Interferon Regulatory Factor Family Genes: At the Crossroads between Immunity and Head and Neck Squamous Carcinoma

Abstract

Objective. This study is aimed at investigating the regulating mechanisms of the interferon regulatory factor (IRF) family genes in head and neck squamous cell carcinoma. Methods. Based on the HNSC data in the ‘The Cancer Genome Atlas (TCGA)’ database, the expression pattern of IRF family genes was investigated. The association of IRFs family genes and survival outcomes were analyzed by Kaplan–Meier plotter web portal. The relation of IRF genes and tumor stages was evaluated by using stage plots and based on GEPIA portal. 50 genes interacting with IRFs were identified using the NetworkAnalyst’s protein-protein interaction (PPI) network construction tool. The top 200 correlated genes with similar expression patterns in HNSC were obtained by the similar gene detection module of GEPIA. Furthermore, functional enrichment analysis was performed to determine the biological functions enriched by the interacting and correlated genes. The potential implication of IRFs in tumor immunity was investigated in terms of tumor-infiltrating immune cells, a pair of immune checkpoint genes (CD274 and PDCD1), and ESTIMATE-Stromal-Immune score. Results. The unpaired sample analysis shows that all of the IRF family genes were highly expressed in HNSC tumor samples compared to control samples. The survival analysis results showed that the overexpression of IRF1, IRF4, IRF5, IRF6, IRF8, and IRF9 was associated with better overall survival in HNSC, while the other IRFs genes (IRF2, IRF3. and IRF7) did not show prognostic values for overall survival outcome of HNSC. Four genes (STAT1, STAT2, FOXP3, and SPI1) were overlapping among 50 interacted genes in the PPI network and top 200 correlated genes identified by GEPIA. The 50 interacting genes in the PPI network and top 200 correlated genes were integrated into 246 genes. These 246 genes were found to be overrepresented in multiple KEGG pathways, e.g., Th17 cell differentiation, T cell receptor signaling pathway, cytokine-cytokine receptor interaction, natural killer (NK) cell-mediated cytotoxicity, FOXO signaling, PI3K-Akt signaling, and ErbB signaling. Most correlations between IRF gene members and TIICs were positive. The strongest positive correlation was identified between IRF8 and T cells ( $r=0.849$ , $p<0.001$ ). The majority of correlation between IRF family genes and ESTIMATE-Stromal-Immune score was found to be positive. The highest positive correlation was found to be between IRF8 and Immune score ( $r=0.874$ , $p=1.09E-158$ ). Most correlations between IRFs and two immunoinhibitor genes (CD274 and PDCD1) were positive. IRF1 and PDCD1 were found to show the highest positive correlation ( $r=0.764$ , $p<2.2e-16$ ). Conclusions. The current analysis showed IRFs were differentially expressed in HNSC, indicated significant prognostic values, were involved in tumor immunity-related signaling pathways, and significantly regulated tumor-infiltrating immune cells. IRF family genes could be potential therapeutic biomarkers in targeting tumor immunity of head and neck cancer.