Abstract 550: Patient-derived organoids as a screening platform to identify mechanisms of osimertinib resistance in non-small cell lung cancer

Abstract Background: Epidermal growth factor receptor (EGFR)-positive non-small cell lung cancer (NSCLC) represents a large subset of NSCLC, accounting for approximately 30% of cases. Despite the initial effectiveness of osimertinib, the frontline modality in EGFR-positive NSCLC, resistance remains a major therapeutic challenge. To address osimertinib resistance, preclinical models that accurately represent the complexity of tumor biology are needed. Patient- and xenograft-derived organoids (PDOs and XDOs) offer a promising option as they preserve the genetic and phenotypic characteristics of the original tumor, allowing for more clinically relevant studies. We developed a panel of EGFR-positive PDOs and XDOs with fully annotated molecular, genetic, clinical and treatment profiles, and will utilize them to investigate osimertinib resistance. Method: To demonstrate the utility of these models as a screening platform to identify osimertinib resistance mechanisms, we conducted a pilot single-agent screen in 6 PDOs/XDOs using osimertinib and a panel of small-molecule inhibitors that target relevant targets in NSCLC including Aurora kinase B, CDK4/6, c-MET, ALK, RET, ROS1, MEK, and BRD4. Viability has been assessed by the ATP luminescence assay CellTiter-Glo. Subsequently, we performed a high-throughput combinatorial screen where 2 osimertinib resistant models (LPTO357 & LPTO362) were treated with osimertinib alone and in combination with a library of 1144 FDA-approved and investigational drugs. Result: We identified several compounds that displayed efficacy and potency in osimertinib-resistant models. The c-MET inhibitor capmatinib was specifically potent in LPTO245 but not in other models. This finding was confirmed using savolitinib, another selective c-MET inhibitor, suggesting the observed effects were due to on-target c-MET inhibition. Using whole exome sequencing, we confirmed the presence of c-MET amplification in LPTO245. Additionally, we identified an inverse sensitivity correlation between osimertinib and the Aurora kinase B inhibitor AZD2811 across all 6 models (r = -0.64; p = 0.03). The following analysis of our RNAseq data has shown a positive correlation between AURKB RNA expression and osimertinib IC50 values in 9 PDOs/XDOs. These data corroborate previously reported role of c-MET and Aurora kinase B as osimertinib resistance mechanisms and demonstrate the utility of the platform. Subsequently, our high-throughput combinatorial screen has identified 125 hits that potentiate/synergize the action of osimertinib including drugs targeting topoisomerases, JAK, PLK1, and PI3K. Conclusion: Our screening platform can provide novel biological insights into osimertinib resistance. Moreover, it can potentially identify drug combinations with immediate clinical utility in NSCLC, particularly with approved and clinical-grade drugs. Citation Format: Yifan Yu, Samir H. Barghout, Nikolina Radulovich, Geoffrey Liu, Ming Tsao. Patient-derived organoids as a screening platform to identify mechanisms of osimertinib resistance in non-small cell lung 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 550.