In an attempt to define the influence of prior hormonal treatments upon aminoglutethimide activity in advanced cancer of the breast, 42 heavily pretreated postmenopausal patients received aminoglutethimide, 4 X 250 mg daily, with hydrocortisone or cortisone. Twenty-six received high doses of medroxyprogesterone before entering this study. There was no significant difference in patients' characteristics with or without medroxyprogesterone pretreatment. A comparison of patients with and without prior medroxyprogesterone shows a significant difference in the response rate to aminoglutethimide-hydrocortisone (4 vs 32%, P = 0.02). In patients pretreated with tamoxifen but not with medroxyprogesterone the response rate to aminoglutethimide was 36%. These results suggest that aminoglutethimide has a low activity in breast cancer patients previously exposed to medroxyprogesterone, an agent with glucocorticoid-like activity inducing adrenal suppression.