Preexisting diabetes, metformin use and long-term survival in patients with prostate cancer

Abstract

Objective

To assess prostate cancer-specific and overall survival in prostate cancer patients with or without preexisting type 2 diabetes mellitus (T2DM) with regards to metformin use.

Methods

Patients diagnosed with prostate cancer in the Lithuanian population between 2001 and 2005 were identified through the Lithuanian Cancer Registry and followed until 2016, date of death, loss to follow-up or whichever came first. Information regarding the diagnosis of T2DM and antihyperglycemic medications were obtained from the National Health Insurance Fund database. Prostate cancer-specific and overall survival outcomes were analysed using univariate and multivariate Cox proportional hazard models.

Results

Out of 6689 men included, 254 (3.8%) had preexisting T2DM. There were 4807 deaths during follow-up, including 2084 from prostate cancer. No differences were found in prostate cancer-specific survival between men with or without T2DM. The risk of overall mortality was higher (HR = 1.24, 95% CI = 1.07–1.43) in diabetic men. Univariate analysis showed cancer stage at diagnosis and age to be significant predictors of survival. After adjustment for age and stage at diagnosis, there was no difference in prostate-specific survival between non-diabetic patients compared to metformin users or metformin non-users. However, overall survival was lower in T2DM patients, with a higher mortality risk for metformin non-users (HR = 1.63, 95% CI = 1.27–2.10). Prostate cancer-specific mortality risk was insignificantly lower in diabetic men on metformin (HR = 0.74, 95% CI = 0.54–1.02).

Conclusion

There was no difference in long-term prostate cancer-specific survival in patients with or without T2DM. Overall survival was lower in T2DM patients not treated with metformin.

Keywords:

• Diabetes mellitus
• metformin
• prostate cancer
• survival
• hazard ratios

Ethical approval

The study was approved by the Vilnius Regional Biomedical Research Ethics Committee.

Informed consent

Due to the design of the study, informed consent of individuals was not obtained.

Author contributions

Study conception and design: G. Smailyte, L. Zabuliene, D. Linkeviciute-Ulinskiene, A. Patasius. Data Analysis and Interpretation: Giedre Smailyte, A. Patasius, M. Kincius. Manuscript writing: D. Linkeviciute-Ulinskiene, A. Patasius, G. Smailyte. Project management: G. Smailyte. All authors reviewed, edited and approved the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).