Abstract
BACKGROUND: Glioma represents the most prevalent and aggressive primary brain tumor in humans. Tumor heterogeneity, the immunosuppressive tumor microenvironment, and therapeutic resistance contribute to the inevitable recurrence of gliomas, posing significant clinical challenges. Understanding the risk factors and molecular mechanisms underlying glioma recurrence and progression is critical for improving patient outcomes. In this study, we aimed to develop a recurrence-associated gene signature to predict clinical recurrence and survival outcomes while elucidating potential molecular mechanisms driving glioma recurrence. METHODS: Gene expression profiles and clinicopathological data were obtained from the Chinese Glioma Genome Atlas (CGGA) database. The CGGA-693 cohort served as the training set, while the CGGA-325 cohort and TCGA database were used for validation. A prognostic model was constructed using LASSO regression analysis. Cox regression and Kaplan-Meier survival analyses were employed to assess prognostic significance. Functional enrichment analyses, including Gene Ontology (GO), Gene Set Variation Analysis (GSVA), and Pearson correlation analysis, were conducted to explore biological pathways. We applied the T-test to analyze the expression levels of apoptotic molecules in primary versus recurrent gliomas, low-grade versus high-grade gliomas, as well as in the high versus low recurrence score groups. Furthermore, correlation analysis was performed to elucidate the relationship between six classic apoptotic genes and the recurrence score. By utilizing the STRING protein-interaction network, we systematically investigated the correlations between these six classic apoptotic genes and the 9-gene signature. RNA expression levels of CASP8 and FADD across various tissues were obtained from the NCBI database and the Human Protein Atlas database. Additionally, the protein levels of CASP8 and FADD in normal brain tissues were retrieved from the Human Protein Atlas database. Statistical analyses and visualization were performed using R software. RESULTS: A 9-gene recurrence-associated signature (AC062021.1, CCT7P2, CTB-1I21.1, DGCR6, RP11- 158M2.5, SLC22A6, SLC25A48, ADAM12, and FAM225B) was established, demonstrating robust predictive performance. Multivariate analysis confirmed that the recurrence score serves as an independent prognostic factor for glioma patients. Functional annotation revealed a significant association between the signature and apoptotic pathways. Subsequent analysis indicated that extrinsic apoptosis-related molecules (FADD and CASP8), rather than intrinsic apoptotic molecules (BCL2 and CASP9), were strongly correlated with glioma recurrence. Additionally, we characterized the expression patterns of key extrinsic apoptotic mediators, FADD and CASP8, in both normal and tumor tissues. CONCLUSIONS: Our study successfully developed a predictive model based on 9 recurrence-related genes, enabling accurate stratification of glioma patients into high- and low-risk recurrence groups. Furthermore, we identified apoptosis, particularly the extrinsic apoptotic pathway involving FADD and CASP8, as a critical mechanism associated with glioma recurrence. These findings provide valuable insights into the molecular basis of glioma recurrence and may facilitate the development of targeted therapeutic strategies.