Bioinformatic Analyses Identify a Prognostic Autophagy-Related Long Non-coding RNA Signature Associated With Immune Microenvironment in Diffuse Gliomas.

Kui Yang,Shengchao Xu,Zhixiong Liu,Lu Tang,Quan Cheng

doi:10.3389/fcell.2021.694633

Abstract

BackgroundAutophagy and long non-coding RNA (lncRNA) play a critical role in tumor progression and microenvironment. However, the role of autophagy-related lncRNAs (ARLs) in glioma microenvironment remains unclear.MethodsA total of 988 diffuse glioma samples were extracted from TCGA and CGGA databases. Consensus clustering was applied to reveal different subgroups of diffuse gliomas. Kaplan-Meier analysis was used to evaluate survival differences between groups. The infiltration of immune cells was estimated by ssGSEA, TIMER, and CIBERSORT algorithms. The construction of ARL signature was conducted using principal component analysis.ResultsConsensus clustering revealed two clusters of diffuse gliomas, in which cluster 1 was associated with poor prognosis and enriched with malignant subtypes of gliomas. Moreover, cluster 1 exhibited high apoptotic and immune characteristics, and it had a low purity and high infiltration of several immune cells. The constructed ARL signature showed a promising accuracy in predicting the prognosis of glioma patients. ARL score was significantly elevated in the malignant subtype of glioma and the high ARL score indicated a poor prognosis. Besides, the high ARL score notably indicated low tumor purity and high infiltration of macrophages and neutrophils.ConclusionOur study developed and validated a novel ARL signature for the classification of diffuse glioma, which was closely associated with glioma immune microenvironment and could serve as a promising prognostic biomarker for glioma patients.

Highlights

Gliomas are the most common type of primary cancer in the central nervous system (Ostrom et al, 2019)
Single-sample Gene Set Enrichment Analysis, Tumor Immune Estimation Resource (TIMER), and CIBERSORT algorithms were adopted to evaluate the abundance of infiltrated immune cells in each sample (Barbie et al, 2009; Newman et al, 2015; Li et al, 2020)
With the application of Least Absolute Shrinkage and Selection Operator (LASSO) analysis, 38 autophagy-related lncRNAs (ARLs) were selected for the analysis

Summary

Introduction

Gliomas are the most common type of primary cancer in the central nervous system (Ostrom et al, 2019). Diffuse glioma is termed as a variety of brain cancers including WHO grade II to grade IV tumors. Several biomarkers are found to be associated with the prognosis of glioma patients. 1p19q codeletion is considered as a favorable prognostic factor for glioma patients (Kaloshi et al, 2007; Eckel-Passow et al, 2015). Age is considered as a prognostic factor for glioma patients, and the younger age indicates a favorable prognosis (Li et al, 2009; Lin et al, 2020). Current management for gliomas has achieved great progress, the prognosis of glioma patients remains unfavorable because of tumor recurrence and chemoresistance. The role of autophagy-related lncRNAs (ARLs) in glioma microenvironment remains unclear

Methods

Results

Discussion

Conclusion