Abstract 7437: AI-guided discovery of small molecules targeting eIF4F complex to inhibit oncogenic mRNA translation in castration-resistant prostate cancer

Abstract Effective treatment of castration-resistant prostate cancer (CRPC) remains a significant clinical challenge. Our structure-guided biochemical studies have revealed that the interaction between eukaryotic initiation factors eIF4A and eIF4G within the eIF4F complex plays an essential role for prostate cancer cell survival and CRPC tumor growth. Genetic disruption of the eIF4A-eIF4G interaction in cancer cells selectively inhibits oncogenic mRNA translation, significantly reduces prostate cancer cell survival, and suppresses the in vivo growth of xenografted CRPC tumors. Despite its promise as a therapeutic target, no chemical drug has been developed that effectively disrupts the eIF4A-eIF4G complex to block oncogenic mRNA translation for cancer therapy. To enhance the likelihood of discovering highly biologically active compounds targeting the eIF4A-eIF4G complex, we are developing and utilizing artificial intelligence (AI)-guided “Deep Docking (DD)” algorithms with Schrödinger Molecular Docking platform. This approach enables to develop 'end-to-end' algorithms for structure-based virtual screening (SBVS) of ultra-large chemical libraries containing billions of synthesizable molecules, including novel and unchartered areas of the chemical universe. Building on our resolved structure of the human eIF4A-eIF4G complex, we performed AI-guided SBVS of a chemical library containing 1.5 billion of molecules. By integrating our optimized computational filtering and validation approaches, we identified and selected top-ranked promising candidates for chemical synthesis. Experimental validation of the synthesized compounds demonstrated their abilities to disrupt the eIF4A-eIF4G interaction, block the translation of oncogenic mRNAs such as MYC, MDM2, BCL2, and MCL1, and significantly reduce the survival of CRPC cells. These results demonstrate the effectiveness of our AI-guided platform for structure-based ultra-high chemical library screening, highlighting its potential to accelerate the discovery of drug-like molecules targeting the eIF4F complex. Translationally, the experimentally validated compounds could be advanced into first-in-class therapies for castration-resistant prostate cancer (CRPC) characterized by aberrant activation of the eIF4F complex. Citation Format: Fengze Jin, Puja Singh, Hanyong Chen, Yibin Deng. AI-guided discovery of small molecules targeting eIF4F complex to inhibit oncogenic mRNA translation in castration-resistant prostate cancer [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 7437.