The standard of care for locally advanced breast cancer (LABC) is neoadjuvant chemotherapy,¹ with LABC including clinical stages IIA, IIB, and IIIA. The goals of preoperative chemotherapy are to downstage so as to render breast conservation feasible, to eradicate disease in the axillary nodes, and to allow in vivo testing of tumor drug sensitivity, all with the ultimate aim of improving prognosis. Clinical trials have demonstrated that the pathologic in-breast response generally correlates with pathologic response in the lymph nodes. Furthermore, nodal status at the time of surgery correlates with overall survival (OS) and disease-free survival (DFS).²,³ A combined analysis of two large prospective neoadjuvant chemotherapy trials demonstrated significantly higher 5-year OS and DFS in patients achieving in-breast pathologic complete response (pCR), compared with those who did not (OS, 89% vs 64%; DFS, 87% vs 58%, respectively).⁴

At the start of this trial, the most effective neoad- juvant regimen remained in question. Even now, National Comprehensive Cancer Center guidelines suggest that any recommended adjuvant regimen can be used in the neoadjuvant setting.¹ Numerous phase II and III trials have evaluated single-agent^5–8 and combination^{9– 32} chemotherapies, most of which are anthracycline- based, with pCR rates reported between 7% and 36%. In the NSABP-B27 study, patients treated preoperatively with four cycles of doxorubicin and cyclophosphamide (AC) followed by four cycles of docetaxel (Taxotere) had a 26% pCR rate versus a 13% pCR rate in those receiving preoperative AC and postoperative docetaxel. Despite the doubling of pCR with neoadjuvant docetaxel, there was no difference in DFS or OS.9 However, as reported by Kuerer et al, patients achieving a pCR after completion of neoadjuvant chemotherapy appeared to have superior survival.⁴

Many previous trials (including the study reported here) did not exclude patients with human epidermal growth factor receptor 2 (HER2)-positive disease. It is now well established that such patients should be treated with neoadjuvant regimens incorporating HER2-targeted therapy. In fact, an early neoadjuvant study of paclitaxel followed by fluorouracil, epirubicin, and cyclophosphamide with or without 24 weeks of concurrent trastuzumab (Herceptin) in patients with HER2-positive tumors was closed early because patients receiving trastuzumab had a pCR rate of 65%, compared with 26% in those who did not receive it.³³ Expanded clinical trials of this approach are in progress.

The selection of capecitabine (Xeloda) and docetaxel in the present trial was based on the hypothesis that the upregulation of thymidine phosphorylase by docetaxel should increase the activity of capecitabine. ^34–36 Single-agent docetaxel in the neoadjuvant setting has yielded pCR rates of 7%–20%.^6–8 Treatment with docetaxel and capecitabine together has been reported to produce pCR rates of 10%–21%.^37–39 The addition of carboplatin was based on studies by Hurley et al at the University of Miami^{39– 41} suggesting that platinum salts appeared quite active in the neoadjuvant setting, with the combination of docetaxel and cisplatin producing a pCR rate of 20%, with no residual disease in the breast or axilla.40 Other regimens incorporating cisplatin or carboplatin have pCR rates ranging from 16% to 24%.^27,42–44

Patients and methods

Study design

In this phase II multicenter study, patients were assigned to receive docetaxel (30 mg/m2 IV) and carboplatin (AUC 2 IV) on days 1, 8, and 15 of each 28-day cycle plus capecitabine (625 mg/m2 PO) twice daily on days 5–18. The capecitabine dose was based on observations that this dose was effective and relatively nontoxic in metastatic breast cancer (C.L. Vogel, empirical observations). Patients were to receive four cycles prior to surgical resection.