Abstract 6019: AI-guided development of CAR-T drugs

Le résumé fourni par la source

Abstract Background: Chimeric Antigen Receptor T-Cell (CAR-T) therapy is an innovative cancer treatment that involves genetically engineering a patient’s T-cells to recognize and attack cancer cells.Artificial Intelligence (AI) is playing an increasingly important role in drug development, accelerating drug discovery, optimizing treatment protocols, and enhancing efficacy.Building on the in-depth research of Nobel laureates Demis Hassabis and John Jumper in structural prediction, AI technology has been combined with antibody library techniques. Under the guidance of AI, antibodies targeting the second epitope of FOLR1 were successfully identified from hundreds of candidates. Methods: In this study, we performed phage display panning for FOLR1, a tumor target, followed by monoclonal identification and sequencing. This resulted in a large number of antibodies targeting the same epitope. Due to the structural characteristics of FOLR1, most antibodies bind to its folate binding site. To obtain antibodies targeting different epitopes and enhance therapeutic efficacy, we combined AI-guided selection with antibody library screening. We conducted NGS sequencing of the antibody library and utilized AI for extensive modeling analysis. Results: Through next-generation sequencing (NGS) of the antibody library, we obtained a large number of sequences distinct from those obtained through monoclonal sequencing. We modeled and docked these antibodies to predict their targeted epitopes. After analyzing over 300 antibodies with the highest binding potential, we identified one antibody that targets a different epitope. This antibody demonstrated significant binding efficacy and does not compete with our previously identified candidate antibodies. Conclusions: By integrating AI technology with our antibody library, we successfully identified a second epitope-targeting antibody against FOLR1. This antibody exhibits properties comparable to those obtained through wet lab experiments. As AI technology continues to advance, its role in antibody development will become increasingly significant, transitioning from AI-assisted drug development to AI-guided drug development. Finally, we extend our heartfelt gratitude to Demis Hassabis and John Jumper for their contributions to AI research. Without their foundational work, it would have been very difficult for us to achieve these results. Citation Format: Chao Cheng, Yanjun Ge, Ermin Xie, Huajing Wang, Xiaowen He, Haiyan Zhu. AI-guided development of CAR-T drugs [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 6019.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.