Identification of an epigenetic prognostic signature for patients with lower-grade gliomas.

Introduction Glioma is the most common malignant primary brain tumor with survival outcome for patients with lower-grade gliomas (LGGs) being quite variable. Epigenetic modifications in LGGs appear tightly linked to patient clinical outcomes but are not commonly used as clinical tools. Aims We aimed to derive an epigenetic enzyme gene signature for LGGs that would allow for improved clinical risk stratification. Results The study employed transcriptomic data of 711 lower-grade gliomas from three publically available data sets. Based on least absolute shrinkage and selection operator (LASSO) Cox regression analysis, we discovered a 13-gene epigenetic signature that strongly predicts poor overall survival in LGGs. The robust prediction ability for survival was further verified in two independent validation cohorts. The signature was also significantly associated with malignant molecular signatures including wild-type IDH, unmethylated MGMT promoter, and non-codeletion of 1p19q together with linkage to multiple oncogenic pathways. Interestingly, our results showed that immune infiltration of MDSCs together with mRNA expression of immune inhibition biomarkers was also positively correlated with the epigenetic signature. Lastly, we confirmed the oncogenic role of SMYD2 in glioma tumor cells in functional assays. Conclusions We report a novel epigenetic gene signature that harbors robust survival prediction value for LGG patients that is tightly linked to activation of multiple oncogenic pathways.