We randomized 122 premenopausal women to receive tamoxifen or to undergo a surgical oophorectomy. Of 54 evaluable women treated with tamoxifen, 24% had an objective response, as compared with 21% of 53 women having an oophorectomy. The median duration of response for tamoxifen (20 months) was longer than that for surgical oophorectomy (7 months), but this did not achieve statistical significance (P = .056). Overall median survival was 15 months for 58 patients receiving tamoxifen and 25 months for 53 patients undergoing oophorectomy (P = .18). Toxicity was greater in those undergoing oophorectomy, though both treatments were well tolerated. In those premenopausal women for whom hormonal therapy is indicated, tamoxifen is a suitable alternative to surgical oophorectomy.