Positive Expression of IGFBP2 as a Novel Prognostic Biomarker for Patients with Lower-Grade Gliomas

Abstract

Background: Previous studies have identified insulin-like growth factor-binding protein 2 (IGFBP2) as a target gene associated with the prognosis of various malignant cancers. This study aimed to explore the role and mechanisms of this prognostic signature in patients with low-grade gliomas (LGGs).Method: A total of 217 patients with LGGs were retrospectively obtained from the Chinese Glioma Genome Atlas as the training group, whereas an additional 190 cases (GSE107850) were collected from the Gene Expression Omnibus as validation data. The Kaplan–Meier method evaluated the overall survival (OS) between the high IGFBP2 and low IGFBP2 expression groups. Univariate and multivariate Cox analyses were used to identify independent prognostic factors associated with survival. Gene set enrichment analysis (GSEA) was conducted to investigate signaling pathways influencing glioma cell proliferation at the transcriptional level of IGFBP2. Statistical analyses and data visualization were performed using R language (version 3.6.3) and Perl software (version 5.38.1), with significance set at P < 0.05.Results: Kaplan–Meier survival analysis suggested that the group with decreased IGFBP2 expression may have improved OS as compared with the group with high IGFBP2 expression. Increased IGFBP2 expression in gliomas significantly correlated with isocitrate dehydrogenase mutation-wild type. GSEA results revealed that five differential pathways involved in collagen binding, collagen-containing extracellular matrix, collagen metabolic process, collagen trimer, and extracellular structure organization were significantly enriched in patients with glioma with high IGFBP2 expression.Conclusion: Our study is the first to show that overexpression IGFBP2 could be an independent glioma biomarker. For patients with LGG overexpressing IGFBP2, radiotherapy may be a preferable choice over chemotherapy.

Keywords

• igfbp2
• glioma
• prognosis
• bioinformatics
• genomics