Phase I dose escalation study of gemcitabine and paclitaxel plus colony-stimulating factors in previously treated patients with advanced breast and ovarian cancer.

Gemcitabine and paclitaxel (PTX) are among the most active new drugs in advanced breast and ovarian cancer. In this Phase I study, we used fixed doses of gemcitabine administered on days 1 and 8 and escalating doses of paclitaxel on day 1 of a 21-day cycle in patients with pretreated metastatic breast or ovarian cancer. The dose of gemcitabine was fixed at 1,000 mg/m2; PTX was commenced in the first small patient group at a dose of 90 mg/m2, which was then escalated in subsequent groups by 30 mg/m2 per step. From the third dose level onwards, all patients received granulocyte colony-stimulating factor 300 microg by subcutaneous injection on days 5 and 6, and granulocyte macrophage colony-stimulating factor on days 15-18. Cohorts of at least 3 patients were treated at each dose level. Dose escalation was stopped if at least a third of the patients in a given cohort had dose-limiting toxicity (DLT), which was defined as grade 4 neutropenia or thrombocytopenia, or grade 3-4 non-haematological toxicity. The maximum tolerated dose (MTD) was defined as the dose level immediately below that causing DLT in one-third of the patients or more. Evaluation of the tumour response was performed every three cycles. Forty-five patients (31 with breast cancer, 14 with ovarian cancer) were treated at seven different dose levels. Only at the seventh PTX dose level was DLT observed after the first course of therapy: three grade 4 neutropenia, one grade 4 thrombocytopenia, and one grade 4 anaemia. DLT occurred in 5/6 patients at at PTX dose of 270 mg/m2; therefore dose escalation was stopped at that level and the dose immediately before it (PTX 240 mg/m2) was considered as the MTD and recommended for further studies. No toxic deaths occurred. Grade 3-4 uncomplicated neutropenia was observed in four patients. Three had uncomplicated grade 3-4 thrombocytopenia. One patient had grade 3 and one grade 4 anaemia. Nonhaematological side effects were generally mild. Among 30 evaluable patients with metastatic breast cancer, four complete responses (CR) (13%) and 12 partial responses (PR) (40%) were observed, for an overall response rate of 53% (95% confidence interval (CI) 34-72). The median duration of response was 31 weeks. Among 13 evaluable patients with advanced ovarian cancer, one CR (8%) and five PRs (38%) were observed, for an overall response rate of 46% (95% CI 19-78). The median duration of response was 32 weeks. Our study shows that gemcitabine and PTX can be administered in combination in patients with breast and ovarian cancer without unexpected toxicities and with encouraging therapeutic results.