Mutations in exon 12 of the nucleophosmin (NPM1) gene occur in about 60% of adult AML with normal karyotype. By exploiting a specific feature of NPM1 mutants, that is insertion at residue 956 or deletion/insertion at residue 960, we developed highly sensitive, real-time quantitative (RQ) polymerase chain reaction (PCR) assays, either in DNA or RNA, that are specific for various NPM1 mutations. In all 13 AML patients carrying NPM1 mutations at diagnosis, cDNA RQ-PCR showed >30 000 copies of NPM1-mutated transcript. A small or no decrease in copies was observed in three patients showing partial or no response to induction therapy. The number of NPM1-mutated copies was markedly reduced in 10 patients achieving complete hematological remission (five cases: <100 copies; five cases: 580–5046 copies). In four patients studied at different time intervals, the number of NPM1 copies closely correlated with clinical status and predicted impending hematological relapse in two. Thus, reliable, sensitive RQ-PCR assays for NPM1 mutations can now monitor and quantify MRD in AML patients with normal karyotype and NPM1 gene mutations.
A large body of evidence shows minimal residual disease (MRD) in human leukemia significantly correlates with clinical outcome, making MRD monitoring important for clinical decisions.1, 2, 3 Although immunophenotypical investigation of MRD predicts outcome in patients with acute myeloid leukemia (AML),3 large panels of monoclonal antibodies are needed to cover all myeloid lineages, several blast cell subpopulations may coexist at diagnosis and the method has never been applied in large multicenter clinical protocols. Thus, in AML, easy and reliable MRD monitoring is possible in only about 30% of patients with genetic markers that polymerase chain reaction (PCR) detects, including those carrying inv(16),4, 5 t(8;21),5 and t(15;17)2 balanced chromosomal translocations.
For the 40–50% of AML patients in different clinical trials who, at conventional cytogenetics/FISH, show normal karyotype no satisfactory method is available for monitoring MRD. Detection of FLT3 mutations6, 7 and overexpression of the Wilms' tumor (WT1) gene8 have been proposed. However, FLT3 mutations, which are usually considered secondary abnormalities, are not stable during follow-up and are detectable in only 30% of AML with normal karyotype.6 Drawbacks are also associated with monitoring WT1 gene overexpression, since the assay has limited sensitivity and can be applied to a minority of AML patients.8
We recently identified mutations occurring at exon-12 of the nucleophosmin (NPM1) gene as the most specific and frequent genetic lesion in adult9 and pediatric10 AML patients with normal karyotype. NPM1 mutations cause alterations at the C-terminus of the NPM mutant proteins9, 11 which are responsible for their aberrant cytoplasmic expression. About 40 heterozygous NPM1 mutations have been identified to date.12, 13, 14, 15, 16 Mutation A, due to a TCTG tetranucleotide duplication at position 956, accounts for over 75% of adult cases and another 15–20% show mutation B or D due to insertion of a CATG or CCTG tetranucleotide at position 959. By exploiting features of the mutated DNA/RNA NPM1 sequence, we developed highly sensitive, specific, reliable real-time quantitative (RQ) PCR assays for both DNA and RNA, which are able to monitor all the variant NPM1 mutations in adult AML with normal karyotype.
Materials and methods
Patients and samples
Fresh or thawed bone marrow and/or peripheral blood mononuclear cells from AML patients at diagnosis and/or at different follow-up time points were separated using Ficoll–Hypaque (Pharmacia LKB, Uppsala, Sweden) and subjected to RNA extraction using the RNAwiz reagent (Ambion, Austin, TX, USA) or to DNA extraction by the standard phenol–chlorophorm method.
A PCR amplification product, containing mutation A, generated from patient RNA by RT–PCR using primers NPM1_390_F (5′-IndexTermIndexTermGGTCTTAAGGTTGAAGTGTGGT-3′) and NPM1_1043_R (5′-IndexTermIndexTermTCAACTGTTACAGAAATGAAATAAGACG-3′)9 was cloned into plasmid vector pCR II-TOPO (Invitrogen, Groningen, The Netherlands), as described.17 Plasmid DNA concentration was determined by absorbance measurement; five serial plasmid dilutions (105, 104, 103, 102, 10 copies) were prepared. Sequential dilutions were amplified by RQ-PCR to construct a standard curve for the absolute quantitative assessment of copy number. The NPM1 mutation A value was normalized on the number of Abelson (ABL) transcripts and expressed as the number of NPM1 mutation A copies every 104 copies of ABL.17
Real-time quantitative-polymerase chain reaction assays for cDNA
Real-time quantitative-polymerase chain reaction assays shared a common forward primer and probe, but differed on the mutation specific reverse primer. We focused on mutations A and B, which account for almost 90% of all NPM mutations in AML. Probe and primers were designed by Primer Express software (Applied Biosystems, Foster City, CA, USA): the forward primer in exon 11 (cNPM1F), probe in exon 11/exon 12 junction (5′-FAM-3′-MGB) and reverse primers, in exon 12 (cNPM-mutA-R and cNPM-mutB-R) (Figure 1). Reverse transcription was performed on 1 μg RNA according to EAC protocol.17 Real-time quantitative-polymerase chain reaction mixture reaction contained 12.5 μl Taq Man Universal PCR Master Mix (Applied Biosystems), 300 nM Primers, 200 nM probe and 5 μl cDNA (1/10 of RT product), in a total volume of 25 μl. We used ABI PRISM 7700 Sequence Detection System (Applied Biosystems), for sample amplification and analysis. Amplification conditions were: 2 min at 50°C (UNG enzyme activation), 10 min at 95°C (UNG enzyme inactivation and AmpliTaq polymerase activation) followed by 50 cycles at 95°C for 15 s and at 62°C for 1 min for mutation A and at 59 for 1 min for mutation B. To correct for quantity and quality of RNA samples, the ABL gene RNA was amplified. A threshold value of 0.1 was used and baseline was set to 3–15 either for ABL or for NPM1;17 all analyses were performed in duplicate. To assess sensitivity and specificity, 10-fold serial dilutions were performed mixing RNA from bone marrow cells carrying NPM1 mutation A or B and RNA from a bone marrow pool of patients without NPM1 mutations.
Real-time quantitative-polymerase chain reaction assay for genomic DNA
Specific forward primers for NPM1 mutations were designed using Primer Express software (Applied Biosystems) to anneal to the mutated region of NPM1 exon 12. To simplify and optimize the RQ-PCR procedure, a common reverse primer and probe were designed on the exon 12 sequence (Primer Express software). A specific reverse primer was designed for the NPM1 type A mutation (Figure 1). The TaqMan PCR core reagent kit was used (Applied Biosystems). Reaction mixtures of 25 μl contained 2.5 μl TaqMan buffer A, 5 mM MgCl2, 0.3 mM dATP, 0.3 mM dCTP, 0.3 mM dGTP, and 0.6 mM dUTP, 5 μ M primers, 1.25 U AmpliTaq Gold and 500 ng DNA. The amplification protocol consisted of 10 min at 95°C (denaturation of target DNA, and activation of AmpliTaq Gold), followed by target amplification via 50 cycles of 15 s at 95°C and 1 min at 61°C. Real-time analysis was performed on the ABI PRISM 7900 Sequence Detection System (SDS) containing a 96-well thermal cycler (Applied Biosystems); all samples were tested in triplicate. Real-time quantitative-polymerase chain reaction was optimized for each NPM1 mutation by modifying annealing temperature to increase sensitivity and specificity. For all NPM1 mutations, the best amplifications were obtained using 61°C as annealing temperature. To correct for quantity and quality of DNA in follow-up samples, the Albumin gene DNA was amplified (TaqMan® Control Genomic DNA, Applied Biosystems). Standard curves for NPM1 and Albumin were established by amplifying a 10-fold serial dilution of target DNA in DNA from healthy donor peripheral blood mononuclear cells or water, respectively. The values for NPM1 mutations were divided by the albumin value. The diagnostic value was set as 1.0. NPM1 specific quantities in follow-up samples were related to quantity at diagnosis.
In accordance with the European Study Group current guidelines on MRD in leukemia,18 in all results ‘maximal reproducible sensitivity’ was defined as the lowest dilution in which all replicates were positive within a 1.5 cycle threshold (Ct) range, and the highest Ct of replicates was at least 3.0 Ct lower than the lowest background amplification replicate. ‘Maximal sensitivity’ was defined as the lowest dilution with at least one positive replicate, and at least 1.0 Ct lower than the lowest background amplification replicate. Amplifications in 1 of 2 wells, below the maximal reproducible sensitivity but still 1.0 Ct lower than the lowest background replicate, are classified as ‘positive, not quantifiable’.
Real-time quantitative-polymerase chain reaction assays
The feasibility of using NPM1 mutations as a marker for MRD detection was assessed in RNA and DNA.
Real-time quantitative-polymerase chain reaction sensitivity and specificity were tested in 10-fold serial plasmid dilutions. Figure 2 shows Ct values for each replicate and curve slopes for each plasmid amplification experiment. Plasmid standard curves showed a mean slope of −3.38 and intercept of 39.5±0.45 Ct. High correlation coefficients (>0.99 in all experiments) allowed accurate assessment of NPM1 mutation A copies in unknown samples. Maximal reproducible sensitivity corresponded to 10 plasmid molecules (Figure 2). To mimic real MRD assessment, cDNA RQ-PCR sensitivity and specificity were tested in 10-fold serial dilutions of diagnostic RNAs from five AML cases (four mutations A; one mutation B) (Table 1). Maximal reproducible sensitivity was 10−4 in all. Maximal sensitivity was 10−6 in three of four AML cases with mutation A and 10−5 in the other. Patient 5, carrying mutation B, had a 10−6 sensitivity (Table 1). The assay was highly specific as the background was amplified in only one case, at very high Ct (Ct>48).
Real-time quantitative-polymerase chain reaction sensitivity and specificity was tested on DNA from 15 AML patients (nine mutations type A; two type D; and one of each type B, E, G and H).9, 10 We performed 10-fold serial dilutions of diagnostic DNA in DNA from a pool of peripheral blood mononuclear cells from five healthy donors. The same fluorescent probe was used for all mutations combined with a specific forward primer for each NPM1 mutation type (see Figure 1). A common reverse primer was developed for all NPM1 mutations, except for mutation A for which we designed a specific primer, in order to improve specificity and sensitivity (see Figure 1). All amplification plots were analyzed by positioning the Ct at 0.1. In all cases, this value was in the linear range of all dilutions. At least 10−4 sensitivity was reached in all cases but one; maximal reproducible sensitivity was at least 10−4 for eight patients and 10−3 for seven patients (Table 1). Maximal reproducible sensitivity was lowest for mutation A, probably because of its 4 nucleotide tandem duplication. Cases with a longer nucleotide insertion displayed a higher sensitivity, as expected. The assay was highly specific as the background was amplified in only two cases, at very high Cts (Table 1). The mean Ct of undiluted DNA samples was 23.3 (range 22.1–24.8). The mean dilution curve slope was −3.6 (range: −2.9–4.0). High correlation coefficients (0.99 in all but one) allowed accurate assessment of the quantity of NPM1 in unknown samples.
Monitoring minimal residual disease in NPMc+ acute myeloid leukemia
Using the plasmid calibration curve (see Figure 2), we analyzed samples from 13 AML patients carrying NPM1 mutation A at diagnosis and after induction therapy who were similar in terms of FAB type; % of blasts and clinical features. Results were expressed as the NPM1 mutation A copy number for every 104 copies of ABL (Figure 3a). More than 55 000 mutated copies were found in all but one (>30 000 copies) of these 13 cases. After induction therapy, the number of copies dropped markedly in 10 patients who achieved complete hematological remission. Five reached <70 copies; the other five had between 580 and 5046 copies (Figure 3a). A small or no decrease in NPM1-mutated copies was observed in three patients (one partial hematological remission; two chemoresistant cases) (Figure 3a). In three representative cases (Figure 3b) of these 13 patients with NPM1 mutation A, cDNA RQ-PCR showed <10 copies after the first or second cycle of consolidation therapy in all three but the kinetics were different in each patient. A few persistent NPM1-mutated copies were associated with hematological remission in one. In the other two, the number of copies rose again and hematological relapse was diagnosed at different time-points.
In four cases we monitored MRD by both NPM1-mutated allele expression and WT1 gene expression at the same follow-up time-points. A representative experiment is shown in Figure 3c. Similar profiles were observed; however, costitutive WT1 expression was always detected during follow-up, while NPM-mutated expression was further decreasing in intermediate time points, increasing the range of tumor cells discrimination (Figure 3c).
In patient 15 with mutation H (Table 1), genomic RQ-PCR showed the NPM1-mutated clone decreased during therapy (Figure 3d). At complete hematological remission (3 months after induction therapy), the NPM1-mutated clone resulted ‘positive, not quantifiable’ (below 10−4 maximal reproducible sensitivity, with 10−5 maximal sensitivity). After autologous bone marrow transplantation, MRD was negative (<10−5). The patient is still in molecular remission after 1.5 years.
We describe highly sensitive, specific, reliable assays for assessing MRD in AML patients carrying mutations in exon 12 of the NPM1 gene.9 This subgroup, named NPMc+ AML, accounts for about one-third of adult AML9 and 7% of childhood AML,10 that is, respectively, 60% and 25% of adult and pediatric AML with normal karyotype. NPMc+ AML exhibits distinct clinical and pathological features: frequent M4 and M5 FAB morphology, multilineage involvement, no CD34 expression, and high frequency of FLT3 mutations9. NPMc+ AML shows distinct gene expression profiling,19 better response to induction therapy,9 and more favorable long-term survival than AML with normal karyotype without NPM1 mutations.14, 15, 16
Although we know at least 40 NPM1 mutation variants have been identified in 1839 sequenced AML cases,12, 13, 14, 15, 16 the RQ-PCR assays we developed allow us to analyze >95% of NPMc+ AML patients because mutation A accounts for 70–80% of all NPM1 mutations,12, 13, 14, 15, 16 and mutations B and D for an additional 15–20%. The same RQ-PCR assays can be adapted to AML patients carrying sporadic (<5%) NPM1 mutations, by using forward or reverse mutation-specific primers. Since mutation A is most prevalent, a plasmid containing the type A sequence was used as calibrator for absolute quantification in cDNA PCR assay. To simplify and standardize the method for routine use, a single fluorescent MGB-probe was designed on NPM1 exon 11-exon 12 junction, and a common forward primer was designed on NPM1 exon 11. Using a exon 12 mutation-specific reverse primer and a probe positioned on the exon junction limits amplification to the cDNA target, excluding contaminating genomic DNA.
Our RQ-PCR assays can be applied on RNA and DNA. The cDNA assay seems to be slightly more sensitive than the genomic perhaps because of the relatively high expression of NPM transcripts. cDNA RQ-PCR can be applied to samples collected in routine diagnostic testing for common translocations while genomic DNA RQ-PCR can be related directly to the number of residual leukemic cells. Therefore, independently of the type of material available, both approaches should allow retrospective analysis of AML samples. Moreover, results with the two methods can be compared in paired samples, in attempts to understand the correlation between the number of residual leukemic cells during follow-up and mutated NPM1 gene expression.
Our RQ-PCR assays are an excellent tool for detecting the presence, persistence and reappearance of leukemic cells in adult and pediatric AML with normal karyotype. In 13 NPMc+ AML patients (Figure 3a) the number of NPM1 type A-mutated transcripts closely correlated with tumor burden at diagnosis and with response to induction therapy. At diagnosis, the number of NPM1-mutated copies was >30 000 in all patients. No drop was observed in one patient with refractory disease. Only a slight decrease occurred in another nonresponder and in one case achieving partial remission. All 10 patients who obtained complete hematological remission showed a marked reduction in NPM1-mutated copies. They were heterogeneous in terms of the tumor reduction kinetics, since fewer than 100 copies persisted in five patients, while residual NPM1-mutated copies ranged in number from 580 to 5046 in the others. Interestingly, three of the five cases with under 100 copies were long-term survivors (data not shown). Many more cases need to be analyzed to establish whether the number of persistent NPM1 transcripts copies will serve (possibly in association with other parameters such as the absence of FLT3-ITD mutations)14, 15, 16 as a predictor for long-term favorable outcome in NPMc+ AML.
The reliability of NPM-based MRD detection was further supported by the simultaneous detection of mutation-specific NPM gene expression and WT1 gene expression during follow-up of four cases. Although similar profiles were observed, these preliminary data suggest that NPM could be more appropriate than WT1 for MRD monitoring in AML cases with normal karyotype. In all cases, the NPM1-mutated allele was expressed at levels higher than WT1. Moreover, at intermediate time points the constitutive WT1 expression is always detected, while NPM1-mutated allele is specific for the leukemic cell and its expression can further decrease (Figure 3c).
Finally, our results in three NPMc+ AML patients for whom information was available at diagnosis and throughout follow-up suggest our RQ-PCR assays predict impending hematological relapse (Figure 3b). This information might be useful for stratifying AML patients with normal karyotype to specific risk groups and/or for deciding to increase therapy because of early evidence of disease progression.
In our study, NPM status emerges as a stable marker of MRD. Confirming this observation, Boissel et al.13 reported 15/15 AML patients (10 NPM+; 5 NPM−) maintained the same NPM mutation pattern at diagnosis and relapse. Although Suzuki et al.12 reported 2/17 AML patients lost NPM positivity, one had normal karyotype at diagnosis but at relapse presented an acquired clonal del(20)(q1?) abnormality, indicating the relapse clone had another origin. Although these data indicate NPM mutations tend to be stable, stability needs to be confirmed in a large cohort of paired AML patients who relapse.
Our specific, sensitive RQ-PCR assays for cDNA and genomic DNA quantitatively assess NPM1-mutated gene copies and provide, for the first time, a reliable system for monitoring MRD in the majority of AML patients with normal karyotype. Future studies in large prospective clinical trials should focus on whether these assays will serve as predictors of imminent hematological relapse and long-term survival.
van Dongen JJ, Seriu T, Panzer-Grumayer ER, Biondi A, Pongers-Willemse MJ, Corral L et al. Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood. Lancet 1998; 352: 1731–1738.
Diverio D, Rossi V, Avvisati G, De Santis S, Pistilli A, Pane F et al. Early detection of relapse by prospective reverse transcriptase-polymerase chain reaction analysis of the PML/RARalpha fusion gene in patients with acute promyelocytic leukemia enrolled in the GIMEMA-AIEOP multicenter ‘AIDA’ trial. GIMEMA-AIEOP Multicenter ‘AIDA’ Trial. Blood 1998; 92: 784–789.
Vidriales MB, San-Miguel JF, Orfao A, Coustan-Smith E, Campana D . Minimal residual disease monitoring by flow cytometry. Best Pract Res Clin Haematol 2003; 16: 599–612.
Guerrasio A, Pilatrino C, De Micheli D, Cilloni D, Serra A, Gottardi E et al. Assessment of minimal residual disease (MRD) in CBFbeta/MYH11-positive acute myeloid leukemias by qualitative and quantitative RT–PCR amplification of fusion transcripts. Leukemia 2002; 16: 1176–1181.
Krauter J, Gorlich K, Ottmann O, Lubbert M, Dohner H, Heit W et al. Prognostic value of minimal residual disease quantification by real-time reverse transcriptase polymerase chain reaction in patients with core binding factor leukemias. J Clin Oncol 2003; 21: 4413–4422.
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.
Beretta C, Gaipa G, Rossi V, Bernasconi S, Spinelli O, Dell'Oro MG et al. Development of a quantitative-PCR method for specific FLT3/ITD monitoring in acute myeloid leukemia. Leukemia 2004; 18: 1441–1444.
Cilloni D, Gottardi E, De Micheli D, Serra A, Volpe G, Messa F et al. Quantitative assessment of WT1 expression by real time quantitative PCR may be a useful tool for monitoring minimal residual disease in acute leukemia patients. Leukemia 2002; 16: 2115–2121.
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.
Cazzaniga G, Dell'Oro MG, Mecucci C, Giarin E, Masetti R, Rossi V et al. Nucleophosmin mutations in childhood acute myelogenous leukemia with normal karyotype. Blood 2005; 106: 1419–1422.
Nakagawa M, Kameoka Y, Suzuki R . Nucleophosmin in acute myelogenous leukemia. N Engl J Med 2005; 352: 1819–1820; author reply 1819–1820.
Suzuki T, Kiyoi H, Ozeki K, Tomita A, Yamaji S, Suzuki R et al. Clinical characteristics and prognostic implications of NPM1 mutations in acute myeloid leukemia. Blood 2005; 106: 2854–2861.
Boissel N, Renneville A, Biggio V, Philippe N, Thomas X, Cayuela JM et al. Prevalence, clinical profile and prognosis of NPM mutations in AML with normal karyotype. Blood 2005; 106: 3618–3620.
Dohner 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.
Schnittger S, Schoch C, Kern W, Mecucci C, Tschulik C, Martelli MF et al. Nucleophosmin gene mutations are predictors of favourable prognosis in acute myelogenous leukemia with a normal kayotype. Blood 2005; 106: 3733–3739.
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.
Gabert J, Beillard E, van der Velden VH, Bi W, Grimwade D, Pallisgaard N et al. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia – a Europe Against Cancer program. Leukemia 2003; 17: 2318–2357.
van der Velden VH, Hochhaus A, Cazzaniga G, Szczepanski T, Gabert J, van Dongen JJ . Detection of minimal residual disease in hematologic malignancies by real-time quantitative PCR: principles, approaches, and laboratory aspects. Leukemia 2003; 17: 1013–1034.
Alcalay M, Tiacci E, Bergomas R, Bigerna B, Venturini E, Minardi SP et al. Acute myeloid leukemia bearing cytoplasmic nucleophosmin (NPMc+ AML) shows a distinct gene expression profile characterized by up-regulation of genes involved in stem-cell maintenance. Blood 2005; 106: 899–902.
This work was supported by the Associazione Italiana per la Ricerca sul Cancro (AIRC), Ministero Istruzione Università e Ricerca, Associazione Umbra Studio e Terapia Leucemie e Linfomi (AULL), Fondazione Cassa di Risparmio di Perugia Fondazione Tettamanti, Fondazione Cariplo. Federica Alberti's work was supported by the Progetto Lagrange-Fondazione CRT e Fondazione ISI Turin. The authors would like to thank Marco Cappelletti (Applera Italia) and Amplimedical SpA (Buttigliera Alta, Turin, Italy) for collaboration and technical support.
About this article
- acute myeloid leukemia
- normal karyotype
- minimal residual disease
- quantitative polymerase chain reaction.
Case Reports in Hematology (2019)
Utility of Immunophenotypic Measurable Residual Disease in Adult Acute Myeloid Leukemia—Real-World Context
Frontiers in Oncology (2019)
Surgical Pathology Clinics (2019)
European Journal of Haematology (2019)