Two phase II breast cancer studies have been completed with gemcitabine in patients with locally advanced or metastatic breast cancer. Gemcitabine was administered as a 30 minute intravenous infusion on days 1, 8, and 15 of a 28-day cycle. In a European study of 44 patients, 40 patients were evaluable for response, 26 having received chemotherapy (seven in the adjuvant setting). The mean number of completed cycles administered was 2.7 and 81% of doses were delivered as scheduled. There were three complete responses and seven partial responses, giving an overall response rate of 25.0% (95% confidence interval, 12.7% to 41.2%). Four patients were not evaluable for efficacy: one had insufficient therapy, two had no bidimensionally measurable disease, and one had insufficient therapy and no bidimensionally measurable disease. The median duration of survival was 11.5 months. Hematologic toxicity was generally mild, with World Health Organization grade 3 and 4 leukopenia occurring in 6.8% and 2.3% of patients and neutropenia in 23.0% and 7.0% of patients, respectively. Nonhematologic toxicity was minimal. Flu-like symptoms were mild and transient. Only one patient developed alopecia. In a US study, 18 of 21 heavily pretreated patients were evaluable, all of whom had stage IV disease. The median number of cycles administered was two, with 12% of injections omitted and 31% reduced by 50%. No responses were observed in this smaller study. The safety profile was similar to that in the European study, although myelosuppression was greater in the US study, in which patients were heavily pretreated. The differing results observed from single-agent studies of gemcitabine in advanced breast cancer may be explained in part by the amount of prior chemotherapy received and the lower mean dose of chemotherapy administered in the US study. Responses have been observed in both chemotherapy-naive and previously treated patients. The drug was extremely well tolerated in both studies, even in heavily pretreated patients. In view of its modest toxicity profile and its novel mechanism of action, gemcitabine deserves further evaluation in breast cancer patients and, in particular, because of its relative lack of myelotoxicity would be an ideal candidate for combination chemotherapy.