Previous studies have evaluated whether metformin is associated with prostate cancer incidence and outcomes with conflicting conclusions. In this study, we evaluate the incidence of prostate cancer in diabetic patients treated with and without metformin compared with nondiabetic patients.Materials and Methods:
One thousand thirty-four patients underwent transperineal template-guided mapping biopsy secondary to either an elevated prostate-specific antigen (PSA) or a prior biopsy finding of atypical small acinar proliferation/prostatic intraepithelial neoplasia. The cohort included 881 nondiabetic men, 65 diabetic men treated with metformin, and 88 diabetic men not receiving metformin. In metformin-treated patients, the median duration of usage was 6.0 years. Differences in prostate cancer diagnosis, histologic grade, and tumor volume were compared across the 3 cohorts.Results:
There was no statistically significant differences discerned between the 3 cohorts in patient age, prebiopsy PSA, prostate volume, PSA density, PSA doubling time, PSA velocity, or the total number of prior transrectal ultrasound biopsy sessions. Five hundred eighty-four patients were diagnosed with prostate cancer. There was no difference in prostate cancer diagnosis (P=0.153), Gleason score (P=0.960), the number of positive biopsy cores (P=0.764), or risk group stratification (P=0.877) between the 3 cohorts. In multivariate analysis, only older age predicted for prostate cancer diagnosis. In terms of Gleason score ≥7, patient age, PSA velocity, and body mass index predicted for more aggressive histology. Neither diabetes, metformin use or duration was of statistical consequence.Conclusion:
Metformin did not impact incidence of prostate cancer diagnosis, Gleason score distribution, or volume of disease.