Despite immunotherapy's promise in cancer treatment, patient responses vary substantially because of the individual nature of the immune system and the lack of reliable biomarkers. To address this issue, we developed a precision ex vivo platform that integrates patient-specific tumor and immune cells to study the mechanisms of antitumor immune response, predict immunotherapy outcomes, and identify effective treatments. This platform revealed unique single-cell immune response mechanisms and sensitivities to standard-of-care immunotherapies. Furthermore, we were able to identify a synergistic combination of anti-programmed cell death protein 1 (anti-PD-1) together with a Casitas B lineage lymphoma-b inhibitor that overcame anti-PD-1 resistance in selected patient samples. Activation of the interferon-γ-stimulated cytokines predicted combination efficacy, while immunosuppressive cytokines were associated with poor response. Our findings underscore the platform's potential in tailoring immunotherapies and advancing drug development, offering avenues for personalized cancer treatment.