P4-02-04: Androgen Receptor (AR) Expression in a Cohort of Patients (pts) with Triple Negative Breast Cancer (TNBC).

Abstract

Abstract Background TNBC, defined by the absence of ER, PR, and HER2, is associated with higher risk of recurrence and BC-related mortality, earlier age at diagnosis, menarche, and 1st pregnancy, increased parity, higher BMI, and African-American/Hispanic race. TNBC is a heterogeneous group. Using gene expression analysis, our group described a subset of AR+ ER/PR- BC that exhibits androgen-dependent growth. In vitro studies confirmed the functional role of AR and showed that growth could be abrogated by antiandrogens.(Doane et al 2006) We translated this work into a phase II trial of bicalutamide in pts with AR+ ER/PR- metastatic BC (MBC). (NCT00468715) We now describe the prevalence and clinicopathological characteristics of AR+TNBC in primary disease in a single-institution retrospective cohort. Methods: We identified 1,032 pts with resectable, TNBC (ER/PR<1%; HER2<2+/FISH<2.2) who had surgery at MSKCC from 1998–2006. Exclusion criteria: neoadjuvant chemotherapy, prior radiation, inflammatory/MBC. IRB approval was obtained. We constructed tissue microarrays (TMA) from 210 primary tumors (> 1 cm) with each tumor represented by three 0.6mm cores. AR was tested with DAKO antibody (Clone AR441; dilution 1:500). TMAs were digitized with a Mirax scanner. MetaMorph image analysis software was used to quantify the ratio of DAB staining to hematoxylin signal. A ratio >1 SD above mean was defined as AR+. AR+ cores were manually reviewed; false positives due to core artifact were excluded. To evaluate clinicopathological variables and differences in recurrence-free survival (RFS) and overall survival (OS) by AR status we used chi-square/t-tests and Kaplan-Meier methods/log-rank test, respectively. Results: 169 pts had adequate cores for image analysis/quantification of AR. 10% of pts tested AR+ (17/169). Median (med) followup: AR+=6 years (yr), AR-=5.6yr. Demographic/clinicopathological variables: Table 1 (ages in med yr). Overall med age=54yr (29-84). Adjuvant chemotherapy received: AR+ 82%, AR- 87%, p =0.40; 77% received anthracycline/taxane-based therapy. Med time to distant metastasis (DM)=2.1yr (0.2−6.2yr). We were unable to demonstrate a difference in 5yr RFS (69% vs. 77%; p=0.37) or OS (68% vs. 84%; p=0.25) between AR+ and AR- TNBC. Conclusions: Consistent with our prospective study, AR is expressed in ∼10% of TNBC tumors in this retrospective cohort. The pts in our dataset may be older, postmenopausal, more likely to self-report white race and have T1-2/N0-1 BC. No statistically significant differences were observed in demographic/clinicopathological variables or survival outcomes between AR+ and AR- TNBC. Additional TMA data from our database will be presented. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-02-04.