Abstract 4287: GBC-11004: an AI-driven novel kinase target with potential to overcome osimertinib resistance in NSCLC

Abstract Osimertinib resistance poses a significant clinical challenge in the treatment of non-small cell lung cancer (NSCLC), with diverse mechanisms complicating patient outcomes. Conventional next-generation sequencing (NGS) analysis methods often fall short in identifying effective therapeutic targets due to the complexity and heterogeneity of resistance mechanisms. To address this issue, we have developed an artificial intelligence (AI)-driven target discovery platform designed to identify novel and effective drug target genes capable of overcoming Osimertinib resistance in NSCLC, thereby surpassing the capabilities of traditional NGS analysis. Our platform integrates three key components: deep learning (G-TAC), statistical significance testing (G-SET), and a large language model (G-LAT). G-TAC and G-SET evaluate and rank genes according to their responsiveness to Osimertinib and tumor-specific expression. G-LAT assesses these genes based on publications and clinical trials to ensure novelty and efficacy of the identified targets. We identified a novel kinase target named GBC-11004, as one of the top-ranked target genes that were found to be overexpressed in patient-derived organoids (PDOs) resistant to Osimertinib. To ascertain the functional impact of GBC-11004, target validation was conducted using PDOs and CRISPR/Cas9-based gene editing. Gene editing in Osimertinib resistant PDOs resulted in a significant decrease in cell viability corresponding to increased indel frequency. Furthermore, we have initiated lead compound optimization by preliminary IC50 analyses using compounds targeting GBC-11004 and observed significantly enhanced sensitivity in the combination therapy group (Osimertinib + GBC-11004 inhibitor) compared to the Osimertinib monotherapy group in resistant PDO models. Our results demonstrate the potential of GBC-11004 as a novel therapeutic target for overcoming Osimertinib resistance in NSCLC treatment and emphasize the capability of our PDO-based AI-driven target discovery platform in identifying high-priority novel targets. Citation Format: Hat Nim Jeon, Ji Young Bae, Hyunsun Na, Ji Hwan Park, Seung Hyun Ahn, Bada Pyo, Jihye Ryu, Jinguen Rheey. GBC-11004: an AI-driven novel kinase target with potential to overcome osimertinib resistance in NSCLC [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 4287.