Gulf and Texaco petroleum plants in the Beaumont-Pon Arthur area of Texas.18 However, a series of industrv based studies of the same three refineries found no increased risk.19-21 No association between brain tu mors and specific work categories was demonstrated in a follow-up case-control study by Thomas et al21 In 1980, J. William Lloyd of OSHA and Irving Selikoff of the Mount Sinai School of Medicine organized a conference through the offices of the New York Acad emy of Sciences in order to evaluate the rapidly grow, ing but conflicting evidence on brain tumors and expo sure to petrochemical production. Several studies were presented including four by epidemiologists from OSHA, NIOSH, and NCI and four by investigators with industry-backed support (Dow Chemical, Gulf Oil, Du Pont, and a group representing some of the United Kingdom oil refineries). The disparities were marked with the government studies invariably indicating a twofold increased risk and the industrially supported studies showing no elevation of risk. The industrybacked studies were criticized for including all em ployees such as office workers who were most likely not significantly exposed. By diluting the study popula tion, certain subgroups who actually had significant exposures and potential risks may not have been identi fied. On the other hand, some of the criticisms pertain ing to the government studies that were brought out included the use of PMR methodology which can inflate the calculated risk for cancer. Some of the problems faced by all the investigators included the relatively low number of identified cases, the varied degree of expo sures among employees, and the retrospective analysis that involved tracing employee records back through 40 or more years.23 Most recendy, a five-year follow-up of a previous mortality study on 1,207 men employed in a Canadian oil refinery again revealed an excess risk ofbrain malig nancies (SMR " 519) among men who died within 20 years of first exposure.24 The study, however, failed to identify any hazardous exposures in the workplace. Although conflicting results exist concerning the in creased risks for brain malignancies in the oil refining and petrochemical production process, there is enough OLI 3259 CANCERS OF THE NERV OU S SVSTEM--Kessler. Brandt Rauf evidence to suggest such a risk. The extremely large size of the work force together with the controversial studv results place a great demand on investigators to proceed with further research, OTHER INDUSTRIAL AND NONINDUSTRIAL STUDIES The results of some of the above studies have prompted investigations in other industries including the nuclear fuels and weapons fabrication industry. In one study, the mortality risk was found to be elevated (SMR = 2.40) and the incidence rate was found to be significantly elevated (SIR = 2 6") for brain tumors among male employees at 3 nuclear fuel fabrication plant. However, radiation, industrial, or previous em ployment exposures could not be correlated.3' Ura nium fuel is contained in a controlled area and the radiation levels are assumed to be low, however, the exposures to metal dust, fumes, and oil mists are not unlike a metal fabrication plant and have to be consid ered as well as possible sources of carcinogen expo sure in nuclear fuel fabrication. At the Rocks- Flats Nu clear Plant. VC'ilkenson et al noted an increased risk for brain tumors, however, when tumors of benign and unspecified classifications were excluded, the increased risk was no longer statistically significant.36 Other studies of plutonium processing workers failed to dem onstrate an excess brain cancer risk among employees 2However, because the nuclear industry is still relatively new, the latency- period may be insufficient to accurately examine the risks in these cohorts Follow-up investiga tions on already studied cohorts would certainly be worthwhile, particularly because studies of other groups nonoccupationally exposed to radiation for medical-diagnostic and therapeutic purposes support the concept of radiation-induced nervous system tu mors.3 Several other occupationally or environmentally ex posed groups have been targeted as potential risks for brain malignancies. Professional artists, pathologists, embalmers, and farmers are examples of the nonindustrially exposed groups with suspected brain tumor risks.38 Chemicals common to these groups include formaldehyde and organic solvents although definitive correlations with exposure have not been established. In addition, farmers are exposed to pesticides, and Italian studies have suggested that this may be contribu tory to the increasing incidence of gliomas in this occu pational group.39 Other potential associations that have been noted include: trauma and the development of brain tumors, particularly meningiomas: lead exposure and the development of astrocuic gliomas in children and animals; dietary consumption of nitrites and me ningiomas in women; exposure to farm animals and the development of childhood brain tumors (suggesting a possible role of animal viruses )3. and occupational or environmental exposure to power frequency magnetic fields and brain rumors-*011 One of the most serious problems with occupational studies of nervous system tumors in the past has been die failure to segregate out the various and diverse histologic t\pes owing to lack of adequate pathologic evidence and the small numbers involved. However, it is probably erroneous to consider all primary nervous system tumors as a single disease in terms of suscep tibility to induction by occupational exposures. The most recent studies have attempted to address this issue For example, a case-control stud)' of "8 astro cytoma patients showed an elevated risk for working at an airfield or living near a petrochemical plant, sewage treatment plant, or paper or saw mill.*2 A study based on the Swedish Cancer-Environment Registry exam ined 1,092 cases of meningioma and found statistically significant standardized incidence ratios between 4 ~ and 6 0 for glass, porcelain or ceramic w orkers of both sexes, and for male workers in headgear fabrication and book publishing Significant SIRs between l" and 3 0 were observed for men employed in health care, railroad and trolley construction, sheet and plate metal fabrication, moving equipment operations, and ele mentary schools.33 These findings were consistent with previous studies.383--3- In a parallel studs- of intra cranial gliomas, a significant twofold excess risk was also found for glass, porcelain, and ceramic workers38 This group is of interest because of potential exposure to arsenic, chromium, and lead, and hat makers are of interest because of potential exposure to mercury a well-known neurotoxin. Finally note should be made of the concern for the potential development of childhood brain tumors due to parental occupational exposures.39 It has been clearly demonstrated in animal models that carcinogen exposure in the mothers can cause brain tumors in the offspring. For example, the offspring of pregnant rats injected intravenously with ethyl nitrosourea during the latter half ofpregnancy have an increased incidence ofgliomas and neuromas; in fact, the nervous systems of fetal rats have a 50-times-greater sensitivity to the carci nogenic effects of ethyl nitrosourea compared to adults.80 The first indication of a possible effect in hu mans was provided by the study of Fabia and Thuy which raised the possibility that children of fathers in hydrocarbon-related occupations were at elevated risk for childhood cancer, a large fraction of which occur in the nervous system.31 Hemminki et al. found an associa tion between brain cancer in children under 15 years old and specific jobs held by the father, including ma chine repairmen (relative risk 4.4) and painters (rela tive risk = 2 6), once again raising the possibility of a link to hydrocarbon and other chemical exposures.-3 Other studies have supported an association with paren tal chemical exposure or work in particular industries OLI 3260 313 SEMINARS IN OCCUPATIONAL MEDICINE Volume 2, Number 4 December 1987 (meial and construction work) with the development of childhood tumors.'15-'' Design problems common to these studies have included confounders, histologic diagnosis and the \ agueness of job category. The inde pendent risk factors not addressed have included preor postnatal radiation exposure, cancer m siolings. viral diseases, and congenital defects. Although none of the findings warrant a firm conclusion, clearly, as is true with occupational associations for nervous system can cers in general, further investigation is warranted. REFERENCES 1 Kessler E, Brandt-Rjuf PW Occupational cancers of the brain and bone, Occup Sled 198". 2 155-163 2 Schoenherg B5 Nervous svstem. m Schottenfeld D, Fraumeni JF Jr teds) Cancer Epidemiology and Pretention Phila delphia. WB Saunders Co, 1982. pp 968-983 3. Mancuso T Epidemiologic 4ud\ of tumors of the central ner vous svstem in Ohio Ann VY Acad Sci 1982. 381 P-39 4 Monson RR.\jkjno KK Mortalitv among rubber workers. Am J Epidemiol 19*6. 103 284-296 5 Monson RR Fine LI Cancer mortality and morbidity among rubber workers I Nat Cancer Inst 19'8 61 1OP-10S3 6 Simons MJ. Andielkovich DA Spirti-, R. Flerman DR Brain and central nervous system cancer mortality in L S rubber work ers .Ann NA Acad Sci 1982, 381 1*6-159 ' McMichael Al. Spirtas R. Kupper LL .An epidemiologic study of mortality within a cohort of rubber workers, 196*-19'2 J Occup Med 19'* I64S8-464 8 Andielkoc ich D Taulbee J. Blum 5 Mortalitv of female workers in a rubber manufacturing plant J Occup Med 19--8; 20 -*09-1)3 9. Waxweiler RJ. Stringer W Wagoner JK. Jones J, Falk H, Carter C Neoplastic risk among workers exposed to vinvl chloride, Ann NT Acad Sci 19"6. 2"1 -tO-48 10 Monson RR. Peters JM, Johnson MN. Proportional mortality among \mcl chloride workers Lancet 19"*, 239"'-398 11. Maltom C Vinyl chloride carcinogenicity An experimental model for carcinogenesis studies, in Hiatt HH, Watson JD, Winsten JA (eds). Origins of Human Cancer Cold Spring Harbor, N V. Cold Spring Harbor Laboratory, 19T pp 119--1-*6 12 Maltom C. Cihbem A, Carretti D: Experimental contributions in identifying brain potential carcinogens in the petrochemical industry Ann NY Acad Sci 1982. 381,261-249 13. Fox A). Collier PF. Mortality experience of workers exposed to vinyl chloride monomer in the manufacture of polyvinyl chlo ride in Great Britain. Br J Ind Med 1977, 34:1-10 14. Alexander V, Leffingwell SS, Lloyd JW, et al. Brain cancer in petrochemical workers; A case senes report Am J Ind Med I960; 1115-123 15. 'Xfcxweiler RJ, Alexander V, Leffingwell SS, et al: Mortality from brain tumor and other causes in a cohort of petrochemical workers. J Nat Cancer Inst 1983; 70:75-81 16. Reeve GR, Bond GG. Uoyd JW, et al: An investigation of brain tumors among chemical plant employees using a samplebased cohort method. J Occup Med 1983; 25 387-393 17 Bond GG, Cook RR. Shellenberger gj, et al. Case-control study of brain tumor deaths among employees at a chemical manu facturing plant. .Ann NY Acad Sci 1982. 38173-82 18. Thomas TL, Wixw-eilcr g/, Crandall MS. et al: Brain cancer among OCAW members in three Texas oil refineries. Ann NY Acad Set 1982. 381 120-129 19 Wen CP Tsai SP, Gibson RL, A report on brain tumors from a retrospective cohort study of refinery workers. Ann NY Acad Sci 1982, 381 130-138 20 Divine 8J, Barron v, Kaplan SD Texaco Mortalitv y, Houston, Texas. Texaco Inc, 1983 udv 21. Morgan RW, Wong O An Epidemiologic Analysis of the M iry Experience of Mobil 0:1 Employees at the Beaumont t ^ Refinery Oakland. California, Environmental Health a, J'*' lions Inc. 1984 1':u 22. Thomas TL, Waxweiler RJ, Crandall MS, et al Cancer m(1 patterns b\ work categorv in three Texas oil refineries ih Med 1984, 6 3-16 23 Selikoff IJ, Hammond EC Brain tumors in the chemical rvi try Ann NY Acad Sci 1982, 381.1-362 'TMu' 24 Theriault G, Provencher S Mortality study of oil refn workers five-year follow-up. J Occup Med 198', 29 .v- " 25 Hadiimichael OC, Ostfeid AM, DAin DA Brubaker re \j, , iry and cancer incidence experience of employees in a mici fuels fabrication plant J Occup Med 1983; 25*8-6] ^ 26. Wilkenson GS, Voelz GL, Acquaveila JR, et al, Mortalitv Jm,, plutonium and other workers at a nuclear facilm in Epiden^n ogy Applied to Health Physics. Albuquerque, New ^ Health Phvsics Society, 1983. pp 328-337 l" 27, Gilbert ES, Marks S; An analysis of the mortality of workers nuclear facility. Radiat Res 19^9; 79-122-1*8 J 28 Thomas TL, Waxwetler iy. Brain tumors and occupan, ,,,ai rut factors. Scand J Work Em-iron Health 1986; 12 1-is 29 Musicco M. Filippini G. Bordo BM, et al Gliomas and -kvupa tional exposure to carcinogens- Case-control studs vn 1 eb deimol 1982, 116'82-'90 pi 30. Lin RS, Dishinger PC, Conde J, Farrell KP Occupational exp, sure to electromagnetic fields and the occurrence ,4 hrjln tumors. J Occup Med 1985, 2~* 13-419 31. Wertheimer N, Leeper E Magnetic field exposure related 1,. cancer subtypes. Ann NY Acad Set 198'. 502 *3-5* 32. Olm RG, Ahlbom A Undberg-Navier I. et al Oc,upan.,nj] factors associated with astrocytomas A case-control Muds y-, 1 Ind Med 1987 11615-625 33- McLaughlin JK, Thomas TL, Stone BJ. et al Occupant xial n-y, for meningiomas of the CNS in Sweden. J Occup Med I98* -9 66-68 34 Thomas TL, Fontham ETH. Norman SA et ai Occupai tonal mk factors for brain tumors. A case-referent death-cettihcaie anal ysis Scand J Work Environ Health 1986. 12 121-12" 35 Home GR, Fraser D, Lindsay J, et al. Cancer morealm |-S19 ) in relation to diesel fume and coal exposure m a o hon of retired railway workers. J Nat Cancer Inst 1983 '9 lot51019 36. Dubrow R, Wegman DH. Setting priorities for ocoupjiioru) cancer research and control- Synthesis of the results of scu,na tional disease surveillance studies. J Nat Cancer Inst 19*) r 1123-1U2 3' Walrath J, Rogot E, Murray J. Mortality Patterns knv ir 1 ' Veterans bv Occupation and Smoking Status MH Pun u jn.-n No. 85-2756 Washington. D C., DHHS, 1985 38. McLaughlin JK, Malker HSR, Bloc 9tJ. Occupational n-k- ,n Sweden for intracranial gjiomas, Cancer Res 194" IP Hf-fV. 39. Savitz Da Childhood cancer. Occup Med 1986 1 ii'-i24 40. SwenbetgjA Koestner A Wechsler W Denlmger rh Quuuhli live aspects of transplacental tumor induction with etm inur., sourea in rats. Cancer Res 1972; 2656-2660 41. Fabta J. Thuv TD: Occupation of father at time 2 Nnh < children dying of malignant disease. Br J Pres Sot sted ivt 28 98-100' 42. Hemminki K. Saloniemi I. Salonen T Childhood ,-aixef jini parental occupation in Finland. J Epidemiol c. mrr...mi' Health 1981. 3511-15 43- Peters JM, Preston-Manin S, Yu MC Bran lumors m , "'id.-vr and occupational exposure of parents. Science I4"i -1* 235-237 44. Preston-Manin S, Yu MC, Benton B, Henderson BE n r,,,r- ~ compounds and childhood brain tumors a case *onn.) j,- Cancer Res, 1982; 42 5240-5245 45. WilkinsJR HI. Koutras RA Parental occupation and Nun j> f in offspring Joint Conference on Occupational Heaim v1 can Academy of Occupational Medicine. 1983 314 OLI 3261