Introduction: Physical activity (PA) may be a protect factor against prostate cancer (PCa). We aim to investigate and establish the potential relationship between leisure time PA and PCa. Evidence acquisition: PubMed, Embase databases were searched for eligible studies (cohort) before Aug-11, 2016. We conducted a meta-analysis to pool the dose-specific effects between leisure time PA and PCa. Subgroup analysis was employed to investigate the potential variation among each group. Sensitivity analysis was used to examine whether our results were stable. Evidence synthesis: Twenty-one cohort studies were identified, with a total of 803,872 subjects and 28,707 PCa cases. In our meta-analysis, no obvious evidence of non-linear relationship was observed. The RRs of each 500 MET-s/week increment of leisure time PA was 1.00 (95% CI: 0.99-1.01) for total PCa, 1.00 (95% CI: 0.98-1.01) for local PCa, and 1.00 (95% CI: 0.98-1.02) for advanced PCa. In subgroup analysis, for Americans, the RR of each 500 MET-s/week increment of leisure time PA was 1.00 (95% CI: 0.99-1.01); and for Europeans the RR was 1.00 (95% CI: 0.99-1.01). Sensitivity analysis suggested our results were stable. Conclusions: Current evidence based on observational studies suggests no obvious association between leisure time activity and risk of prostate cancer, regardless of total PCa, local PCa, or advanced PCa.