A Novel Ferroptosis-related Gene Signature for Prognosis Prediction in Glioma

Abstract Background Glioma is a lethal brain tumor characterized by its invasive nature, fast growth, and malignant recurrence. Despite aggressive surgical resection, concomitant concurrent radiation, and chemotherapy, the overall survival (OS) of patients with glioma remains low. Ferroptosis is a unique modality that regulates programmed cell death and is associated with multiple steps of tumorigenesis in a variety of tumors. As ferroptosis is an oxidative-stress-induced form of cell death and is related to the intracellular iron level that is intricately linked to cell metabolism, cancer cells may have a higher proclivity for it. Methods Genes relevant to ferroptosis were retrieved, and differential analysis was performed, followed by prognosis-related univariate and multivariate Cox regression analyses. A ferroptosis-involving model was also identified. Each sample’s risk score was statistically weighted by gene expression and stratified into high-risk and low-risk groups based on the cut-off value of the ROC curve. GO, KEGG, and GSVA analyses were used to detect probable biological functions and signaling pathways. All independent risk factors were included to build nomograms, and their accuracy and practicability were tested using the concordance index (C-index) and ROC curves. Immune cell infiltration was quantified using CIBERSORT software. Drug response and tumor mutation burden (TMB) were used to verify the effectiveness of the clinical use of immune-targeted therapy. Results The patient samples were stratified into two risk groups based on a four-gene signature. The high-risk group had poorer overall survival. The results of the functional analysis indicated that the extracellular matrix-related biological and interferon-gamma-mediated signaling pathway functions were enriched in the high-risk group, and multiple immune pathways were saturated in the GSVA analysis. The infiltration of immunocytes differed between the two groups. At multiple immune checkpoints, the risk score showed a strong positive correlation. The analysis of drug treatment response and TMB showed that the high-risk group had a better treatment response and better curative effect on immune target therapy. The CNV data analysis showed a correlation between the four genes in glioma and different CNV subtypes related to OS in glioma. Besides, we compared the published signatures of ferroptosis-related genes and found that our model performed well in evaluating glioma patients' prognosis in the new WHO CNS5 classification. Conclusions A novel ferroptosis-related gene signature, which can classify glioma patients into subgroups, can be used for prognostic prediction in glioma. The four-gene model in the clinical application could serve as a potential method to assess glioma patients' prognosis and response to therapy.