To explore novel genetic abnormalities occurring in myelodysplastic syndromes (MDS) through an integrative study combining array-based comparative genomic hybridization (aCGH) and next-generation sequencing (NGS) in a series of MDS and MDS/myeloproliferative neoplasms (MPN) patients. 301 patients diagnosed with MDS (n = 240) or MDS/MPN (n = 61) were studied at the time of diagnosis. A genome-wide analysis of DNA copy number abnormalities was performed. In addition, a mutational analysis of DNMT3A, TET2, RUNX1, TP53 and BCOR genes was performed by NGS in selected cases. 285 abnormalities were identified in 71 patients (23.6%). Three high-risk MDS cases (1.2%) displayed chromothripsis involving exclusively chromosome 13 and affecting some cancer genes: FLT3, BRCA2 and RB1. All three cases carried TP53 mutations as revealed by NGS. Moreover, in the whole series, the integrative analysis of aCGH and NGS enabled the identification of cryptic recurrent deletions in 2p23.3 (DNMT3A; n = 2.8%), 4q24 (TET2; n = 10%) 17p13 (TP53; n = 8.5%), 21q22 (RUNX1; n = 7%), and Xp11.4 (BCOR; n = 2.8%), while mutations in the non-deleted allele where found only in DNMT3A (n = 1), TET2 (n = 3), and TP53 (n = 4). These cryptic abnormalities were detected mainly in patients with normal (45%) or non-informative (15%) karyotype by conventional cytogenetics, except for those with TP53 deletion and mutation (15%), which had a complex karyotype. In addition to well-known copy number defects, the presence of chromothripsis involving chromosome 13 was a novel recurrent change in high-risk MDS patients. Array CGH analysis revealed the presence of cryptic abnormalities in genomic regions where MDS-related genes, such as TET2, DNMT3A, RUNX1 and BCOR, are located.
Citation: Abáigar M, Robledo C, Benito R, Ramos F, Díez-Campelo M, Hermosín L, et al. (2016) Chromothripsis Is a Recurrent Genomic Abnormality in High-Risk Myelodysplastic Syndromes. PLoS ONE 11(10): e0164370. https://doi.org/10.1371/journal.pone.0164370
Editor: Javier S. Castresana, University of Navarra, SPAIN
Received: February 25, 2016; Accepted: September 23, 2016; Published: October 14, 2016
Copyright: © 2016 Abáigar et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files. All files are available from the GEO database (accession number GSE67682).
Funding: This work was partially supported by Grants from the Spanish Fondo de Investigaciones Sanitarias FIS (PI12/00281); Proyectos de Investigación del SACYL (BIO/SA47/13; BIO/SA52/14; GRS/874/A13; GRS 994/A/14); COST Action “EuGESMA” (BM0801); Red Temática de Investigación Cooperativa en Cáncer (RTICC), Instituto de Salud Carlos III (ISCIII), Spanish Ministry of Economy and Competitiveness & European Regional Development Fund (ERDF) “Una manera de hacer Europa” (Innocampus, CEI-2010-1-0010) (RD12/0036/0069; RD12/0036/0029; RD12/0036/0044); and the European Union Seventh Framework Programme [FP7/2007-2013] under Grant Agreement n°306242-NGS-PTL. Mutation analyses were performed using oligonucleotide primer plates designed as part of the IRON-II collaborative network of hematological laboratories applying 454 amplicon next-generation deep sequencing. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. MA was supported by a “Junta para Ampliación de Estudios” fellowship [09-02402] of the Spanish National Research Council (Consejo Superior de Investigaciones Científicas, CSIC), co-funded by the European Social Fund, and by a “Grant from Fundación Española de Hematología y Hemoterapia”. AK is employed by AstraZeneca. The funders provided support in the form of salaries for these authors [MA, AK], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors [MA, AK] are articulated in the ‘author contributions’ section.
Competing interests: The authors have declared no competing financial interests. AK: current employment by AstraZeneca. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
The progressive accumulation of genetic aberrations such as copy number abnormalities, in the form of gains or losses of genetic material affecting certain regions of the genome, and particular gene mutations, provide the basis for cancer development [1,2]. However, recent studies have revealed the presence of an alternative mechanism, termed chromothripsis, in which massive chromosome rearrangements occur in a one-step catastrophic event, indicating that chromosome instability is a central aspect of cancer cell biology. A key feature of chromothripsis is the occurrence of tens to hundreds of clustered genomic rearrangements usually in one or, in some instances, several chromosomes. This complex abnormality can affect an entire chromosome, a chromosome arm, or be confined to a single region of a chromosome [2–9]. These rearrangements usually appear crisscrossing the involved regions , and chromosomes affected by chromothripsis show a characteristic pattern of copy number oscillations between two (or occasionally three) copy number states [4,8]. By far the simplest explanation of such rearrangements is that, at some stage during cancer development, distinct chromosomes or chromosomal regions are broken into many segments and then inaccurately stitched back together by DNA repair mechanisms [2,7,10]. Chromothripsis was initially observed in chronic lymphocytic leukemia (CLL), but it is present in a wide range of human cancers, including multiple myeloma (MM), acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), Hodgkin lymphoma, bone cancers, medulloblastoma, neuroblastoma, colorectal cancer and melanoma. Some congenital disorders also show chromothripsis [3–17].
Myelodysplastic syndromes (MDS) are a clinically heterogeneous group of clonal hematopoietic stem cell disorders characterized by morphological dysplasia, ineffective hematopoiesis and peripheral blood cytopenias . MDS and chronic myelomonocytic leukemia (CMML), an entity sharing features of myelodysplastic syndromes and chronic myeloproliferative neoplasms (MPN) , have a highly variable clinical course . The presence of chromosomal abnormalities is a recurrent hallmark of both MDS and CMML patients, with consequences for their diagnosis, risk stratification and prognosis [18,21,22]. In fact, these genetic changes are directly incorporated into the Revised International Prognostic Scoring System (IPSS-R) for MDS and the CMML-Specific Prognostic Scoring System (CPSS) [21,22]. In addition, gene mutations are also very frequent (80–90%) in MDS and related myeloid neoplasms [23,24]. These mutations affect transcription factors such as RUNX1 and BCOR, epigenetic modulators such as TET2, DNMT3A, IDH1/2, ASXL1 and EZH2, tumor suppressor genes such as TP53, several components of the RNA-splicing machinery such as SF3B1, SRSF2, U2AF1, and ZRSR2, genes involved in DNA replication such as SETBP1, and genes of the cohesin complex such as STAG2, RAD21, SMC1A, and SMC3. The list of genes carrying mutations involved in the pathogenesis of MDS is still growing [18,23–25].
Conventional metaphase cytogenetics (CC) is still the gold standard for karyotypic studies; however, diagnosis and prognostication may be difficult in the 10–15% of patients with non-informative cytogenetics, due to the absence of mitosis, or the 40–60% with a normal karyotype [26,27]. Additionally, the presence of complex karyotypes with three or more chromosomal abnormalities may hinder the identification of the chromosomes involved in these changes . Therefore, these CC techniques are not sufficient for a thorough study of these myeloid malignancies. The use of molecular genome-wide scanning techniques allows the identification of cryptic abnormalities in patients with a normal karyotype and the better characterization of unbalanced genetic changes [23,24,26,27,29,30]. Thus, array-based karyotyping revealed MDS and related myeloid neoplasms with a normal karyotype to have one or more genomic abnormalities, including deletions of TET2 and RUNX1 genes (40%) [26,27,29]. These studies suggest the potential clinical utility of such genome-wide scanning techniques for the management of MDS and related diseases.
To gain insight into the characterization of the molecular changes in MDS and to explore novel genetic abnormalities occurring in these disorders, an integrative study combining array-based comparative genomic hybridization (aCGH) and next-generation sequencing (NGS) in a series of MDS and MDS/MPN patients was carried out. The results demonstrated the presence of infrequent but recurrent chromothripsis involving chromosome 13 in these diseases and also showed that aCGH could be used in a clinical setting as a complementary method to conventional cytogenetics for identifying copy number changes in MDS and CMML patients.
Materials and Methods
A total of 301 patients diagnosed with MDS (n = 240) or MDS/MPN (n = 61) at the time of diagnosis were studied. The main clinical characteristics of the patients are summarized in Table 1. The median age was 77 years (range, 11–93 years), and 191 patients (63.5%) were male. Diagnoses were established according to the 2008 World Health Organization criteria  (Table 1). This study was performed in accordance with the Declaration of Helsinki guidelines, and was approved by the Local Ethical Committees “Comité Ético de Investigación Clínica, Hospital Universitario de Salamanca”. All patients provided written informed consent.
Array-based comparative genomic hybridization studies
Genome-wide DNA copy number abnormalities (CNAs) were analyzed in bone marrow samples from all patients using the Human CGH 12x135K Whole-Genome Tiling v3.0 Array (Roche NimbleGen, Madison, WI, USA). For sample preparation and hybridization the NimbleGen CGH array standard protocol was followed  (See S1 Methods). All detected CNAs were carefully reviewed to identify regions overlapping those previously reported to be copy number variants (CNVs) in the Database of Genomic Variants (http://dgv.tcag.ca/); these were excluded from subsequent analysis. Genomic abnormalities were interpreted and reported in accordance with the International System for Human Cytogenetic Nomenclature (ISCN 2013) guidelines . All the array data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE67682.
Additionally, conventional metaphase cytogenetic results were available from all patients. Based on CC results, patients were divided into three groups: patients with non-informative cytogenetics (n = 40, 13.3%) due to the absence of mitosis, patients with an abnormal karyotype (n = 45), and patients with a normal karyotype (n = 216, 71.7%). The latter group was further divided into three categories according to the number of good-quality metaphases evaluated: ≥20 metaphases (n = 164), between 11 and 19 metaphases (n = 38), and ≤10 metaphases (n = 14). Detailed information about the cytogenetic groups is summarized in Table 1, and the cytogenetic abnormalities found in the whole series are listed in S1 Table.
All genomic changes found by aCGH but not detected by conventional metaphase cytogenetics were validated by interphase fluorescence in situ hybridization (FISH), in the case of large recurrent deletions and gains, or by using an independent genome-wide analysis of DNA copy number changes with the SurePrint G3 Human CGH Microarray (8x60k) (Agilent Technologies, Palo Alto, CA, USA) for small recurrent and individual abnormalities.
Next-generation sequencing studies
Mutations in DNMT3A, TET2, RUNX1, TP53 and BCOR genes were screened by amplicon-based next-generation sequencing (NGS) in selected cases using 454 Titanium amplicon chemistry (454 Life Sciences, Branford, CT, USA). Briefly, all coding exons of TET2, RUNX1 and BCOR, exons 7–23 of DNMT3A and exons 4–11 of TP53 were covered by 27, 7, 29, 16 and 8 amplicons, respectively (S3 Table). Amplicon libraries were prepared following the manufacturer’s recommendations and previously described methods  (See S1 Methods).
For the detection of variants, all amplicon reads were analyzed with the Sequence Pilot software (v3.5.2; JSI medical systems, Kippenheim, Germany) and GS Amplicon Variant Analyzer Software (v2.9; 454 Life Sciences). Single nucleotide polymorphisms (SNPs) (http://www.ncbi.nlm.nih.gov/SNP/) and variants within introns were not scored. In addition, all mutations were validated by resequencing of PCR products from new independent PCRs.
Chromothripsis on chromosome 13 is a recurrent abnormality in high-risk MDS
The analysis of copy number profiles derived from aCGH data identified complex genomic rearrangements on the aCGH chromosome plots, showing multiple non-contiguous CNAs, with the hallmarks of chromothripsis. Based on previous studies, evidence of chromothripsis was defined as the presence of at least ten changes in segmental copy number between two or three copy number states on an individual chromosome . Using these criteria, three MDS cases (1.2%) exhibited chromothripsis (Fig 1A and 1B), with more than 11 copy number changes involving exclusively chromosome 13. The copy number states rapidly alternate between one (deletion), two (normal) and three (gain) copies. The patterns of genomic alteration were different between the three high-risk MDS patients. However, it should be pointed out that involvement of a total of 91 genes mapping on chromosome 13 were common to the three patients. As examples XPO4, FLT3 and FLT1 were commonly amplified genes; BRCA2 and RB1 were commonly deleted genes (S6 Table). All these results were validated using an independent microarray (SurePrint G3 Human CGH Microarray, 8x60k, Agilent Technologies). The three patients with chromothripsis were diagnosed as high-risk MDS (RAEB 3/40; 7.5%), two of them were RAEB-1, with 6% and 8% of BM blasts, respectively, and the remaining patient was RAEB-2, with 12% of BM blasts. All three patients died within one year. All of them had a complex karyotype revealed by aCGH (3/17; 17.6%; patients #026, #027, #072; Fig 1A), with a median of 21 CNAs (range, 19–33) throughout the whole genome. The three patients showed genomic losses on 5q23.2-q35.3, two of them also carried losses on 7q22.3-q36.3 and 15q11.1–21.2 (S1 Table).
(A) Whole genome view ratio plots derived from aCGH data of MDS patients (#026, #027 and #072) showing chromothripsis on chromosome 13, indicated by the red-shaded box. The three patients had complex karyotypes: patient #026 had 33 aberrations and affecting seven chromosomes; patient #027 had 19 abnormalities affecting five chromosomes; patient #072 had 21 aberrations affecting six chromosomes. The Y-axis represents the log2 ratio values of MDS:control signal intensities for each probe. The X-axis illustrates all the probes in the array sorted by chromosome and physical mapping position. Chromosome numbers are indicated below the X-axis. (B) Detailed view of the whole chromosome 13 in patients #026, #027 and #072 showing a complex pattern of alternating copy number gains and losses. The grey area represents the thresholds of signal values (log2 ratio) to call CNAs, red lines indicate segmented copy number profiles, and boxes shaded in pale-red and pale-blue depict copy number losses and gains, respectively. Patient #026 had 18 alternating copy number changes, patient #027 had 11, and patient #072 had 16 changes along the whole chromosome 13. Copy number profiles differed between these patients. The Y-axis represents log2 ratios and the X-axis shows all probes of chromosome 13 sorted by chromosome position. Genomic location (Mb) is indicated below the X-axis. (C) Distribution of TP53 mutations identified by amplicon-based deep sequencing in the three MDS patients with chromothripsis. All TP53 mutations were located in the sequence-specific DNA binding domain. One patient had two mutations in heterozygosis, while the other two patients had one mutation each in homozygosis. The variant allele frequencies (VAFs) are represented in brackets. Each circle represents a mutation found in one patient. Green and red circles depict missense and nonsense mutations, respectively. Each patient is illustrated by a different-colour triangle. The complete coding region of TP53 is illustrated and the respective exons and amino acid (AA) positions are indicated at the bottom. The following protein domains are shown: TAD1 and TAD2, amino-terminal transactivation domains 1 and 2; DBD, sequence-specific DNA-binding domain; NLS, nuclear localization signalling domain; 4D, carboxy-terminal tetramerization domain; Neg, negative regulation domain. Gene variants were represented using the R package “R453PlusToolbox” .
All cases showing chromothripsis carried TP53 mutations as revealed by NGS (Fig 1C). Specifically, two missense mutations (p.Lys132Arg, p.Pro278Leu) and two nonsense mutations (p.Gln136*, p.Gln167*) were identified. The p.Gln167* and p.Lys132Arg mutations, located in exon 5, were observed in one patient each with a variant allele frequency (VAF) of 93.5% and 92.5%, respectively. However, the p.Gln136* and p.Pro278Leu mutations were observed in the same patient with VAFs of 46.5% and 43.5%, and were located on exon 5 and exon 8, respectively. Taken together, most of these mutations affected exon 5 of TP53, and all of them were located in the sequence-specific DNA binding domain, which plays a central role in transcriptional transactivation (Table 2).
aCGH and NGS allow the identification of hidden recurrent genetic CNAs and gene mutations in MDS
A total of 285 abnormalities were identified (1–33 changes per patient) in 71 of the entire series of 301 patients (23.6%): 61 of 244 (25%) were MDS patients, while 10 of 58 (17.3%) had a diagnosis of CMML. Copy number losses (72.6%) were more frequent than gains (27.4%). The detected CNAs were distributed amongst all chromosomes except for chromosomes 14 and 16.
Among the global series, CNAs were present in 9.3% of the normal karyotype patients, in 86.7% of cases with abnormal cytogenetics, and in 30% of patients with unsuccessful cytogenetic analyses (Fig 2A). The most frequent large recurrent aberrations were del(5q) (35.2%), del(20q) (18.3%), -Y (14.1%), trisomy 8 (14.1%), del(7q) (12.7%), +1/+1q (7%), -18/del(18p) (5.6%), del(17p) (5.6%), del(11q) (5.6%), del(4q) (4.2%), del(15q) (4.2%), and del(12q) (4.2%) (Fig 2B). The most frequent aberrations in CMML were–Y (5.2%) and gains on 1q (3.4%), while in RCMD losses on 5q (6.1%), 20q (5.4%), and–Y (3.4%) predominated. By contrast, in RAEB patients, losses involving 5q (21.7%), 7q (13%) and 17p (6.5%), and trisomy 8 (10.9%) were frequently observed (Fig 2B). In addition, in 38 of 301 cases (12.6%), aCGH revealed the presence of 81 small aberrations (≤5 Mb), which were below the detection limit of CC. In 21 of these cases only one cryptic CNA was detected: 17 deletions and four gains. In the other 17 cases, two or more cryptic aberrations were observed, consisting of 49 deletions and 11 gains. Notably, these cryptic CNAs involved regions such as 2p23.3, 4q24, 5q33.1, 7q22.1, 21q22.12, 21q22.3 and Xp14, where genes implicated in the pathogenesis of MDS and MDS/MPN are located, including DNMT3A, TET2, SPARC, CUX1, RUNX1, U2AF1 and BCOR, respectively (Fig 2B).
(A) Proportion of the whole series of patients with normal and abnormal aCGH profiles. Each aCGH category is then divided by the cytogenetic subgroups detected by CC studies: normal, abnormal and non-informative karyotype. Percentages represent the proportion of patients from the total number of patients within each cytogenetic subgroup. (B) Frequency of large recurrent genomic abnormalities and frequency of cryptic recurrent CNAs involving genes of known significance in MDS and MDS/MPN patients only seen by aCGH. All abnormalities are classified by MDS and MDS/MPN subtypes and color-coded as indicated on the right panel of the figure.
Furthermore, an in-depth analysis by NGS of the regions with genomic deletions of >100 kb by aCGH, where recurrently mutated genes in MDS and CMML are located, was carried out. Thus, aCGH analysis identified two cases with a deletion in 2p23.3 (DNMT3A) (Fig 3A), seven cases with a deletion in 4q24 (TET2) (Fig 3E), six cases with a 17p13 deletion (TP53) (Fig 3G), five cases with a deletion in 21q22 (RUNX1) (Fig 3C) and two cases with an Xp11.4 deletion (BCOR) (Fig 3D). NGS studies detected that one patient with a DNMT3A deletion carried a missense mutation (p.Gly654Asp) in the other allele with a VAF of 73.5%. This mutation was located in the methyltransferase domain (Fig 3B). Three patients harboring a TET2 deletion carried one nonsense mutation (p.Arg550*), one missense mutation (p.Ser1284Phe) and one frameshift deletion (p.Leu1855Trpfs*32) each in the non-deleted allele. The variant allele frequencies were 96.5%, 3.5% and 89%, respectively. The latter two mutations affected the two evolutionarily conserved domains in the TET family proteins (Fig 3F). Of the six patients with a 17p13 deletion affecting the TP53 locus, four carried one missense mutation each (p.Cys275Tyr, p.Ile195Phe, p.Val274Asp, p.Tyr220Ser) with VAFs of 52%, 76.5%, 45% and 54%, respectively. These TP53 mutations were located in the sequence-specific DNA binding domain. None of the other studied genes showed mutations in those patients with losses in the regions of interest (Fig 3H). All mutations are summarized in Table 2.
(A, C, D, E, G) Detailed view of the whole chromosomes 2, 4, 17, 21 and X, where recurrent regions of deletion where found by aCGH, indicated by the red-shaded box. A magnified view of the minimal deleted regions with a schematic diagram showing the genes included within the deletion. For all figures, genomic locations are indicated in Mb, and the chromosome position (bp) and size (kb) of the minimal deleted regions are indicated in the upper part of each chromosome view ratio plots. The Y-axis represents the log2 ratio values and all probes for each chromosome are sorted by genomic position along the X-axis. (A) A 533.4-kb deletion on 2p23.3 affecting the DNMT3A locus. (C) A 298.7-kb deletion on 21q22.12 affecting the RUNX1 locus. (D) A 697.1-kb deletion on Xp11.4 affecting the BCOR locus. (E) A 381.2-kb deletion on 4q24 affecting the TET2 locus. (G) An 11.14-Mb deletion on 17p13.3-p12 affecting the TP53 locus. Genes were represented using the R package “GenomeGraphs”. (B, F, H) Distribution of DNMT3A, TET2 and TP53 mutations identified by targeted amplicon-based deep sequencing. The variant allele frequencies (VAFs) are represented in brackets. The complete coding regions of DNMT3A, TET2 and TP53 are illustrated and the respective exons and amino acid (AA) positions are indicated below. Each circle represents a mutation found in a single patient. Green, red and blue circles depict missense, nonsense and frameshift mutations, respectively. (B) One patient carried a DNMT3A missense mutation located in the MTase domain. The following protein domains are shown: PWWP, proline-tryptophan-tryptophan-proline domain; ZNF, zinc finger domain; MTase, methyltransferase domain. (F) Three patients with a TET2 deletion harbored one nonsense, one missense and one frameshift mutation each. The two evolutionarily conserved domains, boxes 1 and 2, are shown. (H) Four patients carried one TP53 missense mutation each. The following protein domains are shown: TAD1 and TAD2, amino-terminal transactivation domains 1 and 2; DBD, sequence-specific DNA-binding domain; NLS, nuclear localization signalling domain; 4D, carboxy-terminal tetrame-rization domain; Neg, negative regulation domain. Gene variants were represented using the R package “R453PlusToolbox” .
Relationship between cytogenetic results and aCGH data
Array CGH results were compared with the cytogenetic data from each patient. 56 of 83 chromosomal imbalances previously identified by CC were detected by aCGH. A remarkably high correlation between CC and aCGH results was observed in this series. In addition, given that patients from different groups ascertained by CC were included in the study, we decided to analyze the concordance between CC and aCGH, but considering each cytogenetic group of patients separately: the non-informative cytogenetic, normal and abnormal karyotype groups.
Focusing on the 216 patients with a normal karyotype as determined by CC, the aCGH results were in excellent agreement with the cytogenetics of those patients with ≥20 and 11–19 metaphases studied (92.1 and 89.5%, respectively). However, only 78.6% of those patients with ≤10 successful metaphases and no changes by CC displayed no copy number changes by aCGH (S1 Fig). Thus, 20 patients (9.3%) with a normal karyotype as determined by CC showed at least one genomic abnormality by aCGH. Only one chromosome was affected in 16 of these patients. Considering those patients with an aberrant karyotype, aCGH revealed the same genomic abnormalities as previously identified by CC in 86.7% of cases. Indeed, only six of the 45 (13.3%) cases with an abnormal karyotype established by G-banding analysis showed no copy number abnormalities with aCGH. One of these patients had a balanced translocation, which was not detectable by aCGH, and four patients had chromosomal abnormalities in three metaphases of the analyzed cells, clonal cell populations below the detection limit of the aCGH. Detailed information about these discordant cases is presented in S1 Table. With respect to the patients with unsuccessful cytogenetic analyses, 70% of cases displayed a normal aCGH profile, while 30% had at least one copy number aberration. Four patients (three high-risk MDS and one CMML) had a complex karyotype, defined by the presence of at least five copy number changes, as revealed by aCGH.
In addition, a correlation study between the aCGH profile and age, as well as WHO-based risk category, and risk stratification following the IPSS-R for MDS cases and the CPSS for the CMML patients was carried out. We found that age (P = 0.662) was not statistically associated with the presence of normal or abnormal aCGH findings, and that the presence of an abnormal aCGH profile was associated with high-risk MDS cases (P<0.05).
The presence of specific chromosomal abnormalities and genetic changes is a hallmark of MDS [18,23–25,35,36]. In this study, we analyzed a large cohort of MDS patients by integrating two genetic methodologies: array-based comparative genomic hybridization (aCGH) and amplicon-based deep sequencing (NGS). Our results demonstrated the presence of chromothripsis, inferred from aCGH profiles, as an infrequent but recurrent genomic abnormality in high-risk MDS. In addition, our results showed that the combination of these conventional and genome-wide scanning approaches enables a better characterization of MDS and related neoplasms, and provides new information that could improve the current diagnostic and treatment of these patients.
Chromothripsis is a genetic abnormality in which tens to hundreds of clustered genomic rearrangements occur in a one-step catastrophic event. [2–9]. In the present study, chromothripsis was observed in three cases of high-risk MDS (two cases of RAEB-1 and one patient with RAEB-2), two entities known to progress gradually to a more aggressive disease. The occurrence of chromothripsis in myeloid malignancies has been demonstrated in AMLs by SNP array-based copy number profiling, in which 8% of AML patients carried massive and complex rearrangements consistent with chromothripsis [6,37]. The molecular basis of this genomic chaos has been long studied in the last years. Therefore several mechanisms such as ionizing radiation, premature chromosome compaction, DNA replication stress, telomere shortening, abortive apoptosis, TP53 mutations, micronuclei collapse, and hyperploidy formation have been suggested as implicated in this chromothripsis. [2,6–8,10,38–42]. In addition, approximately 50% of AML patients carrying TP53 mutations and approximately 40% of complex karyotype AML patients displayed chromothripsis, while only 1% of AML with wild-type TP53 and no cases with non-complex karyotype showed this aberration [6,37]. Moreover, it was also demonstrated that almost all medulloblastomas showing evidence of chromothripsis had TP53 mutations . These findings reinforce the link between somatically acquired TP53 mutations and the presence of complex karyotypes with chromothripsis. It is of particular interest that the three MDS cases with chromothripsis reported in this study had complex karyotypes as revealed by aCGH and carried TP53 mutations, a previously described association . Additionally, the outcome of the three cases was poor, a feature that is in accordance with high-risk MDS patients, the presence of a complex karyotype or TP53 mutations [18,21]. This complex genomic abnormality has not previously been comprehensively described in MDS patients, perhaps because the previous high-resolution copy number studies in MDS mainly concerned cases with a normal karyotype [23,26,27,29]. Other copy number studies in MDS with abnormal and complex karyotypes have been reported. These studies aimed to analyze the relationship of copy number to the CC data and their prognostic impact, and they did not show the presence of chromothripsis [28,30,43–45]. It should be noted that, in our study, chromothripsis was seen exclusively to affect the entire chromosome 13 in all three MDS patients. The presence of this genomic chaos restricted to a single chromosome has been described. Chromothripsis is typically found to affect different chromosomes at random, but was reported to involve chromosome 16p in 3/7 MM patients, and chromosome 21 in 5/9 iAMP21 ALL patients [5,17]. However, the presence of chromothripsis involving chromosome 13 has not been previously reported in myeloid malignancies [6,12,37,46]. In this study, chromothripsis affected only chromosome 13, with patterns of genomic alteration differing between the three high-risk MDS patients. FLT3 was commonly amplified, while BRCA2 and RB1 were commonly deleted in the cases with chromothripsis. FLT3 is an oncogene that regulates hematopoietic stem cell differentiation, proliferation and survival. FLT3-activating mutations were recurrently described in myeloid malignancies, mainly in AML, and are associated with poor prognosis . RB1 and BRCA2 abnormalities play a role in the development of several cancers and are considered to be tumor suppressor genes [48,49]. In addition, two non-chromothripsis MDS cases (#074 and #126) showed CNAs affecting chromosome 13. One of these cases, diagnosed as RAEB-1, carried a deletion involving the RB1 locus. Our results suggest the involvement of the FLT3 gene and BRCA2 and RB1 inactivation in the pathogenesis of some cases of MDS.
The present study revealed the presence of cryptic abnormalities that were not targeted by FISH and that were below the threshold of resolution by conventional cytogenetics. In fact, 12.6% of MDS patients showed cryptic changes. Some of these submicroscopic CNAs involved regions with genes of known significance in MDS pathogenesis and were deletions in 2p23.3 (DNMT3A), 4q24 (TET2), 5q33.1 (SPARC), 7q22.1 (CUX1), 21q22.12 (RUNX1) and Xp11.4 (BCOR), and gains in 21q22.3 (U2AF1), that were detected in 19 patients [18,25,26,30]. These regions were equally likely to be involved in low- and high-risk MDS or CMML. In addition, the genes included in these recurrent cryptic deletions were further investigated by NGS to identify whether mutations occurred in the other allele. The sequencing results showed that only two of the seven cases with the TET2 deletion carried TET2 mutations, while only one of the two cases with losses in 2p23.3 showed DNMT3A mutations. Our results did not indicate any correlation between the presence of deletions and mutations in these MDS patients, which is in accordance with previously reported data . Therefore, our results support the idea that conventional cytogenetic and aCGH studies could be complemented by the sequencing of multi-gene panels, which have already been described for MDS and related myeloid neoplasms, instead of single genes, in routine workflows for the study of these group of patients.
The presence of cytogenetic changes is a keystone in the prognosis of MDS and CMML patients. In fact, conventional cytogenetics is essential in most prognostic systems to stratify these hematological disorders [18,21,22]. However, in some subsets of MDS and CMML, conventional cytogenetic techniques fail to provide any results due to a lack of cell growth during culturing, and consequently some aberrations may be missed, making the diagnosis and prognostic stratification very difficult. Our study showed that 30% of karyotype failures carried genomic abnormalities revealed by aCGH. Four patients had complex karyotypes, two had trisomy 8 and one showed a del(5q), while the others showed deletions in 9p, 12q, 17p, 21q22, and -Y, and gains in 1q and 15q (S1 Table). Therefore, identifying these clinically relevant lesions is significant in patients with failed CC results. The clinical utility of SNP-A as a karyotyping tool in a series of MDS patients with unsuccessful cytogenetics has been previously demonstrated . This method was also useful in those MDS patients with a normal karyotype when fewer than 20 good-quality metaphases are available for analysis. We demonstrated that in 10% of cases with 11–19 successful metaphases and in 21% of those with ≤10 harbored genomic aberrations aCGH will provide additional information that could redefine the prognostic risk of these patients, as previously suggested . Thus, at least 20 metaphases need to be analyzed for a karyotype to be considered normal . Therefore, the use of aCGH enabled the prognostic stratification according to the IPSS-R that could change the clinical management of this group of patients. Additionally, the presence of an abnormal aCGH profile was associated with high-risk patients.
In summary, the present report describes the presence of a high incidence of genomic changes in MDS and CMML patients by the integrative analysis of several molecular genetic methodologies. In addition to well-known copy number defects, the presence of chromothripsis involving chromosome 13 was a novel recurrent change in high-risk MDS patients. Moreover, aCGH analysis revealed the presence of cryptic abnormalities in genomic regions where MDS-related genes, such as TET2, DNMT3A, RUNX1 and BCOR, are located. Thus, the integrative analysis of conventional cytogenetics, aCGH and NGS in MDS will provide a better understanding of the molecular abnormalities occurring in these patients, and could improve the clinical management of MDS. The potential diagnostic and prognostic value of these new genomic abnormalities should be studied in prospective studies.
S1 Fig. Relationship between aCGH and CC studies in the normal karyotype group.
Normal karyotype patients are divided into three categories on the basis of the number of good-quality metaphases evaluated: ≤10, 11–19 and ≥20. Patients with normal and abnormal aCGH results within each category are represented by different shades of blue.
S1 Methods. Supplementary Methods.
Additional information on patients, array-based comparative genomic hybridization studies and next-generation sequencing studies.
S1 Table. All genomic copy number changes detected by aCGH and all conventional cytogenetic information for the whole series.
S2 Table. Ensembl gene and transcript IDs for the five genes selected for NGS.
S3 Table. PCR primer-pair sequences for all amplicons representing DNMT3A (16), RUNX1 (7), TET2 (27), TP53 (8) and BCOR (29) genes.
S4 Table. PCR amplification mixes for NGS.
All PCR reactions were prepared from a starting material of 30–60 ng of genomic DNA, according to the manufacturers’ recommendations. Amplification mixes #1, #2 and #4 were used with the FastStart High Fidelity PCR System, dNTPack (Roche Applied Science) and amplification mix #3 was used with the GC-RICH PCR System, dNTPack (Roche Applied Science).
S5 Table. PCR protocols for NGS.
All PCR reactions were performed in the 96-Well GeneAmp® PCR System 9700 (Applied Biosystems, Foster City, CA).
The authors would like to thank Irene Rodríguez, Sara González, Teresa Prieto, Mª Ángeles Ramos, Almudena Martín, Ana Díaz, Ana Simón, María del Pozo, Vanesa Gutiérrez and Sandra Pujante from the Centro de Investigación del Cáncer, Salamanca, Spain, for their technical assistance.
- Conceived and designed the experiments: MA JMHR.
- Performed the experiments: MA.
- Analyzed the data: MA CR RB JLG JMHR.
- Contributed reagents/materials/analysis tools: CR RB AK.
- Wrote the paper: MA FR AK MCdC JMHR.
- Treated patients, contributed patient samples, and provided the information about the patient: FR MDC LH JSdR JMA RC MM JNR GMN CA MV AAM.
- 1. Suela J, Alvarez S, Cigudosa JC. DNA profiling by arrayCGH in acute myeloid leukemia and myelodysplastic syndromes. Cytogenet Genome Res. 2007;118(2–4):304–9. pmid:18000384
- 2. Forment JV, Kaidi A, Jackson SP. Chromothripsis and cancer: Causes and consequences of chromosome shattering. Nat Rev Cancer. 2012;12(10):663–70. pmid:22972457
- 3. Kloosterman WP, Hoogstraat M, Paling O, Tavakoli-Yaraki M, Renkens I, Vermaat JS, et al. Chromothripsis is a common mechanism driving genomic rearrangements in primary and metastatic colorectal cancer. Genome Biol. 2011;12(10):R103. pmid:22014273
- 4. Stephens PJ, Greenman CD, Fu B, Yang F, Bignell GR, Mudie LJ, et al. Massive genomic rearrangement acquired in a single catastrophic event during cancer development. Cell. 2011;144(1):27–40. pmid:21215367
- 5. Magrangeas F, Avet-Loiseau H, Munshi NC, Minvielle S. Chromothripsis identifies a rare and aggressive entity among newly diagnosed multiple myeloma patients. Blood. 2011;118(3):675–8. pmid:21628407
- 6. Rausch T, Jones DT, Zapatka M, Stütz AM, Zichner T, Weischenfeldt J, et al. Genome sequencing of pediatric medulloblastoma links catastrophic DNA rearrangements with TP53 mutations. Cell. 2012;148(1–2):59–71. pmid:22265402
- 7. Maher CA, Wilson RK. Chromothripsis and human disease: Piecing together the shattering process. Cell. 2012;148(1–2):29–32. pmid:22265399
- 8. Korbel JO, Campbell PJ. Criteria for inference of chromothripsis in cancer genomes. Cell. 2013;152(6):1226–36. pmid:23498933
- 9. Kloosterman WP, Koster J, Molenaar JJ. Prevalence and clinical implications of chromothripsis in cancer genomes. Curr Opin Oncol. 2014;26(1):64–72. pmid:24305569
- 10. Meyerson M, Pellman D. Cancer genomes evolve by pulverizing single chromosomes. Cell. 2011;144(1):9–10. pmid:21215363
- 11. Edelmann J, Holzmann K, Miller F, Winkler D, Bühler A, Zenz T, et al. High-resolution genomic profiling of chronic lymphocytic leukemia reveals new recurrent genomic alterations. Blood. 2012;120(24):4783–94. pmid:23047824
- 12. Mackinnon RN, Campbell LJ. Chromothripsis under the microscope: A cytogenetic perspective of two cases of AML with catastrophic chromosome rearrangement. Cancer Genet. 2013;206(6):238–51. pmid:23911237
- 13. Nagel S, Meyer C, Quentmeier H, Kaufmann M, Drexler HG, MacLeod RA. Chromothripsis in hodgkin lymphoma. Genes Chromosomes Cancer. 2013;52(8):741–7. pmid:23630094
- 14. Boeva V, Jouannet S, Daveau R, Combaret V, Pierre-Eugène C, Cazes A, et al. Breakpoint features of genomic rearrangements in neuroblastoma with unbalanced translocations and chromothripsis. PLoS One. 2013;8(8):e72182. pmid:23991058
- 15. Hirsch D, Kemmerling R, Davis S, Camps J, Meltzer PS, Ried T, et al. Chromothripsis and focal copy number alterations determine poor outcome in malignant melanoma. Cancer Res. 2013;73(5):1454–60. pmid:23271725
- 16. Kloosterman WP, Cuppen E. Chromothripsis in congenital disorders and cancer: Similarities and differences. Curr Opin Cell Biol. 2013;25(3):341–8. pmid:23478216
- 17. Li Y, Schwab C, Ryan SL, Papaemmanuil E, Robinson HM, Jacobs P, et al. Constitutional and somatic rearrangement of chromosome 21 in acute lymphoblastic leukaemia. Nature. 2014;508(7494):98–102. pmid:24670643
- 18. Cazzola M, Della Porta MG, Malcovati L. The genetic basis of myelodysplasia and its clinical relevance. Blood. 2013;122(25):4021–34. pmid:24136165
- 19. Such E, Cervera J, Costa D, Solé F, Vallespí T, Luño E, et al. Cytogenetic risk stratification in chronic myelomonocytic leukemia. Haematologica. 2011;96(3):375–83. pmid:21109693
- 20. Malcovati L, Porta MG, Pascutto C, Invernizzi R, Boni M, Travaglino E, et al. Prognostic factors and life expectancy in myelodysplastic syndromes classified according to WHO criteria: A basis for clinical decision making. J Clin Oncol. 2005;23(30):7594–603. pmid:16186598
- 21. Greenberg PL, Tuechler H, Schanz J, Sanz G, Garcia-Manero G, Solé F, et al. Revised international prognostic scoring system for myelodysplastic syndromes. Blood. 2012;120(12):2454–65. pmid:22740453
- 22. Such E, Germing U, Malcovati L, Cervera J, Kuendgen A, Della Porta MG, et al. Development and validation of a prognostic scoring system for patients with chronic myelomonocytic leukemia. Blood. 2013;121(15):3005–15. pmid:23372164
- 23. Haferlach T, Nagata Y, Grossmann V, Okuno Y, Bacher U, Nagae G, et al. Landscape of genetic lesions in 944 patients with myelodysplastic syndromes. Leukemia. 2014;28(2):241–7. pmid:24220272
- 24. Papaemmanuil E, Gerstung M, Malcovati L, Tauro S, Gundem G, Van Loo P, et al. Clinical and biological implications of driver mutations in myelodysplastic syndromes. Blood. 2013;122(22):3616–27; quiz 3699. pmid:24030381
- 25. Bejar R, Stevenson K, Abdel-Wahab O, Galili N, Nilsson B, Garcia-Manero G, et al. Clinical effect of point mutations in myelodysplastic syndromes. N Engl J Med. 2011;364(26):2496–506. pmid:21714648
- 26. Thiel A, Beier M, Ingenhag D, Servan K, Hein M, Moeller V, et al. Comprehensive array CGH of normal karyotype myelodysplastic syndromes reveals hidden recurrent and individual genomic copy number alterations with prognostic relevance. Leukemia. 2011;25(3):387–99. pmid:21274003
- 27. Merkerova MD, Bystricka D, Belickova M, Krejcik Z, Zemanova Z, Polak J, et al. From cryptic chromosomal lesions to pathologically relevant genes: Integration of SNP-array with gene expression profiling in myelodysplastic syndrome with normal karyotype. Genes Chromosomes Cancer. 2012;51(5):419–28. pmid:22250017
- 28. Martínez-Ramírez A, Urioste M, Melchor L, Blesa D, Valle L, de Andrés SA, et al. Analysis of myelodysplastic syndromes with complex karyotypes by high-resolution comparative genomic hybridization and subtelomeric CGH array. Genes Chromosomes Cancer. 2005;42(3):287–98. pmid:15611930
- 29. Heinrichs S, Kulkarni RV, Bueso-Ramos CE, Levine RL, Loh ML, Li C, et al. Accurate detection of uniparental disomy and microdeletions by SNP array analysis in myelodysplastic syndromes with normal cytogenetics. Leukemia. 2009;23(9):1605–13. pmid:19387468
- 30. Kolquist KA, Schultz RA, Furrow A, Brown TC, Han JY, Campbell LJ, et al. Microarray-based comparative genomic hybridization of cancer targets reveals novel, recurrent genetic aberrations in the myelodysplastic syndromes. Cancer Genet. 2011;204(11):603–28. pmid:22200086
- 31. Vardiman JW, Thiele J, Arber DA, Brunning RD, Borowitz MJ, Porwit A, et al. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: Rationale and important changes. Blood. 2009;114(5):937–51. pmid:19357394
- 32. Robledo C, García JL, Benito R, Flores T, Mollejo M, Martínez-Climent JÁ, et al. Molecular characterization of the region 7q22.1 in splenic marginal zone lymphomas. PLoS One. 2011;6(9):e24939. pmid:21957467
- 33. Shaffer LG, McGowan-Jordan J, Schmid M. ISCN 2013: An International System for Human Cytogenetic Nomenclature (2013): Recommendation of the International Standing Committee on Human Cytogenetic Nomenclature. Karger, Basel 2013 ed.
- 34. Kohlmann A, Klein HU, Weissmann S, Bresolin S, Chaplin T, Cuppens H, et al. The Interlaboratory RObustness of Next-generation sequencing (IRON) study: A deep sequencing investigation of TET2, CBL and KRAS mutations by an international consortium involving 10 laboratories. Leukemia. 2011;25(12):1840–8. pmid:21681191
- 35. Haase D, Germing U, Schanz J, Pfeilstöcker M, Nösslinger T, Hildebrandt B, et al. New insights into the prognostic impact of the karyotype in MDS and correlation with subtypes: Evidence from a core dataset of 2124 patients. Blood. 2007;110(13):4385–95. pmid:17726160
- 36. Mallo M, Cervera J, Schanz J, Such E, García-Manero G, Luño E, et al. Impact of adjunct cytogenetic abnormalities for prognostic stratification in patients with myelodysplastic syndrome and deletion 5q. Leukemia. 2011;25(1):110–20. pmid:20882045
- 37. Zemanova Z, Michalova K, Buryova H, Brezinova J, Kostylkova K, Bystricka D, et al. Involvement of deleted chromosome 5 in complex chromosomal aberrations in newly diagnosed myelodysplastic syndromes (MDS) is correlated with extremely adverse prognosis. Leuk Res. 2014;38(5):537–44. pmid:24636338
- 38. Jones MJ, Jallepalli PV. Chromothripsis: Chromosomes in crisis. Dev Cell. 2012;23(5):908–17. pmid:23153487
- 39. Crasta K, Ganem NJ, Dagher R, Lantermann AB, Ivanova EV, Pan Y, et al. DNA breaks and chromosome pulverization from errors in mitosis. Nature. 2012;482(7383):53–8. pmid:22258507
- 40. Zhang CZ, Spektor A, Cornils H, Francis JM, Jackson EK, Liu S, et al. Chromothripsis from DNA damage in micronuclei. Nature 2015, Jun 11;522(7555):179–84. pmid:26017310
- 41. Mardin BR, Drainas AP, Waszak SM, Weischenfeldt J, Isokane M, Stütz AM, et al. A cell-based model system links chromothripsis with hyperploidy. Mol Syst Biol 2015, Sep;11(9):828. pmid:26415501
- 42. Maciejowski J, Li Y, Bosco N, Campbell PJ, de Lange T. Chromothripsis and kataegis induced by telomere crisis. Cell 2015, Dec 17;163(7):1641–54. pmid:26687355
- 43. O'Keefe CL, Tiu R, Gondek LP, Powers J, Theil KS, Kalaycio M, et al. High-resolution genomic arrays facilitate detection of novel cryptic chromosomal lesions in myelodysplastic syndromes. Exp Hematol. 2007;35(2):240–51. pmid:17258073
- 44. Slovak ML, Smith DD, Bedell V, Hsu YH, O'Donnell M, Forman SJ, et al. Assessing karyotype precision by microarray-based comparative genomic hybridization in the myelodysplastic/myeloproliferative syndromes. Mol Cytogenet. 2010;3:23. pmid:21078186
- 45. Tiu RV, Gondek LP, O'Keefe CL, Elson P, Huh J, Mohamedali A, et al. Prognostic impact of SNP array karyotyping in myelodysplastic syndromes and related myeloid malignancies. Blood. 2011;117(17):4552–60. pmid:21285439
- 46. Cancer Genome Atlas Research Network. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med. 2013;368(22):2059–74. pmid:23634996
- 47. Bains A, Luthra R, Medeiros LJ, Zuo Z. FLT3 and NPM1 mutations in myelodysplastic syndromes frequency and potential value for predicting progression to acute myeloid leukemia. Am J Clin Pathol. 2011;135(1):62–9. pmid:21173125
- 48. Di Fiore R, D'Anneo A, Tesoriere G, Vento R. RB1 in cancer: Different mechanisms of RB1 inactivation and alterations of pRb pathway in tumorigenesis. J Cell Physiol. 2013;228(8):1676–87. pmid:23359405
- 49. Lee H. Cycling with BRCA2 from DNA repair to mitosis. Exp Cell Res. 2014;329(1):78–84. pmid:25447315
- 50. Arenillas L, Mallo M, Ramos F, Guinta K, Barragán E, Lumbreras E, et al. Single nucleotide polymorphism array karyotyping: A diagnostic and prognostic tool in myelodysplastic syndromes with unsuccessful conventional cytogenetic testing. Genes Chromosomes Cancer. 2013;52(12):1167–77. pmid:24123380
- 51. Coleman JF, Theil KS, Tubbs RR, Cook JR. Diagnostic yield of bone marrow and peripheral blood FISH panel testing in clinically suspected myelodysplastic syndromes and/or acute myeloid leukemia: A prospective analysis of 433 cases. Am J Clin Pathol. 2011;135(6):915–20. pmid:21571964
- 52. Haferlach C, Rieder H, Lillington DM, Dastugue N, Hagemeijer A, Harbott J, et al. Proposals for standardized protocols for cytogenetic analyses of acute leukemias, chronic lymphocytic leukemia, chronic myeloid leukemia, chronic myeloproliferative disorders, and myelodysplastic syndromes. Genes Chromosomes Cancer. 2007;46(5):494–9. pmid:17311250
- 53. Klein HU, Bartenhagen C, Kohlmann A, Grossmann V, Ruckert C, Haferlach T, et al. R453Plus1Toolbox: An R/bioconductor package for analyzing roche 454 sequencing data. Bioinformatics. 2011;27(8):1162–3. pmid:21349869