P3-14-11: Comparison of Two Nomograms To Predict Pathologic Complete Response to Neoadjuvant Chemotherapy – Evidence That HER2 Positive Tumors Need Specific Predictors.

Abstract

Abstract Aim: The purpose of this study was to compare two different nomograms to predict pathologic complete response (pCR) to preoperative chemotherapy in an independent cohort of 200 patients with breast cancer. The first model was the MDACC nomogram published in 2005 and the other one was a nomogram based on the number of preoperative courses, Ki-67 and steroid hormone receptors expression published by Colleoni et al. in 2010 Patients and methods: Data from 200 patients with breast carcinoma treated with preoperative chemotherapy and operated at Tenon Hospital from 2001 to 2009 were collected. We calculated pCR rate predictions with the two nomograms and compare those predictions with outcome. Patients received between 4 and 8 course of anthracycline/taxanes based chemotherapy. More than 90% of patients with HER2 positive tumors received concomitant trastuzumab with taxanes. Model performances were quantified with respect to discrimination (evaluated by the areas under the receiver operating characteristics curves (AUC)) and calibration. Results: In the entire population, the AUC for the MDACC nomogram and the Colleoni nomogram were respectively 0.74 and 0.75. Both of them underestimated the pCR rate (p=0.02 and 0.0005). When excluding patients treated with trastuzumab, the AUC were 0.78 for both of them with no significant difference between the predicted and the observed pCR (p=0.14 and 0.15). When analyzing the specific population treated with trastuzumab as preoperative treatment, the AUC for the MDACC nomogram and the Colleoni nomogram were respectively 0.52 and 0.53. Conclusion: The MDACC and the Colleoni nomograms are accurate to predict the probability of pCR after preoperative chemotherapy in HER2 negative population but did not correctly predict pCR in HER2 positive patients who received trastuzumab. This suggests that response to preoperative chemotherapy including trastuzumab is biologically driven and that a specific nomogram or predictor for HER2 positive patients has to be developed. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-14-11.