Construction and validation of a novel redox-related immune signature for prognostic and immunotherapeutic prediction in low-grade glioma

Abstract

Abstract Objectives Low-grade glioma (LGG) is a common subtype of brain tumors, and limitations in therapeutic strategies lead to unfavorable survival outcomes. This study was designed to discover a novel biomarker based on LGG-related redox hallmarks. Methods Redox-related genes were retrieved from the Molecular Signatures Database. The transcriptional and clinical profiles of LGG patients were achieved from The Cancer Genome Atlas (TCGA) and the Chinese Glioma Genome Atlas (CGGA). Differentially expressed and prognostic genes were identified and included in feature selection by least absolute shrinkage and selection operator regression. A multivariate Cox model was executed to develop a novel redox immune signature (RIS) to predict patient prognosis and immunotherapeutic response. Patients were classified into low- or high-RIS groups at the median cutoff. The differences in immune infiltration levels and immune checkpoint expression between the two RIS subgroups were also investigated. The ANXA1 protein was detected by immunohistochemistry in LGG tissues. Results The RIS was calculated, according to the expression levels of ANXA1, FZD6, PPIC, SLC2A10, VASN, and VEPH1. Patients with higher RIS had significantly worse prognoses than those with lower RIS in both the TCGA and CGGA cohorts. A high RIS was positively associated with immunocyte infiltration, such as CD4+/CD8+T cells, macrophages, and myeloid dendritic cells. Additionally, the high-RIS subgroup had a greater tumor mutation burden and a lower “tumor immune dysfunction and exclusion” score. 56 % (14/25) cases of LGG patients showed high expression of ANXA1 protein. Conclusions The RIS is an encouraging biomarker for determining LGG prognosis, immune characteristics, and immunotherapy response.