Bioinformatics molecular analysis of prognostic-related genes and drug sensitivity and resistance in head and neck squamous cell carcinomas

Abstract

One of the most heterogeneous collections of malignancies probably is head and neck cancers that in this study, we focus on the head and neck squamous cell carcinomas (HNSCCs). The aim of this study was to recognize the prognosis-related genes and their pathways related to HNSCC and to further analyze its molecular mechanism and drug sensitivity and resistance. Furthermore, the correlation between the expression of candidate genes with immune infiltration and IC50 of some drugs was evaluated. The cancer genome atlas (TCGA) HNSCC was utilized for differential expression analyzing. Differentially expressed genes (DEGs) were indicated by R software. By applying univariate and multivariate Cox regression analysis, and using prognostic risk model, we identified the key prognostic genes in HNSCC based on PPI network, and enrichment analysis of DEGs was performed by Gene Set Enrichment Analysis (GSEA) and Enrichr database. The Tumor Immune Estimation Resource (TIMER) was used to assess the relationship between selected genes and immune infiltration in cancer. Finally, the CCLE and GDSC were used to assess the correlation between the level of selected genes with drug resistance and sensitivity. Our results indicated that four hub genes including NUF2, BUB1, TPX2, and DLGAP5 were selected as poor prognostic genes in HNSCC. Our findings revealed that drugs like Glucosamine, Ribavirin and Tocetrapib could lower the expression of selected genes in HNSCC. In doing so, NUF2, BUB1, DLGAP5, TPX2 were identified as hub-related prognostic genes associated with patients’ survival, drug resistance and sensitivity, and could be effective in the cancer progression.