Identification of m6A methyltransferase-related genes predicts prognosis and immune infiltrates in head and neck squamous cell carcinoma

Abstract

Background Head and neck squamous cell carcinoma (HNSCC) accounts for 90% of head and neck malignant tumors. As the early symptoms of HNSCC are not obvious, and it is prone to recurrence and metastasis, making the overall survival (OS) rate of patients very low. Existing studies have shown m6A methylation plays a crucial role in various cancers, but it is rarely studied in HNSCC. This study aimed to explore the expression of m6A methylation-related genes in HNSCC and its correlation with prognosis, and to explore its relationship with immune infiltration. Methods The gene expression data of HNSCC patient tumor samples (tumor =510) and adjacent normal tissue samples (normal =50) were extracted from The Cancer Genome Atlas (TCGA) database, and the expression characteristics of m6A regulatory factors were described. Kaplan-Meier survival analysis was used to analyze the relationship between m6A regulatory factors and OS and disease-specific survival (DSS). Least absolute shrinkage and selection operator (LASSO) regression was used to construct the m6A regulatory factor-HNSCC risk prediction model. In addition, the relationship between m6A methylation-related genes and tumor immune infiltration were discussed. Results The differential expression of 20 genes were identified by TCGA, and 18 genes (IGF2BP2, IGF2BP1, IGF2BP3, VIRMA, YTHDF1, YTHDF2, YTHDF3, ZC3H13, METTL14, ALKBH5, METTL3, RBMX, WTAP, YTHDC1, FTO, HNRNPC, HNRNPA2B1, and RBM15) were overexpressed in HNSCC. The survival rate of different gene expression levels was different. The high expression of YTHDC1 and YTHDC2 indicated better OS. Furthermore, for DSS, increased expression of YTHDC2 was also correlated with better clinical outcomes (P<0.05). At the same time, we drew a 3-gene risk score model in the TCGA-HNSCC cohort, and the survival curve showed compared with low-risk patients, high-risk patients had significantly worse OS (P<0.05). Gene enrichment analysis showed EPITHELIAL_MESENCHYMAL_TRANSITIO, MTORC1_SIGNALING, MYC_TARGETS_V1, MYC_TARGETS_V2, MYOGENESIS pathways, high TP53 mutations, and suppressive immunity were related to the high-risk group. The low-risk group was related to ALLOGRAFT_REJECTION, COMPLEMENT, IL6_JAK_STAT3_SIGNALING, INTERFERON_ALPHA_RESPONSE, INTERFERON_GAMMA_RESPONSE pathways, low TP53 mutations, and active immunity. Conclusions The m6A methyltransferase-related genes can predict the prognosis of HNSCC and are related to immune infiltration.