In a multicenter trial, 123 patients with advanced breast cancer who had been treated with tamoxifen and/or chemotherapy were randomized to receive medroxyprogesterone acetate (MPA) orally 300 mg X 3 daily or im 500 mg daily for 4 weeks and 500 mg X 2 weekly thereafter. All case histories were reviewed extramurally by the criteria of the International Union Against Cancer. Five and 11 patients were not eligible and evaluable for response, respectively. Pretreatment characteristics were well balanced in both treated groups. Twenty-five of all 107 (23%) evaluable patients achieved an objective remission, whereas in a further 15% the disease became stable after previous progression. Results in both treatment arms did not differ significantly. The median duration of objective remission was 12 and 14 months for orally and im treated patients, respectively (P greater than 0.10). No statistically significant differences in the survival times of orally and im treated patients were found. Pretreatment characteristics positively correlated with an objective remission during MPA therapy in both groups were age greater than 50 years (P less than 0.02) and no previous chemotherapy (P less than 0.01). Toxicity included an increase in body weight, cushingoid effects, muscle cramps, and tremors in both groups. In four patients on im therapy, local infections developed. Mean serum MPA levels reached values above 100 ng/ml in nine orally and eight im treated patients (P greater than 0.10), and neither differed significantly in the patients responding to or failing therapy. In both MPA arms, plasma cortisol levels were suppressed. The drop in plasma cortisol levels was more pronounced in patients with objective remissions than in patients who failed (P = 0.04). In conclusion, oral and im MPA in the given doses had similar activity. Im administration of MPA should be reserved for patients not able to take oral medication.