Introduction:
This study aimed to explore the prognostic value of immune-related long non-coding RNAs (lncRNAs) in glioblastoma (GBM).

Material and methods:
Expression and clinical data were acquired, including the GSE111260 dataset: 67 GBM and 3 normal brain samples; the GSE103227 dataset: 5 GBM and 5 normal brain samples; and TCGA data: 187 GBM samples. Immune-related genes were retrieved from the ImmPort database. Immune-related differentially expressed genes (DEGs) and lncRNAs were screened. Prognostic lncRNAs were then screened to establish a prognostic risk score model. Survival analysis and differential expression analysis were performed in high- vs. low-risk groups, followed by the protein-protein interaction network and the lncRNA-mRNA co-expression network.

Results:
A total of 251 immune-related DEGs were screened. After correlation analysis, 387 immune-related lncRNAs that co-expressed with 140 immune-related DEGs were screened. Univariate analysis identified 18 lncRNAs that were significantly associated with prognosis. The prognostic risk score was able to stratify GBM patients into high- and low-risk groups, and patients with high risk scores displayed worse outcomes than those with low risk scores in both the training set and the validation set. A total of 272 genes had abnormal expression between high- and low-risk groups. Of these, 22 genes were immune-related, such as SNAP25, SNAP91, SNCB, and RAB3A. These genes were mainly enriched in synaptic vesicle cycle/exocytosis and insulin secretion. The co-expression network contained 22 genes and 11 lncRNAs, and lncRNA LINC01574 co-expressed with the great number of mRNAs.

Conclusions:
We identified 18 immune-related prognostic lncRNAs, and the established lncRNA-based prognostic risk model could stratify GBM patients into different risk levels.