Document k3k2oMJ7DvZqMd7B34LpmvpE

DOW CHEMICAL U.S.A. 1803 Building i March 1985 MIDLAND. MICHIGAN ABS*C Dr. Hasmukh Shah " Chemical Manufacturers Association 2501 M Street, NW Washington, D.C. 20037 Dr. M. N. Johnson B. F. Goodrich Company 500 S. Main Street Akron, OH 44318 Mr. R. N. Wheeler Union Carbide Corporation P.0. Box S004 South Charleston, WV 25303 The enclosed series of papers on vinyl chloride and styrene came to . my attention through Brian Bennett of ICI. They obviously will have to be included in the review by Sir Richard Doll and probably should ; be sent to Environmental Health Associates. t Sincerely yours. Theodore R. Torkelson, Sc.D. Health & Environmental Sciences 517/636-5197 Encs t, A*1 OPERATtNO UNIT OF THE DOW CHEMICAL COMPANY CHA 007600 Occupational Hazards of Vinyl Chloride and Styrene William J. Nicholson, Fh.D. y ENVIRONMENTAL SCIENCES LABORATORY MOUNT tIMAt SCHOOL OF MCO<CiNC OF Tw* OlTV UNlVlUflTY OF NW YQK 006 CMA 007&01 Occupational Hazards of Vinyl Chloride and Stvrene Trends in Cancer Mortality Among Workers in Che Synthetic Polymers Industry William J. Nicholson, Paul K. Henneberger and Diane Xarr Occupational Eazards in the VC-PVC Industrv William J. Nicholson, Paul 1C. Henneberger and Herbert Seidman Occupational Hazards in Production of Processing of Styrene Polymers - Epidemiclogic Findings William J, Nicholson and Diane Tarr Lectures presented at a course on occupational hazards of plastics and synthetic elastomers. Institute of Occupational Health, Helsinki, Finland, November 22-27, 1982 ' Published in:Industrial Hazards of Plastics and Synthetic Elastomers Eds. J. Jarvisalo, P. Pfaffli, H. VainiO (1984) Progress in Clinical and 3iological Research: Volume 141 Alan R_ Liss, Inc., Nev York, pp. 65--78, 155--176,263--278. ENVIRONMENTAL SCIENCES LABORATORY haOUN' IC.OOl O* MtDiCiNI C * Tf* c:tv umiviij.tv O* "Cw vo CMA 007602 TRENDS IN CANCER MORTALITY AMONG WORKERS IN THE SYNTHETIC POLYMERS INDUSTRY William J. Nicholson, Paul K. Heoneberger and Diane Tarr Environmental Sciences Laboratory, Mount Sinai School of Medicine of City University of New- York New York, New York 10029, U.S-A. INTRODUCTION , The reactive double bonded structure of ethylene-like molecules allows a wide variety of chemicals to undergo polymerization. Unfortunately, this same.structure has been found capable of transformation to an epoxide by the mamma1ian mixed function oxidase system (Bdnse and Henschler, 1976). These epoxides or their reactive metabolites can bind to cellular macrocode rules and may be responsible for the carcinogenicity of the parent molecule. Epoxide forma tion has been suggested as an intermediate in the carcino genic action of vinyl chloride (Van Duuren, 1975) and vinylidene chloride (Maltoni, 1977), and in the mutagenic action of styrene (Kilvy and Garro, 1976). The epoxides of ethyl ene, styrene and vinyl chloride have been shown to be carci nogenic, as well as directly mutagenic in bacterial test systems without the need for activation. The potential for conversion of ethylene-like molecules to the epoxides is greeter for unsymmetrlcal structures such as vinyl chloride and vinylidene chloride than for symmetrical structures, such as ethylene, l,2-dichloroethylene or cetrichloroethy lene . It is beyond the scope of this review to discuss the structure-activity relationships of the monomers used in the plastics industry. Nevertheless, available data suggest that carcinogenicity depends on the metabolism of these monomers to reactive intermediates and that these reactions may be non-linear. However, when the metabolism of a com pound is understood, a coherent picture of the dose and time dependence of cancer should emerge- 007&03 At this time, date are available on both experimental and human carcinogenesis from exposure to vinyl chloride (VC) aad on its metabolism that provide information import ant for the understanding of observed dose-response rela tionships . This paper will consider these data on VC in detail as they provide estimates o the trends in future disease potential from past exposures and information on the efficacy of current occupational standards. As human and animal data accumulate on the effects of exposure to other monomers, the approach suggested by VC can be applied to their evaluation. dose-eesponse relationships VC is one of the best studied chemicals in animal systems. The magnificent research by Haltonl and associates (1981) on nearly 7,000 animals over a ten year period is virtually unmatched in experimental carcinogenesis. A principal feature of their results is summarised in figure 1 which shows the dose-response relationship for the percent age of animals that developed heaisnfiositcom* (HSA) of the lives from 4 hr/day, 5 day/wk, 52 wk exposures to different concentrations of VC. As can be seen, the relationship is s non-linear one with evidence of saturation at high eon- AOMINISTERED CONCENTRATION OF VINYL CHLORIDE (ppm) Figure 1. The percentage of rats developing liver hemangiosareoma from 52 wk exposures to VC for 4 hr/day, 5 day/wk. CMA 007*04 f ceatratiDcs. However, at concentrations of VC less than 500 ppm, a reasonably linear dose-response relationship obtains. Cehriag et *1 (1973) have explained the .non-linearity in terms of hichaelis-Henten kinetics, in which the trans- - formation of VC to a reactive intermediate follows the equation, 4 V = vJus/a9 + S) (1) V and are the rate and maximum rate, respectively, for the biotransformaLion of VC, S is the concentration of VC in inspired air, and K, the Kichaelis constant. K was deter mined experimentally to be 360 Mg/1 and V to be 5,706 pg/4 hr. figure ZA displays the dose-response- relationship between the percentage of animals with liver BSA and the quantity of VC metabolized according to q. 1. As can be teen, a direct liner relationship exists with no evidence of a threshold or altered slope at low dotes. The possibility of a non-linear dose-response relationship from detoxifica tion kinetic steps has been postulated (Gehring and Blau, 1S77); and discussed in detail (Hoel t al, 1933), but no evidence exists for such non-linearity in the data yet available. The unweighted least squares regression equation for the dose-response relationship is % HSA = -0.066 + 0.0039 V U1 OV JH <3 m V log V V (//g of vinyl chloride metabolized/ 4 hr) Figure 2. Linear and probit dose-response relationships for the quantity of VC metabolized/4 hr exposure (5 day/wk,52 wfc) * CMA 007605 Gehriog et *1 (1978) fitted the early data of Maltoai ad Lefemine (1975) to * log-probir model. Figure 23 chows the log-probit plot using all available data from the stud ies from Haltoai et al (1981). The unweighted least squares regression line is Probit = 0.24 v 1.01 log V (3) linear dose-response relationship fits the data somewhat better (r = 0.77). Further, there i* very Halted biologi cal rationale for the use of a log-probit relationship - in carcinogenesis and its use as a means of extrapolation to predict effects at very low exposures would appear to be more an act of faith than of science. On the other hand, a linear dose-response relationship between the incidence of HSA and the quantity of VC metabolized 'is biologically plausible and fits all available data. Its use is strongly suggested. '' TIME COURSE OF CAKCER 'v Much of human, cancer has been found to follow a power ;law relationship with age (Armitage and Doll, 1961; Cook et al, 1969), R = bfk (4) where R is the incidence rate of cancer at a specific sice, t is age, and b and k are constants specific to site. In general, k is between 4 and 6 for most epithelial malignan cies. While data for exposures to specific carcinogens are limited, bronchogenic carcinoma from cigarette smoking and mesothelioma from asbestos exposure also follow a power law of time from onset of exposure with an exponent between 3 and 5 (Doll and Peto, 1978; Hewhouse and .Be.rry, 1976; Peto et al, 1982). These findings have been interpreted in terms of multistage model of carcinogenesis, the implications of which have been discussed by Peto (1977), Whittemore and Teller (1978), and Day and Brown (1980), among others. Deviations from the above time course occur with exposures to carcinogens that interact symergistlcally, such as asbes tos and cigarette smoking in the production of lung cancer. Thit interaction can be incorporated in the multistage mo del, but a more complicated, relationship obtains. However, CHA 007606 far a rare tumor, such as HSA, interactive effects may not be important and a power lav relationship should adequately describe the time course of risk following exposure. Some data are available from the use of Thorotrast in Japan and Denmark that Indicate " the incidence rate of HSA does follow Eq. 4 (Mori et al, 1979a; Mori et al, 1979b; Taber, I97S). The material was used in these countries over a limited period of time, so the incidence per calendar year and estimates of the population at risk can be used to estimate incidence rates by time from onset of exposure. While the data are very limited, they are consistent with a power law dependence of risk and suggest an exponent -of ap proximately 3. Three is also compatible with the incidence of liver HSA in the mortality study of polymerization work ers described elsewhere in this volume (Nicholson et al, 1983). However, only nine cases are available for analysis. - PROJECTIONS OP FUTURE MORTALITY PROM PAST VC EXPOSURE Sufficient data have accumulated *- on the pattern of mortality from past VC exposures to allow an estimate future mortality from these exposures of using a linear dose-re sponse relationship and time course for risk of death from liver ESA given by Eq. 4. Figure 3 shows the number of eases of ESA according to various measures of time that have been identified in the United States, Western Europe and the world (NI05H, 1982).- The distributions shows in Pigure 3 are the result of the exposure to VC of various groups of individuals in different periods of time since 1935. Equa tion 4 indicates that the incidences (not incidence rates) according to calendar year, year of exposure, and year from onset of exposure, respectively, are: h `Hi ci (Kort> - ci jh *U Vi * '!-i1 ci Tiw"> t5,) t5b) '*=> where i represents the quinqueniuo of exposure and j, the qulnquenium of observation. i runs from 1 to 8, represent ing the years 1935-1974 and j from 1 to 9, extending the observations through 1979. The F.(Mort) are the appropriate age and calendar year adjustments to the population in Jtr CM* 00760? 809^00 VWO .Aiaa*2da* azneodxa spxzotqa tauta zaaqa pn paeodxa saaqzoA jo aaqnma aqa Aonq 2on pan uo iaansodxa uoxaeTndod T*aoa aqa 02 T*noxa -z odoa <3 ex VSH jo sjezz qa `sazneodxa jaqzoA atom jo attirx qa tit aeaoxi *T 3A pait-[oq*aaia pw pa2T<Jrnt qqoq } dxq* -noxiTxas a*aodsaa-assp qa aynxs *tiota*3atta3B09 o& 1i3ba* aqa s aorta pojjad aorta naAjB ax paiodxa szaqzofl 50 zaqwmi wSeaaA* aqa 4-a-t `ajnsodxa noxacxnded inoa qa oa xraoT? -zodozd aa* e,o aqx *datx*X3cw xnuea o*J f Ktton6trpti6 '' \ -rpaaaxaa aorta T*a -aas 1 oa Soxp^oaa* pxza sqa put adozng naaqaa^ `"S'fl *qi 07 aaAXX aqa 50 TutoaaasoxSnraaq 50 cam jo zaqamti aqx ' aanSt uie*g ot wfttowj jo iiuq mwj. hi>*4 p a *p*i*a a c I e A :i ' . i ftr * i . nm 0 * 9 uniQtfi^ |Q Ja, ame4x3 itjy |fi to9jr 1 It M t ft* Relative values for the C.'s can be determined from two sets of data. The first is tke incidence of USA according to calendar period of first exposure Cl,.). Here the C^'s are directly proportional to the incidence-in a given calendar period and available data are sufficient to establish rea sonable values of C. for the time period 1935-195S. Addi tional data on C. can be developed from published data on the production of3"VC monomer. Figure 4 displays the avail able information on production in the United States (S.P.I., 1975-1978; U.S. Tariff Commission, 1946-1966) and Western Europe (O.E.C.D., 1971). A first approximation to the population exposure in different years would be to consider the C.'s to be proportional to VC production. However, average VC concentrations changed over the years of concern (Table 1) and an adjustment for the different relative exposures in different times must be made. This adjustment is indicated in Table 1 and on Figure 4. Further, an ad justment must be made to take into account the different number of workers required to produce a metric ton of VC In different time periods. As it would be expected that more workers were employed per tonne of VC produced during ear lier years, an adjustment is required to account for produc,tivity. Initial estimates of this factor'are also indicated in Figure 4. The relative population exposure, taken to be the product of production, the workforce productivity ad justment, and the exposure adjustment is shown by the solid w*' AELAMYC POrta MI<1M IK m ilfllE Figure 4. The production of VC in the U.S. and Western Europe along with estimates of the population exposures to VC poly merization workers in different calendar periods. A CHA 007609 Table 1 Snurii and aatteatad maoaures ta vinyl chlerlda eo BQlv3ritrior. witm lr. nanam cln ?rloc* Calendar Period ' Approiclnac* <* Vliyrl Chloride axt>c<ur Cose} ralatlvt kersei (1976'} Oct *t el (1975) Sue In ec el (1975) exposure Batora 1950 1930-34 1953-59 1960-64 1963-69 1970-74 1000 1000 400 - 500 300 - 400 300 - 400 150 - 300 100 - 400 100 - 400 10 - so 10 - 30 100 - 900 40 - 30 40 - 60 5 s 3 1 1.3 1 lines across each quiaquenimn. The dashed lines across each quiaqueniua during earlier years are those determined by fitting the observed HSA incidence to q. Sa and matched to the value estimated from production data in the quinquenius 1955*1959 - As can be seen, the comparison of the two sets of data suggested that the population exposure prior to 1960 vas slightly less in some quiaquenia thaq that estimated by the use of the adjustment factors indicated in Figure 4. The procedure of estimating the relative values for C., particularly in the years after 1960, is clearly an approxi mate one. To consider how sensitive any projections of future mortality are to the choices of C.'s, alternate choices are shown by the light solid lines Lin Figure 4a. Any realistic estimates of the C.'s must lie between tbe two lines. ~ Relative values of I., and I._. were calculated using the relative values of C. shown, it1 Figure 4, values of k between 2 and 4, and absolute values determined by matching to the incidence data of HSA found in Figure 3. In this calculation, the age distribution used for time of first exposure was: 15-19, 3.5*; 20-24/ 26*; 25-29,-26*; 30-34, 15*; 35-39, 111; 40-44, 7*; 45-49, 4*; 50-54, 2.5*. This distribution was that of 760 VC workers examined by Mount Sinai School of Hediciae personnel during 1974. The pattern of duration of employment was assumed to be a decreasing exponential with an average employment time of 12 years. Thia corresponds to typical patterns of employment for long-term workers in the chemical industry (Hicbolsoc et al, 1982; Wong, 1982). Separate calculations were made for the CMA 007610 United States and Western Europe. The results of this procedure, combining the data for the United States and Western Europe, ere shown in Figure 5. As can be seen I. ^, the incidence according to years from onset of exposure is best fit by a value o fc * 2. A value of 3 is" compatible with, the data, but values greater than 4 can he ruled out. I. is relatively*insensitive to the choice of k but a value of 4 fits the data best. An interesting feature of this calculation is that the separate determination of the C.'s for Western Europe and the United States indicates that the population exposures per tonne of VC produced were approximately four times greater in Western Europe than the United States. This would sug* gest that more intense exposures occurred 'in some European plants or that more workers were exposed per" tonne of VC produced- year of Ffftsr exposure YEARS SINCE ONSET OF EXPOSURE Figure 5. A comparison of the calculated incidence of heaangiosarcoma o the liver with that observed in the U.5. and Western Europe according to several time criteria and models for calculation. CHfii 007611 One set of data that differs significantly from that calculated is the distribution of cases according to years of VC exposure. As mentioned previously, ue assumed the distribution of employment times in the VC industry would be. decreasing exponential with a mean employment time of 12 years. The significant deficit of cases with employment times less than 10 years suggest that the available informa tion on duration of VC exposure may not be correct, that our assumed employment distribution may be in error, that there may be an naderascertainment of cases with shorter exposures, or that there may be proportionately less risk for shorter exposures than would be predicted on a linear dose-response relationship. It should be mentioned that duration of em ployment is not an important variable in assessment of popu lation risk. Shorter employment times would have required more men to be exposed, but their average exposure would be proportionately lower. _ Using the values of C,*s determined by the preceding analysis and values of k from 2 to 4, the mortality from liver HSA is calculated to the year 2040, using Eq- fa. These data are listed in Table Z, separately for Western Europe and the United States. Also shown in .the data for the United States are projections using-values of C, "s indi; eated by the solid curves ef Figure 4 and projections assum ing that the risk of HSA will increase quadratitally with age of exposure. This age dependence was suggested by expe rimental results of Groth et al C1931). .We also considered a time course for HSA that increased as t"5 for only 45 years aud remained constant thereafter. As can be seen, the pro- Tabl* 2 Trolactions of narttliffi i.- th C-lnd Su*ti and Waacara Europe to tha vast 2040 Iron wwmr so vinyl ctil.arl.da Briar ta 1575 ' Modal ' Halt*d Scatas* t* t* X* t1. t*. ts. e3. to S3 yaars free on*at of axpoaur* upper axpoaur* eurva, rig. lowar axposura eurva, Tig. + Afa* 19D 3*0 630 310 260 260 ' Uastarr Eurcua^ 540 1190 27 g0 1120 26 daash* hava occurred through 1979 39 deaths hava occurrad through 1979 CMA 00761 jected numbers of KSA fox the United States range from 200 to 600 and, for Western Europe, from 550 to 2,S00. (The greater range for Europe is the result of the more recent usage of pattern.) The most probable projection for future' disease is felt to be that represented by a power of 3, a choice suggested by Thoratrast data and the very limited mortality data on HSA in the study by Nicholson et al (1933). Lower values are also reasonable, but the fit to the data would suggest that the use of a power of A may be inappro priate . Obviously, many caveats exist in the consideration of these projections. The estimates strongly depend upon a reasonable ascertainment of eases through 1979. The concerns for VC-induced HSA in recent years would suggest that ascer tainment was fairly good, at least for long term employees and pensioners. However, some cases in short term workers may have been missed. The projections also depend on the choices of the C. and the k. We have projected mortality based os reasonable choices for these parameters. However, other choices cannot be absolutely excluded. While these uncertainties exist, the data indicate tSat, within a factor of 2 or 3, future HSA mortality from exposures prior to 1975 will be about 350 deaths in the United States and 1,200 in Vetters Europe. Further, these deaths will occur in a relatively small population. In the United States, the group at highest risk would be comprised of fewer than 5,000 individuals. Among this heavily exposed group, HSA may account for 10% of all deaths (Nicholson et al, 19S3). Clearly, any intervention techniques that might be developed to reduce this projected risk could be efficiently applied. OCCUPATIONAL STANDARDS FOR VC Nicholson et al (1933) have shown that-liver KSa accounts for at least 50% of all. VC-induced malignancies. Thus, it would appear that average exposures of 200-500 ppm in pre vious years will lead to 1,000-1,000 excess cancer deaths in all workers exposed to VC in Western Europe and the United States prior to 1975. If a standard of 1 ppm is met, the average exposure of all the workers would be between 0.2*0.5 ppm, 1,000 times less than that which existed previously. One would expect the VC-induced malignant risk to be reduced by corresponding amount. This implies that, if the VC industry complies with a 1 tea standard, cancer from emplov- 00PAIS neat, therein would be -virtually eliminated. However, there wilj. still remain a rick of developing USA of the order of 10 per individual for a working lifetime, based os a United States or European workforce of about 10,000 workers. SUMMARY A high risk of death from liver HSA has been documented from past exposures to VC. Similar to other carcinogens, the risk of VC-induced liver HSA appears to increase as the second or third power of time from onset of exposure. It is possible to project future mortality using this power rela tionship, estimates of VC exposure, and observed mortality to 1950. These projections suggest that 200-500 deaths may occur in the United States and 550-2,300 in Western Europe from liver HSA. These projections also suggest that a 1 ppm standard in the VC industry will go far to protecting workers from future malignant disease. REFERENCES Armitage ?, Doll R (1961). Stochastic models for carcinoge nesis. In: Proceedings of the Fourth Berkeley Symposium on Hsthesetical Statistics and Probability. (Ed. Reyman J) Univ Calif Press, Berkeley pp. 19-38. Sames AW (1976). Vinyl chloride and the production of PVC. Proc Roy Soc Med 69:277-280. Sense G, Hanschler D (1976). Chemical reactivity, biotrans formation and toxicity of polychloriaated aliphatic com pounds. CRC Crit Rev Toxicol 5:395. Cook FJ, Doll R, Fellingham SA (1969). A mathematical model for the age distribution of canter in man. Int J Cancer 4:93-112- Day RE, Brown CC (1980). Multistage models and primary prevention of cancer. J Hatl Cancer Inst 64:977-989. Doll R, Peto R (1978). Cigarette smoking and bronchial carcinoma: dose and time relationships among regular smokers and lifelong non-smokers. 32:303-313. * J Epidem Comm Health Faber M (1978). Malignancies in Danish Thorocrast patients. Health Physics 35:153-158. Gehring PJ, Blsu GE (1977). Mechanises of carcinogenesis: dose response. J Environ Path Toxicol 1:163-179. CMA 007614 Ge tiring PJ, Watanabe PG, Park CN (1978). Resolution of dose-response toxicity data for chemicals requiring meta bolic activation: example - vinyl chloride. Appl Pharmacol 44:551-591. . J Toxicol Grotb DH, Coate WB, TJIIaad BM, Hornuag, Rtf (1981). Effects of aging on the induction of angiosarcoma. Environ Health Persp 41:53-57. Hoel 06, Kaplan KL, Anderson MW (1563). Implication of non linear kinetics on risk estimation in carcinogenesis. Sci 219:1032-1037. Maltoai C, Lefemine G (1975). Carcinogenicity assays of vinyl chloride: current results. Ann H7 Acad Sci 245: 195-224. Maltoai, C (1977). Recent findings on the carcinogenicity of chlorinated olefins. Environ Health. Persp 21:1-5. Maltoai C, lefemine G, Ciliberti A, Cottl G, Carrettl 0 (1981). Carcinogenicity bioassays of ' vinyl chloride monomer: a model of risk assessment on an experimental basis. Environ Health Persp 42:3-29. . Hilvy p., Garro AJ (1976). Mutagenic activity of styrene oxide C1,2-epoxyethylbenzene)r a presumed styrene metabo lite. Mutat Res 40:15-18. V Mori T, Kato 7, Shimamine T, Vatanabe S (1979a). Statisti cal analysis of Japanese Thorotoast-administered autopsy cases. Environ Res 18:231-244. Mori T, Maruyame T, Kato Y, Tahahashi S (1979a). Epidemio logical follow-up study of Japanese Thorotrast cases. Environ Res 13:44-54. National Institute of Occupational Safety and Health (U.S.) (October,1982). Reported cases of angiosarcoma of the liver.among vinyl chloride polymerization workers. Newhou.se KL, Berry G (1976). Prediction of mortality from mesothelial tumors in asbestos factory workers. Brit J Indus Med 33:147-151. Hicholson WJ, Perkel 6, exposure to asbestos: Selikoff IJ (1982). Occupational population at risk and projected mortality - 1980-2030. Am J Indust Med 3:259-311. Nicholson WJ, Henneberger P, Seidman H. Occupational ha zards in the VC-PVC industry.. This volume. Organization for Economic Cooperation and Development, Chemical Industry (1971). Quoted in: Levinson C. Work ha zard: vinyl chloride. ICF Geneva. Ott MG, Langner RR, Holder BB (1975). Vinyl chloride expo sure in a controlled industrial environment. Arch Environ Health 30:333-339. * Crtrt 007615 Feto S (1977). Epidemiology, multistage models and short term mutagenicity texts. Is: Origins of Human Cmncer {Eds. Hiatt HH, Watson JD, Wins ten JA). Cold Spring Harbor Laboratory pp. 1403-1430. The Society of the Plastics Industry, Ine^ (1975-1982). Facts and Figures of the U.S, Plastics Industry, New York. Suciu I, Prodfn El, Paduraru A, Pascu L (1975). Clinical manifestations is vinyl chloride poisoning. Ann NY Acad Sei 266:53-69. ` U.S. Tariff Commission (1948-1968). Polyvinyl chloride and copolymer production data. Tan Ihxuren B (1975). On the possible mechanism of carcino genic action of vinyl chloride. Ann NY Acad Sci 246:258-267. " Wfcittemore AS, Keller JB (1973). Quantitative theories of carcinogenesis. Society for Industrial and Applied Mathe matics Review 20:2-30.'_ - - '-* Wong 0 (1981). An epideooilogie study of workers potenti ally exposed to brominated chemicals: with'a discussion of a multifactor adjustment.1 In: Quantification of Occupa tional Cancer (Eds. Peto R, Schneiderman H). Banbury Report 9 Cold Spring Harbor Laboratory^pp. 359-378. CMA 007616 / ;: r,. `-di' ^ OCCUPATIONAi HAZARDS IN THE VC-PVC INDUSTRY William J. Nicholson, Paul K. Henneberger aad Herbert Seidmaa. Environmental Science* Laboratory, Mount Sinai School of Hedicine of CUNY, Hew York, Newark 10029 (WJN, PH) and American Cancer Society, A W. 35th Street, New York, Hew York 10001 (HS>. INTRODUCTION '. On January 24, 1974, The Wall Street Journal publish ed an article describing the occurrence of three deaths from hemangiesarcoma of the liver among '^polyvinyl chloride (PVC) production workers at the B.F. Goodrich Tire and Rubber Company plant in Louisville, Kentucky. This announ cement shattered the relatively complacent view toward health effects associated with plastic production in general and PVC production in particular. At the time, U.S. and Western European production of vinyl chloride (VC) exceeded 6 x 10 . metric tons. Numerous mortality and clinical studies were undertaken in the major producing countries in an attempt to establish the extent of the carcinogenic risk aad to identify clinic*! parameters useful for surveillance of exposed groups. Because of tbe immediate concern in 1974, most of these studies were com pleted between 1974 and 1977- Several reviews and sympo sia on human health effects from VC exposure have been published recently. A superb one is byLelbach and Marsteller (1951). The exposures were bigh that led to the disease observed in these various studies. Typical concentrations in the industry were estimated to be about 1,000 ppm prior to 1955, from 300-500 during 1955-1970, and from 100-200 during 1970-1974 (Barnes, 1976). However, variations from such exposures would have occurred in specific plants (Rowe, 1975). While historical average exposures were generally less than 1,000 ppa, peak exposures often ex- fir CHrt 007617 ceeded 5-10,000 ppm (where workers lost consciousness) and, on occasion, 40,000 ppm (where plants exploded). During 1974, exposures were reduced to about 10-20 ppm in the U.S. industry (Jones, 1931) and even further, follow ing the promulgation of a 1 ppm standard by the Occupa tional Safety .and Health Administration in 1974. MORTALITY STUDIES OF VC-EXPOSED WORKERS Table 1 above the populations observed and the follow up characteristics of twelve cohort studies of vinyl chlo ride exposed workers. The studies were independent with the exception that the portions of the population reported in thp Equitable Environmental Health Study (197S) were included in some other U.S. studies. The proportionate mortality study of Henson e al (1974) is not included as the VC-exposed individuals studied therein were included in the cohort mortality study of Waxweiler et al (1976). The size of the cohorts varied greatly, from 255 in the study of Hicholson et al (1975) to 9,677 in the Equitable Environmental Health study. A notable .feature of all of the studies is that the populations followed were rela tively young or recently employed, even- though many plants in the studies started production in the 1940s. Host workers were hired after 1950, when U.S. and Western European production Increased sixfold in ten years (Hicholson and Heuneberger, 1933). Thus, few deaths occurred among most of the groups observed and data on effects 25 or more years from onset of exposure are li mited. The total mortality exceeded 10% of the observa tion cohort in only three studies. Further, the inclusion of recently employed individuals or those with short employment diluted the effects from VC exposure. Only five studies limited consideration to individuals with more than one year of exposure. In all cases, however, some Individuals. with more than 20 years from onset of employment were availsble for observetion. ' The follow-up terminated In the mid-1970s for all studies. Table 2 compares the results for esneer of sll sites and chronic liver disease in all 12 studies. Cancer is elevated in most of the studies, although it does not achieve a 0.05 level of significance except in the studies by Waxweiler et al (1976) and Hicholson et al (1975). In the study by Ott et al (1975), a highly exposed subgroup with 15 years latency bad 8 cancer deaths compared to 3.2 CMft 007618 Ay*'-: i 'niBOp |q aaaiis? aaoi jo| pi*HtK>3 s|3i*]| 9(Mr| * (161 II 0Y6I 1161 DC mi 9161 YH YH 1(61 U ZMI MCI ft . mi 1.bl 01 mi fftl St 0V6I 1161 St (SC6I) S161 tit 16l MAI H SMI S16I 11 .f 1161 zt Jn-no|]i>) jo Mai An-M||OJ W| miHM B961 ZS6I ainBodan *01 tf IK. Ot *01 1 *0< *1 *0( *0< *t'ft CO S S 0< ,J 5 V o< t1 s 1 CK 1 IK OC Z*0 S'O (cm**) sjHaotlaa 9'01 9U s*o III CO i. rti ' Cf ft . m r 1 ' Cl v s - ^. tt . 0*0 -' r* K . CO SB . 9Z' CO (*S tst ri , Cl mi n ,. C9 911 CO Cl as IW| 0'9 at 0-l> Cl iwn |o luitaja,! 19 p9lA|*w 41p jo B'El l>HO|1|pp* JU31IJ MCI Mti yso '! ib ii||wK'A mi >>*119 m HT3 jm l|nB|JMU, 1 ^c; 1 j MCI st ^' l ; - i* iia y . MCI Its iwn l no '1 : SUI SSt : YSO *|t h wii|ih))|a 90C ' <ir ' Mb! pnij| mi *im COW in >11100 l"* *J ^ til U wn AM) BI^BIIMb) MCI CHI t i> na -* 0S1 i WOT , -* ' 9161 *I IB l>l| 6141 999 wn1: *1* 1 i*I||na 6141 tut ITU *( If FI l|* TfiTwwo 1101(03 i|i<py IpSl# pmutxti apj jojijif y/u"|A fo t*|pn} m|mi jo i |i | in'-m[ | | pw Table 2 Obsar-rad and xt>acred daachl <i chlorida sxooaed workeri In pulw atudiss Sestlis stuir Obaer. Exoee- l*rum u el 30 30.9 Ssfflsr c si 5 rr. latency S 4 5.19 4.34 lyres. sc ml -- Duck, sc el 35 36.44 EqnltS&l* 139 161.39 fee 4 Collier 115 126.77 haauds 8 5.8 Slcholsos 9 3.9 oee sc el 13 13 yr. latency 9 ' 16.0 9.2 IsLai sc si 94 90.6 Theriault 4 Allard 20 16.37 WssusUst ac si 15 yr. latency 35 31 23.3 16.9 SH* 97 154 138 96 1040 91 138 230 . 81 98 112* 122 149+ 184++ Deaths Obaer. Exoec. 3- 0 0 14 `1 5 1 3 26.43 2.68 1.00 (0.6) -1.7 14 18.4 -- *2 . V4.0 - * adjusted for unknown. causes of death ( ) " Satinatad as a ;>ere*ntja o a.3. ikm + ? * 0.05 ++ p < 0.01 * 90. of control population squally hijh. - SKX - - S* -54t 37 300* 167 m 82 50 expected (p < 0.05). The ebieo.ee of significant findings in other studies mey be Attributed to their low power. The study of Bertezzi et el (1979) ey be biased beeeuse of lov follow-up in the group. Fourteen percent of the population were untraced end person-yeers et risk were calculated for these individuals as . if they were alive. The lew SHE of 44 for ell esuaea of death suggests that proportionately- nore deaths occurred in untraced groups than in the traced. The studies by Buffler et al (1979), Byren et al <1976), Masuda (1979), and Theriault snd Allard (1981) had very few deaths sveliable for analysis. That of Ott et al (1975) also was limited by the number of deaths and further by virtue of a study group with rela tively lower exposure (through better industrial hygiene control). While having more deaths available for analysis (136), the study by Ouch et al (1975) was significantly CMA 007620 diluted by the inclusion of many individuals with very short sad recent periods of exposure. Turning to chronic liver disease, one remarkable finding is the absence of significantly elevated mortality from this cause in most Of the populations under observe* tion. The only Study with a significant elevation is that ' of Maiuda (1979) in which five deaths from chronic liver disease occurred where only one was expected. However, this must be considered in the light of an equally high mortality from liver disease (6 observed vs. 1.4 expected) in a comparison population followed for control purposes. Five of 62 deaths from chronic liver disease seen in,, the study by Bertazzi et al (1979) are unusual, but the limi tations of this study and lack of details make evaluation difficult. The generally benign results in other studies contrast sharply with the severe liver disease from VC exposure documented in clinical studies (Marsteller et al, 1975). Hepatomegaly, hepatic fibrosis, portal hyper tension, and bleeding esophageal varices have commonly been found in individuals heavily exposed to VC, even without concomitant exposure to alcohol- - Table 3 lists the mortality data for primary cancer of the liver and biliary passages and for cancer of the lung, trachea and bronchus. In the case of liver cancer, the overall data are consistent and dramatic. Hemangiosarcomas of the liver were found is eight of the twelve studies. In each of the eight, a very large and highly significant SMS for liver cancer was seen. Methodological limitations can account for negative data in the other four studies. The large SMR's observed, however, are largely the result of low values for the expected number of cases rather than a high incidence of observed cases. Only 29 separate liver bemangiosarcomas were identified in ell twelve studies. As the overall excess number of deaths from liver and biliary cancer in all studies was 47, some hemangiosarcomas may not have been identified. The low numbers must also be considered in light of the limited follow-up times in most studies. The evidence for lung cancer is less clear. There is an elevation in some studies, but at a level that does not achieve statistical significance, except in the 15 year latency population of Waxweiler et al (1976). This, in part, may be the result of the low power of many of the QQ7&21 Table 3 . ' J .< i IftMrwi and oreacted daathe Iran eelected mwt ^laaoi vinyl cmagldttx^aitd yoyfcirt - Cue*? of thi liver end bUlatv oaei ' Cancer of the lun*. evict** ud bmehut Obe.* In. SMC MlftflO* ureoUK Oba. Mt lirtwii Bufflac 5 JT. la coney ycon . 10 yr- la coney Buck 19 ye. laconey Equitable 15 ye. latency Fox and Collier 15 yr. latency B 0 4 4_ 10 4 11.CJ* 0.17) 0.97* 0.66 4.37 0.71 (M0)" i* 7 1 (7.7)c -- 0 ` 5 1.73 4 1 * ' -1.49 *ut 909 1" 3 ' "1.7B 2 - -^ ' 0: - - It -' 19.53 14 . 10.69 563 " '5 *' 45 41 ''^*6 21 ''44.29 37. D 91.13 _ 36.0 bands Bicholun Otc 1 Xoial ' Thorn ill t L5 ye. laeaacy i 3" o12 0.1 (0.12) fO-JJ 0.9 (0.5) 167 (2500)" t, 1913 *** (1600) *** 0 3 0 4 s ` 1 (0.8) 0 tl. U *57) .2(3.2 11 ; - 24. 6 1 ` "^i.TB .4.25 (lamellar 19 yr- latency T 7 0.6 0.4 1135 1606 6 11 ' ; -7.7 6 11 3.7 Xiul of adeduplicated heaeotloearco 29 <911 161 161 103 131 1071 111 90 10 S 123) -- 77(96?) 95f 35 47 156 1J4T < Q.05 . <0.01' ' <0.001 ' All vuxifiad Liver weir <utki. iceludint chose established by mi all available Intonation. , ,,. ' :f > tariff if feiatha ittUiMi on eh* -bail*' of 1T5&-1..9J67 C-S. 4]uau4 catan. ICS 13JAC0 1*0-10J._ ^ ( ) > Iqicttf deaths wtiMUl cm tbie balls of national | adjusted ratal. ICS 111-143/ICS 140-205. .' Om beam iloaax cone oecixmd in a tvt libTicun. " Includes cancer of the a*ncrni. adjusted for wnknoua cauaea of death. CMA 007622 studies. Only two have an SOX power to detect an overall risk of 1-5 (Beaumont and Breslov, 1931). Of signifi cance, however, are the very low SKR's in the groups studied by Theriault and Allard (1981), Reinl, et al (1979), and Nicholson et al (1975), cohorts that would be expected to manifest a high risk on the basis of the many hemangiosarcomas that were found- The four largest stu dies, although in some cases limited by inclusion of short-term and recently employed workers, also are note worthy for the SKR's close to 100- Where available, data on snbcohorts with longer latency (> 15 yr) suggest some increased risk. Waxweiler et si (1931) undertook a detailed analysis of the exposure of those with lung cancer in their previ ously published study (Waxweiler et al, 1976) in as at tempt to identify particular etiological agents. The analysis used a serially additive expected dose model (Smith et al, 1960) in which a dose measure during each year of exposure was accumulated for each study individual for a variety of potentially carcinogenic agents. The cumulative doses for those with lung mincer were compared with those of other individuals in the plant under study. The results showed that the greatest correlation of lung cancer was with exposure to PVC dust- Secondarily, expo sure to vinylidene chloride appeared to be important, but only for large cell and adenocarcinoma. The serially additive dose far VC monomer differed little in those with lung cancer compared to others in the plant, except, possibly, for large cell cancers - Thus, evidence to date does not establish that VC r monomer is an important lung carcinogen in exposed worker populations, although it is recognized that limited long term observation has so far been available- In all stu dies .considered, here, a slight deficit.of cases was seen compared to the number expected. In the subcohorts with more than 15 years from onset of exposure, an overall excess of 10% was observed. If, in addition, one consi ders "healthy worker effect," any excess lung esncer would still be considerably less than the excess of liver cancer. A qualification to this conclusion is that no study specifically considered cigerette usage. If cigar ette smoking was much less common among VC workers than the general population, higher SMR's would have been seen If smoking specific data were available. However, this ir ChA 007-523 d possibility is unlikely, considering the many different populations studied. The uncertainty in human data is also reflected in animal studies. Increased lung cancers have been seen in mice but not in rats or hamsters (Maltoni at al, 1981). ' Table 4 "shows the results for brain and central nervous system cancers and for cancers of the lymphatic and hematopoietic systems. Cancers of the brsin and central nervous system were significsntly clevsted in a number of studies, although the results differed consider ably across studies. Again, negative data may be simply the result of limited long-term follow-up or the low power of tfre study. In such esses the information is onlysufficient to set an upper limit on relative risk of brsin cancer. In contrast to lung cancer, however, the largest study group has s significsntly elevated risk of brain and central nervous system malignancy. As with lung cancer, the data on brain and CSS cancer in animals are equivocal. Jfetxrobl a stomas and brain malignancies are observed in rats exposed to VC, but not among mice or hamsters (Kaltoni et al, 1981). The human data are also''mitigated by the recent finding of brain and central nervous system tumors in a variety of chemical plant exposure circumstances (Alexander et al, 1980; Selikoff et al, 1982). Excess brain malignancies, but not. the etiological agents, have been identified in several Texas and Louisiana chemical/ petrochemical plants. VC exposure was documented for some esses, but it could not explain the overall findings. As individuals in many of the VC studies considered here were exposed to other chemicals and petrochemicals, the pos sible sole of these agents cannot be excluded, further, it has been suggested that some working groups, with employer-paid medical plans, msy have better case ascer tainment than is generally available (Greenwmld et al, 1981) and, thus, sore brain malignancies identified. In any case, the number of excess aalignsncies of the brain and central nervous system (approximately 10) in all studies le considerably less than the number of hemangio- sarcomas identified in the same populations. Similar results are obtained for malignancies of the lymphatic and hematopoietic system. Here again, the analysis is limited by the few deaths and disparate re sults which occurred in different studies. Overall, there would appear to be an elevated risk, but the influence of CMA 007A24 Table 4 Obr?d and ernactad death* fron lctd cau**e waK rlnvl chlorlda 3tao*d worker* Cancer af Cha brain 4 eeaxHl nervous rriua Ciatf of cha lymphatic tad hegacggelcclc *v*tem Obin. atutU 1 Sufflar 0 Byrea Suck 2 - Equitable Fox S Colllar 12 2 Haevidx e Xlchelioa i ott l Balal 2 Theriault 0 Tiaaoiailer 3 15 yr. latency 3 Em act. EKE (0.8)* (0.1) 0.33 5.90 125 a* 612* SP 203+ 3. Si 35 (0.15) - (0.1) acoo) 0.4 . (250) 1.3 162 0.6 - 0.9 329 o.s 49 r Gfeeer. 4 0 0 20 9 0 2 1 15 1 4 Expect- SHE (3.0)b (133) (0.5) - - -' 17.01 124 9.01 100 (0.3) - (0.4) (300) (l.S) 7.7 (63) 2147+ 1.67 60 2.5 159 + < 0.05 f* * 0.01 '- * C > Expected oeluctd fees the ntls of xi standardised . 0.5. rate* SCO 193/ICS 140-Z05. b ( 1 Expected estimated fees the ratio of 1950-1909 ~.S. r:u ICS 200-203/ICS 140-230. coafounding exposures precludes definitive statements. The overall excess of such malignancies (about IQ) is also much leas those from primary hemangiosarcomas of the t liver. ' EFFECT OF REDUCTION OF EXPOSURE TO VC As mentioned previously, most mortality studies followed populations only to the 1972-1975 period. Ho deta exist on the risk to previously exposed populations after cessation of exposure in 1974, although heinaagio*arcomas have been noted among retirees. We have recently completed a follow-up through 1981 of the population reported in 1975 (Nicholson et al, 1975) to determine whether a high risk of liver cancer continues, following significant reduction in exposure. The original group OH* employed at a VC polymerization plant in Niagara Falls, New York, baa been expanded by 40 additional workers, all exposed for five years, who achieved ten years from onset of exposure subsequent to'April 1974. Additionally, 195 Individuals employed at a'VC polymerization plant in South Charleston, Vest Virginia, with five yeara of exposure and ten years from onset in December, 1966, were identified and traced through 1980. Table 5 lists the observed and expected deaths by cause for both groups with the deaths occurring after 1974 separately identified. (These are preliminary data; full hy. r\ t , ' < i`. ^ _ ~x t * - . Sable 3 Observed nui aspect*! death* u*ot Tiny! chloride puljMiintloa verier* Blxttr* TU* Ueaaary 1, 155* Dataaber 31, 17(1} ObMnri v feHag gf dtACh^56*?S^7S^lMTjffit___WE All UIUH Zb 20 kU *a.iT ioa All cancer ( 1 14 ). 172* tua# 0 2 2 3.25 2 Caloe/rtcra 1 Z 3 1.39 214 M* 1 ,0 0.33 M3, Liver 3 3 4 0.19 313Ss fsZnu 2 1 3 0.35 . 543* '1 0 1 0.30 200 CWteela of liver 1 1 I 1.41 1(2 Cardlcmaculnc dlaeaee 13 1 21 u.ia 104 .. " feuxb CZarlaaton, CV <w - 193) tXliLeeS r 1, 1944 December 31, 19(0) Com of (Such motived 60-13 Tt~tQ All eaeiaa 14 All cacir 2 1 blx/mn* 0 0 Liver 0 iTWkMt 0 o Clntnu if liver 0 Cwdlmiiulir disease 10 22 10 1 0 .0 4 0 1. 1 u .IiL 34 12 Z 0 0 4 0 1 1 20 Exrvctad SHU (A,74 46 10.43 4.07 , .-A U43, i. 0.43 0.Z3 1739* 0.39 0.47 130 1.(1 35 27.24 73 a 9 0.63 b v * 0.001 or* o.oooi CHA 007626 pathological review of all available specimens ha* not been completed.) Among the 64 deaths that occurred in the Niagara Falla cohort, 6 were from primary cancer of the liver, including 5 hemangiosarcomaa. Three of the hemangioaarcooa* occurred in the period prior to 1974 and 2 ubsequently. Similar findings occurred among the smaller group in Vest`Virginia. Here, of 36 deaths, 4 were from heaangiosarcoma, all of which occurred subsequent to 1974. Thus, the risk of neoplastic VC disease continues undiminiched, even though exposures to the monomer have been significantly reduced. The combined data from both groups are shown in Table 6 and demonstrate an excess risk of cancer, which is totally accounted for hy the enormously increased risk of liver malignancy observed in each time from onset of exposure category. The excess lymphoma* which achieved significance at the p e 0.OS level in the Niagara Falla group lose significance when combined with the data from South Charleston. A deficit of lung cancer was observed in both study groups and brain malignancies were about equal to the number expected. It is not certain whether the resblts of these two plants will be reflected in the results .of other plants in future years. The South Charleston plant was the first facility to commercially produce VC. The New Fork plant opened insnediately following the cessation of Vorld War II, Thus, we are observing effects in populations that include many individuals with long times from onset of exposure. There is no information on whether the expo sures in these two plants were significantly different from those of the majority of other VC polymerization facilities. It is known that pre-1976 exposures in the New York plant were sufficiently high to cause loss of consciousness to some individuals (4.5X of those examined in the clinical survey of 1974) (Lilia et al, 1975). MORBIDITY AND CLINICAL FINDINGS AMONG VC-EXPOSED WORKERS Clinical abnormalities from VC exposure predated by 25 years the documentation of its carcinogenicity. Vari ous VC-related abnormalities were reported in Eastern European literature, including hepatomegaly (Tribukh et al, 1949), angiooeuroais (Filatova and Gronsberg, 1957), osteolytic lesions of distal phalanges (Smirnova, 1961), Raynaud's phenomenon and sclerodermalike akin lesions (Suciu et si, 1963). However, VC disease was not seri- it CHA 007427 * 7 4 tibia 6 Otrrj and *xnct*d Aaatha mant rlnrl chlorld* sxpgaad vorttrt In eno oolvmrltaclap fac.il 1 cm ' br Claw from1 onsac of amoaura Tiara slat* enact of axpeaorc Cass* of 4sach 10 - 19 20 - 29 30+ Total Ob*. All cauit* 24 19.13 All csseir 7 3.72 tuna 1 1.36 Id.TM* ! o.oa 3 rain 1 lywphosa 3 0.27 Cirrhosis of llvar 0 0.76 Cardlooas colat 14 8.74 dlsaasa farses years 2924 Ob*. JR: 30 34.33 9 7.98 1 3.96 3* 0-18 0 1 0.46 3 1.24 13 17.38 2734 Ob*. *P- 26 31.64 13 8.03 2 3.OS !l> 0.17 0 0 a.4i 0 o.jj 12 16.40 ' 1404 0b. SXB . SO 83.61 28 19.66 4 7.31 10 0.42 i 0.76 3 1.143 2.83 41 42.73 M 93 142 33 2381 132 263 103 96 a. bssa&flassrcsaa b. 4 baatancleaartDsaa and 1 baystsaa ously considered in the Vest until the published descrip tion of Raynaud's syndrome, acroosteoTysis, sad pseudoscleroderm* in two Belgium VC reactor cleaners (Cordier et al, 1966). Additional eases were soon noted (Wilson et si, 1967) and a comprehensive epidemiological study of 5,011 U.S. workers employed in production and polymerisa tion was undertaken. It showed that 11.9% had possible X-ray signs of scroosteolysis > compared with 3.2% in a Hichigtn general population control group, with 2% defi nitely having Raynaud's phenomenon or X-ray evidence of acroosteolysis (Dinmsn et al, 1971). The conditions were clearly associated with the cleaning of reactors, in which a heavy exposure to VC occurred. Only one case of Ray naud's phenomenon occurred among 557 workers employed in PVC fabrication. During the early 2970's, VC liver disease was de scribed in detail by Marsteller et al (1973, 1975). Observations on selected workers shewed bepato- and sple nomegaly to he common. Peritoneoscopy and guided liver biopsy identified severe portal hypertension in some, generally without cirrhotic fibrosis, although perisinusoldal and focal or diffuse capsular fibrosis were common ly seen. The portal hypertension could lead to bleeding esophageal varices, with possible fatal consequences. In CMft 007628 heavily exposed individuals', the portal hypertension and hepatic fibrosis often progressed after cessation of exposure (Martin et al, 1974). The histology of malignant and nonaaligoant liver disease has been veil described by Popper and Thomas (1975; Thomas et al, 19753, who suggest ed the possibility of an interrelationship between hemsn- giosarcom* and the proliferation of sinusoidal lining cells and bepstocytes seen in VC fibrosis. Lelbach and Marsteller (1981) have also noted that the vast majority of hemangiosarcona cases have appeared, on a background of some degree of hepatic fibrosis. The implications of these suggestions for a hemangioaarcoiaa dose-response relation are uncertain. " During 1974, extensive studies were undertaken by the Environmental Sciences laboratory of the total workforces of three polymerization plants in the states of New York, Michigan and West Virginia. The results from the New York plant (Lilis et al, 1975) indicated the presence of acroosteolysis in heavily exposed individuals. Hepato- and splenomegaly or hepatic tenderness va*^ commonly observed and aaaodated with duration of exposure and elevated alkaline phosphatase levels. 5ixty-four of 354 had an enlarged or tender liver or spleen and of these, 41% had elevated alkaline phosphatase. Liver function tests were not particularly revealing, except for a correlation of elevated alkaline phosphatase levels with duration of exposure. Additionally, carcinogenic embryonic antigen titers were slightly higher among vinyl chloride exposed groups than in a smoking matched control population (Anderson et al, 1978). Tasburro and Greenberg (1981) have evaluated the effectiveness of federally mandated screening tests for vinyl chloride exposed workers. Figure 1 shows the re sults' on specificity and sensitivity for- 78 individuals with hepatic status deternined by biopsy. ICG clearance bad the highest combined sensitivity and specificity, with SOFT tha second most useful test. Elevated alkaline phosphatase had the greatest specificity of all testa, particularly for chemically--induced liver injury, but was lacking in aensitivity. SGOT and GGPT were of limited use because of their low specificity for chronic liver disease. They recoEsnended the use of ZCG clearance for screening, to he followed with alkaline phosphatase determinations for those with altered clearance. CMA 00762? SENS u UK: SPEC 5 3 < (0 -ta9 ' -0.5 TESTS Figur* 1: Sensitivity and specificity of various biochemical screening eases and chair sensitivity and specificity sum values (5 & 5} based on 78 with biopsy documentation of their hepatic status. Three of the seven individuals vho died after 1974 with hemangiosarcoma in the previously described mortality followup were examined in 1974. One, who died 22 months after examination, had no noteworthy abnormalities oh examination (alkaline phosphatase was 83, slightly high).' A second, who died three years sfter examination, had a slightly enlarged, palpable liver (11 x 6 cm) with normal blood counts and chemistry. Only one of the above drank alcohol at all and he only drank 2-3 beers/month. The third, who died 22 months after examination, had a slight ly enlarged liver (11 x 8 cm) and spleen (13 x 8 cm), and slightly elevated alkaline phosphatase (93), SCOT (52) and CEA (4.7). Thrombocytopenia was also present (75,000). Vo data are available on later clinical parameters, but the above results are clearly not' sufficiently specific for Identification of a special risk. Pulmonary abnormalities also have been associated with VC/PVC exposure. Small opacities, predominantly irregular, of profusion 1/0 or greater were found in 20 of 1,216 workers employed at FVC production in an Italian plant (Maatraagelo et al, 1981). -All had been exposed to high levels of FVC dust (>10 mg/m'5). Lilis et al (1976) CA 007630 w reported that approximately 20% of VC/PVC workers with high exposures to PVC dust bad abnormal X-ray*, which correlated with duration of exposure and. also, with cigarette smoking. In contrast, only 4.7% of individuals in a PVC plant with low dust levels had abnormal X-rays. In addition to "typical pneumoconiosis," a granulatomous reaction to PVC dust has been reported (Amend et al, 1974). Hiller et al (1975) have observed pulmonary func tion abnormalities (a reduction in the ratios FEV^/FVC and KKF/predicted HHF) in both smokers and non-smokers heavily exposed to PVC dust (and also to VC monomer). Hsltoni and Lodi (1941), observed greater percentage of abnormal spu tum cytological results among VC exposed workers compared to several other groups of manufacturing workers or miners Only workers in the chromium industry demonstrated a greater proportion of abnormal cells. Dueatman et al (1975) have observed an increased frequency of chromosome abnormalities in the lymphocyte cultures of VC workers. Host of the abnormalities were "unstable" changes, such as fragments'-^ dicentrics, and rings. This was confirmed by Purchase et al (1978), among others. Seme of the group studied by Purchase were resam pled 18 and 42 months later (Anderson et al, 1980). In those studied during January 1976, the frequency of abnor malities was increased in those who continued VC/PVC employment, hut decreased in those who left the industry. In January 1978, no increased frequency was found in any worker. The authors attributed the decrease to the reduc tion in VC exposure. HEALTH HAZARDS IN THE PVC PROCESSING INDUSTRY t Prior to identification of hemangiosarcoma in VC polymerisation workers, little effort was made to control 'either the concentration of residual monomer in PVC dust or exposures to dust and VC that occurred in the various forming operations of the PVC fabricating industry- VC concentrations In excess of 10 ppm occurred frequently. While these concentrations were significantly lower than those of the polymerization industry, the much greater employment in the processing industry (hundreds of thou sands vs. tens of thousands in the polymerization work) raised concern for population health effects, particularly for malignant disease for which no threshold was known. However, only two hemangiosarcomas have been documented in # CMA 007631 r.I ? At*' the PVC processing Industry, one in an accountant in a plant ukins PVC fabric and one in an Italian plant making PVC sacks. A third case nay have occurred in an electri cal wire insulator, but' the ' pathological diagnosis is uncertain (Lloyd, 1975}.' This is in contrast to 85 cases known to have occurred among polymerization workers (NTOSH, 1982}. This is somewhat comforting and indicates a signi ficantly lower total VC-related neoplastic risk among fabrication workers. However, it should be noted that ease finding is likely to be poorer in this group than in polymerization workers. A proportionate mortality study has been conducted of A,341 deaths of former employees of 17 PVC fabricators (Chiazze Jr., et al, 1977). The direct PMR's suggested an excess in total cancer mortality among' both white men and white women with the major excesses concentrated in can cers of the digestive organs. An excess of breast cancer was also seen in women, but not confirmed in a case-con trol study (which was of very low power and could only detect a threefold increased risk} (Chiazze, Jr. et al, I960). The results of the proportionate mortality study must be considered cautiously. In such studies, elevated cancer risks and are typically seen because of a "healthy worker effect," which leads to a reduction in cardiovascu lar death* relative to those of cancer. If FCKR's (pro portionate cancer mortality ratios) had been calculated, rather than PMR's, digestive cancer would still be elevat ed but not at an 0.05 level of significance. Interest ingly, an excess of stomach cancer was seen in the propor tional mortality study of Baxter and Fox (1976). SUMMARY Overall, the results of the analysis of 12 studies of VC production and polymerization workers demonstrate an enormously elevated risk of liver, malignancies, the possi bility of a twofold increased risk of brain and central nervous system tumors and perhaps, also, of malignancies of the lymphatic and hematopoietic system. However, the role of other agents cannot bs excluded in the etiology of nonhepatic malignancies. Bronchogenic carcinoma does not appear to be increased from exposures to VC monomer, although a relationship to PVC dust was suggested in one study. These conclusions swat be considered in light of limited data on workers followed more than 25 years from CMft 007632 onset of exposure. Considering, the numbers of observed end expected deetbs in ell studies, it would xppexr that the excess of malignancies et aonbepetlc sites is less then the excess of liver,tumors, Date presented elsewhere in this volume (Nicholson end Henneberger, 1983) suggest thet exposure reductions in 1974 may hsve virtually elimi nated the VC-associated risk of liver cancer if the current U.S. standard is met. To the extent that VC exposure is associated with other cancers, a similar risk reduction would be expected. Raynaud's phenomenon, ecroosteolysis, sclerodermalike skin lesions, hepato- end splenomegaly with noncirrhotic hepatic fibrosis, and severe portal hypertension have been associated with past heavy exposures to VC. Evidence exists that the liver disease and portal hypertension may progress following cessation of exposure. However, all of the above syndromes were found largely in heavily exposed individuals - Their occurrence would be much less likely in workers exposed only to concentrations currently allow ed. Pulmonary deficits. X-ray abnormalities, and, per haps, lung cancer have been associated with VC/PVC expo sure. Because of the possible contribution of PVC dust to these findings, engineering controls during polymer dry ing, bagging and usage are warranted. ' REFERENCES Anderson HA, Snyder H$, Levinson T, Woo C, Lilis R, Selikoff 1J (1978). Levels of CEA among vinyl chloride and poly vinyl chloride exposed workers. Cancer 42:1360-1567. Anderson 0, Richardson CR, Height HI, Purchase XFH, Adams VGF (1980). Chromosomal analyses in vinyl chloride exposed workers: Results from analysis 18 and 42. months after an initial sampling. Nutation Res 79:131-162. Alexander V, Leffingvell SS,' Lloyd JV," tfaxweiler RJ, Hiller RL (1980). Brain cancer in petrochemical workers: A case aeries report. Am J Ind Bed 1:115-123. Arnsud A, Foamier de Santi ?, Garbs L, Psyan H, Charpin J (1978). Polyvinyl chloride pneumoconiosis. Thorax 33:19-25. Barnes AH (1976). Vinyl chloride and the production of PVC. Free R Soc Bed 69:277-281. Baxter PJ, Fox AJ (1976). Angiosarcoma of the liver in P.V.C. fabricators. Lancet 1:245. CMA 007*33 Beaumont JJ, Brealow NS (1981). Power considerations is epidemiologic studies of vinyl chloride workers. As J Epidem 114:725-734. Bertaxti PA, Villa A, Foa V, Sals B., Fabbri L, Mapp C, Marcer C, Hanno M, March! M, Marisal - F, Bottasso J (1979). A* epidemiological study of vinyl chloride exposed workers la Italy. Arb big rads teksikol 30:379-397 (Suppl). Baffler PA, Wood S., Eiffler C, Suarez L, Kilisa DJ (1979). Mortality experience of workers ia a vinyl chloride aonoser production plant. J Oce Had 21:195*202. . Bysen D, Engholm G, England A, Westcrholm P (1976). Mor tality and cancer morbidity in a group of Swedish VCM and PCV production workers. Environ Health Persp 17:167-170. Chiaxze Jr L, Nichols WE, Wong 0 (1977). Mortality among employees of PVC fabricators. J Oce Med 19:623-628. Cbiazze Jr L, Wong 0, Nichols WE, Fereace LD (1980). Breast cancer mortality among PVC fabricators. J Cec Med 22:677-679. Cordier JM, Fieve* C, Lefevre MJ, Sevrin A (1966)- Acro- osteolysis combined with skin lesions in two workers exposed in cleaning autoclaves. Chhierx Med Travail 4:14* 3-39. Dinmaa BD, Cook VA, Whitehouse WM, Magnuaon HJ, Ditcheck T (1971). Occupational Acroosteolysis. I. An epide miological study. Arch Environ Health 22:61-73. Ducatmaa A, Hirschhoro X, Selikoff IJ (1975). Vinyl chlo ride exposure and human chromosome aberrations. Mutation Bea 31:163-168. Duck BW, Carter JT, Coombes J (1975). Mortality study of workers in a polyvinyl-chloride production plant. Lancet 11:1197-1199. . Equitable Environmental Health, Inc, (1978). Epidemio logical study of vinyl chloride workers. Equitable Environments! Health, Inc., 6000 Executive Blvd, Boekville HD 20852. '' Filatova VS, Gronaberg S (1957). Hygienic working condi tions in the production of polyvinyl chloride resins end measures for improvement. Gig Sanit 1:38-42 (Russian text). Fox AJ, Collier PF (1976). Low mortality rates in indus trial cohort studies due to selection for work and survival in the industry. Brit J Prev Soc Med 30:225-230. Fox AJ, Collier PF (1977). Mortality experience of work ers exposed to vinyl chloride monomer in the manufacture of polyvinyl chloride in Great Britain. Brit J lad Med 34:1-10. .' Greeauald F, Friedlander BR, Lawrence CE, He a me T, Earle K (1981). Diagnostic sensitivity - an epidemiologic explanation 'for an apparent brain tumor excess. J Occ Med 23:690-694. Jones JH (2981). Worker exposure to vinyl chloride and polyvinyl chloride. Environ Health Persp 41:129-136. Lelbach VK, Marsteller HJ (19B1). Vinyl chloride-asso ciated disease. In: Ergebaisse der Ianeren Medizin und Einderheilkunde, Bd 47, Advances in Internal Medicine and Pediatrics. Berlin. P. Frick et al Eds. Springer-Verlag, . ' ; lf Lills R, Anderson H, Nicholson W, Daunt S, Fiscbbein AS, Selikoff IJ (1975). Prevalence of disease among vinyl chloride and polyvinyl chloride workers. Ann NY Acad Sci 246:22-41. Lills R, Anderson H, Miller A, Selikoff IJ (1976). Pul monary changes among vinyl chlroide polymerization workers. Chest 69:299S-303S (suppl). '* Lloyd JW (1975). Angiosarcoma of the liver in vinyl chloride/polyvinyl chloride workers. J Occ Med 17:333-334. Maltoni C, Lodi P (1981). Results of sputum cytology among workers exposed to vinyl chloride monomer and poly(vinyl chloride). Environ Health Persp 41:85-88. Maltoni C, Lefemine G, Ciliberti A, Cotti Gr Carretti D (1981). Carcinogenicity bioassays of vinyl chloride monomer; A model of risk assessment on an experimental basis. Environ Health Persp 41:3-29. * Marsteller HJ, Lelbach WK, Muller R, Juhe S, Lange CE, Rohner HG, Veltman G (1973). Chronic toxic liver damage in workers of FVC producing plants. Deut Med Wochachr 98:2311-2314. .. Marateller HJ, Lelbach WE, Muller R, Gedigk P (197S). Unusual aplenomegalic liver disease ss evidenced by peritoneoscopy and guided liver biopsy amoog polyvinyl chloride production workers. Ann NY Acad Sci 246:95-134. Mastrangelo G, Saiu B, Harcer G, Piazza G (1981). Epi demiological study of pneumoconiosis in the Italian poly(vinyl chloride) industry. Environ Health Persp 41:153-157. Hesuda Y (1979). Long-term mortality study of vinyl chloride and polyvinyl chloride workers in a Japanese plant. Arh big rede tokxikal 30:403-409 (suppl). Hiller A, Teirstein AS, Chuang M Selikoff IJ, Warshaw R (1975). Changes in pulmonary function in workers expos ed to vinyl chloride and polyvinyl chloride. Ann NY Acad Sci 246:42-52. Hanson Hit, Peters JH, Johnson KN (1974). Proportional mortality among vinyl chloride workers. Lancet 11:397-398. National Institute for Occupational Safety and Health (tt.S.) (October, 1982). Reported cases of angiosareos of the liver among vinyl chloride polymerization workers. (Unpublished). Nicholson WJ, Hammond EC, Seidnaa H, Selikoff IJ (1975). Mortality experience of a cohort of vinyl chloride- polyvinyl chloride workers. Ann NY Aead Sci 246:225-230. Nicholson WJ, Hcnaeberger P (1983). Trends in cancer mor tality among workers in the synthetic polymers industry. This volume. Ott tlG, Laagner NR, Holder BB (1975). Vinyl chloride exposure in a controlled industrial '"environment. Arch Environ Health 30:333-339. .. Popper H, Thomas LB (1975). Alterations of liver and spleen among workers exposed to vinyl chloride. Ann NY Aead Sci 246:172-194. ... Purchase ITH, Richardson CR, Anderson D, Paddle GM, Adams VGF (1378). Chromosomal analysis in' vinvyl chloride exposed workers. Mutation Res 57:325-334. Reial W, Weber H, Greiaer (1979). The mortality of German vinyl chloride (VC) and polyvinyl chloride (PVC) workers. Arh hig rads tokslkol 30:399-402 (suppl). Rowe VJC (197S). Experience in industrial exposure con trol. Ann NY Acad Sci 246:306-310. Selikoff ZJ, Hammond EC, Eds. (1982). Brsln tumors in the chemical industry. Ana NY Acad Sci 381:1-364. Saisanbva HA (1961). On the question of bona lesions due to chronic intoxication by olefins and vinyl chloride. Vests Renigenol Radiol 36:63-66 (Russian text). Smith AH, Waxweiler RJ, Tryler HA (1980). Epidemiologic investigation of occupational carcinogenesis using a sarially additive expected dose model. Am J Epidca 112:787-797. CMA 007636 Sucui I, Drejman I, Valaskai M (1963). Contribution to the study of vinyl chloride disease. Med lotersi 15:967978. Tamburro CH, Greenberg R' (1981). Effectiveness of Fede rally required medical laboratory screening In the de tection of qfaemlcal liver Injury. Environ Health Persp 41:317-122. Theriault G, Allard P (1981). Cancer mortality of a group of Canadian workers exposed to vinyl chloride monomer. J Oee Med 23:671-676. Thomas LB, Popper H, Berk PD, Selikoff IJ, Falk H (1975). Vinyl-chloride-induced liver disease. From idiopathic portal hypertension (Banti's syndrome) to angiosarcomas. H Engl J Med 292:17-22. Trlbukh SR, Tikhomirova HP, Levina 5V,.Koslov LA (1969). Working conditions and measures for their sanitation in the production and utilization of vinyl chloride plas tics. Gigiena Sanit 10:38-44. Vaxveiler RJ, Stringer W, Wagoner JK, Jones J (1976). Neoplastic risk among workers exposed to vinyl chloride. Ann NT Acad Sci 271:40-48. - Waxveiler FJ, Smith AH, Falk H, Tryoler &A (1981). Excess lung cancer risk in a synthetic chemicals plant. Environ Health Persp 41:159-165. Wilson RE, McCormick WE, Tatum CF, Creech JL (1967). Occupational acroosteolysis. J Am Med Assoc 201:577-581. 4**"' * CHrt 007637 OCCUPATIONAL HAZARDS IN PRODUCTION AND PROCESSING OF STYRENE POLYMERS - EPIDEMIOLOGIC FINDINGS WilliAO J. Nicholson and Dime Tarr Environmental Sciences Laboratory, Mount Sine! School of Medicine of CUNY, New York 10029 INTRODUCTION ' Of the major plastic monomers, styrene is exceeded only by ethylene, propylene and vinyl chloride in terms of produc tion. In 1977, approximately 3 x 1G metric tans were produced is the United States and 7 x 10 metric tons world wide (IAKC, 1979}. Approximately 60% of the monomer pro duced was used in hcmopolymers, largely for the packaging industry. Ocher important uses of styrene are in the pro duction of copolymers with acrylonitrile (SAN) and acryloni trile and butadiene (ASS). Styrene also finds widespread use as a copolymer with butadiene in the production of the synthetic elastomer, styrene-butadiene rubber CSSR}, which forms the basis of approximately SOL of U.S. rubber products. Finally, it is extensively used as a solvent and cross-link ing agent for polyester resins In the fiber reinforced plastic (FRP) industry. Estimates of the number of workers employed in the various industries using styrene-based polymers are given in Table 1 along with typical exposure levels (Tossavaines, 1973}- In additiou to occupational exposure, low-level environmental contamination can occur from combustion of styrene-based products, as the thermal decomposition of polystyrene lends to evolution of the monomer, in contrast to other polymer materials. MORTALITY STUDIES OF STYRENE-EXPOSED WORKERS Studies an the mortality of populations exposed to styrene are fraught with difficulty because of esnfeund- tr CMA 007A3S TbU 1 Occupational *xpo*ur** to (errant PfOegM fircat of iCTTau arorfuodaii Kuzbas of wrlun lmtlirW Typical axpaaura Sam froSucGiM (M.ass.nx) lalafercirf plajilcs frorfnrrton (TXT) hlfwr frneuilat in ,m,m) too to * 70 10,000 30.000 10-30 100.000 5-10 1,,000,000 1-20 1-20 10 - 300 O.fll - 1 Pron ItiiinlMt, 1971 ini exposures to other known carcinogenic materials. Benzene exposures can occur in the production of styrene monomer, as the principal production process utilizes benzene to produce ethylbenzene, vhich, in turn, is dehydrogenated to form styrene. Copolymerlzation invol ves exposures to acrylonitrile end/or butadiene, which are carcinogenic in animals (Buff, 2983). Finally, same af the additives used in styrene products may be carcino genic, as veil as other chemicals^ used in facilities producing styrene or polystyrene. Only three studies - provide data on poxaible humsn carcinogenicity of styrene, each having some of , the - confounding exposures mentioned above. All were of groupa of Individuals employed in monomer production, polymerization or polymer fabrication, where exposures were relatively limited. Ho data exist on the mortality of individuals employed in the TRP industry, where sty rene concentrations were (and are) commonly ten times higher. Table 2 lists the cohort* observed and some of Che characteristics of the three studies. \ As can b* **en from Table 2, the large number of indivi dual* lost to follow-up in the' study by Frentrel-Seyme et *1(1978) severely limits its usefulness. Of those exposed, ?T of the German workers and 71%' of foreign "guest workers" were untraced. Of those traced, 74 had died, 12 from cancer. Only 37 deaths occurred in those with five or more years of exposure. The overall result did not demonstrate excess mortality for any cause of death. However, the limitations of the study are clear. CA 007A3? Tibi* 2 Fcpulstloe sad lollow-ttp chrct rt*clc* *f tbra* smdlaa at leyrsa* expoltd uorltar* Stud* " Aaslyrir cohere Conner* is* Uvavbtr of Ttrctne Foresee atich* of traced snsl*r*<! tots! Ore t *1. HU Siebolsos *c *1. 197S Tracers!-m*m i si. 1ST* 051 03* ezr 390* 3<0 i960 97.* 100.0 93,0Car 19.0?or JiiaiMOM Malta xpdiurt latency Tttti <>> fviiri) ffllldW-UP Ote sc si. 19SO aicholson *c *1 1971 Trnessl-S*ya* tt si. 19 7 i i 1940-1373 3 10 1940--1975 1 00. 1 - 1955-1976 301 10.4 53 1*.S 73 3.7 Sxoocuru *10 FP* 31* <10 <1 pp ppl1 51 * { ) -- Eiclcitid ivr*|i uepaiuTi Current Muuttunt *<e*r lettsliseica at entrala The study of Nicholson et ml -(1973) successfully traced all c 563 men employed in styrene production, polymerization and polymer processing ' who had 5 years of employment on Bay 1, 1960 and were 10 years from onset of wort in a large tT.S. production facility. The basic mortality data are ahown in Table 3 and demonstrate no exceaa mortality from any cauae ~ of death. Analytes according to years from onset of exposure and calendar years of observation did not reveal any pattern of excess mortality. However, because of the limited number of deaths, ghi data can be used only to establish upper limits of risk. For example, the data are only suffici ent to indicate that the SHR for lymphoma or leukemia is less than 280 at the 0.05 level of significance and that of lung cancer,, less than 220, While no excess mortality was identified la the cohort observed, 'the' above publica tion mentioned the existence of 7 deaths from leukemia and 5 of malignancy of the lymphatic system among 444 deaths known to have occurred in the plant workforce. While the ages of death were not available for exact proportionate mortality calculations, the number of lymphomas is in line with expectations, while leukemia appears to be in excess by as much as a factor of two. However, the possibility of high exposures to benzene in the facility during earlier years weakens the likelihood Ar CMfy 007640 r>hi* 3 Zxpactad ui obaaraad crtllt7 arparlaasa* uf 360 lnd3.ndua.la aegls-ad La *37tana production tod rolyaarlsatiqn ftior to 1 Kay IJ33, follovad tarn 7tars tfcar ottaac of axpoauT* Cl il*7 13*0 -- 21 DacMfetr 1)753 Causa of Daadb XxpaecaA Thaarrad sa All etna*a Cdacar Ciaaar *t eat lung laakada Lrwfewa ffth.cr cuctr Itaarc *od circuitcar? die. StiplriEorr Alia**a> Debar eaaiai nf Atari 104.il 11.01 (.)* 9.71 1.23 11. IS 34.33 4.44 21.(1 S3 17 4 1 t 1 32 1 13 7* >1 117 124 SO 73 12 15 33 fro* Hlcholaoa at al. 1771 1i'l of association of any possible excess of leukemia with styrene exposure. The final study of styrene mortality is that of Ott et al (1980) who described the experience of 2,904 individuals with potential exposure to styrene prior to January .1, 1976. Three hundred three deaths occurred, 292 among production and nonprofessional research employees 1 The mortality experience for thia latter group is shown in Table 4 and is compared to the expected deaths calculated from B.S. white male rates and rates from observations on other company employees. As can be seen, the mortality of styrene-exposed individuals compares favorably with each group; the only excess of note being six leukemias compared to 2.9 expected t using B.S. rates and 1.6 from company rates. Lymphomas were also elevated, but not at an 0.05 level of significance. The excess leukemia was further investigated in an analyis of cancer incidence in the styrene-exposed population com pared to that expected from rates of the Third National Cancer Survey. In this analysis, it was found that a signi ficant number of lymphatic leukemias (5 observed vs. 0.26 expected) occurred in individuals who were exposed to sty rene (< 5 ppm), ethylbenzene (< 5 ppm), polystyrene extru sion fumes, and colorants. Indeed, four of the five eases worked in the seme general ares during the period of time, 1947-1948, although their dates of death and other exposures varied widely. Interestingly, there were no deaths of lymphocytic leukemia among 442 individuals exposed to higher CMA 0076 *i Me A Obiarved and expected daatlii >7 cause for tacal ^reduction end losfntittlnil reaeateh employes* (2119 *#a). 13*0--137* Causa* All c*u*a* Hsllfaaai seoplsras lasplrstory Sltweii* one Lyaphatlc ud fceaatopolatic *7>tc* tiupt laukeata l*kU 4Ut slew Cardiovascular disease XotBslicuae respiratory dls. all athar causes Observed deaths 2*2 si id i* c a u 1*3 12 72 Autht ro vtiiti mil** 397.* *1.2 20.1 It. 9 1.3 2.1 11-0 172.* I*-3 10*.7 SMS 17 *7 133 207 72 S3 54 *7 Fra* oet at al. USO lipacted deaths, Caapaay eoaaariiTM 2*7.i - *3.0 23.3 21.2 2.1 ha 3* 13 33 *7 230 l. 13.7 MU 10.0 . 71.1 373 3 101 120 101 concentrations o styrene (5-9 ppm). Because of lack of a definitive exposure-response-relationsflip and the presence of possible confounding exposures, the authors refrained from drawing any conclusions on an etiological relationship. In 1976, nine cases of various types of letikeoi* were identified in two SBR plants and reported to the U.S. National Institute for Occupational Safety and Health (Heinhardt et al, 1978). All occurred after 1971 in a papulation of 5,600 workers. No data were presented on the expected numbers of deaths from leukemia in the group. Concern generated by the findings is the two plants led to reports on the leukemias present in two large ongoing studies of rubber workers. McJiichael et al {1976) reported a relative risk of 6.2 for lymphatic and hematopoietic malignancies among employees producing elastomers, including SBR. However, this was based upon only - 6 cases, 3 leukemias and 3 lymphomas. In a subsequent case control study of the saee plant, a relative risk of 2.4 was found for the same exposure group (Spirtss et al, 1976). The difference in the two values reflect the uncertainties associated with small numbers of eases. A similar investigation by Honson et al (1978) showed an excess of leukemia to be present in calendering, extrusion, tire building and rubberized fabrics. However, the excess wss associated with exposure to solvents and not to styrene. tr CMft 007642 Since 1976, considerable interest has existed in poten* till carcinogenicity of styrene. This bee been heightened by the data on mutagenicity and possible carcinogenicity of styrene and styrene oxide (Huff, 1983). Some suggestive human data are available from studies of polymerization worker* and SSR production facilities. The irony of the situation is that no studies have been conducted of groups exposed to the enormously higher concentrations found in the FRP industry. The data available on styrene carcinogenicity re anch that they can provide no assurance of safety at these higher exposures. CARCINOGENIC RISK FROM COMONOMERS USED VITO STYRENE ' The two principal comonomers used in- styrene-based copolymers have each been shows to be carcinogenic in ani mals- these, scrylonitrile has also been associated with lung cancer in humans. In a group of 1,345 male employees, with potential exposure to acrylonitrile and followed from 1956 through 1976, 8 cases of lung cancer occurred compared with 4.4 expected from company rates CQ'Berg, 1980). Fur ther, there was a correlation of increased risk with inten sity and duration of exposure. Among production workers employed between 19S0 and 1952, 6 lung cancer deaths were observed vs. l.S expected tp < 0.01). A second study of 327 eitployces of a rubber chemicals plant with potential expo sure to acrylonitrile identified 9 deaths of lung cancer compared to 5.9 expected, based on mortality rates for U.S. white males (4.7 based on mortality rates for other rubber workers from the same city) (Delzell and Monson, 1982) The excess was greatest among those who had worked for more than 5 years in the facility. These data, while limited by the small numbers, strongly suggest that acrylonitrile is carci nogenic for humans. - While butadiene has been demonstrably.carcinogenic In animals (Huff, 1983), no data are available from human exposures. ClDTICAL FINDINGS AMONG STYSEXE-E3CPC5ED WORKERS A variety of symptoms have been reported from styrene exposure, dating from the rapid increase in production during World War II. Eye, nose and throat irritation. CMrt 007643