BACKGROUND Postsurgical treatment with tamoxifen has been shown to improve overall survival among patients with early stage breast cancer. However, the optimal duration of tamoxifen treatment remains controversial. OBJECTIVE A multicenter, randomized trial was initiated in Sweden in the early 1980s to compare 2 years with 5 years of adjuvant tamoxifen in the treatment of postmenopausal women younger than 75 years of age who had early stage, axillary lymph node-negative or -positive, invasive disease. METHODS The trial was planned and organized by the Swedish Breast Cancer Group, and it involved five regional breast cancer study organizations (South Sweden, South-East Sweden, Stockholm, Uppsala-Orebro, and North Sweden). During the period from 1983 through 1991, a total of 3887 patients were entered in the trial; 3545 (91%) women remained alive and recurrence free at 2 years and could thus contribute meaningful information to the 2-year (n = 1801) versus 5-year (n = 1744) comparison. Primary surgery consisted of either modified radical mastectomy or breast-conserving surgery. Radiation therapy was indicated for patients with lymph node-positive disease and was generally offered to all women who were treated with breast-conserving surgery. Only 89 (2.5%) of the 3545 women who were recurrence free at 2 years received adjuvant chemotherapy concurrently with tamoxifen. Twenty-milligram daily doses of tamoxifen were used at two centers, and 40-mg daily doses were used at the remaining three centers. Estrogen receptor status of the tumor was known for 2987 women (77% of the entered patients). Primary end points in the trial were event-free survival (local-regional recurrence, distant metastasis, contralateral breast cancer, or death) and overall survival. Survival curves were estimated by use of the life-table method. The Cox proportional hazards model was used to make comparisons between the 2- and 5-year treatment groups. RESULTS Patients assigned to receive 5 years of tamoxifen, compared with 2 years of tamoxifen, experienced statistically significant improvements in event-free survival (relative hazard = 0.82; 95% confidence interval [CI] = 0.71-0.96) and overall survival (relative hazard = 0.82; 95% CI = 0.69-0.99). These findings translate into an 18% relative reduction in both first events (95% CI = 4%-29%) and mortality (95% CI = 1%-31%) with the longer treatment. Overall survival at 10 years was estimated to be 80% among patients in the 5-year tamoxifen group who were alive and recurrence free at 2 years, compared with 74% among corresponding patients in the 2-year treatment group. The benefit associated with the longer treatment extended to women with lymph node-positive as well as lymph node-negative disease, but it appeared to be restricted to women whose tumors were classified as estrogen receptor positive. CONCLUSIONS Five years of adjuvant tamoxifen is more beneficial than 2 years in the treatment of postmenopausal women with estrogen receptor-positive, early stage, invasive breast cancer.