The potential of DEirlncRNAs: A novel approach to predict glioblastoma prognosis.

Background: Despite tremendous evolution in therapies, the prognosis of glioblastoma (GBM) remains grim, which calls for innovative approaches to optimize chemotherapy efficacy and predict risk. Methods: The transcriptome and clinical data of GBM were acquired from the Cancer Genome Atlas (TCGA), followed by the identification of differentially expressed immune-related long noncoding RNAs (DEirlncRNAs) with Pearson correlation and limma packet analyses. Survival-related DEirlncRNA pairs were screened with univariate Cox proportional hazard regression. Prognostic markers were obtained, and risk scores were calculated with Lasso regression and multivariate Cox risk regression analyses. The association of the prognostic risk model with immune cell infiltration was evaluated by comprehensively analyzing tumor-infiltrating immune cells with TIMER, XCELL, CIBERSORT, QUANTISEQ, and EPIC. Differences in half-maximal inhibitory concentration (IC50) values between the high- and low-risk groups were assessed with the Wilcoxon signed-rank test Results: A total of 276 DEirlncRNAs were identified, followed by the visualization of their expression patterns. Two prognosis-related DEirlncRNA pairs were screened, with high accuracy and reliability. The constructed prognostic risk model effectively distinguished between high- and low-risk patients, and significant differences were observed in survival outcomes between the high- and low-risk groups. Furthermore, risk scores were associated with tumor-infiltrating immune cells and DEirlncRNA expression. Additionally, the risk model had a correlation with the effectiveness of commonly used chemotherapeutic agents, providing clues into potential treatment responses. Conclusions: In our study, a novel signature was constructed with paired DEirlncRNAs (regardless of their expression), which holds significant clinical predictive value and is a potential breakthrough for personalized management of GBM.