Prospective randomized trial evaluating weekly paclitaxel (WP) versus docetaxel in combination with capecitabine (DC) in operable breast cancer

Abstract

542 Background: Paclitaxel has scheduled-dependant efficacy, and WP is associated with better therapeutic index (J Clin Oncol. 2005;23:5983–5992). The DC combination had higher response rate, longer control of disease, and improved survival in metastatic disease (J Clin Oncol. 2002;20:2812–2823). This randomized study compared the efficacy and safety of WP to DC in patients (pts) with operable breast cancer. Methods: Pts were randomized 1:1 and stratified by the timing of therapy (preoperative vs adjuvant). The primary endpoint was disease-free survival (DFS), and the secondary endpoint in pts receiving preoperative therapy was complete pathologic response (pCR). WP arm was paclitaxel 80 mg/m2 weekly for 12 weeks, followed by 4 cycles of FEC. DC arm included docetaxel 75 mg/m2 day 1, and capecitabine 1500mg/m2 daily for 14 days of each 3 week cycle for a planned 4 cycles of DC followed by 4 cycles of FEC. The study was designed to include 930 pts to have 80% power to detect a difference in 4 year DFS from 85% on the WP arm to 92% on the DC arm, based on a two-sided test at the 0.05 significance level. Results: 601 pts were randomized, 304 on WP arm and 297 on DC arm. 37% of pts on each arm received therapy in the preoperative setting. Pts characteristics were balanced between the two arms. Median follow-up was 38 months. 4-year DFS on DC arm was 91.0% (95% confidence interval [CI], 84.8- 94.1) and the 4-year DFS on the WP arm was 89.3% (95% CI, 84.8- 94.1) (p = 0.957). A total of 107 pts have been treated with DC and 109 have been treated with WP in the preoperative setting and are evaluable for pCR. The pCR rates were 18.7% and 17.4% on each arm respectively (p = 0.81). The DC arm had higher incidence of hand foot syndrome, and myelosuppression, and WP treatment was associated with higher neurotoxicity. Interim data was presented June 2008 to the institutional data monitoring committee and the study was closed due to futility. If the trial had continued as planned, the predictive probability of concluding in favor of the DC arm for DFS and pCR were 0.005 and 0.001, respectively. Conclusions: WP and DC as utilized in this study had similar efficacy in pts with early stage breast cancer. WP was a better tolerated therapy, associated with little myelotoxicity. [Table: see text]