Axillary ultrasonographic criteria to predict chemotherapy response in breast cancer patients receiving neoadjuvant chemotherapy.

Abstract

48 Background: The purpose of the study is to evaluate ultrasonographic criteria to predict response of axillary LNs in patients with breast cancer receiving neoadjuvant chemotherapy. Methods: We retrospectively reviewed data of 402 patients who underwent neoadjuvant chemotherapy from August 2003 to December 2012. Of these, 202 breast cancers in 199 patients (bilateral in 2 patients) in whom axillary LN metastasis was confirmed using axillary ultrasonography (US) and US-guided fine-needle aspiration biopsy (FNAB) were included in the study. US evaluation was done for the long and short axes, and cortical thickening before and after chemotherapy. The ratio of decreased long and short axes and cortical thickening on prechemotherapy US was also calculated. Statistical analysis with receiver operating characteristic (ROC) curve was used to determine the valuable ultrasonographic criteria for predicting pathologic complete remission (CR) using Medicalc program. Results: Final pathologic diagnoses of LNs were yNo in 57, yN1 in 76, yN2 in 50 and yN3 in 19 cases. The mean size of LNs were 1.8cm in long axis, 1.0cm in short axis and 0.9cm in cortical thickening on prechemotherapy US and 1.1cm in long axis, 0.6 cm in short axis and 0.4 cm in cortical thickening on postchemotherapy US. The area under the ROC curve (AUC) was 0.578 for cortex thickening, 0.523 for the long axis, and 0.523 for the short axis of LN on postchemotherapy US. The AUC was 0.643 for the decreased cortex thickening ratio, 0.585 for the long axis ratio, and 0.585 for the short axis ratio. The most significant criteria was decreased cortex thickening ratio to the long and short axes (p < 0.05). The sensitivity and specificity of the decreased cortex thickening ratio for predicting CR were 43.7% and 80.7%, respectively, with a 0.5 cut off value. Conclusions: The most significant criteria for predicting CR was decreased cortex thickening ratio to the long and short axes. It is applicable for selecting sentinel LN biopsy candidates from among breast cancer patients who have undergone neoadjuvant therapy.