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(5_Suppl_1):Abstract nr A020.