Acute myeloid leukaemia with NPM1 mutation: no longer having an absolute blast (count).

Increasing recognition of diagnostic and prognostic implications of molecular abnormalities in acute myeloid leukaemia (AML) have led to the incorporation of entities with disease-defining mutations in recent revisions of the World Health Organization (WHO) classification of haematolymphoid tumours.1Swerdlow S.H. Campo E. Harris N.L. et al.World Health Organisation Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Revised 4th ed. Vol 2. Lyon2017Google Scholar,2Khoury J.D. Solary E. Abla O. et al.The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms.Leukemia. 2022; 36: 1703-1719Crossref PubMed Scopus (362) Google Scholar While nucleophosmin (NPM1) mutation is considered a founding mutation of AML, its presence alone has not hitherto defined AML in absence of blast percentage >20%,1Swerdlow S.H. Campo E. Harris N.L. et al.World Health Organisation Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Revised 4th ed. Vol 2. Lyon2017Google Scholar although this blast threshold is set to be abolished in the fifth revision of the WHO classification.2Khoury J.D. Solary E. Abla O. et al.The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms.Leukemia. 2022; 36: 1703-1719Crossref PubMed Scopus (362) Google Scholar We present a case which demonstrates the diagnostic challenges haematopathologists and clinicians face in AML with NPM1 mutation and supports the proposed elimination of blast percentage requirements for this entity. A 44-year-old otherwise well female presented with sepsis requiring inotropic support, secondary to a large vulval abscess. She had reported the onset of new skin lesions on the face (Fig. 1) and vulval region 2 months prior to her presentation. The vulval lesion, in particular, had rapidly progressed, requiring debridement by the gynaecology-oncology team. Histopathological examination of both vulval and facial lesions were consistent with neutrophilic dermatosis (Sweet syndrome). The patient was treated with a course of pulse methylprednisone 500 mg IV for 5 days. Whilst at time of admission the full blood count (FBC) demonstrated only normocytic anaemia and mild neutrophilia [haemoglobin (Hb) 116 g/L, mean corpuscular volume (MCV) 91fL, total white cell count (WCC) 17.6×109/L, neutrophils 15.6×109/L, platelets 176×109/L], over the course of several days this evolved to a leukoerythroblastic blood picture with 4–6% circulating blasts (Hb 89 g/L, WCC 28.7×109/L, neutrophils 22.9×109/L, myelocytes 0.5×109/L, blasts 1.1×109/L, platelets 206×109/L). A bone marrow biopsy was performed to exclude a haematological driver underlying Sweet syndrome. This demonstrated a markedly hypercellular marrow and marked granulocytic dysplasia (Fig. 2) with accompanying dysmegakaryopoiesis to a lesser degree. Eleven percent blasts were enumerated by morphology on the aspirate sample, though <1% CD34+ cells were identified on flow cytometry and trephine immunohistochemistry. Canonical myeloproliferative neoplasm (MPN) mutations (JAK2, BCR-ABL, MPL and Calreticulin) were negative. The karyotype was normal. The provisional conclusion was that this was suggestive of an evolving myelodysplastic syndrome (MDS) or a MPN, with its interpretation impacted by the significant inflammatory response at the time of sampling. Given this diagnostic dilemma and to confirm the presence of a clonal process, next generation sequencing (NGS) for myeloid-associated mutations was forwarded off-site for further processing. The patient was discharged on a tapering dose of oral prednisone initially at 60 mg daily. Her Sweet syndrome improved on moderate steroid dosing, however, re-occurred each time when weaned below 10 mg daily. The haematological parameters gradually normalised 12 weeks following initial presentation (Hb 120 g/L, WCC 5.2×109/L, neutrophils 4.1×109/L, platelets 244×109/L) although occasional dysplastic neutrophils were still noted on film review. Despite normalisation of the FBC parameters, a progress bone marrow biopsy enumerated 7% blasts with improvement in the degree of dysplasia seen in granulopoiesis and megakaryopoiesis. Of note, at the time of conducting this progress bone marrow biopsy, the myeloid NGS results returned from the initial aspirate. This demonstrated a type A frameshift mutation in the NPM1 gene NM_002520.6: c.860_863dup; p.(Trp288Cysfs∗12) at variant allelic frequency (VAF) of 40–45%, as well as a missense mutation in DNA methyltransferase 3 alpha (DNMT3A) gene NM_022552.4: c.2644C>T; p.(Arg882Cys) with VAF 30–35%. At the time, the entity of AML with NPM1 mutation was recognised, however this still required an absolute blast percentage >20%.1Swerdlow S.H. Campo E. Harris N.L. et al.World Health Organisation Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Revised 4th ed. Vol 2. Lyon2017Google Scholar The case was discussed at the local haematology multidisciplinary meeting and further opinion was sought internationally. A consensus was reached that the patient should be treated as AML with NPM1 mutation, despite her normal blood count parameters and blast percentage <20% on the bone marrow sample. The patient was admitted for her first cycle of induction chemotherapy with daunorubicin-cytarabine (DA 3+8) 4 months following her original presentation with Sweet syndrome. Following cycle 1, bone marrow biopsy demonstrated complete remission (CR). Bone marrow biopsy following cycle 2 of induction therapy continued to show CR with incomplete platelet recovery. NPM1 minimal residual disease (MRD) analysis showed no detectable NPM1 in peripheral blood and low level MRD (1.332 copies per 105 ABL copies) in the bone marrow in keeping with CR with molecular MRD detection at low level (CR-MRD-LL). She received three cycles of high dose cytarabine (HIDAC) consolidation therapy. Bone marrow NPM1 following cycle 2 HIDAC continued to show CR-MRD-LL with 3.87 NPM1 transcripts per 105 ABL, reaching CR with negative MRD (CR-MRD) at end of consolidation therapy. She continues to remain in a CR-MRD negative state, now 12 months following completion of therapy. Three-monthly bone marrow MRD is planned to continue for 2 years following completion of therapy in line with current European Leukaemia Net 2021 guidelines.3Heuser M. Freeman S.D. Ossenkoppele G.J. et al.2021 update on MRD in acute myeloid leukemia: a consensus document from the European LeukemiaNet MRD Working Party.Blood. 2021; 138: 2753-2767Crossref PubMed Scopus (125) Google Scholar The NPM1 mutation has been considered specific to AML since it was originally reported close to two decades ago, seen in approximately 30% of all adult de novo AML.1Swerdlow S.H. Campo E. Harris N.L. et al.World Health Organisation Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Revised 4th ed. Vol 2. Lyon2017Google Scholar Whilst multilineage dysplasia (MLD) can often be observed in AML with NPM1 mutation, the morphological detection of MLD has not been shown to bear any clinicopathological or prognostic implications in AML patients found to have an NPM1 mutation.4Heath E.M. Chan S.M. Minden M.D. Murphy T. Shlush L.I. Schimmer A.D. Biological and clinical consequences of NPM1 mutations in AML.Leukemia. 2017; 31: 798-807Crossref PubMed Scopus (118) Google Scholar Therefore, presence of the NPM1 mutation has been an accepted disease-defining feature of AML, with the presence of MLD being subsidiary to the detection of the NPM1 mutation.1Swerdlow S.H. Campo E. Harris N.L. et al.World Health Organisation Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Revised 4th ed. Vol 2. Lyon2017Google Scholar In the case illustrated, the initial diagnostic challenge related to the presence of both MLD features and NPM1 mutation despite a blast count <20%, so should this be diagnosed and treated as MDS or AML? Myeloid neoplasms with NPM1 mutation presenting with blast percentage <20% are uncommonly reported, with NPM1 mutations detected in <2–3% of MDS or CMML cases.5Forghieri F. Nasillo V. Paolini A. et al.NPM1-mutated myeloid neoplasms with <20% blasts: a really distinct clinico-pathologic entity?.Int J Mol Sci. 2020; 21: 8975Crossref PubMed Scopus (20) Google Scholar Our case shared pathological features of MDS bearing NPM1 mutation as reported in the literature to date, including normal karyotype and CD34 negativity.6Montalban-Bravo G. Kanagal-Shamanna R. Sasaki K. et al.NPM1 mutations define a specific subgroup of MDS and MDS/MPN patients with favorable outcomes with intensive chemotherapy.Blood Adv. 2019; 3: 922-933Crossref PubMed Scopus (61) Google Scholar MDS harbouring NPM1 mutations have been reported to bear an aggressive clinical course, rapidly evolving to AML within 0.5–16 months.6Montalban-Bravo G. Kanagal-Shamanna R. Sasaki K. et al.NPM1 mutations define a specific subgroup of MDS and MDS/MPN patients with favorable outcomes with intensive chemotherapy.Blood Adv. 2019; 3: 922-933Crossref PubMed Scopus (61) Google Scholar Indeed, MDS with NPM1 mutation may simply be AML with NPM1 mutation caught early in its natural history. Whilst acquisition of the NPM1 mutation occurs during leukaemic transformation, most patients already harbour NPM1 mutation during their ‘MDS phase’.7Schnittger S. Haferlach C. Nadarajah N. et al.In AML secondary to MDS NPM1 mutations are late events, less frequent, and associated with a different pattern of molecular mutations than in de novo AML.Blood. 2014; 124: 700Crossref PubMed Google Scholar In the case presented, evolution to AML (defined on the basis of blast percentage alone) was not apparent in the 4 months from initial presentation to induction, though the clinical course was modified by the multidisciplinary decision to commence induction chemotherapy despite normalisation of FBC parameters and marrow findings not meeting original 2017 WHO criteria for AML with NPM1 mutation. We believe this illustrative case supports the move to allow diagnosis of AML with NPM1 mutation irrespective of blast count, whilst exercising ‘judicious clinicopathological correlation’.2Khoury J.D. Solary E. Abla O. et al.The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms.Leukemia. 2022; 36: 1703-1719Crossref PubMed Scopus (362) Google Scholar Furthermore, this case also illustrates pitfalls in morphological blast enumeration with significant interobserver variability amongst expert haematopathologists.8Naqvi K. Jabbour E. Bueso-Ramos C. et al.Implications of discrepancy in morphologic diagnosis of myelodysplastic syndrome between referral and tertiary care centers.Blood. 2011; 118: 4690-4693Crossref PubMed Scopus (82) Google Scholar This, combined with the finding that two-thirds of AML with NPM1 mutation are CD34-negative and another third negative for HLA-DR,1Swerdlow S.H. Campo E. Harris N.L. et al.World Health Organisation Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Revised 4th ed. Vol 2. Lyon2017Google Scholar potentially contribute to under-appreciation of blast percentage AML with NPM1 mutation and misdiagnosis of MDS rather than AML based on the blast percentage requirement prior to the latest proposed iteration of the WHO classification.2Khoury J.D. Solary E. Abla O. et al.The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms.Leukemia. 2022; 36: 1703-1719Crossref PubMed Scopus (362) Google Scholar Indeed, our case was subsequently forwarded to an external laboratory for a second opinion, which yielded a significantly higher blast count on the initial diagnostic marrow that would have met the >20% threshold for AML diagnosis from the outset. Notwithstanding potential bias arising from knowledge of the NPM1 mutation that the external reviewers were privy to (which the original reporting haematopathologists were not), this case highlights the implications of interobserver discrepancies in blast enumeration. Immunohistochemical detection of cytoplasmic NPM1 on bone marrow trephines has been shown to be predictive of NPM1 mutation9Falini B. Martelli M.P. Bolli N. et al.Immunohistochemistry predicts nucleophosmin (NPM) mutations in acute myeloid leukemia.Blood. 2006; 108: 1999-2005Crossref PubMed Scopus (156) Google Scholar and could be considered as a surrogate measure to determine blast percentage or even identify rare non-exon 12 NPM1 mutations.10Martelli M.P. Rossi R. Venanzi A. et al.Novel NPM1 exon 5 mutations and gene fusions leading to aberrant cytoplasmic nucleophosmin in AML.Blood. 2021; 138: 2696-2701Crossref PubMed Scopus (16) Google Scholar Further supporting the proposed WHO changes,2Khoury J.D. Solary E. Abla O. et al.The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms.Leukemia. 2022; 36: 1703-1719Crossref PubMed Scopus (362) Google Scholar the rapid evolution of previously recognised MDS with NPM1 mutation to frank AML underscores a clear clinical argument to treat these patients as per AML regardless of blast percentage. In younger/fit patients diagnosed with MDS bearing NPM1 mutation, treatment with intensive chemotherapy led to better outcomes than use of hypomethylating agents (HMA) alone.6Montalban-Bravo G. Kanagal-Shamanna R. Sasaki K. et al.NPM1 mutations define a specific subgroup of MDS and MDS/MPN patients with favorable outcomes with intensive chemotherapy.Blood Adv. 2019; 3: 922-933Crossref PubMed Scopus (61) Google Scholar Furthermore, in the age of venetoclax/HMA combinations and novel agents in older patients, it can be argued that the diagnosis of AML rather than MDS would afford this group of patients greater access to effective therapy for their disease. Whether or not a downward revision in the blast percentage should apply more broadly to patients in the absence of recurrent genetic abnormalities remains contentious. Historical outcomes of patients with MDS with excess blasts appear dependent on the decision to employ intensive therapy rather than the absolute blast percentage.11Estey E. Thall P. Beran M. Kantarjian H. Pierce S. Keating M. Effect of diagnosis (refractory anemia with excess blasts, refractory anemia with excess blasts in transformation, or acute myeloid leukemia [AML]) on outcome of AML-type chemotherapy.Blood. 1997; 90: 2969-2977Crossref PubMed Google Scholar There are calls for the establishment of a distinct MDS/AML category with blasts of 10–30% to allow patients better access to appropriate treatment and clinical trials. Preliminary presentations on the International Consensus Classification of AML suggest the use of an MDS/AML category with 10–19% blasts in presence of certain mutations such as TP53, whilst the use of >10% blasts is required for AML with mutated NPM1.12Arber D.A. Orazi A. Hasserjian R.P. et al.International consensus classification of myeloid neoplasms and acute leukemia: integrating morphological, clinical, and genomic data.Blood. 2022; 140: 1200-1228Crossref PubMed Scopus (260) Google Scholar This case represents a timely reminder regarding the challenges for haematopathologists and clinicians alike in the face of rapidly evolving developments in our understanding of the diagnostic and prognostic implications of molecular changes in the AML landscape. The authors state that there are no conflicts of interest to disclose.