Abstract
In this study, we sought to determine the predictors of pathological complete response (pCR) and compare the chemotherapeutic regimens administered to breast cancer patients with and those without pCR. We retrospectively reviewed the data of 879 patients treated at the Alvin J. Siteman Cancer Center between 2006 and 2010, to identify patients who were diagnosed with primary stage II or III breast cancer and received neoadjuvant chemotherapy. Patients who received only neoadjuvant endocrine therapy were considered to be ineligible. Patient, tumor, and treatment characteristics, including type of chemotherapy, were compared between patients who did and those who did not achieve pCR using Chi-square or Fishers exact tests and multivariate logistic regression analysis. Two-sided P-values of <0.05 were considered significant. Of the 333 patients who met the inclusion criteria, 61 (18.3%) had documented pCR. Compared with patients not achieving pCR, a greater proportion of patients with pCR had stage II disease (80.3 vs. 68%, P=0.057), had poorly differentiated (grade 3) tumors (82 vs. 59.2%, P<0.001), had negative lymph node involvement (41 vs. 34%, P=0.0004) and had tumors that were HER2-amplified (41 vs. 23.5%, P=0.0054). A greater proportion of patients with pCR received taxane-based chemotherapy (23 vs. 12.5%, P=0.016) or trastuzumab in conjunction with chemotherapy (41.0 vs. 16.9%, P<0.001). No patients receiving solely anthracycline-based therapy achieved pCR in our study. Our study demonstrated that, for stage II and III breast cancer, lower stage, negative lymph node involvement and HER2 receptor amplification were each associated with pCR. Taxane therapy and the concurrent use of trastuzumab were also associated with a higher likelihood of pCR.