Immunological network analysis in HPV associated head and neck squamous cancer and implications for disease prognosis

Human papillomavirus-positive (HPV +) head and neck squamous cell cancer (HNSCC) exhibits a better prognosis than HPV-negative (HPV -) HNSCC. This difference may in part be due to enhanced immune activation in the HPV + HNSCC tumor microenvironment. To characterize differences in immune activation between HPV + and HPV - HNSCC tumors, we identified and annotated differentially expressed genes based upon mRNA expression data from The Cancer Genome Atlas (TCGA). Immune network between immune cells and cytokines was constructed by using single sample Gene Set Enrichment Analysis and conditional mutual information. Multivariate Cox regression analysis was used to determine the prognostic value of immune microenvironment characterization. A total of 1673 differentially expressed genes were functionally annotated. We found that genes upregulated in HPV + HNSCC are enriched in immune-associated processes. And the up-regulated gene sets were validated by Gene Set Enrichment Analysis. The microenvironment of HPV + HNSCC exhibited greater numbers of infiltrating B and T cells and fewer neutrophils than HPV - HNSCC. These findings were validated by two independent datasets in the Gene Expression Omnibus (GEO) database. Further analyses of T cell subtypes revealed that cytotoxic T cell subtypes predominated in HPV + HNSCC. In addition, the ratio of M1/M2 macrophages was much higher in HPV + HNSCC. The infiltration of these immune cells was correlated with differentially expressed cytokine-associated genes. Enhanced infiltration of B cells and CD8 + T cells were identified as independent protective factors, while high neutrophil infiltration was a risk enhancing factor for HPV + HNSCC patients. A schematic model of immunological network was established for HPV + HNSCC to summarize our findings.