COMP-14. MOLECULAR PROFILING AND CELLULAR DECONVOLUTION OF GLIOBLASTOMA BRAIN TUMORS USING CHROMATIN RUN-ON AND SEQUENCING. (11th November 2019)

• Record Type: Journal Article
• Authors: Chu, Tinyi; Rice, Edward; Salamanca, Hans; Wang, Zhong; Longo, Sharon; Corona, Robert; Viapiano, Mariano; Chin, Lawrence; Danko, Charles
• Is Part Of: Neuro-oncology. Volume 21(2019)Supplement 6
• Page Start: vi64
• Page End: vi64
• Publication Date: 2019-11-11
• Subjects: Brain Neoplasms -- Periodicals; Brain -- Tumors -- Periodicals; Brain -- Cancer -- Periodicals; Nervous system -- Cancer -- Periodicals; 616.99481
• DOI: 10.1093/neuonc/noz175.257
• Languages: English
• ISSNs: 1522-8517

Abstract

Abstract: Glioblastoma is among the most heterogeneous malignancies, making difficult the identification of clinically-relevant interactions between tumor cells and their supportive tumor microenvironment. Moreover, whether the heterogeneity of tumor cells is reflected by changes in the composition of the tumor microenvironment remains poorly defined. To further understand the cellular heterogeneity of GBM, we used our previously validated chromatin run-on and sequencing (ChRO-seq) method to analyze 61 GBMs from a retrospective cohort of patients banked at the State University of New York (Upstate Medical Center) between 1987 and 2007 (characteristics: Male:Female ratio= 2:1; median age at diagnosis= 59 years; median KPS=80; median overall survival= 343 days). We developed a new Bayesian statistical model that uses transcription at cell-type specific enhancers to identify the cellular composition of the tumor microenvironment in each patient. We validated our tool using simulations and scATAC-seq data from the same specimens, showing large improvements in sensitivity and accuracy compared with CYBERSORT. Integrative analysis of cellular composition and matching clinical data revealed correlations between the presence of specific cell types in the tumor mass and clinical variables. Finally, our analysis allowed us to identify transcription factors (e.g., NF-kB, C/EBPB) that control gene expression changes, revealing which cell types are controlled by each transcription factor in the GBM microenvironment. Our study uncovers new insights into the cellular heterogeneity of GBM and its impact on clinical progression and survival.