Drug safety is a barrier to the discovery and development of new androgen receptor antagonists

BACKGROUND Androgen receptor (AR) antagonists are part of the standard of care for prostate cancer. Despite the almost inevitable development of resistance in prostate tumors to AR antagonists, no new AR antagonists have been approved for over a decade. Treatment failure is due in part to mutations that increase activity of AR in response to lower ligand concentrations as well as to mutations that result in AR response to a broader range of ligands. The failure to discover new AR antagonists has occurred in the face of continued research; to enable progress, a clear understanding of the reasons for failure is required. METHODS Non‐clinical drug safety studies and safety pharmacology assays were performed on previously approved AR antagonists (bicalutamide, flutamide, nilutamide), next generation antagonists in clinical testing (MDV3100, BMS‐641988), and a pre‐clinical drug candidate (BMS‐501949). In addition, non‐clinical studies with AR mutant mice, and EEG recordings in rats were performed. Non‐clinical findings are compared to disclosures of clinical trial results. RESULTS As a drug class, AR antagonists cause seizure in animals by an off‐target mechanism and are found in vitro to inhibit GABA‐A currents. Clinical trials of candidate next generation AR antagonists identify seizure as a clinical safety risk. CONCLUSIONS Non‐clinical drug safety profiles of the AR antagonist drug class create a significant barrier to the identification of next generation AR antagonists. GABA‐A inhibition is a common off‐target activity of approved and next generation AR antagonists potentially explaining some side effects and safety hazards of this class of drugs. Prostate 77:480–488, 2011. © 2010 Wiley‐Liss, Inc.