Abstract 1645: Developing first-in-class AR-V7/AR-fl molecular glue degrader to revolutionize prostate cancer therapeutics

Abstract Metastatic castration resistance prostate cancer (mCRPC) is a lethal disease due to the resistance of standard-of-care (SoC) treatments, including androgen receptor (AR) pathway inhibitors and taxanes. Treatment resistance occurs partly due to the expression of constitutively active AR splice variants, that hijack AR signaling even after castration. AR-V7 is the most prevalent variant and is expressed in ∼75% of patients with mCRPC, which confers resistance to SoC treatment. Currently, there is no AR-V7 pharmacologic inhibitor, limiting patients’ therapeutic options. Thus, AR-V7 inhibition remains an urgent clinically unmet need. To address this gap, we performed a robust high-throughput phenotypic screen of ∼170K compounds to identify the first AR-V7 protein glue degraders. In addition to degrading AR-V7, our hit compound also degraded full-length AR (AR-fl), having a unique chemotype compared to all existing AR modulators. Hit-to-lead optimization through medicinal chemistry and SAR studies has yielded more potent lead AR-V7/AR-fl degraders (IC50 of <100 nM). Mechanistically, our lead compounds promoted Cullin-RING E3 ligase-mediated proteasomal degradation of AR-V7 and AR-fl within 4 hours, which was blocked by the proteasome inhibitor, bortezomib. Currently, while several AR-fl-PROTAC degraders are in preclinical and early clinical development, no AR-V7 degrader exists. Our lead compounds are the only degraders that target both AR-fl and AR-V7. They bind to the TAU1 and AF-1 functional subdomains within the AR N-terminus (NTD), which is shared by both AR isoforms. Preliminary molecular docking experiments using RoseTTAFold2 further supported this binding. Additionally, Cryo-EM analysis identified conformational changes of AR-fl in the presence of our compounds indicating direct binding, which was also confirmed by cellular thermal shift assays. Importantly, our lead compounds effectively reversed SoC treatment-resistance in preclinical models and demonstrated higher potency than enzalutamide or ARV-110 (AR-fl-PROTAC). Our lead compounds showed excellent on-target activity, favorable drug-likeness, and promising pharmacokinetics, with strong stability, oral bioavailability, and efficacy in enzalutamide-resistant mCRPC xenografts expressing both AR-V7 and AR-fl. We are also conducting an E3-ligase CRISPR screen to nominate the E3 ligase that mediates the degradation activity and using global proteomics to assess potential off-target effects. Collectively, these data support a first-in-class molecular glue degrader mechanism, enabling rapid degradation of both AR-fl and AR-V7 through targeting the AR NTD. In summary, our compounds offer a dual-targeting approach in a single treatment and have the potential to dramatically improve survival for patients with mCRPC and prevent the emergence of AR-V7 expression in patients with earlier-stage disease. Citation Format: CheukMan Cherie Au, Michelle Naidoo, Catrina Estrella, Prerna Vatsa, Ying Wang, Zhao Wang, David M. Nanus, Urko del Castillo, Paraskevi Giannakakou. Developing first-in-class AR-V7/AR-fl molecular glue degrader to revolutionize prostate cancer therapeutics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1645.