INTER AND INTRA-TUMOR HETEROGENEITY OF PEDIATRIC-TYPE DIFFUSE HIGH-GRADE GLIOMA REVEALED BY HIGH-DIMENSIONAL SINGLE-CELL PROTEOMICS

Abstract Paediatric-type diffuse high-grade gliomas (PDHGG) are aggressive brain tumors, affecting children and young adults, with no effective treatments. A main constraint to the development of effective treatment is associated with their highly heterogeneous nature. In order to further dissect their intra and inter tumor heterogeneity, we exploited the mass cytometry technology, an advanced -OMIC approach that, by using metal-tagged antibodies, allows the simultaneous measurement of more than 40 markers, at single-cell level. Here we characterized 8 primary cell lines derived from diffuse pediatric-type high-grade glioma H3-wildtype (DHGG-WT), Diffuse hemispheric glioma H3G34-mutant (DHG-G34) and Diffuse midline glioma H3K27-altered (DMG-K27) patients. The adopted antibody panel was set to recognize antigens expressed by brain and tumor cells, including H3K27M and H3.3G34R variants, and it highlighted important intra- and inter- tumor heterogeneity in the expression of the 16 considered markers. Of these, CD56, CD44, CD29 and NESTIN were more expressed in the hemispheric cell lines, while CD90 was more expressed in the pontine. Even if there was not always a concordance between CyTOF and mRNA expression data from cell lines and tumor samples (e.g. CD90 and GFAP), CyTOF data were in line with the immunohistochemistry analysis for GFAP, whose expression was significantly higher in H3.1K27 compared to H3.3K27. The UMAP analysis allowed us to identify 10 cell clusters, with very minimal overlap between hemispheric and pontine location subgroups and with a peculiar antigenic profile, whose abundance strongly varied according to the mutational subgroups. For example, while the G34 subgroup was enriched for cluster 9 (CD29/CD63/CD56/PDGRFa), the H3.1K27 was enriched for cluster 3 (H3K27M/CD90/CD63/CD56) and cluster 4 (H3K27M/CD63/CD90/CD56/GFAP). In conclusion, single-cell mass cytometry reveals a significant inter and intra-tumoral heterogeneity at protein level, dependent on the molecular alterations. This approach could contribute to the identification of new clinically relevant biomarkers for PDHGG.