Background: The androgen receptor (AR) is expressed in the majority of patients (pts) with ER/PgR+ breast cancer (BC)(Collins LC et al. Mod Pathol. 2011;24:924–931; Loibl S et al. Breast Cancer Res Treat. 2011;130:477–487). In ER+ cell lines, AR over-expression can lead to hormone-resistance (De Amicis F et al. Breast Cancer Res Treat. 2010;121:1-11). Androgen stimulated growth of ER+/AR+ cells lines can be blocked by enzalutamide (ENZA), a potent oral inhibitor of AR. Treatment with exemestane (EXE) can increase androgen levels (Takagi K, et al. Endocr Relat Cancer. 2010;17:415–430), the addition of ENZA to EXE may inhibit any androgen-mediated breast tumor growth. Methods: This randomized placebo (PBO) controlled phase 2 trial will evaluate EXE with or without ENZA (160 mg/day) in 240 patients (pts) with advanced BC (aBC) that is ER/PgR+ and Her2 normal (NCT02007512). Two parallel enrolling cohorts (C) will each evaluate 120 pts who have received either no (C1) or 1 (C2) prior hormone therapy for aBC. Randomization (1:1) is stratified (Y/N) by prior hormone, prior aromatase inhibitor (AI) and hormone resistance. Eligible pts must have ≤ 1 non-hormone regimen and ECOG ≤1; non-measurable bone or skin disease is allowed. Pts with CNS disease or history of seizure are excluded. Tissue submission for determination of AR expression is mandatory. The co-primary endpoint (EP) is progression-free survival (PFS) in all pts and in those with AR+ disease. Cross-over is allowed following RECIST 1.1 progression. Additional EPs include clinical benefit rate (complete/partial response or stable disease at ≥24 weeks); response rate; duration and time to response, safety and tolerability. Efficacy analyses will be conducted on all randomized pts (ITT), a minimum of 90 PFS events/cohort is required to estimate the median PFS and hazard ratio. Approximately 55 PFS events, for a target HR = 0.67, are required in the AR+ subset. The definition of “AR+” will be determined prior to unblinding. Clinical trial information: NCT02007512.