Forty-three postmenopausal females with advanced breast cancer were studied in a prospective comparative trial of estrogen vs. an anti-estrogen (tamoxifen) therapy with a crossover to the alternative hormone with progressive disease. Ten of 19 patients (53%) responded to primary tamoxifen therapy and six of 24 (25%) responded to primary estrogen therapy. Crossover responses were observed in seven of 19 (37%) on the secondary tamoxifen therapy and in two of 10 (20%) on secondary estrogen therapy, and were not related to the response to the primary hormonal maneuver. Responses were related to the presence of estrogen receptor protein (ERP), particularly for tamoxifen therapy, although responses were observed in three of six ERP negative patients receiving estrogen and in seven of 25 (28%) of patients with an unknown ERP status. Complications were observed in 35 instances with estrogen therapy and in only five instances with tamoxifen therapy. Initial hormonal therapy with tamoxifen in postmenopausal patients with advanced breast cancer and ERP status positive or unknown is superior to primary estrogen treatment. Secondary therapy and response to estrogen or tamoxifen is not necessarily predicted by the initial hormone response, and crossover to the alternative therapy is generally indicated.