Metformin and prostate cancer mortality: a meta-analysis

Purpose Observational studies report conflicting results on the association between metformin exposure and prostate cancer outcomes. This meta-analysis summarizes studies reporting overall survival, prostate cancer-specific mortality, and biochemical recurrence. Methods PubMed and Embase were systematically reviewed to identify studies investigating the association between metformin use and clinical endpoints among men with prostate cancer while taking confounding by diabetes diagnosis into account. Pooled risk estimates (hazard ratios, HRs) and 95 % confidence intervals (CIs) were calculated using random-effects models. Sensitivity analyses for quality components and factors for heterogeneity were conducted. Results Of 549 articles identified, nine retrospective cohort studies representing 9,186 patients were included. There was significant heterogeneity between studies, and studies differed in quality. Metformin use was associated with improved overall survival in studies with clear risk window definition (HR 0.88, 95 % CI 0.86-0.90, p < 0.001) and in studies with potential immortal time bias (HR 0.52, 95 % CI 0.41-0.65, p < 0.001). No significant association with prostate cancer-specific mortality was detected (HR 0.76, 95 % CI 0.44-1.31, p = 0.33). Metformin use was associated with a decreased risk of biochemical recurrence (HR 0.79, 95 % CI 0.63-1.00, p = 0.047). Conclusions This meta-analysis suggests a benefit of metformin in men with diabetes and prostate cancer. However, further carefully designed studies are needed to confirm findings and to assess potential generalization to non-diabetic, non-white, and less aggressively treated men with prostate cancer.