One hundred fifty-one postmenopausal women with progressive metastatic breast cancer and no prior hormonal therapy were treated with either diethylstilbestrol (DES) or tamoxifen (TAM). One hundred forty-three eligible patients were followed until death or for a minimum of 14.1 years on the DES arm or 16.7 years on the TAM arm. The overall objective response was 42% for DES and 33% for TAM (p = 0.31) and the median duration of response was 11.8 months for DES and 9.9 months for TAM (p = 0.38). Duration of response and progression-free survival were not found to be significantly different between DES and TAM (p = 0.32 and 0.65, respectively). The median survival was 3.0 years for DES vs. 2.4 years for TAM. The 5-year survival was 35% for the DES arm and 16% for the TAM arm. Survival was significantly better for women on DES than for women on TAM (adjusted p = 0.039). Review of records did not show any difference in pattern of treatment failure or subsequent treatments in the DES and TAM arms. Treatment with DES was more commonly associated with toxicity such as nausea, edema, vaginal bleeding, and cardiac problems, whereas hot flashes were commonly seen with TAM therapy. The initial treatment with DES is associated with increased survival. The basis of this survival advantage is not known. TAM still is the preferred agent in the treatment of metastatic breast cancer, but this trial underscores the fact that estrogens have activity and remain in the armamentarium for treatment of selected patients with metastatic breast cancer.