Mutation of the NPM1 gene contributes to the developmentof donor cell–derived acute myeloid leukemia afterunrelated cord blood transplantation for acutelymphoblastic leukemia☆,☆☆,★
aHaematology Department, Hospital General Universitario Gregorio Marañón, Madrid 28007, SpainbGregorio Marañón Institute for Health Research, Madrid 28007, SpaincBMT Unit, King's College Hospital, London SE5 9RS, UK
Received 18 August 2012; revised 20 December 2012; accepted 3 January 2013
☆ Authorship: conception and design: G.R.-M., I.B., and J.L.D.-M; analysis and interpretation of data: all authors; article drafting: G.R.-M. and I.B.;vision of the article: all authors; final approval: all authors. The first and 2 last authors take primary responsibility for the work reported here in terms ofboratory procedures (I.B.) and care of patients (G.R.-M. and J.L.D.-M.). The order of the remaining authors reflects their degree of participation in the study.
☆☆ The authors of this manuscript have no conflicts of interest to disclose.★ This work has been partially supported by grants from the Fondo de Investigación Sanitaria FIS-ISCIII (Ministry of Economy and Competence, Madrid,
pain; FIS08-90886, FIS08-1463, and FIS11-00708 to I.B.) and a grant from the Spanish Association Against Cancer, Madrid, Spain (AECC to J.L.D.).⁎ Corresponding author. Hematology Department and Bone Marrow Transplant Unit, Hospital General Universitario Gregorio Marañón, c/Doctor Esquerdo
6, 28009 Madrid, Spain.E-mail address: [email protected] (G. Rodríguez-Macías).1 I.B. and J.L.D. contributed equally to this manuscript.
1. Introduction immunomagnetic means (AutoMACS; Miltenyi Biotec,
Donor cell leukemia (DCL) is a rare but well-recognizedcomplication after stem cell transplantation (SCT). Since thefirst case described in 1971 [1], anecdotal clinical cases ofmalignancies derived from donor cells in patients undergo-ing SCT have been published. The incidence is uncertain,varying from 1% to 5% in different series [2], owing to a lackof proper recognition or reporting. In some cases, the donororigin of the leukemia could be misinterpreted because of theabsence of adequate molecular analysis. Unrelated cordblood transplantation (UCBT) has emerged as a feasiblealternative source of hematopoietic progenitors, and at least12 cases of DCL have been reported after UCBT [3,4]. It haseven been postulated that DCL could be more frequent afterUCBT than in recipients of adult hematopoietic progenitors[5]. Nevertheless, follow-up using molecular analysis ofleukemia-specific mutations and chimerism contributes to abetter recognition of this complication [4,6]. In the patientreported here, the molecular approaches for the identificationof the original leukemia, the characterization of specific genemutations, and the evaluation of donor/recipient mismatcheswithin highly polymorphic DNA sequences served as arobust evidence for the diagnosis of leukemia in donor-derived cells with nucleophosmin (NPM1) mutation [7].
Fig. 1 Electropherograms obtained from the DNA fragmentanalysis of the PCR products for the determination of chimerismtargeting STR (short tandem repeat) polymorphisms (D2S1338shown as an example; left panel) and NPM1 mutation (4-base-pairinsertion; right panel) in samples from the CB (cord blood) unit andfrom the recipient pre- and post-UCBT. The patient showedcomplete chimerism (100% donor) throughout the post-UCBTfollow-up including after diagnosis of DCL. NPM1 mutation wasonly present after diagnosis of DCL.
2. Cytogenetic and molecular methods
Conventional cytogenetic analyses were performed usingstandard procedures. Fluorescence in situ hybridization(FISH) was performed on uncultured samples using apanel of probes (Vysis Inc, Downers Grove, IL, USA) forrecurrent alterations in both acute lymphoblastic leukemia(ALL; t(1;19), t(9;22), t(11;var), t(12;21)) and acute myeloidleukemia (AML; t(8;21), inv(16), del(5q), del(7q), +8, t(11;var), del(20q)) per suggestions of the manufacturer.Molecular procedures were performed per previouslypublished approaches (see Supplementary Table for details).Molecular testing for acute lymphoid (t(1;19), t(4;11),t(9;22), t(12;21)) and myeloid (t(8;21), inv(16), FLT3-ITD,NPM1, CEBPA, IDH1) leukemia clinically relevant recur-rent alterations included conventional polymerase chainreaction (PCR; sensitivity 10−5), for t(1;19); real-timequantitative PCR (qPCR; sensitivity 10−5), for all otherchromosome rearrangements and for the follow-up ofNPM1–type A mutation; fluorescent PCR (sensitivity 1%-5%), for the screening of FLT3-ITD, NPM1, and CEBPAmutations; and direct sequencing (sensitivity 15%-20%), forscreening of IDH1. Chimerism was analyzed by STR-PCR(short tandem repeat polymerase chain reaction; AmpFlSTRSGM Plus; Applied Biosystems, Foster City, CA, USA) inbone marrow (BM) and peripheral blood (PB) samples, aswell as in leukocyte lineages (CD3+ T cells, CD19/20+ Bcells, CD15/33+ myeloid cells) isolated (N95% purity) by
Bergisch Gladbach, Germany).
3. Case report
A 43-year-old woman showed a 2-month history ofasthenia, bone tenderness, and fever. BM evaluation showed90% of blasts, with morphologic features of acute lymphoidleukemia. Immunophenotype was CD38+, HLA-DR+,CD19+, CD10+, IgM+, negative TDT, and weak CD15+(88% of cells). Karyotype analysis showed 46,XX,t(1;19)[15]. All molecular and FISH results were negative, exceptfor t(1;19). She was diagnosed as having ALL with t(1;19).After treatment with conventional chemotherapy (PETHEMAALL-AR-03) [8], she achieved complete remission, and SCTwas planned in first complete remission. Neither HLA-identical sibling nor unrelated matched adult donor could befound, and 8 months after diagnosis, the patient underwentUCBT from a 2-antigen HLA-mismatched female donor.Conditioning regimen consisted of thiotepa, fludarabine,busulfan (intravenous) and thymoglobulin, and prophylaxisfor graft-versus-host-disease of metilprednisolone, mycophe-nolate, and cyclosporine A [9]. Cord blood–nucleated cellswere infused. Approximately 6.9 × 107/kg, and CD34+ cellswere 2.4 × 105/kg. Neutrophil engraftment occurred at day+29, and complete chimerism was confirmed in BM at day+30 (Figs. 1, 2, and 3A). The patient developed renalcomplications with dialysis-dependent renal failure andrecurring cytomegalovirus viremias, most likely related to adeficient immune reconstitution, resolved with ganciclovir.No evidence of graft-versus-host-disease allowed immunosu-pression withdrawal 4 months after UCBT. All studiesperformed during the first 15 months after UCBT wereconsistently negative for residual/recurrent disease and
Fig. 2 Genetic monitoring of the patient throughout theobservation period. Karyotype and PCR were positive for t(1;19)at the moment of diagnosis of the initial ALL and remained negativethereafter (no metaphases could be obtained during the treatment ofDCL). Chimerism was complete (100% donor) in all BM samplesanalyzed after UCBT (transient mixed chimerism was observed inPB and T lymphocytes during the first month post-UCBT; seeFig. 3A). qPCR of mutated NPM1 (mutation type A) was positiveonly after diagnosis of DCL. Solid circle indicates positive result;hollow circle, negative result; and dashed circle, no metaphases.
Fig. 3 A, Dynamics of chimerism (% UCB donor) in BM(square), PB (circle), and T lymphocytes (TL; triangle) during thefirst month post-UCBT. B, Response to treatment of DCL asestimated by the percentage of blasts in the BM (square), flowcytometry (circle), and qPCR for NPM1 type A mutation (relativequantification using the sample at diagnosis as calibrator, ie, 100%NPM1, and GUS as control gene; triangle). Gray bars at the bottomof the chart indicate periods of treatment.
1698 G. Rodríguez-Macías et al.
showed complete chimerism (Figs. 1 and 2). Sixteen monthsafter transplant, the patient presented with an upper respiratoryinfection and fever. PB analysis revealed white blood cells70 × 109/L, and BM showed 80% myeloid blasts (FAB-M5-AML). Immunophenotype was MPO+, HLA-DR+, CD38+,CD13+, CD33+, CD14+, CD15+, CD64+, CD11b+, CD56+,and CD4+ (78% of cells) and was negative for the lymphoidmarkers identified at diagnosis. Moreover, cytogeneticanalysis showed a normal karyotype 46,XX[14], andmolecular determinations showed FISH and/or reversetranscriptase PCR negativity for t(1;19) as well as completedonor chimerism (Figs. 1 and 2). Within this scenario,diagnosis of AML with origin in donor-derived cells (DCL)was made. The molecular panel tested revealed mutation ofthe NPM1 gene, which showed a 4-base-pair insertion(Fig. 2), further characterized by DNA sequencing to be atype A mutation [7]. Owing to renal failure, oral chemo-therapy was administered, resulting in hematologic controlfor 8 months (Fig. 3B), after which the patient died ofdisease progression.
qPCR (sensitivity 10−5) was used to test for the presenceof NPM1 mutation in retrospective stored samples from thepatient (Fig. 2) as well as in the UCB unit received for SCT.All results from samples before diagnosis of DCL werenegative. This approach was also used for the molecular
follow-up after DCL, together with cell morphology andflow cytometry (Fig. 3B).
4. Discussion
Leukemia transformation of engrafted human stem cellprogenitors was first described 40 years ago [1]. It has raisinginterest despite is rarity because it might provide insights intothe etiology andmechanisms of leukemogenesis [10]. Etiologyof DCL is unclear, and several mechanisms have beenproposed to explain how it can arise [11], including occultleukemia in the donor, impaired immune surveillance (slowimmune reconstitution after UCBT), stromal abnormalities(inherent or caused by chemotherapy or radiotherapy),transformation of donor cells by antigenic stimulation throughhost tissue, and fusion of donor cells with residual leukemiccells. It is important to consider that both intrinsic cell factorsand external signals from the hematopoietic microenvironmentare known to be involved in leukemogenesis.
The clinicopathologic features of DCL have not been wellcharacterized. Most cases published to date have abnormalkaryotype, and to our knowledge, only 2 cases with normalkaryotype have been reported [12,13]. Such cases wouldhave other underlying molecular aberrations that might leadto DCL [13].
The patient reported here developed AML arising indonor cells after acquiring a de novo somatic mutation. Toour knowledge, this is the first report of NPM1 mutation in aDCL (regardless of the source of progenitors used for SCT).NPM1 is one of the genes most commonly mutated in AML[14]. Patients with AML and this mutation have distinctclinical, cytogenetic, molecular, and immunologic featuresand should be included in a separate provisional entity byrecommendation of the World Health Organization. NPM1is involved in critical cell functions, and NPM1 mutation hasbeen suggested as a founder genetic mutation in AML [14],but the molecular mechanism by which NPM1 mutationinduces leukemia is unknown. The host environment inwhich the original malignancy developed might havetriggered the oncogenic process in donor cells, favored bythe immunosuppressive status after transplantation. In fact,NPM1 mutation appeared 16 months after UCBT, whereasit has not been reported in childhood leukemia under the ageof 3 years [15]. Although we have identified a single-genemutation, which could itself account for the “leukemization”of donor cells, a multiple genetic-hits mechanism of DCL hasbeen proposed recently after a case with multiple mutationsof CEBPA, contributing to the transformation of donor cellsto the leukemic phenotype [13]. Interestingly, underlyingmechanisms overlap between the development of DCL andother forms of leukemia, and continued research tocharacterize DCL will help to better understand cancerstem cells and the stem cell niche [10]. A systematicmolecular screening of relapsed patients would be of key
importance to establish the true incidence of DCL and to geta deeper insight into this serious complication after SCT.
Supplementary data
Supplementary data to this article can be found online athttp://dx.doi.org/10.1016/humpath.2013.01.001
References
[1] Goh K, Fialkow PJ, Thomas ED, Bryant JI, Neiman PE. Leukaemictransformation of engrafted human cells in vivo. Lancet 1971;2:101-2.
[2] Hertenstein B, Hambach L, Bacigalupo A, et al. Development ofleukemia in donor cells after allogeneic stem cell transplantation: asurvey of the European Group for Blood and Marrow Transplantation(EBMT). Haematologica 2005;90:969-75.
[3] Fraser CJ, Hirsch BA, Dayton V, et al. First report of donor cell-derived acute leukemia as a complication of umbilical cord bloodtransplantation. Blood 2005;106:4377-80.
[4] Crow J, Youens K, Michalowski S, et al. Donor cell leukemia inumbilical cord blood transplant patients: a case study and literaturereview highlighting the importance of molecular engraftment analysis.J Mol Diagn 2010;12:530-7.
[6] Spinelli O, Giussani U, Borleri G, et al. Need for an accurate moleculardiagnosis to assess the donor origin of leukemia relapse afterallogeneic stem cell transplantation. Haematologica 2000;85:1153-7.
[7] Falini B, Mecucci C, Tiacci E, et al. Cytoplasmic nucleophosmin inacute myelogenous leukemia with a normal karyotype. N Engl J Med2005;352:254-66.
[8] Ribera J, Oriol A, Morgades M, et al. Treatment of high-risk (HR)Philadelphia chromosome-negative (Ph−) adult acute lymphoblasticleukemia (ALL) according to baseline risk factors and minimalresidual disease (MRD). Results of the PETHEMA ALL-AR-03 trialincluding the use of propensity score (PS) method to reduceassignment bias. Blood 2009;114:322.
[9] Sanz J, Boluda JCH, Martín C, et al. Single-unit umbilical cordblood transplantation from unrelated donors in patients withhematological malignancy using busulfan, thiotepa, fludarabine andATG as myeloablative conditioning regimen. Bone MarrowTransplant 2012;47:1287-93.
[10] Flynn CM, Kaufman DS. Donor cell leukemia: insight into cancerstem cells and the stem cell niche. Blood 2007;109:2688-92.
[12] Wang E, Hutchinson CB, Huang Q, et al. Donor cell-derivedleukemias/myelodysplastic neoplasms in allogeneic hematopoieticstem cell transplant recipients: a clinicopathologic study of 10 casesand a comprehensive reviewof the literature. Am JClin Pathol 2011;135:525-40.
[13] Xiao H, Shi J, Luo Y, et al. First report of multiple CEBPAmutations contributing to donor origin of leukemia relapse afterallogeneic hematopoietic stem cell transplantation. Blood 2011;117:5257-60.
[14] Falini B, Martelli MP, Bolli N, et al. Acute myeloid leukemia withmutated nucleophosmin (NPM1): is it a distinct entity? Blood 2011;117:1109-20.
[15] Hollink IHIM, Zwaan CM, Zimmermann M, et al. Favorableprognostic impact of NPM1 gene mutations in childhood acutemyeloid leukemia, with emphasis on cytogenetically normal AML.Leukemia 2009;23:262-70.
