Document gDv9qoYX45mkmNo0jDk76mYEa

2020 Dow Center January 8, 1993 Dow U.S.A "o Ci'ierTM.-..: u'vmoanv '/irtldiin V.r*' ,:an JH67J Michael Marshall Vice President of Environment, Health & Safety B.F. Goodrich 6100 Oak Tree Blvd Clevland, OH 44131 cc: R. B. Uenhart, 2020 Dow Center R. L. McCreedy, 2020 Dow Center J. J. McDade, 2020 Dow Center W. J. Steams, 2020 Dow Center CANCER EPIDEMIOLOGY STUDIES ON VINYL CHLORIDE WORKERS As per our telephone conversation earlier today, I am sending you a copy of a 1988 publication by Sir Richard Doll which presents an authoritative, interpretive summary of the evidence for a link between vinyl chloride exposure and human cancer. A list of studies published since Doll's review is also attached. I cannot verify the completeness of the list, but it should be sufficient to meet your stated need of "getting started". Let me know if I can be of further help. Sincerely yours, Gregory G. Bond, Ph.D., M.P.H., F.A.C.E. Manager, Environmental, Health and Regulatory Affairs Chemicals and Performance Products (517) 636-9063 Attachments t.e e e H S 'Sti .'0 histories were available for all factories except two. Typical exposures" to VC in air were estimated as me-weighted averages by industrial hygienists using everal sources of variable quality. In most factories, ccasional measurements of VC provided the basis for ast typical exposures, supplemented by knowledge of xposure conditions, processes, and technological hanges over time. Systematic measurements taken ince the mid-1970s provided the basis for more recent .xposure assessments, and an indication of the variibility of exposure levels between job titles. In terms of agents to which the workers were ex posed in VCM, PVC, and VCM/PVC production, VC was the main exposure, and virtually the only exposure n many of these factories. Use of butadiene was rare. In PVC processing (one factory in this study), how ever, additional exposures could have included PVC Just, asbestos, and other agents. All the job-exposure matrices referred only to VC exposure in air. Each job-exposure matrix was checked and validated by two independent industrial hygienists, who were able to provide, prior to the statistical analysis, an in dex for the ranked level of exposure (low: <50 ppm; intermediate: 50--449 ppm; and high: > 500 ppm) in which the classification of the subjects was based on highest level to which the workers were potentialIxposed, specific to their jobs and the years in which 'ey worked, according to levels of VC recorded in the iob-exposure matrices. The same information, that is, job histories and jobexposure matrices, was used to calculate cumulative exposure in parts per million-years (exposure level from the job-exposure matrices multiplied by duration of employment) for the subjects. In some of the analyses for cancer of the liver an estimated job-exposure ma trix was used for the four factories in the United King dom which were unable to provide their own matrices. The estimated job-exposure matrix was based on the matrices provided for the other factories in the United Kingdom. This matrix was checked with the industrial hygienist from the United Kingdom. In the following text, it is clearly noted if the analyses under discus sion include this estimated matrix. For liver cancer only, the Poisson regression analy sis was performed to assess the significance of several variables simultaneously. This analysis used observed deaths and the person-years distribution for crossclassified categories of temporal and exposure varia bles and performed an internal comparison with the base-line categories (ie, within the cohort only) (17, 18). Mortality results se-specific mortality for the total cohort is preted in table 3. A statistically significant deficit for all-cause mortality was apparent (1438 deaths observed versus 1636.4 expected, SMR 88, 95 Cl 83--93). The following four main causes of death contributed to this deficit: (i) diseases of the circulatory system, (ii) diseases of the respiratory system, (iii) accidents, poisonings and violence, and (iv) other known causes. Ftor all malignant neoplasms, the SMR was 104 (95 Vo Cl 95--114), with 445 deaths observed. How ever, there were two causes of death, both cancers, which showed statistically significant excesses. They were cancer of the liver with 24 observed deaths and 8.4 expected (SMR 286, 95 Vo Cl 183--425) and can cer of an unspecified site with 24 observed deaths and 12.9 expected (SMR 187, 95 Vo Cl 120--278). Increases which were not statistically significant were apparent for bladder cancer (21 deaths observed, SMR 146, 95 Vo Cl 91--224), malignant melanoma (7 deaths ob served, SMR 163, 95 Vo Cl 65--335), and lymphosar coma (7 deaths observed, SMR 170, 95 Vo Cl 69-- 351). In this report the results are not given by process. The statistically significant excess of liver cancer evi dent in the total cohort was mainly due to the excess in VCM/PVC production (19 deaths observed, SMR 311, 95 Vo Cl 187--486). Twenty-three sites were selected for analysis ac cording to the years since First exposure, and the results for 10 of these sites are presented in table 4. Apart from liver cancer, which will be discussed in detail, there were no noteworthy patterns in risk according to this variable. Four sites were investigated in detail for relation ships with temporal and exposure variables. Of the two sites with excess risk for the total cohort, cancer of an unspecified site was not analyzed further due to the diversity of cancers found in this category; they are however discussed in a descriptive fashion in the text. Liver cancer, lung cancer, brain cancer, and lym phosarcoma were chosen a priori for further analysis. Excesses of bladder cancer in PVC production (in the United Kingdom) and of melanoma (in Norway) were not investigated further for the total cohort since they were confined to one country. Liver cancer No liver cancer deaths occurred before 15 years since first exposure, after which the SMR was 483 (95 Vo Cl 208--951) for 15--19 years since first exposure, and it did not vary greatly from this level thereafter, the value always being statistically significant. In table 4, the pattern by years since first exposure is seen in 10-year groups. When only those with 15 years since first exposure or more were included in the analysis (15-year latency), the overall SMR for liver cancer was 445 (95 Vo Cl 285--663). Table 5 shows the SMR values for liver cancer ac cording to the four exposure variables, without and with a 15-year latency analysis. According to job title, dichotomous as ever autoclave worker (suspected a priori as the highest risk job) versus never an autoclave worker, very high risk was experienced by those who R&S149992 161 Table 4. Mortality by time since first exposure for selected sites. (0 = observed number of deaths. E = expected number ot deaths, SMR = standardized mortality ratio, 95 % Cl = 95 % confidence interval) Years since first exposure 1-9 1 0-19 20-29 30 O E SMR 95 % Cl O E SMR 95 V* Ct O E SMR 95 % Cl O E SMR 95 % Cl O Total a SMR 95 % Cl au causes (000--999) All malignant neoplasms (140--207) Liver and intrahepatfc bite ducts (155) Trachea, bronchus and lung (162) Bladder (188) Brain (191) Lympho sarcoma (200) Circulatory system (390-458) Respiratory system (460--5191 Chronic liver disease, cirrhosis (571) 207 313.9 66 57--76 529 582 5 91 83-99 471 502.6 94 85--103 231 237.4 97 85-- 111 1438 1636.4 88 83-93 56 73.5 76 58-99 173 155.6 111 95-129 151 138 6 109 92--128 65 60.2 108 83--138 445 427,8 104 95--114 1 5 0 0-245 8 3.2 253 109--499 11 2 8 388 194--694 5 0.9 561 182--1310 24 84 266 183--425 16 22.7 71 40--114 50 53 9 93 69-122 60 50.4 119 91-153 18 21.3 84 50--133 144 148.3 97 82--114 2 1.8 109 13-395 7 4 8 146 59--301 6 5 1 117 43-255 6 2.6 231 85-504 21 143 146 91--224 2 3.4 59 7-213 6 53 113 42-247 2 3.4 59 7--212 4 1 0 407 111 -- 1041 14 13.1 107 59-180 3 1.3 225 46--657 4 1.6 258 70-662 1 0 0 0--381 0.3 0 0--1430 7 4.1 170 69-351 67 108,0 62 48--79 223 246.2 91 79-103 2Q9 236.9 88 77-101 123 121 7 101 84--121 622 7128 87 81 ^94 8 19.9 40 17-79 30 45 0 67 45-95 47 47,5 99 73-132 23 28 6 80 51-121 108 140 9 77 63--93 2 8.4 24 3-87 17 17,0 too 58--160 14 122 115 63-193 2 2.1 95 11--342 35 39.7 88 61 -- 123 i Coos ot the International Classification of Diseases (eighth revisionl in parentheses, Table 5. Mortality data for liver cancer according to the exposure variables* (0 = observed number of deaths, SMR = standardized mortality ratio, 95 % Cl = confidence interval) Exposure variable 15 years of latency* 0* SMR 95 % Cl SMR 95 % Cl Job title Ever autoclave worker Never autoclave worker6 Duration of employment (years) 1--9 10-14 15-19 20--24 >25 Ranked level of exposure (ppm) Low (< 50) Intermediate (50--499) High (a 500) Unknown Cumulative exposure (ppm-years) 0--1999 2000--5999 fiOOO--9999 > 10 000 Unknown Total 11 13 4 5 4 6 5 3 (4) 3 (7) 12 (12) 6 (D 4 (9) 4 (4) 4 (7) 3 (3) 9 (1) 24 896 181 94 327 310 714 1111 119 161 567 317 99 351 800 1429 357 286 447--1603 97-130 ' 26--239 106-763 84-794 262-1555 361-2593 25-347 33-471 293-991 117-691 27-254 96-898 218-2048 295-4175 163-678 183-425 1358 284 678-2430 151-485 205 602 310 714 1111 56-525 196--1406 84--794 262--1555 361--2593 227 (244) 250 (551) 719 (719) 486 (125) 47--664 (67-625) 52-731 (222-1136) 371 -- 1255 (371-1255) 182-1079 (3-697) 191 460 851 1667 536 (348) (400) (1429) (1667) (100) 52--490 (159-662) 125--1177 (109--1024) 232-2179 (574-2943) 344--4871 (344 -- 4871) 245--1017 (2-557) 445 285-663 * The values in parentheses were determineo m analyses including the estimated |Ob-exposure matrices. J In Norway, the longest-held |Ob was used In Sweden ioD rotation was Dracticed. and no one was classified as an autoclave worker 163 Table 8. Cases of liver cancer on the basis of cancer incidence and other information (not Included as liver cancer deaths in tne mortality analysis). (ICO = International Classification of Diseases) Sut>|ect Age at hire (years) Year of hire Dura- Length non of of exposure latency (years) (years) Year of death Cause4 ICO revision Site of incidence or other information l 24 1951 19 19 1970 197 8 8 Cancer incidence: 1CD 155.0 2 45 1961 6 a 1970 197 8 8 Death certificate: carcinoma of liver 3 28 1966 8 8 1974 197 8 B Death certificate: carcinoma of liver, angiosarcoma, hepatic cirrhosis 4 34 1965 10 12 1977 197 8 8 Death certificate: hepatic failure due to liver cancer 5 18 1965 7 ? Emigrated -- Cancer incidence: ICO 155-0 6 36 1957 7 7 1964 199.0 Best evidence (clinical): ICO 155.0 7 50 1954 n 11 1965 199.0 8 Best evidence (clinical). ICO 155.0 8 54 1969 9 10 1980 199.0 9 Cancer incidence: ICO 155.2 9 29 I94t 31 31 1972 157.9 8 Cancer incidence: ICO 155.0 10 32 1953 20 22 1975 159.0 8 Best evidence (pathology): ICO 155.0 tt 43 1944 22 23 1972 227.0 3 Death certificate: hemangioendothelioma of liver, natural causes 12 42 1963 8 17 1980 571 5 9 Death certificate: hepatic failure due to hepatic fibrosis ' According to the ICO. Angio sarcoma of the liver Histoiogi* Total cal confir cumulative mation by exposure pathology (ppm-years) Yes Unknown Yes No 8 203 Yea Yes 1 749 Unknown No Unknown Unknown Unknown Yes Yes Yes No Yes No No No Yes Yes Yea 28 156 7 673 7114 Yes Yes 2 658 liver, age at first exposure and calendar period of ex* posure did not. Very clear exposure-response relation ships were evident between the cumulative exposure to VC and the risk of liver cancer and angiosarcoma of the liver. Finally, an effect of misclassification was demonstrated when the estimated job-exposure matrix was included, although the effect was minimal for the angiosarcoma results. Lung cancer, brain cancer and lymphosarcoma The SMR for trachea, bronchus, and lung cancer was 97 (95 Cl 82--114) for the total cohort, on the basis of 144 observed deaths. The SMR values did not show any remarkable association with a particular process, and although no pattern was evident for years since first exposure, there was a statistically significant in crease at 25--29 years since first exposure on the ba sis of 33 observed deaths (SMR 147, 95 Cl 101 -- 207), mainly from an excess in VCM production in this time period (5 deaths observed, SMR 486, 95 % Cl 158-- 1134). Calendar period at exit and at hire did not reveal any consistent pattern. Fourteen deaths from brain cancer occurred in the cohort, and, although the overall SMR was not in creased (SMR 107, 95 ro Cl 59--180), there was a statistically significant excess at >30 years since first exposure on the basis of four observed deaths (SMR 407, 95 <ro Cl 111 -- 1041) (table 4). The excess was con fined to the calendar period of hire of 1945--1954 and was the most evident for VC.M/PVC production. Ana lyses by calendar period of exit, age at hire, and age at exit did not reveal anv patterns of risk for brain can cer mortality. Table 9. Maximum likelihood estimates for the final model with cumulative exposure and years since first employment for the deaths from angiosarcoma of the liver (N = 22). (95 % Cl = 95 % confidence interval) Variable Relative risk 95 % Cl Cumulative exposure (ppm-years) <2000 2000--5999 6000--9999 a 10 000 - Years since first employment 0-19 20-24 25 1.0 6 S 1.1--41.7 24.7 4.1 -- 150.1 45.4 7.3--281.1 1.0 4.7 1.0--22.8 6.2 1.4--29.0 Table 10, Absolute risk of angiosarcoma of the liver per 100 000. Years since first employment 0--19 20--24 = 25 Cumulative exposure (ppm-years) ------------------------------------- ------------------<2000 2000 -- 5999 6000 -9999 s 10 000 1 0 6.8 4 7 32 0 6 2 42.2 24 4 115.6 152 3 44 8 212 S 280.0 A detailed analysis of the seven deaths from lym phosarcoma showed no pattern in the SMR values ac cording to years since first exposure. All seven deaths occurred in VCM/PVC production, and for this process alone there was no excess apparent by calen dar period of hire or exit or age at hire or exit. 165 Cancer of unspecified site Twenty-four deaths in the cohort were classified as malignant neoplasms of unspecified sites (ICD 199), giving a statistically significant excess (24 deaths ob served, SMR 187, 95 % Cl 120--278). Additional in formation, such as cancer incidence data, was avail able for 20 of the 24 deaths. Three subjects had liver cancer and were therefore included in table 8. It was unknown, however, whether they had angiosarcoma of the liver since no histological information was avail able. No other clear excess of a particular cancer among those classified as cancer deaths of an unspeci fied site was apparent. Cancer incidence results For the 2643 subjects from the four factories in Nor way and Sweden included in the cancer incidence anal ysis, the total number of cancers observed was 127 (SIR 107, 95 9b Cl 89--127), and their distribution by site is shown in table 13 (reported for one or more observed deaths). The only statistically significant excess was for liver cancer, on the basis of seven observed cases (SIR 303, 95 9o Cl 122--623). Suggestive increases were found for stomach cancer (13 cases observed, SIR 150, 95 % Cl 80--256), lung cancer (22 cases observed, SIR 152, 95 9o Cl 95--230), melanoma (8 cases ob served, SIR 184, 95 9o Cl 79--362), and brain cancer (8 cases observed, SIR 159, 95 9o Cl, 68--312). Although the lung cancer increase was not statisti cally significant, it was investigated in more detail as other studies have suggested increased risk for lung cancer. There was no excess according to process type or category of years since first exposure. A slight ex cess was suggested for <15 years of employment, while the SIR values were close to 100 for 15--19 and >20 years of employment. For the ranked level of exposure, the risks in the high and low categories were virtually identical. According to cumulative exposure, no ex posure-response was apparent (results not presented in tabular form). Without statistical significance and with little apparent relationship to VC exposure, some indication of increased lung cancer risk remained for one PVC-processing plant and one Norwegian VCM/ PVC production plant. The national investigator for the PVC-processing plant attributed the excess to ex posure to asbestos, which was utilized in the process (7), while the excess remained unexplained in Norway. Discussion This collaborative study was carried out with the main purpose of analyzing exposure-response relationships between exposure to VC and liver cancer and inves tigating whether exposure to VC could increase can cer risk for sites other than the liver. The results confirmed the association between ex posure to VC and liver cancer. The excess of liver can cer mortality was associated with duration of employ ment, and a clear association with ranked level of ex posure was found. The results were strengthened by the regression analyses, which indicated that the risk of liver cancer depended on cumulative exposure and years since first exposure. Twenty-two subjects had histologically confirmed angiosarcoma of the liver, and the regression analyses demonstrated that the risk was mostly influenced by cumulative exposure to VC. The relative risks were higher at each level of cumulative exposure than those for all liver cancer deaths, but it must be remembered that the same 16 angiosarcoma deaths were included in both analyses. The approximate incidence rate of angiosarcoma of the liver in Norway, for example, based on 1953--1988 data, was 1 in 10 million per year (personal communication from A Andersen, 1989). Others have estimated the annual incidence at 1 to 2 in 10 million (20, 21). Given 222 746 person-years at risk accumulated by this cohort, with an annual incidence of angiosarcoma of the liver of 2 per 10 million in the general popula tion, the overall expected figure for the cohort would be 0.045. The rarity of this tumor supports the use of internal comparisons to assess the significance of ex posure variables. Table 13. Cancer incidence, based on data tor four factories in Norway and Sweden, by detailed cause. (O = observed num ber of cases, E = expected number of cases, SIR = standard ized incidence ratio, 95 % Cl =95 % confidence interval) Cancer site* ' O E SIR 95 % CIS Buccal cavity and pharynx (140--148) Stomach (151) Intestine, except rectum (152-153) Rectum (154) Liver and mtrahepatic bile ducts (155) Pancreas (157) Larynx (161) Trachea, bronchus and lung (162) Melanoma of skin (190) Prostate (177) Testis (178) Bladder (181) Kidney 1180) Brain (193) Thyroid (194) Lympnosarcoma and other lymphoma (200, 202) Multiple myeloma (203) Other malignant neoplasms aii malignant neoolasms i UO--205) 5 4 2 119 39-277 13 87 150 80-256 8 89 89 39--176 2 59 34 4--123 7 23 303 122--623 3 4 3 70 14--203 2 1 6 122 15--441 22 14 5 152 95-230 8 4 4 164 79--362 16 180 89 51 -- 144' 1 2 2 45 1-252 7 7 9 88 36-182 4 5 4 74 20-188 8 5 1 159 68--312 3 0 9 327 67-955 1 36 28 1 -- 154 1 1 9 53 1-297 16 11 9 135 77-219 127 119 0 107 89-127 Code of the International Classification of Diseases in parentheses Based on a Poisson oisfriOution 167 Health, Tokvo: Japan Organizing Commiuee, 1969: 7 Barbin A, Uartsch H. Mutacemc and promutagemc properties ol ON A adducts formed bv vinvl chloride metabolites. In: Singer B, Bartsch H. ed. The role ot cyclic nucleic acid adducts in carcinogenesis and muta genesis. Lyon: International Agency tor Research tor Cancer. 1986:345--358. UARC scientific publication; no 70.) Jones RD, Smith DM, Thomas PG. A mortality study ot \ms I chloride monomer workers employed In the U nited Kingdom in 1940--1974. Scand J Work Environ Health 1988;14:153-60, Wu W. Steenland K, Brown D, Wells V, Jones J, Schulte P. Halperin W. Cohort and case-control analyses of workers exposed to vinvl chloride: an update. J Qccup Med 1989:31:518--23. Hagmar L, Akesson B, Nielson J, et al. Mortality and Lancer morbidity m workers exposed to low levels of sinyl chloride monomer at a polyvinyl chloride process ing plant. Am j Ind Med 1990:17:553--66. Doll R. Effects ol exposure to vinyl chloride: an assess ment ot the evidence. Scand J Work Environ Health, 1988:14:61--78. L'Abbe KA, Ferro G, Winkelmann R, Saracci R, Simonato L. Mortality and cancer incidence results of the European muiti-cenirtc cohort study of workers em ployed in the vinyl chloride industry. Lyon: International Agency for Research on Cancer, 1989. (IARC internal report: no 89/007.) Byren D, Engholm G, Englund A, Westerholm P. Mor tality and cancer morbidity in a group of Swedish VCM and PVC production workers. Environ Health Perspect 1976:17:167--70. Molina G, Homberg B, Elofsson S, Holmlund L, Maasing R, Westerholm P. Mortality and cancer rates among workers in the Swedish PVC processing indus try. Environ Health Perspect 1981;41:145-51. 12. Storetsedt Heldaas S, Langard SL. Andersen A. Inci dence ot cancer among \ myl chloride and poiwmvl cltlo, ride workers. Br J Ind Med 1984:41:25--40. 13. Jones JH. Worker exposure to vinyl chloride and po- ly(vmvlchloride). Environ Health Perspect 1981:41: 129--36. 14. Belli S. Bertazzi PA, Comba P, et al. A cohort study on vinyl chloride manufacturers in Italv: studv design and preliminary results. Cancer Lett 1987;35:253-61. 15. Pirastu R. Comba P. Reggiant A. Foa V, Masina A, Maltoni C. Mortality from liver disease among Italian vinvl chloride monomer/polvvmyl chloride manufac turers. Am J Ind Med 1990:17:155--61. 16. Coleman MP. Hermon C, Douglas A. Person-years (PVRS); a Fortran program tor cohort study analvsis. Lyon: International Agencv for Research on Cancer. 1989. (IARC internal report; no 89/006.) 17. Breslow NE. Dav NE. Statistical methods in cancer Re search: \ol II (The design and analvsis of cohort studies). Lyon: International Agencv tor Research on Cancer, 1987. (IARC scientific publications; no 82.) 18. Baker RJ. Glim 3.77: Reference manual. Oxford: Nu merical Algorithms Group, 1985. 19. Creech JL. Johnson MN. Angiosarcoma of liver in the manufacture of polvvmvl chloride. J Occup Med 1974; 16:150--1. 20. Brady J. Liberaiore F, Harper P. et al. Angiosarcoma ot the h'er: an epidemiologic survey. J Natl Cancer Inst 19^7:59:1383--5. 21. Byren D. Holmberg B. Two possible cases of angiosar coma of the liver in a group of Swedish vinyl chloride workers. Ann NY Acad Set 1975:246:249--50. Received for puolicalion: 18 October 1990 169 American Journal of Industrial Medicine 17:155-161 (1990) Mortality From Liver Disease Among Italian Vinyl Chloride Monomer/Polyvinyl Chloride Manufacturers R. Pirastu, msc, P. Comba, dsc, A. Reggiani, mo, V. Foa, md, A. Masina, md, and C. Maltoni, mo The possible association in humans between nonangiosarcoma primary liver tumors (PLC-non-A), particularly hepatocellular carcinoma (HCC). and exposure to vinyl chlo ride monomer (VCM) is supported by both experimental and human data. This article presents a review of the information regarding 253 deaths that occurred in seven plants manufacturing VCM/PVC and one plant extruding PVC. The retrieval of clinical and pathological data, in addition to the information from death certificate, is referred to as "best evidence'' (BE). BE has been carried out for 63 deaths. Atotal of 14 primary liver cancer (PLC) were detected: seven were angiosarcoma (PLC-A). and two of the re maining seven were hepatocellular carcinoma (HCC). In our series of 14 PLC cases, there was no significant difference between PLC-A and PLC-non-A as to length of exposure and latency. There was no noticeable difference in terms of job title between ASL and non-ASL cases. The list of longest held jobs shows the presence of various job titles, different from autoclave cleaner, for primary liver cancer. PLC-A and PLCnon-A. In conclusion, our observations show that VCM may have a broader carcino genicity action on the liver and that exposure lower than that occurring in autoclave cleaning can cause primary liver cancer, both angiosarcoma and nonangiosarcoma. Key words: primary liver cancer. VCM/PVC, occupational exposure, angiosarcoma liver (ASL) INTRODUCTION Vinyl chloride monomer (VCM) has been shown to be a muitipotential carcin ogen in rodents, i.e., capable of causing a variety of tumors, among which angiosarcoma of the liver (ASL) [Viola et al., 1971; Popper et al.. 1977; IARC, 1987; Maltoni et al.. 1984; Doll. 1988; Maltoni and Cotti, 1988]. In humans, a causal relationship has been found between occupational exposure to VCM and ASL. In exposed workers, the association between the chemical and other malignant tumors--namely, brain neoplasms, lung carcinomas, and malignant University of Rome. "La Saptenza ' Department ot Animal and Human Biology (R.P.). and Istituto Supenore di Sanita (P.C . A.R.). Rome. Italy Institute ot Occupational Health "L Devoio," University ol Milan (V.F ), Milan. I (al v Institute of Oncology F, Addaru Bologna, Bologna. Italy (A M.. C.M.) Address reprint requests to Dr. Comba. Istituto Superiore di Sanaa'. Viale Regina Elena 299. (X)I61 Rome, Italy. Accepted for publication August 4. 1989 c; 1990 Wilev-Liss, Inc. 156 Pirastu et al. primary liver cancer (PLC) other than angiosarcoma--has been pointed out by some investigations, but it is still debated [IARC, 1987]. The possible association in humans between nonangiosarcoma liver tumors (PLC-non-A), particularly hepatocellular carcinoma (HCC), and exposure to VCM is supported by three orders of data, dealing with: 1) the experimental results of long term carcinogenicity bioassays in rats (Maltoni et al., 1984]; 2) the analogies between VCM and Thorotrast carcinogenesis [Popper et al., 1977]; and 3) the observation of cases of HCC among workers exposed to VCM [Gokel et al., 1976; Koischwitz et al., 1981; Evans et al., 1983; Dietz et al., 1985], In an experimental study, pregnant female rats were exposed by inhalation to 2,500 ppm VCM for 76 weeks (group 1). Concurrently, one group of male and female offspring was exposed in the same way for 76 weeks (group II) and another for 15 weeks (group III). Adequate control groups of breeders (group IV) and offspring (group V) were available. The incidence of ASL and HCC in the five groups was. respectively, as follows: 50.9% and 9.4% in group I; 65.1% and 56.5% in group II; 44.4% and 72.6% in group III, 0 and 0 in group IV; 0 and 0.3% in group V. No sharp differences were found in the incidence of these two neoplasms between male and female offspring [Maltoni et al.. 1984], Both VCM and Thorotrast have been shown to induce ASL and HCC in ex perimental rodents and ASL in humans; moreover, Thorotrast exposure has been documented in a variety of liver neoplasms other than ASL in humans [Keplinger et al., 1975; Maltoni and Lefemine, 1975; Maltoni, 1977; Maltoni et al., 1984; Com mission of the European Communities, 1984; Maltoni and Cotti. 1988]. In a com prehensive epidemiological investigation on 5,000 Thorotrast-injected patients and on 5,000 controls, a total of 152 PLC occurred among the exposed, whereas no cases were detected in controls. These neoplasms included both ASL and HCC [Commis sion of the European Communities, 1984], In addition, steroids have been suggested as causative agents of both hepato cellular carcinoma and liver angiosarcoma. Women taking oral contraceptive steroids have a small increased risk for developing hepatocellular carcinoma [Klatsin. 1977], The use of androgenic anabolic steroids has been implicated in the development of hepatocellular carcinoma (Johnson et al., 1972] and angiosarcoma (Falk et al., 1979]. In both ASL and HCC, the common accompanying precursor stage is a lesion con sisting of varying degrees of mixed hyperplasia of hepatocytes and sinusoidal cells and sinusoidal dilatation (Popper et al., 1977]. Few cases of HCC in workers exposed to VCM have been reported in the literature [Gokel et al., 1976; Koinschwitz et al., 1981; Evans et al., 1983; Dietz et al., 1985], In the Federal Republic of Germany, Dietz et al. [19851 reported three cases of HCC in VCM-exposed workers with no history of exposure to other known liver carcinogens (such as Thorotrast and alcohol) and no signs of hepatitis B infec tion. The present article analyzes data on the mortality from various liver diseases and provides some preliminary observations that strengthens the evidence that VCM can cause HCC in humans. MATERIALS AND METHODS We reviewed information regarding 253 deaths occurring in seven plants man ufacturing VCM/PVC and one plant extruding PVC. The seven plants represent all Mortality in VCM/PVC Manufacturers 157 TABLE I. Description of "Best Evidence" Data and Agreement With Death Certificate Information for Workers in Three Plants in Italy Manufacturing VCM/PVC Best evidence data Degree ot No. of Clinical Pathological agreements City Plant # deaths Total data data DC/BE Ravenna Rosignano Ferrara 1 3 17 10 10 II 8 8 56 45 45 7 80% (8/10) 3 100% (8/8) 25 71% (32/45) Italian production facilities, with the exception of two industries in Sardinia, where the investigation is still ongoing. The results of the study, in terms of standardized mortality ratios, have been published for three cohorts [Belli et al., 19871 and is ongoing in the remaining six plants. Causes of death, noted on death certificates and from hospital records (when available), age at death, duration of exposure, latency, and occupational history are reported for each death. Retrieval of clinical and pathological data, in addition to information from the death certificate, is referred to as "best evidence'' (BE). BE was compiled for deaths that occurred in three of the seven plants. The type of additional data and its fre quency are given in Table 1. '"Liver disease" cases comprised 39 of 253 deaths. Deaths from liver disease were defined on the basis of cither death certificate (coded according to the Interna tional Classification of Disease, ICD) or BE-ICD (i.e., ICD after considering the BE), when available. Two categories were identified: 1. Primary liver cancer: ICD or BE-ICD = 155.0, which includes ASL and non-ASL cases (VIII and IX ICD Revision). 2. Liver cirrhosis and other chronic liver disease: ICD or BE-ICD 570 = acute and subacute necrosis of the liver: 571.5 = cirrhosis of the liver without mention of alcohol: 571.9 = unspecified chronic liver disease without mention of alcohol; 573.0 = chronic passive congestion of the liver (VIII and IX ICD Revision). The seven ascertained Italian cases of ASL in VCM/PVC manufacturers have been reported previously in the literature [Maltom. 1974; Maltoni and Rondinella. 1980; Chiappino et al., 1982: Maltoni et al.. 1984; Brugnami et al., 1988: Chiozzini et al., 19881, and six are included in the Register of ASL cases maintained by the Imperial Chemical Industries. Two cases occurred in PVC extruders, four in poly merization workers, one in a monomer worker. All seven cases are included in the present contribution. RESULTS Table II describes the 39 liver-disease deaths in terms of cause ol death (ICD and BE-ICD), age at death, duration of exposure, latency, and longest-held |ob. The letter y\ labels histologically confirmed angiosarcoma cases. We detected a total of 14 primary liver cancers (PLC): seven were angiosarcoma (PLC-A). two of the remaining seven were HCC: for five subjects, histological data were not available, but their clinical history was not compatible with the diagnosis of 158 Pirastu et al. TABLE II. Occupations in Deaths Due to "Liver Disease" in the Italian V.C. Study* Duration of Years since Case # Age al death exposure first exposure Prevalent |ob Best ICD evidence 1 62 11 11 PVC loader 571.9 T 42 20 20 PVC loader 199.1 155.0 3 55 5 5 PVC bagger 571.9 4 74 16 29 Control 571.5 5 48 2 19 Autoclave cleaner 571.9 6 55 i 14 Dryer area helper 571.9 7 60 3 4 Reactor area operator 156,0 157.9 8 62 11 11 Dryer area operator 199.0 155.0 9 62 28 31 Maintenance 155.0 to 68 9 25 Dryer area helper 426.0 573.0 It 46 23 23 Reactor area helper 571.9 12 44 II 7 PVC bagger 199.0 155.0 13 61 21 Autoclave operator 250.0 571.9 14 55 22 >1 Autoclave cleaner 159.0 155.0 A 15 55 29 32 Autoclave cieaner 196.9 155.0 16 47 17 18 Autoclave operator 155.0 17 48 11 23 Autoclave cleaner 571.9 18 68 19 28 Autoclave operator 571.5 19 52 19 23 Maintenance 571.9 20 66 25 25 Clerk 571,9 21 45 6 7 Compounds 571.5 22 43 14 15 Maintenance 155.0 155.0A 23 50 4 ->2 Compounds 571.5 24 50 20 21 Maintenance 571.5 25 50 25 26 Autoclave operator 571.5 155.0 A 26 27 __28 46 37 44 4 2 22 16 PVC Extruder 155.0 A 2 PVC Extruder 155.0 A a Reactor area operator 155.0 155.0 A 29 41 18 19 Reactor area operator 571.0 30 59 11 26 Reactor area helper 571,9 31 49 11 12 Autoclave operator 571.9 \1 65 10 15 Monomer operator 573.0 33 50 9 23 Autoclave operator 571.9 34 43 9 9 Reactor area helper 571 9 35 55 14 19 PVC bagger 571.9 36 61 V*> 28 PVC bagger 155.0 A 37 56 1 21 PVC bagger 571.9 38 61 3 19 Autoclave operator 571.9 39 62 9 20 PVC bagger 155.0 * A = Liver angiosarcoma. ASL. Twenty of 39 "liver disease deaths" were coded as liver cirrhosis, and the remaining five were unspecified chronic liver disease without mention of alcohol (ICD 571.9) and other disorders of the liver (ICD 573.0). The characteristics of the 14 primary liver cancer cases are presented in Table III. Mean age at death was 48 years for angiosarcoma (PLC-A). 53 years for primary nonangiosarcoma liver cancer (PLC-non-A), and 60 for liver cirrhosis and other chronic liver diseases. Mean duration of exposure was 16 years for PLC-A. 18 years tor PLC-non-A, and 12 years for the other pathological entities. Mean latency was. respectively. 19. 20, and 20 years. Mortality in VCM/PVC Manufacturers 159 TABLE III. Demographics and Job Titles Among Primary Liver Cancer Cases in the Mortality Study of Italian Manufacturers of VCM/PVC Liver angiosarcoma Primary liver cancer No. ot cases Mean duration ot exposure (yearsj Mean latency (vears) Mean age at first exposure Mean age at death Job titles Case no. 1 Case no 2 Case no .5 Case no. 4 Case no 5 Case no. 6 Case no, 7 7 16 19 29.9 48 Autoclave cleaner Autoclave operator Maintenance Autoclave operator Controller PVC extruder PVC extruder Reactor area operator Warehouse and transport PVC bagger 7 18 20 33.6 53 Autoclave cleaner Reactor area helper Maintenance Autoclave operator Maintenance PVC loader Drver area operator PVC bauger PVC bagger The list of longest held jobs in Table III shows the presence of job titles such as PVC loader, dryer operator, and maintenance worker for primary liver cancer both PLC-A and PLC-non-A. DISCUSSION Experimental and epidemiological data from the international literature suggest that VCM is a multipotentiai carcinogen; the analogies between Thorotrast and VCM carcinogenesis, case repons of HCC in VCM exposed workers, and our present results strongly suggest that VCM can cause PLC-non-A. panicularly HCC. In Italy, only monalitv data are available on a routine basis; lor the entire country', there are only four tumor registries presenting incidence data with the details of the histotvpe. As a consequence, it is not possible to carry out a proportional mortality analysis or any other measure of the expected number of PLC and HCC cases. In the present study, the histotvpe was ascertained for 64% of primary liver cancer (PLC) cases; in the four tumor registries named above, the percentages were 8%. 12%. 23%. and 50%, Therefore, in the present investigation, the histotvpe of PLC was ascertained in greater detail than in any tumor registry. The percentages of hepatocellular carcinoma (of all PLC) were 14% in this study and 2%. 8%. 14%. and 26%. respectively, in the tumor registries. There is no sound reason to establish a differential and biased histologic ascertainment for ASL and HCC. In our study, ASL appeared to occur at younger age; mean age at death was 48` years for PLC-A cases and 53 for PLC-non-A. These results are comparable with U.S. data | Wong et al.. 1986] reporting 54 years and 61 years as mean age at death, respectively, for PLC-A and PLC-non-A. A report on the Register of ASL cases among VCM-exposed workers, notified to the Imperial Chemical Industries since 160 Pirastu et al. 1974, gives 53 years as the mean age at death tor the 138 ASL cases recorded worldwide (Forman et al., 1985], Mean age at First exposure was 29.9 for PLC-A and 33.6 for PLC-non-A. The same observation of younger age at first exposure tor PLC-A was made in the last (J.S. study: 27.2 and 36.6 (Wong et al.. 1986). The lower age at death for ASL cases both in the U.S. and Italian series is in agreement with the hypothesis of an earlier occurrence of ASL in the frame of liver carcino genesis, but it may also be due to an early age at the time of exposure. In our series of 14 PLC cases, there was no significant difference in length of exposure and latency between PLC-A and PLC-non-A. There was no noticeable difference in terms of job title between ASL and non-ASL cases. Vinyl chloride has been shown to induce ASL at very high exposure levels, as shown by the fact that most ASL cases worldwide occurred among autoclave clean ers. In the present study, the job title of the cases of both PLC-A and PLC-non-A show that these tumors may develop in VCM-exposed workers with jobs other than autoclave cleaner. In conclusion, our observations indicate that VCM may have a broader carci nogenicity spectrum on the liver than known bclorc and that exposure lower than that occurring in autoclave cleaning can cause primary liver cancer, both angiosarcoma and nonangiosarcoma. REFERENCES Belli S. Bertazzi PA, Comba P. Foa V. Maltom C. Masina A. Pirastu R. Reggiam A. Vigotti MA (1987): A cohort mortality study on vinyl chloride manufacturers in Italy: studv design and preliminary results. Cancer Lett 35.253-261, Brugnami G, Abbritti G. Morelli A, Maltom C, Soltnas A. Pelliccioli G (I98X): Liver angiosarcoma in a vinyl chloride polymerization plant worker. Report of the seventh Italian ease. Med Lavoro 79:34-41. ChiappinoG. Bertazzi PA. Barom M. Masim T 119X2)' Hepatic angiosarcoma from vtnvl chloride report of a new Italian case Med Lavoro 6 555-563. Chiozzini G, Saggtoro A. Pallini P. Valsecchi M. Magarotto GC. Scattolin A, Nardin M. Polacco A. Oliboni E (1988): Angiosarcoma epalico da cloruro di vinile. Med Lavoro 79 24-U Commission ot the European Communities iCEC) (19X4): "The German Hiorotrast Studv " Directorate General Science. Research and Development. EUR 9504 EN Dietz A. Langbein G. Permatter W 119X5)' Das Vmychloride-induziertc hepatoccllulare kar/inorn Klin Wochenschr 63-325 -331. Doll R (19X8): Efleets ol exposure to vinyl chloride. An assessment of the evidence. Seand J Work Environ Health 14 61-78 Evans DMD. Jones WW. Rung 1TM (19X3): Angiosarcoma and hepatocellular carcinoma in vinyl chloride workers. Histopathology 7,377-388. Falk H. Thomas LB, Popper H. Ishak KG (1979)' Hepatic angiosarcoma associated with androgemcanabolic steroids Lancet 21120-1123. Forman D. Bennet B. Stafford J, Doll R (1985) Exposure to vinyl chloride and aneiosarcoma ol the liver. a report ol the register ot cases. Br J Ind Med 42.750-753 Gokd JM. Liebezeit E, Liier M (1976): Hemangiosareoma and hepatocellular carcinoma ot the liver following vinyl chloride exposure: A report of two cases. Virchows Arch |A| 372:195-203 IARC 11987)' "Monographs on the Evaluation ol Carcinogenic Risks to Humans ` Suppi. 7 Lyoh Johnson EL, Lcrner KG. Siegal M (1972). Association ot androgemc-anaholic steroid therapy with development ot hepatocellular carcinoma. Lancet 2.1273-1276. Kepliner ML. Goode JW, Gordon DE. Calandra JC (1975) Interim results ol exposure ol rats, hamsters and mice to vinyl chloride. Ann NY Acad Sci 246 219 -224. Klatsin G (1977). Hepatic tumors: possible relationship to use ol oral contraceptives Gastroenterolocv 73-386-394 R&S150006 Mortality in VCM/PVC Manufacturers 161 Koischwitz VD. Lelbach WK, Lackner K, Hermanuntz D (1981): Das vinylchlondinduzierte Leberanaiosarkom und hcpatozellulare karzinom. Fortschr Rontgenstr 134:283-290. Maltom C (1974): Angiosarcoma epatico in operai esposti a cloruro di vinile in Italia. Med Lav 65:1 l-l 2. Maitoni C (1977): Vinyl chloride carcinogenicity: An experimental model for carcinogenesis studies. In Hiatt HW, Watson JD. Winstar JA (eds): "Origins of Human Cancer." Cold Spring Harbor, NY: Cold Spnng Harbor Laboratory, pp 119-146. Maltom C, Cotti G (1988); Carcinogenicity of vinyl chloride. Ann NY Acad Sci 534:146-159. Maltom C, Lefemine G (1975); Carcinogenicity bioassays of vinvl chloride: Current results. Ann NY Acad Sci 246:195-217. Maltom C, Rondtnella R (1980): Angiosarcoma epatico in lavoraton esposti a cloruro di vinile in Italia, Acta Oncol 1:35-49. Maltom C, Clint C, Vicint F, Masina A (1984): Two cases of liver angiosarcoma among polyvinyl chloride extruders of an Italian lactory producing PVC bags and other containers. Am J Ind Med 5:297-302. Maltom C, Letcmine G, Ciliberti A. Cotti G. Carretti D (1984): Experimental research on vinyl chloride carcinogenesis. In: "Archives of Research on Industrial Carcinogenesis," vol II, Princeton. NJ: Princeton Scientific Publishers. Popper H (1974): Hepatic angiosarcoma in man and rodents following prolonged exposure to vinvl chloride Gastroenterology 67:794 Popper H. Selikotf IJ. Maltom C, Squire RA, Thomas LB (1977): Comparison ot neoplastic hepatic lesions in man and experimental animals. In Hiatt HW. Watson JD, Winstar JA (eds); "Origin ot Human Cancer." Cold Spnng Harbor, NY: Cold Spring Harbor Laboratory, pp 1359-1382. Viola PL, Bigottt A. Capuio A < 1971 ) Oncogenic response of rat skin, lungs and bones to vinyl chloride. Cancer Res 32:516-522. Wong O. Whorton MD, Ragland D. Klassen C, Samuels D. Claxton K (1986): "An Update ot An Epidemiologic Study ot Vinyl Chloride Workers 1942-1982. Oakland, CA: Environmental Health Associates, Inc.