Asbestos as a systemic carcinogen: the evidence from eleven cohorts

Our plain-English summary of Asbestos as a systemic carcinogen: the evidence from eleven cohorts, by J.R. Goldsmith (American Journal of Industrial Medicine, 1982).

The study

• Author: J.R. Goldsmith
• Journal: American Journal of Industrial Medicine
• Published: 1982;3(3):341-8
• Cited by: 19 articles, as counted by NIH iCite
• Read it: PubMed · Publisher
• In J-M's library: Johns-Manville's company medical library held a copy (J-M no. 001759).

Abstract

Most known occupational carcinogens are site-specific, which implies that they are "complete" carcinogens with both "initiating" and "promoting" properties (Berenblum's terminology). Excess cancer at gastrointestinal sites among cohorts occupationally exposed to asbestos has been interpreted as reflecting additional site-specific effects, although excess at other sites has also been observed in some studies. The hypothesis that excess cancer at gastrointestinal sites cannot be distinguished from excess cancer at all nonpulmonary sites is tested by data from New York-New Jersey insulation workers working in 1943; similar workers employed after 1943; U.S.-Canadian insulation workers; London factory workers, male and female; Quebec miners and millers; retired U.S. factory workers; U.S. shipyard insulators; Italian shipyard workers in Genoa; Amosite factory workers; and U.S. factory workers. Excluding lung cancer and mesothelioma, observed-expected ratios for nonpulmonary cancer mortality range from 0.97 to 2.78, and do not differ significantly from gastrointestinal ratios. A dose-response gradient is observed for both ratios, when dose is estimated from lung cancer ratios, or in some studies, measured exposures. Site-specificity is unlikely for nonpulmonary cancer associated with asbestos exposure more than 20 years previously. Systemic carcinogenesis may be an example of promotion or impairment by asbestos of some cancer defense mechanism; immunological mechanisms have been suggested by Turner-Warwick and Parkes. Prospective studies are indicated.

In plain English

This single-author paper re-examined published cancer death data from 11 groups of asbestos-exposed workers in four countries, totalling 32,503 people. They included insulation workers, London asbestos factory workers, Quebec miners and millers, retired US factory workers, shipyard workers in the US and Italy, and amosite and other US factory workers. Leaving out lung cancer and mesothelioma, the author tested whether excess gastrointestinal cancer could be told apart from excess cancer at other sites. The paper gave observed-to-expected ratios for these cancers of 0.97 to 2.78, though its own table showed 0.92 for London men with low exposure for two years or more. Combined, there were 582 gastrointestinal cancer deaths against 445.9 expected, and 906 other-site cancer deaths against 717.1 expected; the difference was not statistically significant. The author counted 325 excess nonpulmonary cancer deaths, 777 excess lung cancer deaths, 322 mesothelioma deaths and an estimated 421 asbestosis deaths, partly approximated from excess chronic respiratory disease deaths. The author concluded that a systemic carcinogenic role for asbestos was more likely than site-specific effects, and called for prospective studies.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Re-analysis of previously published mortality data from several occupational cohorts, testing the hypothesis that gastrointestinal cancer increases are not distinguishable from increases at all nonpulmonary sites
• Who was studied: 32,503 asbestos-exposed workers in total, from four countries. The groups were New York-New Jersey insulation workers (623 employed as of 1943, and 523 first employed after 1943), 11,662 US-Canada insulation workers, London asbestos factory workers (2,887 men, 1,268 laggers and mates, 693 women), 10,939 Quebec miners and millers, 1,075 retired US asbestos company workers, US shipyard insulators, 2,190 Genoa shipyard workers, 933 amosite factory workers, and 689 US asbestos factory workers.
• Exposure: Occupational asbestos exposure in insulation work, asbestos factory production, mining and milling, shipyard work and amosite factory work, with varying types and durations. London exposures were classed as low-moderate or severe, and as less than or more than two years. Quebec exposures were graded by cumulative dust (millions of particles per cubic foot-years). In the Italian group, asbestos exposure was inferred.
• Main results:
  • Excluding lung cancer and mesothelioma, the paper stated that observed-to-expected ratios for nonpulmonary cancer deaths ranged from 0.97 to 2.78 and did not differ significantly from the gastrointestinal ratios.
  • Table I nevertheless lists an all-nonpulmonary O/E ratio of 0.92 for London male production workers with low exposure for two years or more (12 observed against 13.1 expected). This is below the stated lower bound of 0.97.
  • When each group was tested separately, in no case did gastrointestinal and other sites differ from all nonpulmonary sites at a probability of 0.1 or less. The test was not significant for the total either.
  • All cohorts combined: gastrointestinal sites had 582 observed deaths against 445.9 expected, and other nonpulmonary sites had 906 observed against 717.1 expected. All nonpulmonary sites together had 1488 observed against 1163 expected (O/E 1.279). The chi-square for the difference was 0.37 with 1 degree of freedom, not significant.
  • The ratio for gastrointestinal sites was higher in 13 of the 19 comparisons.
  • Significant elevations of nonpulmonary cancer were found in five of the nine groups with the largest numbers of observed nonpulmonary cancers. Of the groups with 30 or more observed cases, only the Italian shipyard workers and the Quebec miners did not show a significant excess.
  • Across all groups, there were 325 excess nonpulmonary cancer deaths: 136 at gastrointestinal sites and 189 at other sites.
  • For comparison, there were 777 excess respiratory (lung) cancer deaths (1220 observed against 443 expected) and 322 mesothelioma deaths. There were an estimated 421 asbestosis deaths. For the London and Italian groups, this estimate was approximated from excess chronic respiratory disease deaths.
  • Lung cancer O/E ratios were correlated with nonpulmonary cancer O/E ratios (r = 0.807).
  • The London data for both sexes appeared to show a crude dose-response relationship, and the Quebec data also suggested this.
• The authors' conclusion: The author concluded that the data cast doubt on site-specificity of asbestos-related cancer, and that a systemic carcinogenic role was more likely. A long-acting effect on cancer defense, possibly involving immune mechanisms, could not be excluded. The author called for prospective studies of exposed persons whose immune status can be assessed.
• Limitations the authors state:
  • In the Italian shipyard study, both the exposures and the duration of follow-up were uncertain.
  • The Italian study showed no data for mesothelioma, which may have been included in respiratory or other cancer data. The mesothelioma count may therefore be a slight underestimate.
  • For the London and Italian groups, asbestosis deaths were not reported. They were approximated from excess chronic respiratory disease deaths, and for the Italian workers this could include emphysema or other pneumoconioses.
  • Not all of the London male factory workers had more than 20 years of exposure, though all had at least 10.
  • The early excess of gastrointestinal cancer might reflect the many asbestos workers who lived in or came from countries with high stomach cancer rates. If comparison populations had been matched by country of origin, the higher gastrointestinal ratios might not have been seen.

Topics

Medical terms in this study, explained in our Lexicon: Insulator, Amosite, Lung Cancer, Mesothelioma, Dose-response relationship, Asbestos Exposure, Carcinogenesis.

Related studies

• High prevalence of occupational exposure to solvents or silica in male systemic sclerosis patients: a Belgian cohort analysis — E. De Decker, M. Vanthuyne, et al. (2018)
• A single-center retrospective cohort study of perioperative systemic chemotherapy in diffuse malignant peritoneal mesothelioma — X. Wang, S. Katz, et al. (2022)
• Mechanisms of multistep carcinogenesis and carcinogen risk assessment — J.C. Barrett (1993)
• Human carcinogens so far identified — L. Tomatis, A. Aitio, et al. (1989)
• Risk-based indicators of Canadians' exposures to environmental carcinogens — E. Setton, P. Hystad, et al. (2013)
• The future excess fraction of occupational cancer among those exposed to carcinogens at work in Australia in 2012 — R.N. Carey, S.J. Hutchings, et al. (2017)

Browse more research in American Journal of Industrial Medicine, or return to medical research.