免疫重编程：胃癌中的个体化T细胞策略

Gastric cancer (GC) is the fourth most common malignancy and a leading cause of cancer-related deaths worldwide. While PD-1/PD-L immunotherapy has become part of the standard of care, only a fraction of patients responds. Therefore, there is a strong rationale to explore alternative immunotherapeutic strategies. Since T cells are central to anti-tumor immunity, expanding tumor-reactive lymphocytes and studying their interactions with the tumor microenvironment (TME), including cancer cells and cancer-associated fibroblasts (CAFs) is crucial. In other cancers, co-culturing tumor organoids (PDO) with peripheral blood lymphocytes (PBLs) has been shown to generates effective tumor-specific T cells. Based on this rationale, the program aims to develop a personalized immunotherapy for GC by generating and expanding antigen-specific cytotoxic T cells from patient blood and using advanced 3D tumor models incorporating PDOs, CAFs, and immune cells to assess T cell-mediated killing, evaluate therapeutic combinations, and identify biomarkers and neoantigens predictive of immunotherapy response, while exploring patient-specific PDO-immune co-cultures to guide personalized T cell-based treatments. Tumor-reactive T cells were generated by co-culturing IFNgamma-stimulated GC organoids with autologous lymphocytes. Their activity was measured by high-throughput imaging and flow cytometry for tumor specificity and immune profiling. Single-cell RNA and TCR sequencing will map T cell activation, differentiation, and clonal expansion, while AI models will identify neoantigens driving patient-specific T cell responses. Functional assays conducted in our lab using the Operetta High Content Imaging System and flow cytometry showed that a two-week co-culture of tumor organoids with autologous PBLs (n=5) generated functional CD8+ T cells with high expression of activation markers (e.g. CD137) and tumor-specific cytotoxicity, including organoid killing. However, when exposed to CAF-released factors, total CD8+ T cells increased while tumor-reactive subsets became fewer and less active. Tumor-specific T lymphocytes were subsequently expanded using the Rapid Expansion Protocol (REP) and reintroduced into 3D models to better mimic the native TME. We established a physiologically relevant co-culture platform that “trains” patient-derived immune cells to selectively kill gastric tumor cells. This system enables the study of tumor-immune interactions, functional testing of novel therapies alone or in combination with immunotherapy, and provides insights into mechanisms of drug resistance and immune evasion. Next, we will define clonal expansion and antigen specificity of reactive T cells to identify safe targets and guide precise immunotherapy development.