First analysis of associations between residential radon exposure and cancer risks in the CONSTANCES cohort

Radon exposure is an established risk factor for lung cancer, but its potential role in risks of other cancers remains unclear. A recent systematic review recommended more in-depth studies of specific cancer sites as potential long-term effects of radon exposure, especially during childhood. Here we investigated the association between lifelong residential radon exposure and the risks of various cancers, and the potential role of age at exposure. Study participants were selected from the French population-based CONSTANCES cohort recruited at adult age over the period 2012-2019. Radon exposure was reconstructed by linking lifelong residential histories with estimated municipality-level indoor radon concentrations. We fitted for each cancer site studied a time-dependent Cox proportional hazards model to assess the association between cumulative annual average residential radon exposure and the age at diagnosis of specific primary incident cancer, expressed as a Hazard Ratio per 1000 Bq.m^- 3-years exposure and the corresponding 95% confidence interval. The models were adjusted for birth cohort, diploma, time-varying region of residence, and where appropriate, time-varying smoking status. At the end of follow-up in year 2022, 7433 (12%) participants over 61,366 were diagnosed with incident primary cancer. The median age at time of censoring was 55 years old (Q1, Q3 = 44, 65). Based on current follow-up, no robust association was observed between radon exposure and any of the targeted cancers. The main limitations of this study to date are the semi-ecological approach used for exposure reconstruction which should be improved in the future, and the still limited duration of follow-up after inclusion. The richness of the CONSTANCES cohort combined with extended follow-up and planned improvements in radon exposure assessment will further increase the potential of future investigations.