ASL expression predicts shorter overall survival in glioblastoma, linked to MIF-associated myeloid programs

Glioblastoma (GBM) is an aggressive primary brain tumor characterized by profound intratumoral heterogeneity, immune evasion, and therapeutic resistance, leading to poor prognosis. Current standard-of-care treatments often fall short due to the tumor's ability to suppress the immune system and develop resistance. Arginine metabolism plays a crucial role in GBM progression and immune regulation, influencing the tumor microenvironment. However, specific prognostic biomarkers within this pathway, particularly those with defined cellular contexts, remain insufficiently characterized, representing a critical gap in understanding and targeting GBM aggressiveness.

Researchers conducted an integrative analysis combining bulk and single-cell RNA sequencing (scRNA-seq) data to identify arginine metabolism-related prognostic genes in glioblastoma. They analyzed 18 arginine metabolism-related genes in the TCGA GBM cohort using LASSO Cox regression with tenfold cross-validation, followed by univariate and multivariate Cox models to pinpoint prognostic indicators. Prognostic findings were validated in independent CGGA and GEO datasets. Predictive performance was assessed via ROC and decision curve analysis (DCA). For single-cell resolution, scRNA-seq data from five GBM samples were processed using Seurat, quantifying arginine-metabolism activity in 38,263 high-quality cells. Myeloid cells were subclustered, and state dynamics were examined using pseudotime, RNA velocity, and CellChat analyses.

The study prioritized ASL, FAH, and NAGS as key prognostic candidates, with ASL emerging as the most significant indicator. High expression of ASL was explicitly associated with high-risk molecular subtypes of glioblastoma, including IDH-wildtype and 1p/19q non-codeletion tumors. In the TCGA cohort, high ASL expression predicted shorter overall survival, a finding consistently validated across CGGA and GEO datasets. ASL demonstrated superior clinical prognosis discrimination and net benefit in DCA compared to other candidates. Single-cell RNA sequencing revealed that ASL expression and overall arginine-metabolism activity were significantly enriched in monocyte/macrophage populations within the GBM tumor microenvironment. Trajectory analyses placed MKI67⁺ monocytes upstream, showing a clear trend of ASL expression decreasing along pseudotime, supported by RNA velocity for directional transitions.

CellChat analysis highlighted strong macrophage migration inhibitory factor (MIF)-, MHC-II-, and SPP1-associated signaling within myeloid networks, consistent with Spearman correlation patterns between ASL and MIF-axis genes.