Identification of Necroptosis-related Molecular Subtypes and Construction of Necroptosis-related Gene Signature for Glioblastoma Multiforme.

Abstract

Necroptosis is a highly regulated and genetically controlled process, and therefore, attention has been paid to the exact effects of this disorder on a variety of diseases, including cancer. An in-depth understanding of the key regulatory factors and molecular events that trigger necroptosis can not only identify patients at risk of cancer development but can also help to develop new treatment strategies. This study aimed to increase understanding of the complex role of necroptosis in glioblastoma multiforme (GBM) and provide a new perspective and reference for accurate prediction of clinical outcomes and gene-targeted therapy in patients with GBM. The objective of this study was to analyze the gene expression profile of necroptosis regulatory factors in glioblastoma multiforme (GBM) and establish a necroptosis regulatory factor-based GBM classification and prognostic gene signature to recognize the multifaceted impact of necroptosis on GBM. The necroptosis score of the glioblastoma multiforme (GBM) sample in TCGA was calculated by ssGSEA, and the correlation between each gene and the necroptosis score was calculated. Based on necroptosis score-related genes, unsupervised consensus clustering was employed to classify patients. The prognosis, tumor microenvironment (TME), genomic changes, biological signal pathways and gene expression differences among clusters were analyzed. The gene signature of GBM was constructed by Cox and LASSO regression analysis of differentially expressed genes (DEGs). Based on 34 necroptosis score-related genes, GBM was divided into two clusters with different overall survival (OS) and TME. A necroptosis-related gene signature (NRGS) containing 8 genes was developed, which could stratify the risk of GBM in both the training set and verification set and had good prognostic value. NRGS and age were both independent prognostic indicators of GBM, and a nomogram developed by the integration of both of them showed a better predictive effect than traditional clinical features. In this study, patients from public data sets were divided into two clusters and the unique TME and molecular characteristics of each cluster were described. Furthermore, an NRGS was constructed to effectively and independently predict the survival outcome of GBM, which provides some insights for the implementation of personalized precision medicine in clinical practice.