This study aimed to investigate whether testosterone mediates or confounds the effect of obesity-related traits on prostate cancer (PCa) using Mendelian randomization (MR) analysis.Data of obesity-related traits (body mass index [BMI], waist-to-hip ratio [WHR], and waist-to-hip ratio adjusted for body mass index [WHRadjBMI]) were obtained from up to 806,834 people of European ancestry; data of testosterone (bioavailable testosterone [BT], total testosterone [TT], and sex hormone-binding globulin [SHBG]) were extracted from up to 194,453 participants in the UK Biobank; and the summary-level data of PCa (79,194 cases and 61,112 controls) were obtained from the PRACTICAL consortium.The results supported the causal relationship between higher BMI and a reduced risk of PCa (OR = 0.91, 95% confidence interval [CI]: 0.86-0.96). Furthermore, increased BT levels were associated with an elevated risk of PCa (OR = 1.15, 95% CI: 1.06-1.24). Importantly, our analysis revealed a unidirectional causal effect-higher BMI was linked to lower BT levels (beta = -0.27, 95% CI: -0.3--0.24), but not the other way around. This suggests that BT may mediate the effect of BMI on PCa rather than confound it. Our multivariable MR results further demonstrated that considering BT as a mediator led to the weakening of BMI's effect on PCa risk (OR = 0.97, 95% CI: 0.90-1.05), while the impact of BT on PCa remained unchanged when accounting for BMI. Moreover, we identified a significant indirect effect of BMI on PCa risk (OR = 0.96, 95% CI: 0.94-0.98).Our study provided genetic evidence that serum BT can mediate the effect of BMI on the risk of PCa, indicating the possible mechanism by which obesity reduces PCa risk.