Acute lymphoblastic leukemia (ALL) is the most common cancer in childhood, accounting for about 20% of all pediatric malignancies with well-known genetic subtypes. The recent advances in multiomic techniques and multiparametric flow-cytometry (MFC) have allowed for distinguishing novel subtypes of B-cell precursor (BCP) ALL with a propensity to a transient immunophenotypic switch towards the myelomonocytic lineage (mm-SW) early during Induction therapy (e.g. at day 15 [D15]). This phenomenon consists of the appearance of a blast population in bone marrow characterized by the downregulation of CD19, strong expression of CD34, CD58, and CD45, and high SSC (Side Scatter), besides the blast population with the same immunophenotype observed at diagnosis. The biological basis driving this phenomenon is still unclear, but -apart from the possibility of subclonal selection - in particular cellular transdifferentiation driven by steroids and/or chemotherapeutic agents is under discussion.

Interestingly, while most tumors exhibit global DNA hypomethylation, ALL is characterized by a hypermethylated genome. However, no studies have explored the methylome landscape of mm-SW–positive (mm-SWpos) BCP-ALLs or its relationship with the transcriptomic profiles. Therefore, it is necessary to dissect the cell type composition of mm-SWposBCP-ALLs and identify the molecular mechanisms underlying mm-SW, eventually unveiling potential blast phenotypic plasticity and its epigenetic regulation.

For this purpose, a single-cell multiomic profiling was performed on mm-SWpos and mm-SWneg BCP-ALLs. Bone marrow samples collected at diagnosis (Dx, n=11) and D15 (n=6) were processed using BD Rhapsody technology to examine whole transcriptomes and epitopes at single-cell level. Applying a Seurat-based custom bioinformatic pipeline, we projected blast cells on an atlas spanning from hematopoietic stem cells along B-cell development to investigate the differences in the maturation stages between mm-SWpos and mm-SWneg blasts. Subsequently, we leveraged different bioinformatic approaches to trace evolution trajectories and cell-state dynamics between Dx and D15. Finally, by applying the minfi and CHAMP R packages, the methylome of 19 samples at Dx was investigated to highlight the possible epigenetic involvement in driving the mm-SW.

Single-cell data revealed a differential abundance of cells in the immature compartment between the two groups of patients. In detail, the mm-SWpos subgroup was characterized by an enrichment of stem cells expressing also myeloid markers, suggesting a more immature phenotype and myeloid priming, already at diagnosis. The investigation of cell-state densities identified the presence of a highly dynamic and dense cell subpopulation, linking Dx and D15 in mm-SWpos, pointing to a transcriptional reprogramming leading to mm-SW. Furthermore, DNA methylation profiling showed distinct patterns in mm-SWpos BCP-ALLs at Dx, particularly in regulatory regions such as TSS200, TSS1500, and gene bodies, implying a divergent epigenetic landscape from mm-SWneg cases. Moreover, apart from mm-SW status, a genetic subgroup-specific methylation profile was identified, besides the mm-SW status.In conclusion, our findings suggest that transient mm-SW may be due to an intrinsic biological plasticity, supported by immaturity features characterizing this BCP-ALL subgroup. Indeed, a strong immature phenotype at Dx indicates that leukemic cell fate is still not determined, and myeloid priming could sustain the switching phenomenon, under the pressure of chemotherapy and together with a different basal epigenetic status in mm-SWpos.

This content is only available as a PDF.
2025
Sign in via your Institution