Background: To evaluate the association between use of metformin and other antidiabetic drugs with tumor characteristics and survival in surgically managed prostate cancer (PCa) patients.
Methods: The study population included 1314 men who underwent radical prostatectomy at the Tampere University Hospital during 1995-2009. Causes of deaths were collected from the Finnish Cancer Registry. Individual-level data on medication use during 1995-2009 was obtained from national prescription database. Fasting blood glucose and hemoglobin A1c values during the study period were gathered from hospital district database. Gleason grade and pathological stage were compared by drug use before surgery and separately by metformin usage. Risk of biochemical recurrence, all-cause death and PCa-specific death were calculated using Cox proportional hazard regression with adjustment for age, tumor characteristics, glycemic control and use of other drug groups.
Results: High-grade tumors were more common among antidiabetic drug users (P=0.032), including metformin users (P=0.012). Despite this, no difference in PSA levels was observed. Men who had used antidiabetic drugs before surgery had an increased risk of Gleason 7-10 disease (odds ratio (OR) 1.83, 95% confidence interval (CI) 1.04-3.23). The risk of high-grade PCa was higher among metformin users compared with other antidiabetic drug users (OR 3.11, 95% CI 1.16-8.33). During the median follow-up of 8.6 years after surgery, 551 men had biochemical recurrence and 244 died, 32 owing to PCa. Generally, no association with risk of disease recurrence was observed. Risk of death was increased by preoperative use of antidiabetic drugs (hazard ratio 1.81 95% CI 1.03-3.19), but no survival associations for postoperative use of antidiabetic drugs or metformin were observed.
Conclusion: Diabetic men have more high-grade PCa at lower PSA levels, but that does not have a clear impact on disease-specific survival in the short term even when glycemic control is being considered.