B-cell progenitor acute lymphoblastic leukemia (BCP-ALL) is the most common childhood malignancy. The established subtypes of BCP-ALL, such as BCR::ABL1-positive, ETV6::RUNX1-positive, and high hyperdiplod (HeH) ALL, are important for diagnosis, prognostication, and treatment stratification. However, we and others have described novel subtypes, of which DUX4-rearranged (DUX4-r) ALL is among the most common. The subtype-specific rearrangements of DUX4 result in aberrant expression of a truncated version of this gene, which normally encodes an embryonic transcription factor. ALL is typically associated with a good prognosis, and DUX4-r ALL represents a favorable subtype. However, the cytotoxic treatment is associated with significant short- and long-term side effects. By an increased understanding of the cellular and molecular characteristics of ALL, the development of more specific treatment alternatives with high efficacy and less toxicity should be facilitated. We have used a multimodal single cell approach to delineate the transcriptional, epigenetic and phenotypic characteristics of a cohort of childhood BCP-ALLs (n=23) with a focus on the novel DUX4-r ALL subtype (n=11). The cohort also included three established subtypes, with four samples each of BCR::ABL1, ETV6::RUNX1, and HeH. The leukemic blast cell populations of the DUX4-r cases (blast count at diagnosis: 70 - 96 %) displayed distinct gene expression profiles. To deduce where the blast cells were located along a normal differentiation axis, we projected the blast cells onto a reference force directed knn-graph of normal bone marrow (NBM) cells from five healthy donors (Fig. 1). The projection of cells allowed inferring the most likely stage of maturation for each cell, as well as visualizing the altered gene expression profile compared to their healthy counterparts using a "dissimilarity score". Several DUX4-r ALL (4/11) samples exhibited a lack of B-cell progenitors after a distinct point in the graph, suggesting a distinct maturation block. For the remaining samples, the blast cells were dispersed throughout the B-cell maturation axis, indicating a partial or a more heterogenous arrest among the leukemic blasts. We also classified the cell type of each cell in the ALL samples according to a NBM reference. With this method, most of the blast cells were characterized as early progenitor B-cells. However, other cell classifications also occurred. In fact, in five of eleven DUX4-r ALLs, more than 35% of the blast population were classified as a cell type distinct from B-cell progenitors. Interestingly, the classification of blast cells as granulocyte-macrophage progenitors (GMP) was particularly common among DUX4-r ALLs (2-30% of blast cells) compared to the other subtypes (0-12%). However, such cells displayed an altered gene expression profile compared to GMP cells in NBM, as manifested by a high dissimilarity score. Nevertheless, this could suggest a higher plasticity or a more myeloid-biased transcriptional state among the blast cells within the DUX4-r subtype compared to other subtypes. To further delineate altered pathways in DUX4-r blast cells, differential gene expression analyses were performed, both between DUX4-r blast cells and blast cells of other subtypes, as well as between DUX4-r blast cells and NBM progenitor B-cells. The DUX4-r blast cells were characterized by high expression of distinct outlier genes, including ANGPT2, MCAM, EGR1, PDGFRA, NR4A1, AFT3 and PTPRM. Notably, GSEA highlighted upregulation of gene sets involved in angiogenesis- and vasculature development. In conclusion, by single cell multi-omic profiling we demonstrate that BCP-ALL blast cells are associated with distinct patient-specific gene expression patterns that are strongly influenced by subtype. DUX4-r ALL blast cells are associated with a heterogenous maturation block, transcriptional features associated with angiogenesis- and vasculature development, and possibly also higher plasticity with a myeloid-biased transcriptional state. The results should, hopefully, form a basis towards improved future diagnostics and treatment stratification of BCP-ALL, and aid the discovery of targeted treatment alternatives. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal