Document omK10E9vJxZpBVwoOkO1zVjdX

r} Reprinted from CANCER. Vol. 2G. No. 4. Octo ber 1970. Copyright, 1970, by the American Cancer Society, Inc. J. B. Lippiiicotl Gompain. Primed in I'.S.A. EPIDEMIOLOGY OF PRIMARY MALIGNANT MESOTHELIAL TUMORS IN CANADA* A. D. McDonald, A. Harper, O. A. El Attar, and J. C. McDonald All fatal malignant mesothclial tumors known to pathologists in Canada be tween 1959 and mid-1968 were registered. They numbered 165 (1 per million population per annum); two thirds were in males. Occupational and residen tial histories were obtained "blind" from relatives and friends of 90% of the cases and 2 matched control series. An association with definite or probable occupational exposure to asbestos was clearly demonstrated, but only 20% of male cases and one female case had any such contact. Almost all the excess was in the manufacture and industrial application of asbestos rather than in mining or milling. No association was found with lesser degrees of occupa tional exposure or residence in asbestos-mining areas, but there was a small excess of possible domestic exposures. The smoking histories in the mesothelial tumor and main control groups were almost identical and unlike those for cases of primary lung cancer. vidence for an association between ma- of Pathologists or the Quebec Association of E lignant mesothelioma and asbestos ex Laboratory Physicians inquiring whether they posure rests mainly on the remarkable conh ad seen any fatal cases of primary malignant centration of cases reported by Wagner et al.5 tumors of the pleura or peritoneum diagnosed in the crocidolite mining areas Of South Africa, by autopsy or biopsy since 1959. In due course the case-control studies of Elmes et al.1 in Bel we obtained a reply from them all, and, if this fast, and Newhouse and Thompson2 in Lon was in the affirmative, a personal visit was don, and the considerable excess mortality due made by one of us to review the records and to this disease observed by Selikoff et al.3 in to exclude cases which did not meet these American insulation workers. The survey de criteria or about which the pathologist had scribed here was undertaken to obtain a more serious doubt. In all, 165 cases were found; generally representative view of the problem 113 were pleural, 45 peritoneal, 3 pleural and in one of the 2 major chrysotile-producing peritoneal, and 4 pericardial. The diagnosis countries of the world. Our aim has been to was supported by autopsy in two thirds of the study all fatal cases known to have occurred cases and by biopsy in the remainder. Later, in Canada since 1959. we plan to have material from as many as possible reviewed by a panel of pathologists. Method During the visit to the pathologist, 2 fatal control cases--one primary and one second At the end of 1967 we wrote to the 423 ary lung cancer--were selected from the au members of either the Canadian Association topsy or biopsy records, matched for sex and, as closely as possible, for age and date of From the Department of Epidemiology and Health, McGill University, Montreal, Canada. A preliminary report on these results was pre sented at the International Conference on Pneumo coniosis at Johannesburg in April, 1969. Supported by a grant from the Institute of Occu pational and Environmental Health of the Quebec Asbestos Mining Association. This survey was made possible by the great help given us by pathologists and members of provincial and other health departments throughout Canada. t\'e are also much indebted to all those who did the field work and to the friends and relatives of the deceased for their cooperation. Received for publication April 9, 1970. death. In 2 instances, no adequately match ing case of primary lung cancer could be found. Two control series were chosen in or der to examine 2 types of question. Primarily, we wished to know whether persons who had died from malignant mesothelial tumors had differed in their past exposure to asbestos and certain other materials from comparable per sons in the general population. Comparabil ity demanded that the control series should comprise persons matched with the cases for 914 ]!' i t l!f ASARCO ALV 0007256 No. 4 Malignant Mesothelial Tumors McDonald ct al. 915 age, sex, date and place of death, and post mortem diagnostic procedure. To insure that information obtained from relatives would be of equal quality, it was important that the terminal illness should also be similar. Sec ondary lung cancer seemed to meet these cri teria reasonably well and had the additional advantage that several primary cancer sites were included, but none were in the lung. A second main question was whether primary mesothelial tumors were related to tobacco smoking in the same way as primary lung can cer, and, for this reason, a matched series of cases of this disease was also selected. A field survey was arranged in which in vestigators (doctors, public health nurses, and public health inspectors) sought out the relatives and friends of the deceased and com pleted a detailed questionnaire in English or French relating to occupation, residence, in direct household exposure to dust, family his tory, and smoking habits. Special emphasis was given to the occupational history. So far as possible, all employers or employing com panies were listed with dates and a note made of the exact nature of the work done. The relative was then asked whether each job in volved contact with wood, rubber, asbestos, cement, copper, nickel, or fiber glass. Finally, a check list was used, to inquire whether the subject had ever been employed in any of the following: locomotive repair shop; shipbuild ing, overhauling, or breaking; installation of large boilers or steam pipes; insulation of do mestic heating equipment; building--sound proofing, fireproofing, heat retention, assem bling prefabricated buildings, and demolish ing buildings; manufacture of brake linings or clutch plates;-mining industry; textile in dustry; cement industry; dock or transport industry. No interviewer knew to which group the case under investigation belonged, and the information recorded in the questionnaire was also coded blind. Questionnaires were completed for 447 (91%) of the 493 cases and controls. Findings The distribution of cases by province and year of death is shown in Table 1, and by age, sex, and site in Table 2. More cases were found in 1966 and 1967 than in previous years. Recent interest in these tumors may have contributed to this, and, as the records in some hospitals were poorer for the earlier years, memory may also have been a factor. With these points in mind, the true rate of increase is difficult to assess. During the 8 com plete years, 1960-1967, the average annual in cidence was 1.0 per million of population. The rate was appreciably higher in Quebec (1.5) than in Ontario (0.8) or in the rest of Canada (0.9), 1966 and 1967 again account ing for much of the difference. Three quar ters of the cases were aged 50 or more at death, and males predominated by 2.4 to 1. Under 50, there was little difference in the sex distribution. The case and 2 control series were very similar in age distribution and in the propor tion of each for which the questionnaire was completed (Table 3). Table 4 shows the number of men aged 18 years or more who were occupationally exposed up to 10 years before death to asbestos and certain other ma terials in dust or vapor form. Whereas all 3 Table 1. Distribution of Cases by Year of Death and Province Year Ontario 1960 1961 1962 1963 1964 1965 1966 1967 1968 (1st half) - 1 4 9 6 7 8 6 5 ----46 Incidence/million/year 1960-1967 0.8 Quebec 6 6 4 3 8 7 14 13 9 70 1.5 Other Provinces 9 3 S 7 3 5 9 6 2 -- 49 0.9 Canada 15 10 13 19 17 19 31 25 16 ii165 1.0 ASARCO ALV 0007257 916 Cancer October 1970 Table 2. Pleura Age (years) M F 0- _ _ 10- 2 1 20- -- 4 30- 7 4 40- 9 7 50- 18 3 60- 26 S 70- IS 5 80- 3 4 Distribution of Mesothelial Tumors by Arc, Sex, and Site Peritoneum Pleura and peritoneum Pericardium Total M FM FMFM F -- 1 ---------- 1 ------------ 2 1 -- -- -- -- `--- -- -- 4 1 1 ------ 1 86 4 1 -- -- -- 1 13 9 11 4 1 -- 1 1 31 8 2 5 -- 2 -- -- 28 12 7 6 -- -- -- -- 22 11 1 1---- --4 S All 80 33 26 19 1 2 1 3 108 57 Total 113 45 3 4 165 Vol. 26 groups were equally exposed to wood dust, relatively more men in the mesothelial tumor group were exposed to all the other materials, the greatest excess being for asbestos. Men who were exposed to asbestos tended also to be exposed to other materials, especially ce ment and fiber glass (Table 4), but these were only responsible for part of the excess exposure to the other materials. The excess of "definite" or "probable" exposure to as bestos (Table 5) in the mesothelial tumor group compared with both control groups was substantial (P < .001), but there was virtuallyno difference in the frequency of jobs in which exposure was classified as "possible." In 60% of the cases, however, occupational exposure to asbestos was considered "un likely." Only 2 women in all had any occupa tional exposure to asbestos; there was a definite exposure in the mesothelial tumor group and a possible exposure in the second ary lung cancer group. The occupations involving definite or prob able exposure to asbestos are listed in Table 6. Men employed in asbestos textile manu facture, the installation of brake linings, or as insulators were responsible for most of the excess of cases over controls. Mine or mill workers made little contribution to the differ ence, and only one person, a hardware storeman, had worked with asbestos-cement prod ucts. The time between first exposure and death for the 21 cases in the definite or prob able class ranged from 16 to 50 years; 4 men and one woman who worked on the job for less than 4 years had intervals of 18, 26, 26, 29, and 33 years. Materials other than asbestos also deserve further consideration since several men had worked with more than one of them. Exclud- Table 3. Distribution by Age and Sex of Cases and Controls Males Females Mesothelial Secondary Primary Mesothelial Secondary Age tumors lung cancer lung cancer tumors lung cancer 0- 1 CD 1 0) 1 (1) 1 (1) 10- 2(2) 1 (1) 20- 2(2) 4(3) 2(2) 30- 8(8) 6(6) 2(2) 6(6) 6(5) 40- 13(10) 14(11) 24 (20) 9(8) 10(10) 50- 31 (31) 26 (22) 24 (23) 8(7) 7(7) 60- 27 (25) 35 (32) 35 (33) 12(11) 15(14) 70- 23 (22) 20(19) 17(15) 11 (10) 13(11) 80- 4(3) 4(4) 4(4) 5(4) 3(3) All Mean SD 108(101) 58.7 13.9 108 (97) 59.1 13.3 107 (98) 59.1 12.5 No. of questionnaires completed shown in parentheses ( ). 57 (51) 55.8 19.3 57 (53) 57.4 17.3 Primary lung cancer 1 0) 3 (2) 14 (13) 8(7) 12(7) 15 (15) 3(2) 56 (47) 5S.8 15.5 ASARCO ALV 0007258 No. 4 Malignant Mesothelial Tumors McDonald et al. 917 Table 4. Occupational Exposure of Men to Certain Materials* Mesothelial tumors Secondary lung cancer . Primary lung cancer Asbestos Cement Copper Fiber glass Nickel Rubber Wood (100 men) 20 17(4) 6(2) 6(3) 5(2) 3(1) 18(1) (96 men) 3 9(0) . 1 (0) 0 0 0 19(0) (97 men) 4 8(0) 1 (0) 0 0 0 16(0) Total exposures recorded 75(15) 32 (0) 29 (0) * Definite or probable exposure to the substances named in dust or vapor form in men aged 18 or more. The number of men who were also definitely or probably exposed to asbestos is given in parentheses. ing wood, for which there was evidently no difference between the 3 groups, 37 exposures to cement, copper, fiber glass, nickel, or rub ber were recorded for the cases, of which 12 were in men definitely or probably exposed to asbestos. The 25 remaining exposures were in 19 men, and the comparable figures for the 2 control groups were 10 and 9, respectively. An analysis of these exposures is shown in Table 7, where it may be seen that the excess appeared to be in occupations involving con tact with copper, nickel, fiber glass, or rubber (9:1:1) rather than with cement only, and seemed unrelated to possible exposure to as bestos. The relevant occupations of the 9 men in the mesothelial group were as follows: copper: lathe operator (2), munition manu facture, brass foundry; nickel: lathe operator (2), mine worker; rubber: tire manufacture, plantation work; fiber glass: municipal main tenance work, wood-caulking (ship). Residential histories indicated that 45% of male cases and 38% of female cases had spent their childhood entirely in a rural area com pared with 40% of males and 38% of females with secondary lung cancer, but the primary lung cancers were less often reared in rural areas (males 31%, females 28%). No differ ence was found between cases and both con trol groups after the age of 15 years. Exclud ing 5 cases (3 mine or mill workers and 2 textile workers) and one mine and mill worker in each control group, who lived in mining towns, only 3 men and 2 women had Table 5. Distribution of Cases and Controls According to Their Occupational Exposure to Asbestos" Males Mesothelial tumors Secondary lung cancer Primary lung cancer Exposure to asbestos Definite Probable Possible Unlikely No. % 1 20.0 9) 20 20.0 60 60.0 No. % '} 3.1 21 22 22.9 71 74.0 No. % '] 4.1 3) 16 16.5 77 79.4 All* 100 100.0 96 100.0 97 100.0 Males: Distribution of definite and probable groups compared with possible and unlikely: x1 = 21.5, 2 d.f.. P < 0.001 Females Mesothelial tumors Secondary lung cancer Primary lung cancer Exposure to asbestos Definite Probable Possible Unlikely No. % 1 2.0 -- -- 49 98.0 No. ____ -- 1 51 % 1.9 98.1 No. % -- -- -- 46 100.0 All* 50 100.0 52 100.0 * One child in each group dying under the age of 18 has been excluded. 46 100.0 ASARCO ALV 0007259 918 Cancer October 1970 Table 6. Distribution of Occupations Classified Under Definite or Probable Exposure to Asbestos Mesothelial Secondary Primary tumors lung cancer lung cancer Asbestos mine or mill Asbestos textile manufacture Brakelining installation Insulation Other occupations involving contact with insulation materials Hardware store (handling asbestos sheet and pipe) Filled gas masks with asbestos textile Total 3 4* 2 5 S 1 1 -- 21 1 0 0 0 2 0 0 -- 3 1 0 0 0 3 0 0 4 * Includes one person who worked in brakelining manufacture. Vol. 26 ever lived within 20 miles of an asbestos mine. One of the 5 was a case, 3 had second ary lung cancer, and one had primary lung cancer. The possibility of contact as a result of some other member of the family bringing home dusty clothes from occupations involv ing the use of asbestos was also examined. Ex cluding 7 persons (5 cases and one in each control group) who were occupationally ex posed themselves (definitely or probably), there was one case in a female with definite home_exposure and one male with probable home exposure. In addition to 5 persons (one case and 4 with primary lung cancer) who were possibly occupationally exposed to as bestos themselves, there were 26 persons with a possible home exposure: 14 cases, 6 with secondary lung cancer, and 6 with primary lung cancer. The category of possible home exposure was given in households where a person possibly.exposed at work to asbestos brought home dusty clothes. Thus, excluding those persons with definite or probable occu pational exposure, there were 15 persons with mesothelial tumors, 8 with secondary lung cancer, and 10 with primary lung cancer, who had had at least possible home exposure to asbestos dust. The frequency of cancer reported in close relatives did not differ from that in the con trol groups. An analysis of cigarette smoking habits is shown in Table 8. In both males and females, persons in the primary lung cancer group had smoked considerably more than those in either the case or secondary lung cancer groups. The latter 2 groups had almost an identical distribution of smoking habits. The 20 men with definite or probable occupational exposure to asbestos showed no great difference in smoking habits from cases without occupational exposure. Discussion Though our study is not directly compa rable with those of Elmes et al.1 in Belfast, and Newhouse and Thompson2 in London, it seems fairly clear that the difference between cases and controls regarding asbestos exposure was less in our findings than in theirs. There are several possible explanations for this. The first is that our study was based on a na tional sample of cases derived from all parts of a very large country, whereas theirs were from 2 specific industrial areas, one a ship building city and the other from a hospital close to a large asbestos factory. A second Table 7. Occupational Exposure to Certain Materials of Men without Definite or Probable Exposure to Asbestos Cement only Copper, nickel, rubber, or fiber glass Mesothelial Secondary Primary tumors lung cancer lung cancer 10(5) 9(6) 8(2) 9* (2) 1(0 1 (0) 19 (7) 10 (7) 9(2) * Including three men also exposed to cement. The number of men possibly exposed to asbestos is given in parentheses. ASARCO ALV 0007260 No. 4 Malignant Mesothelial Tumors McDonald el at. 919 Table 8. Average Daily Number of Cigarettes Smoked Male Female Cigarettes per day Mesothelial tumors Secondary' lung cancer Mesothelial tumors with Primary asbestos lung cancer exposure Mesothelial tumors Secondary lung cancer Primary lung cancer Nil 10 or less 11-29 30 or more 21 22 17 IS 40 41 19 20 19 20 21 22 37 39 18 19 44 10 11 47 SO 34 36 2 10 4 20 13 65 15 29 58 9 18 10 20 24 36 71 5 10 7 14 36 21 47 9 20 12 27 37 Total* 97 100 95 100 95 100 20 100 50 100 51 too 45 100 * Excluding 6 persons under IS years of age and 6 men and 2 women whose smoking history was not recorded possibility is that our series was diluted by cases which they would not have accepted. A third point is that we took great care to in sure that the histories of case and control were collected and coded without knowledge of the group to which they belonged. This was not so stated in the Belfast survey and not possible in the study of Newhouse and Thompson. The findings of Selikoff et al.s are not comparable with ours since their investi gation was based on an occupational group selected for its exposure to asbestos and not on a population sample. It was not our aim to assess the role of as bestos in the etiology of primary lung cancer. Several studies have shown that occupational exposure may be associated with increased risk of this tumor, especially in smokers.4 Since our series of primary lung cancer was selected to match the mesothelial cases in age, sex, and geographical distribution, it cannot be considered representative in itself. Neverthe less, it was probably sufficiently so for it to be worth noting that the history of exposure to asbestos was' very similar to that for secondary lung cancer. Any contribution from asbestos in the causation of this series of primary lung cancers was, therefore, not apparent. Conclusion The findings from this national survey in dicate that primary malignant mesothelial tu mors are rare in Canada, and it is somewhat doubtful whether there has been a true in crease in incidence since 1959. An association between these tumors and definite or prob able occupational exposure to asbestos was clearly demonstrated, but only a minority of male cases (20%) and one female case had any such exposure. No difference was found between cases and the 2 control groups re garding lesser degrees of occupational expo sure or residence in asbestos-mining areas, but, in 10% of the mesothelial tumors com pared with 5% of the secondary lung cancer group, there was at least a possible domestic contact with asbestos dust. Probably our most noteworthy finding was that almost all the excess in occupational exposure was in the manufacture and indus trial application of asbestos rather than in mining or milling. Taken in conjunction with the results of other epidemiologic stud ies, this suggests either that chrysotile is less associated with mesothelial tumors than other forms of asbestos or that something in addi tion to asbestos is necessary. If there is an additional factor, our findings imply that it is not cigarette smoking. Nine men with mesothelial tumors had worked in occupations unrelated to asbestos but involving contact with copper, nickel, rubber, or fiber glass compared with one in each of the control groups. The interpreta tion of this finding is doubtful. REFERENCES 1. Elmes, P. C., McCaughey, W. T. E., and Wade, O. L.: Diffuse mesothelioma of the pleura and asbestos. Bnt. Med. J. 1:350-353, 1965. 2. Newhouse, M. L., and Thompson, H.: Mesothe lioma of pleura and peritoneum following exposure to asbestos in the London area. Brit. J. Industr. Med. 22-.2GI-269, 1965. 3. Selikoff, I. J., Churg, J., and Hammond, E. C.: Relation between exposure to asbestos and mesotheli oma. New Eng.J. Med. 272:560-565, 1965. 4. Selikoff, I. J., Hammond, E. C,, and Churg, J.: Asbestos exposure, smoking and neoplasia. JAMA 204:106-112,1968. 5. Wagner, J. C., Sleggs, C. A., and Marchand, P.: Diffuse pleural mesothelioma and asbestos exposure in the north western Cape Province. Brit. J. Industr. Med. 17:260-271, 1960. asarco alv 0007261