ABSTRACT Background NSUN5 is a conserved RNA methyltransferase whose oncogenic role has been demonstrated in various cancers. However, its function and prognostic value in gliomas remain unclear. Methods In this study, we systematically analyzed the expression and functional associations of NSUN5 in glioma using data from The Cancer Genome Atlas (TCGA) and the Chinese Glioma Genome Atlas (CGGA) databases. A total of 117 machine learning algorithm combinations were employed to construct and validate a prognostic model for glioma patients. In addition, in vitro experiments were performed to further validate the expression and biological functions of NSUN5. Results NSUN5 expression is significantly upregulated in glioma and is positively associated with tumor malignancy and poor prognosis. Immune infiltration analysis revealed a marked increase in M2 macrophages in the NSUN5 high‐expression group, and NSUN5 levels were positively correlated with the expression of multiple inhibitory immune checkpoints. In addition, drug sensitivity analysis and molecular docking suggested that NSUN5 may influence the response to Olaparib. Finally, based on NSUN5‐associated genes, we constructed 117 machine learning models and identified the optimal prognostic model, STRICOM, which demonstrated robust predictive performance for patient survival. Conclusion High NSUN5 expression is closely associated with poor prognosis in glioma patients, highlighting its potential as a prognostic biomarker and therapeutic target.
Wenhao et al. (2026) studied this question.