Benjamin Kendzia, Ann Olsson, Jack Siemiatycki, Karl-Heinz Jöckel, Kurt Straif, Joachim Schüz, Hans Kromhout, Thomas Brüning, Thomas Behrens, and the SYNERGY study group

Abstract

Objective: Socioeconomic status (SES) is an important confounder of the association between occupational exposures and lung cancer. However, the extent of SES effects on lung cancer, including mediating effects via smoking habits, remained unclear as several biases were often indicated. Therefore, we aimed to quantify the impact of potential biases on the SES-lung cancer association.

Material and Methods: We used data from the SYNERGY project (https://synergy.iarc.who.int), including 16,550 cases and 20,147 controls from European and Canadian case-control studies, to investigate bias effects in the association of occupational SES (in quartiles) and lung cancer. We estimated odds ratios (OR) with 95% confidence intervals (CI) by logistic regression adjusting for age, study centre, and smoking, stratified by sex. We also estimated natural direct SES effects and natural indirect effects via smoking by inverse odds ratio weighting. In a quantitative bias analysis, we considered impacts of misclassification of smoking status, selection bias, and unmeasured mediator-outcome confounding by genetic disposition, calculating 95% simulation intervals (SI) by bootstrap (n=2500 repetitions).

Results: Adjustment for smoking as well as natural effects estimation showed that nearly 50% of lung cancer risks for lower SES groups in men and up to one third in women were attributable to smoking. Consideration of all types of bias reduced lung-cancer risks in the fully adjusted logistic regression models: For the 4th versus 1st (highest) SES quartile OR decreased from 1.83 (1.69-1.98 CI) to 1.50 (1.32-1.69 SI) in men, and OR 1.48 (1.27-1.72 CI) to 1.20 (1.01-1.45 SI) in women.

Conclusion: Direct lung-cancer risks of lower SES groups were lowered by multiple bias adjustment but remained elevated. As long as it remains unclear to what extent these effects are attributable to occupational hazards, SES should be considered in the analysis of occupational exposures and lung cancer.

Leon AM Berge, Niki Marjerrison, Nita K Shala, Ronnie Babigumira, Tom K Grimsrud, Marit B Veierød, Jo S Stenehjem

Abstract

Objectives: Benzene has been recognised as a cause of acute myeloid leukaemia by the International Agency for Research on Cancer (IARC) since 1979. In their 2017 evaluation of benzene, IARC concluded that positive associations were observed also for other histological subtypes of lymphohematopoietic (LH) cancers: chronic myeloid leukaemia, non-Hodgkin lymphoma, chronic lymphoid leukaemia, and multiple myeloma. There is a need to further clarify benzene’s association with these LH cancer subtypes, using causal inference methodology to assess the role of exposure time-windows. We aim to investigate the time-dependent effect of benzene exposure on LH cancer subtypes in offshore petroleum workers using a target trial emulation framework.

Materials and methods: We will use the Norwegian Offshore Petroleum Workers cohort and the Heliport cohort, encompassing 100,000 workers employed offshore 1965–2023. Inclusion criteria will be specified according to the causal roadmap framework. Between 1999–2023, 823 LH cancer cases were identified by linkage to the Cancer Registry of Norway. Start of follow-up (time zero) will be 1999 or at the start of employment and continue until cancer diagnosis, death, or end of follow-up. Benzene exposure will be assessed using an expert-developed job-exposure matrix. Data on potential confounding factors (e.g. smoking and non-offshore benzene exposure) will be retrieved from questionnaire and register data by linking the cohort to national registries.

Results: Results will be presented as cumulative incidence (absolute and relative risks) calculated by multi-state models accounting for death as a competing event and adjusted for confounders using inverse probability treatment weights producing propensity scores.

Conclusion: These cohorts, with time-dependent benzene exposure, enable application of a target trial emulation framework, which may provide new insights and contribute to important clarifications regarding benzene exposure and risk of the major histological subtypes of LH cancers, and hence follow up on the latest IARC evaluation of benzene.

Dario Consonni, Slvia Fustinoni

Abstract

Objective: Tetrafluoroethylene (TFE) and perfluorooctanoic acid (PFOA) are used for the synthesis of fluoropolymers. The International Agency for Research on Cancer classified TFE as probably carcinogenic and PFOA as carcinogenic to humans (based on limited evidence for kidney and testis cancer). Within a multicentre cohort study, mortality (1960-2008) had been evaluated of workers employed (1960-1999) in a plant producing fluoropolymers in Piedmont, North West Italy. We updated the follow-up through 2023.

Material and Methods: Workers’ job histories were merged with a semi-quantitative job-exposure matrix (JEM) to calculate cumulative exposure to TFE (still used) and PFOA (phased-out in 2013). Person-years were calculated for the period 1960-2023. Expected deaths were calculated using national and regional reference rates. We calculated standardised mortality ratios (SMR) and 95% confidence intervals (CI).

Results: Among 505 male workers elevated SMRs (reference: regional rates) were found for cancer of lung (SMR 1.50, CI: 1.00-2.26, 23 deaths) and testis (SMR 7.03, CI: 0.99-49.9, 1 death), and for non-Hodgkin’s lymphoma (NHL, SMR 2.84, CI 1.28-6.33, 6 deaths). Lung cancer morality was not associated with length of employment or with exposure to TFE or PFOA. The testis cancer death occurred among workers exposed to TFE and unexposed to PFOA. NHL was positively associated with length of employment, time since first employment, and cumulative exposure to TFE (all 6 deaths exposed), while 3 deaths occurred among workers unexposed to PFOA. The NHL excess occurred among older workers (65+ years) first employed before 1980 and last employed before 2000. Pleural cancer was also in excess (SMR 3.33 using national rates, CI: 1.07-10.3, 3 deaths).

Conclusion: This study found elevated mortality from NHL. The available information suggests an association with TFE rather than PFOA exposure, although exposure misclassification cannot be ruled out.