Paclitaxel-Based Combination Chemotherapy for Breast Cancer
Paclitaxel-Based Combination Chemotherapy for Breast Cancer
The initial clinical trials with single-agent paclitaxel (Taxol) demonstrated the marked antitumor efficacy of this agent in ovarian and breast cancer.[1,2] Subsequent clinical trials expanded the experience with this agent and showed that paclitaxel retained substantial antitumor efficacy in previously treated patients and that paclitaxel was active against metastatic breast cancer at a variety of doses and schedules of administration.[3-8] Comparative clinical trials to determine the optimal dose and schedule of administration of this agent are completing accrual. Paclitaxel has been extensively evaluated in combination with other cytotoxic agents with demonstrated activity against metastatic breast cancer (Table 1).
Soon after the first report of its antitumor activity, the evaluation of paclitaxel-based combinations started. Since doxorubicin, until the appearance of the taxanes, was considered the most active antitumor agent for metastatic breast cancer, the anthracycline/taxane combination was a logical first step. The University of Texas M. D. Anderson Cancer Center and the Medicine Branch of the National Cancer Institute (USA) initiated paclitaxel/doxorubicin combination phase I trials simultaneously.[9-13] Both groups used the longer durations of infusion of paclitaxel, since that was the schedule of administration used for single-agent trials. In addition, to limit cardiac and other nonhematologic toxicity related to anthracyclines, infusion schedules of administration were used for this agent, too (Table 2).
Dose-limiting toxicities of these phase I trials included severe mucositis, diarrhea, tiflitis, and neutropenic fevers. Reversing the sequence of administration of doxorubicin and paclitaxel in the M. D. Anderson Cancer Center clinical trials allowed dose escalation by two or three levels, but eventually, the same dose-limiting toxicity was reached. These initial paclitaxel/doxorubicin combinations were clearly effective, with overall response rates ranging from 62% to 80% and with 95% confidence intervals ranging from 38% to 98%. Furthermore, it is noteworthy that no clinically relevant cardiac toxicity was observed; specifically, no instances of congestive heart failure were reported.
As the first reports of these combination trials appeared, other investigators designed new combinations with the same two agents or with other cytotoxic drugs. Shortening the duration of administration of doxorubicin permitted the administration of higher doses, yet a schedule-dependent interaction was noted and confirmed by pharmacokinetic studies.[14-16] Investigators at the Istituto Nazionale Tumori in Milan, Italy, developed a combination based on bolus administration of doxorubicin and a 3-hour administration of paclitaxel.
When the two drugs were given in this schedule of administration, there was no apparent schedule-dependent interaction of clinical relevance, although these authors also confirmed the previously described pharmacokinetic variations related to the sequence of administration. The doses administered could be increased to 60 mg/m² and 200 mg/m² for doxorubicin and paclitaxel, respectively, and the antitumor activity increased markedly, with an overall response rate exceeding 90%, and complete remission rate now greater than 40% (Table 3).
Observers of this trial were concerned by the appearance of congestive heart failure in 20% of patients and a greater than 15% decrease in left-ventricular ejection fraction in as many as 75% of patients. In fact, in 22% of patients who did not develop clinical heart failure, ejection fraction dropped below 50%. A second study that used the same doses and schedule of administration as the Milan group confirmed these results, both in terms of antitumor activity and cardiac toxicity. Subsequent experience from the Milan group, limiting the total cumulative doxorubicin dose to 360 mg/m² (six cycles), demonstrated the safety of this regimen, with no episodes of heart failure in nearly 40 patients. [L. Gianni, personal communication, December 1996]
Currently, several confirmatory trials are ongoing. These trials employ the same schedule of administration of paclitaxel and anthracycline (in some studies, epirubicin was substituted for doxorubicin), and although substantial antitumor activity has been reported, these studies are not mature enough to fully assess the results in terms of antitumor activity and toxicity. A randomized phase II study is being conducted at The University of Texas M. D. Anderson Cancer Center, in which doxorubicin (60 mg/m²) is administered as an intravenous bolus followed immediately by paclitaxel (200 mg/m²), either by 1- or 3-hour administration. The study's major focus is close cardiac monitoring to identify whether this combination can be administered safely, to determine the regimen's degree of cardiac toxicity, and to see if the associated cardiac toxicity is the result of synergistic effects of these drugs. Elements of the study include clinical evaluation, noninvasive cardiac functional tests, and endomyocardial biopsies.
Cisplatin (Platinol) is reported to have marked antitumor activity in previously untreated metastatic breast cancer. In addition, cisplatin is less myelosuppressive than many other useful agents against metastatic breast cancer and is considered a good candidate for combination with paclitaxel. One group developed a regimen in which cisplatin was given at 60 mg/m² with the appropriate premedication and hydration, while paclitaxel was administered at 90 mg/m² as a 3-hour infusion. The combination was repeated at 14-day intervals.
In an exploratory study that included mostly patients who had previously received an anthracycline, there was an overall response rate of 85%, including an 11% complete response rate (Table 4). Tolerance was acceptable. There were no episodes of grade 4 nonhematologic toxicity, although grade 2 to 3 fatigue, peripheral neuropathy, and nausea were reported. Myelosuppression and neutropenic fevers were also seen.
Wasserheit et al used higher dosages of both cisplatin and paclitaxel at 3-week intervals. The activity of this regimen was less impressive than that reported by Gelmon and collaborators, and the toxicity was considerably higher. Other investigators have also reported successful combinations of cisplatin and paclitaxel. The optimal dose and schedule of administration for this combination remain to be established.
Because the mechanisms of action of alkylating agents and taxanes are different, as are many of the side effects and mechanisms of resistance, and because cyclophosphamide (Cytoxan, Neosar) is an integral part of most commonly used combination regimens for breast cancer, this was another logical combination to explore. Some of these combinations used standard doses of cyclophosphamide, while others increased the dose of the alkylating agent to two- to sixfold the standard dose[22-25] (Table 5). Because myelosuppression is an overlapping toxicity of the two agents, hematopoietic growth factors were used in all these trials.
Tolcher et al reported the results of his dose-escalation study, which demonstrated a 62% overall response rate in 56 patients treated with 160 mg/m² of paclitaxel infused over 72 hours and 3,300 mg/m² of cyclophosphamide. Although the combination is clearly active, myelosuppression is quite considerable, and the need for the systematic addition of granulocyte colony-stimulating factor (G-CSF [Neupogen]) makes this combination less attractive than other paclitaxel-based regimens.
Paclitaxel has also been combined with vinorelbine (Navelbine). This new nor-vinblastine analog is substantially active in metastatic breast cancer, and it is very well tolerated when administered weekly as a single agent. Vinorelbine is a tubulin-active agent, and it was anticipated on the basis of preclinical experiments that the sequence of administration with a taxane would be an important consideration.
Several phase I or exploratory trials of the vinorelbine/paclitaxel combination are in progress. The University of Texas M. D. Anderson Cancer Center group used a simultaneous schedule of administration for vinorelbine and paclitaxel: both were given as a 3-hour IV infusion every 3 weeks. Dose-limiting toxicities were neutropenic fever and cumulative neuropathy. With hematopoietic growth factor (G-CSF) support, additional dose escalation was possible, and neuropathy became the dose-limiting event. This combination was also active, with objective response rates in the 50% to 65% range.
In other centers, paclitaxel and vinorelbine are being evaluated in sequential schedules of administration.[30-32] Although no definitive results have been reported, peripheral neuropathy appears to be a less prominent problem in these trials.