N6-methyladenosine (m6A) RNA methylation, the most common form of mRNA modification and regulated by the m6A RNA methylation regulators ("writers," "erasers," and "readers"), has been reported to be associated with the progression of the malignant tumor. However, its role in glioblastoma (GBM) has been poorly known. This study aimed to identify the expression, potential functions, and prognostic values of m6A RNA methylation regulators in GBM. Here, we revealed that the 13 central m6A RNA methylation regulators were firmly related to the clinical and molecular phenotype of GBM. Taking advantage of consensus cluster analysis, we obtained two categories of GBM samples and found malignancy-related processes of m6A methylation regulators and compounds that specifically targeted the malignant processes. Besides, we also obtained a list of genes with poor prognosis in GBM. Finally, we derived a risk-gene signature with three selected m6A RNA methylation regulators, which allowed us to extend the in-depth study and dichotomized the OS of patients with GBM into high-and low-risk subgroups. Notably, this risk-gene signature could be used as independent prognostic markers and accurate clinicopathological parameter predictors. In conclusion, m6A RNA methylation regulators are a type of vital participant in the malignant progression of GBM, with a critical potential in the prognostic stratification and treatment strategies of GBM.

The progression of most human cancers mainly involves the gradual accumulation of the loss of differentiated phenotypes and the sequential acquisition of progenitor and stem cell-like features. Glioblastoma multiforme (GBM) stem cells (GSCs), characterized by self-renewal and therapeutic resistance, play vital roles in GBM. However, a comprehensive understanding of GBM stemness remains elusive. Two stemness indices, mRNAsi and EREG-mRNAsi, were employed to comprehensively analyze GBM stemness. We observed that mRNAsi was significantly related to multi-omics parameters (such as mutant status, sample type, transcriptomics, and molecular subtype). Moreover, potential mechanisms and candidate compounds targeting the GBM stemness signature were illuminated. By combining weighted gene co-expression network analysis with differential analysis, we obtained 18 stemness-related genes, 10 of which were significantly related to survival. Moreover, we obtained a prediction model from both two independent cancer databases that was not only an independent clinical outcome predictor but could also accurately predict the clinical parameters of GBM. Survival analysis and experimental data confirmed that the five hub genes (CHI3L2, FSTL3, RPA3, RRM2, and YTHDF2) could be used as markers for poor prognosis of GBM. Mechanistically, the effect of inhibiting the proliferation of GSCs was attributed to the reduction of the ratio of CD133 and the suppression of the invasiveness of GSCs. The results based on an in vivo xenograft model are consistent with the finding that knockdown of the hub gene inhibits the growth of GSCs in vitro. Our approach could be applied to facilitate the development of objective diagnostic and targeted treatment tools

BackgroundAlthough mRNA vaccines have been efficient for combating a variety of tumors, their effectiveness against glioma remains unclear. There is growing evidence that immunophenotyping can reflect the comprehensive immune status and microenvironment of the tumor, which correlates closely with treatment response and vaccination potency. The purpose of this research was to screen for effective antigens in glioma that could be used for developing mRNA vaccines and to further differentiate the immune subtypes of glioma to create an selection criteria for suitable patients for vaccination.MethodsGene expression profiles and clinical data of 698 glioma samples were extracted from The Cancer Genome Atlas, and RNA_seq data of 1018 glioma samples was gathered from Chinese Glioma Genome Atlas. Gene Expression Profiling Interactive Analysis was used to determine differential expression genes and prognostic markers, cBioPortal software was used to verify gene alterations, and Tumor Immune Estimation Resource was used to investigate the relationships among genes and immune infiltrating cells. Consistency clustering was applied for consistent matrix construction and data aggregation, Gene oncology enrichment was performed for functional annotation, and a graph learning-based dimensionality reduction method was applied to describe the subtypes of immunity.ResultsFour overexpressed and mutated tumor antigens associated with poor prognosis and infiltration of antigen presenting cells were identified in glioma, including TP53, IDH1, C3, and TCF12. Besides, four immune subtypes of glioma (IS1-IS4) and 10 immune gene modules were identified consistently in the TCGA data. The immune subtypes had diverse molecular, cellular, and clinical features. IS1 and IS4 displayed an immune-activating phenotype and were associated with worse survival than the other two subtypes, while IS2 and IS3 had lower levels of tumor immune infiltration. Immunogenic cell death regulators and immune checkpoints were also diversely expressed in the four immune subtypes.ConclusionTP53, IDH1, C3, and TCF12 are effective antigens for the development of anti-glioma mRNA vaccines. We found four stable and repeatable immune subtypes of human glioma, the classification of the immune subtypes of glioma may play a crucial role in the predicting mRNA vaccine outcome.