Abstract 2313: Characterizing intratumoral cellular heterogeneity in human glioblastoma xenografts by 30+ parameter single-cell mass cytometry

Abstract

Abstract Introduction: Glioblastoma tumors contain dramatic intratumoral heterogeneity at the level of cellular phenotypes. High-parameter single-cell platforms such as mass cytometry offer an opportunity to simultaneously measure tumor subpopulations with unexpected combinations of differentiation markers, as well as stromal cells, vasculature, and infiltrating immune cells. Methods: A subcutaneous patient-derived xenograft from an adult with glioblastoma was enzymatically digested into a single-cell suspension. Half of the cells were treated with epidermal growth factor (EGF) at 20 ng/mL for 15 minutes. Both samples were then stained with a 32-marker antibody panel and measured by single-cell mass cytometry. The antibody panel included surface-expressed markers of stemness and differentiation, as well as intracellular markers to monitor cell cycle, apoptosis, and growth factor signaling. Unsupervised spectral clustering was used to segregate >200,000 tumor cells into 13 tumor cell subsets. Inducible signaling phenotypes were quantified in these subsets by comparing binding of phospho-specific antibodies in the stimulated sample, relative to the unstimulated control. Results: Many previously reported markers of cancer stem cells were detectable, but they did not agree on a single population of cells. Although the CD133+ fraction was rare (<5% of cells), it contained three distinct subpopulations of phenotypically distinct cells. The most informative markers for discriminating the tumor subsets were markers of stemness (i.e. CD133, CD44, Sox2) or integrin expression (i.e. CD49e, CD49f, CD51). The primitive cells (i.e. Sox2+) exhibited higher basal kinase activity across multiple pathways, including markers of cell division (i.e. Histone H3 pSer28). Short-term stimulation with EGF induced distinct responses in each of the different CD133+ subsets. Conclusions: Single-cell phenotyping by mass cytometry produced rich developmental and functional profiles of glioblastoma cell subsets. Subpopulations of primitive (CD133+) cells can exist in a single glioblastoma tumor and exhibit distinct signaling responses to external stimulation. Altogether, these results validate the use of this platform for investigation of the intratumoral cellular heterogeneity of solid neural tumors with this platform. Work is underway to identify correlations between intratumoral cellular phenotypes and upstream driver mutations in a larger cohort. Citation Format: Erin F. Simonds, Garry P. Nolan, William A. Weiss. Characterizing intratumoral cellular heterogeneity in human glioblastoma xenografts by 30+ parameter single-cell mass cytometry. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2313. doi:10.1158/1538-7445.AM2015-2313