PATH-29. GLIOBLASTOMA TUMOR CLASSIFICATION BASED ON SPATIAL ANALYSIS OF ONCOGENIC AMPLIFICATIONS AND PROTEIN EXPRESSION. (12th November 2021)

• Record Type: Journal Article
• Authors: Walentynowicz, Kacper; Engelhardt, Dalit; Yadav, Shreya; Onubogu, Ugoma; Salatino, Roberto; Vincentelli, Cristina; McDonald, Thomas; Michor, Franziska; Janiszewska, Michalina
• Is Part Of: Neuro-oncology. Volume 23: Supplement 6(2021)
• Page Start: vi121
• Page End: vi121
• Publication Date: 2021-11-12
• Subjects: Brain Neoplasms -- Periodicals; Brain -- Tumors -- Periodicals; Brain -- Cancer -- Periodicals; Nervous system -- Cancer -- Periodicals; 616.99481
• DOI: 10.1093/neuonc/noab196.481
• Languages: English
• ISSNs: 1522-8517

Abstract

Abstract: Heterogeneity of glioblastoma (GBM) has been extensively studied in recent years with identification of oncogenic drivers of GBM cellular subtypes. However, little is known about how these cells interact with each other or with the surrounding tumor microenvironment (TME). We employed spatial protein profiling targeting immune and neuronal markers (79 proteins) coupled with single-cell spatial maps of fluorescence in situ hybridization (FISH) for EGFR, CDK4, and PDGFRA on human GBM tissue sections. Several cores from 20 GBM samples were collected to create a tissue microarray, and 96 regions of interests were profiled with 37, 844 nuclei for oncogenic amplification screen. Spatial protein profiling identified strong correlation of certain immune markers, TAU-associated proteins, and oligodendrocyte-enriched protein groups and overall high intratumor heterogeneity of TME. Our single-cell quantification of FISH signals showed differences among tumors based on the prevalence of dual amplification of EGFR and CDK4 within a cell relative to single oncogene amplified cells. High relative frequency of dual amplification was associated with increased expression of immune-related markers and decreased expression of EGFR protein. Moreover, this protein expression signature was associated with survival in another GBM dataset. Here, we present spatial genetic analysis at the single cell level coupled with protein expression profiles associated with tumor microenvironment. Our results suggest that assessment of genetic heterogeneity in GBM could potentially drive improved patient stratification and treatment.