The object of the study was to evaluate the effectiveness of ifosfamide/etoposide and mesna therapy in advanced breast cancer. A total of 44 patients with breast cancer were included in the trial. Eligibility criteria included measurable, refractory disease; prior anthracycline therapy (or its contraindication); a life expectancy of at least 3 months; and adequate hepatic, renal, CNS and bone marrow function. All patients were less than or equal to 70 years of age and had a Karnofsky performance status of greater than or equal to 50%. There were 36 evaluable cases. Sites of metastatic disease included bone (19), skin (18), liver (9), lung (14), lymph node (19), and miscellaneous (7). Treatment consisted of 1,500 mg/m2 ifosfamide given i.v. on days 1-5, 120 mg/m2 etoposide given i.v. on days 1-3, and 400 mg i.v. mesna given with and at 4 and 8 h after ifosfamide. Cycles were repeated every 28 days. Initial doses were reduced by 25% or 50% in patients who had previously undergone both chemotherapy and radiotherapy. A median of 4 cycles (range, 2-8) were given. The myelotoxicity was marked: WHO grades 3/4 leukopenia (n = 37), grades 3/4 thrombocytopenia (n = 12), and grades 2/3 anemia (n = 13). Due to myelotoxicity, dose reduction or prolongation of treatment-free intervals was necessary in 28 cases. Alopecia was seen in 35 patients and CNS toxicity, in 8. Partial remission (PR) was obtained in five cases and complete remission (CR), in three. Sites of response included the lung (5), skin (4), lymph node (5), and peritoneum (1). The duration of response was 4 (n = 2) and 8 (n = 1) months for CR and 2 (n = 2), 6 (n = 2), and 10 (n = 1) months for PR. We conclude that the ifosfamide/etoposide and mesna regimen is effective, but its myelotoxicity is treatment-limiting.