Two randomized clinical studies comparing the efficacy of oral UFT (2 years) with that of classical cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) (six courses) have been conducted in patients with resected early breast cancer. We have performed a pooled analysis of these two randomized studies. A pooled analysis was performed using individual patient data from the two trials. Hazard ratios (HRs) were determined with a Cox model stratified by study and adjusted for clinical characteristics. We preplanned to verify the following two hypotheses: UFT is non-inferior to CMF in all patients (hypothesis 1) or in ER-positive patients (hypothesis 2) with respect to relapse-free survival (RFS). Non-inferiority of UFT versus CMF was established if the upper limit of the two-sided confidence interval (CI) of the HR for RFS did not exceed 1.30. Hochberg multiplicity adjustment for the significance level was performed. A total of 1,057 patients were analyzed (CMF, n = 528; UFT, n = 529). Median follow-up time was 5.6 years. The HR for RFS was 1.04 (95% CI, 0.78-1.40) in all patients and 0.79 (97.5% CI, 0.49-1.27) in ER-positive patients. UFT was shown to be non-inferior to CMF in ER-positive patients. An exploratory subgroup analysis showed that RFS was better with UFT than with CMF in ER-positive patients who were 50 years or older (HR, 0.58; 95% CI, 0.34-1.01). UFT is non-inferior to CMF in terms of inhibiting recurrence of ER-positive, early breast cancer.