DNA Methylation-Driven Genes for Developing Survival Nomogram for Low-Grade Glioma

Area of Science:

• Oncology
• Genetics
• Bioinformatics

Background:

• Low-grade gliomas (LGG) exhibit heterogeneity, posing challenges for current prognostic models.
• Existing predictive tools for LGG are often unsatisfactory or lack user-friendliness.

Purpose of the Study:

• To develop a user-friendly nomogram for predicting LGG prognosis.
• To integrate methylation-driven genes with clinicopathological parameters for enhanced predictive accuracy.

Main Methods:

• Analysis of RNA and methylation sequencing data from 516 LGG patients via The Cancer Genome Atlas (TCGA).
• Utilized differential expression, methylation correlation, and survival analyses.
• Employed LASSO regression for selecting key prognostic genes.
• Validated the nomogram internally (training/testing sets) and externally (Chinese Glioma Genome Atlas database).

Main Results:

• Identified three DNA methylation-driven genes (ARL9, CMYA5, STEAP3) as independent prognostic factors.
• The final nomogram, incorporating these genes with IDH1 mutation status, age, and sex, achieved a high AUC of 0.930.
• Demonstrated stable and consistent predictive performance across internal and external validation datasets.

Conclusions:

• The developed prognostic nomogram accurately predicts individual survival probabilities for LGG patients.
• This tool offers a user-friendly approach for prognostic evaluation, treatment optimization, and patient management in LGG.