Abstract Introduction: Esophageal adenocarcinoma drives most esophageal cancer mortality in North America, with poor 5-year survival (15-25%) despite taxane-based triplet chemotherapy. Extensive inter- and intra-patient genomic heterogeneity often limits the effectiveness of genetically guided targeted therapies and contributes to resistance to chemotherapy and immunotherapy. Currently, no definitive guidelines exist to predict patient-specific treatment response. Conventional animal models and 2D cell lines inadequately recapitulate the tumor microenvironment (TME), underscoring the need for personalized platforms that capture interactions among immune cells, cancer-associated fibroblasts (CAFs), and the vasculature. In this study, we developed an immune-competent esophageal adenocarcinoma-on-a-chip model that incorporates both stromal components and immune cells and enables personalized therapy testing by simulating patient-specific tumor-immune interactions within the TME. Methods: Treatment-naive patient-derived organoids (PDOs), matched CAFs, autologous tumor-infiltrating lymphocytes (TILs), and primary esophageal endothelial cells were integrated into a two-channel chip (Emulate). Tumor organoids were seeded in the epithelial channel, while CAF and endothelial cells occupied the stromal channel, separated by a porous PDMS membrane. Flow at 60 µl/h and stretch (10%, 0. 15 Hz) were applied to mimic perfusion and peristalsis. After stromal monolayer formation, labeled TILs were perfused for 2 hours at 1000 µl/h. 3D microtissue organization and TIL dynamics were monitored for 10 days. Parallel static transwell cultures (-flow/-stretch) using the same patient-derived cells were established to assess the impact of dynamic biomechanical cues. Results: Real-time imaging demonstrated tumor growth and dynamic TME remodeling for up to 10 days under physiologically relevant flow and shear stress. Progressive formation of complex 3D microtissues was observed in both epithelial and stromal channels. Autologous TILs adhered within the stromal channel, proliferated, and formed clusters, with some migrating toward the tumor-containing epithelial channel, indicating active tumor-immune crosstalk. Matched static transwell cultures failed to generate comparable 3D stromal structures or directional TIL migration, confirming the critical role of flow and cyclic stretch in stromal architecture and immune recruitment. Conclusion: This patient-specific, TME-inclusive esophagus-on-a-chip platform offers a robust alternative to animal models by capturing physiological flow and stretch-driven tumor-immune dynamics. Our model enables rapid ex vivo assessment of personalized chemotherapy and immunotherapy responses, offering clinically relevant results to guide precision treatment in esophageal adenocarcinoma. Citation Format: Manal Al Dow, Sanjima Pal, Cedric Julien, Anna Staravoitava, Betty Giannias, Nick Bertos, Kim Ma, Lorenzo Ferri, Jonathan Cools Lartigue. Patient-specific esophageal adenocarcinoma-on-a-chip: Modeling tumor-stromal-immune interactions and therapeutic response abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts) ; 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86 (8Suppl): Abstract nr LB487.
Dow et al. (Fri,) studied this question.