Weekly paclitaxel-cisplatin administration with G-CSF support in advanced breast cancer. A phase II study.

Abstract

In a previous phase I study we found the MTDs of paclitaxel and cisplatin when given together weekly, with or without G-CSF support, in patients with advanced solid tumors. The present study was conducted to define the toxicity and efficacy of this regimen, when used with G-CSF support, in chemotherapy-naive or pretreated patients with advanced breast cancer, and to compare the antiproliferative activity of paclitaxel-cisplatin and paclitaxel-doxorubicin combinations on two human breast cancer cell lines. Patients with metastatic breast cancer received weekly paclitaxel (as a 3-hour i.v. infusion) at the dose of 85 mg/m2 (75 mg/m2 in pretreated women) followed by cisplatin (40 mg/m2) for a minimum of 6 weeks. An additional 6 weekly cycles were delivered in patients showing absence of documented disease progression after the first 6 weeks. After the 12th cycle only patients who had shown a substantial tumor shrinkage received 6 further cycles. G-CSF 5 microg/kg was also given, SC on days 3 to 5 of each week, for the whole duration of chemotherapy. The combination of paclitaxel with cisplatin or doxorubicin was also tested in vitro on two breast cancer cell lines (MCF-7 and MDAMB-231). Forty-three women with metastatic breast cancer entered this trial between June 1995 and January 1997. Twenty-seven patients were previously untreated for their metastatic disease (but 23 had previously received adjuvant chemotherapy). The dominant site of disease involvement was visceral in 23, bone in 13, and soft tissues in 7 patients. Seven complete and 15 partial responses were observed in unpretreated patients, while no complete and 6 partial responses were achieved in the pretreated population. The overall response rate, assessed on an 'intent to treat' basis, was 81% (26% CRs) in patients unpretreated for metastatic disease and 37% in those who had received one or more previous chemotherapy regimens. Eighteen responder patients had previously received anthracyclines either as adjuvant chemotherapy (12) or in the treatment of metastatic disease (6). At a median potential follow-up of 12 (range, 3-21) months, 14/27 unpretreated and 12/16 pretreated patients had shown disease progression. The median time to treatment failure was 13 and 7 months, respectively, in the 2 subgroups. The 1-year survival probability was 95% in unpretreated patients. The treatment showed a moderate toxicity in both subgroups of patients. Both hematological toxicity and peripheral neuropathy occurred more frequently in pretreated patients. Treatment-related deaths did not occur, and severe myelosuppression was observed only in pretreated patients with massive liver involvement. Delays in chemotherapy administration were very uncommon, especially during the first 6 treatment cycles, and the average actually delivered dose intensity exceeded 90% in unpretreated patients. The in vitro data on MCF-7 and MDA-MB-231 human breast cancer cell lines showed that exposure to the combination of cisplatin and paclitaxel produced a tumor cell killing similar to that achievable with equivalent concentrations of doxorubicin and paclitaxel. Weekly paclitaxel and cisplatin with G-CSF support is an active and particularly well tolerated treatment for patients with either unpretreated or pretreated metastatic breast cancer. This approach seems quite effective also in patients relapsing after anthracycline-based adjuvant chemotherapy. In view of the negligible hematological toxicity associated with this regimen, further clinical trials testing the addition of non cross-resistant drugs to this combination should be performed.