Integrative multi-omics analysis reveal novel therapeutic targets for glioblastoma

Background and objectives: Glioblastoma (GBM) is the most common and aggressive primary malignant brain tumor in adults, yet effective therapeutic strategies remain limited. A comprehensive exploration of brain proteomic profiles may offer new insights into GBM pathogenesis and treatment. Methods: We integrated the largest available summary-level GWAS data for GBM, derived from eight studies of European ancestry, with brain proteomic and transcriptomic data. Proteome-wide association studies (PWAS) were performed using dorsolateral prefrontal cortex (dPFC) samples from the Religious Orders Study/Memory and Aging Project (ROS/MAP) and the Banner Sun Health Research Institute. Transcriptome-wide association studies (TWAS) utilized dPFC transcriptomic data from the CommonMind Consortium (CMC). To assess causality, Mendelian randomization (MR) analyses based on protein and expression quantitative trait loci (pQTL and eQTL) were conducted. Bayesian colocalization analyses were performed to determine whether GBM risk and candidate genes shared a common causal variant. Based on the integrative evidence, candidate genes were prioritized and assigned confidence levels regarding their therapeutic relevance. Results: The primary PWAS identified eight candidate genes associated with GBM ( LIME1, EGFR, RHBDF1, SCFD1, FAIM, KDELC2, GPX1, and GMPPB ). Among these, two ( LIME1 and GMPPB ) were replicated in a confirmatory PWAS, and six ( LIME1, EGFR, SCFD1, FAIM, GPX1, and GMPPB ) were validated in TWAS. pQTL-based MR analysis supported causal relationships for SCFD1, GPX1 , and GMPPB , while eQTL-based MR analysis identified EGFR, SCFD1, FAIM , and GMPPB . Colocalization analysis revealed that EGFR, RHBDF1, SCFD1 , and GMPPB share causal variants with GBM risk. Through comprehensive multi-omics integration, three genes – EGFR, SCFD1 , and GMPPB – were identified as high-confidence therapeutic targets. Discussion: By leveraging brain proteomic and transcriptomic data alongside large-scale GBM GWAS, we identified eight candidate genes and prioritized three – EGFR, SCFD1 , and GMPPB – as high-confidence therapeutic targets. Notably, SCFD1 and GMPPB represent novel candidates. These findings advance our understanding of the molecular mechanisms underlying GBM and warrant further functional validation.