Glioblastoma multiforme (GBM), a highly malignant, non-curative brain tumor of the primary central nervous system, has been subclassified in the past years in several distinct subtypes using a multitude of analysis methods. Here, we introduce the MACSima™ imaging platform which allows for fully automated, multiparametric, cyclic immunofluorescence analysis of specimens with hundreds of antibodies. We apply this method to characterize and classify glioblastomas according to published classification schemes and identify novel glioblastoma specific markers. Glioblastoma xenografts derived from primary tumors were dissociated to single cells using the Tumor Dissociation Kit, human and the gentleMACS™ Octo Dissociator (Miltenyi Biotec). Single cells were analyzed for cell surface marker expression by flow cytometry (MACSQuant Analyzer 10) including 371 directly conjugated antibodies (MACS Marker Screen, Miltenyi Biotec). A ranking was applied according to the percentage of positive cells and the stain index, leading to a selection of 96 markers for characterization of eight primary glioblastoma using the MACSima™ imaging platform. Cryosections were fixed by acetone and each specimen was exposed to 96 fluorescent labeled antibodies by cycles of antibody reaction, image acquisition and erasure of signal. The 2D image stacks were analyzed for antigen quantification and pattern recognition using both, pixel and segmented single-cell data. The detection of previously published markers, such as PDGFRα, Olig2, p16, p53, Synaptophysin, CD44, Nestin, Podoplanin, GFAP, MET, Hes-1 and EGFR was used to subclassify the glioblastomas according to i) Motomura (Motomura et al., 2012) into Oligodendrocyte Precursor (OPC), Differentiated Oligodendrocyte (DOC), Astrocytic Mesenchymal (AsMes) or Mixed subtype, and ii) Verhaak (Verhaak et al., 2010) into Proneural, Neural, Classical, or Mesenchymal subtype. The analysis of well-established glioblastoma marker partially already used in CAR T cell based clinical trials such as EGFRvIII, HER2, or IL-13Rα2 revealed a broad inter- and intratumor diversity of expression. Infiltrating immune cells were present in most of the tumors, but showed varying percentages. Finally, segmentation, clustering, and correlation analysis allowed for identification of new marker which might be used for a more robust classification of glioblastomas. In summary, our analysis using the MACSima™ imaging platform reveals a high heterogeneity of protein expression in glioblastomas along with the ability to deeper classify the diverse tumors and identify novel markers that allow selective detection of tumor cells for their potential use in immunotherapy.
Citation Format: Sandy Reiß, Stefan Tomiuk, Jutta Kollet, Jan Drewes, Wolfgang Brück, Melanie Jungblut, Andreas Bosio. Characterization and classification of glioblastoma multiforme using the novel multiparametric cyclic immunofluorescence analysis system MACSima [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 245.