Prognostic value of Ki-67 labeling index according to age in patients with non-metastasic triple-negative breast cancer.

Abstract

e12118 Background: The prognostic value of Ki67 in triple negative breast cancer (TNBC) is yet unclear because the cut-off points employed differ widely and its predictive effect may vary according to age. The purpose of this study was to analyze the role of Ki-67 among patients with TNBC, and determined the optimal Ki-67 cut-off point to demonstrate its prognostic relevance associated with patient age and treatment strategy. Methods: 201 patients treated for primary TNBC from 1999 to 2014 were identified from the breast surgery database. Clinicopathologic characteristics and outcomes were compared between patients treated with neoadjuvant or adjuvant chemotherapy. We used time-dependent receiver operating characteristic (ROC) curve and time-dependent area under the ROC curve (AUC) to evaluate the discriminative ability of Ki-67 at 3 and 5 years of follow up. A Ki-67 cut-off point was set up to maximize sensitivity and specificity. Interaction effect between age and Ki-67 on disease-free survival (DFS) and overall survival (OS) was evaluated by stratified analysis. Results: There was a non-significant Ki-67 cut off value to predict OS and DFS at 3 and 5 years neither in the whole serie (201) nor in the adjuvant group (125), and only a slightly better threshold for DFS in the neoadjuvant one (N = 46) at 5 years (AUC = 0.697). According to the coordinates of the ROC curves, the best cut-off point for Ki-67 was 60%. In the multivariate analysis (COX proportional hazards regression), high Ki-67 ( > 60%), was a poor prognostic factor for DFS in patients > 40 yo. Among the patients < 40 yo, high Ki-67 was a better prognosis factor. Conclusions: Our data suggest that a threshold of Ki-67 of 60% could provide a usefull tool to define patients with significantly different outcome in TNBC.