Identification of Prognostic Markers for Head and NeckSquamous Cell Carcinoma Based on Glycolysis-Related Genes

Abstract

Head and neck squamous cell carcinomas (HNSCCs) comprise a heterogeneous group of tumors. Many patients respond differently to treatment and prognosis due to molecular heterogeneity. There is an urgent need to identify novel biomarkers to predict the prognosis of patients with HNSCC. Glycolysis has an important influence on the progress of HNSCC. Therefore, we investigated the prognostic significance of glycolysis-related genes in HNSCC. Our results showed that ELF3, AURKA, and ADH7 of 20 glycolysis-related DEGs were significantly related to survival and were used to construct the risk signature. The risk score showed high accuracy in distinguishing the overall survival (OS) of HNSCC. The Kaplan–Meier curves demonstrated that the risk score was associated with an unfavorable prognosis in patients with female sex, male sex, grade 3, T1/2 stage, N+ stage, N2 stage, M0 stage, and clinical stage III/IV. Independent prognostic analysis showed that clinical stage and risk score were strongly associated with OS. Moreover, the risk score had higher accuracy in predicting 1-, 3-, and 5-year survival. AURKA and ADH7 were only significantly related to M1 macrophages and neutrophils, respectively, while ELF3 was significantly correlated with M2 macrophages and monocytes (all $p<0.05$ ).The ceRNA network demonstrated that miR-335-5p and miR-9-5p may play core roles in the regulation of these three genes in HNSCC. The risk score constructed based on three glycolysis-related genes showed high accuracy in predicting the prognosis and clinicopathological characteristics of HNSCC.