Abstract Background: Immunotherapies have transformed cancer care, but response variability and the lack of predictive biomarkers limit their broader impact. Tools that improve patient selection and guide immunotherapy design could extend benefits to more patients and cancer types. Optim.AI™ is a functional precision medicine platform with demonstrated clinical utility in identifying patient-specific combination treatments in hematologic cancers and sarcoma. These previous studies lacked immune components, limiting immunotherapy testing. Optim.AI™ 2.0 overcomes this by integrating high-content imaging with tumor–immune co-cultures to enable ex vivo assessment of immunotherapy responses. Here, we evaluated its ability to functionally predict responses to immunotherapy combinations across solid and hematologic tumors. Methods: Fresh peripheral blood mononuclear cells (PBMCs) and tumor cells from non-Hodgkin lymphomas and gynecologic or gastrointestinal cancers were isolated and fluorescently labeled to enable cell-type tracking for high-content screening. Orthogonal array composite design drug combinations were tested in 384-well immune–cancer co-culture models to evaluate indication-specific 12-drug panels, including monoclonal antibodies, antibody-drug conjugates, and bispecifics. High-content imaging of tumor-specific cell death served as the quantitative phenotypic readout for Optim.AI™ 2.0. The platform uses a hybrid experimental–analytical approach that searches 531,441 permutations derived from 155 ex vivo test combinations to rank all clinically actionable treatments within a 12-drug panel. Predicted immunotherapy responses were compared with treatment histories to assess clinical concordance. Results: Optimized effector–target ratios enabled precise measurement of immune-mediated tumor killing after drug treatment. High-content imaging quantified tumor-specific cytotoxicity and visualized key immune–tumor interactions, including immune cell migration and infiltration. The platform demonstrated feasibility and specificity across cancer types by detecting antigen-dependent responses and distinguishing context-specific antibody-based combinations in both solid and hematologic samples. Optim.AI™ 2.0 also accurately predicted sensitivity and resistance to first-line rituximab in treatment-naïve and relapsed/refractory DLBCL, while highlighting novel immunotherapy combinations with potential benefit in later treatment lines. Conclusion: Optim.AI™ 2.0 enables high-content functional evaluation of immunotherapy combinations in patient-derived tumor-immune co-cultures, generating actionable ex vivo data that can complement standard clinical and molecular diagnostics. With further clinical validation, this platform has the potential to guide personalized immunotherapy selection and, when integrated with molecular profiling, to reveal novel biomarker-informed combinations for defined patient subsets. Citation Format: Sharon Chan, Jhin Jieh Lim, Masturah Rashid, Edward Chow. Clinical evaluation of a high-content functional precision medicine platform for predicting immunotherapy combination response in solid and hematological cancers abstract. In: Proceedings of the AACR Immuno-Oncology Conference (AACR IO): Discovery and Innovation in Cancer Immunology: Revolutionizing Treatment through Immunotherapy; 2026 Feb 18-21; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Immunol Res 2026;14(2 Suppl):Abstract nr C018.
Chan et al. (Wed,) studied this question.