Abstract
This study aimed to examine the causal relationship between metformin treatment and the prevention of various cancer types through a 2-sample Mendelian randomization (MR) analysis. Utilizing genome-wide association study summary data, a 2-sample MR approach was employed to assess the association between metformin treatment and multiple cancer types. The analysis further investigated the causal effects of specific metformin targets, including adenosine monophosphate-activated protein kinase (AMPK), mitochondrial complex I, mitochondrial glycerol 3-phosphate dehydrogenase, and growth differentiation factor 15 (GDF15). Causal associations were evaluated using inverse-variance weighting, MR-Egger, weighted median, and Wald ratio methods. Sensitivity analyses were performed using the TwoSampleMR package in R to ensure robustness. inverse-variance weighting estimates demonstrated a protective effect of metformin against esophageal, gastric, and colorectal cancers (all odds ratios <0.1, all P <.05). The Wald ratio analysis identified a significant causal association between GDF15 and colorectal cancer (odds ratios: 0.927, 95% confidence interval: 0.883-0.974, P = .003). However, no significant causal relationships were observed between the other metformin targets and the aforementioned cancers. All single-nucleotide polymorphisms used as instrumental variables exhibited strong associations, and no evidence of horizontal pleiotropy was detected. The findings provide genetic evidence supporting a potential protective role of metformin against the development of esophageal, gastric, and colorectal cancers. Furthermore, our analysis suggests that the protective effect of metformin on colorectal cancer might be partially mediated by GDF15, which warrants further investigation.