Resistance to endocrine therapy is a major clinical problem in estrogen receptor positive (ER+) breast cancer. The androgen receptor (AR) is expressed in ~90% of all ER+ breast cancers and its expression is retained in a majority of endocrine resistant tumours. Generally, AR activation in treatment-naïve breast cancer cell line models antagonizes ER-mediated proliferation which is consistent with observations that high AR expression is associated with a better patient outcome in this subtype. However, opposing effects of AR signalling have been reported in endocrine resistant tumours. This uncertainty surrounding the role of AR in endocrine resistant tumours has led to a number of current clinical trials in which AR-targeted agents with contrasting mechanisms are being investigated. In this study, we sought to investigate the role of AR in endocrine resistant tumours and provide in vivo efficacy data of AR-targeted agents to inform the rational design of future clinical trials. We have characterised the role of AR in endocrine resistant cell lines using siRNA against AR. Knockdown of AR in MCF7 and T47D tamoxifen-resistant (TAMR) and long term estrogen derived (LTED) cells reduced cell proliferation suggesting a pro-survival role of AR. Gene expression profiling post AR knockdown in MCF7 TAMR and MCF7 LTED identified changes in common pathways pertaining to “estrogen response” and PI3K-Akt signalling pathways. Despite increased nuclear ER post AR knockdown, “estrogen response genes” were both up- and down-regulated suggesting that AR may differentially affect ER binding at these sites. In vivo the growth of AI resistant PDX model was inhibited by AR agonist DHT whereas Enobosarm, a selective AR modulator, elicited a heterogeneous response and AR antagonist enzalutamide had no effect on tumour growth.In the DHT-treated and Enobosarm-treated responders, anti-tumour effects were associated with significant decreases in nuclear ER and the proliferation marker ki-67. Collectively, our preliminary data show that while AR appears to be pro-survival in endocrine resistant cells, in vivo the role of AR is more complex and we propose that the optimal approach for targeting AR in patients with endocrine resistant disease may be by an agonistic or modulatory approach rather than an antagonistic approach.