BIOM-25. GRAMMA-GLUTAMYL-HYDROLASE (GGH) AS A TARGETABLE DRIVER IN GLIOBLASTOMA: A COMPREHENSIVE OMICS-DRIVEN BIOINFORMATICS-TO-BENCH PIPELINE INTEGRATING CRISPR, ANIMAL MODELS, AND ADVANCED SPATIAL/MULTIPLEX/3D MOLECULAR IMAGING

Abstract BACKGROUND GBM is lethal, with a median survival of 1 year. Standard therapy includes surgery, radiotherapy, and TMZ, but resistance is common. Novel targets are crucial; our lab has identified a promising biomarker, GGH, for further investigation. METHODS We analyzed TCGA (n=690), CGGA (n=1012), GLASS (n=176), and a single-cell dataset (n=7930 cells from 28 patients) to identify prognostic GBM markers. We integrated the updated WHO classification, survival factors, multi-variate COX, Kaplan-Meier, and sophisticated bioinformatics analyses (Differential analyses, GSEA, CIBERSORT) to pinpoint GGH. Functional validation included siRNA in cell lines, a series of biochemical experiments, patient-derived organotypic cultures (n=7), electron microscopy, Tomocube live-imaging, CRISPR-Cas9 knockout, and in vivo models with IVIS system (n=20), Spatial transcriptome (Xenium, n=10), Phenocycler (previous CODEX, n=20), and CLARITY. RESULTS GGH emerged as a significant biomarker. (1) Its expression was elevated in IDH-wildtype astrocytoma vs. normal tissue, correlating with higher grade and worse prognosis (p=0.008). (2) Single-cell data showed predominant expression in proliferating stem cells. (3) GGH expression correlated with proliferative-stem-tumor component (p=0.0001). (4) Gene set enrichment analysis confirmed associations with oxidative phosphorylation, MYC, and ROS pathways. (5) Electron microscopy showed membrane disruption and cell death post-siRNA. (6) In tumor slice cultures, GGH knockdown led to dystrophic, necrotic changes and reduced Ki67 (p=0.033). (7) CRISPR-Cas9 knockout generated three viable clones, each with luciferase. (8) In the nude mice model, GGH knockout cells showed reduced tumor growth via IVIS investigation and histology evaluation. (9) Through spatial Xenium data and Phenocycler multiplex images, we further confirmed the relationship between GGH and proliferative stem-tumor cells, as well as the associated microenvironments. CONCLUSION Collectively, our bioinformatics-to-bench pipeline revealed GGH as a crucial driver of malignant proliferation, stemness, and poor outcomes in glioblastoma multiforme (GBM). Knockdown/knockout approaches confirmed GGH’s pivotal role in oxidative phosphorylation, MYC regulation, and ROS pathways. These findings underscore GGH’s potential as a future therapeutic target for managing resistant GBM and improving therapies.