Multi-omics and functional dissection of IGFBP-3 as a prognostic and immune driver in Glioma

Glioma remains one of the most lethal brain tumors, demanding the identification of novel biomarkers for improved prognosis and therapy. This study reveals the critical role of IGFBP-3 in glioma by integrating multi-omics analysis and experimental validation. We initially performed systematic bioinformatics analyses to examine its expression patterns, prognostic value, immune infiltration characteristics, and clinical correlations. Subsequent validation was conducted using clinical specimens and serum samples to investigate IGFBP-3's dual potential as both a diagnostic biomarker and therapeutic target. In vitro functional assays were then performed using U87MG cells to determine the impact of IGFBP-3 on critical cellular processes including proliferation, apoptosis, migration, and invasion, while simultaneously exploring the underlying molecular mechanisms. Furthermore, we implemented advanced computational algorithms to identify and quantify IGFBP-3-associated radiological and histopathological features derived from medical imaging and tissue samples. Finally, we developed and rigorously validated an integrated radiopathological model capable of predicting IGFBP-3 expression levels in glioma patients.Models included LR, RF, SVM, MLP, and XGBoost, with XGBoost selected for best performance. Our analyses revealed a significant inverse correlation between IGFBP-3 and IGFALS expression levels. Functional studies demonstrated that IGFBP-3 knockdown suppressed the invasive and migratory capacities of U87MG cells, whereas its overexpression promoted migration/invasion, with no significant proliferation effect, concomitant with inhibition of the PI3K/Akt/mTOR signaling pathway. Furthermore, we developed and validated a robust risk prediction model (RPM) based on IGFBP-3-associated genes, which exhibited consistent prognostic accuracy and predictive utility for drug sensitivity across multiple independent cohorts. IGFBP-3 serves as an independent prognostic biomarker for gliomas. Its role in regulating invasion and migration through the PI3K/Akt/mTOR signaling pathway and also affects the immune microenvironment of glioma.The RPM model can be involved in the individualized decision-making for glioma.