The effect of metformin (M) on overall survival (OS) of patients (Pts) with colorectal cancer (CRC) treated with chemotherapy (CTX).

2608

Background: Prior observations had shown the beneficial effects of M on survival of Pts with CRC. This study was conducted to evaluate the effect of M on Pts treated with CTX. Methods: A retrospective study of 5052 Pts with CRC from 10/01/1996 to 3/31/2010 was conducted from the VA cancer registry data base (VISN16). CRC Pts who received CTX were analyzed for the effect of M on OS using stratified Cox regression with M exposure as a time-dependent covariate. Stratification factors were treatment (TX) modalities (CTX details, S and RT). Results: 1,078 of the 5,052 Pts with CRC received CTX. 571 (33.4%) had DM,141 were on M. There were 37(2.2%) females, 1,207 (70.1%) Caucasians, 408 (23.9%) African Americans, 93 (6.0%) other/unknown. Median age (quartiles) at CRC diagnosis was 64 (57–71) years (Yrs). CTX details: 796 (46.6%) single agent, 833 (48.8%) multi-agent, 79 (4.6%) not specified (NOS). 1362 (79.7%) had S, 639 (37.4%) had RT. 24 deaths occurred during 339 yrs of post-M follow-up (0.07 deaths/year/person[d/y/p]), 879 deaths during 6930 Yrs. of M-free follow-up (0.13 d/y/p). Without stratification, the hazard ratio (HR) (95% CI; p value) was 0.61 (0.41–0.91; p=0.017) for CTX+M compared to CTX alone. Adjusted HR post stratification on S, RT, and CTX details was 0.59 (0.39–0.89; p=0.012) for CTX+M compared to CTX alone. Conclusions: Metformin appears to improve the OS rate of Pts with CRC treated with CTX. There was no influence of the TX modalities (S, CTX, and RT) on the effect of M.