Building an Immune-Related Genes Model to Predict Treatment, Extracellular Matrix, and Prognosis of Head and Neck Squamous Cell Carcinoma

Abstract

Due to the considerable heterogeneity of head and neck squamous cell carcinoma (HNSCC), individuals with comparable TNM stages who receive the same treatment strategy have varying prognostic outcomes. In HNSCC, immunotherapy is developing quickly and has shown effective. We want to develop an immune-related gene (IRG) prognostic model to forecast the prognosis and response to immunotherapy of patients. In order to analyze differential expression in normal and malignant tissues, we first identified IRGs that were differently expressed. Weighted gene coexpression network analysis (WGCNA) was used to identify modules that were highly related, and univariate and multivariate Cox regression analyses were also used to create a predictive model for IRGs that included nine IRGs. WGCNA identified the four most noteworthy related modules. Patients in the model’s low-risk category had a better chance of survival. The IRGs prognostic model was also proved to be an independent prognostic predictor, and the model was also substantially linked with a number of clinical characteristics. The low-risk group was associated with immune-related pathways, a low incidence of gene mutation, a high level of M1 macrophage infiltration, regulatory T cells, CD8 T cells, and B cells, active immunity, and larger benefits from immune checkpoint inhibitors (ICIs) therapy. The high-risk group, on the other hand, had suppressive immunity, high levels of NK and CD4 T-cell infiltration, high gene mutation rates, and decreased benefits from ICI therapy. As a result of our research, a predictive model for IRGs that can reliably predict a patient’s prognosis and their response to both conventional and immunotherapy has been created.