Metformin improves prostate cancer-specific survival and inhibits the development of castrate resistant metastasis

Purpose/Objective(s): Metformin has been shown in population studies to decrease the incidence of prostate cancer. Pre-clinical data suggests that metformin has the ability to inhibit prostate cancer cell growth in vitro via the activation of AMP-kinase pathway and inhibition of mammalian target of rapamycin. Metformin also ameliorates hyperinsulinemia, which may play a role in reducing the development of castrate-resistant prostate cancer (CRPC). The aim of our study was to assess whether the use of metformin in patients with prostate cancer was associated with improved clinical outcomes and reduction in CRPC. Materials/Methods: We screened 2,901 patients with biopsy-proven prostate cancer that were treated with external beam radiation therapy from 1988 to 2008 at a single institution. A total of 157 patients were identified using metformin. The non-metformin control arm consisted of 2,744 consecutive patients. Within the non-metformin control arm, a subgroup of 159 diabetic patients not using metformin was identified. Biochemical relapse-free survival (bRFS), distant metastasis-free survival (DMFS), and prostate cancer-specific survival (CSS) outcomes were analyzed. In addition, development of CRPC was compared between patients with diabetes taking metformin and those who were not. Results: Median follow-up was 13.4 years. After correcting for stage, Gleason score, PSA, androgen deprivation therapy use, dose of radiation, and diabetes, patients taking metformin compared with non-metformin users, the 10-year actuarial rate of bRFS was 67.9% versus 65% (p = 0.006; hazard ratio [HR] = 0.51 [0.32-0.83]), DMFS was 90.3% versus 85.1% (p = 0.001; HR = 0.26 [0.12-0.56]), and CSS was 96.2% versus 92.2% (p = 0.012; HR = 0.26 [0.09-0.75]). The improvements seen were in spite of diabetes predicting for significantly worse DMFS (p < 0.001, HR = 2.03) and CSS (p = 0.002, HR = 1.95). On subgroup analysis of diabetic patients alone, the metformin arm significantly reduced the development of CRPC with our multivariate model (p = 0.001). Conclusions: This is the first study to report an improvement in prostate cancer-specific outcomes including bRFS, DMFS, and CSS in prostate cancer with the use of metformin. Furthermore, it appears that metformin also reduces the formation of CRPC in diabetic patients. These findings strongly support further investigation of metformin in prostate cancer in a prospective randomized study in localized prostate cancer. (Table Presented) .