An integrative molecular map of pediatric B-cell precursor acute lymphoblastic leukemia.
[BACKGROUND] The molecular landscape of pediatric B-cell precursor acute lymphoblastic leukemia (BCP-ALL) has been extensively characterized through single-modality studies.
APA
Krali O, Enblad AP, et al. (2026). An integrative molecular map of pediatric B-cell precursor acute lymphoblastic leukemia.. Communications medicine, 6(1). https://doi.org/10.1038/s43856-026-01568-9
MLA
Krali O, et al.. "An integrative molecular map of pediatric B-cell precursor acute lymphoblastic leukemia.." Communications medicine, vol. 6, no. 1, 2026.
PMID
41965886
Abstract
[BACKGROUND] The molecular landscape of pediatric B-cell precursor acute lymphoblastic leukemia (BCP-ALL) has been extensively characterized through single-modality studies. However, the interplay between molecular modalities and their collective influence on treatment response and outcomes remains poorly understood.
[METHODS] We integrated genomic, epigenomic, transcriptomic, and ex vivo drug response data from 1231 patients diagnosed with BCP-ALL. Using Multi-Omics Factor Analysis, we identified signatures explaining key aspects of the integrative molecular landscape, referred to as cross-modal elements (CMEs). The CME-derived signatures were introduced into pathway and intermodal network analyses, while their impact on patient outcomes was assessed through survival modeling.
[RESULTS] Pathway and network analyses annotate the resulting integrative CMEs, linking them to key biological processes, including disease development, cellular regulatory processes, metabolic pathways, and drug response. By leveraging correlations between DNA methylation and ex vivo response to doxorubicin, we stratify patients with hyperdiploidy into subgroups that differ in relapse-free survival. These signatures are independent of clinical variables. Survival models incorporating CME-selected ex-vivo drug responses combined with clinical data improve risk prediction compared to clinical models alone (FDR < 0.05), demonstrating the potential of integrative multiomics in refining risk stratification.
[CONCLUSIONS] Our study highlights the importance of multimodal data integration in BCP-ALL to provide biological insights with potential relevance for precision medicine.
[METHODS] We integrated genomic, epigenomic, transcriptomic, and ex vivo drug response data from 1231 patients diagnosed with BCP-ALL. Using Multi-Omics Factor Analysis, we identified signatures explaining key aspects of the integrative molecular landscape, referred to as cross-modal elements (CMEs). The CME-derived signatures were introduced into pathway and intermodal network analyses, while their impact on patient outcomes was assessed through survival modeling.
[RESULTS] Pathway and network analyses annotate the resulting integrative CMEs, linking them to key biological processes, including disease development, cellular regulatory processes, metabolic pathways, and drug response. By leveraging correlations between DNA methylation and ex vivo response to doxorubicin, we stratify patients with hyperdiploidy into subgroups that differ in relapse-free survival. These signatures are independent of clinical variables. Survival models incorporating CME-selected ex-vivo drug responses combined with clinical data improve risk prediction compared to clinical models alone (FDR < 0.05), demonstrating the potential of integrative multiomics in refining risk stratification.
[CONCLUSIONS] Our study highlights the importance of multimodal data integration in BCP-ALL to provide biological insights with potential relevance for precision medicine.