Multiple sub-microscopic genomic lesions are a universal feature of chronic myeloid leukaemia at diagnosis

Article metrics

Chronic myeloid leukaemia (CML) is a clonal stem cell disorder, characterized at the cytogenetic level by the presence of a balanced chromosomal rearrangement, the t(9;22) or Philadelphia chromosome (Ph) translocation and at the molecular level by the presence of the BCR-ABL fusion gene.1 Several lines of evidence point to deregulated expression of the BCR-ABL tyrosine kinase as the initial genomic lesion in CML.1

Despite the presence of a consistent genetic abnormality, however, CML patients display considerable clinical heterogeneity, the basis of which is poorly understood. This heterogeneity was well characterized by Sokal et al.2 and is reflected 24 years later by the varying responses to treatment in chronic phase patients treated with a tyrosine kinase inhibitor.3 We therefore used a novel ultra-high-resolution genomic screening assay to search for additional acquired genomic abnormalities that might explain this clinical heterogeneity and help to assess prognosis for individual patients.

DNA was extracted from the polymorphonuclear cells in bone marrow samples from 10 previously untreated chronic phase patients. These patients subsequently received imatinib and achieved complete cytogenetic responses, at which point further polymorphonuclear-derived DNA was prepared. Comparative genomic hybridization (CGH) was performed with a 2.1 million oligonucleotide array (NimbleGen, Milton Keynes, UK; ‘HD2’ 070713_HG18_WG_CGH_HX1 design). The probes on this array were selected to achieve a uniform distribution throughout the genome, with approximately one probe every 1200 bp. Each DNA sample from diagnosis was competitively hybridized against the same patient's remission sample, which avoided detection of constitutional polymorphic copy number variants and limited results to acquired leukemia-related changes. Scanned array images were imported into NimbleScan (NimbleGen) to identify copy number aberrations (CNAs) from HD2 image and intensity data. Nexus 3 software (BioDiscovery Inc., El Segundo, CA, USA) was used to visualize the normalized segmented data. For representative CNAs the CGH result was confirmed by fluorescence in situ hybridization or quantitative real-time PCR.

All 10 CML patient samples harboured detectable genomic imbalances with an average of 53 CNAs per patient (range: 4–166). Of the 530 CNAs detected 381 (72%) were amplifications and 149 (28%) were deletions. Two hundred and fifty two CNAs (48%) involved at least one known gene. Many of the CNAs that involved single genes contained the complete gene with only small quantities of adjacent non-coding DNA. The average size of CNAs was 103 kb (range 9 kb–2 Mb). Seventy different genomic regions were aberrant in two or more patients in the cohort. Of these recurrent CNAs, amplifications of the DUSP1 and PBEF1 genes were most frequently detected; they were present in four and eight patients respectively. Other genes amplified or deleted in more than one patient included DUSP22, MAPK8IP1, MAP3K11, SUPT4H1, PTPRC, GRK6 and several members of the histone gene family.

The HD2 platform provides a sensitivity that is at least an order of magnitude greater than that of those employed in previous studies. Brazma et al.4 used a bacterial artificial chromosome array with a much lower resolution (1 Mb) and discovered 14 common cryptic abnormalities in CML blast crisis samples that were rare in chronic phase. In a comparison between Ph-positive acute lymphoblastic leukaemia and CML, Mullighan et al.5 used a higher-resolution 250 k SNP array and observed up to 8 CNAs per patient (range 0–8) in a cohort of 23 chronic phase CML samples, but no recurrent aberrations were detected and the authors did not specify whether samples had been taken at diagnosis or later in chronic phase. Exclusive analysis of presentation samples in our study demonstrates that many CNAs are early events in CML. Furthermore, the use of the patients' remission DNA as reference material rather than pooled normal DNA confirms all observed imbalances as disease-related.

The findings of our ultra-high-resolution screening have numerous implications. The presence of multiple genomic lesions at diagnosis supports the notion of an increased level of genomic instability in CML cells and raises the possibility that one or more aberrations in addition to BCR-ABL may dictate the CML phenotype. The considerable range in the number of CNAs present at diagnosis would be consistent with a differing level of genomic instability between individual patients. Assessment of the correlation between number of CNAs detected at diagnosis and survival in an unselected group of patients might show that high-resolution genome profiling was a good method for predicting clinical outcome in CML.

The observation that many CNAs are demarcated by the extremities of individual genes, suggests that these genes might be specific targets. Furthermore, the presence of same genetic imbalances in more than one patient in the cohort suggests a role for these genes in the pathogenesis or progression of CML. The most commonly involved gene, PBEF1, has a number of functions including roles in neutrophil proliferation6 and in the NAD anti-apoptotic pathway.7 Expression of PBEF1 is modulated by JUN-B,8 which is commonly downregulated in CML cells.1 It is therefore plausible that PBEF1 could be involved in the pathogenesis of CML. DUSP1, overrepresented in 4 of our 10 patients, might be a cooperating gene as it regulates mitogen-activated protein kinase (MAPK) activity and is itself regulated by p53 in stress responses.9 Further study is necessary to elucidate the precise roles of these and other genes commonly subject to copy number imbalance in the CML genome.

As with most somatic rearrangements, the mechanism by which CNAs occur is unclear, but low-copy repeats such as Alu sequences, and segmental duplications (duplicons) might facilitate aberrant homologous recombination.10 Interestingly, small duplicons are present within the 3′ end of PBEF1 and 5 kb downstream of DUSP1. This raises the possibility that duplicons may play a role in the abnormal amplification of these genes; the existence of similar duplicons close to the normal ABL and BCR genes was recently postulated as a possible mechanism underlying the formation of the t(9;22).11


  1. 1

    Melo JV, Barnes DJ . Chronic myeloid leukaemia as a model of disease evolution in human cancer. Nat Rev Cancer 2007; 7: 441–453.

  2. 2

    Sokal JE, Cox EB, Baccarani M, Tura S, Gomez GA, Robertson JE et al. Prognostic discrimination in ‘good-risk’ chronic granulocytic leukemia. Blood 1984; 63: 789–799.

  3. 3

    Druker BJ, Guilhot F, O′Brien SG, Gathmann I, Kantarjian H, Gattermann N et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 2006; 355: 2408–2417.

  4. 4

    Brazma D, Grace C, Howard J, Melo JV, Holyoke T, Apperley JF, Nacheva EP . Genomic profile of chronic myelogenous leukemia: Imbalances associated with disease progression. Genes Chromosomes Cancer 2007; 46: 1039–1050.

  5. 5

    Mullighan CG, Miller CB, Radtke I, Phillips LA, Dalton J, Ma J et al. BCR-ABL1 lymphoblastic leukaemia is characterized by the deletion of Ikaros. Nature 2008; 453: 110–114.

  6. 6

    Jia SH, Li Y, Parodo J, Kapus A, Fan L, Rotstein OD, Marshall JC . Pre-B cell colony-enhancing factor inhibits neutrophil apoptosis in experimental inflammation and clinical sepsis. J Clin Invest 2004; 113: 1318–1327.

  7. 7

    Revollo JR, Körner A, Mills KF, Satoh A, Wang T, Garten A et al. Nampt/PBEF/Visfatin regulates insulin secretion in beta cells as a systemic NAD biosynthetic enzyme. Cell Metab 2007; 6: 363–375.

  8. 8

    Kendal CE, Bryant-Greenwood GD . Pre-B-cell colony-enhancing factor (PBEF/Visfatin) gene expression is modulated by NF-kappaB and AP-1 in human amniotic epithelial cells. Placenta 2007; 28: 305–314.

  9. 9

    Liu YX, Wang J, Guo J, Wu J, Lieberman HB, Yin Y . DUSP1 is controlled by p53 during the cellular response to oxidative stress. Mol Cancer Res 2008; 6: 624–633.

  10. 10

    Lupski JR, Stankiewicz P . Genomic disorders: molecular mechanisms for rearrangements and conveyed phenotypes. PLoS Genet 2005; 1: e49.

  11. 11

    Saglio G, Storlazzi CT, Giugliano E, Surace C, Anelli L, Rege-Cambrin G et al. A 76-kb duplicon maps close to the BCR gene on chromosome 22 and the ABL gene on chromosome 9: possible involvement in the genesis of the Philadelphia chromosome translocation. Proc Natl Acad Sci USA 2002; 99: 9882–9887.

Download references

Author information

Correspondence to A G Reid.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Khorashad, J., De Melo, V., Fiegler, H. et al. Multiple sub-microscopic genomic lesions are a universal feature of chronic myeloid leukaemia at diagnosis. Leukemia 22, 1806–1807 (2008) doi:10.1038/leu.2008.210

Download citation

Further reading