Between 1978 and 1981, 463 evaluable postmenopausal patients 65 years of age or younger with operable breast cancer and metastases in axillary lymph nodes were entered in Ludwig Breast Cancer Study III (Ludwig III) and randomly allocated to receive chemoendocrine therapy with cyclophosphamide, methotrexate, 5-fluorouracil, low-dose continuous prednisone, and tamoxifen (CMFp + T) for 12 monthly cycles, or endocrine therapy alone with prednisone and tamoxifen (p + T) for 1 year, or no adjuvant treatment after mastectomy (observation). At 60 months' median follow-up, the 5-year disease-free survival (DFS) rates were 59% for CMFp + T, 41% for p + T, and 31% for observation (P less than .0001), and the 5-year overall survival (OS) rates were 71% for CMFp + T, 64% for p + T, and 59% for observation (P = .16; CMFp + T v observation, P = .07). A new quality of life-oriented endpoint was defined to assist in the selection of therapeutic approach after surgery for postmenopausal patients: the time without symptoms of disease and subjective toxic effects of treatment (TWiST). Despite the larger initial discount due to subjective toxicity with chemoendocrine therapy, by 5 years postmastectomy the net difference in average TWiST for treated patients compared with the observation group was positive and approximately equal for both adjuvant treatment programs. Adjuvant chemoendocrine therapy for postmenopausal women appears to be justified due to an emerging OS advantage and increasing TWiST gained for the treated patients.