Identification and validation of an individualized prognostic signature of lower-grade glioma based on nine immune related long non-coding RNA

Clin Neurol Neurosurg. 2021 Feb:201:106464. doi: 10.1016/j.clineuro.2020.106464. Epub 2021 Jan 5.

Abstract

Background: Low-grade glioma (LGG)is one of the most common and aggressive neurological malignant tumors of the central nervous system. Mounting evidence indicates that aberrantly expressed long non-coding RNA (lncRNAs) and immune cell infiltration influence low-grade glioma development. Despite the increasing amount of research on lncRNA, there are very few immune-related lncRNA for LGG studies.

Methods: We evaluated immune cell infiltration in 529 low-grade glioma patient specimens from TCGA and 1152 normal brain tissue samples from GTEx. ssGSEA was used to generate high, medium, and low immune cell infiltration groups and to examine the heterogeneity of the low-grade glioma immune microenvironment. A risk model of immune-related lncRNAs based on immune gene sets was developed. Sequential single-factor Cox regression, Lasso regression, and stepwise multiple Cox regression analyses uncovered immune-related lncRNAs with low-grade glioma prognostic value. Kaplan-Meier analysis, ROC analysis, and nomograms were used to predict low-grade glioma OS. At length, We performed GO term and KEGG enrichment analyses and used standardized enrichment scores (NES) to identify signaling pathways that were significantly enriched.

Results: We identified nine immune-associated lncRNAs with low-grade glioma prognostic value (AC009283.1, AC009227.1, AL121899.1, LINC00174, LINC02166, AC018647.1, AC061961.1, NRAV, and LINC00320).These prognostic lncRNAs were used to establish prognostic markers. Kaplan-Meier Survival analysis revealed a 10-year survival rate of 22.68 % (95 % CI: 13.54-38 %] in high-risk LGG vs. 54 % (95 % CI: 39.04-74.8 %] in low-risk LGG patients. Univariate Cox regression analysis showed that the HR of risk score and 95 % CI were 1.081 and (1.060-1.102) (p < 0.001), respectively. In contrast, those from multivariate Cox regression analysis were 1.066 and (1.046-1.087) (p < 0.001). This indicated that nine LncRNAs are independent prognostic factors for patients with low-grade glioma. GSEA suggests that the identified lncRNAs influence low-grade glioma tumorigenesis and prognosis by modulating immune responses and cancer pathways.

Conclusions: Our data highlight the potential prognostic value of the nine immune-related lncRNA in low-grade glioma and may open new research lines and guide low-grade glioma management.

Keywords: Immune prognostic model; Long non-coding RNA (lncRNA); Low-grade glioma (LGG); Risk score; The Cancer Genome Atlas (TCGA).

Publication types

  • Validation Study

MeSH terms

  • Biomarkers, Tumor / genetics
  • Biomarkers, Tumor / immunology*
  • Brain Neoplasms / immunology*
  • Gene Expression Profiling / methods*
  • Glioma / immunology*
  • Humans
  • Nomograms
  • Precision Medicine / methods
  • Prognosis
  • RNA, Long Noncoding / genetics
  • RNA, Long Noncoding / immunology*
  • Transcriptome / immunology

Substances

  • Biomarkers, Tumor
  • RNA, Long Noncoding