Document gaNxn13nkd179Zz1GJDp4NbZL

AIA-61 INSSFM SYMPOSIA SERIES VoI. 52 IARC SCIENTIFIC PUBLlCA TIONS V* 13 Environment*! Pollution *n4 Ctrcmoqantc Silk* Pollution d* i'tnvironnomont it mqut* carcwoqiooi IHSSFM. J97S. Vo/. 52, (so. 102-116 ASBESTOS CANCERS AS AN EXAMPLE OF THE PROBLEM OF COMPARATIVE RISKS J.C. GILSON MRC Pneumoconiosis Unit, Llandough Hospital Penarth, Glamorgan, CF6 1XW, Wales, U.K. Occupational exposures to several dusts have caused an excess of respiratory tract cancers - for example, arsenic, chromates, fluorspar, haematite, nickel, uranium. Here the agent is thought to be the element or its salt-, or ionising radiation, or .the two operating together. Professor Maltoni and Dr. Mole will be discussing examples of these. I will limit my remarks to asbestos and some other mineral fibres for two reasons. First, present evidence suggests that the physical properties of the fibres rather than their chemical composition are especially important in the cancers they induce; and second, because "asbestos" is a commercial term of a small sub-group of fibrous silicates, but there are many other fibrous silicates and other fibrous minerals* widely distributed over the surface of the earth, so that the dust from these - even though in very small amounts - has been in the general air since time immemorial and may be seen in airborne dust and in lung residues. (11. . These simple observations indicate that even though there may not be a threshold level below which no effect is produced in a strict biological sense, there is a practical level below which no serious disease is produced^^ Our Imowledge of the carcinogenic effects of asbestos is almost entirely derived from occupational and para-occupational exposures in the past. The question is. therefore, do we yet know enough about the-effect of varying intensity and type of exposure in industry to specify what may be acceptable conditions within industry in the future, as well as to the general public? Types of Cancers Table I shows the types of asbestos and the cancers with which they are associated. Nearly all of it is chrysotile (about 4 million tons/year), but it should not be assumed, as used to be done, that the biological effects of all types are the same. The cancers widely accepted as due to asbestos are bronchial and mesotheliomas of the pleura and peritoneum. But an excess of gastro- * These Include calcium silicates (woolastomite. cement dust); sepiolite (hydrous magnesium silicate); hornblende (amphiboie mineral variety); diatomaceous earth (amorphous silica); fibrous clay minerals (kaolinites, bentonites); and naturally occurring fibrous minerals, such as pyroxene minerals; amphiboie minerals (other than commercial varieties); serpentine minerals (antigorite); and oxide minerals (brucite, magnesium hydroxide). Ol-~OSZ13'53S ASBESTOS INFORMATION ASSOC: North America ' 1835 K Street, N. W. Suite Washington, D. C. 2GCC6 AIA-61 intestinal tumours has been reported in several large cohort mortality studies of asbestos workers (2.3,4,), though the excess has oeen much smaller than for the lung cancers. Larynx, pancreas, and lymphomas have also been suggested. There is still uncertainty aoout the causai relationship of these at cancers with asbestos. Cancer Sites Types of Asbestos Bronchial Chrysoule (95%) Mesothelial Amphiboles Pleural Amosite Peritoneal Crocidolite G.I. 1 Larynx / Others ) ? Anthophyllite Table I; Asbestos and Cancers A pattern of recent research The last 15 years has seen the results of many epidemiological studies into tl incidence of asbestos-related cancers (5), and also parallel animal expenme studies to explore mechanisms as to how such remarkably stable and inert minerals are carcinogenic. The pattern of this work illustrates one of the themes Or. Kigginson has been developing at the tARC; a combined epidemiological and experimental programme of research in which the wor' each field stimulates and orientates that in the other. The experimental w has been mainly aimed at discovering physical or chemical factors responsibi for the carcinogenesis, and it has not been aimed at defining a TLV for man t extrapolation from studies in animals. These approaches and their inter relations are shown in Fig. 1. \ Clinical Observation* Bioto deal Taata Epidemiology t Proof of Causation t Effect of 'Pup*' Exposures Importance of Fibre size; shape: compoaitioit Proof of Association 1 -- ] Estimates of Excess { Risks Ji / Dome/Response f Relatione V/ Relative Rilke Assessment of Future Risks using Mineral Flbrea Other Risks Fig. 1: Research into Asbestos Cancers AIA-61 Epidemiology Proof of association between asbestos and cancers of a particular site Is Sas in part on case/control studies of pathological material, and in part on the c: of deaths in cohorts of asbestos-exposed workers. The pathological studies no measure of the magnitude of the risk and are. therefore, of little predict: value. The simple cohort studies do give a measure of the excess risk for e type of cancer, but due to the long latent period for cancer induction, often 2 or more years, they refer to conditions many years in the past. Such studie many of which have received wide publicity in the general press, are also of little use for prediction of risks in the future or of extrapolation to the gener population. Dose response studies For such predictions cohort studies, in which the workers can be sub-divider on the basis of duration and intensity of past exposure, are required, Ideal!; need quantitative information on past dustiness, out useful information o< obtained by the detailed historical study of the factory or mill, making maximum use of the knowledge of long-term employees, and relating these tc such dust measurements as there have been. In this approach the men's job! can often be grouped into those with markedly different dust exposures, so tt mortality experience can be related to dose. The advantage of this approach that groups of employees are identifiable, who almost certainly bad exposure several orders of magnitude greater than the general population, and yet sho no excess cancer risk or a small one compared to those who have been most heavily exposed. ' Men bom 1091-1920 Deatha/1000 to 1969 Puat Exposure fpare! el*/years I <10 to- 100. 200- 400. soo* AU Cancers 91 11 34 41 47 79 Unt " 10 13 13 It 21 13 Abdominal " ho. of Men 11 14 2010 2339 19 1124 12 1007 2S . S3T 29 303 Tnelndtnf Mtsotiltllama* <5) Table II; Chrysofiie Mining and Milling, Quebec Table II Is an example from the ehrysotUe mining and milling industry in Que where about half the world's asbestos is produced. Only in the two highest exposure groups is there an increase of cancer risks relative to the lowest exposed group. For the whoie group of workers there was no lung cancer ex over the general population. Thus the least exposed group were not at a significant excess risk despite being exposed to far more asbestos dust than no Ol AIA-61 Biological tests The cancers proved in man to be caused by asbestos are also induced in rats and other small animals. Tor example, tntra-pieurai injection of all types of asbestos causma high incidence of mesotheliomas (6), and althougn by this route part of the defence mechanisms of the iungs are effectively by-passed, the technique is very us'eful to study the effect of particle snape and composition m the induction of a tumour kr.own to be caused by asDestos in man. Present evidence suggests that to produce mesotheliomas the fibres may have to be ri VQjj m in length and Less than about lyjn diameter. Larger fibres > 3^u m diameter and rounded particles rarely or never cause mesotheliomas by this route. When It is possible to produce narrow sized ranges of fibres for both diameter and length, it should be possible to specify fairly precisely the critical dimensions for the induction of mesotheliomas. Much effort is being put into preparing such test samples of fibres at present because the results of such experiments are likely to influence the way new man-made mineral fibres are produced in the future and also the way fibrous dusts are sampled in air and water. When the fibres are inhaled it Is the diameter rather than the length which controls entry to the periphery of the lung; also curly fibres, such as chrysotile. are more readily arrested in the upper respiratory tract oy impaction than the straight fibres of amosite or crocidolite. This'may be one of the reasons why the incidence of mesotheliomas in those exposed only to chrysotile is much lower than those exposed to crocidolite or amosite (71. The composition and chemical structure does not seem very important because all types of asbestos and some very fine giass fibres and ceramic fibres have all produced mesotheliomas by intra-pieural inoculation. The animal experimental results have given no support to hypotheses that trace elements or adsorbed hydrocarbons might be important in the production of the pleural tumours (3). 3ut an interesting recent observation (9) is that fully magnesium leached chrysotile. even though fibrous, produces very few mesotheliomas compared to the untreated mineral. An important result of the experimental work has been a better understanding of the likely mesothelioma risk of man-made mineral, and other fibres. In gensral the fibres are too large in diameter and the airborne concentrations too low for many to reacn deep into the lung. In addition, man made fibres do not break down within the lung into fine fibrils as occurs with asbestos (10, 11). Thus present evidence indicates the chances of present man made mineral fibres of a critical site reaching the periphery of the iung in significant amounts must be far less than in the case of the natural fibrous minerals. Inhalation of asbestos produces in rats a small incidence of bronchial cancers and even fewer mesotheliomas (12). So far nothing is known about the importance of fibre size in the production of the bronchial tumours. Feeding rats with asbestos, as well as detailed autopsies of rats inhaling the mineral, (during which they will ingest an appreciable amount when cleaning their fur), has shown no excess of gasxro-intestinal tumours 113.14). 109 AIA-61 members of the general population. Similar patterns have been reported in the asbestos cement industry, and tor lung cancer and mesotheliomas in factories manufacturing asbestos products (IS, 16. 17). Differences within the Industry Table III summarizes differences in proportional mortality within the industry for lung-eancers and mesotheliomas (13), As is often tne case in epidemiology, the interpretation of findings is not simple; in this case because the type of asbestos fibre used and the type of industrial process are to some extent confounded. Many manufacturing processes use more than one type of asbestos, often all three of the major types - chryiotile, amosite, and crocidolite. In mining and milling exposures to one type do occur, but unfortunately for the epidemiologist crocidolite and amosite are mined only in South Africa where the opportunities for quantitative epidemiology are slender because of the paucity of records. Aibiitoi Lung Cancer No. of Meeotb; Surveys Total Men Insulation IS - 26 S-9 S 26.30Q * Factories . 6-21 1 -7 3 10.300 Mining and Milling 2 . 10 0 - 0*2 3 13.TOO Gen: Pool: E and W 9 U.S.A. S 0-00 91 Canada i Mixed fibre exposures * Chrysotile (11; Afttftophylllte (t); Table flt; Proportion (Pa) of All Deaths due to Lung Cancer and Mesotheliomas in Cohorts of Asbestos Workers and the General Population The percentage of lung cancers and mesotheliomas is highest in the insulators; tess in the factories making asbestos products; and least In mining and milling, but the mining and milling covers only chrysotile and anthophyllite. In the two large chrysotile-exposed cohorts, one in Quebec and the other in Italy, no general excess of lung cancers was shown, and no or a very few mesotheliomas. If we add to these quantitative cohort studies the results of 15 years* observations In Southern Africa, where chrysotile, amosite. and crocidolite are mined, it is fairly certain that mining and milling crocidolite is - by a factor of 100 or more - more likely to produce mesotheliomas than amosite or chrysotile. Studies in 111 AIA-61 Finland suggest anthophyllite almost never causes mesotheliomas (19). Cofactors Cigarette smoking is most important in the production of bronchial cancer in asbestos workers. The effects of asbestos and cigarette smoke are multi* plicative, not simply additive. Table IV shows this for both sexes (20,21,). Asbestos alone is apparently a weak bronchial carcinogen but it has not been possible to establish this very firmly, because there are few non-smoking asoestos workers with long exposures. It seems likely that stopping cigaretti smoking is likely to have a much bigger effect on the incidence of lung cancer asbestos workers than small changes in exposure to asbestos dust. No cofact affecting the incidence of mesothelioma have been firmly identified. Ob**rvt4 Insulation Cigarette* Non-Smoker* 24 0 Expect<1 Non Oceup; 2-98 Non Oeeup; and Oeeup; *X 13-9 22-3 0-QS 2-3 0-4 Factor!** Male Cigarette* Non-Smoksri 2S-S 0 9-9 25- 23*4 0-0 0-9 0-1 Ftmnl* Cigarette* ' Non-Smoker* 15-5 l-T 1-4 12-5 1J-J 0*2 4-T 1-3 Table IV: Lung Cancer Deaths and Smoking fit with Additive (*) and Multiplicative 00 Hypotheses Relative risks These asbestos cancers should be seen in perspective in relation to other occupational cancers and occupational accidents (Table V). The information this precisely is not available, but the general pattern is well summarized b Pochin (22,23). All the figures are deaths per million per year. Cancers v from about 700 for the nasal cancers in woodworkers to 24.000 for the betanaphthalene manufacturers, with asbestos lung cancers at about 3,000. The occupational fatalities range from 3 in clothing manufacturers up to 11,000 f the professional divers (24). For both occupational cancers and accidents tt is a big range of risk for different types of job. In general the cancer risks the bigger, but affect a selected group of workers. 112 ) AIA-61 type of fibre used. In the case of mesotheiiomas there is evidence of a major effect of the fibre type in the order of risk, crocicoiite > amostte >chrysotile > anthophyllite. Differences of risk within an industry indicate that there is a dose response relation for both Bronchial cancers and mesotneliomas. Also that in some instances it has been possible to identify lightly exposed groups in which r.o excess risks were aetectable. As these least exposed groups are likely to ha had several orders of magnitude heavier exposure than the general population the risks to the general public are Likely to be negligible. Bronchial cancers in absolute numbers are the major excess risk and are hig smoking-related. Stopping cigarette smoking is likely to be of paramount importance in reducing the excess cancer risks in asbestos-exposed individui Cancers in other sites which may possibly be related to asbestos exposure n-further study even though the magnitude of the excess risk has been small compared to the Lung cancers. In my view it is now possible to use the epidemiological evidence and experimental results to predict with fair confidence the physical and chemica characters and dose of natural and man-made fibres which will .not cause a significant hazard in the future. 'References (1) Pooley F. D.; Personal communication. (2) Elmes P.C. and Simpson M. J. C. : Insulation workers in Belfast. 3r J. industr. Med., 19T1. 23. 3, 226-236. (3j Selikoff I. J. , Hammond E.C. . and Seidman H.: Cancer risk of insula; workers in the United States:in "Biological Effects of Asbestos", IARC Sci. Pub. No. 3, 1973. pp. 209-216. (4) McDonald J. C.: Cancer in chrysotile mines and mills. Ibid. pp. 189- (3) Bogovski : Ibid. pp. 139-226. (6) Wagner J. C. and 3erry G.: Mesotheiiomas in rats following inoculati with asbestos. Brit. J. Cancer. 1969, 23,3, 567-331. (7) Timbrell V.: Physical factors as etiological mechanisms in "3ioiogxc Effects of Asoesxos", IARC Sci. Pub. No. 3, 1973, pp. 295-303. (8) Wagner J.C. , Berry G. . and Timbrell V.: Mesotheliomata in rats alter inoculation with asbestos and other minerals. Brit. J. Cancer, 1973, 28, 2, 173-185. (9) Wagner J. C. : Personal communication. 114 AIA-61 i Cancers ! Accidents Wood Workers Nasal too Clothing Mf; 3 Asbt Industry Lung 2.000 M A,000 F 3rlcks and Cement Ship Building ao ISO Rubber Werners Bladder 7.000 Coal face 600 Nickel Refining (up to 13251 Lung 15.000 Company Directors 1.300 . Deep eea fishing 3.000 JNaph: Mf: Bliddar 24.000 Prof: Divers 11.000 Table V: Estimated Occupational Mortality/Vt/ year Finally, compare these occupational risks with risks run by ail from accidents, our personal habits, and our age (Table VII. Traffic accidents are about o.ne quarter that of working on the coal face, but cigarette smoking at 20/day is about double that of heavy asbestos exposure. This is. of course, due to cigarette smoking affecting several diseases. The last column shows how rapidly age creeps up on us and perhaps indicates that those who research into occupational hazards should be in the 30's to see things In perspective! General Accident*; Person*! Habits; Your Ate (Melel Aceidtfit* Your Age T raffle ISO 12 330 Home 130 30 1.000 Suicide 30 41 3.000 AU 480 S3 10,000 Cltirtctts S3 30.000 20/day 5,000 ?? too.aoo Table VT: Estimated Mortality/M/year in U.K. SUMMARY Major differences in excess cancer risks have occurred in the asbestos industry in the past; part of this difference is probably related to dustiness, part to the 113 i 0/ ***** AIA-61 do) Corn M. and Sansone E. 3.: Determination of total suspended particulate matter and airborne fitter concentrations at three fibrous glass manufacturing facilities. Envlronm, Res., 137 , 3, 37-52. dil Assuncaa J. and Corn M. : The effects of milling on diameters ar.d lengths of fibrous glass and chrysotile asoestos fibers. Paper presentee at the .Annual Meeting of the .American Industrial Hygiene .Association. Minneapolis, Minnesota, 1975. (To be puoiished). (12) Wagner J. C. , 3erry G. , Skidmore J. W. . and Timbrel! V. : The effects of the inhalation of asbestos in rats. Brit. J. Cancer, 1974, 29, 3, 252-269. (13) Gross P. , Harley A.R. , Swinburne L.M. . Davis J.M.G. , and Grane W. B.: Ingested mineral fibres: Do they penetrate tissue or cause cancer1 Arch, envlronm. Hlth. , 1974, 29, 6, 341-347. (14) Wagner J.C.: Personal communication. (15) McDonald J. C.: Cancer in chrysotile mines and mills: in "3iological Effects of Asbestos". IARC Sci. Pub. No. 8, 1973, pp. 139-t94, , (!6) Mewhouse M. L.: Cancer among workers' in the asbestos textile industry. Ibid, pp. 203-203. (17) Enterline P. E. , de Couile P. . and Henderson V.: Respiratory cancer in relation to occupational exposures among retired asbestos workers. Brit. J. industr. Med., 1973, 30, 2. 162-166. (18) McDonald A. and McDonald J. C.: Paper to International Congress of Occupational Health, 3righton, 1975. (To be published). (19) Wagner J. C., Gilson J. C. , Berry G. . and Tlmbrell V.: Epidemiology of asbestos cancers. Brit. med. Bull., 1971, 27, l, 71-76. (20) Doll R.: Practical steps towards the prevention of bronchial carcinoma. Scot. med. J. . 1970. IS. 433-447. (21) Berry G. , Newhouse M. L., and Turok M.: Combined effect of asbestos exposure and smoking on mortality from lung cancer in factory workers. Lancet, 1972, 2, 47S-479. 122) PochinE.E.: Occupational and other fatality rates. Community Health. 1974, 6. 2-13. (23) PochinE.E.: The acceptance of risk. Brit. med. Bull.. 1975. 31, 3, 184-190. (24) Crockford G. W. and Dyer D.: Annual Report of the TCC Centenary Institute of Occupational Health. London, 1973- 1974, pp. 19-20. ns 0i >o