The effect of metformin exposure on colorectal cancer incidence according to tumor sidedness

Background: Colorectal cancer (CRC) is a major cause of cancer-related morbidity and mortality. Tumor sidedness influences treatment decisions for metastatic disease due to differences in biology and therapeutic response. Metformin, an anti-diabetic medication, may reduce CRC risk. This study aims to evaluate the relationship between metformin exposure and CRC risk based on tumor-sidedness. Methods: A nested case-control study was conducted using the Veterans Administration (VA) database (1999–2020). The cohort included individuals with diabetes mellitus and at least 3 years of follow-up. CRC cases were identified and classified by tumor sidedness. Controls were selected via incidence-density sampling, matched on age, sex, index-date, and first VA encounter. Exposure of interest was cumulative metformin use prior to the index-date. Conditional logistic regression was used to estimate adjusted odds-ratios (ORs) and 95 % confidence intervals (CIs). The analysis was adjusted for race, BMI, smoking, aspirin, statins, and other anti-diabetic medications. Results: The study included 31,078 CRC cases and 310,621 matched controls among diabetic individuals. Metformin exposure showed no effect on right-sided CRC incidence (adjusted OR 1.03, 95 %CI 0.92–1.16 for 1–3 years; 0.96, 95 %CI 0.83–1.12 for 3–5 years). In contrast, in patients with left-sided CRC metformin use was associated with a reduced risk of CRC (adjusted OR 0.90, 95 %CI 0.82–0.98 for 1–3 years; 0.87, 95 %CI 0.77–0.98 for 3–5 years). Conclusions: Metformin was associated with a decrease in the incidence of left-sided CRC. These results suggest the influence of tumor sidedness, not only on treatment effect, but on prevention strategies as well.