Specific detection of Flt3 point mutations by highly sensitive real-time polymerase chain reaction in acute myeloid leukemia

ORIGINAL ARTICLES Specific detection of Flt3 point mutations by highly sensitive real-time polymerase chain reaction in acute myeloid leukemia SEBASTIAN SCHOLL, CLAUDIA KRAUSE, IVAN F. LONCAREVIC, ROUVEN MÜLLER, CHRISTA KUNERT, ULRICH WEDDING, HERBERT G. SAYER, JOACHIM H. CLEMENT, and KLAUS HÖFFKEN JENA, GERMANY

Among activating class III receptor tyrosine kinase (Flt3) mutations, internal tandem duplications of Flt3 (Flt3-ITD) are detected in about 25% of patients with acute myeloid leukemia (AML). In contrast, mutations within the tyrosine kinase domain of Flt3 (Flt3-TKD mutations) are less frequent (approximately 7%), and there are only limited data on the frequency of recently demonstrated activating Flt3 point mutation at codon 592 (Flt3V592A mutation). We evaluated a new approach for rapid screening of Flt3-TKD and Flt3-V592A mutations using the fluorescence resonance energy transfer (FRET) principle in a group of 122 patients. Based on individual Flt3-TKD mutations, we designed patientspecific primers to perform a highly sensitive polymerase chain reaction (PCR) assay for rapid detection of minimal residual disease (MRD). We also used a model system with MonoMac-6 cells carrying the Flt3-V592A mutation to establish a mutation-specific real-time PCR approach also for this molecular aberration. We identified 9 cases (8%) of Flt3-TKD mutations (5 cases of mutation D835Y, 3 cases of mutation D835H, and 1 case of mutation Del836), and no cases of Flt3-V592A mutation. Screening for Flt3-TKD mutations with fluorescent probes is equivalent to conventional screening using standard PCR followed by EcoRV restriction. We present a real-time PCR protocol that can be used for MRD analyses based on individual Flt3-TKD mutations. Examples of MRD analyses are presented for all 3 subtypes of Flt3-TKD mutation identified in this study. In summary, we demonstrate new methodological approaches for rapid screening of Flt3 point mutations and for detection of MRD based on patient-specific Flt3-TKD mutations. (J Lab Clin Med 2005;145:295–304) Abbreviations: AML ⫽ acute myeloid leukemia; FL ⫽ Flt3 ligand; Flt3 ⫽ class III receptor tyrosine kinase; Flt3-TKD ⫽ tyrosine kinase domain of Flt3; FRET ⫽ fluorescence resonance energy transfer; MRD ⫽ minimal residual disease; PBSCT ⫽ peripheral blood stem cell transplantation; PCR ⫽ polymerase chain reaction

F

lt3 is a class III tyrosine kinase receptor that confers proliferative and anti-apoptotic effects on normal and leukemic hematopoietic stem cells.1– 6 Mutations of the Flt3 gene represent the most frequent and one of the best-characterized alterations in acute

myeloid leukemia (AML).7,8 Flt3-ITD mutations (ie, internal tandem duplications within the juxtamembrane domain) can be detected in about 25% of AML patients at primary diagnosis.9 –11 In contrast, mutations within the kinase domain of Flt3 are less frequent, with an

From the Department of Internal Medicine II (Oncology and Hematology) and Institute of Human Genetics, Friedrich Schiller University, Jena, Germany.

Reprint requests: Sebastian Scholl, Department of Internal Medicine II, Erlanger Allee 101, 07740 Jena, Germany; e-mail: sebastian. [email protected]. 0022-2143/$ – see front matter © 2005 Mosby, Inc. All rights reserved. doi:10.1016/j.lab.2005.03.005

Supported by the Dr. Rainald Stromeyer Foundation, Germany. Submitted for publication September 16, 2004; accepted for publication March 5, 2005.

295

296

J Lab Clin Med June 2005

Scholl et al

incidence of about 7%.11–13 Among these aberrations, point mutations of codon 835 (eg, single base exchange, leading to replacement of aspartate by tyrosine or histidin) and deletion of codon 836 can be regularly observed. These mutations are known as Flt3-TKD mutations. Flt3-ITD and Flt3-TKD mutations confer a constitutive activity of Flt3 in a Flt3 ligand (FL)-independent manner, leading to permanent stimulation of different signalling pathways, such as activation of Stat5 and Akt and different modulation of bcl-2 family members.14 –19 Furthermore, several investigations have demonstrated that activating mutations of Flt3 lead to a worse prognosis.10 –12,20 –24 Until now, the conventional assays used for screening of Flt3-TKD mutations as well as the test principle introduced by Small’s group were based on classical polymerase chain reaction (PCR) techniques followed by EcoRV digestion and electrophoretic analysis of cleavage products.12,25 Both methods represent longlasting procedures and thus are rather time-consuming. Furthermore, false-positive results can arise from incomplete EcoRV restriction of larger amounts of amplified primary fragments.26 EcoRV cleavage-based methods can detect changes only within codons 835 and 836 of the Flt3 gene, representing the EcoRV recognition sequence. In the meantime, other mutations, including some polymorphisms within the Flt3 gene, were identified. In particular, silent mutations of exons 13 and 14 were demonstrated in 4 of 34 AML patients by denaturing high-pressure liquid chromatography (dHPLC); the same study also demonstrated the first point mutation within the N-terminal kinase domain of Flt3 (Flt3A680V mutation).26 The functional importance of the latter aberration has not yet been elucidated. In contrast, other mutations of Flt3 besides Flt3-ITD or Flt3-TKD mutations have been evaluated with respect to functionality and frequency in AML patients.27 Thus Spiekermann et al28 identified a new activating point mutation of Flt3 within the juxtamembrane domain (codon 592), termed the Flt3-V592A mutation, in MonoMac-6 cells and its parental cell line MonoMac-1. So far, only 1 study has investigated the frequency of the Flt3-V592A mutation in AML patients.29 In contrast with Flt3-TKD mutations, the point mutation at codon 592 (gtt ¡ gct) does not lead to the loss of a recognition site of a restriction enzyme, and a new cleavage motif is not acquired by this point mutation. Importantly, a simple screening for this Flt3 mutation is not possible by enzymatic digestion of PCR products generated with codon 592 flanking primers. In general, analysis of minimal residual disease (MRD) in patients with leukemia may represent a pow-

erful and independent prognostic indicator.30,31 Individual long-term analyses of residual leukemia cells by PCR techniques has been associated with an increased risk of relapse. Thus measurement of MRD can lead to a better and more reliable stratification of patients in terms of further therapeutic options, such as allogeneic stem cell transplantation. Until now, different methodological approaches for detecting only single base mutations have been described; real-time PCR techniques enable precise and specific quantification of such alleles and transcripts.32–34 In this article we report a new method for rapid detection of Flt3-TKD and Flt3-V592A mutations, as well as a sensitive and specific PCR assay especially for quantifying different Flt3-TKD mutations. We also discuss how this new approach can be used for monitoring of MRD in patients with AML. METHODS Patients and patient samples. This retrospective study was carried out according to the principles of the Declaration of Helsinki. We obtained informed consent from 122 patients treated for AML at our clinic. Bone marrow aspirate specimens from patients with AML at diagnosis or at different time points during therapy were drawn regularly for routine diagnostic tests. Remaining bone marrow cells of this material underwent erythrocyte lysis, and pellets were frozen at ⫺ 80°C. The DNA Blood Mini Kit (Qiagen, Hilden, Germany) was used to isolate genomic DNA. In some cases, slides with bone marrow smears were also used as sources for DNA isolation. Conventional screening for Flt3-TKD mutations. Investigation of point mutations and deletions within codons 835 and 836 creating an EcoRV cleavage site were carried out as described by Yamamoto et al.12 In contrast to that study, however, we used an alternative primer pair flanking codon 835 for amplification by PCR, with primers located in intron 16 (D835fw) and exon 17 (D835rv). The primer sequences are given in Table I, and the PCR conditions are indicated in Table II. The PCR setup was as follows (final concentrations): 100 to 200 ng of genomic DNA, 0.25 mmol each of dNTP’s, 10 ⫻ PCR buffer containing 2.5 mmol MgCl2, 1 ␮mol each of D835fw and D835rv primers, 2.5 U of Taq polymerase, and sterile water up to a final 25 ␮L. Subsequently, 10 ␮L of PCR product was incubated with 2.5 U EcoRV at 37°C for 1 hour, and a total of 20 ␮L was separated on 2% agarose gel. Primary PCR products, as well as fragments refractory to EcoRV, were expected at 228 bp. Screening for Flt3-TKD mutations using real-time PCR based on fluorescence resonance energy transfer. For

rapid detection of mutations within codons 835 and 836 as well as within adjacent codons of the Flt3 gene, we established a LightCycler assay (Roche, Penzburg, Germany) based on the FRET principle using a wild-type homologous probe. The amplification of fragments containing codon 835 of exon 17 was done using the same primers applied for EcoRV screening. Figure 1 illustrates a scheme demonstrating the principle of probe design and location of primers for

J Lab Clin Med Volume 145, Number 6

Scholl et al

297

Table I. Primers and probes: Overview of primers and fluorescence-conjugated probes used for screening of Flt3-TKD and Flt3-V592A mutations and analyses of minimal residual disease

1 2 3 4 5 6 7 8 9 10 11 12

Primers and probes

Sequences 5= ¡ 3=

Modification

D835fw D835rv V592fw V592rv D835anchor D835sensor V592anchor V592Asensor D835Y D835H De1836 V592Amutrv

CC GGTACCTCCT ACTGAAG GCA GCC TCA CAT TGC CCC TTC ATT GTC GTT TTA ACC CT CAA ATG GTG AGT ACG TGC GGG AAA GTG GTG AAG ATA TGT GAC TTT GG G GCT CGA GAT ATC ATG AGT GAT T GGT ACA GGT GAC CGG CTC CTC AGA T ATG AGT ACT TCT ACG CTG ATT TCA GAG AA A GTT GGA ATC ACT CAT GAT GTA A GTT GGA ATC ACT CAT GAT GTG ATA GTT GGA ATC ACT CAT ATC TAT TCA TAT TCT CTG AAA TGA G

No No No No 3= - FL 5= -LC640 3= - FL 5=-LC640 No No No No

“GAT ATC” of the D835 sensor probe reflects the EcoRV recognition site; “CTG” within the V592A sensor probe represents the mutation of codon 592. The underlined nucleotides of the mutation-specific primers 9 and 10 indicate the point mutation of codon 835 (terminal nucleotide) and the additional nucleotide exchange at position 3.

Table II. PCR conditions for screening and MRD analyses

Conventional screening for Flt3-TKD mutation Real-time PCR screening (TKD and V592A) Standard PCR for MRD (TKD1 and V592A2) Real-time PCR for MRD (TKD1 and V592A2)

Denaturation

Annealing

Elongation

Cycles

92°C / 30 sec 92°C / 20 sec 92°C / 30 sec

56°C / 30 sec 58°C / 15 sec 62°C / 30 sec1 56°C / 30 sec2 62°C / 10 sec1 56°C / 10 sec2

72°C / 30 sec 72°C / 20 sec 72°C / 30 sec

35 45 35

72°C / 20 sec

45

92°C / 20 sec

Characteristic temperature and incubation time for each step of amplification is indicated for all PCR techniques applied. Real-time PCR for screening of point mutations was carried out by means of fluorescence-labeled sequence-specific probes, whereas for real-time PCR for MRD analyses, point-mutation–specific primers were used in a SYBRGreen-based PCR setting.

detection of mutations by fluorescence resonance energy transfer (FRET). The PCR conditions are as given in Table II, and the PCR setup was as follows (final concentrations): 20 to 100 ng of genomic DNA, 2.4 mmol of MgCl2, 0.4 mmol of anchor probe, 0.2 mmol of sensor probe, 0.5 ␮mol each of D835fw and D835rv primers, 10 ⫻ master mix, and sterile water up to a final 20 ␮L. A minimal amount of at least 20 ng of genomic DNA was sufficient for qualitative analysis of Flt3-TKD mutations at diagnosis. During acquisition, the FL2/FL1 mode was used. Screening for Flt3-V592A mutation by real-time PCR based on the FRET principle. The PCR setup was as follows

(final concentrations): 20 to 100 ng of genomic DNA, 2.4 mmol of MgCl2, 0.4 mmol of anchor probe, 0.2 mmol of sensor probe, 0.5 ␮mol each of V592fw and V592rv primers, 10 ⫻ master mix, and sterile water up to a final 20 ␮L. The PCR conditions were as given in Table II (FL2/FL1 mode). THP-1 cells were used as a negative control, and MonoMac-6 cells were used as a positive control to establish the assay system. The existence of the wild-type codon 592 in THP-1 cells and the homologous Flt3-V592A mutation in MonoMac-6 cells was determined by amplification of a genomic PCR fragment with the primer pair V592fw/V592rv and confirmed by sequencing. While the Flt3-TKD sensor probe matched the

wild-type sequence of the Flt3 gene, the Flt3-V592A sensor probe was designed to be homologous with the Flt3-V592A mutation. Cloning and sequencing of individual Flt3-TKD mutations. For identification of patient-specific point mutations or

deletions within codons D835 and 836 of the Flt3 gene, conventional detection of these mutations (via PCR, EcoRV digestion, or gel electrophoresis) was repeated, and the PCR fragment within the upper band was excised using the QIAEX kit (Qiagen) according to the manufacturer’s instructions. The PCR fragments were cloned into the pCRII vector (Invitrogen, Breda, The Netherlands). Blue-white screening was performed before plasmids were isolated after overnight culture using a MiniPrep Kit (Biozyme, Hess, Germany) as recommended by the manufacturer. Plasmids were further analyzed by HindIII and EcoRV restriction. Positive clones were sequenced with M13 forward primer and M13 reverse primer using the CEQ3000 system (Beckman-Coulter, Krefeld, Germany). Flt3-TKD–specific MRD analysis using real-time PCR in a SYBRGreen-based setting. Characteristics of the patients

selected for retrospective MRD analyses are given in Table III. For detection of residual leukemia cells, we designed mutation-specific primers for D835Y and D835H mutations

298

J Lab Clin Med June 2005

Scholl et al

Examples of MRD analyses with patient samples. Genomic DNA from bone marrow samples from patients with Flt3TKD mutations during or after therapy was isolated as described earlier; approximately 500 ng genomic DNA was used for each assay. The PCR conditions and PCR setup correlated with those for the real-time PCR assay using SYBRGreen for MRD analysis based on Flt3-TKD mutations described earlier. Establishment of a Flt3-V592A– based model system for MRD. THP-1 and MonoMac-6 cells were obtained from the

Fig 1. Prinicipal of direct analyses of Flt3 mutations. Schematic illustration of the FRET principle in detecting mutations using PCR techniques with fluorescence-conjugated probes and localization of primers and probes (A). Localization of primers and probes for detection of Flt3-TKD mutations (B) and the Flt3-V592A mutation (C). The sensor probe for screening of different subtypes of Flt3TKD mutations was designed to be homologous to the wild-type sequence indicated by the unchanged EcoRV recognotion site “gatatc” spanning codon 835 and 836. In contrast, to detect the Flt3-V592A mutation, we used a sensor probe that matched to the mutation-specific sequence (gtt ¡ gct point mutation).

and for the deletion of codon 836 (see Table I). For evaluation of sensitivity and specificity, each of the plasmids with a Flt3-TKD mutation was serially diluted with wild-type genomic DNA. For D835Y mutation, the efficiency of PCR using plasmid dilutions was compared with serially diluted genomic DNA from the diagnosis as indicated. Conditions for standard PCR and real-time PCR were as given in Table II. The PCR setup for standard PCR was as follows (final concentrations): 200 or 500 ng of template DNA (dilution of plasmid or genomic DNA), 0.25 mmol each of dNTP’s, 10 ⫻ PCR buffer containing 2.5 mmol of MgCl2, 0.5 ␮mol each of D835fw primer and the mutation-specific primer, 2.5 U of Taq polymerase, and sterile water up to a final 25 ␮L. For real-time PCR, the PCR setup was as follows: 200 or 500 ng of template DNA (dilution of plasmid or genomic DNA), 0.5 ␮mol of D835fw primer and the mutation-specific primer, 5 ⫻ master mix containing SYBRGreen, and sterile water up to 20 ␮L.

DSMZ (Braunschweig, Germany) and maintained in RPMI1640 supplemented with 10% FCS. For evaluation of the potential role of Flt3-V592A mutation for MRD analysis, MonoMac-6 cells were added to peripheral blood of volunteers in a dilution series, as shown in Figure 2. After lysis of erythrocytes, genomic DNA was isolated as described earlier. Standard PCR conditions and conditions for real-time PCR using SYBRGreen were as given in Table II. The setup for standard PCR was as follows (final concentrations): 500 ng of template DNA, 0.25 mmol each of dNTP’s, 10 ⫻ PCR buffer containing 2.5 mmol of MgCl2, 0.5 ␮mol each of V592fw and V592Amutrv primers 2.5 U of Taq polymerase, and sterile water up to a final 25 ␮L. The PCR setup for real-time PCR was as follows: 500 ng of template DNA, 0.5 ␮mol each of V592fw and V592Amutrv primers, 5 %times; master mix, and sterile water up to 20 ␮L. Data analysis and control PCR for MRD analysis. Meltingcurve analysis in screening for Flt3 mutations and real-time PCR were calculated using the LightCycler software (Roche). For MRD analyses, a genomic fragment of the ␤-globin gene was amplified as an internal control with the DNA control kit (Roche). A standard curve could be constructed using the threshold cycles resulting from the dilution series with genomic DNA at diagnosis for each Flt3-TKD mutation. RESULTS Frequency of Flt3-TKD mutations: Conventional screening versus melting-curve analysis. We screened genomic

DNA obtained at primary diagnosis from 122 AML patients for Flt3-TKD mutations with both methods. We constructed new primers to achieve a better separation of PCR products after EcoRV cleavage on agarose gels (Fig. 2A). Both conventional screening and the new screening method for detection of Flt3-TKD mutations using the FRET principle were performed. One additional case in conventional screening generated an upper band after enzymatic digestion, reflecting a mutated allele. In contrast, screening with the LightCycler PCR technique did not identify this patient carrying a Flt3-TKD mutation, and we repeated the conventional analyses while extending the EcoRV restriction.26 Thus the mutation-specific band disappeared. Characteristics of the LightCycler-based assay for detection of Flt3-TKD mutations. We found characteristic

melting curves for different kinds of Flt3-TKD mutations that we identified in AML patients. The melting

J Lab Clin Med Volume 145, Number 6

Scholl et al

299

Table III. Characteristics of patients carrying Flt3-TKD mutations selected for retrospective MRD analyses Flt3-TKD mutation

Patient

FAB

1 G.S.

M1

GAT Del836

2 G.T.

M1

3 H.K.

M2

4 A.E.

M4

5 F.R.

M4

CAT ATC D835H CAT ATC D835H TAT ATC D835Y CAT ATC D835H

Bone marrow

MRD

1. Before allo PBSCT 2. ⫹27 days after PBSCT 3. ⫹144 days after PBSCT 1. After induction 2. At relapse 1. After 2 ⫻ LODAC

CR CR CR PR (6%) ⬎ 30% CR

Negative Negative Positive Negative Positive Positive

1. After consolidation 2. 18 months later 1. After induction 2. Under palliative chemotherapy

CR CR PR (7%) PR

Negative Negative Negative Positive

Time of sample acquisition

Follow-up

Relapse Relapse

Persistent CR Relapse

LODAC ⫽ Low dose cytarabin. Subtype of Flt3-TKD mutation, clinical data, bone marrow status, corresponding MRD results, and follow-up are indicated for each patient.

codon 835) or histidine (D835H, gat ¡ cat) led to additional melting peaks at 59.5°C and 59.0°C, respectively. Interestingly, using the melting-curve– based PCR technique, we could even detect a Flt3-Del836 mutation. An extra peak within the melting curve analysis (53°C) could still be detected even if the whole codon was lost. In consideration of one of the largest studies of Flt3-TKD mutations by Thiede et al,11 the Flt3-TKD mutations identified in our patients represent about 3 of 4 mutations (74.7%) within both of these codons of the Flt3 gene. Analyses of Flt3-V592A mutations in AML at diagnosis.

Fig 2. Conventional screening for Flt3-TKD and detection of Flt3 mutations by melting-curve analyses. (A) PCR products and enzymatic digestion before separation on 2% agarose gel including one patient (case 7) with a Flt3-TKD mutation. (B) Examples of screening patient samples for Flt3-TKD mutations. One patient has an additional peak at a lower temperature, representing the mutated allele. (C) Characteristic melting curves in the analysis of Flt3V592A mutation. The melting peak of MonoMac-6 cells carrying the Flt3-V592A mutation is localized at a higher temperature compared with wild-type genomic DNA of THP-1 cells.

peak of the wild-type allele homologous to the Flt3TKD sensor probe was detected at 64°C, whereas Flt3TKD mutations generated additional and characteristic melting peaks. Specifically, replacement of aspartate by tyrosine (D835Y, gat ¡ tat point mutation within

We next aimed to establish a rapid fluorescence-based screening for the Flt3-V592A mutation as well. Similar to Flt3-ITD mutations, the Flt3-V592A mutation is also localized in the juxtamembrane region, indicating the potential key function of valin at codon 592 in inhibiting Flt3 activity in the absence of FL. We could detect this aberration in MonoMac-6 cells and confirmed the functional aspect of FL-independent catalytic activity of Flt3 carrying the Flt3-V592A mutation by immunoprecipitation and phosphotyrosine western blot (data not shown). We established a fluorescence probe– based PCR assay using the FRET principle to detect Flt3-V592A mutations. Figure 2C shows the characteristic differences in melting peak analysis between the wild-type cell line THP-1 and MonoMac-6 cells carrying only Flt3-V592A alleles. We determined the frequency of Flt3-V592A mutations in patients with AML and found that, interestingly, none of the 122 AML patients we investigated carried an Flt3-V592A mutation. Sensitivity and specificity of PCR with mutant-specific

After establishing our new screening method for rapid detection of Flt3-TKD mutations, we were interested in a specific PCR approach applicable to mutation-specific primers (Flt3-D835Y, Flt3-D835H, and Flt3-Del836).

300

Scholl et al

J Lab Clin Med June 2005

Fig 3. Sensitivity and specificity of PCR with mutation-specific primers. For each detected Flt3-TKD mutation [(A) D835Y, (C) D835H, (D) Del836], different copy numbers of each cloned plasmid were added to 200 ng wild-type genomic DNA (lanes 1 to 5) as indicated. In lane 6, the PCR template was 500 ng of wild-type genomic DNA, representing a specificity control. (B) A dilution series (%) of genomic DNA (500 ng each) obtained at diagnosis of one patient with a Flt3-D835Y mutation to compare the efficiency of PCR using different templates (A vs B). The lengths of the specific PCR products were 201 bp for D835H and D835Y and 200 bp for Del836. (E) Real-time PCR for different copy numbers of Flt3-D835H plasmids (see C) including the control reaction. (F) Product specificity represented by melting analysis.

detection of residual cells. Consequently, all EcoRV refractory PCR fragments from patients carrying mutations within codons 835 and 836 of the Flt3 gene were cloned, and mutations were confirmed by sequencing. We found that 5 patients had a D835Y mutation, 3 patients had a D835H mutation, and 1 patient had a deletion of codon 836. We demonstrated that detection of all 3 identified mutations can be performed in a highly sensitive and very specific manner (Figs. 3A, C, and D). Furthermore, we compared the efficiency of different kinds of “gene equivalents” as templates (plasmid DNA vs genomic DNA); this efficiency for the Flt3-D835Y mutation is shown in in Figure 3 (A vs B). Our data demonstrate that genomic DNA also can be used for such MRD approaches. Figures 3E and F show the real-time PCR curves as well as the high specificity of mutation-specific PCR for the Flt3D835H mutation. Analogous results, including an equivalent specificity, were achieved with PCR assays

for detection of Flt3-D835Y and Flt3-Del836 mutations (data not shown). Figure 4 shows an example of MRD quantification in a patient with an Flt3-D835H mutation. Examples of MRD analyses in patient samples using mutant-specific primers for Flt3-D835Y, Flt3-D835H, and Flt3Del836 mutations. We next attempted to evaluate the

potential clinical relevance of the detection of residual cells. Thus we selected patients with bone marrow samples acquired at several time points after diagnosis, and performed MRD analyses with mutation-specific primers. Table III gives an overview of the relevant clinical data for the patients that we investigated for remaining leukemia cells at different time points during or after therapy. Note, for example, that patient 1 underwent an allogeneic peripheral blood stem cell transplantation (PBSCT). We can show that the malignant clone was already detectable about 5 months after the PBSCT—about half a year before the relapse of AML was observed. Patient 2 already had a relapse when the

J Lab Clin Med Volume 145, Number 6

Scholl et al

301

Fig 4. Example of MRD determination in a patient with Flt3-D835H mutation. A standard curve was generated using the number of threshold cycles based on a dilution series containing different amounts (10, 30, 100, 300, and 1000 copies) of the plasmid containing the specific point mutation in a background of 500 ng wild-type genomic DNA. The mutation-specific MRD was calculated by means of the threshold cycle, as determined by real-time PCR of the patient specimen.

MRD status was positive, whereas the Flt3-D835H– positive clone could not be detected during good partial remission (PR). Another patient reached complete remission (CR) in the bone marrow while the peripheral reconstitution was not complete. In this case MRD status was already positive with ⬍ 5% bone marrow blasts, and the further clinical course resulted in a relapse. The fourth patient was in CR, including reconstituted peripheral blood and MRD-negative in both bone marrow samples. This patient is still free of relapse. Patient 5 developed a positive MRD status during a period of good PR before relapse of AML. MRD model system using Flt3-V592A mutation. We also established a mutation-specific PCR for sensitive and specific detection of the Flt3-V592A mutation. Figure 5A shows a MRD model system for evaluating the “cutoff” for reliable analyses of residual cells carrying this mutation. A sensitivity of at least 0.01% can be reached while no signal is detected using wild-type DNA. Figures 5B and C show that this assay can also be performed using the real-time PCR technique. Thus the Flt3-V592A mutation can be used principally for its evaluation in MRD monitoring if this point mutation can be found in patients with AML. DISCUSSION

In the present study we intended to establish not only an improved screening for Flt3-TKD mutations, but also a method to identify patients carrying a Flt3V592A mutation. Toward both goals, we designed

PCR-based methods for rapid and reliable screening. Detection of Flt3-TKD mutations with fluorescent probes is not only time-saving (about 1 hour vs at least 4 hours), but also has equal sensitivity and possibly even higher specificity than conventional screening. The phenomenon of false-positive results by incomplete restriction analyses can be avoided by using fluorescent probe– based PCR. During the preparation of this manuscript, Bagrintseva et al35 also introduced melting-curve analysis for detecting Flt3 mutations. They used a fluorescence probe spanning the region around codon 840 in their work on acquired Flt3 mutations that confer drug resistance after treatment with protein tyrosine kinase inhibitors. In contrast, in the present study we provide the first evaluation (to our knowledge) of this screening method in routine diagnostic testing. The detection of Flt3-V592A mutations was established with MonoMac-6 cells carrying the underlying point mutation and based on the same principle regardless of whether or not a mutation-specific probe was used or Mono Mac-6 cells were diluted to mimic a model for MRD. Interestingly, not a single Flt3V592A–positive patient was identified among the 122 AML patients investigated. These findings are in line with the recently published data from Stirewalt et al,29 showing only 1 of 140 AML patients carrying a V592A point mutation detected by screening with a PCR/single-strand conformation polymorphism technique. This latter study confirms our findings indicating that Flt3-

302

Scholl et al

J Lab Clin Med June 2005

Fig 5. Synopsis of mutation-specific detection of Flt3-V592A mutations. (A) Detection of Flt3-V592A by mutation-specific primers (PCR product of 223 bp) after a mixture of different ratios (%) of MonoMac-6 cells and leukocytes in whole blood (lanes 1 to 5) and demonstration of PCR specificity (50 ng DNA of MonoMac-6 in lane 6 vs 500 ng DNA of leukocytes (wild-type) in lane 7 and dilution of 50 ng MonoMac-6 in 500 ng wild-type DNA in lane 8). (B and C) LightCycler analyses, including melting curves confirming product specificity (inserts), for 1% and 0.1% MonoMac-6 cells (B) and a negative control (C) (500 ng wild-type DNA).

V592A mutation is a rare event in patients with AML, at least at primary diagnosis. Thus Flt3-V592A mutations are much less clinically significant than Flt3-TKD mutations. Independent of this, we have reported a highly specific and sensitive PCR technique that can be used for MRD analyses in AML patients carrying a Flt3-V592A mutation. Mutations within codons 835 and 836 of the Flt3 gene reflect the second most frequent subtype of Flt3 mutations. Thus evaluating point mutation–specific PCR for performing MRD analyses based on different kinds of Flt3-TKD mutations was a special focus of this work. The Flt3-TKD mutations of the 9 of our 122 patients found to carry such aberrations reflect the most frequent mutations of this region and can be observed in about 75% of AML patients with Flt3-TKD mutations.11 We demonstrated highly specific and sensitive real-time PCR assays for detection of minimal residual disease based on mutation of D835Y, D835H, or Flt3, reflected by deletion of codon 836. Furthermore, we provided the first results of retrospective MRD analyses for all 3 kinds of Flt3-TKD mutations. Two of our patients had a negative MRD status despite 6% to 7% bone marrow blasts, suggesting a limitation of our test in terms of sensitivity. During early hematopoietic reconstitution after initial chemotherapy, patients with up to 7% bone marrow blasts can be regularly observed. Patients in such very good PR either develop a persistent

CR that remains stable after the chemotherapy has been continued or have a really low level of residual leukemia cells, which can lead to a relapse of AML. In consideration of the assay characteristics, we interpret the findings as indicating that in both cases there were only very low MRD levels, with rapid growth of the clone after specimen collection. Despite the relapse of both patients, this does not necessarily imply a limited sensitivity of the assay in these cases. Our data demonstrate preliminary but promising examples of MRD analyses based on Flt3-TKD mutations. Thus quantitative MRD detection could be possible in prospective time course analyses. Our retrospective measurement of Flt3-TKD– based MRD demonstrates that such monitoring may be useful in early prediction of relapse. We are aware that in some cases Flt3-TKD mutations are lost at relapse. Of course, in most of these cases the inability to detect the Flt3-TKD–mutated clone will be due to a clonal selection until relapse.36 Shih et al36 used conventional detection of Flt3-TKD mutations to screen for these aberrations. Thus, whether or not the sensitivity of this PCR approach is acceptable for evaluation of leukemia cells still harboring Flt3-TKD mutations at relapse and for MRD analysis must be considered. We were not able to detect mutant alleles less than 10% of wild-type allele using conventional PCR followed by EcoRV detection.

J Lab Clin Med Volume 145, Number 6

The possibility of highly sensitive detection of residual leukemia cells carrying Flt3-TKD mutations via a rapid, simple, and unique PCR might open new doors in the monitoring of this special subset of AML patients. Prospective individual time course analyses need to be made to evaluate the role of mutation-specific PCR in MRD detection and its predictive value. In summary, Flt3-TKD mutations represent specific diagnostic markers for sensitive detection of residual leukemia cells. Analysis of Flt3-TKD mutations by a fluorescence probe– based PCR technique provides a rapid and reliable new method for primary screening. Moreover, using mutation-specific primers for each subset of Flt3-TKD mutations has the capacity of provide a highly sensitive and specific PCR method for quantitative MRD analysis. This new approach might aid early prediction of relapse in leukemia patients carrying Flt3-TKD mutations. We are grateful to Dario Papi (TIB MOLBIOL, Berlin, Germany) for his advice in the design of the fluorescent probes and optimizing primer design.

REFERENCES

1. Rosnet O, Marchetto S, de Lapeyriere O, Birnbaum D. Murine Flt3, a gene encoding a novel tyrosine kinase receptor of the PDGFR/CSF1R family. Oncogene 1991;6:1641–50. 2. Lyman SD, James L, Vanden Bos T, de Vries P, Brasel K, Gliniak B, et al. Molecular cloning of a ligand for the flt3/flk-2 tyrosine kinase receptor: a proliferative factor for primitive hematopoietic cells. Cell 1993;75:1157– 67. 3. Rusten LS, Lyman SD, Veiby OP, Jacobsen SE. The FLT3 ligand is a direct and potent stimulator of the growth of primitive and committed human CD34⫹ bone marrow progenitor cells in vitro. Blood 1996;87:1317–25. 4. McKenna HJ, de Vries P, Brasel K, Lyman SD, Williams DE. Effect of flt3 ligand on the ex vivo expansion of human CD34⫹ hematopoietic progenitor cells. Blood 1995;86:3413–20. 5. Maroc N, Rottapel R, Rosnet O, Marchetto S, Lavezzi C, Mannoni P, et al. Biochemical characterization and analysis of the transforming potential of the FLT3/FLK2 receptor tyrosine kinase. Oncogene 1993;8:909 –18. 6. Birg F, Courcoul M, Rosnet O, Bardin F, Pebusque MJ, Marchetto S, et al. Expression of the FMS/KIT-like gene FLT3 in human acute leukemias of the myeloid and lymphoid lineages. Blood 1992;80:2584 –93. 7. Gilliland DG, Griffin JD. The roles of Flt3 in hematopoieses and leukemia. Blood 2002;100:1532– 42. 8. Levis M, Small D. Flt3: ITDoes matter in leukemia. Leukemia 2003;17:1738 –52. 9. Nakao M, Yokota S, Iwai T, Kaneko H, Horiike S, Kashima K, et al. Internal tandem duplication of the flt3 gene found in acute myeloid leukemia. Leukemia 1996;10:1911– 8. 10. Kottaridis PD, Gale RE, Frew ME, Harrison G, Langabeer SE, Belton AA, et al. The presence of a FLT3 internal tandem duplication in patients with acute myeloid leukemia (AML) adds important prognostic information to cytogenetic risk group and response to the first cycle of chemotherapy: analysis of 854 patients from the United Kingdom Medical Research Council AML 10 and 12 trials. Blood 2001;98:1752–9.

Scholl et al

303

11. Thiede C, Steudel C, Mohr B, Schaich M, Schakel U, Platzbecker U, et al. Analysis of Flt3-activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood 2002;99:4326 –35. 12. Yamamoto Y, Kiyoi H, Nakano Y, Suzuki R, Kodera Y, Miyawaki S, et al. Activating mutation of D835 within the activation loop of FLT3 in human hematologic malignancies. Blood 2001;97:2434 –9. 13. Abu-Duhier FM, Goodeve AC, Wilson GA, Care RS, Peake IR, Reilly JT. Identification of novel FLT-3 Asp835 mutations in adult acute myeloid leukaemia. Br J Haematol 2001;113:983– 8. 14. Kiyoi H, Ohno R, Ueda R, Saito H, Naoe T. Mechanism of constitutive activation of FLT3 with internal tandem duplication in the juxtamembrane domain. Oncogene 2002;21:2555– 63. 15. Hayakawa F, Towatari M, Kiyoi H, Tanimoto M, Kitamura T, Saito H, et al. Tandem-duplicated Flt3 constitutively activates STAT5 and MAP kinase and introduces autonomous cell growth in IL-3– dependent cell lines. Oncogene 2000;19:624 –31. 16. Mizuki M, Fenski R, Halfter H, Matsumura I, Schmidt R, Muller C, et al. Flt3 mutations from patients with acute myeloid leukemia induce transformation of 32D cells mediated by the Ras and STAT5 pathways. Blood 2000;96:3907–14. 17. Tse KF, Mukherjee G, Small D. Constitutive activation of FLT3 stimulates multiple intracellular signal transducers and results in transformation. Leukemia 2000;14:1766 –76. 18. Fenski R, Flesch K, Serve S, Mizuki M, Oelmann E, KratzAlbers K, et al. Constitutive activation of FLT3 in acute myeloid leukaemia and its consequences for growth of 32D cells. Br J Haematol 2000;108:322–30. 19. Minami Y, Yamamoto K, Kiyoi H, Ueda R, Saito H, Naoe T. Different anti-apoptotic pathways between wild-type and mutated FLT3: insights into therapeutic targets in leukemia. Blood 2003;102:2969 –75. 20. Kiyoi H, Naoe T, Nakano Y, Yokota S, Minami S, Miyawaki S, et al. Prognostic implication of FLT3 and N-RAS gene mutations in acute myeloid leukemia. Blood 1999;93:3074 – 80. 21. Abu-Duhier FM, Goodeve AC, Wilson GA, Gari MA, Peake IR, Rees DC, et al. FLT3 internal tandem duplication mutations in adult acute myeloid leukaemia define a high-risk group. Br J Haematol 2000;111:190 –5. 22. Rombouts WJ, Blokland I, Lowenberg B, Ploemacher RE. Biological characteristics and prognosis of adult acute myeloid leukemia with internal tandem duplications in the Flt3 gene. Leukemia 2000;14:675– 83. 23. 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. 24. Fröhling S, Schlenk RF, Breitruck J, Benner A, Kreitmeier S, Tobis K, et al. Acute myeloid leukemia. Fröhling S, Schlenk RF, Breitruck J, et al. Prognostic significance of activating Flt3 mutations in younger adults (16 to 60 years) with acute myeloid leukemia and normal cytogenetics: a study of the German AML Study Group, Ulm. Blood 2002;100:4372– 80. 25. Murphy KM, Levis M, Hafez MJ, Geiger T, Cooper LC, Smith BD, et al. Detection of Flt3 internal tandem mutations by a multiplex polymerase chain reaction and capillary electrophoresis assay. J Mol Diagn 2003;5:96 –102. 26. Bianchini M, Ottaviani E, Grafone T, Giannini B, Soverini S, Terragna C, et al. Rapid detection of Flt3 mutations in acute

304

27.

28.

29.

30. 31.

Scholl et al

myeloid leukemia patients by denaturing HPLC. Clin Chem 2003;49:1642–50. Spiekermann K, Bagrintseva K, Schoch C, Haferlach T, Hiddemann W, Schnittger S. A new and recurrent activating length mutation in exon 20 of the Flt3 gene in acute myeloid leukemia. Blood 2002;100:3423–5. Spiekermann K, Dirschinger RJ, Schwab R, Bagrintseva K, Faber F, Buske C, et al. The protein tyrosine kinase inhibitor SU5614 inhibits Flt3 and induces growth arrest and apoptosis in AML-derived cell lines expressing a constitutively activated Flt3. Blood 2003;101:1494 –1504. Stirewalt DL, Meshinchi S, Kussick SJ, Sheets KM, PogosovaAgadjanyan E, Willman CL, et al. Novel Flt3 point mutations within exon 14 found in patients with acute myeloid leukaemia. Br J Haematol 2004;124:481– 4. Campana D. Determination of minimal residual disease in leukaemia patients. Br J Haematol 2003;121:823–38. Jaeger U, Kainz B. Monitoring of minimal residual disease in AML: the right time for real time. Ann Hematol 2003;82:139 – 47.

J Lab Clin Med June 2005

32. Stirewalt DL, Willman CL, Radich JP. Quantitative, real-time polymerase chain reactions for FLT3 internal tandem duplications are highly sensitive and specific. Leuk Res 2001;25:1085– 8. 33. Bi W, Stambrook J. Detection of known mutation by proofreading PCR. Nucl Acids Res 1998;26:3073–5. 34. Song P, Li S, Meibohm B, Gaber AO, Honaker MR, Kotb M, et al. Detection of MDR1 single nucleotide polymorphisms C3435T and G2677T using RealTime polymerase chain reaction: MDR1 single nucleotide polymorphism genotyping assay. AAPS PharmSci 2002;4(4):E29. 35. Bagrintseva K, Schwab R, Kohl TM, Schnittger S, Eichenlaub S, Ellwart JW, et al. Mutations in the tyrosine kinase domain of Flt3 define a new molecular mechanism of acquired drug resistance to PTK inhibitors in Flt3-ITD–transformed hematopoietic cells. Blood 2004;103:2266 –75. 36. Shih LY, Huang CF, Wu JH, Wang PN, Lin TL, Dunn P, et al. Heterogeneous patterns of Flt3 Asp835 mutations in relapsed de novo acute myeloid leukemia: a comparative analysis of 120 paired diagnostic and relapsed bone marrow samples. Clin Cancer Res 2004;10:1326 –32.