Single‐Cell Transcriptomics Identifies Novel Prognostic Signatures in HNSCC Immunotherapy Response

ABSTRACT Head and neck squamous cell carcinoma (HNSCC) poses a major therapeutic challenge. In this study, we aimed to analyze tumor immune microenvironment changes and develop a prognostic model based on immunotherapy response. We analyzed single‐cell RNA sequencing data from three HNSCC patients receiving TLR8 agonist and anti‐PD1 combination therapy, identified cell subpopulations before and after treatment with a focus on six major immune cell types, and developed a LASSO‐Cox risk stratification model using combined single‐cell and bulk RNA sequencing data. We identified 19 pre‐treatment and 13 post‐treatment cell subpopulations. Analysis of six major immune cell types revealed differential gene expression patterns. Based on treatment‐induced differential genes, we developed a LASSO‐Cox model with 51 survival‐associated genes, which showed robust predictive performance (AUC: 0.749–0.800) across different timepoints for both HPV‐positive and HPV‐negative patients. High‐risk groups had elevated MDSCs and CAFs, decreased immune cell infiltration (except Th2 CD4+ T cells and common lymphoid progenitors), and increased expression of ICB‐related genes. In conclusion, our model effectively captures patients' immune status and provides insights for optimizing HNSCC immunotherapy strategies.