Emerging mechanisms of resistance to androgen receptor inhibitors in prostate cancer

Philip A. Watson,Vivek K. Arora,Charles L. Sawyers

DOI: https://doi.org/10.1038/nrc4016

IF: 78.5

2015-11-13

Nature Reviews Cancer

Abstract:Key PointsProstate cancer pathogenesis is dependent on signalling through the steroid nuclear hormone androgen receptor (AR), which is activated after binding of the androgen ligand testosterone or dihydrotestosterone. Ligand-bound AR translocates to the nucleus, where it serves to induce or repress gene expression through binding to chromatin at cis androgen response elements.Medical castration to substantially deplete serum testosterone is the mainstay of therapy for advanced prostate cancer that recurs following surgical removal of the prostate (prostatectomy) or radiotherapy. However, castration therapy is not curative, and patients will eventually progress to lethal castration-resistant prostate cancer (CRPC).Despite a castrate level of testosterone, CRPC almost uniformly remains dependent on AR signalling. Next-generation hormonal therapies for prostate cancer, abiraterone and enzalutamide, are now in widespread clinical use; abiraterone attacks AR signalling through inhibition of extra-gonadal androgen biosynthesis and enzalutamide interferes directly with androgen binding to AR.Resistance mechanisms to these drugs have been identified that result in restoration of AR signalling through gain-of-function AR mutations, upregulation of constitutively active AR splice variants or increased intratumoural androgen biosynthesis. Another resistance mechanism bypasses AR by switching to the related glucocorticoid receptor (GR) to maintain transcriptional regulation of a subset of the same genes.At resistance, a subset of patients are now presenting with low or no AR in their tumours, suggesting that evolution to complex genomic states completely independently of AR could increasingly become a cause for concern.Comprehensive analyses of late-stage CRPC are uncovering multiple genetic lesions in this patient cohort that indicate that it may eventually be possible to stratify patients based on the genomic profile of their cancer. These efforts will aid in clinical trial design and facilitate the use of rationally designed combination strategies to improve patient outcomes.

oncology