Abstract
Medulloblastoma, a common pediatric malignant tumor, has been recognized to have four molecular subgroups [wingless (WNT), sonic hedgehog (SHH), group 3, group 4], which are defined by the characteristic gene transcriptomic and DNA methylomic profiles, and has distinct clinical features within each subgroup. The tumor immune microenvironment is integral in tumor initiation and progression and might be associated with therapeutic responses. However, to date, the immune infiltrative landscape of medulloblastoma has not yet been elucidated. Thus, we proposed MethylCIBERSORT to estimate the degree of immune cell infiltration and weighted correlation network analysis (WGCNA) to find modules of highly correlated genes. Synthesizing the hub genes in the protein-protein interaction (PPI) network and modules of the co-expression network, we identify three candidate biomarkers [GRB2-associated-binding protein 1 (GAB1), Abelson 1 (ABL1), and CXC motif chemokine receptor type 4 (CXCR4)] via the molecular profiles of medulloblastoma. Given this, we investigated the correlation between these three immune hub genes and immune checkpoint blockade response and the potential of drug prediction further. In addition, this study demonstrated a higher presence of endothelial cells and infiltrating immune cells in Group 3 tumor bulk. The above results will be conducive to better comprehending the immune-related pathogenesis and treatment of medulloblastoma.