Trends in Cancer OPEN ACCESS Opinion Aneuploidy in Cancer: Lessons from Acute Lymphoblastic Leukemia Oscar Molina,1,2,* Maria Alba Abad,3 Francesc Solé,4 and Pablo Menéndez1,2,5,* Aneuploidy, the gain or loss of chromosomes in a cell, is a hallmark of cancer. Highlights Although our understanding of the contribution of aneuploidy to cancer initiation Although aneuploidy is a hallmark of and progression is incomplete, significant progress has been made in uncovering cancer, our understanding of its contri- butions to cancer initiation and progres- the cellular consequences of aneuploidy and how aneuploid cancer cells self- sion remains limited. adapt to promote tumorigenesis. Aneuploidy is physiologically associated with sig- nificant cellular stress but, paradoxically, it favors tumor progression. Although Aneuploidy is physiologically associated more common in solid tumors, different forms of aneuploidy represent the initiating with drastic cellular stress, but paradoxi- cally favors tumor progression. Signifi- oncogenic lesion in patients with B cell acute lymphoblastic leukemia (B-ALL), cant progress has been achieved over making B-ALL an excellent model for studying the role of aneuploidy in tumori- recent years in deciphering how aneu- genesis. We review the molecular mechanisms underlying aneuploidy and ploid cancer cells self-adapt to promote discuss its contributions to B-ALL initiation and progression. tumorigenesis. Different forms of aneuploidy represent the initiating oncogenic lesion in ~35% Aneuploidy in Cancer of patients with B-cell acute lympho- The gain or loss of whole chromosomes, termed aneuploidy (see Glossary), was identified as blastic leukemia (B-ALL), especially in a distinct feature of cancer cells more than a century ago by the German zoologist Theodor childhood B-ALL. Boveri [1], and is now recognized as a major genomic insult in human cancers [2,3]. Aneuploidy Hyperdiploid B-ALL cells display a typi- is observed more frequently than any other oncogenic or tumor-suppressor mutation, and is cal aneuploid-stress signature in which found in ~90% of solid tumors and ~60% of hematological malignancies [4]. High-resolution, mitotic and chromosomal segregation genome-wide analysis has revealed that ~25% of the genome in each cancer cell is affected defects are associated with impair- ments of the condensin complex and by either chromosome-arm or whole-chromosome copy-number alterations [5]. Importantly, the aurora B kinase/chromosomal although recent studies have regarded both chromosome-arm gains and losses as passenger complex (CPC), leading to aneuploidy, these are generated by different mechanisms. Thus, according to the classical cohesion defects and mitotic slippage. definition of aneuploidy, we hereafter use the term 'aneuploidy' to refer to whole-chromosome gains and losses. Despite its high incidence, our understanding of the contribution of aneuploidy to cancer initiation and progression is limited for several reasons. First, large chromosomal changes lead to alterations in hundreds of genes and/or pathways, making it challenging to precisely identify those genes 1 involved in tumorigenesis. Second, aneuploidy is often associated with chromosome instability Josep Carreras Leukaemia Research Institute, School of Medicine, University (CIN) and/or microsatellite instability (MIN) [6,7] that generate heterogeneous aneuploid karyo- of Barcelona, Barcelona, Spain types, making it difficult to distinguish the alterations that drive cancer growth from the numerous, 2 Centro de investigación Biomédica en apparently random, genetic alterations that occur during tumorigenesis. Third, aneuploidy is highly Red de Cáncer (CIBER-ONC), Instituto de Salud Carlos III (ISCIII), Barcelona, Spain dependent on the cellular context and can both promote and inhibit tumor development in a tissue- 3 Wellcome Center for Cell Biology, specific manner [2]. Indeed, specific chromosome gains and losses have been shown to be linked University of Edinburgh, Edinburgh, to specific cell types and tumors [8,9]. Finally, the generation of models with specific aneuploidies Scotland, UK 4 Myelodysplastic Syndrome (MDS) remains technically challenging despite recent advances in microcell-mediated cell transfer [10] Research Group, Josep Carreras and in Cre/loxP [11] and CRISPR/Cas9 technology [12]. Leukaemia Research Institute, Catalan Institute of Oncology (ICO), Hospital Germans Trias i Pujol, Badalona, Spain Given the widespread aneuploidy found in cancer, there has been long-standing debate on 5 Institució Catalana de Recerca i Estudis whether it is the cause or consequence of cancer. Because cancer cells are often defective in Avançats (ICREA), Barcelona, Spain Trends in Cancer, January 2021, Vol. 7, No. 1 https://doi.org/10.1016/j.trecan.2020.08.008 37 © 2020 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Trends in Cancer pathways associated with cell-cycle control and genome stability, one possibility is that aneuploidy *Correspondence: omolina@carrerasresearch.org arises as a consequence of disruption of these pathways [13,14]. Another possibility is that (O. Molina) and aneuploidy per se acts as a tumor-initiating event. In this line, it was shown that introduction of pmenendez@carrerasresearch.org specific chromosomes into cells induces replication stress and further genomic instability [15], (P. Menéndez). which may trigger cell transformation [11,16–18]. Of note, aneuploidy has been observed in some preneoplastic conditions such as Barrett’s esophagus [19]. In addition, Down syndrome (DS) individuals harboring constitutional trisomy of chromosome 21 have a 200-fold higher risk of developing hematological malignancies [20]. Aneuploidy may clearly exert a tumor-promoting role in some cancers. We review here the cellular and molecular mechanisms underlying aneuploidy, and discuss recent findings highlighting how aneuploidy contributes to the origin and progression of B-cell acute lymphoblastic leukemia (B-ALL). B-ALL represents an excellent model for studying the role of aneuploidy in tumor development because frequent chromosome gains and losses are observed as the sole genomic abnormalities in aneuploid B-ALL. Aneuploidy Is Caused by Chromosome Segregation Defects The karyotype of most aneuploid tumors is within the range of diploid cells (46 chromosomes [4]), suggesting that chromosome imbalances occur largely through sequential losses and gains of chromosomes through CIN from an initial diploid state or by chromosome losses after endoreduplication of genetic material. Mechanisms of CIN include those that reduce mitotic fidelity leading to chromosome segregation defects [21]. Improper Kinetochore–Microtubule Attachments During prometaphase, nuclear envelope breakdown allows chromosomes to attach to microtubule- based 'spindles' via kinetochores (Figure 1). During metaphase, chromosomes are concentrated in a region at the mid-plane of the bipolar mitotic spindle known as the metaphase plate. The back- to-back geometry of sister kinetochores on each chromosome favors the attachment of each chro- matid to microtubules arising from opposite poles (amphitelic attachments), thus allowing proper chromosome bi-orientation and segregation [22]. However, the stochastic nature of kinetochore– microtubule attachments frequently leads to chromosome misattachments to the spindle [23], such as syntelic, monotelic, and merotelic attachments (Figure 1). These attachment defects occur naturally in early mitosis and, in healthy cells, they are detected and corrected before ana- phase onset by a system that includes different tension-sensing factors, including aurora B kinase and Mps1, to ensure faithful chromosome segregation [24]. Merotelic attachments are the most common cause of chromosome segregation errors because they are less likely to be detected by these factors and are often not corrected in cancer cells [21,25]. The persistence of merotelic attach- ments in human cancer cells is mediated by two nonexclusive mechanisms – a decreased rate of error correction and an increased rate of merotely. Notably, some cancer cell lines display hyperstable merotelic kinetochore–microtubule attachments in early mitosis [25], suggestive of inherent defects in correcting erroneous kinetochore–microtubule attachments. Accordingly, de- fects in several centromere/kinetochore proteins have been observed to reduce chromosome segregation efficiency by increasing the frequency of merotelic attachments and lagging chro- mosomes in anaphase [26,27]. However, mutations of kinetochore proteins in cancer are rare, presumably because these proteins are essential for life. However, imbalances in the levels of these proteins might act dominantly [23]. Correspondingly, other mechanisms such as epigenetic abnormalities in centrochromatin, such as loss of histone H3 lysine 4 dimethylation, that is typically found at the core centromere interspersed with the centromere-specific histone H3 CENP-A, were shown to lead to abnormal kinetochore protein levels resulting in chromosome mis-segregation [28,29]. 38 Trends in Cancer, January 2021, Vol. 7, No. 1 Trends in Cancer OPEN ACCESS Glossary Aneuploidy: gain or loss of one or more chromosomes resulting in an unbalanced chromosome number that is not an exact multiple of the haploid complement. Bridge-fusion-breakage cycle: a mechanism of chromosome instability in which broken ends from different chromatids fuse, leading to the formation of a dicentric chromosome. This chromosome bridge breaks apart in mitosis as centromeres are pulled in opposite directions, generating broken ends that can reinitiate the cycle. Centrosomes: microtubule organizing centers in metazoans; they consist of a pair of cylindrical centrioles surrounded by a matrix of proteins called the pericentriolar material, where most centrosome-associated proteins localize. Chromosomal passenger complex (CPC): a heterotetrameric protein complex consisting of an enzymatic component aurora B kinase together with three regulatory components: INCENP (inner centromere protein), survivin, and borealin. The CPC dynamically localizes to different subcellular localizations throughout mitosis to regulate key events such as kinetochore–microtubule attachments, activation of the SAC, and assembly and maintenance of the outer kinetochore. Chromosome instability (CIN): persistent gains and losses of chromosomes during cell division. It is important to underscore the difference between aneuploidy (the 'state' of the karyotype) and CIN (the 'rate' of karyotypic changes). Chromothripsis: a mutational process by which up to thousands of clustered chromosomal rearrangements occur following the formation of micronuclei containing mis-segregated chromosomes. Cohesin: a multiprotein ring-shaped complex that maintains replicated sister chromatids together from S-phase until anaphase onset. Euploidy: a chromosome number that is an exact multiple of the haploid complement (n). It includes diploidy, Trends in Cancer triploidy, tetraploidy, and polyploidy. Figure 1. Mechanisms of Chromosome Mis-Segregation. Newly replicated chromatids are held together by cohesin, Ploidies other than diploidy decrease the a multiprotein complex that in humans is composed of four subunits: two structural maintenance of chromosomes (SMC) fitness of cells but, compared with ATPases (SMC1 and SMC3), the kleisin sister-chromatid cohesion protein 1 (RAD21), and stromal antigen (SA). During aneuploid cells, polyploid cells are prophase, the bulk of cohesin is removed from chromosome arms by Pds5–WAPL and only remains at centromeres that relatively fit because their gene are protected by shugoshin 1 (Sgo1). In prometaphase, kinetochore–microtubule attachments are initiated, including expression remains balanced and they amphitelic attachments (sister kinetochores are attached to microtubules from opposite poles), syntelic attachments (both therefore represent a more stable state than any aneuploidy. (Figure legend continued at the bottom of the next page.) Trends in Cancer, January 2021, Vol. 7, No. 1 39 Trends in Cancer Defects in Chromosome Cohesion Kinetochore: a large protein complex In eukaryotes, newly duplicated sister chromatids are held together by cohesin [30]. Cohesin is that assembles at centromeres and binds to microtubules and directs also important to ensure the back-to-back orientation of kinetochores that facilitate chromosome chromosome segregation [90]. amphitely (Figure 1), and is involved in other fundamental cellular processes such as DNA repair Li–Fraumeni syndrome: a rare and transcription regulation [31]. Cohesin is removed from chromosomes in two stages during autosomic dominant disorder caused by cell division to allow physical separation of the sister chromatids and accurate transfer of the mutations in the TP53 gene; the syndrome is associated with an genetic material into the two daughter cells. The bulk of cohesin is removed from the chromo- increased risk of developing different some arms during early mitosis in a process driven by the cohesin-associated factor Pds5 and types of cancer. the cohesin-release factor WAPL that form the Pds5–WAPL complex (Figure 1). Residual cohesin Microsatellite instability (MIN): clonal is concentrated at centromeres where it is protected from removal by shugoshin-1 (from changes in the number of repeated DNA nucleotide units in microsatellite the Japanese 'guardian spirit'), resulting in the classical X-shaped mitotic chromosome [32] sequences. It arises in some tumors (Figure 1). Once all chromosomes have aligned at the metaphase plate, the protease separase owing to inactivation of mismatch repair removes centromeric cohesion and triggers anaphase onset, allowing separation of the sister chro- genes. matids to opposite spindle poles (Figure 1) [30,33]. Defective sister chromatid cohesion can result Mitosis: a highly orchestrated cellular process for faithful segregation of in chromosome mis-segregation and aneuploidy, and both mutations and abnormal expression of replicated chromosomes into two cohesin-related genes have been identified in different human cancers [34]. Given the essential role identical daughter cells. of cohesin in genome stability and transcription regulation, cohesion dysfunction might affect Spindle-assembly checkpoint tumorigenesis in different ways; further studies will be necessary to shed light on the functional (SAC): an evolutionarily conserved surveillance mechanism that detects the importance of mutations in cohesin and its interacting partners. presence of misattached kinetochores in mitosis and pauses division until all pairs Defective Spindle-Assembly Checkpoint of sister chromatids achieve the Eukaryotic cells have developed an evolutionarily conserved surveillance mechanism to prevent bioriented geometry. chromosome mis-segregation as a result of erroneous kinetochore–microtubule attachments, termed the spindle-assembly checkpoint (SAC). The SAC pathway serves to restrain the protease separase, which cleaves the cohesin complex in anaphase (Figure 1). Impairment of the SAC allows precocious anaphase onset, and consequently significantly increases the frequency of chromosome mis-segregation [35]. Complete SAC ablation leads to massive chromosome mis-segregation and catastrophic cell death in cell lines [35,36] and is embryonic lethal in mice [37], but not in Drosophila, where chromosome attachment to the mitotic spindle occurs extremely rapidly [38]. By contrast, heterozygous mutations in different SAC proteins induce aneuploidy and tumorigenesis in mice [39–41]. Of note, mutations in BUB1B, that encodes the SAC protein BubR1, cause mosaic variegated aneuploidy in humans – a rare and recessive syndrome that is associated with mosaic aneuploidy and predisposition to cancer development [41]. Supernumerary Centrosomes Centrosomes are responsible for the formation of the mitotic spindle that is needed for proper chromosome segregation [42]. Coordination of the duplication and segregation cycles of centro- somes is essential to prevent chromosome segregation defects, because the number of centro- somes in a cell determines the number of spindle poles [43]. Duplication of the centrosome that is present in a G1 cell begins during S phase. In early mitosis the duplicated centrosomes nucleate sister kinetochores are attached to microtubules from the same pole), monotelic attachments (only one sister kinetochore is attached to the spindle), and merotelic attachments (both sister kinetochores are simultaneously attached to microtubules emanating from the same pole). Erroneous attachments are sensed by aurora B (AURKB) and Mps1 kinases that retain the mitotic checkpoint complex (Bub proteins, Mad2, and Cdc20) at unattached kinetochores and keep the spindle assembly checkpoint (SAC) activated, blocking the activity of the anaphase-promoting complex/cyclosome (APC/C). In metaphase, the tension established by amphitelic attachments causes the chromosomes to align at the metaphase plate. Correct kinetochore–microtubule attachments inactivate the SAC and allow APC/C activation, releasing separase to cleave cohesin rings at centromeres and promoting anaphase onset. Merotelic attachments are usually not sensed by aurora B, and these chromosomes frequently lag behind in anaphase, giving rise to either euploid or aneuploid daughter cells. Lagging chromosomes can be trapped in the cleavage furrow in cytokinesis, leading to chromosome breakage, or they segregate intact to a daughter cell where they can eventually encapsulate as a micronucleus. Abbreviations: PCM, pericentriolar material; Ub, ubiquitin. 40 Trends in Cancer, January 2021, Vol. 7, No. 1 Trends in Cancer OPEN ACCESS the spindle microtubules and, later on in mitosis, the centrosomes from opposite spindle poles segregate to the daughter cells together with a complete set of chromosomes. An abnormal number of centrosomes can arise from defects in the centrosome replication cycle, segregation failure during mitosis, or cell fusion, leading to genome doubling and polyploidy. Remarkably, several observations support the hypothesis that polyploid cells represent early steps in tumor formation because centrosome duplication represents a source of merotelic attachments leading to chromosome mis-segregation [44–46]. Consequences of Aneuploidy and the 'Aneuploidy Paradox' in Cancer The deleterious effects of aneuploidy in physiology are well established from yeast to mammalian cells [16,47]. Aneuploidy is often lethal in multicellular organisms, and the few aneuploidies that do not cause lethality can reduce growth and trigger developmental abnormalities, leading to a substantial fitness cost under most circumstances [2]. Indeed, aneuploid mice generated by chromosome transfer die in utero at early stages [16]. Aneuploid cells share a set of phenotypes that are collectively known as the 'aneuploidy-stress phenotype' [48], which includes specific cellular responses such as lower proliferation rates, genetic and metabolic stress, and organismal effects including immune system activation (Box 1). Despite the detrimental consequences of aneuploidy for cell fitness, the fact that it is a strikingly common feature in cancer creates a paradox in terms of its contribution to tumorigenesis, the so-called 'aneuploidy paradox' [2]. It is now clear that the effects of aneuploidy on tumorigenesis are much more complex than was initially proposed. Aneuploidy and CIN can act both as tumor- suppressors and tumor-initiators depending on the context [2,18]. A pan-cancer genomic analysis based on the potency and distribution of oncogenes and tumor-suppressor genes suggested that cumulative haploinsufficiencies and triplosensitivities drive aneuploidy patterns Box 1. Aneuploidy-Stress Phenotype Cellular Effects of Aneuploidy The most evident effect of aneuploidy is a reduction in cell proliferation and delayed cell-cycle progression at the G1–S transition [16,47,71]. Of note, the proliferation defects are not observed in aneuploid mouse embryonic stem cells or human pluripotent stem cells [17,72,73], suggesting a variable tolerance to aneuploidy depending on the developmental stage of the cell. Lower proliferation rates are most likely a response to other cellular stresses associated with aneuploidy, such as gene expression and proteotoxic stresses. In addition, aneuploid cells have a characteristic gene expression signature in which gene expression networks are globally disturbed, similarly to the environmental stress response in yeast [47]. In mammalian cells, the aneuploidy-stress response involves the conserved downregulation of pathways involved in nucleic acid metabolism, including replication, DNA repair, transcription, and RNA processing, as well as upregulation of pathways associated with autophagy, lysosomal pathways, and membrane metabolism, metabolic pathways such as glycolysis, and pathways associated with the inflammatory response [71]. The increased gene expression triggers proteotoxic stress as cells attempt to cope with the unbalanced excess of proteins generated by additional chromosome numbers. Proteotoxic stress is characterized by impaired protein folding, activation of degradation pathways, and accumulation of cytoplasmic protein aggregates with consequent accumulation of autophagosomes [74,75]. These pathways allow aneuploid cells to maintain protein homeostasis against a background of elevated gene expression. In addition to metabolic stresses, aneuploidy generates further genomic instability, leading to cancer genome evolution and adaptation. Genomic instability in aneuploid cells is triggered by different mechanisms such as replication stress [15], chromosome bridge-fusion-breakage cycles [76,77], and chromothripsis [78]. Organismal/Physiological Effects Aneuploidy and CIN activate the innate immune response through the cGAS–STING pathway (cyclic GMP-AMP synthase– stimulator of interferon genes), a cellular defense against viral infection [79,80]. The cGAS–STING pathway detects DNA double-strand breaks (DSBs) that leak from micronuclei into the cytoplasm and induce the transcriptional activa- tion of inflammatory pathways, such as type I interferon signaling and the senescence-associated secretory phenotype (SASP), thus providing a link between chromosome mis-segregation, aneuploidy, and innate immune signaling [80]. Indeed, the type I interferon pathway is consistently upregulated in aneuploid mammalian cells, and the expression levels of proinflammatory cytokines increase upon chromosome mis-segregation [79,80]. Thus, aneuploidy triggers signals for cell self-elimination that may serve as a means for cancer cell clearance. Trends in Cancer, January 2021, Vol. 7, No. 1 41 Trends in Cancer and shape the cancer genome in human primary tumors [49]. These genetic effects are tissue- specific [9], and this may explain the tumor type-specificity of the observed patterns of chromo- some gains and losses. Collectively, these data highlight the balance between the deleterious and beneficial consequences of aneuploidy and CIN during the early steps of tumor evolution. Aneuploidy in Hematological Neoplasms Dramatically different biases of individual chromosome gains and losses have been reported between solid tumors and hematological neoplasms, and these could reflect their fundamentally different etiologies [4]. Whereas solid tumors preferentially lose chromosomes, this relationship is not so clear in hematological neoplasms, which show a more balanced rate of chromosome gains and losses [4]. Notably, the most frequently gained chromosome in hematological neoplasms is chromosome 21, which has a clear bias towards loss in solid tumors. Consistently, patients with DS, with constitutional trisomy 21, have a >200-fold increased susceptibility to hematological neoplasms, including acute megakaryocyte-erythroid leukemia and B-ALL, but appear to be protected from solid tumors [50]. Importantly, trisomy 21 and its mouse homolog (trisomy 16) have been shown to perturb the development of hematopoietic stem and progenitor cells (HSPCs), specifically B lymphoid differentiation [20,51,52]. These data suggest that trisomy 21 influences cell-autonomous differentiation and transformation phenotypes in progenitor B cells, thus linking specific aneuploidies to the development of hematological neoplasms. Because hematological neoplasms are characterized by the presence of recurrent chromosomal abnormalities, cytogenetic studies at diagnosis remain essential to (i) confirm clonality, (ii) classify patients based on World Health Organization (WHO)-defined subgroups, and (iii) stratify patients according to their prognosis to determine the most suitable treatment (revised international prog- nostic scoring system, IPSS-R) [53]. Beyond structural chromosome abnormalities, aneuploidy is frequent in some hematological neoplasms and is associated with specific clinical outcomes (Table 1). Although it is generally challenging to define which chromosomal abnormality represents the primary change, chromosome gains and losses are mostly accompanied by additional cytoge- netic aberrations that reflect disease progression in most hematological neoplasias. Nevertheless, aneuploidy represents the primary event in some hematological neoplasms (Table 1), where ALL Table 1. Main Aneuploidy Cytogenetic Abnormalities in Hematological Neoplasmsa Neoplasm Cytogenetic alterations Clinical outcome Refs Chronic myelomonocytic leukemia (CMML) −Y Favorable [81] +8, CK Poor Myelodysplastic syndromes (MDS) −Y Very favorable [53] −7b Poor CK Very poor Acute myeloblastic leukemia (AML) −5b, −7, +8, −17 or abn(17p), CK, MK Poor [82] b Chronic lymphocytic leukemia (CLL) +12 , +18 NPA [83] Multiple myeloma (MM) +1, +3, +7, +11, +15 NPA [84,85] Non-Hodgkin lymphoma (NHL) +3, +5, +7, +12, +18 NPA [86] Myeloproliferative neoplasms (MPN) -7 , +8 , +9 , −Y b b b b NPA [53] Acute lymphoblastic leukemia (ALL) +21, hyperdiploidy (>50 chr)b Favorable [87,88] b Hypodiploidy (<40 chr) Poor a Abbreviations: abn, abnormality; chr, chromosomes; CK, complex karyotype; del, deletion; MK, monosomal karyotype; NPA, no prognosis associated. b As a single abnormality. 42 Trends in Cancer, January 2021, Vol. 7, No. 1 Trends in Cancer OPEN ACCESS is the best example, where highly aneuploid karyotypes are observed without other detectable cytogenetic abnormalities (Box 2). The fact that chromosome gains and losses represent the unique cytogenetic abnormality in most aneuploid B-ALL cases makes this an excellent model for studying the pathogenic effects of aneuploidy on the initiation and progression of hematological cancers. Etiology of Aneuploidy in Childhood B-ALL Genomic studies mapping the temporal relationship between chromosomal gains or losses and specific mutations in aneuploid chromosomes of B-ALL have revealed that aneuploidies arising in utero are the initiating event in both hyperdiploid and hypodiploid childhood B-ALL [54,55]. These findings are consistent with studies showing that hyperdiploidy is present at birth, as shown by genetic analysis of monozygotic twins with concordant and discordant high hyperdiploid (HeH) B-ALL [56], the presence of hyperdiploid cells in cord blood samples [57], and the identifica- tion of clonotypic IGH rearrangements in Guthrie cards (neonatal blood tests that are also known as 'heel-prick tests') from patients who later developed HeH-B-ALL [58]. Overall, these data strongly suggest that aneuploidies are early initiating events acquired prenatally during fetal hematopoiesis, as has been observed for other common reciprocal translocations in B-ALL, and also that a secondary postnatal mutational event is necessary to promote leukemia development [59,60]. This requirement for secondary cooperating oncogenic insults is suggested by the variable post- natal latency period of the disease and by the inability of common B-ALL fusion genes to promote leukemia in either transgenic mice per se [61] or human HSPCs [62]. Accordingly, aneuploid B-ALL cases typically have recurrent secondary mutations or copy-number alterations involving genes such as CDKN2A, PAX5, IKZ1, and ETV6, as well as mitogen-activated protein kinase (MAPK) signaling pathway genes (RAS, FLT3, and PTPN11) and histone modifiers [63,64]. The cellular and molecular mechanisms leading to aneuploidy in B cell progenitors are largely unknown because the aberrant mitosis that generates the gross chromosome gains and losses Box 2. Aneuploidies in B Cell ALL Aneuploidies are common cytogenetic abnormalities in B-ALL, particularly in pediatric cases that account for ~30% of B-ALL cases. The most frequent aneuploidies are hyperdiploid karyotypes, generally with several chromosome gains, and these represent the largest cytogenetic subgroup in childhood B-ALL (25–30% of cases) [67]. Hyperdiploid B-ALL is classified according to the chromosome gains into low-hyperdiploid (HeL-B-ALL), with 47–50 chromosomes; high- hyperdiploid (HeH-B-ALL), with 51–67 chromosomes; and near-triploid/tetraploid, with >67 chromosomes (Figure IA– C). Among hyperdiploid B-ALL cases, the most common group is HeH-B-ALL, which is associated with a favorable prognosis and has a long-term overall survival (OS) of >90% [67]. Chromosome gains in B-ALL are not random, and HeH-B-ALL shows preferential gains of chromosomes 4, 6, 10, 14, 17, 18, 21, and X. Further research will be neces- sary to establish the individual contributions of each chromosome gain in both leukemia origin/initiation and (sub)clonal intratumor heterogeneity. Although HeH-B-ALL is the most common aneuploidy in childhood B-ALL, chromosome losses are also observed and are classified as hypodiploid B-ALL, which represents <10% of patients with B-ALL [89]. Based on the number of chromo- some losses, hypodiploid B-ALL cases are classified as high-hypodiploid (HoH-B-ALL), with 40–45 chromosomes; low-hypodiploid (HoL-B-ALL), with 30–40 chromosomes (Figure ID); and near-haploid (Figure IE), with <30 chromosomes. Most hypodiploid cases are HoH-B-ALL, but this group is genetically heterogeneous and does not have a clear prognostic value. By contrast, HoL-B-ALL and near-haploid B-ALL, that account for ~2% of pediatric B-ALL cases, have an important prognostic value but an extremely dismal prognosis (5 year OS of <25%) [89]. Similarly to hyperdiploid B-ALL, cases of hypodiploid B-ALL display a preferential loss of chromosomes. In near-haploid B-ALL, retained disomies principally com- prise chromosomes X/Y, 8, 10, 14, 18, and 21, whereas HoL-B-ALL shows preferential retention of disomies X/Y, 1, 5, 6, 8, 10, 11, 14, 18, 19, 21, and 22. Half of all patients with HoL-ALL frequently harbor 'inherited' TP53 mutations, suggesting that TP53-mutated HoL-B-ALL is a clinical evolution of Li–Fraumeni syndrome [64]. Notably, half of all hypodiploid B-ALL cases show chromosomal doubling in the hypodiploid clone, resulting in clones with 50–78 chromosomes (Figure ID). The doubled clones frequently represent the major leukemic clone at diagnosis, leading to 'masked-hypodiploidy' [89], which is clinically challenging because these patients could be erroneously classified and be treated as hyperdiploid B-ALL while being at high risk of treatment failure. Trends in Cancer, January 2021, Vol. 7, No. 1 43 Trends in Cancer Trends in Cancer Figure I. Aneuploid Karyotype Configurations Found in B-ALL. (A) G-banded karyotype of pseudodiploid B-ALL blasts, where the only aneuploidy observed is +21. (B) G-banded karyotype of high-hyperdiploid B-ALL blasts. (C) R-banded karyotype of near-tetraploid B-ALL blasts. (D) G-banded karyotype of low-hypodiploid B-ALL blasts (above) and a doubled-up clone observed in the same patient sample with 74 chromosomes (below). Both clones were seen at disease presentation. (E) R-banded karyotype of near-haploid B-ALL blasts. Abbreviation: B-ALL, B cell acute lymphoblastic leukemia; mar, structurally abnormal marker chromosome in which no part can be identified. in aneuploid B-ALL cannot be directly observed. The evidence for the cellular mechanisms leading to aneuploid B-ALL is thus purely circumstantial. Studies of microsatellite sequence allelic ratios in tetrasomic chromosomes and uniparental disomies (UPDs) in HeH-B-ALL cases strongly suggest that hyperdiploid karyotypes arise as a result of simultaneous chromosome gains in a single abnormal mitosis in most cases (~70%), or by chromosome losses from a tetraploid inter- mediate (~30%) [65]. However, the cellular mechanisms leading to these mitotic defects remain unknown. 44 Trends in Cancer, January 2021, Vol. 7, No. 1 Trends in Cancer OPEN ACCESS Whole-genome sequencing analyses in large cohorts of patients with aneuploid B-ALL have not Outstanding Questions identified recurrent mutations or defects in genes associated with mitosis or cell division that Is there a common pathogenic could explain the causative mechanisms for these aneuploidies in B cell progenitors [63,64]. mechanism that is shared between hyperdiploid and hypodipoid B-ALL? However, no studies on the mitotic mechanisms involved in aneuploid B-ALL have been per- formed. We recently reported the first and most comprehensive analysis of mitotic progression What is the contribution of mitotic and of HeH-B-ALL using primary blasts and primograft-derived xenograft models [66]. The studies chromosome segregation defects to the revealed that hyperdiploid B-ALL blasts show a delay in early mitosis in prometaphase as a result origin and progression of aneuploidy in B-ALL? of defects in chromosome alignment that ultimately lead to chromosome segregation errors, as observed by the presence of a higher frequency of anaphase bridges and lagging chromosomes What are the precise cellular and in late mitosis compared with non-hyperdiploid B-ALL samples. Accordingly, we observed molecular mechanisms that underlie disease initiation and progression in karyotype variability in HeH-B-ALL samples with a major clone and a series of smaller clones, aneuploid subtypes of B-ALL? suggesting a mild CIN phenotype. Although the presence of CIN in HeH-B-ALL remains contro- versial [67], the analysis is supported by the direct visualization of chromosome mis-segregation What secondary alterations are in mitosis, and favors the presence of CIN observed by single-cell analyses such as interphase- necessary (if any) to develop leukemia in aneuploid B-ALL? FISH (fluorescence in situ hybridization) in HeH-B-ALL samples [68,69]. Biochemical analysis of dividing B-ALL blasts revealed that the condensin complex is functionally impaired in HeH-B- What is the contribution of each ALL blasts, leading to chromosome hypocondensation, loss of centromere stiffness, and chromosome gain or loss in both leukemia initiation and progression? mislocalization of the chromosomal passenger complex (CPC) that contains aurora B kinase Are chromosome-number alterations as its enzymatic core. Chromosome architecture defects have recently been reported by other stochastic or hierarchical? authors using Hi-C (high-throughput sequencing and chromatin conformation capture) analysis of primary HeH-B-ALL samples [70] and can explain very well the poor chromosome morphology Does intratumor heterogeneity contribute to progression of aneuploidy B-ALL? that clinical cytogeneticists historically encountered in metaphase spreads of HeH-B-ALL. HeH- B-ALL blasts showed chromatid cohesion defects and loss of SAC efficiency, leading to mitotic Why is aneuploidy preferentially observed slippage, perhaps because aurora B kinase fails to localize to the inner centromere to protect in pediatric B-ALL rather than in adult B-ALL or acute myeloid leukemia chromatid cohesion and allow proper SAC activity; this provides a direct link between chromo- (AML)? Is there a developmental link some misalignment and segregation defects in late mitosis. Overall, these data constitute direct between aneuploidy and B cell evidence for mitotic and chromosome defects in HeH-B-ALL cells and highlight the molecular leukemia? mechanisms involved in the pathogenesis of this frequent subgroup of childhood B-ALL. Whether these mechanisms are involved in the origins of hyperdiploidy or are the consequence of chromo- some gains in B cell progenitors remains an open question that requires further investigation. Concluding Remarks An important caveat to these studies is the precise identification of causative versus consequential defects with a view to establishing the mechanism responsible for the different aneuploidies in B-ALL. In this sense, mouse models can provide important insights into the mechanisms underlying mitotic defects in B cell progenitors and the effects of aneuploidy on leukemia initiation and progression. A better understanding of the biology of hyperdiploid and hypodiploid B-ALL subtypes will open up new avenues for in vivo modeling aneuploidy by genetically engineering HSPCs. This will be crucial for answering several key questions about the role of aneuploidy in leukemia initiation and progression (see Outstanding Questions). The new data obtained from these studies will translate into better diagnosis and treatment of patients with these common forms of leukemia. Acknowledgments We are indebted to all the members of the laboratory of P.M. for their insights. Special thanks to Dr Isabel Granada (Hematology laboratory at Germans Trias i Pujol Hospital, Catalan Institute of Oncology, Badalona, Spain), Dr Paola Ballerini (Hopital Trousseau, Paris, France), and Drs Nassera Abermil and Chrystele Bilhou-Nabera (Saint Antoine Hospital, Paris, France) for providing the karyotype images. We thank Dr A.A. Jeyaprakash (Wellcome Trust Centre for Cell Biology, University of Edinburgh, Edinburgh, UK) for critical revision of the MS. We thank Centres de Recerca de Catalunya (CERCA)/Generalitat de Catalunya, the Fundació Josep Carreras, and the Obra Social 'la Caixa' for their institutional support. Work in the laboratory of P.M. is supported by Spanish Ministry of Cience and Innovation (PID2019-108160RB-I00/AEI/10.13039/501100011033), the European Research Council Trends in Cancer, January 2021, Vol. 7, No. 1 45 Trends in Cancer (CoG-2014-646903 and PoC-2018-811220), the Spanish Ministry of Economy and Competitiveness (SAF2016-80481-R), the Fundación Uno entre Cienmil, and the Fundación Leo Messi. F.S. is supported by Instituto de Salud Carlos III, the Spanish Ministry of Economy and Competitiveness (PI/17/0575), and SGR288/2017 (GRC). We also thank the Wellcome Trust for supporting M.A. A. through a Senior Research Fellowship to A.A. Jeyaprakash (202811). P.M. is an investigator of the Spanish Cell Therapy Coop- erative Network (TERCEL). References 1. Boveri, T. (2008) Concerning the origin of malignant tumours by 25. Bakhoum, S.F. et al. (2009) Deviant kinetochore microtubule Theodor Boveri. Translated and annotated by Henry Harris. dynamics underlie chromosomal instability. Curr. Biol. CB 19, J. Cell Sci. 121, 1–84 1937–1942 2. Ben-David, U. and Amon, A. (2020) Context is everything: aneu- 26. Abad, M.A. et al. (2014) Structural basis for microtubule recognition ploidy in cancer. Nat. Rev. Genet. 21, 44–62 by the human kinetochore Ska complex. Nat. Commun. 5, 2964 3. Gordon, D.J. et al. (2012) Causes and consequences of aneu- 27. Cimini, D. et al. (2006) Aurora kinase promotes turnover of kineto- ploidy in cancer. Nat. Rev. Genet. 13, 189–203 chore microtubules to reduce chromosome segregation errors. 4. Duijf, P.H. et al. (2013) Cancer cells preferentially lose small Curr. Biol. CB 16, 1711–1718 chromosomes. Int. J. Cancer 132, 2316–2326 28. Molina, O. et al. (2016) Epigenetic engineering reveals a balance 5. Beroukhim, R. et al. (2010) The landscape of somatic copy- between histone modifications and transcription in kinetochore number alteration across human cancers. Nature 463, maintenance. Nat. Commun. 7, 13334 899–905 29. Molina, O. et al. (2017) Using human artificial chromosomes to 6. Bolhaqueiro, A.C.F. et al. (2019) Ongoing chromosomal instabil- study centromere assembly and function. Chromosoma 126, ity and karyotype evolution in human colorectal cancer 559–575 organoids. Nat. Genet. 51, 824–834 30. Uhlmann, F. and Nasmyth, K. (1998) Cohesion between sister 7. Gryfe, R. et al. (2000) Tumor microsatellite instability and clinical chromatids must be established during DNA replication. Curr. outcome in young patients with colorectal cancer. N. Engl. Biol. CB 8, 1095–1101 J. Med. 342, 69–77 31. Remeseiro, S. et al. (2012) A unique role of cohesin-SA1 in gene 8. Ben-David, U. et al. (2011) Large-scale analysis reveals acquisi- regulation and development. EMBO J. 31, 2090–2102 tion of lineage-specific chromosomal aberrations in human adult 32. Morales, C. and Losada, A. (2018) Establishing and dissolving stem cells. Cell Stem Cell 9, 97–102 cohesion during the vertebrate cell cycle. Curr. Opin. Cell Biol. 9. Sack, L.M. et al. (2018) Profound tissue specificity in proliferation 52, 51–57 control underlies cancer drivers and aneuploidy patterns. Cell 33. Uhlmann, F. et al. (2000) Cleavage of cohesin by the CD clan 173, 499–514 protease separin triggers anaphase in yeast. Cell 103, 375–386 10. Liskovykh, M. et al. (2016) Moving toward a higher efficiency of 34. Mannini, L. and Musio, A. (2011) The dark side of cohesin: the microcell-mediated chromosome transfer. Mol. Ther. Methods carcinogenic point of view. Mutat. Res. 728, 81–87 Clin. Dev. 3, 16043 35. Michel, L.S. et al. (2001) MAD2 haplo-insufficiency causes pre- 11. Thomas, R. et al. (2018) Whole chromosome loss and associ- mature anaphase and chromosome instability in mammalian ated breakage-fusion-bridge cycles transform mouse tetraploid cells. Nature 409, 355–359 cells. EMBO J. 37, 201–218 36. Kops, G.J. et al. (2004) Lethality to human cancer cells through 12. Zuo, E. et al. (2017) CRISPR/Cas9-mediated targeted chromo- massive chromosome loss by inhibition of the mitotic check- some elimination. Genome Biol. 18, 224 point. Proc. Natl. Acad. Sci. U. S. A. 101, 8699–8704 13. Hernando, E. et al. (2004) Rb inactivation promotes genomic 37. Wang, Q. et al. (2004) BUBR1 deficiency results in abnormal instability by uncoupling cell cycle progression from mitotic megakaryopoiesis. Blood 103, 1278–1285 control. Nature 430, 797–802 38. Buffin, E. et al. (2007) Flies without a spindle checkpoint. Nat. 14. Manning, A.L. et al. (2010) Loss of pRB causes centromere dys- Cell Biol. 9, 565–572 function and chromosomal instability. Genes Dev. 24, 39. Sotillo, R. et al. (2007) Mad2 overexpression promotes aneu- 1364–1376 ploidy and tumorigenesis in mice. Cancer cell 11, 9–23 15. Passerini, V. et al. (2016) The presence of extra chromosomes 40. Foijer, F. et al. (2014) Chromosome instability induced by Mps1 leads to genomic instability. Nat. Commun. 7, 10754 and p53 mutation generates aggressive lymphomas exhibiting 16. Williams, B.R. et al. (2008) Aneuploidy affects proliferation and spon- aneuploidy-induced stress. Proc. Natl. Acad. Sci. U. S. A. 111, taneous immortalization in mammalian cells. Science 322, 703–709 13427–13432 17. Zhang, M. et al. (2016) Aneuploid embryonic stem cells exhibit 41. Hanks, S. et al. (2004) Constitutional aneuploidy and cancer pre- impaired differentiation and increased neoplastic potential. disposition caused by biallelic mutations in BUB1B. Nat. Genet. EMBO J. 35, 2285–2300 36, 1159–1161 18. Weaver, B.A. et al. (2007) Aneuploidy acts both oncogenically 42. Prosser, S.L. and Pelletier, L. (2017) Mitotic spindle assembly in and as a tumor suppressor. Cancer cell 11, 25–36 animal cells: a fine balancing act. Nat. Rev. Mol. Cell Biol. 18, 19. Persons, D.L. et al. (1998) Interphase cytogenetics of esopha- 187–201 geal adenocarcinoma and precursor lesions. Cancer Genet. 43. Nigg, E.A. (2002) Centrosome aberrations: cause or conse- Cytogenet. 106, 11–17 quence of cancer progression? Nat. Rev. Cancer 2, 815–825 20. Roy, A. et al. (2012) Perturbation of fetal liver hematopoietic stem 44. Bielski, C.M. et al. (2018) Genome doubling shapes the evolution and progenitor cell development by trisomy 21. Proc. Natl. Acad. and prognosis of advanced cancers. Nat. Genet. 50, 1189–1195 Sci. U. S. A. 109, 17579–17584 45. Fujiwara, T. et al. (2005) Cytokinesis failure generating tetraploids 21. Cimini, D. et al. (2001) Merotelic kinetochore orientation is a promotes tumorigenesis in p53-null cells. Nature 437, 1043–1047 major mechanism of aneuploidy in mitotic mammalian tissue 46. Ganem, N.J. et al. (2009) A mechanism linking extra centro- cells. J. Cell Biol. 153, 517–527 somes to chromosomal instability. Nature 460, 278–282 22. McIntosh, J.R. (2016) Mitosis. Cold Spring Harb. Perspect. Biol. 47. Torres, E.M. et al. (2007) Effects of aneuploidy on cellular physiology 8, a023218 and cell division in haploid yeast. Science 317, 916–924 23. Thompson, S.L. and Compton, D.A. (2011) Chromosome 48. Chunduri, N.K. and Storchova, Z. (2019) The diverse conse- missegregation in human cells arises through specific types of quences of aneuploidy. Nat. Cell Biol. 21, 54–62 kinetochore–microtubule attachment errors. Proc. Natl. Acad. 49. Davoli, T. et al. (2013) Cumulative haploinsufficiency and Sci. U. S. A. 108, 17974–17978 triplosensitivity drive aneuploidy patterns and shape the cancer 24. DeLuca, K.F. et al. (2011) Temporal changes in Hec1 phosphory- genome. Cell 155, 948–962 lation control kinetochore–microtubule attachment stability during 50. Choi, J.K. (2008) Hematopoietic disorders in Down syndrome. mitosis. J. Cell Sci. 124, 622–634 Int. J. Clin. Exp. Pathol. 1, 387–395 46 Trends in Cancer, January 2021, Vol. 7, No. 1 Trends in Cancer OPEN ACCESS 51. Pfau, S.J. et al. (2016) Aneuploidy impairs hematopoietic stem 71. Stingele, S. et al. (2012) Global analysis of genome, tran- cell fitness and is selected against in regenerating tissues scriptome and proteome reveals the response to aneuploidy in in vivo. Genes Dev. 30, 1395–1408 human cells. Mol. Syst. Biol. 8, 608 52. Lane, A.A. et al. (2014) Triplication of a 21q22 region contributes to B 72. Ben-David, U. et al. (2014) Aneuploidy induces profound cell transformation through HMGN1 overexpression and loss of his- changes in gene expression, proliferation and tumorigenicity of tone H3 Lys27 trimethylation. Nat. Genet. 46, 618–623 human pluripotent stem cells. Nat. Commun. 5, 4825 53. Greenberg, P.L. et al. (2012) Revised international prognostic 73. Catalina, P. et al. (2008) Human ESCs predisposition to karyo- scoring system for myelodysplastic syndromes. Blood 120, typic instability: is a matter of culture adaptation or differential 2454–2465 vulnerability among hESC lines due to inherent properties? 54. Paulsson, K. (2016) High hyperdiploid childhood acute lympho- Mol. Cancer 7, 76 blastic leukemia: chromosomal gains as the main driver event. 74. Donnelly, N. et al. (2014) HSF1 deficiency and impaired HSP90- Mol. Cell. Oncol. 3, e1064555 dependent protein folding are hallmarks of aneuploid human 55. Safavi, S. et al. (2013) Loss of chromosomes is the primary event in cells. EMBO J. 33, 2374–2387 near-haploid and low-hypodiploid acute lymphoblastic leukemia. 75. Santaguida, S. et al. (2015) Aneuploidy-induced cellular stresses Leukemia 27, 248–250 limit autophagic degradation. Genes Dev. 29, 2010–2021 56. Bateman, C.M. et al. (2015) Evolutionary trajectories of 76. Tanaka, H. and Watanabe, T. (2020) Mechanisms underlying re- hyperdiploid ALL in monozygotic twins. Leukemia 29, 58–65 current genomic amplification in human cancers. Trends Cancer 57. Maia, A.T. et al. (2004) Identification of preleukemic precursors of 6, 462–477 hyperdiploid acute lymphoblastic leukemia in cord blood. Genes 77. Umbreit, N.T. et al. (2020) Mechanisms generating cancer Chromosom. Cancer 40, 38–43 genome complexity from a single cell division error. Science 58. Gruhn, B. et al. (2008) Prenatal origin of childhood acute 368, eaba0712 lymphoblastic leukemia, association with birth weight and 78. Crasta, K. et al. (2012) DNA breaks and chromosome pulveriza- hyperdiploidy. Leukemia 22, 1692–1697 tion from errors in mitosis. Nature 482, 53–58 59. Greaves, M. (2018) A causal mechanism for childhood acute 79. Santaguida, S. et al. (2017) Chromosome mis-segregation lymphoblastic leukaemia. Nat. Rev. Cancer 18, 471–484 generates cell-cycle-arrested cells with complex karyotypes 60. Bueno, C. et al. (2019) Natural history and cell of origin of TC F3- that are eliminated by the immune system. Dev. Cell 41, ZN F384 and PTPN11 mutations in monozygotic twins with con- 638–651 cordant BCP-ALL. Blood 134, 900–905 80. Mackenzie, K.J. et al. (2017) cGAS surveillance of micronuclei 61. Andreasson, P. et al. (2001) The expression of ETV6/CBFA2 (TEL/ links genome instability to innate immunity. Nature 548, AML1) is not sufficient for the transformation of hematopoietic cell 461–465 lines in vitro or the induction of hematologic disease in vivo. Cancer 81. Such, E. et al. (2011) Cytogenetic risk stratification in chronic Genet. Cytogenet. 130, 93–104 myelomonocytic leukemia. Haematologica 96, 375–383 62. Montes, R. et al. (2011) Enforced expression of MLL–AF4 fusion 82. Grimwade, D. et al. (2010) Refinement of cytogenetic classifica- in cord blood CD34+ cells enhances the hematopoietic tion in acute myeloid leukemia: determination of prognostic sig- repopulating cell function and clonogenic potential but is not suf- nificance of rare recurring chromosomal abnormalities among ficient to initiate leukemia. Blood 117, 4746–4758 5876 younger adult patients treated in the United Kingdom 63. Paulsson, K. et al. (2015) The genomic landscape of high Medical Research Council trials. Blood 116, 354–365 hyperdiploid childhood acute lymphoblastic leukemia. Nat. 83. Dohner, H. et al. (2000) Genomic aberrations and survival in Genet. 47, 672–676 chronic lymphocytic leukemia. N. Engl. J. Med. 343, 1910–1916 64. Holmfeldt, L. et al. (2013) The genomic landscape of hypodiploid 84. Drach, J. et al. (1995) Multiple myeloma: high incidence of chro- acute lymphoblastic leukemia. Nat. Genet. 45, 242–252 mosomal aneuploidy as detected by interphase fluorescence in 65. Paulsson, K. et al. (2005) Evidence for a single-step mechanism situ hybridization. Cancer Res. 55, 3854–3859 in the origin of hyperdiploid childhood acute lymphoblastic leuke- 85. Walker, B.A. et al. (2014) Intraclonal heterogeneity is a critical mia. Genes Chromosom. Cancer 44, 113–122 early event in the development of myeloma and precedes the 66. Molina, O. et al. (2020) Impaired condensin complex and Aurora development of clinical symptoms. Leukemia 28, 384–390 B kinase underlie mitotic and chromosomal defects in 86. Wang, S.S. et al. (2008) Chromosomal aberrations in peripheral hyperdiploid B-cell ALL. Blood 136, 313–327 blood lymphocytes and risk for non-Hodgkin lymphoma. J. Natl. 67. Paulsson, K. and Johansson, B. (2009) High hyperdiploid child- Cancer Inst. Monogr. 78–82 hood acute lymphoblastic leukemia. Genes Chromosom. Can- 87. Moorman, A.V. et al. (2010) Prognostic effect of chromosomal cer 48, 637–660 abnormalities in childhood B-cell precursor acute lymphoblastic 68. Talamo, A. et al. (2010) Clonal heterogeneity and chromosomal leukaemia: results from the UK Medical Research Council instability at disease presentation in high hyperdiploid acute lym- ALL97/99 randomised trial. Lancet Oncol. 11, 429–438 phoblastic leukemia. Cancer Genet. Cytogenet. 203, 209–214 88. Schwab, C. and Harrison, C.J. (2018) Advances in B-cell precur- 69. Alpar, D. et al. (2014) Sequential and hierarchical chromosomal sor acute lymphoblastic leukemia genomics. HemaSphere 2, e53 changes and chromosome instability are distinct features of 89. Safavi, S. and Paulsson, K. (2017) Near-haploid and low- high hyperdiploid pediatric acute lymphoblastic leukemia. hypodiploid acute lymphoblastic leukemia: two distinct subtypes Pediatr. Blood Cancer 61, 2208–2214 with consistently poor prognosis. Blood 129, 420–423 70. Yang, M. et al. (2019) Proteogenomics and Hi-C reveal transcrip- 90. Fukagawa, T. and Earnshaw, W.C. (2014) The centromere: tional dysregulation in high hyperdiploid childhood acute lym- chromatin foundation for the kinetochore machinery. Rev. Dev. phoblastic leukemia. Nat. Commun. 10, 1519 Cell 30, 496–508 Trends in Cancer, January 2021, Vol. 7, No. 1 47