O8A.5 Total energy expenditure and risk of lymphoma subtypes: results from the european epilymph case-control study

IntroductionPhysical activity is known to convey protection against several cancers, but results on risk of lymphoma and its subtypes have been inconsistent. A possible reason might be confounding by workplace exposures associated with occupational energy expenditure, which was not considered in studies of recreational physical activity. It is also unclear whether energy expenditure acts directly, or through preventing obesity.ObjectivesTo investigate the role of energy expenditure, including occupational and recreational physical activity, on risk of lymphoma subtypes.Materials and methodsBased on questionnaire information on lifetime recreational physical activity and lifetime occupational history available for all participants to the multicenter EpiLymph case-control study, we estimated energy expenditure at work by occupational ISCO68 code, and we applied it to the work histories of study subjects. We also categorized hours of lifetime recreational physical activity into quartiles. We calculated risk of lymphoma subtypes with unconditional polytomous regression analysis, associated with increasing categories of lifetime energy expenditure at work (EEW), increasing categories of recreational physical activity (RPA), and their interaction term (total energy expenditure, TEE), adjusting by age, gender, education, body mass index (BMI), and history of farm work and solvents use.ResultsRisk of lymphoma overall, diffuse large B cell lymphoma and multiple myeloma was not associated with EEW, RPA and TEE. Risk of follicular lymphoma and chronic lymphocytic leukemia associated were elevated with medium and high EEW (OR 3.1, 95% CI 1.5–6.1; (OR 2.5 95% CI 1.2–5.1, respectively), but there was not a significant upward trend.ConclusionsFurther epidemiologic and mechanistic research is warranted to assess the role of physical activity in the etiology of lymphoma subtypes. New standardized energy expenditure assessment methods, as the ones herein developed, might contribute to a better understanding of the nature of the observed inconsistent findings.