Mass Cytometry Of Human Glioblastoma Characterizes More Than 99 Percent Of Cells And Reveals Intratumoral Cell Subsets Defined By Contrasting Signaling Network Profiles

Background: Glioblastoma (GBM) remains largely incurable despite intense study of resected tissue. Prior studies have revealed GBM cell subsets (Patel et al., Science 2014) and have implicated subset emergence as a potential mechanism of poor outcome in other cancer types. Signaling in rare cells or a mix of cell subsets may enable therapy resistance and recurrence of GBM. For example, STAT3 RNA expression has been previously shown to correlate with poor outcome in GBM (Jahani-Asl et al., Nat Neurosci 2016 and TCGA). The complexity of GBM, combined with the interconnectedness between cancer and host cells in the microenvironment, means that a single cell biology approach is needed to comprehensively characterize patient biopsy cells and determine how protein expression, signaling, and functional capabilities impact treatment response. Methods: We developed a novel mass cytometry approach to characterize human GBM that identified ~90-95% of tumor cells (Leelatian u0026 Doxie et al., Cytometry B 2016). Here, we applied this approach using a newly created 35-antibody mass cytometry panel focused on basal phospho-protein signaling. The published panel of 16 identity proteins included SOX2, CD44, Nestin, PDGFRα, S100B, and NCAM. This panel was augmented to measure 10 additional proteins and 9 phospho-proteins including p-STAT3, p-EGFR, and p-NFκB. Signaling measurements were chosen to match prior single cell studies of signaling networks that stratified clinical outcomes in blood cancers (Irish et al., Cell 2004; PNAS 2010, Levine et al., Cell 2015). Between 10,000 and 250,000 viable cells were characterized for each tumor (N = 7). Tumors were collected with informed consent and in accord with the Declaration of Helsinki. Results: This new 35-antibody mass cytometry panel positively identified u003e99% of GBM cells. Subsets of GBM cells displayed protein expression that matched previously observed transcriptional molecular subclasses (Verhaak et al., Cancer Cell 2010 and TCGA). Strikingly, this panel revealed novel GBM cell subsets defined by contrasting basal signaling profiles. An inverse correlation was observed between baseline STAT3 phosphorylation and the abundance of CD45 + leukocytes. Additionally, similar signaling patterns were seen in cells that expressed proteins associated with distinct functions, such as proliferation and migration. Conclusions: The correlation between low STAT3 signaling and high immune cell abundance provides evidence for the idea that an intimate relationship exists between immune cells and GBM tumor growth and survival. Moreover, single cell analysis may reveal biomarkers of treatment response and allow prediction of clinical outcomes. The abnormal signaling mechanisms observed here in some GBM cell subsets should be studied further as potential targets for novel cancer-selective combination therapies. Citation Format: Nalin Leelatian, Justine Sinnaeve, Bret C. Mobley, Akshitkumar M. Mistry, Daniel Liu, Kyle D. Weaver, Reid C. Thompson, Lola B. Chambless, Rebecca A. Ihrie, Jonathan M. Irish. Mass cytometry of human glioblastoma characterizes more than 99 percent of cells and reveals intratumoral cell subsets defined by contrasting signaling network profiles [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 364. doi:10.1158/1538-7445.AM2017-364