Acute myeloid leukaemia (AML) carrying mutations of the nucleophosmin (NPM1) gene, which lead to aberrant cytoplasmic dislocation of the mutated protein (NPMc+ AML),1 accounts for about one-third of all adult AML, usually harbouring a normal karyotype,2 and shows distinctive molecular, immunophenotypic and clinical features.3 In the absence of a concomitant FLT3-ITD mutation, this leukaemia is characterized by a relatively favourable outcome.4 For these reasons, AML with mutated NPM1 has been recently included as new provisional entity in the fourth edition of the World Health Organization (WHO) Classification of Tumors of the Hematopoietic and Lymphoid Tissues.5
AML with mutated NPM1 is regarded as the typical de novo leukaemia because it does not usually occur following myelodysplastic syndromes or chronic myeloproliferative disorders.3 We describe for the first time a unique case of Philadelphia-positive chronic myeloid leukaemia (CML) presenting as myeloid blast crisis carrying cytoplasmic mutated NPM1.
In June 2007, a 70-year-old female patient was diagnosed for pulmonary tuberculosis. At that time, peripheral blood counts showed: white blood cell (WBC) 20 × 109 per litre, with normal differential count; haemoglobin (Hb) 13.0 g/100 ml; platelet (PLT) count 250 × 109 per litre. She received multi-antibiotic therapy achieving complete remission of her disease. In December 2007, an increase of WBC count was documented (39 × 109 per litre, with myeloid precursors), which was regarded as consistent with a myeloproliferative neoplasm, likely neglected at the time of tubercular infection. However, the patient refused further investigation. In January 2008, she was referred to our institution for haematological evaluation due to progressive worsening of the performance status. Physical examination showed only mild splenomegaly. Peripheral blood counts were as follows: WBC 165.1 × 109 per litre (neutrophils 44%, lymphocytes 2%; monocytes 8%; basophils 1%; metamyelocytes 7%; myelocytes 2%; promyelocytes 3% and myeloid blasts 33%); Hb 12.0 g/100 ml; PLT 104 × 109 per litre. Bone marrow aspirate and biopsy were then performed. On aspirate, the cellularity was very high, the erythropoiesis was severely impaired, megakaryocytes were present and the granulopoiesis mainly consisted of medium-sized granulated blastic elements. By flow cytometry, the blastic population turned out to be CD33+, CD13+, CD117+, MPO7+, CD19−, CD14−, CD64+/− and CD34−. The bone marrow trephine showed a 95% cellularity and was diffusely infiltrated by myeloblasts, which demonstrated irregular, round to oval nuclei containing one or more small well-defined nucleoli. Reticulin fibrosis was absent. At immunohistochemistry performed as previously described,6 leukaemic cells were positive for myeloperoxidase (MPO), nucleophosmin (nuclear plus aberrant cytoplasmic positivity), CD56 and negative for CD34, CD68 (PGM1) and CD7, as well as for B- and T-cell markers (Figure 1). Conventional cytogenetics showed 46XX, t(9;22)(q34; q11) in 20/20 metaphase, whereas fluorescent in situ hybridization (FISH) analysis demonstrated BCR-ABL1 rearrangement in 100% analyzed cells. Additional molecular analyses were then performed to evaluate BCR-ABL1 rearrangement, NPM1 mutational status (based upon the presence of NPMc+ leukaemic cells in the bone marrow) and possible FLT3 internal tandem duplication or mutations. The b3-a2 BCR-ABL1 transcript variant was detected by quantitative real-time PCR. Denaturing high performance liquid chromatography and direct sequencing documented the presence of the NPM1 mutation. In particular, the so-called mutation A 1 was found with duplication of a TCTG tetranucleotide at positions 956–959. Conversely, the FLT3 gene turned out to be not mutated.
Although no molecular study could be performed at the time of initial diagnosis, the clinical features, the increasing WBC count during the preceding months and the morphological findings of our patient were consistent with a diagnosis of CML. The results presented in this study raise the important issue of how the newly developed AML with mutated NPM1 was linked with the preceding chronic myeloproliferative disorder. As the patient was comprehensively studied only in blast crisis, it was not possible to definitely assess whether the NPM1 mutation was present ab initio or only in the clonal blastic evolution. The latter possibility appears to be the most likely because none of the 450 CML cases investigated in chronic phase by the authors (SP and BF) carried cytoplasmic mutated NPM1. Another important question concerning our patient was how the NPM1 mutation correlated with the BCR-ABL1 rearrangement in AML cells of blast crisis. AMLs raising in the course of chronic myeloproliferative neoplasms are molecularly heterogeneous, for example they may derive from the clone carrying a Philadelphia chromosome (or a JAK2-V617 mutation), or even from a normal haemopoietic stem cell. In a recent case of AML with mutated NPM1, which developed in a patient with JAK2/V617+ primary myelofibrosis, the origin of AML cells remained controversial.7 In our patient, molecular and immunohistochemical findings were consistent with the presence of concomitant NPM1 mutation and BCR-ABL1 rearrangement in the same population of leukaemic cells during blast crisis. In fact, immunohistochemistry of the bone marrow trephine showed aberrant cytoplasmic expression of nucleophosmin in almost the totality of leukaemic cells (Figure 1, right bottom), and FISH analysis revealed BCR-ABL1 rearrangement in all blastic cells.
NPM1 gene mutations usually occur as a primary oncogenic event in de novo AML with normal karyotype 1 and have been found to be, in general, mutually exclusive with other AML recurrent cytogenetic abnormalities.2 In this particular case, it seems that the NPM1 gene mutation had a function in the blastic transformation of CML. This must be an exceptional event because none of additional 70 CMLs in myeloid blast crisis investigated by the authors (BF and SP) showed aberrant cytoplasmic expression of nucleophosmin. The results in our patient does not contradict the concept of NPM1 mutation being a founder genetic lesion, because other AML recurrent cytogenetic abnormalities, including t(8;21), inv(16) and t(15;17), 8, 9 have been implicated in the genesis of blast crisis of CML. Interestingly, clinicopathological features of blastic transformation of CML with the t(8;21), inv(16) or t(15;17) resemble those of de novo AML carrying the same cytogenetic abnormalities 8 but show poorer prognosis. Similarly, CML blast crisis in our patient exhibited several distinctive features of typical de novo AML with mutated NPM1, such as CD34 negativity and absence of other chromosomal abnormalities, apart the t(9;22) typical of underlying CML, but very dismal prognosis. In fact, the patient did not show any significant response after either imatinib and dasatinib therapy, and died for progressive disease 10 months after diagnosis.
Conflict of interest
The authors declare no conflict of interest.
This work was supported by Centro Interdipartimentale per la Ricerca sul Cancro ‘G. Prodi’, European LeukemiaNet, COFIN 2002-2003 (Professor SA Pileri), FIRB/RFO (Professor SA Pileri and Professor PL Zinzani), AIRC, Progetto Strategico di Ateneo 2006 (Professor SA Pileri and Dr Piccaluga), Fondazione Cassa di Risparmio in Bologna, and BolognAIL grants.