Identification of gene expression and DNA methylation of and as novel prognostic markers in lower-grade gliomas.

BACKGROUND

Lower-grade gliomas (LGGs) is characteristic with great difference in prognosis. Due to limited prognostic biomarkers, it is urgent to identify more molecular markers to provide a more objective and accurate tumor classification system for LGGs.

METHODS

In the current study, we performed an integrated analysis of gene expression data and genome-wide methylation data to determine novel prognostic genes and methylation sites in LGGs.

RESULTS

To determine genes that differentially expressed between 44 short-term survivors (<2 years) and 48 long-term survivors (≥2 years), we searched LGGs TCGA RNA-seq dataset and identified 106 differentially expressed genes. and were selected for further study. Kaplan-Meier plots showed that and expression were significantly correlated with overall survival (OS) and relapse-free survival (RFS) in TCGA LGGs patients. We next validated the correlation between the candidate genes expression and clinical outcome in CGGA LGGs patients. Multivariate analysis showed that mRNA expression had a significant prognostic value independent of other variables (HR = 4.825, 95% CI = 1.370-17.000,  = 0.014). Then, differential methylation sites were identified from differentially candidate gene expression groups, and all four methylation sites were significantly negatively correlated with gene expression (spearman  <  - 0.5, < 0.0001). Moreover, hyper-methylation of four methylation sites indicated better OS ( < 0.05), and three of them also shown statistical significantly association with better RFS, except for cg15509705 ( = 0.0762).

CONCLUSION

Taken together, these findings indicated that the gene expression and methylation of and may serve as prognostic predictors in LGGs and may help to precise the current histology-based tumors classification system and to provide better stratification for future clinical trials.