BACKGROUND: Whether metformin therapy affects incident prostate cancer risk in Asian patients with type 2 diabetes mellitus (T2DM) has not been investigated. METHODS: The National Health Insurance reimbursement database of Taiwanese male patients with new-onset T2DM between 1998 and 2002 and aged 40years (n=395,481) were retrieved to follow up prostate cancer incidence until the end of 2009. Metformin was treated as a time-dependent variable. Of the patients studied, 209,269 were never-users and 186,212 were ever-users. A time-dependent approach was used to calculate prostate cancer incidence and estimate hazard ratios using Cox regression for ever-users, never-users and subgroups of metformin exposure (tertiles of cumulative duration and cumulative dose). Sensitivity analyses were conducted in various subgroups, using time-dependent and non-time-dependent approaches. RESULTS: During the follow-up, 2776 metformin ever-users and 9642 never-users developed prostate cancer, representing an incidence of 239.42 and 737.10 per 100,000 person-years, respectively. The hazard ratio (95% confidence intervals) after adjustment for propensity score (PS) for ever- versus never-users was 0.467 (0.446-0.488). The PS-adjusted hazard ratios for the first, second and third tertiles of cumulative duration of metformin therapy were 0.741 (0.698-0.786), 0.474 (0.441-0.508) and 0.231 (0.212-0.253), respectively (P-trend<0.001); and were 0.742 (0.700-0.786), 0.436 (0.406-0.468) and 0.228 (0.208-0.251) for the respective cumulative dose (P-trend<0.001). Sensitivity analyses consistently supported a protective effect of metformin on incident prostate cancer. CONCLUSIONS: Metformin use is associated with a decreased risk of incident prostate cancer in Taiwanese male patients with T2DM.
Date:
2014-11
Relation:
European Journal of Cancer. 2014 Nov;52(16):2831-2837.
