Abstract
Medulloblastoma represents the most common malignant brain tumor in children, characterized by distinct molecular variants that suggest varied prognosis and treatment responses. Genome-wide methylation profiling could be used to characterize these aforementioned variants. Infinium EPIC [850k] analyzes 850 000 sites within the genome of the medulloblastoma to establish the unique pattern of DNA methylation at CPG sites. This study conducted a genome-wide methylation analysis on six pediatric medulloblastoma samples using Illumina Human Methylation 850 K BeadChip arrays. We integrated methylation data with copy number variation (CNVs) profiles to assign molecular subgroups (WNT, SHH, Group 3, and Group 4) to our medulloblastoma cases. Three cases were successfully classified using the EPIC arrays, while the rest of the cases were partially assigned. Group 3-4 tumors were the most common molecular subgroup identified in two cases with a methylation score of >0.9 and in one case with a score of 0.62. An SHH molecular group was also represented. Furthermore, CNV profiles were provided for all cases. Patients classified with Group 3-4 tumors were shortly deceased after diagnosis, while all the remaining patients were alive and well after 48-108 months of follow-up. Group 3-4 tumors were the most common group in this limited series of our patients with medulloblastoma and contributed to mortality in three patients. This study highlights the use of genome-wide methylation analysis and CNVs in categorizing pediatric medulloblastoma in the country and presents a new tool in diagnostic neuro-oncology, prognostic classification, and personalized treatment strategies.