Abstract
Molecular sub-characterization of triple-negative breast cancer (TNBC) has great therapeutic and possibly prognostic implications. The primary aim of this study was to investigate the incidence of luminal androgen receptor (LAR) subtype of TNBC and secondary aims were sub-categorization and clinico-pathologic correlation of LAR breast cancers. Retrospective study (January 2008 and 31st of December 2018) consisting of 157 TNBC patients. Androgen receptor (AR) expression was measured by immunohistochemical analysis. One percent cutoff was set as a positive expression. Sub-categorization was done on the basis of EGFR (> 15% of tumor cells) and Ki-67 expression (low- < 11%, intermediate- 11-20%, and high- > 21%). AR expression was correlated with various clinico-pathologic features and outcomes of the patients. The incidence of AR expression in TNBC was 24.8%. Considering different thresholds of > 5%, > 10%, and > 20% immunostaining, the incidence of AR positivity was 18.4, 15.2, and 11.5% respectively. The incidence of Ki-67 (p = 0.89) and EGFR (p = 0.643) expression did not differ significantly in AR-positive and -negative TNBC. Based on EGFR expression 19, 67 and 14% patients were categorized as low, intermediate, and high risk respectively. Low-risk (p ≤ 0.001) and low-grade (p = 0.014) tumors were more likely to have > 10% AR expression. Clinico-pathological profile, response to neoadjuvant chemotherapy, disease-free survival (p = 0.458), and overall survival (p = 0.806) did not significantly differ between AR expressing and negative TNBC. On multivariate analysis, only tumor staging was a significant predictor of survival (p = 0.012) and AR expression of > 10% revealed a trend towards improved survival (p = 0.07). When considering only AR-positive TNBC, AR expression of > 10% (p = 0.038), distant metastases (p = 0.003), and EGFR status (p = 0.024) were significantly associated with survival. AR expression does not seem to very strongly correlate with prognosis in TNBC and further studies could focus more on its predictive role in deciding anti-androgen therapy.