A new predictive marker for predicting response after neoadjuvant chemotherapy in hormone receptor positive/HER2-negative patients: a logarithmic model

Abstract

Purpose: Estrogen receptor (ER) and progesterone receptor (PgR) levels as well as Ki-67 expression levels are independent predictive markers in patients with hormone receptor-positive breast cancer. In this study, we investigated the predictive significance of the formula of log (ER)*log (PgR)/Ki-67, which was created using 3 independent predictive markers, for the pathological complete response of the Hormone Receptor (HR)-positive/HER2-negative breast cancer patients receiving neoadjuvant chemotherapy (NACT). Methods: This retrospective study included 126 patients with HR-positive/HER2-negative breast cancer and axillary lymph node metastasis who received NACT. The log (ER)*log (PgR)/Ki-67 value was calculated from the pre-NACT pathological evaluation results in all patients. We determined the ideal predictive cut-off value, which separates patients into 2 groups according to pathological complete response (pCR) and pathological non-complete response (non-pCR), using Receiver Operating Characteristic (ROC) curve analysis. According to this cut-off point, patients were divided into 2 groups as cut-off ratio high and cut-off ratio low and were compared using logistic regression analysis along with clinicopathological features. Results: According to the predictive model, we estimated the ideal cut-off value that distinguishes patients as pCR and non-pCR to be 0.12 (p=0.015). According to this cut off value, %54.8 of the patients were categorized as cut-off value high and %46.2 were cut-off value low. The non-pCR rates of the groups were 91.3% and %71.9, respectively(p=0.004). A cutoff value of 0.12 provided the feature of being a predictive marker in the univariate analysis for distinguishing between pCR and non-pCR (OR=4.09 95% CI 1.48-11.33, p=0.007), and it preserved this feature in the multivariate analysis. (OR=3.27, 95% CI 1.12-9.56, p=0.030). Conclusion: The formula of log (ER)*log (PgR)/Ki-67 can be used as a simple and easy-to-use predictive marker for response to neoadjuvant therapy in patients with HR-positive/HER2-negative breast cancer receiving NACT. © 2021 Zerbinis Publications. All rights reserved.