Identification and Validation of an Immune-Associated RNA-Binding Proteins Signature to Predict Clinical Outcomes and Therapeutic Responses in Glioma Patients

Abstract

The prognosis of patients with glioma is largely related to both the tumor-infiltrating immune cells and the expression of RNA-binding proteins (RBPs) that are able to regulate various pro-inflammatory and oncogenic mediators. However, immune-associated RBPs in glioma remain unexplored. In this study, we captured patient data from The Cancer Genome Atlas (TCGA) and divided them into two immune subtype groups according to the difference in infiltration of immune cells. After differential expression and co-expression analysis, we identified 216 RBPs defined as immune-associated RBPs. After narrowing down processes, eight RBPs were selected out to construct a risk signature that proven to be a novel and independent prognostic factor. The patients were divided into high- and low-risk groups on the basis of risk score. Higher risk scores meant worse overall survival and higher expression of human leukocyte antigen and immune checkpoints such as PD1 and CTLA4. In addition, analyses of pathway enrichment, somatic mutation, copy number variations and immuno-/chemotherapeutic response prediction were performed in high- and low-risk groups and compared with each other. For the first time, we demonstrated a novel signature composed of eight immune-associated RBPs that was valuable in predicting the survival of glioma patients and directing immunotherapy and chemotherapy.

RNA Binding Protein-Mediated Immunosuppressive Tumor Microenvironment Contributes to Poor Prognosis in High-Grade Gliomas

“…On the contrary, the Cluster B re ected the immune in amed phenotype, and the tumor present contained a higher abundance of immune cells, such as activated NK cells, monocytes, activated dendritic cells, activated mast cells, eosinophils, while stromal score and immune score were relatively signi cantly decreased compared with Cluster A. Both expression clusters were identical to the phenotypes that have been previously reported (15,23).…”

“…A number of previous studies have con rmed that RBPs play critical roles in the tumor development, progression, and tumor immunity across various human tumors (13,14). The regulation of immune microenvironment by RBPs in glioma has been previously reported (15). However, there are few studies on the validation of this regulation mode and its regulatory mechanism from the perspective of RBPs and RBP-regulated immune genes.…”

Identification and Validation of the Pyroptosis-Related Molecular Subtypes of Lung Adenocarcinoma by Bioinformatics and Machine Learning

“…Consensus Clustering, an unsupervised clustering method, is a common cancer subtype classification method in which the clustering framework incorporates results from multiple runs of an inner-loop clustering algorithm on sub-sampled subjects ( Wilkerson and Hayes, 2010 ; Seymour et al, 2019 ). The “ConsensusClusterPlus” R package was used for consensus clustering and distinguishing different molecular subtypes based on the mRNA expression data of 33 pyroptosis-related genes ( Tian et al, 2021 ). Pairwise consensus values are calculated and stored in a consensus matrix (CM) for each k ( Zheng J. et al, 2020 ).…”

Table_1.DOCX

“…Consensus Clustering, an unsupervised clustering method, is a common cancer subtype classification method in which the clustering framework incorporates results from multiple runs of an inner-loop clustering algorithm on subsampled subjects (Wilkerson and Hayes, 2010;Seymour et al, 2019). The "ConsensusClusterPlus" R package was used for consensus clustering and distinguishing different molecular subtypes based on the mRNA expression data of 33 pyroptosisrelated genes (Tian et al, 2021). Pairwise consensus values are calculated and stored in a consensus matrix (CM) for each k (Zheng J. et al, 2020).…”

“…With the development of high-throughput sequencing, the understanding of tumor gene expression profiles is more comprehensive and in-depth, which will help in classifying tumors and precise treatment. Many studies have been conducted to identify subtypes, predict prognosis or predict drug resistance based on the expression of different genes and clinical features (Chen et al, 2021;Tian et al, 2021;Zhang S. et al, 2021). In this study, we did a prior study on identifying the pyroptosis-related molecular subtypes of LUAD, based on the high-throughput sequencing data.…”