Causal association of physical activity with lymphoma risk: a Mendelian randomization analysis.

Abstract

Controversial relationship of physical activity with lower lymphoma risk has been reported in observational studies. The purpose of this study was to explore the causal correlation of physical activity with lymphoma risk using two-sample Mendelian randomization (MR). Genetic variants associated with physical activity (moderate-to-vigorous physical activity (MVPA), average acceleration physical activity, number of days/week of moderate physical activity 10+ min, and number of days/week of vigorous physical activity 10+ min) and lymphoma [overall lymphoma, Hodgkin lymphoma, mature T/NK-cell lymphomas, diffuse large B-cell lymphoma (DLBCL), and follicular lymphoma] were obtained from published genome-wide association studies (GWAS) and the FinnGen database and used as instrumental variables. Primary results were based on inverse variance-weighted (IVW) analysis and were described as odds ratio (OR) and 95% confidence interval (CI). Higher levels of genetically predicted MVPA (OR = 0.079, 95% CI: 0.021-0.300, P = 0.0002) and number of days/week of vigorous physical activity 10+ min (OR = 0.237, 95% CI: 0.098-0.573, P = 0.0014) were negatively associated with Hodgkin lymphoma risk. There was a weak negative association between high levels of genetically predicted MVPA (OR = 0.114, 95% CI: 0.015-0.856, P = 0.0348) and average acceleration physical activity (OR = 0.830, 95% CI: 0.705-0.976, P = 0.0243) and risk of DLBCL. No causal relationship was observed between physical activity and the risk of overall lymphoma, mature T/NK-cell lymphomas, and follicular lymphoma (P > 0.05). This study supported the causal relationship between higher physical activity levels and lower risks of Hodgkin lymphoma and DLBCL.