Th17-mediated antitumor immunity in patient-derived organoid and autologous immune cell cocultures predicts response to immunotherapy in head and neck cancer

Background Immune checkpoint inhibitor (ICI) therapy improves survival in head and neck squamous cell carcinoma (HNSCC), yet only some patients benefit, highlighting the need for early identification of responders and functional evaluation of tumor–immune interactions. We developed an autologous coculture platform combining patient-derived organoids (PDOs) with peripheral blood mononuclear cells (PBMCs) to assess immune responses and correlate preclinical findings with clinical outcomes. Material and methods PDOs from HNSCC specimens were expanded and cocultured with PBMCs, followed by treatment with ICIs targeting programmed cell death protein 1 (PD-1), lymphocyte-activation gene 3 (LAG-3), or their combination. Results PBMCs were enriched for CD4 + and CD8 + T cells (∼90%) and CD16 + CD56 + natural killer cells, with baseline exhaustion (LAG-3 + , CTLA-4 + ). T cell–mediated tumor killing occurred in 44% of cases and was accompanied by Th1/Th17 cytokine signatures. ICI treatment modestly shifted T-cell phenotypes, reducing exhaustion (PD-1) and enriching cytolytic subsets (CD4 + IFN-γ + , CD8 + TOX + ). Interleukin (IL)-17A emerged as a relevant cytokine, correlating with FasL secretion and suggesting activation of the Fas/FasL axis. High circulating IL-17A levels in treatment-naïve HNSCC patients were associated with improved survival, whereas other cytokines showed no prognostic value. In a noninterventional observational trial, increases in IL-17A positively correlated with favorable ICI responses, indicating its potential as a biomarker. PDO viability decreased in response to ICIs in a patient-specific manner, consistent with cytokine signatures and clinical outcomes. Conclusion Collectively, this autologous PDO–PBMC coculture platform, combined with longitudinal cytokine profiling, captures patient-specific tumor–immune interactions and interpatient heterogeneity, providing a framework for real-time response prediction, biomarker discovery, and personalized immunotherapy in HNSCC.