The two-stage clonal expansion model in occupational cancer epidemiology: results from three cohort studies

Our plain-English summary of The two-stage clonal expansion model in occupational cancer epidemiology: results from three cohort studies, by A. Zeka, R. Gore, D. Kriebel (Occupational and Environmental Medicine, 2011).

The study

• Authors: A. Zeka, R. Gore, D. Kriebel
• Journal: Occupational and Environmental Medicine
• Published: 2011;68(8):618-24
• Cited by: 10 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Objectives: The objective of this work was to apply the two-stage clonal expansion model, with the intention to expand the literature on epidemiological applications of the model and demonstrate the feasibility of incorporating biologically based modelling methods into the widely used retrospective cohort study.

Methods: The authors fitted the two-stage clonal expansion model model to three occupational cohort studies: (1) a cohort of textile workers exposed to asbestos and followed for lung cancer mortality; (2) a cohort of diatomaceous earth workers exposed to silica and also followed for lung cancer mortality; and (3) a cohort of automotive manufacturing workers exposed to straight metalworking fluid (MWF) and followed for larynx cancer incidence. The model allowed the authors to estimate exposure effects in three stages: cancer initiation (early effects), promotion or malignant transformation (late effects).

Results: In the first cohort, the authors found strong evidence for an early effect of asbestos on lung cancer risk. Findings from analyses of the second cohort suggested early and less evidently late effects of silica on lung cancer risk. In the MWF (third) cohort, there was only weak evidence of straight MWF exposure effects on both early and late stages. The authors also observed a late birth cohort effect on larynx cancer risk.

Conclusions: The findings for asbestos and silica were essentially confirmatory, supporting evidence for their early effects on lung cancer from a large body of literature. The effect of straight MWF on larynx cancer was less clear.

In plain English

This study reanalysed three existing groups of workers using a mathematical model of cancer development, the two-stage clonal expansion model. The model tests whether an exposure fits the data better as acting early (initiation) or late (promotion or malignant transformation) in cancer development. The asbestos group was South Carolina asbestos textile factory workers employed at least 1 month between 1940 and 1965, followed to 1990, with yearly exposure estimates per worker. Among these workers, models converged only when asbestos exposure was allowed to affect the parameter the authors linked to early effects. The best of these improved the fit measure (minus two log likelihood) by 9.0 over a model without exposure, which the authors called substantial. Models with exposure acting on the other parameters fitted poorly. In the other two groups, silica-exposed diatomaceous earth workers and autoworkers exposed to straight metalworking fluid, adding exposure improved the fit by only about 2. The authors concluded that asbestos and silica appear to act early in lung cancer development. They noted that smoking information was not available for all three cohorts.

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: Reanalysis of three existing occupational cohort datasets, fitting a biologically based two-stage clonal expansion (TSCE) cancer model that estimates exposure effects on cancer initiation (early effects) or on promotion or malignant transformation (late effects); described as methods development research
• Who was studied: Asbestos cohort: all workers in asbestos textile operations at a factory in South Carolina, employed at least 1 month between 1 January 1940 and 31 December 1965, followed for vital status through 31 December 1990. The other two cohorts were diatomaceous earth workers (silica) and automotive parts workers (metalworking fluid).
• Exposure: Asbestos in textile operations, treated as lifetime cumulative exposure built from annual exposure estimates for each worker. The body of the article calls the exposure only 'asbestos' and does not name a fiber type.
• Main results:
  • Asbestos cohort: models converged only when exposure was allowed to modify parameter C, the parameter the authors associated with early (initiation) effects.
  • Asbestos cohort: the best model (log exposure form acting on C) had a minus two log likelihood of 928.1, compared with 937.1 for the baseline model without exposure. The authors called this reduction of 9.0, with one degree of freedom, substantial; by their own scale a reduction above 6.6 counts as strong evidence.
  • Asbestos cohort: models in which exposure modified A (net proliferation rate of intermediate cells) or B (second mutation rate) fitted poorly.
  • Asbestos cohort: the authors reported strong evidence for an early effect and no evidence for a late-stage effect.
  • Silica cohort (diatomaceous earth workers, lung cancer): the best model, a linear exposure effect on C, had a minus two log likelihood of 1045.9, compared with 1048.0 for the baseline model, a reduction of about 2.
  • Metalworking fluid cohort (larynx cancer): exposure models improved the fit by at most about 2. Year of birth (birth cohort) showed strong evidence of an effect.
• The authors' conclusion: The authors concluded that both asbestos and silica appear to act early in the process of lung cancer development. They described these findings as essentially confirmatory of earlier literature, and found the effect of straight metalworking fluid on larynx cancer less clear.
• Limitations the authors state:
  • Smoking information, the main risk factor for lung and laryngeal cancer, was not available for all three cohorts.
  • No formal hypothesis testing was performed.
  • Not all models converged; the authors treated non-convergence as poor fit.
  • The underlying biological rates cannot be estimated directly from incidence or mortality data, so combined parameters were estimated instead.
  • Most cancers involve more than two irreversible steps, so a two-stage model simplifies the real process.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer.

Medical terms in this study, explained in our Lexicon: Epidemiology, Cohort Study, Lung Cancer, Laryngeal Cancer, Malignant Transformation.

Related studies

• Lung cancer in chrysotile asbestos workers: analyses based on the two-stage clonal expansion model — D.B. Richardson (2009)
• Phase 1 cohort expansion study of LY3023414, a dual PI3K/mTOR inhibitor, in patients with advanced mesothelioma — M.G. Zauderer, E.W. Alley, et al. (2021)
• Tumor suppressor alterations cooperate to drive aggressive mesotheliomas with enriched cancer stem cells via a p53-miR-34a-c-Met axis — C.W. Menges, Y. Kadariya, et al. (2014)
• Multipotent cancer stem cells derived from human malignant peritoneal mesothelioma promote tumorigenesis — S. Varghese, R. Whipple, et al. (2012)
• Mesothelioma mortality in asbestos workers: implications for models of carcinogenesis and risk assessment — J. Peto, H. Seidman, et al. (1982)
• A 3-dimensional in vitro model of epithelioid granulomas induced by high aspect ratio nanomaterials — V.C. Sanchez, P. Weston, et al. (2011)

Browse more research in Occupational and Environmental Medicine, or return to medical research.