Ki-67 labeling index as a predictor of response to neoadjuvant chemotherapy in breast cancer.

Abstract

To investigate Ki-67 index with regard to its ability to predict achievement of pathologic complete response (pCR) to neoadjuvant chemotherapy (NACT) in breast cancer patient. It was a prospective observational study, conducted in Department of Medical Oncology, Rajiv Gandhi Cancer Institute & Research Center (RGCIRC), New Delhi from February 2014 to March 2016. A total of 134 patients with Stage II/III breast cancer who underwent NACT followed by surgery at our center were enrolled and analyzed. Before starting the treatment, clinical, tumor-related and treatment-related factors were recorded. Response evaluation was done clinically and radiologically after completion of NACT and pathologically on the surgical specimen. We calculated Ki-67 cut-off of 35% to label it as high by area under Receiver operating characteristic curve analysis for prediction of pCR. Clinical complete response (cCR) was observed in 35/134 (26.1%) patients while pCR was observed in 32/134 (23.9%) patients. On univariate analysis, higher grade (III), high Ki-67 index (>35%) and number of chemotherapy cycles (>3) were associated with better CCR rates. On multivariate analysis, number of chemotherapy cycles (>3) and high Ki-67 index (>35%) were independent predictive factors. For the predictive factors of pCR, univariate analysis showed grade (III), estrogen receptor/progesterone receptor negativity, HER-2 positivity, number of chemotherapy cycles (>3), TNBC and high Ki-67 index (>35%) to be associated with higher pCR rates. On multivariate analysis, Ki-67 index >35% and HER-2 positivity were the only independent predictive factors of pCR. We suggest 35% as best cut-off for Ki-67 expression for predicting response to NACT and achievement of pCR. Validation of this cut-off is required in larger studies.