Identification of Nine M6A-Related Long Noncoding RNAs as Prognostic Signatures Associated with Oxidative Stress in Oral Cancer Based on Data from The Cancer Genome Atlas.

Abstract

Objective Although the expression of long noncoding RNAs (lncRNAs) and N6-methyladenosine (M6A) is correlated with different tumors, it remains unclear how M6A-related lncRNA functioning affects the initiation and progression of oral squamous cell carcinoma (OSCC). Materials and Methods Gene expression and clinical data were retrieved from The Cancer Genome Atlas. The prognostic value of M6A-related lncRNAs was determined using univariate Cox regression analyses. Gene set enrichment analysis of OSCC patient clusters revealed the pathways that elucidate the mechanism. Furthermore, a risk-based model was established. The difference in the overall survival (OS) between groups, including low-/high-risk groups, was determined by Kaplan-Meier analysis. Relationships among immune cells, clusters, clinicopathological characteristics, and risk scores were explored. Results Among 1,080 M6A-related lncRNAs, 36 were prognosis-related. Patients with OSCC were divided into two clusters. T stage and the pathological grade were noticeably lower in cluster 2 than in cluster 1. Epithelial-mesenchymal transition showed greater enrichment in cluster 1. Nine hub M6A-related lncRNAs were identified for the model of risk score for predicting OSCC prognosis. The OS was longer in patients with a low-risk score than in patients with a high-risk score. The risk score was an independent prognostic factor of OSCC and was associated with the infiltration of different immune cells. T stages and the American Joint Committee on Cancer (AJCC) stages were more advanced in the high-risk score group than in the low-risk score group. Finally, expression correlation analysis showed that the risk score is associated with the expression of oxidative stress markers. Conclusion M6A-related lncRNAs play an important role in OSCC progression. Immune cell infiltration was related to the risk score model in OSCC and could accurately predict OSCC prognosis.