110P - Predictive value of tumor response rate for axillary response of neoadjuvant chemotherapy in patients with clinically node-positive breast cancer

Date 18 December 2016
Event ESMO Asia 2016 Congress
Session Poster lunch
Topics Anticancer Agents
Breast Cancer, Locally Advanced
Surgical Oncology
Presenter Hyung Suk Kim
Citation Annals of Oncology (2016) 27 (suppl_9): ix30-ix34. 10.1093/annonc/mdw576
Authors H.S. Kim, Y.H. Eom, T.K. Yoo, B.J. Song, B.J. Chae
  • Surgery, Seoul St. Mary's Hospital, of the Catholic University, KS013 - Seoul/KR



Neoadjuvant chemotherapy (NAC) is the standard treatment for patients with clinically node-positive breast cancer. An axillary pathologic complete response (pCR) is associated with excellent prognosis, and patients who achieve axillary pCR can be spared axillary lymph node dissection (ALND). The aim of this study was to assess the factors that predicted axillary pCR and evaluated a model predicting of axillary pCR in our patient population.


We retrospectively identified 201 patients with clinically node-positive breast cancer who were treated with NAC and underwent ALND between 2010 and 2015 at Seoul St. Mary’s Hospital, Catholic University of Korea. We analyzed the Baseline patient and tumor characteristics, clinical tumor response rate, pathologic nodal responses. The tumor response rate was calculated by the rate of tumor and nodal size reduction by the Response Evaluation Criteria in Solid Tumors ver. 1.1. The overall prediction of the model including tumor response rate was assessed by the discriminative performance by receiver operating characteristic (ROC) curve analysis.


Axillary pCR was achieved for 68 patients (33.8%) who underwent ALND after NAC. Patients presenting with high nuclear grade [grade 3 vs. 1 and 2, odds ratio (OR) 2.59], higher Ki-67 value [≥14% vs.〈14%, odds ratio (OR) 1.97] and tumor response rate [≥47.1% vs. < 47.1%, odds ratio (OR) 3.86] were more likely to achieve nodal pCR. In multivariate analysis, tumor response rate was the only independent predictor of a pCR (p = 0.0006). The analysis of tumor response rate values revealed that 47.1% was a reasonable cutoff value for predicting the response to nodal pCR. The discrimination of the model using tumor response rate status versus excluding tumor response rate status [area under the curve (AUC) 67.5%, 95% CI, 0.59-0.76] was significantly improved using tumor response rate status [area under the curve (AUC) 74.9%, 95% CI, 0.68-0.82, P = 0.01].


Tumor response rate can predict axillary pCR in node-positive patients receiving NAC. This prediction model including tumor response rate shows reasonable accuracy for predicting axillary pCR and may have utility for informing treatment decisions.

Clinical trial indentification

Legal entity responsible for the study

Seoul St. Mary\'s Hospital


Seoul St. Mary\'s Hospital


All authors have declared no conflicts of interest.