Individual and joint effects of metformin and statins on mortality among patients with high-risk prostate cancer

Xiang Lin Tan, E. Jian-Yu, Yong Lin, Timothy R. Rebbeck, Shou En Lu, Mingyi Shang, William K. Kelly, Anthony D'Amico, Mark N. Stein, Lanjing Zhang, Thomas L. Jang, Isaac Yi Kim, Kitaw Demissie, Anna Ferrari, Grace Lu-Yao

Research output: Contribution to journalArticlepeer-review

23 Scopus citations


Background: Pre-clinical studies suggest that metformin and statins may delay prostate cancer (PCa) metastases; however, data in humans are limited. To the best of our knowledge, this is the first human study aimed to quantify the individual and joint effects of statin and metformin use among patients with high-risk PCa. Methods: This population-based retrospective cohort study identified patients from the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database. Exposure to metformin and statins was ascertained from Medicare Prescription Drug Event files. The association with all-cause and PCa mortality were evaluated using Cox proportional hazard model with competing causes of death, where propensity scores were used to adjusted imbalances in covariates across groups. Results: Based on 12 700 patients with high-risk PCa, statin alone or in combination with metformin was significantly associated with reduced all-cause mortality (Hazard Ratio [HR]: 0.89; 95% Confidence Interval [CI]: 0.83, 0.96; and HR: 0.75; 95% CI, 0.67-0.83, respectively) and PCa mortality (HR, 0.80; 95% CI: 0.69, 0.92) and 0.64; 95% CI, d 0.51-0.81, respectively. The effects were more pronounced in post-diagnostic users: combination use of metformin/statins was associated with a 32% reduction in all-cause mortality (95% CI, 0.57-0.80), and 54% reduction in PCa mortality (95% CI, 0.30-0.69). No significant association of metformin alone was observed with either all-cause mortality or PCa mortality. Conclusions: Statin use alone or in combination with metformin was associated with lower all-cause and PCa mortality among high-risk patients, particularly in post-diagnostic settings; further studies are warranted.

Original languageEnglish
Pages (from-to)2379-2389
Number of pages11
JournalCancer Medicine
Issue number7
StatePublished - 1 Apr 2020
Externally publishedYes


  • all-cause mortality
  • high-risk prostate cancer
  • metformin
  • population-based cohort
  • prostate-cancer mortality
  • statins
  • time-varying Cox proportional hazard models


Dive into the research topics of 'Individual and joint effects of metformin and statins on mortality among patients with high-risk prostate cancer'. Together they form a unique fingerprint.

Cite this