Background: Prevention of colorectal adenomas (CA) can reduce colorectal cancers (CRC). Epidemiological and experimental data suggest that the green tea catechin epigallocatechingallate has an antineoplastic effect in the large bowel. MIRACLE is the largest trial so far to examine the effect of three-year daily intake of green tea extract (GTE) on the incidence of metachronous CA in a Caucasian population. Methods: Prospective, parallel group, double-blinded, placebo-controlled, randomized multicenter trial (40 German centers, recruitment 11/2011-6/2015). Patients (n = 1001, age 50-80y), polypectomy ≤ 6 months and tolerating GTE well (one-month run-in) were randomized to receive decaffeinated GTE standardized to EGCG (150 mg bid, capsules) or placebo (P) for 3 years. Primary endpoint: Incidence of metachronous CA at the 3-year follow-up colonoscopy. Secondary endpoints: Occurrence, number, localization, size, histological subtype of CA, frequency of CRC, biomarker and safety. Strata: study center, low-dose aspirin (≤100 mg/d). Results: Clinical parameters were well balanced. CA incidence at the 3-year follow-up colonoscopy was analyzed in the modified ITT set (modITT; n = 309 patients (GTE), n = 323 (placebo), timely follow up colonoscopy) and the per protocol set (PP, modITT set without major protocol violations). Incidence of CA was 55.7 % (P) and 51.1% (GTE), (modITT, adj. RR 0.905, one sided, p = 0.081), respectively 54.3 % (P) and 48.3% (GTE) (PP, adj. RR 0.883, one sided, p = 0.058). These differences did not reach statistical significance. In the preplanned exploratory analysis regarding gender incidence of CA in females was 47.9% (P) and 47.6% (GTE) in the modITT-set (adj. RR 0.989; 95%-CI: 0.753,1.299; p = 0.935), respectively 45.4% (P) and 46.9% (GTE) in the PP-set (adj. RR 1.014; 95%-CI: 0.748, 1.373; p = 0.930). In contrast, in the male population incidence of CA in the follow-up colonoscopy was 60.4% (P) and 52.9% (GTE) in the modITT-set (adj. RR 0.846; 95% CI 0.717, 0.999); p = 0.048), respectively 59.1% (P) and 49.1% (GTE) in the PP-set (RR 0.803, 95% CI: 0.666, 0.969; p = 0.022). Thus, GTE intake was associated with a significant, 12.4 relative and 7.5% absolute reduction of metachronous CA in the male modITT population. There were no differences with respect to safety between the groups. Conclusions: GTE reduced the incidence of metachronous CA. However, a significant effect was only observed in the in the male population whereas there was no effect in the female population. Clinical trial information: NCT 01360320.