Lung cancer risk at low cumulative asbestos exposure: meta-regression of the exposure-response relationship

Our plain-English summary of Lung cancer risk at low cumulative asbestos exposure: meta-regression of the exposure-response relationship, by S. van der Bij, H. Koffijberg, V. Lenters, et al (Cancer Causes & Control, 2013).

The study

• Authors: S. van der Bij, H. Koffijberg, V. Lenters, L. Portengen, K.G. Moons, D. Heederik, R.C. Vermeulen
• Journal: Cancer Causes & Control
• Published: 2013;24(1):1-12
• Type of article: Meta-analysis, Review
• Cited by: 27 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Purpose: Existing estimated lung cancer risks per unit of asbestos exposure are mainly based on, and applicable to, high exposure levels. To assess the risk at low cumulative asbestos exposure, we provide new evidence by fitting flexible meta-regression models, a notably new and more robust method.

Methods: Studies were selected if lung cancer risk per cumulative asbestos exposure in at least two exposure categories was reported. From these studies (n = 19), we extracted 104 risk estimates over a cumulative exposure range of 0.11-4,710 f-y/ml. We fitted linear and natural spline meta-regression models to these risk estimates. A natural spline allows risks to vary nonlinearly with exposure, such that estimates at low exposure are less affected by estimates in the upper exposure categories. Associated relative risks (RRs) were calculated for several low cumulative asbestos exposures.

Results: A natural spline model fitted our data best. With this model, the relative lung cancer risk for cumulative exposure levels of 4 and 40 f-y/ml was estimated between 1.013 and 1.027, and 1.13 and 1.30, respectively. After stratification by fiber type, a non-significant three- to fourfold difference in RRs between chrysotile and amphibole fibers was found for exposures below 40 f-y/ml. Fiber-type-specific risk estimates were strongly influenced by a few studies.

Conclusions: The natural spline regression model indicates that at lower asbestos exposure levels, the increase in RR of lung cancer due to asbestos exposure may be larger than expected from previous meta-analyses. Observed potency differences between different fiber types are lower than the generally held consensus. Low-exposed industrial or population-based cohorts with quantitative estimates of asbestos exposure a required to substantiate the risk estimates at low exposure levels from our new, flexible meta-regression.

In plain English

This review article was a meta-regression combining 19 earlier studies of asbestos and lung cancer. Eighteen were industry-based cohort studies (one a nested case–control study) and one was a population-based case–control study. The authors took 104 risk estimates covering cumulative exposures of 0.11 to 4,710 fiber-years per milliliter (f-y/ml). A flexible curve called a natural spline fit better than straight-line models. With it, the relative risk was 1.013 (95% CI 0.791–1.296) at 4 f-y/ml and 1.133 (95% CI 0.888–1.444) at 40 f-y/ml. A version without an intercept gave 1.027 and 1.301, with confidence intervals above 1. In the intercept-corrected model, the rise in risk above 1 was three to four times larger for amphibole and mixed fibers than chrysotile: at 4 f-y/ml, relative risks were 1.006 for chrysotile, 1.022 for amphiboles and 1.018 for mixed fibers. This difference was not statistically significant and was strongly influenced by a few studies. The authors concluded that at low cumulative exposures the increase in lung cancer risk may be larger than expected. They noted possible publication bias and limited smoking information.

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: Review article: meta-regression of published exposure-response data, comparing linear and natural spline models
• Who was studied: 19 published studies: 18 industry-based cohort studies (one of them a nested case–control study) and one general population-based case–control study (Stockholm County)
• Exposure: Cumulative asbestos exposure in fiber-years per milliliter (f-y/ml); 104 risk estimates spanning 0.11 to 4,710 f-y/ml; results also split by fiber type (chrysotile, amphibole, or mixed)
• Main results:
  • The natural spline model with intercept fit better than the linear model (deviance 105.6 versus 111.7; p = 0.01)
  • Natural spline model, corrected for intercept, all studies combined: relative risk 1.013 (95% CI 0.791–1.296) at 4 f-y/ml and 1.133 (95% CI 0.888–1.444) at 40 f-y/ml
  • Natural spline model fitted without intercept, all studies combined: relative risk 1.027 (95% CI 1.020–1.034) at 4 f-y/ml and 1.301 (95% CI 1.215–1.392) at 40 f-y/ml
  • Natural spline model corrected for intercept, by fiber type: relative risk 1.022 at 4 f-y/ml and 1.232 at 40 f-y/ml for amphiboles, 1.006 and 1.064 for chrysotile, and 1.018 and 1.194 for mixed fibers
  • The authors described the difference between fiber types as a non-significant three- to fourfold difference in potency (the increase in risk per unit of exposure) between chrysotile and amphibole fibers
  • The fiber-type potency differences were strongly influenced by a few studies
  • In the spline model without intercept, amphiboles had a statistically significant 8 to 12-fold higher risk than chrysotile; the authors described this exposure-response relationship as seemingly unrealistic and uncertain at higher exposures
  • Predicted risk was about three times higher in studies that used a 10-year latency (RR 1.030 at 4 f-y/ml and 1.329 at 40 f-y/ml) than in studies using no latency
  • The intercepts suggested about a 50% excess lung cancer risk among workers compared with the general population that is attributable to factors other than asbestos; the authors considered this at least partly due to measurement error
• The authors' conclusion: The authors concluded that for low cumulative exposures, the increase in relative risk of lung cancer due to asbestos may be larger than expected from previous results, and that additional research is required
• Limitations the authors state:
  • Correcting for the intercept assumes it fully reflects a difference in baseline risk, which may not be true
  • Publication bias may have affected the risk estimates
  • Limited information on smoking prevalence and length of follow-up prevented exploring the bias these may cause
  • Risks could not be examined by exposure intensity, only by cumulative exposure
  • Because of differences in study quality, the size of potency differences between fiber types at low exposure could not easily be determined

Topics

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

Medical terms in this study, explained in our Lexicon: Lung Cancer, Asbestos Exposure, Cumulative Exposure, Relative Risk, Chrysotile, Amphibole Group.

Related studies

• A meta-analysis of asbestos and lung cancer: is better quality exposure assessment associated with steeper slopes of the exposure-response relationships? — V. Lenters, R. Vermeulen, et al. (2011)
• Risk of mesothelioma after cessation of asbestos exposure: a systematic review and meta-regression — P. Boffetta, F. Donato, et al. (2019)
• Potential pitfall in using cumulative exposure in exposure-response relationships: demonstration and discussion — M.M. Finkelstein (1995)
• A meta-analysis of the relation between cumulative exposure to asbestos and relative risk of lung cancer — T.L. Lash, E.A. Crouch, et al. (1997)
• Health Effects of Occupational and Environmental Exposures to Nuclear Power Plants: A Meta-Analysis and Meta-Regression — R.T. Lin, H. Boonhat, et al. (2024)
• A meta-analysis of asbestos-related cancer risk that addresses fiber size and mineral type — D.W. Berman, K.S. Crump (2008)

Browse more research in Cancer Causes & Control, or return to medical research.