Abstract Familial adenomatous polyposis (FAP) is a hereditary disease characterized by the development of numerous colorectal adenomas in young adults. Metformin, an oral diabetic drug, has been shown to have antineoplastic effects and a favorable safety profile. We performed a randomized, double-blind, controlled trial to evaluate the efficacy of metformin on the regression of colorectal and duodenal adenoma in patients with FAP. Thirty-four FAP patients were randomly assigned in a 1:2:2 ratio to receive placebo, 500 mg metformin, or 1,500 mg metformin per day orally for 7 months. The number and size of polyps and the global polyp burden were evaluated before and after the intervention. This study was terminated early based on the results of the interim analysis. No significant differences were determined in the percentage change of colorectal and duodenal polyp number over the course of treatment among the three treatment arms (P = 0.627 and P = 1.000, respectively). We found no significant differences in the percentage change of colorectal or duodenal polyp size among the three groups (P = 0.214 and P = 0.803, respectively). The overall polyp burdens of the colorectum and duodenum were not significantly changed by metformin treatment at either dosage. Colon polyps removed from the metformin-treated patients showed significantly lower mTOR signal (p-S6) expression than those from patients in the placebo arm. In conclusion, 7 months of treatment with 500 mg or 1,500 mg metformin did not reduce the mean number or size of polyps in the colorectum or duodenum in FAP patients (ClinicalTrials.gov ID: NCT01725490). Prevention Relevance: A 7-month metformin treatment (500 mg or 1,500 mg) did not reduce the number or size of polyps in the colorectum or duodenum of FAP patients as compared to placebo. These results do not support the use of metformin to promote regression of intestinal adenomas in FAP patients.