Bridging the translational gap in HNSCC immunotherapy: From resistance mechanisms to high-fidelity preclinical models.

Head and neck squamous cell carcinoma (HNSCC) remains a therapeutic challenge, with immune checkpoint inhibitors (ICIs) benefiting only a minority of patients. This review addresses the translational gap between promising preclinical results and suboptimal clinical outcomes. We critically analyze biological resistance drivers, including the immunosuppressive tumor microenvironment, TGF-β-mediated exclusion, hypoxia-driven metabolic checkpoints (e.g., adenosine signaling), and the distinct immune landscapes of HPV-positive versus HPV-negative disease. We argue that conventional models, such as 2D cell lines, syngeneic mice, and standard immunodeficient xenografts, inadequately recapitulate these human-specific mechanisms and stromal complexities. Consequently, we advocate for a paradigm shift toward high-fidelity platforms, specifically Air-Liquid Interface (ALI) autologous patient-derived organoids (PDOs) and humanized mouse models. While acknowledging technical challenges like establishment efficiency and time-to-answer constraints, we propose that integrating these avatars into biomarker-driven co-clinical trials is essential. Ultimately, integrating these high-fidelity avatars into co-clinical trials will enable precise patient stratification and the rational design of biomarker-driven trials, moving HNSCC therapy from empirical selection to data-driven clinical decision-making. Furthermore, standardizing these high-fidelity protocols to meet clinical turnaround times is the next frontier for ensuring their practical implementation in routine precision oncology for HNSCC.