Metformin, Diabetes, and Survival among US Veterans with Colorectal Cancer

Background: Metformin has been associated with improved colorectal cancer survival, but investigations are limited by small numbers of patients and confounding by diabetic severity. We examined the association between metformin use and overall survival (OS) in patients with diabetes and colorectal cancer in a large population of U.S. veterans, while adjusting for measures of diabetic severity. Methods: Patients diagnosed with colorectal cancer from January 2001 to December 2008 were identified from the Veterans Affairs Central Cancer Registry. Multivariable models were used to examine the adjusted association of OS with diabetes and use of antidiabetic medications. Results: There were 21,352 patients diagnosed with colorectal cancer identified (n = 16,355 nondiabetic patients, n = 2,038 diabetic patients on metformin, n = 2,136 diabetic patients on medications other than metformin, n = 823 diabetic patients not on antidiabetic medication). Diabetic patients had a significantly worse OS than nondiabetic patients, but metformin users had only a 10% increase in death (HRadj 1.10; 95% CI, 1.03-1.17, P = 0.004), as compared with 22% for users of other antidiabetic medications (HRadj 1.22; 95% CI, 1.15-1.29, P < 0.0001). Among colorectal cancer patients with diabetes, metformin users had a 13% improved OS versus patients taking other antidiabetic medications (HRadj 0.87; 95% CI, 0.79-0.95, P = 0.003), while diabetic patients not on any antidiabetic medications did not differ with respect to OS (HRadj 1.02; 95% CI, 0.90-1.15, P = 0.76). Conclusions: Among diabetics with colorectal cancer, metformin use is associated with improved survival, despite adjustments for diabetes severity and other risk factors. Impact: These data lend further support to the conduct of randomized studies of possible anticancer effects of metformin among patients with colorectal cancer. (C) 2016 AACR.