Abstract A020: Applying a functional precision medicine platform, Optim.AI™, to identify novel KRAS inhibitor-based combinations in pancreatic cancer

Abstract Purpose: KRAS is one of the most frequently mutated oncogenes in pancreatic cancer, with over 90% of patients harboring such mutations. Increased drug development efforts have recently led to FDA approvals for KRAS G12C inhibitors. However, this is beneficial for only a small proportion of pancreatic cancer patients as G12C mutations are rare (1%) while the majority (89%) are G12D, G12V and G12R. Hence, there is a need to develop strategies to target a broader range of KRAS mutations, including combination therapies. Optim. AI™ is a functional precision medicine platform that combines small data analytics with biological experiments to identify optimal drug-dose combinations based on functional response. It has demonstrated clinical utility in identifying effective combination treatments for hematological malignancies and sarcomas. In this study, we explored the application of Optim. AI™ to identify effective novel drug combinations containing KRAS inhibitors in pancreatic cancer. Methods: Tumor cells were isolated from solid tissues, pleural effusion or ascites of primary and advanced pancreatic cancer samples. Short-term patient-derived organoids (PDOs) were formed before combinatorial treatment with a drug panel comprising FDA-approved chemotherapy, targeted agents, and investigational RAS inhibitors. Post-drug treatment, cell viability was quantified as the phenotypic dataset for Optim. AI™ analysis, which maps drug interactions and ranks all possible combination therapies for each patient sample based on predicted efficacy. Samples were also sequenced to determine their KRAS mutation status. Results: Through Optim. AI™ analysis, pancreatic cancer PDOs demonstrated a range of sensitivities towards different KRAS inhibitors, which correlated with their KRAS mutational profile. Tumors with G12V or G12R mutations showed greater overall susceptibility to pan-RAS inhibitors, such as RMC-6236, as compared to G12C- or G12D-targeting agents. Notably, while published clinical data of RMC-6236 mainly show single agent activity, Optim. AI™ results suggested enhanced activity in combination with gemcitabine. KRAS inhibitors also paired well with defactinib among the top-ranked combinations in several PDOs, with potentially synergistic interactions. In contrast, standard of care regimens such as gemcitabine/paclitaxel and FOLFIRINOX showed minimal efficacy towards the PDOs, consistent with prior line resistance or clinical progression observed in some patients. Conclusion: This study demonstrated the utility of Optim. AI™ in identifying distinct sensitivities of KRAS inhibitors in pancreatic cancer PDOs, concordant with KRAS mutation status. Optim. AI™ highlighted novel synergistic partners with KRAS inhibitors which could result in greater anti-tumor activity. These preliminary findings could be used to expand and stratify KRAS inhibitor-sensitive patients and identify suitable biomarkers to drive precision medicine. Future work will include evaluating the clinical significance of these combinations by validating them across a larger patient sample cohort. Citation Format: Masturah Rashid, Jhin Jieh Lim, Sharon Chan, Edward K. -H. Chow. Applying a functional precision medicine platform, Optim. AI™, to identify novel KRAS inhibitor-based combinations in pancreatic cancer abstract. In: Proceedings of the AACR Special Conference in Cancer Research: RAS Oncogenesis and Therapeutics; 2026 Mar 5-8; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Res 2026;86 (5Suppl₁): Abstract nr A020.