Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Aberrations identified by genomic arrays in normal karyotype CMML can be detected in 40% of patients, but do not add prognostic information to molecular mutations

Chronic myelomonocytic leukemia (CMML) is a chronic clonal disease characterized by myelodysplastic (MDS) and myeloproliferative (MPN) features.1 It is defined by the occurrence of peripheral blood (PB) monocytosis (>1 × 109/l) without BCR-ABL1 fusion oncogene or PDGFRA or PDGFRB rearrangements. Cases are further subclassified in CMML-1 (<10% blasts in the bone marrow (BM)) and CMML-2 (10–19% blasts), accounting for 80 and 20% of cases, respectively. The genetic landscape is highly heterogeneous. Most frequently mutated genes are TET2 (59–60% of cases), SRSF2 (50–51%), ASXL1 (40–45%), RAS (30%), CBL (15–21%) and SETBP1 (6–15%), whereas TP53 mutations are rare (1%).2, 3 Majority of patients (up to 70%) show normal karyotype (NK), whereas clonal cytogenetic abnormalities have been identified in almost 30% of cases.4 Most frequent abnormalities are trisomy 8 (23%), loss of the Y chromosome (20%), monosomy 7 or 7q deletion (14%), complex karyotype (10%), trisomy 21 (8%), abnormalities of chromosome 3 (8%) and 20q deletion (8%). On the basis of the cytogenetic findings, the Spanish group defined three risk classes, that is, low (–Y and NK), intermediate (karyotype not classified as low or high risk) and high risk (+8 or monosomy 7).4 Balanced rearrangements have been described rarely.5

However, chromosome banding analysis (CBA) shows some limits related to the necessity of in vitro replication of the abnormal clone.6 Moreover, spatial resolution is limited to 10 MB size. Therefore, submicroscopic cytogenetic aberrations cannot be detected. Genomic arrays do not rely on proliferating cells and provide a very high spatial resolution, up to few Kb. Furthermore, arrays can detect not only copy number abnormalities (CNAs), that is, losses and gains, but also copy neutral loss of heterozygosity (CN-LOH).7 However, at least 15% of cells should harbor the genomic abnormality to allow a reliable detection. On the basis of these premises, we aimed to evaluate genomic abnormalities in CMML patients with normal CBA through the application of genomic array.

One hundred and forty patients have been analyzed. Diagnosis has been performed on morphological evaluation of PB and BM smears according to the WHO classification (2016).1 Cytogenetic and genomic array evaluation was performed on BM, and gene sequencing has been performed only on BM in 137 cases and BM/PB in three cases. Samples were referred to MLL Munich Leukemia Laboratory between October 2005 and February 2015. Morphology and cytogenetic evaluation as well as survival information were available for every patient. Median follow-up was 2.4 years. Cases with normal CBA were further studied with genomic array (SurePrint G3 ISCA CGH+SNP (4 × 180 K), Agilent, Waldbronn, Germany) and data were analyzed using Nexus Copy Number Software v. 6.1 (Biodiscovery, Inc., El Segundo, CA, USA) according to manufacturer’s recommendations. Amplicon-based next generation sequencing or polymerase chain reaction were applied to detect TET2, SRSF2, ASXL1, RUNX1, CBL, SETBP1, KRAS, NRAS, JAK2, EZH2, KIT, U2AF1, SF3B1, DNMT3A, SMC1A and CSF3R mutations in all patients. Library preparation was performed with Access Array Technology (Fluidigm, South San Francisco, CA, USA) and sequencing with MiSeq Instrument (Illumina, San Diego, CA, USA). Part of patients was the validation set published by Itzykson et al.8 and were also included in Meggendorfer et al.2, 3, 9 All patients gave written informed consent for use of data for scientific evaluations. The study was approved by the Internal Review Board and adhered to the tenets of the Declaration of Helsinki. Dichotomous variables were compared between different groups using χ2-test and continuous variables by Student’s t-test. Significance was set at P<0.05. Reported P-values are two-sided. Kaplan–Meier method was applied for overall survival (OS) curves and results were analyzed applying the two-sided log-rank test. Cox regression analysis was performed for multivariate analysis, including parameters significant (P<0.05) at univariate analysis.

Median age was 74 years and 95 patients out of 140 (68%) were male. Mean white blood cell (WBC) count was 17 × 109/l (±20 s.d.) with a mean percentage of monocytes of 29% (±12). The mean absolute monocyte count (AMC) was 5.2 × 109/l (±6.1). Thirty-one patients (22%) showed CMML-2 morphology. Ninety-five out of one hundred and forty patients (71%) showed a NK by CBA, whereas the remaining 41 patients (29%) showed CBA abnormalities. These two groups did not differ for age, gender, WBC, AMC, hemoglobin and CMML subtype. However, the group with abnormalities showed lower platelet count (mean 186 vs 131 × 109/l, P=0.004). In NK subgroup, mutations were identified at the following frequencies: TET2 (76/99, 77%), SRSF2 (55/99, 56%), ASXL1 (47/99, 48%), RUNX1 (20/99, 20%), CBL (14/99, 14%), SETBP1 (8/99, 8%), KRAS (12/99, 12%), NRAS (10/99, 10%), JAK2 (10/99, 10%), EZH2 (5/99, 5%), KIT (5/99, 5%), U2AF1 (5/99, 5%), SF3B1 (4/99, 4%), DNMT3A (2/99, 2%) and SMC1A (2/99, 2%). No mutations were detected on CSF3R gene. TET2 mutations were more frequent in NK subset (77 vs 56%, P=0.024), whereas SETBP1 mutations were more frequent in abnormal CBA subset (8 vs 22%, P=0.04). No statistical differences were evident regarding remaining genes.

Applying genomic array in all 99 cases with NK, we detected 50 CNA/CN-LOH in 40/99 patients (40%). These encompassed 28 CNA (22 deletions and 6 gains) in 24 patients and 22 CN-LOH in 20 patients (Figure 1a). The median CNA size was 0.79 MB (range 0.07–159). Only in one case, the abnormality was >10 MB, that is, a case with monosomy 7 not detected by CBA, probably due to insufficient in vitro proliferation of aberrant clone. The median CN-LOH size was 66.6 MB (range 10.14–135.44). In 19 cases, CN-LOH ranged to telomere and in three cases, it was interstitial. Majority of patients had one aberration (31/40, 78%), eight patients showed two aberrations and one patient showed three aberrations (all gains) (further details in the right panel of Figure 1a).

Figure 1

Genomic array findings and prognosis in normal karyotype (CBA) setting. (a) On the left, pie chart showing the distribution of the 50 genomic array abnormalities (22 CN-LOH, 6 gains and 22 deletions) and Venn diagram (on the right) indicating the co-occurrence of genomic array abnormalities among the 40 patients: 1 patient (2.5%) showed simultaneously 1 gain and 1 loss; 2 patients (5%) showed simultaneously 1 gain and 1 CN-LOH; 2 patients (5%) showed simultaneously 1 CN-LOH and 1 loss; 1 patient (2.5%) showed exclusively gains, in particular, 3 gains; 18 patients (45%) showed exclusively a loss, 1 of them 2 losses; 16 patients (40%) showed exclusively CN-LOH, 2 of them showed 2 CN-LOH. Percentages are referred to the subgroup with abnormal genomic array (n=40). (b) Kaplan–Meier plots showing overall survival analysis according to ASXL1 (left graph), EZH2 (left middle graph), NRAS (right middle graph) and RUNX1 (right graph) mutations. (c) Kaplan–Meier plots showing the overall survival analysis according to genomic array findings (left graph), number of aberrations (<2 vs 2) (left middle graph), occurrence of copy number abnormalities (i.e., gains and/or losses) (right middle graph) and occurrence of CN-LOH (right graph).

Except for mutual exclusion between NRAS mutations and CNA/CN-LOH, no other association with recurrently mutated genes was evident. Moreover, the number of additional mutations did not differ between normal and abnormal genomic array subgroups.

Most frequently recurrent aberration was the CN-LOH of chromosome 4q22.3q32.2 (97 791 611–190 803 979) in six patients involving TET2 gene (five harbored TET2 mutations and one patient showed SRSF2 mutation), followed by 11q13.5q25 CN-LOH (76 150 205–134 195 978) involving CBL gene (all of them showing CBL mutation) in four patients and 17q22q24.1 CN-LOH (53 404 930–63 868 162) in three patients, involving MPO gene. Moreover, 4q24 deletion (105 479 140–106 215 692), also involving TET2 gene, was found in three patients, all showing TET2 mutation. Two patients showed 7q21.3q36.3 CN-LOH (7 368 205–159 118 566) and other two patients showed 7q22.1q22.1 deletion (101 262 074–101 951 902) encompassing the CUX1 gene, a tumor suppressor gene frequently deleted in myeloid neoplasms.10 Two further patients showed Xp22.2p22.2 deletion (15 748 556–16 686 778) involving the ZRSR2 gene. The aforementioned findings are in line with the study by Tiu et al.,11 reporting a rate of aberrations of 37% in 55 CMML patients with normal CBA and Gondek et al.,12 indicating a rate of CN-LOH, among 24 unselected MDS/MPN patients, of 35%. Furthermore, we confirmed the recurrence of 11q, 7q, 17q CN-LOH.11

No specific molecular profile related to recurrent abnormalities, except for CBL mutation (n=14) that was strongly related to 11q CN-LOH (10 out of 95 patients without 11q CN-LOH, 11% vs 4 out of 4 with 11q CN-LOH, 100%, P=0.001). From the 14 patients with mutated CBL, four patients (28.6%) showed 11q CN-LOH, whereas only 2 of the 10 patients without 11q CN-LOH (20%) had two CBL mutations. This is in line with data stating that 11q CN-LOH strictly relates to CBL mutations.13 Among the 76 patients with mutated TET2, eight patients showed 4q CNA/CN-LOH. Five of these eight patients with CN-LOH showed TET2 homozygosity and three patients with deletions showed TET2 hemizygous status, whereas, among the 69 patients without 4q aberrations, 48 (70%) showed two TET2 mutations, corroborating the finding that TET2 acts as tumor suppressor gene.14 Noteworthy, only one patient did show neither any gene mutation nor any genomic array aberration.

Overall survival analysis showed a negative impact of ASXL1 (median 5.6 vs 1.7 years, P=0.008), EZH2 (4.3 years vs 4 months, P=0.03), RUNX1 (6.6 vs 1.5 years, P=0.014) and NRAS (4.3 vs 1.4 years, P=0.015) mutations (Figure 1b). However, at Cox multivariate analysis, only ASXL1 mutation retained statistical significance (hazard ratio 2.1; 95% confidence interval 1.02–4.4, P=0.04), in line with previous observations.8 On the other hand, no differences in OS were detected between normal and abnormal genomic array subgroups, also considering numbers of aberrations or recurrent lesions (Figure 1c). Recently, Palomo et al.15 investigated the impact of genomic array in 117 low-risk cytogenetic CMML (NK and loss of chromosome Y) and 11 cases with failed karyotyping. Even though they found aberrations at higher frequency compared with our series (i.e., 67%), they did not find any correlation with OS, regarding either the occurrence of any aberrations or their number, confirming our results.

In conclusion, genomic array analysis can detect aberrations in 40% of CMML patients with NK at CBA, enriching the genomic scenario (Figure 2). Recurrent genomic array aberrations were detected at 4q, 7q, 17q, 11q and Xp. However, the prognostic role of molecular genetic markers, in particular, ASXL1 mutations, overcomes the prognostic significance of any genomic aberration in terms of OS.

Figure 2

Genomic, molecular and morphological landscape in CMML cohort. Each column represents one patient. Rows correspond to analyzed genes and cytogenetic and morphological parameters. First, patients are clustered according to normal karyotype by CBA and normal genomic array (light red), genomic array aberrations in patients with normal CBA (red) and occurrence of CBA abnormalities (dark red); then, the 16 genes analyzed are listed (light gray indicates wild-type and dark gray mutation), at the end, the occurrence of CMML-1 is indicated in dull green, whereas CMML-2 morphology is indicated in black. On the top, the clinical-molecular score, according to Itzykson et al.,8 is indicated (yellow=low risk, light green=intermediate risk and orange=high risk) and the cytogenetic risk according to Such et al.4 is showed (light blue=low risk, pink=intermediate risk and blue=high risk).


  1. 1

    Arber DA, Orazi A, Hasserjian R, Thiele J, Borowitz MJ, Le Beau MM et al. The 2016 revision to the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia. Blood 2016; 127: 2391–2405.

    CAS  Article  Google Scholar 

  2. 2

    Meggendorfer M, Roller A, Haferlach T, Eder C, Dicker F, Grossmann V et al. SRSF2 mutations in 275 cases with chronic myelomonocytic leukemia (CMML). Blood 2012; 120: 3080–3088.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  3. 3

    Meggendorfer M, Haferlach T, Alpermann T, Jeromin S, Haferlach C, Kern W et al. Specific molecular mutation patterns delineate chronic neutrophilic leukemia, atypical chronic myeloid leukemia, and chronic myelomonocytic leukemia 1. Haematologica 2014; 99: e244–e246.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  4. 4

    Such E, Cervera J, Costa D, Sole F, Vallespi T, Luno E et al. Cytogenetic risk stratification in chronic myelomonocytic leukemia. Haematologica 2011; 96: 375–383.

    Article  PubMed  PubMed Central  Google Scholar 

  5. 5

    Costa D, Munoz C, Carrio A, Nomdedeu M, Calvo X, Sole F et al. Reciprocal translocations in myelodysplastic syndromes and chronic myelomonocytic leukemias: review of 5,654 patients with an evaluable karyotype. Genes Chromosomes Cancer 2013; 52: 753–763.

    CAS  Article  PubMed  Google Scholar 

  6. 6

    Maciejewski JP, Mufti GJ . Whole genome scanning as a cytogenetic tool in hematologic malignancies. Blood 2008; 112: 965–974.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7

    Schoumans J, Suela J, Hastings R, Muehlematter D, Rack K, van den BE et al. Guidelines for genomic array analysis in acquired haematological neoplastic disorders. Genes Chromosomes Cancer 2016; 55: 480–491.

    CAS  Article  PubMed  Google Scholar 

  8. 8

    Itzykson R, Kosmider O, Renneville A, Gelsi-Boyer V, Meggendorfer M, Morabito M et al. Prognostic score including gene mutations in chronic myelomonocytic leukemia. J Clin Oncol 2013; 31: 2428–2436.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  9. 9

    Meggendorfer M, Bacher U, Alpermann T, Haferlach C, Kern W, Gambacorti-Passerini C et al. SETBP1 mutations occur in 9% of MDS/MPN and in 4% of MPN cases and are strongly associated with atypical CML, monosomy 7, isochromosome i(17)(q10), ASXL1 and CBL mutations. Leukemia 2013; 27: 1852–1860.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. 10

    McNerney ME, Brown CD, Wang X, Bartom ET, Karmakar S, Bandlamudi C et al. CUX1 is a haploinsufficient tumor suppressor gene on chromosome 7 frequently inactivated in acute myeloid leukemia. Blood 2013; 121: 975–983.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  11. 11

    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: 4552–4560.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  12. 12

    Gondek LP, Tiu R, O'Keefe CL, Sekeres MA, Theil KS, Maciejewski JP . Chromosomal lesions and uniparental disomy detected by SNP arrays in MDS, MDS/MPD, and MDS-derived AML. Blood 2008; 111: 1534–1542.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  13. 13

    Makishima H, Cazzolli H, Szpurka H, Dunbar AJ, Tiu R, Huh J et al. Mutations of e3 ubiquitin ligase cbl family members constitute a novel common pathogenic lesion in myeloid malignancies. J Clin Oncol 2009; 27: 6109–6116.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  14. 14

    Tefferi A, Lim KH, Levine RL . Mutation in TET2 in myeloid cancers. N Engl J Med 2009; 361: 1117–1118.

    CAS  Article  PubMed  Google Scholar 

  15. 15

    Palomo L, Xicoy B, Garcia O, Mallo M, Adema V, Cabezon M et al. Impact of SNP array karyotyping on the diagnosis and the outcome of chronic myelomonocytic leukemia with low risk cytogenetic features or no metaphases. Am J Hematol 2016; 91: 185–192.

    CAS  Article  PubMed  Google Scholar 

Download references


We would like to thank all physicians for sending samples to MLL laboratory and for providing clinical information. In addition, we would like to thank all the coworkers at the MLL for the great work in performing technical evaluation.

Author contributions

CV performed statistical analysis and wrote the manuscript; CH was responsible for cytogenetics and genomic array; TH was responsible for cytomorphologic analysis; MZ performed cytogenetics and genomic array; NN provided bioinformatic support; WK contributed to statistical analyses; MM investigated molecular mutations and was involved in the collection of clinical data as well as manuscript preparation. All authors read and contributed to the final version of the manuscript.

Author information



Corresponding author

Correspondence to M Meggendorfer.

Ethics declarations

Competing interests

CH, WK, and TH declare equity ownership of MLL Munich Leukemia Laboratory. NN, MZ and MM are employed by MLL Munich Leukemia Laboratory. CV declares no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Vetro, C., Haferlach, C., Haferlach, T. et al. Aberrations identified by genomic arrays in normal karyotype CMML can be detected in 40% of patients, but do not add prognostic information to molecular mutations. Leukemia 30, 2235–2238 (2016).

Download citation


Quick links