BACKGROUNDThe only established treatment for preventing colorectal cancer in patients with familial adenomatous polyposis (FAP) is colectomy, which greatly reduces patient quality of life. Thus, an alternative method is warranted. In this trial, we aimed to clarify the individual and joint effects of low-dose aspirin and mesalazine on the recurrence of colorectal polyps in Japanese patients with FAP.METHODSThis was a randomised, double-blind, placebo-controlled, multicentre trial with a two-by-two factorial design done in 11 centres in Japan. Eligible patients were aged 16-70 years and had a history of more than 100 adenomatous polyps in the large intestine, without a history of colectomy. Before the study, patients underwent endoscopic removal of all colorectal polyps of at least 5·0 mm in diameter. Randomisation was done with a minimisation method with a random component to balance the groups with respect to the adjustment factors of sex, age (<30 years vs ≥30 years), or smoking status at the time of entry. Patients and researchers were masked to the treatment group. There were four groups: aspirin (100 mg per day) plus mesalazine (2 g per day), aspirin (100 mg per day) plus mesalazine placebo, aspirin placebo plus mesalazine (2 g per day), or aspirin placebo plus mesalazine placebo. Treatment was continued until 1 week before 8 month colonoscopy. The primary endpoint was the incidence of colorectal polyps of at least 5·0 mm at 8 months and was assessed in the intention-to-treat population. Safety was assessed in the ITT population. We also did a per-protocol analysis including only patients who took at least 70% of the allocated study drug. This trial is registered with the UMIN Clinical Trials Registry, number UMIN000018736, and is complete.FINDINGSBetween Sept 25, 2015, and March 13, 2017, 104 patients were randomly assigned to receive either aspirin or aspirin placebo (n=52) or mesalazine or mesalazine placebo (n=52). Two patients withdrew from the aspirin plus mesalazine placebo group. 26 (50%) of 52 patients who received no aspirin had colorectal polyps of at least 5·0 mm at 8 months, as did 15 (30%) of the 50 patients who received any aspirin, 21 (42%) of the 50 patients who received no mesalazine, and 20 (38%) of the 52 patients who received any mesalazine. The adjusted odds ratio for polyp recurrence was 0·37 (95% CI 0·16-0·86) in the patients who received any aspirin and 0·87 (95% CI 0·38-2·00) in any who received mesalazine. The most common adverse events were grade 1-2 upper gastrointestinal symptoms in three (12%) of 26 patients who received aspirin plus mesalazine, one (4%) of 24 patients who received aspirin plus mesalazine placebo, and one (4%) of 26 patients who received mesalazine plus aspirin placebo. There was one grade 4 event in the mesalazine plus aspirin placebo group, but not related to the treatment.INTERPRETATIONLow-dose aspirin safely suppressed the recurrence of colorectal polyps larger than 5·0 mm in patients with FAP. These results suggest an effect of low-dose aspirin for FAP and could be an alternative method for preventing colorectal cancer in FAP.FUNDINGJapan Agency for Medical Research and Development.