Introduction:
Intratumoral heterogeneity is a major determinant of progression and treatment failure in Head and Neck Squamous Cell Carcinoma (HNSCC). While genetic diversity has been extensively characterized, the contribution of extracellular matrix (ECM)-mediated signaling to functional heterogeneity remains less well defined. Given the central role of the tumor microenvironment in shaping cellular behavior, we sought to examine how ECM ligands influence pathway activity across distinct cellular compartments at single-cell resolution.

Methods:
Single-cell RNA sequencing data from HNSCC tumors were analyzed using a curated NABA core matrisome gene set. The Matrix Influence Scoring Algorithm (MASC) was applied to quantify the regulatory influence of ECM ligands on hallmark cancer pathways, including epithelial–mesenchymal transition (EMT), angiogenesis, and hypoxia. Influence scores were computed per cell and subsequently aggregated by annotated cell types to evaluate compartment-specific signaling patterns.

Results:
TGFBI, PAPLN, and MFGE8 consistently emerged as the most influential ECM ligands across pathways. Their regulatory impact varied markedly between cellular populations. Cancer-associated fibroblasts and myofibroblasts exhibited elevated ECM-associated EMT influence scores, endothelial cells displayed higher angiogenesis-related ECM influence, and malignant epithelial cells showed variable hypoxia-associated signaling patterns across pathways. Correlation analyses revealed substantial cross-pathway consistency in ligand influence, suggesting coordinated upstream regulation. Importantly, ECM-driven signaling was spatially and functionally restricted rather than uniformly distributed across the tumor ecosystem.

Conclusions:
These findings highlight the ECM as a critical contributor to functional heterogeneity in HNSCC. Single-cell influence modeling uncovers compartment-specific signaling programs that may shape tumor plasticity and therapeutic response, underscoring the potential of targeting microenvironmental signaling networks in precision oncology.