Abstract:Background: Medulloblastoma is the most common malignant brain tumor in children accounting for about 10% of all pediatric cancer deaths. According to the 2016 WHO Classification, four molecular subtypes of medulloblastoma including WNT, SHH, Group 3 and Group 4 were characterized by high-throughput gene expression profiling. These molecular subtypes display distinct clinical, demographic and genetic features that are associated with prognostic and therapeutic differences. Because it is currently impractical to perform microarray analysis in clinical settings, distinguishing the four molecular subgroups of medulloblastoma in the daily treatment of patients, as well in the setting of clinical trials, remains an important challenge. Methods: Three medulloblastoma microarray datasets were curated to perform an integrative analysis. To identify a reliable biomarker, a training-validating approach was adopted in this study. The gene expression profiles of 103 samples were selected as a training set for signature identification. Additional 358 samples were used for signature validation. The Gene Ontology and KEGG pathway analysis were performed to reveal the biological features of candidate genes. Results: A 23-gene expression signature derived from the training set was strongly associated with the molecular subtypes of medulloblastoma. The 23-gene expression signature was validated in 8 WNT, 61 SHH, 62 Group 3 and 227 Group 4 samples. With the 23-gene expression signature, 8 samples were classified as WNT, 63 as SHH, 73 as Group 3 and 214 as Group 4. The gene expression-based assignments reached a 95.5% overall agreement with the reference diagnoses (342 of 358; 95% CI: 0.928 to 0.974). Sensitivity ranged from 94% to 100%, while specificity ranged from 96% to 100%. The functional enrichment analysis showed that 23 genes were significantly associated with signal transduction and WNT signaling pathway. Conclusions: A 23-gene expression signature that could accurately discriminate molecular subtypes of medulloblastoma was identified in this study. Our results may prompt further development of this gene expression signature into a molecular assay amenable to routine clinical practice. Legal entity responsible for the study: Qinghua Xu. Funding: Has not received any funding. Disclosure: J. Chen, C. Chen, Y. Sun, L. Chen: Employment: Canhelp Genomics Co. Ltd. Q. Xu: Employment, stock ownership: Canhelp Genomics Co. Ltd. All other authors have declared no conflicts of interest.

Comprehensive Analysis of a Microrna Expression Profile in Pediatric Medulloblastoma.

Identification of Differentially Expressed Micrornas by Microarray: a Possible Role for Micrornas Gene in Medulloblastomas.

Prognostic Significance of MicroRNAs in Medulloblastoma: A Systematic Review and Meta-Analysis

Identification of Aberrant Microrna Expression Pattern in Pediatric Gliomas by Microarray

Examining the Biomarkers and Molecular Mechanisms of Medulloblastoma Based on Bioinformatics Analysis

Microarray Base Mirnas Profiling in Human Medulloblastomas

Integrated Analysis of Transcriptome Data Revealed AURKA and KIF20A as Critical Genes in Medulloblastoma Progression

Identification of Core Genes and Pathways in Medulloblastoma by Integrated Bioinformatics Analysis

Translational Applications of Microrna Genes in Medulloblastomas

Exosomal miR-101-3p and miR-423-5p inhibit medulloblastoma tumorigenesis through targeting FOXP4 and EZH2

miRNA-211 maintains metabolic homeostasis in medulloblastoma through its target gene long-chain acyl-CoA synthetase 4

Upregulation of miR-20a and miR-106b is involved in the acquisition of malignancy of pediatric brainstem gliomas.

PDTB-08. Mir-449a, A POTENTIAL DIAGNOSTIC BIOMARKER FOR WNT GROUP OF MEDULLOBLASTOMA

Identification Of Prognostic Genes In Paediatric Medulloblastoma From Mrna Expression Profiles

In-depth Inference of Transcriptional Regulatory Networks Identifies NPM1 as a Druggable Master Regulator in Medulloblastoma

miR-106b is overexpressed in medulloblastomas and interacts directly with PTEN.

miR-124-3p and miR-194-5p regulation of the PI3K/AKT pathway via ROR2 in medulloblastoma progression

Preliminary study on aberrant expression of miRNAs related to the formation of the distinction between pediatric and adult types of brainstem gliomas

Circ_103128 is Associated with the Tumorigenesis of Medulloblastoma.

Identification and Validation of a 23-Gene Expression Signature for Subtype Classification of Medulloblastoma.

Multi-omics analysis of intertumoral heterogeneity within medulloblastoma uncharted-pathway subtypes