Distinct factors determine the kinetics of disease relapse in adults transplanted for acute myeloid leukaemia.

BACKGROUND: Disease recurrence remains the major cause of death in adults with acute myeloid leukaemia (AML) treated using either intensive chemotherapy (IC) or allogenic stem cell transplantation (allo-SCT). AIMS: The timely delivery of maintenance drug or cellular therapies represent emerging strategies with the potential to reduce relapse after both treatment modalities, but whilst the determinants of overall relapse risk have been extensively characterized the factors determining the timing of disease recurrence have not been characterized. MATERIALS AND METHODS: We have therefore examined, using a series of sequential landmark analyses, relapse kinetics in a cohort of 2028 patients who received an allo-SCT for AML in CR1 and separately 570 patients treated with IC alone. RESULTS: In the first 3 months after allo-SCT, the factors associated with an increased risk of relapse included the presence of the FLT3-ITD (P < 0.001), patient age (P = 0.012), time interval from CR1 to transplant (P < 0.001) and donor type (P = 0.03). Relapse from 3 to 6 months was associated with a higher white cell count at diagnosis (P = 0.001), adverse-risk cytogenetics (P < 0.001), presence of FLT3-ITD mutation (P < 0.001) and time interval to achieve first complete remission (P = 0.013). Later relapse was associated with adverse cytogenetics, mutated NPM1, absence of chronic graft-versus-host disease (GVHD) and the use of in vivo T-cell depletion. In patients treated with IC alone, the factors associated with relapse in the first 3 months were adverse-risk cytogenetics (P < 0.001) and FLT3-ITD status (P = 0.001). The factors predicting later relapse were the time interval from diagnosis to CR1 (P = 0.22) and time interval from CR1 to IC (P = 0.012). DISCUSSION AND CONCLUSION: Taken together, these data provide novel insights into the biology of disease recurrence after both allo-SCT and IC and have the potential to inform the design of novel maintenance strategies in both clinical settings.

Addition of the nuclear export inhibitor selinexor to standard intensive treatment for elderly patients with acute myeloid leukemia and high risk myelodysplastic syndrome.

Treatment results of AML in elderly patients are unsatisfactory. In an open label randomized phase II study, we investigated whether addition of the XPO1 inhibitor selinexor to intensive chemotherapy would improve outcome in this population. 102 AML patients > 65 years of age (median 69 (65-80)) were randomly assigned to standard chemotherapy (3 + 7) with or without oral selinexor 60 mg twice weekly (both arms n = 51), days 1-24. In the second cycle, cytarabine 1000 mg/m2 twice daily, days 1-6 with or without selinexor was given. CR/CRi rates were significantly higher in the control arm than in the investigational arm (80% (95% C.I. 69-91%) vs. 59% (45-72%; p = 0.018), respectively). At 18 months, event-free survival was 45% for the control arm versus 26% for the investigational arm (Cox-p = 0.012) and overall survival 58% vs. 33%, respectively (p = 0.009). AML and infectious complications accounted for an increased death rate in the investigational arm. Irrespective of treatment, MRD status after two cycles appeared to be correlated with survival. We conclude that the addition of selinexor to standard chemotherapy does negatively affect the therapeutic outcome of elderly AML patients. (Netherlands Trial Registry number NL5748 (NTR5902), www.trialregister.nl ).

Correction: Lenalidomide added to standard intensive treatment for older patients with AML and high-risk MDS.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Lenalidomide added to standard intensive treatment for older patients with AML and high-risk MDS.

More effective treatment modalities are urgently needed in patients with acute myeloid leukemia (AML) of older age. We hypothesized that adding lenalidomide to intensive standard chemotherapy might improve their outcome. After establishing a safe lenalidomide, dose elderly patients with AML were randomly assigned in this randomized Phase 2 study (n = 222) to receive standard chemotherapy ("3 + 7") with or without lenalidomide at a dose of 20 mg/day 1-21. In the second cycle, patients received cytarabine 1000 mg/m2 twice daily on days 1-6 with or without lenalidomide (20 mg/day 1-21). The CR/CRi rates in the two arms were not different (69 vs. 66%). Event-free survival (EFS) at 36 months was 19% for the standard arm versus 21% for the lenalidomide arm and overall survival (OS) 35% vs. 30%, respectively. The frequencies and grade of adverse events were not significantly different between the treatment arms. Cardiovascular toxicities were rare and equally distributed between the arms. The results of the present study show that the addition of lenalidomide to standard remission induction chemotherapy does not improve the therapeutic outcome of older AML patients. This trial is registered as number NTR2294 in The NederlandsTrial Register (www.trialregister.nl).

Long-term outcome of older patients with newly diagnosed de novo acute promyelocytic leukemia treated with ATRA plus anthracycline-based therapy.

Treatment outcome in older patients with acute promyelocytic leukemia (APL) is lower compared with younger patients, mainly because of a higher induction death rate and postremission non-relapse mortality (NRM). This prompted us to design a risk- and age-adapted protocol (Programa Español de Tratamientos en Hematología (PETHEMA)/HOVON LPA2005), with dose reduction of consolidation chemotherapy. Patients aged ⩾60 years reported to the PETHEMA registry and were treated with all-trans retinoic acid (ATRA) plus anthracycline-based regimens according to three consecutive PETHEMA trials that were included. We compared the long-term outcomes of the LPA2005 trial with the preceding PETHEMA trials using non-age-adapted schedules (LPA96&LPA99). From 1996 to 2012, 389 older patients were registered, of whom 268 patients (69%) were eligible. Causes of ineligibility were secondary APL (19%), and unfit for chemotherapy (11%). Median age was 67 years, without relevant differences between LPA2005 and LPA96&LPA99 cohorts. Overall, 216 patients (81%) achieved complete remission with no differences between trials. The 5-year NRM, cumulative incidence of relapse, disease-free survival and overall survival in the LPA2005 vs the LPA96&99 were 5 vs 18% (P=0.15), 7 vs 12% (P=0.23), 87 vs 69% (P=0.04) and 74 vs 60% (P=0.06). A less intensive front-line regimen with ATRA and anthracycline monochemotherapy resulted in improved outcomes in older APL patients.

An in vitro model for cytogenetic conversion in CML. Interferon-alpha preferentially inhibits the outgrowth of malignant stem cells preserved in long-term culture.

IFN-alpha has been shown to prolong survival in chronic myeloid leukemia patients, but its mechanism of action is still not understood. The human cobblestone area-forming cell (CAFC) assay allows for the measurement of the concentration of normal as well as malignant stem cells, while their progeny can be measured in parallel long-term culture (LTC) in flasks. Using CAFC and LTC assays, we have examined direct effects of IFN-alpha (500; 5,000 IU/ml) on the maintenance and outgrowth of CD34-enriched normal and malignant stem cells, obtained from six patients with an established major cytogenetic response to IFN-alpha and from four nonresponding patients. CAFC concentrations were not affected by IFN-alpha. In contrast, IFN-alpha strongly inhibited the clonogenic output in flask LTC. Nucleated cells (NC) produced in LTC were evaluated by fluorescent in situ hybridization (FISH) for the presence of the Philadelphia (Ph) translocation. After 8 wk of LTC, the percentage of Ph+ NCs produced was significantly more inhibited by IFN-alpha in responding patients than in nonresponders. Control LTC without IFN-alpha showed no significant differences of Ph+ NC production between responders and nonresponders. These findings provide the first in vitro model for cytogenetic conversion and suggest that direct antiproliferative effects of IFN-alpha account for the cytogenetic response observed clinically.

Tumor necrosis factor downregulates granulocyte-colony-stimulating factor receptor expression on human acute myeloid leukemia cells and granulocytes.

Tumor necrosis factor (TNF) inhibits granulocyte-colony-stimulating factor (G-CSF)-induced human acute myeloid leukemia (AML) growth in vitro. Incubation of blasts from three patients with AML in serum-free medium with TNF (10(3) U/ml), and subsequent binding studies using 125I-G-CSF reveal that TNF downregulates the numbers of G-CSF receptors by approximately 70%. G-CSF receptor numbers on purified blood granulocytes are also downmodulated by TNF. Downregulation of G-CSF receptor expression becomes evident within 10 min after incubation of the cells with TNF at 37 degrees C and is not associated with an apparent change of the dissociation constant (Kd). The TNF effect does not occur at 0 degrees C and cannot be induced by IL-2, IL-6, or GM-CSF. TNF probably exerts its effect through activation of protein kinase C (PKC) as the TNF effect on G-CSF receptor levels can be mimicked by 12-O-tetradecanoylphorbol-13- acetate. The PKC inhibitor Staurosporine (Sigma Chemical Co., St. Louis, MO) as well as protease inhibitors can completely prevent G-CSF receptor downmodulation. Thus, it appears TNF may act as a regulator of G-CSF receptor expression in myeloid cells and shut off G-CSF dependent hematopoiesis. The regulatory ability of TNF may explain the antagonism between TNF and G-CSF stimulation.

High-vs low-dose cytarabine combined with interferon alfa in patients with first chronic phase chronic myeloid leukemia. A prospective randomized phase III study.

A prospective randomized phase III study was performed to evaluate whether intensified cytarabine would induce a higher response rate and longer event-free interval as compared to low-dose cytarabine in chronic myeloid leukemia (CML). One hundred and eighteen patients with CML in early chronic phase entered the study. Twenty-eight out of 32 patients assigned to group A received two cycles of a combination of intensified cytarabine and idarubicin followed by interferon alfa (IFN-alpha) maintenance, 28 patients in group B received standard treatment by a combination of low-dose cytarabine and IFN-alpha. Forty-nine patients with a human leukocyte antigen-identical sibling donor proceeded to allogeneic stem cell transplantation (allo-SCT) and nine patients were excluded from the analysis. Hematological response was observed in 97% of the patients in group A vs 86% of the patients in group B during the first year of treatment. In group A, 16 patients (50%) achieved a major cytogenetic response, which compared to seven patients (25%) with a major cytogenetic response in group B. With a median follow-up of 58 months (range 34-76), event-free survival was not significantly different between arms A and B. The estimated 5-year survival rate was 56% in the intensified arm and 77% in the low-dose arm (P = 0.05). Recipients of allo-SCT showed a 5-year estimated survival rate of 55%. Although intensified cytarabine induced a higher initial percentage of major and complete cytogenetic responses, responses were not sustained by IFN-alpha maintenance therapy.

Distinct cytoplasmic regions of the human granulocyte colony-stimulating factor receptor involved in induction of proliferation and maturation.

The granulocyte colony-stimulating factor receptor (G-CSF-R) transduces signals important for the proliferation and maturation of myeloid progenitor cells. To identify functionally important regions in the cytoplasmic domain of the G-CSF-R, we compared the actions of the wild-type receptor, two mutants, and a natural splice variant in transfectants of the mouse pro-B cell line BAF3 and two myeloid cell lines, 32D and L-GM. A region of 55 amino acids adjacent to the transmembrane domain was found to be sufficient for generating a growth signal. The immediate downstream sequence of 30 amino acids substantially enhanced the growth signaling in the three cell lines. In contrast, the carboxy-terminal part of 98 amino acids strongly inhibited growth signaling in the two myeloid cell lines but not in BAF3 cells. Truncation of this region lead to an inability of the G-CSF-R to transduce maturation signals in L-GM cells. An alternative carboxy tail present in a splice variant of the G-CSF-R also inhibited growth signaling, notably in both the myeloid cells and BAF3 cells, but appeared not to be involved in maturation.

Comparative value of post-remission treatment in cytogenetically normal AML subclassified by NPM1 and FLT3-ITD allelic ratio.

Post-remission treatment (PRT) in patients with cytogenetically normal (CN) acute myeloid leukemia (AML) in first complete remission (CR1) is debated. We studied 521 patients with CN-AML in CR1, for whom mutational status of NPM1 and FLT3-ITD was available, including the FLT3-ITD allelic ratio. PRT consisted of reduced intensity conditioning (RIC) allogeneic hematopoietic stem cell transplantation (alloHSCT) (n=68), myeloablative conditioning (MAC) alloHSCT (n=137), autologous hematopoietic stem cell transplantation (autoHSCT) (n=168) or chemotherapy (n=148). Favorable overall survival (OS) was found for patients with mutated NPM1 without FLT3-ITD (71±4%). Outcome in patients with a high FLT3-ITD allelic ratio appeared to be very poor with OS and relapse-free survival (RFS) of 23±8% and 12±6%, respectively. Patients with wild-type NPM1 without FLT3-ITD or with a low allelic burden of FLT3-ITD were considered as intermediate-risk group because of similar OS and RFS at 5 years, in which PRT by RIC alloHSCT resulted in better OS and RFS as compared with chemotherapy (hazard ratio (HR) 0.56, P=0.022 and HR 0.50, P=0.004, respectively) or autoHSCT (HR 0.60, P=0.046 and HR 0.60, P=0.043, respectively). The lowest cumulative incidence of relapse (23±4%) was observed following MAC alloHSCT. These results suggest that alloHSCT may be preferred in patients with molecularly intermediate-risk CN-AML, while the choice of conditioning type may be personalized according to risk for non-relapse mortality.

T-lymphocyte reconstitution following rigorously T-cell-depleted versus unmodified autologous stem cell transplants.

We compared the kinetics of T-cell recovery after extensive ex vivo and in vivo T-cell depleted autologous stem cell transplantation (SCT) for multiple sclerosis (MS; n=8) with unmodified SCT for hematological malignancies (HM; n=39). Both patient group showed a very protracted recovery of 'naive' CD4(+), 45R0(-) ( approximately CD45RA(+)) T-cells. Within the 'primed' CD4(+), 45R0(+) T-cells, the 'central memory' cells expressing the CD62L and CD27 markers were the slowest to recover. The repopulating T-cells were highly activated, as shown by increased expression of HLA-DR and the apoptosis marker CD95. The capability of CD4(+) and CD8(+) T-cells to produce IFN-gamma, IL-2 and TNF-alpha had reached normal ranges from 2 months post SCT onwards. Unexpectedly, the kinetics of T-cell recovery between 3 and 12 months post transplant was similar in T-depleted and unmodified SCT. Before SCT, the HM patients showed lymphopenia of all T-cell subsets, upregulated HLA-DR and CD95 expression and increased cytokine responses. We suggest that the similar kinetics of T-cell recovery in the two patient groups may be explained by the susceptibility to apoptosis of the activated CD4(+) T-cells in the autografts of the HM patients. This susceptibility to apoptosis would interfere with a swift and sustained CD4(+) T-cell regeneration post SCT.

ID1 and ID2 are retinoic acid responsive genes and induce a G0/G1 accumulation in acute promyelocytic leukemia cells.

Acute promyelocytic leukemia (APL) is uniquely sensitive to treatment with all-trans retinoic acid (ATRA), which results in the expression of genes that induce the terminal granulocytic differentiation of the leukemic blasts. Here we report the identification of two ATRA responsive genes in APL cells, ID1 and ID2. These proteins act as antagonists of basic helix-loop-helix (bHLH) transcription factors. ATRA induced a rapid increase in ID1 and ID2, both in the APL cell line NB4 as well as in primary patient cells. In addition, a strong downregulation of E2A was observed. E2A acts as a general heterodimerization partner for many bHLH proteins that are involved in differentiation control in various tissues. The simultaneous upregulation of ID1 and ID2, and the downregulation of E2A suggest a role for bHLH proteins in the induction of differentiation of APL cells following ATRA treatment. To test the relevance of this upregulation, ID1 and ID2 were overexpressed in NB4 cells. Overexpression inhibited proliferation and induced a G0/G1 accumulation. These results indicate that ID1 and ID2 are important retinoic acid responsive genes in APL, and suggest that the inhibition of specific bHLH transcription factor complexes may play a role in the therapeutic effect of ATRA in APL.

Nature of the delayed graft-versus-host reactivity of fetal liver cell transplants in mice.

Experiments were designed to determine which actual differences in the cellular composition between fetal liver and bone marrow account for the distinct types of graft-versus-host (GvH) disease. The assay of reactive lymphocytes (by in vitro mitogenic stimulation) in fetal liver transplants in mice, the purification of hemopoietic stem cells (HSC) of the transplants, and the quantitation of HSC numbers in the grafts traced the basis for the distinctly weak type of GvH disease after fetal liver cell grafts. It was found that transplantation of purified HSC concentrates did not modify the severity of GvH mortality. The moderate character of the delayed GvH disease was shown to depend on the presence of an HSC population in fetal liver with different qualities and not on numerical differences between the HSC in fetal liver and bone marrow. Data collected also demonstrated that when GvH disease occurred in the recipients of transplants of fetal liver, it shared the characteristic histologic features of the bone marrow GvH syndrome. The recovery of mitogen responsiveness of spleen cells may have been delayed in fetal liver allotransplantation as compared to syngeneic grafting. By supportive infusion of lymphoid cells, it was suggested that the immunodeficiency coinciding with GvH disease represented a secondary manifestation of the disease rather than a primary impairment in lymphoid differentiation.

[Views on the financing of biomedical scientific research; the report 'Erop of eronder' (Sink or swim) from the KNAW (Royal Netherlands Academy of Arts and Sciences)]. / Visie op de financiering van het biomedisch wetenschappelijk onderzoek; het KNAW-rapport 'Erop of eronder'.

In a recent report, entitled 'Erop of eronder. Financiering van (bio)medisch wetenschappelijk onderzoek' [Sink or swim. The financing of (bio)medical scientific research], the Royal Netherlands Academy of Arts and Sciences has reviewed various aspects of the current financing system. The recommendations concern a funding process with emphasis on open programmes focusing on researcher-initiated research, creating an earmarked budget for matching obligations, creating technology transfer centres offering broad expertise for the validation ofthe scientific results, and giving priority to the establishment of a European Research Council that encourages excellence in fundamental research within the European research area.

Feeder cell requirements for leukemia cell colony formation in cultures supplemented with phytohemagglutinin.

The use of phytohemagglutinin-supplemented colony cultures has offered new opportunities recently for studying acute myeloid leukemia (AML) cell growth in vitro. The active stimulator cells for AML colony-forming cells have not been identified, although this could be important for optimal application of the technique and for elucidating differences in growth between normal and leukemic progenitor cells. In this study, feeder layers were prepared from subpopulations of normal peripheral blood leukocytes which were obtained by centrifugation through Ficoll-Isopaque, erythrocyte rosette sedimentation, and adherence separation. Underlayers containing lymphocytes (B, T, or B plus T) or adherent monocytes failed to stimulate AML colony formation. The colony stimulation capacity of total mononuclear cells was decreased significantly following depletion of T-lymphocytes. The highest AML colony numbers were obtained when adherent monocytes and T-lymphocytes in combination were added to phytohemagglutinin-containing cultures. Stimulation of AML colony formation depended on the quantitative interrelationship of monocytes and T-lymphocytes in the cultures. Thus, AML colony-forming cells, unlike normal marrow granulocyte-monocyte colony-forming cells, do not respond to monocyte stimulation alone and require for their proliferation an inducing factor derived from phytohemagglutinin-exposed T-lymphocytes and monocytes.

Influence of interferon preparations on the proliferative capacity of human and mouse bone marrow cells in vitro.

The toxicity of interferon to bone marrow was studied by the use of in vitro colony forming assays for hemopoietic cells. In the same study the relative inhibitory effects of two clinically common interferon preparations, leukocyte and fibroblast interferons, were compared with regard to their effect on both myeloid [colony-forming unit, culture (CFUc)] and erythroid [colony-forming unit, erythroid (CFUe)] progenitor cells. CFUe formation in human bone marrow cells in vitro appeared to be fairly resistant to both interferons. Only high doses of both interferons gave a marked inhibition of CFUe. However, the toxicity of leukocyte and fibroblast interferon was divergent for CFUe in human bone marrow. Leukocyte interferon appeared to be considerably more inhibitory for CFUe than was fibroblast interferon. The effects of mouse interferon, induced in L929 cells, on the growth of CFUc and CFUe in murine bone marrow cells were comparable with those of fibroblast interferon on human cells. The toxicity of human and murine interferon was species specific. Except for the toxicity of leukocyte interferon to CFUc in human bone marrow, the toxicity of interferon was marked only with concentrations on interferon far exceeding the amount necessary to produce an antiviral state in vitro.

Expression of a passenger miR-9* predicts favorable outcome in adults with acute myeloid leukemia less than 60 years of age.

In double-stranded miRNA/miRNA* duplexes, one of the strands represents an active miRNA, whereas another, known as a passenger strand (miRNA*), is typically degraded. MiR-9* is not detectable in normal myeloid cells. Here we show that miR-9* is expressed in 59% of acute myeloid leukemia (AML) cases and we investigate its clinical impact in 567 adults with de novo AML (age⩽60 years). AML cases with detectable miR-9* included a lower percentage of cases with favorable risk (P<0.001) as compared with those with no detectable miR-9*. High levels of miR-9* expression independently predicted for higher complete remission (odds ratio=1.28, P=0.013) and better event-free survival (EFS) (hazard ratio (HR)=0.86, P=0.001), relapse-free survival (RFS) (HR=0.84, P=0.008) and overall survival (OS) (HR=0.86, P=0.002). Among the subgroup of adverse risk patients, high miR-9* expressers had strikingly longer median survival than low miR-9* expressers (EFS: 16 vs 5 months, P=0.020; RFS: 12 vs 4, P=0.060; OS: 23 vs 8, P=0.021). Comparative transcriptome analysis suggests that miR-9* regulates genes involved in leukemogenesis, for example, MN1 and MLLT3. This is the first report showing that an miRNA* has prognostic value in AML.

OC-10 - Disseminated intravascular coagulation at diagnosis strongly predicts both arterial and venous thrombosis in acute myeloid leukemia patients.

INTRODUCTION: Acute myeloid leukemia (AML) is associated with a slightly increased risk of VTE with an incidence of 1.7-8.9%, but only limited data are available. The mechanism of the occurrence of thrombosis in hematological disorders is still unresolved. Disseminated intravascular coagulation (DIC) is associated with VTE and bleeding in acute promyelocytic leukemia and acute lymphoblastic leukemia. Although DIC has also been reported in AML, no data exist on the relationship between DIC and VTE in AML patients. AIM: We hypothesized that the presence of DIC at diagnosis of AML may contribute to the risk of both venous and arterial thrombosis in AML. Therefore we studied a large cohort of adult patients with newly diagnosed AML aged <65 years by measuring DIC parameters at diagnosis prior to treatment and assessing the occurrence of both venous and arterial thrombosis during follow up. The findings of this study were validated in a second large cohort of patient with newly diagnosed AML aged >60 years. MATERIALS AND METHODS: In a prospective study we analysed markers of DIC and their association with the occurrence of thrombosis during follow up in a cohort of 272 young AML patients (aged 18-65) and a validation cohort of 132 elderly AML patients (aged >60) patients that were all treated with intensive chemotherapy. DIC parameters (fibrinogen, D-dimer, alpha-2-antiplasmin, antitrombin, prothrombin time and platelets) were measured at presentation with AML before start of induction chemotherapy. The DIC score according to the International Society of Thrombosis and Haemostasis DIC scoring systemwas calculated of all patients. RESULTS: The prevalence of thrombosis was 8.7% (4.7% venous, 4.0% arterial) in young patients over a median follow up of 478 days and 10.4% (4.4% venous, 5.9% arterial) in elderly patients. Most thrombotic events (66%) occurred before start of the second course of chemotherapy. The calculated DIC score significantly predicted venous and arterial thrombosis with a Hazard Ratio (HR) for a high DIC score (=>5) of 4.79 (1.71-13.45) in the cohort of young AML patients. These results were confirmed in our validation cohort of elderly AML patients. (HR 11.08 (3.23-38.06)). Of all DIC parameters D-dimer levels are most predictive for thrombosis with a HR of 12.3 (3.39-42.64) in the cohort of young AML patients and a HR of 7.82 (1.95-31.38) in the elderly cohort for a D-dimer >4.0 mg/L. CONCLUSIONS: It is concluded that both venous and arterial thrombosis occurs in around 10% of AML patients treated with intensive chemotherapy, which can be predicted by the presence of DIC, or individual DIC parameters at time of AML diagnosis.

Aberrant expression of miR-9/9* in myeloid progenitors inhibits neutrophil differentiation by post-transcriptional regulation of ERG.

Aberrant post-transcriptional regulation by microRNAs (miRNAs) has been shown to be involved in the pathogenesis of acute myeloid leukemia (AML). In a previous study, we performed a large functional screen using a retroviral barcoded miRNA expression library. Here, we report that overexpression of miR-9/9* in myeloid 32D cell line (32D-miR-9/9*) had profound impact on granulocyte colony-stimulating factor-induced differentiation. Further in vitro studies showed that enforced expression of miR-9/9* blocked normal neutrophil development in 32D and in primary murine lineage-negative bone marrow cells. We examined the expression of miR-9/9* in a cohort of 647 primary human AMLs. In most cases, miR-9 and miR-9* were significantly upregulated and their expression levels varied according to AML subtype, with the highest expression in MLL-related leukemias harboring 11q23 abnormalities and the lowest expression in AML cases with t(8;21) and biallelic mutations in CEBPA. Gene expression profiling of AMLs with high expression of miR-9/9* and 32D-miR-9/9* identified ETS-related gene (Erg) as the only common potential target. Upregulation of ERG in 32D cells rescued miR-9/9*-induced block in neutrophil differentiation. Taken together, this study demonstrates that miR-9/9* are aberrantly expressed in most of AML cases and interfere with normal neutrophil differentiation by downregulation of ERG.