A novel immune-related gene signature stratifies prognosis and characterizes the tumor immune microenvironment in head and neck squamous cell carcinoma.

Head and neck squamous cell carcinoma (HNSCC) exhibits heterogeneous responses to immunotherapy, primarily determined by the tumor immune microenvironment (TME). Accordingly, there is an urgent need for reliable biomarkers that can effectively distinguish between immune "hot" and "cold" tumors and predict prognosis and immunotherapy response. We integrated transcriptomic and clinical data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) cohorts to identify immune-related genes (IRGs) associated with TME phenotypes. Unsupervised clustering was used to define immune-hot and immune-cold subtypes. A six-gene prognostic signature (PYGL, SFRP1, FGD3, OLR1, DUSP9, and MASP1) was developed using Cox and Least Absolute Shrinkage and Selection Operator (LASSO) regression, which was validated across four cohorts and evaluated for its association with immune infiltration, tumor mutation burden (TMB), and immunotherapy response. The functional role of PYGL was assessed via siRNA knockdown in HNSCC cell lines. The signature demonstrated good performance in stratifying patients into high- and low-risk groups, characterized by distinct survival, immune landscapes, and genomic mutation profiles. High-risk patients showed an immunosuppressive TME, higher TMB, and poorer response to immune checkpoint blockade. PYGL knockdown suppressed proliferation, migration, and clonogenicity. Transcriptomic analysis revealed downregulation of MYC/E2F targets and EMT pathways, alongside concurrent activation of interferon and immune response programs. This IRG-based signature offers a clinically translatable tool for prognostication and immunotherapy selection in HNSCC. PYGL represents a key oncogenic driver and potential therapeutic target linking cellular metabolism to tumor progression and immune evasion.