We have demonstrated that androgen receptor (AR) agonists durably suppress ER+ breast cancer growth in pre-clinical models [1] and have demonstrated efficacy of an AR agonist for treatment of advanced ER+ breast cancer [2]. Herein, we aimed to identify candidate factors and pathways that influence the tumour suppressive function of AR agonists inn ER+ breast cancer with view to finding novel combination treatment strategies or biomarkers of response.
An unbiased whole-genome CRISPR-CAS9 knockout (KO) screen using the Brunello library [3] was performed in T-47D ER+ AR+ breast cancer cells to identify genes that enhanced or antagonised inhibition of growth following treatment with an endogenous AR agonist, 5α–dihydrotestosterone (DHT, n = 4 biological replicates). The CRISPR-KO screen identified 26 significant sgRNA depletions (KO enhanced DHT-mediated growth suppression; e.g., TFAP2C, SREBF1) and 13 enrichments (KO reduced DHT-mediated growth suppression; e.g., CDK13, RARA). Network analysis using the METASCAPE platform [4] identified pathways necessary for AR activity (e.g. regulation of transcriptional elongation by CDK12/13) or that enhanced AR activity (e.g. lipid metabolism). To date, we found that treatment with a selective CDK12/13 inhibitor (SR-4835) mitigated the anti-proliferative effect of DHT in two ER+AR+ breast cancer cell lines (T-47D, ZR-75-1), validating one of the screen results. In accordance with these findings, loss of CDK12/13 or cofactor cyclin-K, was associated with AR antagonist resistance in prostate cancer [5]. Hence, deficiency in CDK12/13 signalling may be a biomarker of poor response to AR agonist drugs in ER+ breast cancers.
In summary, this project identified multiple candidate factors that are necessary for optimal AR tumour suppression or may enhance this activity in ER+ breast cancers. Further validation of CDK13 and other candidates is underway to characterise the effect of these factors on AR-mediated tumour suppression in this disease.