Identification of prognostic biomarkers related to retinoic acid metabolism in gliomas and analysis of their impact on the immune microenvironment.

Glioma is a primary tumor of the central nervous system. Numerous investigations have demonstrated that retinoic acid (RA) signaling plays an important role in glioblastoma. This research aimed to develop a RA metabolism-related gene signature associated with glioma. The RA metabolism-related differentially expressed genes were obtained through differential analysis of RA metabolism-related genes in GSE4290. The univariate Cox and least absolute shrinkage and selection operator regression analysis were adopted to build a RA metabolism-related glioma prognostic signature. We further conducted immune feature estimation and functional enrichment analysis between 2 risk subgroups. Finally, the potential drug-targeting prognostic genes were predicted through the DrugBank database. A sum of 10 RA metabolism-related differentially expressed genes between normal and tumor groups were identified. Then, a RA metabolism-related prognostic signature was built based on the 7 prognostic genes (ADH4, DHRS3, DHRS9, LRAT, RDH10, RDH12, and RDH5). Glioma patients were separated into 2 risk subgroups (low-risk vs high-risk) based on the median value of the risk score. We found that monocytes were negatively correlated with DHRS9, while activated naive CD4+T cell was positively correlated with RDH10. These prognostic genes participated in some immune-related processes, such as "B cell-mediated immunity." Finally, 4 drugs targeting DHRS3, LRAT, and RDH12 were predicted, including vitamin A, nicotinamide adenine dinucleotide, ethanol, and cyclohexylformamide. The prognostic signature comprised of ADH4, DHRS3, DHRS9, LRAT, RDH10, RDH12, and RDH5 based on RA metabolism was established, which provided a theoretical basis and reference value for the research of glioma.