Gene Model Related to m6A Predicts the Prognostic Effect of Immune Infiltration on Head and Neck Squamous Cell Carcinoma.

Yaping Deng,Kehua Li,Fengwu Tan,Hanbo Liu,Quan Cheng

doi:10.1155/2021/1814266

Abstract

Head and neck squamous cell carcinoma (HNSCC) is a highly aggressive solid tumor. Because most studies have focused on the intrinsic carcinogenic pathways of tumors, we focused on the relationship between N6-methyladenosine (m6A) and the prognosis of HNSCC in the tumor immune microenvironment. We downloaded RNA-seq data from the TCGA dataset and used univariate Cox regression to screen m6A-related lncRNAs. The expression value of LASSO-screened genes was the sum of LASSO regression coefficients. We then evaluated relationships between the risk score and cellular components or cellular immune response. Differences in immune response under various algorithms were visualized with heat maps. The GSVA package in R was used to analyze GO, BP, KEGG, and hallmark gene sets of immune checkpoint clusters and immune checkpoint scores. The GSEA analysis was performed with the cluster profile package, yielding 21 m6A genes. Related lncRNAs were screened with Pearson's correlations, and the resulting 442 lncRNAs were screened using single-factor analysis. Eight lncRNAs closely related to prognosis were identified through survival random forest. Survival analysis showed that patients with a high risk score had a poor prognosis. Low- and high-risk-score groups differed significantly in m6A gene expression. Prognostic scores from different algorithms were significantly correlated with B cells, T cells, and memory cells in the immune microenvironment. Expression of immune checkpoints and signal pathways differed significantly across risk-score groups, suggesting that m6A could mediate lncRNA-induced immune system dysfunction and affect HNSCC development. A comprehensive study of tumor-cell immune characteristics should provide more insight into the complex immune microenvironment, thus contributing to the development of new immunotherapeutic agents.

Highlights

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common malignant tumor worldwide, claiming approximately 350000 lives every year [1]
Singlefactor results of the eight genes in our survival model were significant (P < 0.05). ese eight genes were significantly associated with HNSCC prognosis (Figure 1(c)), and they were linked to patient survival
Survival analysis of the eight genes was plotted (Figure S1). e survival model (TNM stage, grade, sex, age, status, and risk score) was related to the eight genes (Figure 1(d)). e receiver operating characteristic (ROC) curve showed that the risk score had a strong predictive ability, with an AUC of 0.65, 0.65, and 0.629 in 1, 3, and 5 years compared with age factors. e risk model may serve as an important indicator for evaluating the prognosis of HNSCC (Figure 1(e))

Summary

Introduction

Head and neck squamous cell carcinoma (HNSCC) is the sixth most common malignant tumor worldwide, claiming approximately 350000 lives every year [1]. HNSCC includes malignant tumors of the oral cavity, nasopharynx, oropharynx, hypopharynx, larynx, nasal cavity, and salivary gland [2]. Patterns of clinical behavior and treatment response to recurrent and metastatic HNSCC are heterogeneous. Ese include the effects of previous treatments on tumor cells and the infiltrative and multifocal nature of HNSCC, which are typical features of recurrent disease in this region [5]. A review has shown that HNSCC is suitable for immunotherapy. We analyzed the relationship between genetic models related to immune infiltration and biometrics-based prognosis of HNSCC

Objectives

Methods

Results

Conclusion