Paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ), the first taxane used in routine clinical practice, has aroused considerable interest for its high single-agent activity against breast cancer and for its novel mechanism of action. Epirubicin, the 4' epimer of doxorubicin, is another agent with a high activity against breast cancer and is known for its lower rate of toxic side effects, especially toxic cardiac events, compared with its mother compound. The combination of paclitaxel and doxorubicin yielded response rates between 63% and 93% in phase I/II studies. In these studies, however, the investigators reported severe cardiac toxic events. The rationale for the current study was therefore to evaluate the combination of paclitaxel/epirubicin, focusing mainly on cardiac toxicity. In two groups, 85 patients with metastatic breast cancer entered the study. Approximately 20% of the patients had primary metastatic breast cancer with large tumors at the primary site. Half of the patients had received adjuvant chemotherapy. Study medication in group A consisted of epirubicin 60 mg/m2 given intravenously over 1 hour, followed by paclitaxel 175 mg/m2 administered as a 3-hour intravenous infusion after premedication with steroids, antihistamines, and H2-blockers. In group B, epirubicin 90 mg/m2 was combined with paclitaxel 175 mg/m2, given in the same manner as in group A. Dose escalation to 225 mg/m2 paclitaxel was planned in both groups. The main toxicity in both groups was neutropenia (73% World Health Organization grade 3/4 in group A and 93% in group B). Other hematologic side effects were rare. No febrile neutropenia was reported in group A, but two episodes occurred in group B. Peripheral neuropathy, arthralgia, and myalgia were mild (only World Health Organization grades 1 and 2). Alopecia was universal. In group A, the paclitaxel dose could be escalated to 200 mg/m2 in 15 patients and to 225 mg/m2 in seven patients. Dose reduction due to severe neutropenia was necessary in 11 patients. No cardiac events were reported in group A. In group B, the paclitaxel dose could be escalated to 200 mg/m2 in only one patient, and no patient reached 225 mg/m2. Three patients needed a dose reduction. In this group, one patient had a greater than 10% decrease in the left ventricular ejection fraction with no clinical signs. In group A, 43 patients were evaluable for response; in group B, 25 patients were evaluable. Thirteen patients were out of protocol with only bone metastasis, and two patients had more than one prior chemotherapy for metastatic disease. The response rate was identical in both groups, with five complete remissions and 24 partial remissions in group A and three complete responses and 14 partial remissions in group B. The duration of response was 8.2 months in both groups. The median cumulative epirubicin dose was 420 mg/m2 in group A and 630 mg/m2 in group B. The combination of paclitaxel 175 mg/m2 and epirubicin 60 or 90 mg/m2 can be administered safely to patients with metastatic breast cancer. Although response was not the primary end point of this trial, the response data are nonetheless encouraging and suggest that further evaluation of this combination-line treatment of metastatic breast cancer is warranted.