整合功能性生物特征与同细胞空间多组学以预测头颈癌联合检查点治疗的协同获益

Effective immune checkpoint blockade (ICB) requires coordinated activation, spatial organization, and functional cross-talk among immune populations within the tumor microenvironment (TME). Farcast TruTumor™ platform preserves viable tumor, stromal, and immune compartments in their near-native architecture, aligning with the FDA's New Approach Methodologies (NAM) framework for human-relevant cancer models. The model has shown that in head and neck squamous cell carcinoma (HNSCC) anti-PD-1 response is associated with cytotoxic T-cell (CTL) infiltration into tumor nests, and that moderate responders may be rescued by adding anti-CTLA-4, accompanied by increased spatial separation between CTLs and T regulatory (Treg) cells. Here, we integrate TruTumor's functional characterization with the CosMx® Spatial Molecular Imager (SMI) using same-cell spatial multiomics (64-plex IO protein panel plus whole-transcriptome RNA) to elucidate the dynamics of improved efficacy with nivolumab (N) plus ipilimumab (I) leading to predictors of ICB responses with greater accuracy. Patient-derived HNSCC explants were cultured on the TruTumor histoculture platform and treated for 72 hours with N or N+I. Treatment effects were assessed using histopathology, cleaved caspase-3 IHC, multiparameter flow cytometry, and NanoString nCounter bulk transcriptomics. Spatial profiling was performed using the CosMx SMI same-cell multiomic assay, enabling subcellular quantification of immune checkpoints, activation and exhaustion markers, costimulatory molecules, and whole-transcriptome signatures within intact tissue architecture. Analyses included spatial clustering, cell-cell neighborhood modeling, immune-tumor interface mapping, and ligand-receptor interaction inference. CosMx multiomics revealed treatment-dependent remodeling of immune states and spatial niches not captured by bulk assays. N+I responders exhibited: (i) expansion potential of cytotoxic and proliferative CTLs; (ii) depletion or spatial exclusion of suppressive Tregs from tumor nests; (iii) enhanced antigen-presentation and IFN-gamma-response programs in tumor and myeloid cells; and (iv) rewired immune checkpoint and costimulatory ligand-receptor networks at tumor-immune borders. Same-cell integration of protein and RNA markers defined discrete CTL activation and exhaustion trajectories and uncovered microenvironments uniquely sensitized to combination therapy. Combining functional response metrics with spatial features yielded a mechanism-driven biosignature predictive of treatment outcome. Integrating the TruTumor histoculture models with CosMx SMI same-cell spatial multiomics establishes a scalable, NAM-aligned translational platform that enables discovery of predictive biomarkers for next-generation immuno-oncology therapies.