Sunlight exposure during leisure activities and risk of prostate cancer in Montréal, Canada, 2005-2009

Abstract

Doc number: 756

Abstract

Background: Prostate cancer (PCa) is the leading cause of cancer in men in many developed countries, but no modifiable risk factors have been identified. A handful of analytical studies have suggested a possible etiological role for sunlight exposure. We report here on the association between leisure-time sunlight exposure during adulthood and PCa risk in the context of a population-based case-control study.

Methods: In all, 1,904 PCa cases were ascertained across Montreal French hospitals between 2005 and 2009. Concurrently, 1,962 population controls, frequency matched to cases by age (±5 years), were selected from the electoral list for French-speakers in Greater Montreal. Interviews elicited the frequency of engagement in any leisure activity during adulthood. This was used to derive cumulative sunlight exposure indices: a cumulative number of leisure activities events entailing sunlight exposure and a cumulative duration of sunlight exposure during leisure activities. Unconditional logistic regression was conducted to yield odds ratios (OR) and 95% confidence intervals (CI) for estimating the association between sunlight exposure indices and PCa risk, adjusting for age, ancestry, family history of PCa, PCa screening, education, solar protection, body mass index and physical activity.

Results: Compared with men in the upper quartile category for the number of sunlight exposure events, men never exposed during leisure time had an OR of 1.32 (95% CI: 0.82-2.14). ORs were 1.11, 0.91 and 1.00 for the first to the third quartiles of exposure, respectively. Similar results were observed for cumulative duration of exposure to sunlight, and by PCa aggressiveness.

Conclusion: These findings provide little evidence of an association between sunlight exposure during leisure-time and PCa risk. Men with no sunlight exposure appeared at somewhat higher risks but none of the estimates achieved statistical significance.