%0 Generic
%A Blanco Carmona, Enrique
%C Heidelberg
%D 2024
%F heidok:35741
%K ATRT, IDH glioma, Tumor heterogeneity, Single-Cell, Multi-Omics, Transcriptomics, ATAC
%R 10.11588/heidok.00035741
%T Characterizing tumor heterogeneity in ATRT and IDH-mutant glioma tumors using single-cell multi-omics analyses
%U https://archiv.ub.uni-heidelberg.de/volltextserver/35741/
%X Over the course of the last decade, the world health organization (WHO) classification of tumors of the central nervous system (CNS) has started to incorporate different molecular insights as decision criteria for the categorization of different tumor types, which have promoted the surge of novel tumor types and subtypes. While methodology advances have enabled for an easier and more accurate diagnosis of the tumor cases, the underlying biology and tumor heterogeneity between tumor types and subtypes remain to be fully elucidated. This is evidenced by the dismal prognosis some tumor subtypes possess, underscoring the need for more effective and subtype-targeted tumor therapies.    Developing in parallel to the new CNS tumor classification, single-cell technologies have emerged as very powerful approaches to perform comparative analysis of tumor subtypes at different Omics layers, including transcriptomics and chromatin accessibility. Within this context, the two research projects I have worked on during my PhD focused on understanding the tumor heterogeneity depicted by the various IDH-mutant glioma or atypical teratoid/rhabdoid tumor (ATRT) subtypes.     In both cases, single-cell analyses identified a novel tumor cell subpopulation. In the case of IDH-mutant gliomas, this was a non-cycling, ribosomal-enriched tumor cell population harboring a stemness phenotype and exhibiting expression of elongation factors and oncogenes (annotated as RE). For ATRTs, a “rhabdoid ground-state” tumor cell population was identified and characterized across all SMARCB1-deficient ATRT subtypes, which presented high stemness activity, together with an expression profile resembling that of neuroblasts with cycling activity (annotated as IPC-like). Both these tumor cell populations in IDH-mutant gliomas and ATRTs, upon validation in external datasets, hold promise for the development of subtype-specific therapies, albeit further research is still needed.     Further analyses on the IDH-mutant glioma cohort revealed a differential composition of tumor-associated macrophages (TAM) across subtypes, with an increased prevalence of pro-inflammatory TAM states in astrocytomas, for which immunohistochemistry (IHC) staining revealed elevated p-STAT1 expression, suggesting the promotion of a pro-inflammatory microenvironment in astrocytomas. Longitudinal analyses on paired primary-recurrent astrocytomas sample pairs demonstrated that the composition of tumor cell types across patients at tumor recurrence remained consistent, emphasizing their therapeutic potential.     Subsequent analyses on the ATRT cohort are still being carried out. These include the examination of the single-cell chromatin accessibility data, and the characterization of the crosstalk between both tumor cell populations and the tumor and its microenvironment. Additional experiments encompass the validation, in ATRT organoid models, of druggable targets designed to push tumor cells into differentiated cell states within ATRT subtype-specific tumor cell lineages. Other analyses include the validation and spatial distribution of the various tumor and TME cell pupations in ATRTs of all three subtypes using spatial transcriptomics.     Finally, in order to address the increasing need of streamlined alternatives to generate high-quality, publication-ready data visualizations of single-cell transcriptomics data, I developed a software package for R, SCpubr. The software tool provides data visualization one-liner functions, the scope of which range from simpler visualization tasks such as inspecting dimensional reduction embeddings, displaying cell type composition, or assessing the expression or enrichment of selected genes, to inspecting the output of more complex analyses, such as copy number variant analysis or gene set enrichment analysis. Altogether, the scientific community has successfully adopted SCpubr for visualizing single-cell transcriptomic data, as evidenced by its growing number of citations.