Original Article

Leukemia (2011) 25, 615–621; doi:10.1038/leu.2010.299; published online 14 January 2011

Acute Leukemias

Characterization of NPM1-mutated AML with a history of myelodysplastic syndromes or myeloproliferative neoplasms

S Schnittger1, U Bacher2, C Haferlach1, T Alpermann1, F Dicker1, J Sundermann1, W Kern1 and T Haferlach1

  1. 1MLL Munich Leukemia Laboratory, Munich, Germany
  2. 2Department for Stem Cell Transplantation, University Cancer Center Hamburg, Hamburg, Germany

Correspondence: Dr S Schnittger, MLL Munich Leukemia Laboratory, Max-Lebsche-Platz 31, Munich 81377, Germany. E-mail: susanne.schnittger@mll.com

Received 21 June 2010; Revised 2 November 2010; Accepted 9 November 2010; Published online 14 January 2011.



The role of the nucleophosmin (NPM1) mutations in de novo acute myeloid leukemia (AML) is well analyzed, but the impact in secondary AML (s-AML) following myelodysplastic syndromes (MDS) or transformed myeloproliferative neoplasms (MPNs) remains unclear. We investigated 350 patients—283 s-AML after MDS and 67 transformed MPNs—for NPM1mut. NPM1mut was detected in 43/350 patients (12.3%) at diagnosis of s-AML (transformed MDS: 37/283; 13.1%; transformed MPNs: 6/67; 9.0%). Cytogenetic alterations were present in 12/40 cases (30.0%) with available karyotypes. Additional molecular mutations were found in 23/43 NPM1mut s-AML after MDS (53.5%) and in transformed MPN in 18/37 (48.6%): FLT3-ITD: 14/37 (37.8%); FLT3-TKD: 3/28 (10.7%); NRASmut: 4/37 (10.8%), RUNX1mut: 1/16 (6.3%). In NPM1mut-transformed MPNs, five out of six cases showed 1–2 additional molecular mutations (2 × KITD816V, ETV6-PDGFRB, 2 × JAK2V617F, 2 × FLT3-ITD). Backtracking of nine of these cases by quantitative real time PCR showed the NPM1mut already at diagnosis of MDS/MPN, at variable levels and up to 14 months before diagnosis of AML, and at transformation often being preceded or accompanied by other genetic alterations. Thus, NPM1 mutations are involved in the transformation from MDS to AML or MPN to blast phase in single cases, which should be further confirmed in larger studies.


NPM1 mutations; secondary acute myeloid leukemia; leukemogenesis; myelodysplastic syndrome; myeloproliferative neoplasms



Acute myeloid leukemia (AML) carrying mutations of the nucleophosmin (NPM1) gene account for about 35% of all cases (50–60% of AML with normal karyotype).1 NPM1 mutations are most frequently characterized by different four base-pair insertions and, although 80% of these mutations account for one single mutation type (type ‘A’), more than 40 different mutation sub-types have already been identified.1 The NPM1 mutation mediates aberrant localization of the nucleophosmin protein to the cytoplasm that is thought to interfere with the normal chaperone and shuttling functions of the native nucleophosmin protein,2 including regulation of ribosome biogenesis and control of the ARF-P53 tumor suppressor pathway.3, 4 Prognosis of de novo AML patients with isolated NPM1 mutations in normal karyotype has been outlined to be more favorable when compared with many other subgroups of AML. In contrast, presence of an additional FLT3-ITD is associated with worse outcomes.5, 6, 7, 8, 9 For all above reasons, AML with mutated NPM1 was included as a new provisional entity in the 2008 World Health Organization of myeloid neoplasms.10

NPM1 mutations were reported to be typical for de novo AML.11, 12 Falini et al.6 described a 35.2% mutation rate in 591 samples from patients with de novo AML, but identified no mutation carrier in 135 secondary AML (s-AML) cases. In other studies, the mutation was identified in s-AML cases, but with lower frequencies compared with de novo disease.5, 7, 13

Around 25% of patients with myelodysplastic syndromes (MDS) and variable proportions of patients with myeloproliferative neoplasms (MPN) show acute transformation with blasts >20% during the course of disease. This transformation is assumed to be a multistep process being mediated by the acquisition of different genetic events. Besides clonal evolution of cytogenetics several molecular mutations have been implicated in the transformation of MDS to AML, for example, mutations of the NRAS, FLT314, 15 and CEBPA genes,16 as well as the MLL-PTD.17

To our knowledge, no study focused so far specifically on the role of the NPM1 mutations in s-AML after MDS or acute transformation of MPNs. Thus, it remains to be clarified whether the NPM1 mutation has a role for the transformation from myeloid malignancies with blasts below 20% to s-AML or whether its function is restricted to the initiation of de novo AML. To address this issue, we performed analysis for NPM1 mutations in a total of 350 patients with s-AML after MDS (or chronic myelomonocytic leukemia (CMML), respectively) or acutely transformed MPNs, and analyzed the NPM1 mutation by backtracking, and also in the context of other molecular and cytogenetic alterations.



Our cohort included 283 s-AML patients with a history of MDS or CMML, and 67 patients with transformed MPNs. Bone marrow or peripheral blood samples were sent for diagnosis to the Munich Leukemia Laboratory (MLL) in the period from August 2005 to July 2009. This s-AML cohort was compared with a control cohort of 618 NPM1-mutated de novo AML being investigated in the MLL in the same period. Patients gave their informed consent to genetic analysis and use of data for research. The study was performed in accordance with the Declaration of Helsinki.

Investigation of bone marrow and/or peripheral blood samples were carried out as published before in parallel with cytomorphology,18 immunophenotyping, cytogenetics with chromosome banding analysis and fluorescence in situ hybridization,19 and PCR-based techniques.


Screening for the NPM1 mutations was carried out with a Lightcycler-based melting-curve assay at the time of diagnosis of the s-AML.7 Backtracking of the NPM1 mutation in the MDS and MPN samples was realized by mutation sub-type-specific quantitative real-time PCR (RQ-PCR) achieving high sensitivity from 10−4 to 10−6 as previously described.20 Samples with evidence of an NPM1 mutation were additionally analyzed for FLT3-ITD,21 FLT3-TKD,22 MLL-PTD,23 NRAS,24 RUNX1 (=AML1)17 and JAK2V617F25 mutations following previously described procedures. Selected cases were investigated for KIT mutations26 or PDGFRA/PDGFRB rearrangements.27

Bone marrow and peripheral blood samples underwent May Giemsa Gruenwald staining and cytochemistry with myeloperoxidase and nonspecific esterase. Cases with greater than or equal to5% of cup-like blasts in the peripheral blood and/or bone marrow were fulfilling the criterion of ‘cup-like morphology’ according to Kroschinsky et al.28 Chromosome banding analysis and fluorescence in situ hybridization were performed according to standard methods.19 Immunophenotyping was performed as previously described.29

Statistical analysis

Survival curves were calculated for overall survival (OS) and event-free survival (EFS) according to Kaplan–Meier method and compared using the two-sided log-rank test. OS was the time from diagnosis of AML to death or last follow-up. Event-free survival was defined as the time from diagnosis of AML to treatment failure, relapse, death or last follow-up. Relapse was defined according to Cheson et al.30 Time to transformation was calculated from the time point of first diagnosis of MDS or MPN until the time point of the diagnosis of s-AML. Cox regression analysis was performed for OS and event-free survival with different analyzed parameters as covariates. Parameters that were significant in univariate analysis were included into multivariate analysis. Dichotomous variables were compared between different groups using the χ2-test and continuous variables by Student's t-test. Spearman's rank correlation was used to analyze correlations between continuous parameters. For all analyses, results were significant at a level of P<0.05 at both sides. SPSS (version 14.0.1) software (SPSS, Chicago, IL, USA) was used for statistical analysis.



Biological parameters

The frequency of NPM1 mutations in the entire cohort was 43/350 patients (12.3%) at diagnosis of the s-AML or acute transformation of MPNs. This cohort was composed of 23 men and 20 women with a median age of 71.3 years (range: 29.3–87.7 years). The intervals between the first manifestation of the MDS or MPNs, respectively, were available in 25 cases: less than or equal to6 months: n=9 patients; 7–12 months: n=6; >12 months: n=10. The median interval to leukemic transformation was 11 months (range, from 6 weeks to 165 months). The patients from the control cohort consisting of 618 de novo NPM1-mutated AML patients (328 women, 290 men) had a significantly lower median age (63.8 years, 13.3–88.4 years; P<0.001) when compared with the NPM1-mutated s-AML patients.

Frequency and characterization of NPM1 mutations

NPM1 mutations were detected in 37/283 (13.1%) patients with s-AML following MDS or CMML (being classified as ‘s-AML’ for differentiation from the acutely transformed MPN cases). The initial diagnoses of the NPM1-mutated subgroup were as follows (information available for 33 out of 37 cases): refractory anemia with ring sideroblasts: n=1; refractory anemia with excess of blasts: n=4; RAEB-2: n=3; 15 cases were diagnosed with initial MDS subtypes without further subclassification. A total of 10 patients had a previous diagnosis of CMML. Information on the previous disease subtype was missing in four cases.

In the 67 patients with acutely transformed MPNs, 6 patients (9.0%) were identified as carriers of NPM1 mutation. These patients had a history of KITD816-mutated mastocytosis (n=2), atypical BCR-ABL1-negative CML (n=2), one of these with an ETV6-PDGFRB rearrangement, JAK2V617F-mutated polycythemia vera (n=1), or JAK2V617F-mutated primary myelofibrosis (n=1).

NPM1 type A mutations were identified in 30 patients (69.8% of all mutated cases). A total of 13 patients (30.2%) had rare mutations: B (n=4; 9.3%), D (n=5; 11.6%) and single rare types (n=4; 9.3%). In a control cohort of 618 NPM1-mutated de novo AML, type A mutations were identified in 474 cases (76.7%), type B in 41 (6.6%) and type D mutations in 30 cases (4.9%). Thus, non-A types were slightly more frequent in s-AML, but this did not reach statistical significance.

Characterization of additional molecular alterations

Molecular markers in addition to the NPM1 mutation were detected in 23/43 patients (53.5%) (Table 1). In more detail, one additional molecular marker was identified in 20 patients, whereas 3 patients had two additional molecular markers in combination. There was a higher frequency of additional molecular markers in patients with a history of MPN when compared with those with a previous MDS (5/6; 83.3% vs 18/37; 48.6%, not significant).

Secondary AML after MDS

In the cohort of NPM1-mutated s-AML after MDS or CMML, 22 additional molecular markers were identified in 18/37 (48.6%) patients with NPM1mut: evidence of FLT3-ITD was identified in 14/37 (37.8%) of patients, FLT3-TKD in 3/28 (10.7%) patients. NRAS mutations were observed in 4/37 patients (10.8%), and RUNX1 (=AML1) mutations in 1 out of 16 (6.3%) analyzed (the complete summary of mutations is presented in Table 1).

In the control group (618 NPM1-mutated de novo AML), additional molecular markers were identified in 318/618 patients (51.5%), most frequently being represented by the FLT3-ITD: 222/615 (36.1%); FLT3-TKD: in 51/469 (10.9%); NRAS: 20/156 (12.8%); and RUNX1: 1/74 (1.4%). Thus, the frequencies of additional molecular mutations showed no significant differences between de novo and s-AML.

Acute transformation of MPNs

Five out of six cases (83.3%) with NPM1-mutated transformed MPNs showed additional molecular mutations, which had been retained from the earlier disease: Two patients had a history of KITD816V-positive mastocytosis, and one case showed a ETV6-PDGFRB/t(5;12)(q33;p13). Other than that, we observed FLT3-ITD and JAK2V617F, which occurred as well in combination (Table 1).

Detection of the NPM1 mutation during development of s-AML from MDS or acute transformation of MPNs

In eight cases with s-AML following MDS and in one case with transformed MPN, paired samples were available from MDS and from diagnosis of transformed disease. Thus, the NPM1 mutations were re-tracked by mutation-specific RQ-PCR in samples from earlier time points of disease before the transformation to s-AML. Backtracking was also performed for molecular markers being present in addition to the NPM1 mutation at s-AML.

Backtracking of the NPM1 mutation in eight cases with a history of MDS or CMML

In patients with s-AML after MDS or CMML, the NPM1 mutation showed two different patterns (Figures 1 and 2):

  1. In three cases (patients #1, #2, #3), the NPM1 mutation was not detectable at diagnosis of the MDS (35, 11 and 17 months before diagnosis of the s-AML, respectively). In case #2, 4 months after diagnosis of MDS, an %NPM1mut/ABL1 expression level of 1.56% was detected, whereas 11 months after diagnosis of MDS the patient transformed to s-AML, and the %NPM1mut/ABL1 expression level further increased to 2120.3% (Figure 2a). Thus, the NPM1 mutation was clearly emerging during the later MDS course. Case #2, which was observed with a normal karyotype at diagnosis of the MDS, showed in addition clonal cytogenetic evolution leading to a complex aberrant karyotype (+1,der(1;13)(q10;q10),+i(5)(p10),+8) at transformation to the AML. This patient received treatment of the MDS with hydroxyurea and azacitidine (two courses), whereas patients #1 and #3 received supportive treatment only. Patient #3 received two induction chemotherapy courses for treatment of the s-AML.
  2. In five MDS cases, the NPM1 mutation was already detectable at diagnosis of the MDS: (1) In three cases (patients #4, #5, #6), the NPM1 mutation was not detectable using the standard assay (melting curve analysis) at diagnosis of MDS, but with sensitive RQ-PCR, a low mutation level 1–4 log below the s-AML level was detectable at diagnosis of the MDS (with %NPM1mut/ABL1 ranging from 0.857 to 8.785) at 16, 11 and 9 months before onset of s-AML. Thus, the NPM1 mutation was present at low detectable levels already at the MDS stage and then was subsequently increasing (Figure 2b). (2) In two cases (patients #7 and #8), a high NPM1 mutation level comparable to that in s-AML was already detectable in MDS at 2–12 months before the s-AML evolved. These cases gained an FLT3-ITD at the time point of transformation from MDS to AML. Case #8 carried in addition a RUNX1 mutation emerging during the course of MDS, and was thus affected by three molecular events at transformation to the s-AML. Patient #7 from this subgroup received treatment with hydroxyurea for the MDS, whereas all other patients received supportive treatment for the MDS only. Patients #4 and #8 received induction treatment for the s-AML with intensive chemotherapy, whereas patient #7 received allogeneic hematopoietic stem cell transplantation following induction chemotherapy (Figure 2c).

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Backtracking of different mutations in eight cases with NPM1-positive s-AML following MDS (patients #1–8) and in one case with transformed MPN (patient #9). At the y axis the mutation status is shown. The x axis shows the intervals from diagnosis (days). Different mutations have symbols as indicated in the legend. Red symbols indicate a positive mutation status; gray symbols indicate absence of the mutation. Small symbols indicate low-level mutations and larger symbols indicate a higher mutation load (Rel, relapse; SCT, allogeneic stem cell transplantation).

Full figure and legend (69K)

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Quantitative assessments of %NPM1mut/ABL1 ratios in paired MDS/s-AML samples: (a) NPM1 mutation not detectable at diagnosis of the MDS but high in MDS; (b) NPM1 mutation low at diagnosis of MDS and increasing to high levels in s-AML; (c) NPM1 mutation levels already high at diagnosis of MDS. (d) A patient with JAK2V617F-mutated MPN that progressed quickly to s-AML with highly increasing NPM1 mutation levels and slightly decreasing JAK2V617F mutation levels.

Full figure and legend (98K)

In these eight cases, which were followed by NPM1 mutation-specific RQ-PCR, we observed a trend to shorter time to transformation with increasing %NPM1mut/ABL1 expression levels at diagnosis of MDS (Cox regression analysis; P=0.12).

Backtracking of the NPM1 mutation in one patient with transformed MPN

Backtracking of the mutations with RQ-PCR was as well performed in one transformed MPN case (patient #9). In this case, the %NPM1mut/ABL1 expression was 16.739 during the MPN phase, and showed a one-log increase to 711.126 at transformation to the s-AML (Figure 2d). Additionally, both a FLT3-ITD and a JAK2V617F mutation were present at the time of the MPN and accumulated to a higher mutation load at transformation to the s-AML. As a fourth genetic alteration, a newly acquired isochromosome i(X)(p10) was detected with chromosome banding analysis at the transformation to blast phase. The s-AML was treated with two courses of induction chemotherapy.

Characterization of additional cytogenetic alterations

In 40/43 patients from the cohort of NPM1-mutated disease (93.0%), both cytogenetic and molecular data were available: in 34/37 cases (91.9%) with NPM1-mutated s-AML following MDS or CMML and in all six cases with NPM1-mutated transformed MPNs (Table 2).

Additional cytogenetic alterations were present at the stage of transformation in 12/40 cases (30.0%)—in detail in 8/34 (23.5%) of those patients with a history of MDS or CMML and in 4/6 (66.7%) of those with a previous MPN. One patient with a transformed MPN was diagnosed with a t(5;12)(q33;p13)/ETV6-PDGFRB (Tables 1 and 2).

In the control cohort consisting of 618 patients with NPM1-mutated de novo AML, additional cytogenetic abnormalities were significantly less frequent (82/618; 13.3%) when compared with NPM1-mutated s-AML (cytogenetic aberration rate, 12/40; 30.0%; P=0.008).

Frequency of additional cytogenetic and molecular alterations in NPM1-mutated s-AML

We then considered those 40 patients in whom cytogenetics and molecular genetics were available in parallel (NPM1-mutated s-AML following MDS or CMML: n=34; following MPNs: n=6). Most patients of the s-AML cohort (26/40; 65.0%) had additional genetic alterations (s-AML following MDS or CMML: 21/34; 61.8%; following MPNs: 5/6): Eight patients from the s-AML cohort showed additional molecular and cytogenetic alterations in combination (8/40; 20.0%: four s-AML following MDS, four with a history of MPN). A total of 14 out of 40 patients (35.0%) from the whole cohort had additional molecular alterations but no additional cytogenetic alterations (s-AML following MDS: n=13; following MPNs: n=1). Additional cytogenetic alterations but no additional molecular markers were observed in 4/40 of the total cohort (10.0%), all with an MDS history (Table 3).

In the NPM1-mutated de novo AML control cohort, only 35/618 patients showed additional cytogenetic and molecular alterations in combination (5.7%), which was less when compared with the rate in NPM1-mutated s-AML (20%) (P=0.002).

Cup-like morphology

A total of 33 cases were evaluable for cup-like morphology. The median percentage of cup-like blasts was 5% (range, 0–50%). A total of 20 cases (60.6% from the cohort) fulfilled the criterion of cup-like morphology according to the above definition28 (greater than or equal to5% of cup-like positive blasts).

Comparison of the immunophenotypes of NPM1-mutated with NPM1-unmutated s-AML patients

Immunophenotyping was performed in 18 s-AML cases with an NPM1 mutation and in 158 s-AML cases without an NPM1 mutation. The NPM1-mutated cases had a higher expression of CD33 (88.2±8.9% vs 56.3±27.6%; P<0.001) and CD38 (73.3±19.7% vs 60.8±21.9%; P=0.022), whereas s-AML cases without an NPM1 mutation had a higher expression of CD34 (34.4±25.1% vs 9.5±10.2%; P<0.001), CD133 (24.4±21.0% vs 5.0±7.0%; P<0.001), CD2 (21.8±17.8% vs 8.2±5.1%; P<0.001) and CD7 (29.6±21.0% vs 16.9±15.1%; P=0.004). There were no significant differences in the other antigens analyzed.

Survival outcomes

Subsequently, we analyzed OS and event-free survival in the cohort with s-AML after MDS. The median follow-up time was 289 days (for NPM1wt 305 days and 230 days for NPM1mut). In a first analysis, 176 patients with available survival data were compared (NPM1wt: n=154; NPM1mut: n=22). We were unable to detect a favorable prognostic impact of NPM1 mutations in our cohort with s-AML. Overall survival was even worse in those with NPM1 mutations compared with the NPM1wt cases (median: 164 days vs 506 days; P=0.032). In a multivariate analysis including age, white blood cells and bone marrow blast percentages (Table 4), NPM1 mutation was the only parameter with a trend for adverse outcomes (P=0.069). Event-free survival did not differ significantly between both groups (median: 164 vs 376 days; P=0.227).

Subsequently, the FLT3-ITD status was taken into account. For this subgroup analysis, 173 s-AML patients were available: NPM1wt/FLT3wt (n=140), NPM1mut/FLT3wt (n=13), NPM1wt/FLT3-ITD (n=11) and NPM1mut/FLT3-ITD (n=9). No significant difference with respect to outcome was detected between the four different subgroups (medians for OS: 504, 164, 529 and 159 days, respectively; medians for event-free survival: 380, 164, 237 and 159 days, respectively).



NPM1 mutations exhibit features of a founder genetic lesion in AML, showing a close association with de novo AML.6, 12, 31 Few studies only have so far reported on NPM1 mutations in s-AML. Gale et al.13 documented a lower frequency of 27% of NPM1 mutations in s-AML (n=25/93) compared with 43% in de novo AML (n=478/1124; P=0.003). Döhner et al.5 described frequent occurrence of the NPM1 mutations as well in patients with secondary or therapy-associated AML (t-AML). Low incidences of NPM1 mutations had also been identified in patients with MDS in previous studies, for example by Zhang et al.,32 who reported NPM1 mutations in 5.3% of 38 MDS patients.

So far, no study focused specifically on the role of the NPM1 mutations in s-AML or transformed MPN, or was able to backtrack the respective time points and cases. We here performed analyses of NPM1 mutations in a cohort of 350 cases with s-AML following a diagnosis of MDS or MPN and compared the results with a control cohort of 618 patients with NPM1-mutated de novo AML.

An NPM1 mutation was detected less frequent (12.2%) in cases with s-AML following MDS or MPNs (13.1% in s-AML following MDS, 9.0% in transformed MPNs) compared with de novo AML (overall mutation rate of 35%).6 These data allow speculations about the role of NPM1 mutations for secondary leukemogenesis in subsets of patients with MDS or MPN, although they clearly seem to have a minor role in s-AML when compared with de novo AML in which they represent one of the main lesions.

Additional cytogenetic alterations were identified in 30.0% of NPM1 mutation carriers at the time point of transformation from MDS to s-AML, and thus were significantly more frequent compared with 13.3% in the control cohort of NPM1-mutated de novo AML (P=0.008). Coincidence with other molecular markers was observed in 53.5% of the whole cohort of NPM1 mutated s-AML. In patients with an MDS history, most frequent were the FLT3-ITDs (37.8%), FLT3-TKD (10.7%), NRAS (10.8%) and RUNX1 (6.3%) mutations. This was similar to the frequencies of additional molecular markers in our control cohort of NPM1-mutated de novo AML and similar to previous results of Döhner et al.5 and Thiede et al.33

To investigate the evolutionary events during the time from first diagnosis of MDS to s-AML, backtracking of the mutations was carried out in eight MDS cases at the stage of MDS and compared with the time point of s-AML. Two main patterns were observed regarding the leukemogenesis from MDS to NPM1-mutated s-AML: (1) cases with unmutated NPM1 at diagnosis of the MDS acquiring the mutation during the transformation process, (2) cases that showed an NPM1 mutation already in an early phase of the MDS at low expression levels increasing at transformation to AML, or with high NPM1 levels already at MDS phase and acquiring other molecular or cytogenetic events at the time of transformation.

Thus, additional molecular and cytogenetic alterations were either present simultaneously with the NPM1 mutations during the MDS phase or were acquired during the transformation process. This suggests that the NPM1 mutations might participate in a multistep transformation process in subsets of patients. Various other mutations have been implicated in the transformation of MDS towards s-AML, for example, mutations of the NRAS, FLT3,14, 15 CEBPA34 or MLL (MLL-PTD)17 genes. The results from our study suggest that the NPM1 mutations can function as an early genetic event or might occur as a later genetic hit to induce transformation of MDS to NPM1-mutated AML.

It might be discussed whether some cases of ‘transformed’ AML represent AML ab initio rather than MDS when the NPM1 mutation was detectable already at the MDS stage, especially when the interval between diagnosis of the MDS and transformation to the s-AML was short. The occurrence of multilineage dysplasia in cases with NPM1-mutated AML35 also might cause problems in the assignment of cases. On the other hand, the long MDS prephase ranging from 7 to 165 months in 64.0% of patients from this analysis, gave clear support to the potential of the NPM1 mutations to contribute to secondary leukemogenesis in MDS cases.

We further performed backtracking investigation in one transformed MPN case. The NPM1 mutation was detectable at a considerable expression level already at the time of MPN, being accompanied by an additional FLT3-ITD and JAK2V617F mutation, as well as cytogenetic aberrations. Leukemic transformation was associated with a one-log increase of the NPM1 mutation expression and gain of a new cytogenetic alteration. This may be a first example that NPM1 mutations participate in multistep pathways not only in MDS but also in the acute transformation of MPNs.34 It remains speculative whether the MDS cases develop new NPM1-mutated clones, whereas the initial ‘MDS-like’ clone remains NPM1 unmutated or whether the ‘MDS clones’ gain the NPM1 mutations.

In contrast with the data known from de novo AML,5, 7 in our s-AML cohort, we saw no survival benefit for s-AML patients with a sole NPM1 mutation compared with NPM1 wild-type s-AML cases. As the limited size of this cohort has to be seen, the prognostic impact of the NPM1 mutations in patients with s-AML should be further investigated.

Finally, we performed morphological evaluation of the cohort for the phenomenon of cup-like morphology of the blasts. According to the definition of Kroschinsky et al., 60.6% of all cases investigated with NPM1-mutated s-AML fulfilled the criterion of ‘cup-like morphology’. These results suggest that the previously noted association of cup-like morphology and NPM1-mutated AML28, 36 is not restricted to de novo cases, but can be observed as well in s-AML.

In conclusion, the NPM1 mutations may have a relevant role for the leukemic transformation of MDS or MPN either as an early- or as an late-genetic hit. NPM1 mutations can be present at diagnosis of MDS or MPN or can emerge later during follow-up of disease in addition to or in parallel with other molecular or cytogenetic alterations. This supports the idea of a multistep leukemogenesis in MDS and MPN on its way to s-AML.

In conclusion, these data suggest that the NPM1 mutations are not only a key factor in the initiation of de novo AML but may contribute to s-AML following MDS or MPN, confirming previous observations.5, 13 On the other hand, this analysis confirms that the link between the NPM1 mutations and de novo AML is stronger when compared with s-AML in which the NPM1 mutations seem to be relevant in subsets of patients only. Considering that no survival benefit of the NPM1 mutations in s-AML was seen in this analysis, future studies should investigate the prognostic impact of isolated NPM1 mutations in s-AML, and the separation of NPM1-mutated MDS and AML by distinct blast thresholds should be discussed.


Conflict of interest

SS, CH, WK and TH are part owners of the Munich Leukemia Laboratory. TA, FD and JS are employed by the MLL. UB has nothing to disclose.



  1. Falini B, Nicoletti I, Martelli MF, Mecucci C. Acute myeloid leukemia carrying cytoplasmic/mutated nucleophosmin (NPMc+ AML): biologic and clinical features. Blood 2007; 109: 874–885. | Article | PubMed | ISI | ChemPort |
  2. Falini B, Bolli N, Liso A, Martelli MP, Mannucci R, Pileri S et al. Altered nucleophosmin transport in acute myeloid leukaemia with mutated NPM1: molecular basis and clinical implications. Leukemia 2009; 23: 1731–1743. | Article | PubMed | ISI | ChemPort |
  3. Grisendi S, Mecucci C, Falini B, Pandolfi PP. Nucleophosmin and cancer. Nat Rev Cancer 2006; 6: 493–505. | Article | PubMed | ISI | ChemPort |
  4. Bolli N, De Marco MF, Martelli MP, Bigerna B, Pucciarini A, Rossi R et al. A dose-dependent tug of war involving the NPM1 leukaemic mutant, nucleophosmin, and ARF. Leukemia 2009; 23: 501–509. | Article | PubMed | ISI | ChemPort |
  5. Döhner K, Schlenk RF, Habdank M, Scholl C, Rucker FG, Corbacioglu A et al. Mutant nucleophosmin (NPM1) predicts favorable prognosis in younger adults with acute myeloid leukemia and normal cytogenetics: interaction with other gene mutations. Blood 2005; 106: 3740–3746. | Article | PubMed | ISI | ChemPort |
  6. Falini B, Mecucci C, Tiacci E, Alcalay M, Rosati R, Pasqualucci L et al. Cytoplasmic nucleophosmin in acute myelogenous leukemia with a normal karyotype. N Engl J Med 2005; 352: 254–266. | Article | PubMed | ISI | ChemPort |
  7. Schnittger S, Schoch C, Kern W, Mecucci C, Tschulik C, Martelli MF et al. Nucleophosmin gene mutations are predictors of favorable prognosis in acute myelogenous leukemia with a normal karyotype. Blood 2005; 106: 3733–3739. | Article | PubMed | ISI | ChemPort |
  8. Verhaak RG, Goudswaard CS, van Putten W, Bijl MA, Sanders MA, Hugens W et al. Mutations in nucleophosmin (NPM1) in acute myeloid leukemia (AML): association with other gene abnormalities and previously established gene expression signatures and their favorable prognostic significance. Blood 2005; 106: 3747–3754. | Article | PubMed | ISI | ChemPort |
  9. Schlenk RF, Döhner K, Krauter J, Fröhling S, Corbacioglu A, Bullinger L et al. Mutations and treatment outcome in cytogenetically normal acute myeloid leukemia. N Engl J Med 2008; 358: 1909–1918. | Article | PubMed | ISI | ChemPort |
  10. Swerdlow S, Campo E, Lee Harris N, Jaffe E, Pileri S, Stein H et al. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues, 4th edn. IARC press: Lyon, 2008.
  11. Falini B. Any role for the nucleophosmin (NPM1) gene in myelodysplastic syndromes and acute myeloid leukemia with chromosome 5 abnormalities? Leuk Lymphoma 2007; 48: 2093–2095. | Article | PubMed | ISI | ChemPort |
  12. Shiseki M, Kitagawa Y, Wang YH, Yoshinaga K, Kondo T, Kuroiwa H et al. Lack of nucleophosmin mutation in patients with myelodysplastic syndrome and acute myeloid leukemia with chromosome 5 abnormalities. Leuk Lymphoma 2007; 48: 2141–2144. | Article | PubMed | ISI | ChemPort |
  13. Gale RE, Green C, Allen C, Mead AJ, Burnett AK, Hills RK et al. The impact of FLT3 internal tandem duplication mutant level, number, size, and interaction with NPM1 mutations in a large cohort of young adult patients with acute myeloid leukemia. Blood 2008; 111: 2776–2784. | Article | PubMed | ISI | ChemPort |
  14. Bacher U, Haferlach T, Kern W, Haferlach C, Schnittger S. A comparative study of molecular mutations in 381 patients with myelodysplastic syndrome and in 4130 patients with acute myeloid leukemia. Haematologica 2007; 92: 744–752. | Article | PubMed | ISI | ChemPort |
  15. Pinheiro RF, de Sa Moreira E, Silva MR, Alberto FL, Chauffaille ML. FLT3 internal tandem duplication during myelodysplastic syndrome follow-up: a marker of transformation to acute myeloid leukemia. Cancer Genet Cytogenet 2008; 183: 89–93. | Article | PubMed | ISI |
  16. Kaeferstein A, Krug U, Tiesmeier J, Aivado M, Faulhaber M, Stadler M et al. The emergence of a C/EBPalpha mutation in the clonal evolution of MDS towards secondary AML. Leukemia 2003; 17: 343–349. | Article | PubMed | ISI | ChemPort |
  17. Dicker F, Haferlach C, Sundermann J, Wendland N, Weiss T, Kern W et al. Mutation analysis for RUNX1, MLL-PTD, FLT3-ITD, NPM1 and NRAS in 269 patients with MDS or secondary AML. Leukemia 2010; 24: 1528–1532. | Article | PubMed | ISI |
  18. Löffler H, Raststetter J, Haferlach T. Atlas of Clinical Hematology, 6th edn. Springer: Berlin, 2004.
  19. Schoch C, Schnittger S, Bursch S, Gerstner D, Hochhaus A, Berger U et al. Comparison of chromosome banding analysis, interphase- and hypermetaphase-FISH, qualitative and quantitative PCR for diagnosis and for follow-up in chronic myeloid leukemia: a study on 350 cases. Leukemia 2002; 16: 53–59. | Article | PubMed | ISI | ChemPort |
  20. Schnittger S, Kern W, Tschulik C, Weiss T, Dicker F, Falini B et al. Minimal residual disease levels assessed by NPM1 mutation specific RQ-PCR provide important prognostic information in AML. Blood 2009; 114: 2220–2231. | Article | PubMed | ISI |
  21. Schnittger S, Schoch C, Dugas M, Kern W, Staib P, Wuchter C et al. Analysis of FLT3 length mutations in 1003 patients with acute myeloid leukemia: correlation to cytogenetics, FAB subtype, and prognosis in the AMLCG study and usefulness as a marker for the detection of minimal residual disease. Blood 2002; 100: 59–66. | Article | PubMed | ISI | ChemPort |
  22. Bacher U, Haferlach C, Kern W, Haferlach T, Schnittger S. Prognostic relevance of FLT3-TKD mutations in AML: the combination matters—an analysis of 3082 patients. Blood 2008; 111: 2527–2537. | Article | PubMed | ISI | ChemPort |
  23. Schnittger S, Kinkelin U, Schoch C, Heinecke A, Haase D, Haferlach T et al. Screening for MLL tandem duplication in 387 unselected patients with AML identify a prognostically unfavorable subset of AML. Leukemia 2000; 14: 796–804. | Article | PubMed | ISI | ChemPort |
  24. Bacher U, Haferlach T, Schoch C, Kern W, Schnittger S. Implications of NRAS mutations in AML: a study of 2502 patients. Blood 2006; 107: 3847–3853. | Article | PubMed | ISI | ChemPort |
  25. Schnittger S, Bacher U, Kern W, Schroder M, Haferlach T, Schoch C. Report on two novel nucleotide exchanges in the JAK2 pseudokinase domain: D620E and E627E. Leukemia 2006; 20: 2195–2197. | Article | PubMed | ISI | ChemPort |
  26. Schnittger S, Kohl TM, Haferlach T, Kern W, Hiddemann W, Spiekermann K et al. KIT-D816 mutations in AML1-ETO-positive AML are associated with impaired event-free and overall survival. Blood 2006; 107: 1791–1799. | Article | PubMed | ISI | ChemPort |
  27. Cools J, DeAngelo DJ, Gotlib J, Stover EH, Legare RD, Cortes J et al. A tyrosine kinase created by fusion of the PDGFRA and FIP1L1 genes as a therapeutic target of imatinib in idiopathic hypereosinophilic syndrome. N Engl J Med 2003; 348: 1201–1214. | Article | PubMed | ISI | ChemPort |
  28. Kroschinsky FP, Schakel U, Fischer R, Mohr B, Oelschlaegel U, Repp R et al. Cup-like acute myeloid leukemia: new disease or artificial phenomenon? Haematologica 2008; 93: 283–286. | Article | PubMed | ISI |
  29. Kern W, Voskova D, Schoch C, Hiddemann W, Schnittger S, Haferlach T. Determination of relapse risk based on assessment of minimal residual disease during complete remission by multiparameter flow cytometry in unselected patients with acute myeloid leukemia. Blood 2004; 104: 3078–3085. | Article | PubMed | ISI | ChemPort |
  30. Cheson BD, Bennett JM, Kopecky KJ, Büchner T, Willman CL, Estey EH et al. Revised recommendations of the International Working Group for diagnosis, standardization of response criteria, treatment outcomes, and reporting standards for therapeutic trials in acute myeloid leukemia. J Clin Oncol 2003; 21: 4642–4649. | Article | PubMed | ISI |
  31. Liso A, Castiglione F, Cappuccio A, Stracci F, Schlenk RF, Amadori S et al. A one-mutation mathematical model can explain the age incidence of acute myeloid leukemia with mutated nucleophosmin (NPM1). Haematologica 2008; 93: 1219–1226. | Article | PubMed | ISI |
  32. Zhang Y, Zhang M, Yang L, Xiao Z. NPM1 mutations in myelodysplastic syndromes and acute myeloid leukemia with normal karyotype. Leuk Res 2007; 31: 109–111. | Article | PubMed | ISI | ChemPort |
  33. Thiede C, Koch S, Creutzig E, Steudel C, Illmer T, Schaich M et al. Prevalence and prognostic impact of NPM1 mutations in 1485 adult patients with acute myeloid leukemia (AML). Blood 2006; 107: 4011–4020. | Article | PubMed | ISI | ChemPort |
  34. Ernst T, Chase A, Zoi K, Waghorn K, Hidalgo-Curtis C, Score J et al. Transcription factor mutations in myelodysplastic/myeloproliferative neoplasms. Haematologica 2010; 95: 1473–1480. | Article | PubMed | ISI |
  35. Falini B, Macijewski K, Weiss T, Bacher U, Schnittger S, Kern W et al. Multilineage dysplasia has no impact on biologic, clinicopathologic, and prognostic features of AML with mutated nucleophosmin (NPM1). Blood 2010; 115: 3776–3786. | Article | PubMed | ISI |
  36. Chen W, Konoplev S, Medeiros LJ, Koeppen H, Leventaki V, Vadhan-Raj S et al. Cuplike nuclei (prominent nuclear invaginations) in acute myeloid leukemia are highly associated with FLT3 internal tandem duplication and NPM1 mutation. Cancer 2009; 115: 5481–5489. | Article | PubMed | ISI | ChemPort |


We thank Professor Brunangelo Falini (Institute of Hematology, University of Perugia) for fruitful discussion of the manuscript.