Clonal origin and development of high hyperdiploidy in childhood acute lymphoblastic leukaemia

Abstract

High hyperdiploid acute lymphoblastic leukemia (HeH ALL), one of the most common childhood malignancies, is driven by nonrandom aneuploidy (abnormal chromosome numbers) mainly comprising chromosomal gains. In this study, we investigate how aneuploidy in HeH ALL arises. Single cell whole genome sequencing of 2847 cells from nine primary cases and one normal bone marrow reveals that HeH ALL generally display low chromosomal heterogeneity, indicating that they are not characterized by chromosomal instability and showing that aneuploidy-driven malignancies are not necessarily chromosomally heterogeneous. Furthermore, most chromosomal gains are present in all leukemic cells, suggesting that they arose early during leukemogenesis. Copy number data from 577 primary cases reveals selective pressures that were used for in silico modeling of aneuploidy development. This shows that the aneuploidy in HeH ALL likely arises by an initial tripolar mitosis in a diploid cell followed by clonal evolution, in line with a punctuated evolution model.

Chromosomal instability in aneuploid acute lymphoblastic leukemia associates with disease progression

“…1C ). The HS observed across aneuploid cB-ALL samples are well within the range of those observed in previous studies using scWGS of cB-ALL samples (Bakker et al, 2016 ; Woodward et al, 2023 ), demonstrating the reliability of our data. Overall, the results consistently showed moderately higher levels of chr-CNH in all aneuploid cB-ALL samples than in Eup-B-ALL and HSPCs.…”

“…CIN is caused by an increased frequency of chromosome segregation errors, leading to cell-to-cell variability in chromosomal content and in adaptation to diverse cellular stresses (Vasudevan et al, 2021 ). Given the inherent complexity in studying the actual rate of chromosome mis-segregation in actively dividing cB-ALL primary cells, CIN is typically assessed by quantifying chr-CNH within a given population (Alpar et al, 2014 ; Ramos-Muntada et al, 2022 ; Woodward et al, 2023 ). To comprehensively study the presence of CIN and its relationship with both chr-CNH and disease outcome, we generated PDX models using a discovery cohort consisting of 12 primary diagnostic cB-ALL samples, three samples per ploidy group (Table 1 and Fig.…”

“…However, the presence and active role of CIN in cB-ALL is under debate and has been limited to studies of karyotype heterogeneity because primary leukemic cells fail to grow ex vivo (Alpar et al, 2014 ; Elghezal et al, 2001 ; Heerema et al, 2007 ; Paulsson et al, 2010 ; Ramos-Muntada et al, 2022 ; Talamo et al, 2010 ). A recent study applied scWGSeq to directly visualize the cell-to-cell variability of the entire karyotype from nondividing cells in nine HeH-B-ALL samples, and found variable levels of genome-wide HS in the HeH-B-ALL samples, ranging from low-to-mid chr-CNH (Woodward et al, 2023 ). Here, we applied a novel method to infer chr-CNH from bulk WGS data using a large panel of cB-ALL samples (van Dijk et al, 2021 ) and scWGSeq to eight cB-ALL samples with different ploidies (Bakker et al, 2016 ).…”

Constitutional and acquired genetic variants in ARID5B in pediatric B‐cell precursor acute lymphoblastic leukemia

“…Why this results in increased risk specifically for HeH ALL is not known. A characteristic feature of the HeH subtype is the nonrandom gain of chromosomes, with an extra copy of chromosome 10, that carries ARID5B , being seen in ~70% of cases 5,17 . We have previously reported that HeH ALLs that are constitutionally heterozygous for ARID5B risk alleles and with an acquired trisomy 10 more commonly gain the chromosome 10 homolog carrying the risk allele 7 .…”

“…A total of 1335 BCP ALL cases from five different cohorts were included in the investigation, of which 590 were HeH. Cohort 1 comprises 268 HeH ( n = 113) and non‐HeH ( n = 155) cases and is an expansion from a previously published cohort 15–17 . Cohort 2, The Therapeutically Applicable Research to Generate Effective Treatments (TARGET, dbGAP accession number phs000464) comprises 706 cases (HeH; n = 116, non‐HeH; n = 590) genotyped using either the Affymetrix Genome‐Wide Human SNP Array 6.0 or whole genome sequencing (WGS) based on the Complete Genomics technology.…”