Predictive factors of no response during neoadjuvant chemotherapy in breast cancer.

Abstract

e11508 Background: Neoadjuvant chemotherapy (NACT) has been the standard for patients with inflammatory (IBC) and locally advanced breast cancer (LABC) and increases breast conservation rates in patients with operable tumor. Pathologic complete response (pCR) after NACT has been accepted as a marker for improved long-term outcome and predictors of pCR have been reported in many trials. However, predictors of progression or stable disease have been limitedly reported. We aimed to identify predictors of no response during NACT to identify patients who might benefit from an alternative approach. Methods: The medical records of patients with breast cancer who received NACT were reviewed retrospectively at a single center. Between March 2000 and January 2010, 316 patients were enrolled with stage I to III breast cancer. Statistical analyses were performed to compare patients with any response (RG) with patients with progression or stable disease (NRG). Results: Out of 316 patients, 263 patients (83.2%) had some response, 36 (11.4%) had SD, and 17 (5.4%) had PD during NACT. Factors predictive of NRG included nodal (N) status (p=.007) and clinical stage (p=.004), IBC cancer type (p=.02), negative estrogen receptor (ER) status, and negative HER2 status (p=0.04). Pre-NACT N stage, ER status, cancer type, and treatment with trastuzumab were independent predictors of NRG in multivariate analysis. NRG was a negative predictor of disease-free survival (p=.002) and overall survival (p=.026) in multivariate analysis. Conclusions: Factors predicting NRG include advanced nodal stage, ER negativity, IBC and NACT without trastuzumab in HER2-positve tumors. Because these variables are associated with poor prognosis, novel targeted therapies and molecular predictors are needed to improve outcome