Fiber Concentration in Lung Tissue of Patients With Malignant Mesothelioma A Case-Control Study GUNNAR MOWE, MD,' BJ0RN GYLSETH, PHD,' FLORA HARTVEIT, MB, CHB, DR MED,t AND VlDAR SKAUG, MD'

The risk of malignant mesothelioma associated with low-level asbestos exposure is an important unresolved issue today. We have analyzed the asbestos fiber concentration in lung tissue from 14 cases of malignant mesothelioma and 28 case-matched controls by scanning electron microscopy. The cases represent 86% of all mesotheliomas recorded by the Cancer Registry of Norway from the county of Hordaland between 1970 and 1979. Based on 1 million fibers per g of dried tissue as an indicator of cumulated asbestos exposure, the odds ratio (relative risk) was 8.5 (95% confidence limits, 2.3-31.1). Assuming that the risk of malignant mesothelioma is related to mineral fiber concentration in lung tissue, it is concluded that a fiber concentration exceeding 1 million fibers per g of dried tissue is associated with an increased risk of malignant mesothelioma. Furthermore, the results are consistent with a no-threshold response. Cancer 56:1089-1093, 1985.

T

occupational asbestos exposure is a major problem in epidemiologic investigations of asbestos-induced diseases. Consequently, the assessment of the risk of malignant mesothelioma associated with low-level asbestos exposure is complicated. Fiber concentration in lung tissue as analyzed by electron microscopy is probably one of the best indicators of cumulated asbestos exposure,'-3 and such analyses have therefore been a valuable tool in epidemiologic studies of malignant mesothelioma in recent The objective of this investigation was to analyze fiber concentration in lung tissue from patients with malignant mesothelioma compared with case-matched controls. Assuming that the risk of developing the disease is related to the fiber concentration in lung tissue, the investigation gives an indication of the relative risk of malignant mesothelioma associated with low-level asbestos exposure. HE RETROSPECTIVE EVALUATION O f

From the *Institute of Occupational Health, Oslo, and the ?Department of Pathology, the Gade Institute, University of Bergen, Norway. Address for reprints: Gunnar Mow&,MD,Institute of Occupational Health, POB 8149, Dep. Oslo I , Norway. The authors thank the Cancer Registry of Norway, the National Insurance Institution, and the local Social Insurance Office of Bergen for their cooperation. Accepted for publication January 18, 1985.

Material and Methods

Study Design The investigation was based on a case-control design. The source population for both cases and controls was the population of Hordaland county in the western part of Norway from 1970 through 1979. The county has a population of about 390,000, one half of whom live in the city of Bergen. One pathology department serves this population. The major industries in the county comprise shipyards, textile industry, manufacturing, agriculture, water transport, and fishing. In 1978 there were 35 shipyards with about 7000 employees in the county of Hordaland. Most of the shipyards were small, and only two had more than 500 employees.8 The incidence rates of malignant mesothelioma in Hordaland county have been calculated from the total number of mesothelioma recorded by The Norwegian Cancer Registry in 1970 through 1979 and the mean of the census population of Hordaland for 1972 and 1977. The age-adjusted incidence rates have been calculated from the mean of the census populations of 1972 and 1977 for the whole country as standard p o p u l a t i ~ n . ~ -

Selection of Cases A national incidence of 117 men and 24 women with malignant mesothelioma was recorded by the Cancer

1090

CANCERSeptember 1 1985

Registry of Norway in 1970 through 1979; 14 men and 4 women came from Hordaland county." Due to the

small number of women, the investigation has been confined to men. The basis for the selection of the 14 cases included in this investigation was: They died between 1970 and 1979; the diagnosis was verified at autopsy; they were reported to and registered by the Cancer Registry as having malignant mesothelioma; and stored lung tissue for fiber analysis and the autopsy report were available. The cases comprise 86% of all men with malignant mesothelioma in the period from 1970 to 1979 registered by the Cancer Registry from Hordaland county. Two men were excluded, as autopsy was, not performed, and another two men from the neighboring county were included, as autopsy had been performed at the Gade Institute.

Selection of Controls For each case, two controls were selected from the autopsy files. The criteria for selection of the controls included: death between 1970 and 1979; matching for sex, age at death (+5 years), year of death, and county of residence at time of death; verification of the cause of death at autopsy; and availability of stored lung tissue for fiber analysis and the autopsy report. Patients with cancers and chronic noninflammatory pulmonary disease were excluded, because such diseases could be associated with previous asbestos exposure. The first two autopsies in chronological order after the case in the autopsy files that met the selection criteria were chosen as controls. The controls were selected independent of occupation mentioned in the autopsy report.

Exposure Information Information about past occupations for both cases and controls was obtained from the local social insurance offices. Because of the compulsory social insurance system in Norway, the local social insurance offices have information on past and current occupation as well as employers for all employees in the county. The evaluation of the probability of occupational asbestos exposure for the cases and the controls was based on a retrospective evaluation of the occupational histories. The cases and controls were categorized into four groups according to the following criteria of estimated probability of occupational asbestos exposuredefinite exposure (++): cases employed as insulation workers or asbestos cement production workers; probable exposure (+): cases employed in occupations such as in construction industry, workshop and heating trades, ship

Vol. 56

yards, and ship machine crew; possible exposure (+?): cases employed in various jobs or industries involving dust exposure; otcupational asbestos exposure may be involved in many such occupations; and unlikely or unknown exposure (-): cases employed in occupations with no evidence of occupational asbestos exposure; occupational or nonoccupational asbestos exposure, however, cannot be excluded.

Mineral Fiber Analysis After ashing of the dried lung tissue specimens in activated oxygen, the fiber concentration was analyzed by scanning electron microscopy.12 Through extensive interlaboratory comparison^'^.'^ it has been demonstrated that this method is adequate for total number determinations. Based on this method, an analytical sensitivity of 0.1 million fibers per g of dried tissue and a detection limit of 0.4 million fibers per g of dried tissue were obtained."

Statistical Methods The calculation of odds ratio (relative risk), chisquare, and 95% confidence limits were based on methods suggested by Miettinen. "-I8

Results Fourteen men with malignant mesothelioma were recorded by the Cancer Registry in the source population during the period from 1970 to 1979 against 10.4 expected, based on the age-specific incidence rates in the general population of Norway between 1970 and 1979." The 95% confidence interval is 7.7 to 23.5, based on a Poisson distribution of the expected numbers. The age-adjusted incidence rate of malignant mesothelioma in men in 1970 through 1979 was 8.0 per million person-years in Hordaland county and 5.9 per million person-years for the whole country." Detailed information is given about cases and controls in Tables 1 and 2. The mean age at death for the cases was 67.5 years (standard deviation, 10.0), and for the controls 68.1 years (standard deviation, 9.6). Eleven of the cases (79%) against seven of the controls (25%) had been employed in occupations with probable occupational asbestos exposure (odds ratio, 1 1; chi-square, 10.7; 95% confidence limits, 2.6, 46.4). Furthermore, 2 of the cases and 14 of the controls had been employed in occupations with possible asbestos exposure. Thus, for 13 of the cases (93%) and 21 (79%) of the controls occupational asbestos exposure was probable or possible. Table 3 shows the distribution of the fiber concentra-

No. 5

FIBERCONCENTRATION

IN

LUNGAND MESOTHELIOMA*

MOW6

et a/.

109 1

TABLE1. Survey of 14 Men With Malignant Pleural Mesothelioma Recorded in 1970 Through 1979 by the Cancer Registry of Norway from Hordaland County in West Norway

Case no. I 2 3 4 5 6 7 8 9 10 11

12 13 14

Occupation

Year at death

Age at death

Shipstoker (+) Waiter (-) Foundry worker (+) Mine storekeeper (+?) Plumber, shipyard (+) Carpenter, shipyard (+) Plumber, railway workshop (+) Foundry worker* (+) Ship engineer (+) Construction worker (+) Carpenter (+) Plumber (+) Construction worker Goldsmith (+?)

1975 I970 1979 1971 1978 1975 1974 1975 1971 1974 1971 1975 1977 1979

78 81 63 65 78 66 62 61 74 62 41 67 78 69

* Exposed to both silica and asbestos. At autopsy silicosis was diagnosed. Probability of occupational asbestos exposure based on the occupa-

Latency time since first exposure until death

Fiber concentration (million/g dried lung tissue)

37

0.8 1.5 8.5 2.6 36.0 6.5 37.0 2.0 2.1 1.8 4.4 35.0 1.2 0.4

33

31 32 35 27 36 32 18

-

tional history-(++): unknown.

definite; (+): probable; (+?):possible; (-): unlikely/

TABLE2. Survey of 28 Case-Matched Controls

Control no. 1-1

1-2 2- 1 2-2 3- 1 3-2 4- 1 4-2 5- 1 5-2 6- 1 6-2 7- I 7-2 8-1 8-2 9- 1 9-2 10-1 10-2 11-1 11-2 12-1 12-2 13-1 13-2 14-1 14-2

Occupation Construction worker (+) Tram driver (+?) Clerk (-) Fisherman (+?) Printer (+?) Truck driver (+?) Unknown (-) Cashier (-) Military officer (+?) Electrician (+) Stone worker (+?) Clerk, paper mill (+?) Salesman (-) Painter (+?) Stores man, slate mine (+?) Typographer (+?) Construction worker (+) Furnace tapper, ferrochromium alloy factory (+) Painter (+?) Truck driver (+?) Construction worker (+) Packer (unspecified) (-) Factory owner (+) Clerk, railway office (-) Railway worker (+?) Ship electrician (+) Rolling mill worker (+?) Unknown (-)

Year at death

Age at death

Fiber concentration (million/g dried lung tissue)

Senile dementia Chronic pyelonephritis Cystic renal disease Cerebral thrombosis Myocardial infarction Myocardial infarction Senile psychosis Myocardial infarction Myocardial infarction Myocardial infarction Aortic aneurism Myocardial infarction Cerebral hemorrhage Diabetic nephropathy Myocardial infarction Cardiosclerosis Cerebral hemorrhage Cerebral hemorrhage

1975 1975 1970 1971 1979 1979 1971 1971 1978 I978 1975 1975 1974 1975 1975 1975 1971 1972

73 81 83 79 62 64 68 73 81 74 69 62 61 63 63 64 76 74

3.3 0 0 0 0 0.6 0 0 0 0.5 0.6 2.4 0.2 0. I 3.9 0 0.2 1.6

Pulmonary tuberculosis Myocardial infarction Subarachnoidal hemorrhage Subarachnoidal hemorrhage Myocardial infarction Diabetes mellitus Cerebral apoplexia Cardiosclerosis Myocardial infarction Chronic glomerulonephritis

1975 1974 1972 1971 1975 1975 1978 1978 1979 1979

58 63 42 46 67 68 77 76 75 66

0.2 0 0.4 0.3 3.9

Cause of death

Probability of occupational asbestos exposure based on the occupa-

0.1

0.9 4.8 2.9 0.3

tional history-(++): definite; (+): probable; (+?): possible; (-): unlikely or unknown.

1092

CANCERSeptember I

1985

Vol. 56

TABLE3. Fiber Concentration in Lung Tissue of Cases and Case-Matched Controls: Probability of Occupational Asbestos Exposure as Judged by Occupational History Million fibers/g of dried lung tissue

Cases

0 I million fibers/g of dried tissue I

0

0

5

0

2

7 0

2 Cases

=

> I million < I million

Point estimate of the odds ratio (relative risk): 8.5; chi-square (xzI) 10.3, P 5 0.00I (one-sided); 95% confidence interval: 2.3, 3 1.2.

population of Hordaland than in the general population, probably due to the extent of the shipbuilding industry.' The proportion of cases (79%) who had ever been employed in occupations associated with probable occupational asbestos exposure is consistent with other investigation^.^.'^ However, the reported proportion of cases with asbestos exposure varies,20and it is suggested that a significant number of cases are apparently not asbestos-related.2',22The exposure documentation is not standardized, and the evaluation of the occupational history depends on subjective judgment. Consequently, it is difficult to compare results from other epidemiologic studies. The fiber concentration in lung tissue was thus chosen as an indicator of occupational asbestos exposure in this investigation. The fiber concentration in a sample of lung tissue depends both on the cumulated asbestos exposure, a sequence of various mechanism^,'^ and the topographic distribution of fibers in the Furthermore, chrysotile is probably deposited more slowly and eliminated more rapidly than amphibole^.^,^^ Marked variation in fiber concentration in lung tissue from site to site has been found; therefore, low fiber concentrations should always be evaluated with caution.25In this investigation the localization of the lung tissue samples was unknown, but the bias related to this procedure has probably been minimized by the case-control design. In a previous investigation we have shown that the lung fiber concentration is a useful indicator of the cumulated asbestos exposure as judged from the occupational h i ~ t o r y In . ~ workers with high exposure, such as insulation workers and asbestos-cement workers, we have found fiber concentrations between 50 and 500 million per g of dried tissue in some case^.^.^^ Rased on analysis of lung tissue from patients with known occupational history, we consider fiber concentrations between 1 and 10 million per g of dried lung tissue as a result of occupational asbestos exposure of relatively

No. 5

FIBERCONCENTRATION IN LUNGAND MESOTHELIOMA*

low intensity or short d ~ r a t i o nIn . ~ seven of the controls a fiber concentration between 1 and 5 million per g dried tissue were found. This observation reflects the occurrence of probable or possible occupational asbestos exposure in the source population. The current investigation shows that among all controls with unlikely or unknown occupational asbestos exposure, the lung fiber concentration was less than 1 million per g of dried tissue. Recent investigations have shown that, despite high chrysotile exposure, the amphiboles are the dominating fiber type in lung t i s s ~ e . ’ ~ ~We *~~ have * ’ inferred that the much higher deposition and retention of amphiboles in human lungs is probably the most important causal factor associated with malignant me~othelioma.~’ Based on this information, the fibers in this study were not classified according to fiber type. The difference between lung fiber concentrations among cases and nonexposed controls is apparent. The study further demonstrates an increased relative risk of malignant mesothelioma associated with low asbestos exposure corresponding to a critical level of only 1 million fibers per g of dried tissue. However, the number of both cases and controls is small, and the results should therefore be evaluated with caution. A fiber concentration of less than 1 million per g of dried tissue is usually not associated with occupational asbestos exposure, as it obviously occurs frequently among persons in the general population.’ In some of the cases the fiber concentrations overlapped those of the general p o p ~ l a t i o n .If~ ~ ~~ asbestos is considered as the causal agent even in cases with low fiber concentration in lung tissue, the results are also consistent with a no-threshold response. REFERENCES I . Churg A. Fiber counting and analysis in the diagnosis of asbestosrelated disease. Hum Pathol 1982; 13:381-392. 2. Morgan A, Holmes A. Concentrations and characteristics of amphibole fibres in the lungs of workers exposed to crocidolite in the British gas mask factories and elsewhere during the second world war. Br J Ind Med 1982; 39:62-69. 3. Mow6 G, Gylseth B, Hartveit F, Skaug V. Occupational asbestos exposure, lung fiber concentration and latency time in malignant mesothelioma. Scand J Work Environ Health 1984; 10:293-298. 4. Jones JSP, Roberts GH, Pooley FD et al. The pathology and mineral content of lungs in cases of mesothelioma in the United Kingdom in 1976. In: Wagner JC, Davis W, eds. Biological Effects of Mineral Fibres, vol. 1. Lyon: International Agency for Research on Cancer, 1980; 187-199. 5. McDonald AD, McDonald JC, Pooley FD. Mineral fibre content

MOW6

el

d.

1093

in mesothelial turnours in North America. Ann Occup Hyg 1982; 26: 41 7-42 I. 6. Wagner JC, Pooley FD, Berry G et al. A pathological and mineralogical study of asbestos-related deaths in the United Kingdom in 1977. Ann Occup Hyg 1982; 26:423-431. 7. Wagner JC, Berry G, Pooley FD. Mesothelioma and asbestos type in asbestos textile workers: A study of lung contents. Br Mcd J 1982; 285:603-606. 8. Fylkesplanen 1978. Bergen: Fylkesridmannen. 1978; 88-90. 9. Statistical Year Book of Norway 1974. Oslo: Central Bureau of Statistics of Norway, 1974; 10. 10. Statistical Year Book of Norway 1979. Oslo: Central Bureau of Statistics of Norway, 1979; 14. 1 1. Mow6 G. Time trend in the incidence. of malignant mesothelioma in Norway (1970-1979). Proceedings of the International Symposium on the Prevention of Occupational Cancer-International Symposium. In: Occupational Safety and Health Series, No 46. Geneva: International Labour Office, 1982; 213-219. 12. Gylseth B, Baunan RB, Bruun R. Analysis of inorganic fiber concentrations in biological samples by scanning electron microscopy. Scand J Work Environ Health 1981; 7:lOl-108. 13. Gylseth B, Rogers A, Overaae L, Summers G. An interlaboratory exchange of lung tissue samples from workers occupationally exposed to asbestos. In: Eustace IE, ed. XXI International Congress on Occupational Health, Abstract Book. Dublin: ETA Publications Ltd., 1984. 14. Gylseth B, Churg A, Davies JMG et a/. Analysis of asbestos fibers and asbestos bodies in human lung tissue samples: An international interlaboratory trial. Scand J Work Environ Health 1985; 11:107-1 I I . 15. Rogers AJ. Determination of mineral fibre in human lung tissue by light microscopy and transmission electron microscopy. Ann Occup Hyg 1984; 28~1-12. 16. Miettinen 0. Individual matching with multiple controls in the case of all-or-none responses. Biomefrics 1969; 29339-354. 17. Miettinen 0. Estimation of relative risk from individually matched series. Biometrics 1970; 26:75-86. 18. Miettinen 0. Estimability and estimation in case-referent studies. Am J Epidemiol 1976; 100:226-235. 19. Zielhuis RL, Versteeg JPJ, Planteijdt HT. Pleura mesothelioma and exposure to asbestos: A retrospective case-control study in the Netherlands. Inf Arch Occup Environ Health 1975; 36: 1-1 8. 20. Selikoff IJ, Lee DHK. Asbestos and disease. New York: Academic Press, 1978. 2 I . Pelnar PV. Non-Asbestos Related Malignant Mesothelioma: A Review of the Scientific and Medical Literature. Montreal: Canadian Asbestos Information Centre, 1983; 2-12. 22. Peterson JT, Greenberg SD, Buffler PA. Non-asbestos related malignant mesothelioma: A review. Cancer 1984; 54:95 1-960. 23. Rowlands N, Gibbs GW, McDonald AD. Asbestos fibres in the lungs of chrysotile miners and millers: A preliminary report. Ann OCCUPHyg 1982; 26141 1-415. 24. Morgan A, Holmes A. Distribution and characteristics of amphibiole asbestos fibres, measured with the light microscope, in the left lung of an insulation worker. Br J Ind Med 1983; 40:45-50. 25. Churg A, Woods P. Observation on the distribution of asbestos fibers in human lungs. Environ Res 1983; 31:374-380. 26. Gylseth B, Mow6 G, Skaug V, Wannag A. Inorganic fibers in human lung tissue from patients with pleural plaques or malignant mesothelioma. Scand J Work Environ Health 198 1 ; 7: 109-1 13. 27. Gylseth B, Mow6 G, Wannag A. Fibre type and concentration in the lungs of workers in an asbestos cement factory. Br J Ind Med 1983; 40~375-379. 28. Roggli VL, McGavran MH, Subach J , Sybers HD, Greenberg SD. Pulmonary asbestos body counts and electron probe analysis of asbestos body cores in patients with mesothelioma: A study of 25 cases. Cancer 1982; 50:2423-2433.