NME1 and NME2 as markers for myeloid leukemias

The identification of molecular markers associated with good or poor prognosis has dramatically impacted the management and treatment of patients with acute myeloid leukemia (AML). In the most recent version of the World Health Organization (WHO) classification, over half of AML subtypes can be categorized based on the presence of specific genetic defects [1,2]. FLT3 ITD, NPM1 and bi-allelic C/EBPα mutations are frequently sought, and affect the prognosis and therapeutic plan of patients with AML who have normal cytogenetics (CN-AML) [3]. The prognostic relevance of the c-kit mutation is still being explored. The presence of these mutations affects the recommendation for allogeneic stem cell transplant (SCT) in first remission. Likewise, some recommend assessing for the c-kit mutation in the core binding factor (CBF) leukemias, as the traditionally favorable subset of AML is now classified as less favorable in the presence of the c-kit mutation, and consideration should be given for allogeneic SCT in first remission or a tyrosine kinase inhibitor-containing clinical trial [3]. There is also evolving interest in assessing for mutations in IDH1/2, MLL-PTD, WT-1 and DNMT3A and alterations in expression of BAALC and EVI1 [3,4]. It was very recently reported that high-dose daunorubicin is more efficacious than the standard dose of the drug, in terms of overall survival among patients with DNMT3A or NPM1 mutations or MLL translocations [4]. Inhibitors against specific mutations, most notably FLT3 ITD, are being developed. Thus, the presence of specific molecular aberrations can have an impact at multiple steps in the management of patients with AML. It is particularly important in defining the prognosis of patients with AML, and may drive decisions on whether stem cell transplant in first remission should be employed. It can also affect decisions regarding the dose of anthracycline in induction therapy, and potentially promote the use of targeted therapies as an adjunct to chemotherapy.

Identifying new ways to further refine prognosis remains of great importance and of high clinical-translational relevance. In this issue of the journal, the Cross group at the University of Leipzig further refine the roles of NME1 and NME2 in myeloid leukemias. The human NME genes localize to chromosome 17q21.6 and encode enzymes that are responsible for the majority of nucleoside diphosphate kinase activity in mammals [5–7]. Early work has shown that the expression of these genes is inversely correlated with the metastatic potential of experimental rodent cells and some human tumors [8]. Because of their known metastatic potential, there has been interest in assessing the role of these genes in human leukemias. Yokoyama et al. have shown that that the NME1/2 genes are overexpressed in AML, especially in AML-M5, and that the levels of NME1 gene expression correlate with a poor prognosis in AML [9]. Since the 1990s, when the potential role of the NME genes in AML was first recognized, the prognostic stratification of AML has grown, and includes well-defined molecular aberrations in addition to chromosome abnormalities; as described above [10]. However, the recognition that gene expression is routinely measured under optimal steady-state conditions of freshly aspirated material at the time of presentation, while the phenotypic response of the leukemia cells to therapy may influence prognosis, prompted this group to consider that gene expression patterns relevant to outcome may be more apparent when the cells are under stress than when they are analyzed immediately after the cells are obtained from a newly diagnosed patient with AML.

As reported in this issue of the journal, Bach et al. found a prognostic pattern of NME mRNA when the bone marrow was transported and therefore subjected to stress [11]. In prior work, as discussed above, high NME mRNA expression has been associated with poor prognosis in AML [9]. Although Bach did not observe that NME1 and NME2 mRNA levels in fresh bone marrows were prognostic in CN-AML, they found that they were so in bone marrow samples that had not been freshly processed [11]. Over a period of 2 days or more, NME1 mRNA was decreased relative to an internal control in all samples, while NME2 mRNA decreased in some and was maintained in others, suggesting that it may be useful as a potential prognostic factor. In fact, Bach et al. found that maintenance of high NME2 mRNA was an indicator of increased event-free survival (EFS), independent of FLT3 ITD status [11]. Should these results prompt further studies to determine whether delayed processing should be utilized for assessing other mutations? Will the results of this work ultimately change the way samples are procured and processed? Clearly, before any such considerations, the results of this work must be reproduced in other large studies.

Also in this issue of the journal, the same group report the results of studies examining the functional role of NME2 in chronic lymphocytic leukemia (CML). NME2 is a key transcriptional activator of the c-myc gene [12,13]. As c-myc expression has been previously shown to contribute to CML progression [14,15], the authors hypothesized that altered NME2 expression may be a functionally relevant feature of CML. They found that NME2 protein was overexpressed in samples from patients with CML at diagnosis, while its expression was absent in peripheral blood cells from normal donors, the majority of patients with AML, and BCR–ABL negative lymphoid cells from patients with CML [16]. NME2 expression was high in peripheral blood cells of patients with imatinib-resistant CML and low in cells of responding patients [16]. One interpretation of these findings is simply that NME2 protein is just a surrogate marker for BCR–ABL and responsiveness to tyrosine kinase inhibitors, and unlikely to have independent prognostic implications. However, alternative explanations are possible, and further studies to define the mechanisms by which NME2 is regulated by BCR–ABL may be informative.

Altogether, this body of work reminds us of the potential importance of nucleoside diphosphate kinase activity in hematological malignancies and can prompt further questions as to what are the appropriate conditions to assess potential predictive molecular changes in hematological malignancies. It also remains to be defined why the patterns of NME1/2 expression are different in CML cells as compared to AML cells, and whether such differences reflect leukemic cell arrest at different levels of hematopoietic cell differentiation.

Potential conflict of interest

Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.