This prospective randomized study aimed at establishing the optimal postoperative adjuvant chemotherapy regimen for premenopausal n+ breast cancer patients. The treatments were Regimen A, comprising 6 courses of CMF (cyclophosphamide, 100 mg/body on days 1-14; methotrexate, 40 mg/m2 on days 1 and 8; and 5-fluorouracil, 500 mg/m2 on days 1 and 8), and Regimen B, consisting of UFT (300 mg/day) and tamoxifen (30 mg/day) administered orally each day for 2 years. Telephone registration allocated the patients to the treatment groups by the minimization method in relation to the T category, number of n+ lesions and estrogen receptor status. Forty-five patients were registered, and 44 of them were eligible (22 cases each to Regimen A and Regimen B). The principal background factors showed no biases between the groups. The adverse reaction incidence was significantly higher with Regimen A (90.9% vs 22.7%). The 5-year survival rate was 89.8% with Regimen A and 100% with Regimen B, while the 5-year disease-free rates were 64.5% and 76.3%, showing no statistical significance. Regimen B showed a better QOL rating after 6 months of therapy in relation to nausea-vomiting and hair loss, and after 24 months in relation to appetite, sleep, performance status, happiness, anorexia and hair loss.