Construction of a Prognostic Model Using RNA Processing Factor Genes and the Key Role of NSUN6 in Glioma Outcomes

Abstract

Glioma is the most common malignant brain tumor and remains associated with a poor prognosis and limited predictive tools. The dysregulation of RNA processing factor genes has been implicated in glioma development, yet their prognostic relevance remains unclear. This study aimed to construct a robust prognostic model based on RNA processing factor genes and explore their functional roles and therapeutic potential. Transcriptomic and clinical data from glioma patients in the TCGA, CGGA, GEO and Rembrandt cohorts were analysed. Univariate, multivariate and LASSO-Cox regression analyses were performed to establish a prognostic signature. Model performance was assessed using Kaplan–Meier survival curves, time-dependent ROC analysis and C-index evaluation. Key genes were identified via random forest analysis and validated through single-cell datasets and immunohistochemistry. Functional assays were conducted to examine the biological roles of the key gene. Seventy-eight RNA processing factor genes were associated with glioma prognosis, and a 19-gene risk signature was constructed. The model effectively stratified patients into high- and low-risk groups with significantly different survival outcomes (log-rank p < 0.001). The AUCs for 1-, 3- and 5-year survival were 0.812, 0.774 and 0.769 in TCGA and 0.796, 0.758 and 0.741 in CGGA. The model achieved a C-index of 0.781 and was validated as an independent prognostic factor. NSUN6 was identified as a key protective gene whose overexpression inhibited glioma cell proliferation and migration in vitro. RNA processing factor genes have prognostic utility in glioma. The 19-gene model and NSUN6 highlight novel avenues for molecular stratification and targeted therapy.