Document bOV0xzBp4VqozrEBmVE3VGzEZ

August 13, 1980 RECEIVED SEP 11 1SS0 R? N. Wheeler* Jr. REPORT ON MR. JOHN GRASSO Samuel S. Epstein, M.D. ucc 028163 1 . , n TABLE OF CONTENTS - I. HISTORY OF EXPOSURE TO VC/PVC II. early literature on the toxicology AND CARCINOGENICITY OF VC/PVC III. THE CARCINOGENICITY OFVC IV. CONCLUSIONS Page 2 8 1624 ucc 0281S4 X. THE HISTORY OF EXPOSURE TO VC/PVC The potential sources of Mr. Grasso's exposure to VC/PVC over the period from 1964 to 1976 are a complex of the occupational and residential. 1. Occupational* The Du Pont Chamber Works handled VC from 1964 to 1973, during which time Mr. Grasso was continuously employed there; he worked as a chemical process operator in the Ponsol Colors Area, which primarily handled textile dyes, from 1960 to 1974, and in the Monastral Colors Area, which primarily handled non textile dye applications, from 1974 to 1976. Du Pont used VC for two pro cesses in separate locations in the approximately 100 buildings comprising its 600 acre plant. a. The Freon Area. VC was used in the formulation of Freon-propellant mixtures in this area. The VC was obtained from Dow Chemical, in amounts rang ing from 27,000 pounds in 1964 to 600,000 pounds in 1973. The total annual VC emissions in 1973 were about 8,000 pounds, representing about 1.3%- losses, which apparently occurred largely during tank car unloadings. The Freon area was about 1,000 yards (0.7 miles or 1.1 km) from Mr. Grasso's work site. b. The Jackson Lab Area. VC was used here in amounts under 2,500 pounds from 1970 to 1972 for R S D on copolymers. This area was about 850 yards from Mr. Grasso's work site. Based on Du Pont data, emissions appear to have been minima c. Estimated VC Exposure. From 1970 to 1973, it is estimated that Mr. Grasso could have been exposed at his workplace to VC emitted from the Freon area at levels of 0.25 to 2.5 ppm for about 976 hours under favorable wind conditions, and up to 100 ppm for about 98 hours under stable atmospheric conditions (Peskin, 1930). Additionally, based on other modelling data, annual average workplace ex posures from B.F. Goodrich emissions have been estimated to be about 0.26 ppb (Dames s Moore, 1980). *Tt should be noted that there is no occupational or other history of exposure to arsenic or thorium oxide (thorotrast), the only known causes of angiosarcoma of the liver besides VC/PVC. 3 2 * Residential Hr. Grasso moved to Pedricktown in 1965. His home there was located about 6 miles from the Du Pont Chambers Works, and about 1.8 miles (3 km) south-east of the B. F. Goodrich plant which opened in 1970. He was thus potentially exposed to VC emissions from the Du Pont plant from 1965 to 1973, and*' to a much greater extent,to VC/FVC emissions from the B. F. Goodrich plant from 1970 until July 1976, when his angiosarcoma of the liver was diagnosed. Goodrich is a major PVC manufacturing facility, purchasing VC from other com parties including PPG, Diamond Shamrock, Air Products and. Gonnoco. The VC was shipped in railroad tank cars or tank car trucks, and unloaded directed into pressurized storage tanks. PVC production ranged from 22.5 million pounds in 1970, to 130 million pounds in 1973, peaking at 131 million pounds in 1974. Based on Goodrich data, total VcypVC emissions in 1973 were about 6.9 million pounds, 3.6 million pounds of which were VC; this is equivalent to 72,000 lbs/weekly, 12,000 lbs/daily ox 500 lbs/hourly. VC losses in 1973 were in the order of 4.5%. These emissions had three major origins; con tinuous point sources; continuous fugitive sources; and episodic point sources. Some 60 episodic releases of VC from the Goodrich plant are recorded from June, 1973 to December, 1976, many of which --involve large amounts of VC released to the atmosphere" (Dames & Moore, 1980). For instance, on October 6, 1973, about 13,000 lbs (6,083 kg) of VC were released in a 5 minute period; this is equivalent to 156,000 lbs/hour, some, 300-fold in excess of the plant's average hourly 500 lbs emissions in 1973. From 1970 to 1976, based on an 8 hour working day shift of 5 days/week, Mr,. Grasso spent about 14 hours daily during the week and 24 hours daily during the weekend, comprising about 120 hours weekly, in and around his home. Much of this time was spent out-of-doors, including periods when he exercised his hunting dogs and bicycled, particularly after his cardiac surgery in 1973, in UCC locations closer to the Goodrich plant. The failure of Goodrich to have monitored community atmospheric VC levels from 1970 onwards (particularly in view of the plant's known substantial VC losses,and also the substantial information available to Goodrich on the car cinogenicity of VC, especially prior to 1974), perforce necessitates reliance on modelling data in attempts to estimate Grasso',s residential exposure levels. The limitations in such data are, however, well recognized (Dames & Moore, 1980). "....during any specific short-term period signi ficant differences can be expected between the model estimates and actual ambient values. It was not possible to determine from the model evalua tion results whether the model produced values with any consistent bias (either high or low) as com pared to actual VC concentrations, This is parti cularly the case concerning any possible bias in the estimation of VC exposures of Mr. Grasso, be cause numerous differences in details of the calculations exist between the exposure estimates and the model evaluation estimates. These dif ferences include major changes in the source parameters, source-receptor relationships, and averaging times." There are two sets of estimates on Mr. Grasso's residential exposure to VC. The first is largely derived from modelling based on the limiting assump tion that the "suspected effects on health of VC are associated with long term exposures------, therefore the expected annual average concentration pattern was regarded to be of interest" (Dames Moore, 1980). On the basis of such \ data, annual average residential levels were estimated to range from 6.1 to 8.8 ppb, and annual average exposures from 4.4 to 6.2 ppb. Maximum 24 hours exposure from 1974 to 1975 was estimated as 231 ppb, with levels in excess of 10 ppb occurring only in less than 10% of the days in this period. While no estimates were made of residential exposure levels during episodic releases, monitoring on June 29, 1977 (after EPA emission controls were promulgated), detected perimeter levels of 3.36 ppm (Dames Moore, 1980). The possibility UCC 028167 rr.ust be considered that such perimeter levels may underestimate residential ex posures, particularly during episodic releases, due to possible "umbrella" effects. The size of these effects is likely to have reflected stack or release heights? for the "rass" building this was 77 feet (Dames & Moore, 1980). While there are also earlier perimeter monitoring data, these have not yet been made available.--"A limited amount of perimeter monitoring was done prior to 1976, but the quality of the data produced is unknown" (Dames Moore, 1980). Other modelling estimates, based more selectively on daily meteorological data, conclude that from 1970 to 1976, VC levels from 3.9 to 39 ppm occurred over 1680 hours under favorable wind conditions, and up to 100 ppm over 146 hours during stab! atmospheric conditions (Peskin, 1980). Residential exposures to Goodrich emissions include PVC particulates, in addi tion to VC gases. Based on Goodrich data submitted to EPA, PVC and VC emissions in 1973 ware 3.3 and 3.6 million lbs, respectively (Davidson, 1980). Goodrich acknow ledges VC levels up to 1% in unstripped PVC suspension resins manufactured prior to 1976; following stripping, levels up to 400, 2000, and 3000 ppm were recognized in suspension, dispersion and bulk resins, respectively. Significant .amounts of PVC .em from Redricktown were in the 0.1-30 u range, roost being less than 10' y (Goodrich letter to EPA, 9/3/74); of interest, are observations that cars parked overnight i the Grasso neighborhood were often covered with light white dust in the mornings.* 3. EPA Experience The above estimates of Grasso's exposure should be considered in relation to * ERA experience, particularly prior to October, 1976, when VC was regulated as a hazardous air pollutant with emissions restricted to 10 ppm. ,*T'r.e toxic effect of PVC particulates may reflect the elution of entrapped residua] monomer or desorption of adsorbed vc. In spite of a paucity of data, there is suggestive evidence of the carcinogenicity of PVC dust.,- Intrapleural injection of dust induced "sarcomas" of the liver in 5 of 9 rats dying within a subsequent 18 months (Wagner, 1930). In a case-control study in the B.F. Goodrich, Louisville K plant, exposure to PVC dust was shown to be statistically associated with excess 1; cancer mortality, specifically with large cell undifferentiated cancer (Waxwerler, UCC a. EPA Monitoring Data. In preliminary field studies, based on extensive samplings of VC/PVC plants in 1974, average community exposures were 17 ppb in 90% of samplings within 5 miles of plants (EPA, 1975). Based on 24 hour continuous monitoring, 2.5% of community samples had levels in excess of 0.5 ppm, and 1% had levels in excess of 1 ppm.A 10 minute average sample at 3 miles from a plant was 3.4 ppm. Less than 10% of grab samples had levels in excess of 1 ppm. The maxi mum VC level found was 33 ppm at a distance of 0.3 miles from a PVC plant. The grab sarpling data were considered to underestimate real values because of metho dological' problems (EPA, 1975). Of additional interest, is the monitoring of an elementary school in Saugus, California, located near a TVC manufacturing plant (Keysor Century Corporation), where VC levels as. high as 2.8 ppm were recorded (EPA 1977); the levels were found to be closely related to tank car unloadings at night. b. EPA Modelling Estimates. Based on an accidental venting of about 5,000 lb: in 10 minutes (such incidents have occurred about 20 times in most PVC plant in 1975), maximum hourly concentrations of 20 ppm at 0.25 miles were.derived; peak con centrations were considered to be about 5 to 10 times highet as VC pollutant clouds passed over residential areas (EPA, 1975). It was concluded that populations most effected by spills and non-spills resided within about 1.2 mile of plant, and that impacts of a spill "would be minimal" beyond 3 miles. 4. Comparison of Residential and Occupational Exposures In spite of the apparently wide divergencies in the two estimates of occupation and residential exposures, ranging from ppm hourly estimates (Peskin, 1980) to the ppb annual averages (Dames e Moore, 1980), both are in substantial agreement that th residential exposures were one order of magnitude greater than the occupational, even though the latter were more prolonged. Additionally, the residential exposures appear to have been punctuated by frequent and sometimes substantial episodic emissions. uee 023169 7 5. References Danes Moore: "Dispersion Modelling Analyses of VC Exposure to a Specific Individual from the PVC Production Plant Near Pedricktown, New Jersey," January 8, 1980. Davidson, B. D.: "Engineering Investigation Report. Plant Generation of VCM and PVCR into Ambient Air and Across the Factory Fence Line," Kay 22, 1980. EPA, "Scientific and Technical Assessment Report on VC and PVC" (Star) , EPA-600/6-75-004, June, 1975. EPA, "Survey of VC Levels in the Vicinity of Keysor-Century, Saugus, Califor nia." EFA-330/2-77-017, June, 1977. Peskin, R. L.: "Investigation Report. Atmospheric Dispersion of VCM and PVC Resin from Goodrich Plant towards the Grasso House", May 27, 1980 (and subsequent attachments). Kacner, C., NIEHS, NIOSH and OSHA, Conference to Re-evaluate Toxicity, of .VC, ... PVC ana Structural Analogues, ..Bethesda, MD,.1980. Waxweiler, R. J., "An Epidemiologic Investigation of Lung Cancer in a Multixenobiotic Occupational .Environment", PhD. Thesis, 1978. UCC 028170 XX. 8. n EARLY LITERATURE OM THT TOXICOLOGY AD CARCINOGENICITY Q? VC/PVC This section reviews data on the toxicology and carcinogenicity of VC/PVC prior to January 22, 1974, when 3.F. Goodrich first publicly admitted that there had been three deaths from angiosarcoma of the liver since 1971 in its Louisville, Ky. plant- Prior to the former data, there was no open recognition by the chemical industry of the carcinogenicity of VC. 1. Animal Studies The acute toxic effects of VC in experimental animals, particularly ro dents, have been recognized since 1930 (Patty et al., 1930). Several publica tions over the ensuing 20 years established that inhalation of VC in concen trations in excess of 50,000 ppm (5 percent) induced progressive narcosis. cardiac arrythmias, and pulmonary edema. Subsequently, chronic studies in % rodents demonstrated the induction of centrilobulax degeneration of the liver and interstitial and tubular renal changes at VC concentrations as low as 500 ppm, then the U.S. occupational standard (Torkelson et al., 1961). These findings were later confirmed and extended with the demonstration of angiofi-- brosis, cutaneous hyperkeratosis, and periosteal proliferation of bone at higher VC levels r et al Studies from 1966-1969 demonstrated the occurrence of cJu.6IliL libr _______ions in the lungs of guinea pigs dying following chronic exposure to VC over a period of three months (Prodan et al., 'i 1966, 1975). A series of studies from 1961 onwards, demonstrated the persorption of PVC particles, in a size range from 10-40 microns, from the gut of rodents and other animals; the particles were rapidly disseminated in the blood stream, with subsequent embolisation in liver and other org 1961 1972). Individual PVC particles were identified in tissue sections for prolonged subsequent periods. Transplacental passage of PVC particles was also demonstrated. UCC 028171 9 r In May, 1970, an Italian, toxicologist, Pier Luigi Viola (employed by Solvay Manufacturing Corporation), reported at an international cancer congress in Houston, Texas, that long-term intermittent exposure of a small group of rats to 30,000 ppa of VC in air resulted in the production of cancers in a wide range of organs (Viola, 1970; Viola et al,, 1971).. This first report on the carcinogenicity of VC created little impact. This wj because Viola had claimed that the type of. tumors induced were peculiar to rats and without human significance, and also because of the very high levels of VC to which animals were exposed. Actually, Viola had induced tumors- at much lower concentrations, extending down to less than 5,000 ppm. Ihes facts were discussed in detail at a meeting of the Manufacturing Chemists Associa tion (MCA), the major trade association of the U.S. Chemical Industry, in November 1971, when it was concluded that they should not be published, as this would otherwise "lead to serious problems with regard to the vinyl chlo ride monomer and resin industry...and force an industrial upheaval via new laws or strict interpretation of pollution and occupational health laws" / (Wheeler, 1971). Union Carbide further expressed concerns that in view of the large stake it had in "areas most likely to be affected, such as food, food packaging, fiber and aerosols (it) would be seriously hurt by arbitrary or panic-induced government restrictions." A consortium of European chemical industries, led by Italy's semi-govern mental Montedison, and with contributions from British, Belgian, and French firms, then financed a major study initiated in July, 1971, by Cesare Maltoni of the Bologna Cancer Institute (Maltoni Lefemine, 1975). An extensive * series of tests were conducted in which adult,infant, and pregnant mice, rats, and hamsters were exposed to VC in inhalation chambers at concentrations ranging from the relatively low 50 ppm level to the toxic'100,000 ppm level. By August, 1972, Maltoni had confirmed and extended Viola's results. . VC was lj 0 i- 02S172 shown to be a potent carcinogen, inducing the rare angiosarcomas of the liver, as veil as the more common hepatomas, and cancers of other organs, including kidney, brain, and lung. The cancers were subsequently found even at the low est level then tested, 50 ppm. The overall tumor incidence was concentrationdependent, with higher doses increasing the incidence and rate of development of the cancers (Maltoni, 1977). In October, 1972, the MCA entered into an agreement with the European consortium to share their information, but not to disclose it without prior consent (Edsall, 1975; Turshen, 1976), In January, 1973, the U.S. industry representatives visited Maltoni and were given full details- of hj_s experimen tal studies.. The MCA (and Maltoni) subsequently failed to disclose this information, in spite of the fact that NIOSH, in the same month, had publicly requested all available data on the toxic effects of VC. In March,1 1973, the MCA recommended to NIOSH a precautionary label for VC that made no reference to toxic or carcinogenic effects on animals, or humans. .* Four months later, however, representatives of the MCA met in secret with the Director of NIOSH and a small group of his senior staff who were informed that Maltoni had induced tumors in rats exposed to VC levels of 250 ppra. At one stage of the meeting, a Dow Chemical representative adjourned to a separate office for a private meeting with the Director of NIOSH, and subse quently commented that "this private discussion of the carcinogen problem was worth the whole effort" (VTheeler, 1973). Finally, industry concluded that as a result of a meeting "the chanc.es of precipitous action by NIOSH on VC was materially lessened. NIOSH did not appear to want to alienate a cooperative industry nor they did not want to know too much unpublished data. '* In Kerch, 1973, following complaints of an unpleasant taste, Schenley Distillers found 10-20 ppm levels of VC in their liquors sold in PVC bottles. On' the basis of these findings, and still knowing nothing about the carcino genicity data, FDA then banned the use of these liquor bottles. In December, t i"-' IJCC 028173 -sSaViSrt---v 1973, the Society of Plastics Industry, the plastics industry trade associa tion, provided the FDA with data on migration of VC from PVC containers to food, but made no reference to problems of carcinogenicity. According to a recent special conroittee report of the American Association for the Advancement of Science, the MCA "appears to have deliberately deceived NIOSH regarding the true facts -- Because of the suppression of these data, tens of thousands of workers were exposed without warning, for perhaps some two years, to toxic concentrations of vinyl chlorid (Edsall, 1975)7 2. Human Studies Chronic toxic effects of VC/PVC have been recognized in workers for almost three decades. Qironic hepatitis, gastritis, upper respiratory tract irritation, and skin lesions were noted in a group of 73 workers in a Russian plastic facto3jy''(Tribukh et_ al., 1949)^ Toxic angioneuropathy was subsequently noted in workers exposed Lo VC below the then Russian standard of 390 ppm (Filatova et_ al., 1957)y Raynauds-like acroosteolytic disease was also recognized in the same year in VC/PVC workers' (Kubota, 1957) n The best earliest description of the characteristic spectrum of non-neoplastic VC- induced disease was by a Rumanian group of workers, who recognized chronic hepatitis, splenomegaly, contact allergic dermatitis, pseudoscleroderma, Raynauds-like disease, and gastrointestinal and central nervous system disor ders (fSuciu^et jl.7~1963). \By 1966, acroosteolysis, often accompanied by Raynauds disease, was recognized in VC polymerization workers (tCordier et al,', 1966). These findings were confirmed in several subsequent investigations. Further studies over the period of 1962-1970 confirmed the occurrence of VCinduced occupational disease, and demonstrated a dose-response relationship over a range from 36 to 825 ppm, with a highly significant incidence of VC disease at concentrations as low as 40 ppm ((Suciu^et al. , 1972, 1975K^ Several reports have drawn attention to the^develormcntr'of chronic IJCC 028174 pulmonary disability in VC/PVC workers (Vertkin and Mamontov, 1970; Szende et al., 1970). However, a common factor characterizing these and later reports is chronic exposure to PVC dust. The pulmonary changes range from asymptoma.tic decrease in pulmonary function tests and radiological abnormalities to clinical bronchitis and emphysema, and to pneumoconiosis. At least eight fatal cases of primary carcinoma of the liver, subsequently diagnosed as angiosarcoma, have been recognized in VC/PVC workers prior to 1970 ({Heath et al., 1975)^ seven occurred in the USA, including three cases at a B.F. Goodrich plant in Louisville, Ky. (Heath et_al., 1975). Tne eighth case occurred in a Swedish worker in 196y^(Byren et al, 197gJ~^ Studies by Maltoni as early as 1969, but not disclosed until 1975, revealed a high incidence of malignant cells, of a large cell adenocarcinomatous typenot generally seen in smokers, in^the sputum of VC/PVC workers, suggestive of early lung cance^^lHaltoni, 1976K On January 22, 1974, B. F. Goodrich announced that since 1971 three PVC' workers in their Louisville, Kentucky, plant had died of angiosarcoma of the liver. On that same day, the MCA revealed the unpublished Maltoni data to NIOSH. Based on these data, OSHA announced a temporary emergency standard of 50 ppm on April, 1974; this was replaced by the present 1 ppm permanent stan dard on April 1, 1975. 3. Comments It is clear that a significant body of animal and human data had esta blished the carcinogenicity of VC prior to January, 1974, and at least as early as 1970. It is also clear that much of these data had not been published or otherwise released by the chemical industry prior to 1974. The control of'con tinuous fugitive and point source, and episodic emissions by Goodrich in 1970 (instead of waiting until so required by the 1976 EPA regulations), together with the development of community monitoring systems, could have effectively 13 curtailed Kr. Grasso's residential exposure to VC/PVC. In all probability, this would have substantially reduced the chances of his developing angio sarcoma of the liver- ucc 023176 14 4. References 3yren, D., et al. Ann N. Y. Acad. Sci. 246:249, 1975. Cordier, J. M., et al^. Cahier Medecine du Travail (Belgium) January 4, 1966. Idsall, J. J* Science 188:687, 1975. Filatova, V. S., et al. Gigiena Sanit. 1^.38, 1957. Heath, C. W., et al. Ann. N. Y. Acad. Sci. 246:231, 1975. Kubota, R. J. Sci. Labour (Japan) J3:l, 1957. Lester, D. L. Am. Ind. Hyg. Assoc. 24_:265, 1963. Maltoni, C. and Lefenune, G. Ann. N. Y. Acad. Sci. 2.46:195, 1975. Maltoni, C. Ann. N. Y. Acad. Sci. 271:431, 1976, Maltoni, C. Origins of Human Cancer, eds. Hiatt, A. et al., pp-. 119, 1977. Oster, G. Ann. N. Y. Acad. Sci. 246:149, 1975. Ratty, F. A., et al, Pub. Health Rep. V. 45_: 1963. Prodan, L., et al. Congr. Intern at. Med. Travail, Vienne, (5:328, 1966. Prodan, L., et al. Ann. N. Y. Acad. Sci. 246;159, 1975. Suciu, I,, et al. Med. Intemat. 15^:967, 1963. Suciu, I., et_ al_- Simpoz. Intemat. Ludwigshafen, jjp. 520-528, 1972. Sucxu, I., et al. Ann. N. Y. Acad. Sci. 24653, 1975. Szende, B., et al. Med. Lavoro. 1:433, 1970. Torkelson, T. R., et alAm. Ind. Hyg. Assoc. 22;354, 1961. Tribuhh, S. L., et al. Gigiena Sanit. 10:38, 1949. Turshen, M. Kealth/PAC Bulletin, 71:1, 1976. Vertkin, J. and Kamontow, J. R. Gig. Trud. Prof. Zabolevaniya 14_:29, 1970. Viola, P. L. Proc. X Intemat. Cancer congr., Houston, Texas, 1970. Viola, P. L., et al. Cancer Res. .31:516, 1971. Volkheiner, G- F., et al. Deut, Gesundheitsw. _16:1727, 1961. Volkheiner, G. F. Persorption. George Tnieme Verlag, Stuttgart, Germany, 1972. UCC Wheeler, R. N. Onion Carbide Internal Memorandum, MCA Report to NIOSH, November 23, 1971. Wheeler, R. N. Union Carbide Internal Memorandum, MCA-Vinyl Chloride Conference, July 19, 1973. Wilson, R. H, and McCormick, W, E. AKA Arch, Ind, Health 21:536, I960, ucc 023173 HI.. THE CAHCIKOGENICITy O? VC VC is a highly potent multi-system carcinogen which is known to induce a wide range of cancers in a wide range of experimental animals, and in humans. 1. Animal Studies VC induces angiosarcomas of the liver and other organs, and a wide range of other nalignant neoplasms in a wide range of experimental animals, rats, mice and hamsters, following chronic inhalation, oral or pea:enteral exposure. VC also induces hepatocellular carcinomas, carcinoma of the kidney, lung end 2ymbal gland, nephroblastomas, brain neuroblastomas, and miscellaneous other malignant neoplasms, including mammary adenocarcinomas, sebaceous gland carcinomas, lymphomas and osteochondromas^ (Viola et al. , LD7I; MaitonT c. Lefemine, 1975; Keplinger et al., 1975; Maltoni, 1977) ~lft extensive experiments from 1971 to 1978, the carcinogenic effects of VC were tested in rats, mice and hamsters by inhalation, over a dose range from 1 to 10,000 ppm administered intermittently over periods ranging from 5 weeks to 1 year, and also by other route^Jt^toni, 1977, 133Q) ^-inhalation by rats over a dose range from 1 to 2,500 ppm (for 4 hours daily, for 5 days weekly, for 1 year) induced carcinogenic effects down to the lowest doses tested, including breast cancers at 1-5 ppm and liver angiosarcomas, at iq pj>m. Angiosarcomas were also induced in rats and mice by 50 ppm VC administered intermittently for 17 and 30 weeks, respectively. Additionally, carcinogenic effects were induced in hamsters by 50 ppm doses over 30 weeks. In other experiments, angiosarcomas and other neoplasms were induced in r------------- nice at the lowest dose tested, 50 ppm, over a 36 week perxodf (Keplinger, V375)jJ and over a year ^(Kolnberg, 1976; Lee, 1977) v Carcinogenic effects have also been resorted in rats and nice exposed to 1 ppm levels for 32 weeks (Kaloyanova, 1976).) ucc 028179 17. In extensive experiments contracted by the Consumer Product Safety Commission, the carcinogenic effects of VC in mice were demonstrated after a single exposure to 5,000 ppm for 1 hour only^{CPSC, 1979);/borderline carcinogenic effects were, however, also noted following 1 hour's exposure to 500 ppm. It was concluded "that even single exposure of sufficient magnitude may be carcinogenic." The study warns that "a one-time exposure occurring, for example, as a result of an accidental spillage------could threaten------nearby residents." 2. Occupational Exposure During Manufacture of VC/FVC One of the earliest reported occupational epidemiological studies, based on 257 workers exposed to VC for at least 5 years with initial ex posure having occurred more than 10 years past, found a 2.3-fold excess in deaths from cancer of all sites combined; three of the cancer deaths were due to angiosarcoma of the liver/CNicholson et al. , 1974)/ in the same year, hepatic angiosarcomas and an excess of cancers of multiple organs, including mouth, pharynx, digestive system, lung and lymphomas, were reported among VC/FVC workers(^CTabershaw Gaffey, 1974K) a pro- portionate mortality analysis of 161 deceased workers in two plants producing and polymerizing VC" found an 11-fold excess risk^of^ldjzeae-aa^^ biliary tract cancers, and a 4-fold excess of brain cancers (Monson et al., 1974)j In subsequent studies, an excess of cancers of four organ systems, liver, brain, lung and lvmcho-hematopoietic systems, were found among VC polymeriza tion workers (Waxweiler et al., 197&)of fourteen histologically confirmed liver cancers, eleven were hepatic angiosarcomas, a sixteen-fold increased' risk of liver cancer mortality. A statistically significant excess of liver and pancreatic cancer deaths was found among 271 Swedish workers employed in a VC/PVC production plant UCC 023130 (Eyren et al. , 1976)^ this excess appeared v;ithin the first five years after first exposure to VC. In a study on cancer mortality among 7,021 males em ployed in VC production and polymerization, an excess of cancer mortality for multiple organs, liver (12 observed vs. 0.6 expected), stomach (18 observed vs. 9.6 expected), and the lymphatic and hematopoietic systems (15 observed vs. 5.2 expected) , was found in comparison with the West German male population^(Von Keinl et al. , 1977).y In a similar- study on 7,561 males who at some time between 1940 and 1974 were employed at one of four plants producing PVC, an .excess liver cancer mortality was shewn for each group of workers irrespective whether exposure to VC was "high, medium, or low" ^^lTyollier^r~1-57?T;7 although the study provided no evidence of excess mortality from cancers other than of the liver, it was emphaziec the period of follow-up was too short to- permit valid negative inferences. A KIOSH industrial hygiene survey, conducted in thirteen VC/PVC plants during 1974-1975, found that VC .exposures, as determined by personal sampling, ranged fron 0.01 to 84.77 ppm (n=119) for VC monomer plants, and from non-detectable to 245.0 ppm (n=449) for VC polymerization plants (Jones, 1978). 3. Occupational Exposure During PVC Fabrication Levels of exposure to VC are much lower in PVC fabricating plants than in plan manufacturing VC or polymerizing it to PVC. A 197S NIOSH industrial hygiene survey of PVC fabricating plants found that VC exposures ranged from the non-detectable tc. 3 2.44 ppm (Jones, 1578); PVC dust levels ranged from 1.7 to 13.3 mg/m . Two confirmed cases of hepatic angiosarcoma have been reported in Connecticut among workers employed in industries using PVC ^Christine et al_., 1974) one was a t,----------------------------------------------- 47 year old man who had worked for the previous ten years as an accountant in a factory producing vinyl sheets and processing PVC resins, and the second was a 61 year old man who had spent twenty-five years in an electrical plant operating a machine that applied PVC-containing plastic to wires. A review of fourteen cases* *0ther epidemiological studies have confirmed such' short latencies for excess risks of cancer following exposure to VC/PVC. UCC 023181 19 of hepatic angiosarcoma diagnosed in Great Britain during-1963-1973, found-one case who had worked in a plant which used PVC as a raw materia^lBaxter et al., 1977^) - .' A proportionate mortality study of 707 male PVC fabricators and other plastic workers, dying in Great Britain in 1970-1972, found an excess of sto- nach cancer mortality, 924 observed vs. 16.4 expected, p<0.05[ (Baxter s Fox, 1976?) A cross-sectional mortality study of 3,845 deaths during 1964-1973 among current and former white male employees of seventeen PVC fabrication facili ties found an excess of deaths from digestive tract cancer (210 observed vs. 162.373 expected, PMR129), from respiratory tract cancer (205 observed vs. 176.899 expected, PKS--116), and from'.lymphatic cancer. (42 observed vs. 32.435 expected, PMR=129) ^fdhiazze et .al.. 1977) .) For white females, there were excesses of deaths from digestive tract cancer .T53 observed vs. 34.929 expected, PMR-152) , from breast cancer (44 observed vs. 31.907 expected, EM_J^138) , and from urinary tract cancer (11 observed vs. 4.487 expected, PMR=245). 4. Cancers in Communities Wear VC/PVC Plants A wide range of studies haye demonstrated excess cancer risks among resi dents of communities located in proximity to VC/PVC manufacturing facilities. Following the findings of excess brain tumors in VC-exposed workers, a significant excess of these tumors, 38 observed vs. 24 expected, p<0.01, was found in several Ohio communities where PVC manufacturing facilities are lo cated^[Infante, 1976)o Similar findings have been reported in a Canadian com munity where a PVC manufacturing plant had operated for some 30 years (Iturra, 1976) . Two community cases of hepatic angiosarcoma have been reported in Connecticut in individuals with probable residential exposure to VC /(Christine et al . 1974)j^neither had any history of occupational exposure to VC or arsenic, or diagnostic exposure to thorium oxide (the only 3 agents known to induce ucc 028182 hepatic angiosarcoma). One case lived within two miles of a plant producing PVC coated wire, while the second lived within 0-5 miles of a plant producing vinyl sheet; both plants also had cases of hepatic angiosarcoma among their workers. Another case of liver angiosarcoma was reported in Great Britain in a resident who had lived for six years within half a mile of VC/PVC manufacturing On the basis of these observations, a study was undertaken in New York State of 26 confirmed cases of hepatic angiosarcoma et al., e controls, comprised of individuals who had an inter- nal malignant tumor other than primary liver cancer, were matched with index cases on the basis of age at diagnosis, race, sex, place of residence and vital status. A statistically significant association was found between angiosar coma of the liver and direct occupational or therapeutic exposure to arsenic (2 cases), vinyl chloride (3 cases), and thorium dioxide (2 cases). The study further demonstrated that of ten female cases of liver angiosarcoma, without any occupational or therapeutic exposure to VC arsenic or thorium dioxide, one lived within 1,700 feet (0.34 miles) of a VC polymerization plant, and four lived from 500 to 4,500 feet (0.85 miles) of a PVC fabrication plant. In con trast, none of their matched controls lived within one mile of any facility polymerizing VC or fabricating PVC. 5. Comments VC is a very potent carcinogen inducing angiosarcomas and a wide range of other malignant tumors in rodents and humans. In experimental animals, mali gnant tumors have been induced by 1 ppm doses administered intermittently over one year, by 50 ppm doses administered from 17 to 36 weeks, and by 5,000 ppm doses (with marginal effects at 500 ppm), following 1 hour exposure only. Angiosarcomas of the liver and other malignant tumors axe well recognized in workers in plants manufacturing VC and PVC, where levels'of VC exposure are l JCC 028183 21 relatively high. Additionally, an excess of these tumors is recognized among workers involved in PVC fabricating plants, where exposure levels are well under 10 ppm, and probably closer to 1 ppm.** Excess risks erf angiosarcoma of the liver and of brain tumors are also recognized among residents of communi ties located in proximity to VC/PVC manufacturing facilities. However, the chemical industry has not developed systematic monitoring data on ambient levels of VC in communities located in proximity to VC/PVC manufacturing facilities in relation to either continuous or episodic emissions. Over and above these consideration, as overwhelmingly recognized by the informed independent scientific community, there is no evidence for the existence of safe levels or thresholds for the irreversible processes involved in chemical or physical carcinogenesis. 3 * It may be noted that 1 ppm is approximately equivalent-to 2,600 ug/m ; assuming inhaltion by an adult of 1 m3/hri this is equivalent to a dose of 26,000 ug or 26 mg over a 10 hour period. ucc 028184 6. References Baxter, P. J. Fox, A. J., Lancet 2_: 27, 1975. Baxter, P. J. & Fox, A. J., Jancet 1_: 245, 1976. Brady, J. s. et al. J. Nat. Cancer Inst., 59: 1383, 1977. 3yren, D. et al. Env. Health Perspec., 17: 167, 1976. Chiazze, L. et al. J. Occup. Med., 19: 623, 1977. Christine, B. W. et al. Morbidity and Mortality Weekly Report, 23: 210, 1974. Consumer Product Safety Commission, Press Release, "CPSC Cancer Study Suggests New Potential Risks", April 26, 1979; Hehir, R. et al. Prepublication Release of .CPSC. Falk, H. Waxweiler, A. J., Proc. Roy. Soc. Med., 69: 303, 1976. Fiechtner, J. J. Reyes, C. N., J.A.M.A., 236: 1704, 1976. Fox, A. J. Collier, P. F., Brit. J. Ind. Med., 34_: 1, 1977. Hammond, E. C. Selikoff, I. J., Occup. Health Nursing, 22: 17, 1974. Infante, P. , Ann. N.Y. Acad. Sci., 271: 49, 1976. Iturra, H. Proceedings Toxic Substances in the Air Environment: Specialty Conference, p. 96, Air Pollution Control Association, Pittsburgh, PA, 1976. Jones, J. H., Draft NIOSH Report, Unpublished, 1978. Kaloyanova, F. (Bulgaria), Unpublished Draft, Communication to NIOSH, November 1976. Replinger, M. L. et al. Ann. N.Y. Acad. Sci., 246: 195, 1975. Lee, C. C. et al. J. Tox. Env. Health, 4_: 15, 1978. Maltoni, C., Origins of Human Cancer, Cold Spring Harbor Symposium, pp. 119, 1977. Maltoni, C., NIEhs, NIOSH and OSHA, Conference to Re-evaluate Toxicity of VC, PVC and Structural Analogues, Bethesda, Hd., 1980. Maltoni, C. Lefemine, G., Ann. N.Y. Acad. Sci., 246: 219, 1975. Monson, R. R. et al. Lancet 2: 397, 1974. ucc 028185 23 . Nicholson, W. J. et al. Ann. N.Y. Acad. Sci. 246; 225, 1976. Tahershaw, J. R. 6 Gaffey, W. R., J. Occup. Med., 16; 509, 1974. Viola, P. L. et al. Cancer Res., 31; 516, 1971. von Reinl, W. et al. Medichea, p. 2, September 1977. wagonerand Infante, P., VC/PVC: A Review of Carcinogenic and other Toxicologic effects. In Press, 1980. Waxveiler, R. J. et al. Ann. N.Y. Acad. Sci., 271; 40, 1976. ucc 028186 24 . IV. CONCLUSIONS Cn the basis of review of available documentation, there is a very substantial probability that Hr. Grasso's fatal angiosarcoma of the liver resulted from exposure to VC. Based on available documentation, there is also a substantial probability that the major site of his exposure was residential, originating from the Pedricktown facility of B.P. Goodrich, although a contributory role for occupational exposure at the Du Pont Ciamber Works appears likely. These conclusions reflect the followingconsiderations : 1. Hr. Grasso was exposed residentially to VC/PVC emissions from the Pedricktown plant, a major. FVC manufacturing facility, from 1970 to 1976. These included emissions from continuous point and fugitive sources, and episodic plant sources; 6 episodic emissions, some of which resulting in major VC losses, were recorded in the Pedricktown plant from 1973 to 1976. ' ' 2. Mr. Grasso was additionally exposed to lower levels of VC emissions in his workplace from 1964 to 1973, 3. VC is a potent carcinogen, inducing angiosarcoma of the liver and malignant neoplasms of other organs in experimental animals following prolonged intermittent exposure to 1 ppm levels, and following. 1 hour exposure to 500-5000 ppm. 5. VC has induced angiosarcoma of the liver and other malignant neoplasms in PVC fabrication workers exposed to VC levels, generally below 2.44 ppm. 6. VC has induced angiosarcomas of the liver in residents of communities located near major VC/PVC manufacturing facilities. UCC 028187 25* Finally, it should be noted that the industry failed, for several years, to openly recognize information available to them on the carcino genic hazards posed by VC, and to act on such information for the pro tection of surrounding communities and exposed workers. The institution of appropriate corrective action, in 1970, by industries including B.F. Goodrich and Dow Chemical would have substantially reduced Mr. Grasso's exposure to VC and his subsequent risks of developing angiosarcoma of the liver. UCC 028188