Quantification of minimal residual disease in patients with ... - Nature

Leukemia (2002) 16, 1167–1175  2002 Nature Publishing Group All rights reserved 0887-6924/02 $25.00 www.nature.com/leu

Quantification of minimal residual disease in patients with e1a2 BCR-ABL-positive acute lymphoblastic leukemia using a real-time RT-PCR assay H Yokota1, NH Tsuno2, Y Tanaka3, T Fukui4, K Kitamura1, H Hirai3, K Osumi4, N Itou5, H Satoh5, M Okabe6 and K Nakahara1 1

Department of Laboratory Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; 2Department of Transfusion Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; 3Department of Cell Therapy and Transplantation Medicine, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; 4Otsuka Assay Laboratories, Otsuka, Pharmaceutical Co., Ltd, Tokyo, Japan; 5Department of Biomolecular Science, Toho University, Chiba, Japan; and 6Sapporo Hokuyu Hospital, Artificial Organ and Transplantation Hospital, Sapporo, Japan

Using a real-time RT-PCR method, we analyzed the expression of e1a2 BCR-ABL mRNA in bone marrow samples from 13 patients with e1a2 BCR-ABL-positive acute lymphoblastic leukemia (ALL) at different time points during chemotherapy and after bone marrow transplantation (BMT). The detection limit of the method, assessed using serial dilutions of ALL/MIK cells, was found to be 1:105, similar to what is observed for the conventional RT-nested PCR method. The e1a2 BCR-ABL values were normalized with respect to those of the housekeeping gene GAPDH. The decrease in the e1a2 BCR-ABL/GAPDH ratio after remission induction chemotherapy reflects well the response to chemotherapy and consequently correlates with the prognosis. Although molecular remission was achieved by chemotherapy alone, some patients relapsed, and the e1a2 BCR-ABL/GAPDH ratios in these cases progressively increased to the levels seen prior to hematological relapse. Long-term hematological complete remission (more than 30 months) could be achieved in cases in which a more than 4.0 log decrease in the e1a2 BCR-ABL/GAPDH ratio was obtained by chemotherapy alone, and BMT was then performed. In conclusion, real-time RT-PCR allows for an evaluation of the kinetics of e1a2 BCR-ABL/GAPDH expression during the various phases of chemotherapy or after BMT and may be effective for the indication and control of disease relapse in Ph-positive ALL patients. Leukemia (2002) 16, 1167–1175. DOI: 10.1038/sj/leu/2402483 Keywords: e1a2 BCR-ABL; acute lymphoblastic leukemia; minimal residual disease; monitoring; real-time RT-PCR assay

Introduction The Philadelphia (Ph) chromosome [t(9;22)(q34;q11)], a cytogenetic aberration that results from a reciprocal translocation that joins 3⬘ sequences of the tyrosine kinase c-ABL protooncogene on chromosome 9 to 5⬘ sequences of the BCR gene on chromosome 22,1–3 is found in about 20–30% of adult patients with acute lymphoblastic leukemia (ALL).4–6 The majority (approximately 70–75%), of Ph breakpoints in ALL occurs in the first intron of the BCR gene,7,8 which is referred to as the minor breakpoint cluster region, and results in the fusion of BCR exon 1 to ABL exon 2. This fusion produces a 7.0 kb BCR-ABL transcript (e1a2 BCR-ABL) that encodes a 190 kDa protein.1,9,10 In approximately 25–30% of adult patients with Ph-positive ALL,7,8 the breakpoint occurs in the 5.8 kb major breakpoint cluster region of the BCR gene. This BCRABL fusion gene produces an 8.5 kb BCR-ABL transcript (b2a2 or b3a2 BCR-ABL) that encodes a protein of 210kDa.11,12 Ph-positive ALL is an aggressive disease with a poor prog-

Correspondence: H Yokota, Department of Laboratory Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan; Fax: 81-3-5689-0495 Received 30 November 2000; accepted 24 January 2002

nosis and is consequently resistant to conventional chemotherapy.4–7 Although conventional chemotherapy induces hematological complete remissions (CR) in approximately 70% of these patients, the relapse rate is high,13–16 and the 5year leukemia-free survival rate is less than 15%.17 Thus, Phpositive ALL patients are candidates for more aggressive chemotherapies, including bone marrow transplantation (BMT). However, the relapse rate is high even after BMT, ranging from 40 to 80%.18–20 RT-PCR is a technique that allows the detection of a single leukemic cell in a background of up to one million normal cells and is therefore an ideal method for the detection of minimal residual disease (MRD) following various treatment protocols. In particular, it seems to be a very sensitive test for predicting the risk of relapse after BMT and may help in identifying patients who might benefit from therapeutic interventions. In most Ph-positive ALL patients, residual leukemic BCR-ABL-positive cells can be detected in the CR phase after chemotherapy and/or BMT.18,21–25 Recently, a competitive RT-PCR-based method for the quantification of BCR-ABL mRNA was developed and shown to be useful for monitoring the response to treatment, and in facilitating prognoses and early diagnoses of hematological relapse in Ph-positive ALL patients.26,27 This method, however, has some disadvantages such as the difficulty of handling, that makes it difficult to be used as a routine clinical test. At the moment, therefore, little is known about the number of BCR-ABL transcripts in Ph-positive ALL patients at the time of diagnosis and about quantitative changes in transcript levels after various therapeutic interventions. In the present study, we investigated the efficacy of a realtime RT-PCR assay to detect residual e1a2 BCR-ABL transcripts in Ph-positive ALL patients and to define the kinetics of the disease. To our knowledge, this is the first report on the use of a real-time RT-PCR assay for the evaluation and quantification of MRD in e1a2 BCR-ABL-positive ALL patients during various phases of treatment. Materials and methods

Cell Iines/dilution series To examine the sensitivity of RT-nested-PCR and real-time RTPCR assays, the e1a2 BCR-ABL-positive leukemic cell line ALL/MIK28 and the e1a2 BCR-ABL-negative cell line HL6029 were used. The HL60 cell line was obtained from the Japanese Collection of Research Bio-Resources (Tokyo, Japan) and cultured in RPMI-1640 medium supplemented with 10% fetal calf serum (FCS) and 1% antibiotic/antimycotic in an atmosphere of 5% CO2 at 37°C.

Evaluation of e1a2 BCR-ABL-positive ALL using real-time RT-PCR assay H Yokota et al

1168

Patients

geneic BMT from an HLA-identical sibling. CR was defined as ⬍5% blasts in the BM. BM instead of peripheral blood samples were chosen because analysis of peripheral blood (PB) samples may cause an underestimation of the levels of MRD.30 For negative control, BM samples were obtained from 20 Ph-negative ALL patients and PB samples from 50 normal individuals (25 men and 25 women) following informed consent.

Bone marrow (BM) samples were obtained from 13 patients with e1a2 BCR-ABL-positive ALL diagnosed at the department of Laboratory Medicine, the University of Tokyo Hospital, all of whom had achieved hematological complete remission (CR) following remission-induction chemotherapy. Informed consent was obtained from all patients. All these patients were negative for b2a2 and b3a2 BCR-ABL mRNAs as detected by RT-nested-PCR assay. The e1a2 BCR-ABL mRNA was chosen due to its high specificity for Ph-positive ALL patients. The samples were collected between January 1996 and April 2001. The clinical and laboratory features of the patients are shown in Table 1. All patients received intensive remissioninduction chemotherapy (vincristine (VCR), prednisolone (PSL), doxorubicin (ADR), L-asparaginase (L-ASP), cyclophosphamide (CPM)), three courses of consolidation chemotherapy (vindesine (VDS), mitoxantrone (MIT), behenoyl cytarabine (BHAC), etoposide (ETP), methotrexate (MTX), cytosine arabinoside (Ara-C), PSL, L-ASP, CPM); and three courses of maintenance/intensification (6-mercaptopurine (6-MP), PSL, MTX, VDS, CPM, MIT, CPM, BHAC, ETP) chemotherapy. Patients who relapsed received either the same protocol or reinduction chemotherapy with high-dose Ara-C and MIT. Serial quantifications of the e1a2 BCR-ABL transcript could be performed in 13 patients. In 11 of these patients (Nos 1, 2, 3, 4, 5, 7,8, 10, 11, 12 and 13), monitoring started at diagnosis, and two patients (Nos 6 and 9) were included after the first relapse. Five patients (Nos 6, 7, 8, 9 and 12) received alloTable 1

Patient

RNA extraction Mononuclear cells (MNCs) from BM samples were isolated by Ficoll–Hypaque (Lymphoprep; Nycomed Pharma, Oslo, Norway) density gradient centrifugation (400 g for 40 min at 20°C) and suspended in phosphate-buffered saline (PBS), and immediately used for RNA isolation. Total cellular RNA was isolated from MNC samples using Isogen-LS (Nippon Gene, Japan) according to the manufacturer’s guidelines, with alternated acid guanidinium thiocyanateiphenol chloroform method. RNA concentration and purity were determined spectrophotometrically at 260 and 280 nm, and RNA samples were stored at −80°C until use.

cDNA preparation Total RNA (1.5 ␮g) was reverse transcribed into cDNA in a 22.5 ␮l reaction mixture containing 500 ng of random hex-

e1a2 BCR-ABL positive ALL patients’ characteristics, treatment and outcome

Sex/Age

FAB

Phenotype Initial status

Result of induction treatment

Treatment

CD10, 19, 33, 34+ CD10, 19, 20, 34+

Diagnosis

CR

Chemotherapy

Diagnosis

CR

Chemotherapy

Current status

1

M/65

L2

Died (14 months) in second relapse Died (17 months) in 1st relapse

2

M/51

L1

3

F/53

L1

CD10, 19, 33, 34+

Diagnosis

CR

Chemotherapy

Died (25 months) in 1st relapse

4

M/64

L2

Diagnosis

CR

Chemotherapy

5

F/44

L2

CD10, 13, 19, 33, 34+ CD10, 13, 19, 33, 34+

Diagnosis

CR

Chemotherapy

Died (15 months) in 2nd relapse Died (19 months) in 2nd relapse

6

M/26

L1

1st relapse

CR

7

F/20

L2

CD10, 19, 20, 34+ CD10, 19, 20, 33, 34+

Diagnosis

CR

Chemotherapy + Allo BMT at 10.2 months in 2nd relapse Chemotherapy + Allo BMT at 6.4 months in 1st CR

Died (25 months) in 4th relapse Alive 38+ months

8

M/25

L2

CD10, 13, 19, 34+

Diagnosis

CR

Chemotherapy + Allo BMT at 3.5 months in 1st CR

Alive 56+ months

9

M/28

L2

1st relapse

CR

10

M/58

L1

CD10, 13, 19, 20, 34+ CD10, 13, 19, 20, 33, 34+

Diagnosis

CR

Chemotherapy + Allo BMT at 12.2 months in 3rd CR Chemotherapy

Died (14 months)/Acute GVHD Alive 14+ months

11

F/70

L2

CD10, 19, 20+

Diagnosis

CR

Chemotherapy

Died (8 months) in 1st relapse

12

M/25

Ls

Diagnosis

CR

M/69

L1

Diagnosis

CR

Chemotherapy + Allo BMT at 7.1 months in 1st CR Chemotherapy

Alive 14+ months

13

CD10, 19, 20, 34+ CD10, 19, 20, 34+

Alive 10+ months

CR, hematological complete remission; Allo BMT, allogeneic bone marrow transplantation; GVHD, graft-versus-host disease. Leukemia


amer, 200 units of SuperScript II reverse transcriptase (GIBCO BRL, Rockville, MD, USA), 40 units of RNase inhibitor (GIBCO BRL), 4.5 ␮l of 5 × first-strand buffer, 1.5 ␮l of 0.1 M DTT and 3 ␮l of 2 ␮M dNTP mix at 42°C for 1 h.

Nested PCR The following two sets of primers were used for e1a2 BCRABL RT-nested PCR: (external sense primer) 5⬘AACAGTCCTTCGACAGCAGC-3⬘; (external antisense primer) 5⬘-GACCCAGCCTTGGCCATTTT-3⬘; (internal sense primer) 5⬘-CACGCCGCAGTGCCATAAG-3⬘; (internal antisense primer) 5⬘-ACACCATTCCCCATTGTGAT-3⬘, respectively. The first PCR was prepared at a final volume of 50 ␮l and included 2 ␮l of cDNA, 1.25 units of Ex Taq DNA polymerase (TaKaRa, Japan), 1 × PCR buffer (10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgC12), 2 ␮M dNTP, and 0.2 ␮M of each external primer. Amplifications were performed on a Model 9600 thermal cycler (PE Biosystems, Foster City, CA,USA) under the following conditions: 94°C for 1 min, 65°C for 1 min, and 72°C for 1min for 30 cycles, followed by a final extension step of 72°C for 10 min before cooling to 4°C. Two ␮l of the firstround products were subjected to second-round PCR in a 50 ␮l reaction containing 0.2 ␮M of each internal primer. The second-round PCR was performed under the same conditions as for the first-round PCR. Seven ␮l of the second-round PCR products (349 bp) were electrophoresed on a 3.5% agarose gel, stained with ethidium bromide and visualized by UV light, and photographed. To avoid false negative results, the presence of intact RNA and the efficiency of cDNA synthesis were confirmed for each sample by an internal control PCR using ABL mRNA sequence-specific primers, as follows: (external sense primer): 5⬘-ATGTTGGAGATCTGCCTGAA-3⬘ (external antisense primer): 5⬘-GACCCAGCCTTGGCCATTTT3⬘ (internal sense primer): 5⬘-CTGCAAATCCAAGAAGGGGC3⬘ (internal antisense primer): 5⬘-ACACCATTCCCCATTGT GAT-3⬘, respectively. ABL mRNA RT-nested PCR was performed at the same conditions as for e1a2 BCR-ABL protocols, and resulted in the amplification of a 271 bp fragment.

Real-time RT-PCR for e1a2 BCR-ABL mRNA The primers and TaqMan probe for the e1a2 BCR-ABL realtime PCR, which produces a 106 bp amplicon, were designed with the PRIMER-EXPRESS software program (Perkin Elmer, Foster City, CA, USA). Sequences of the forward and reverse primers were: 5⬘-ATCGTGGGCGTCCGCAAGAC-3⬘ (located in exon 1 of the BCR gene) and 5⬘-GCTCAAAGTCAGATGCTACTG-3⬘ (located in exon 2 of the ABL gene ). The TaqMan probe (specific to exon 1 of the BCR gene) was fluorescence labeled at the 5⬘-end with FAM (6-carboxyfluorescein) as the reporter dye, and at the 3⬘-end with TAMRA (6-carboxy-teremethyl-rhodamine) as the quencher (FAMCGCCCTCGTCATCGTTGGGCCAGATCT-TAMRA). The PCR product was detected as an increase of fluorescence with an ABI PRISM 7700 (PE Biosystems). The PCR reaction was performed using TaqMan Universal PCR Master Mix (PE Biosystems). 7.5 ␮l of cDNA derived from BM mRNA was mixed with 25 ␮l of PCR Master Mix, 15 pM of both primers and 10 pM of the TaqMan probe, and distilled water was then added to a total volume of 50 ␮l. The PCR reaction was performed under the following conditions: 1 cycle at 50°C for 2

min and 95°C for 10 min, and 50 cycles at 95°C for 15 s and 61°C for 1 min. A standard was prepared by subcloning the e1a2 BCR-ABL gene into the vector pCRII (Invitrogen, Carlsbad, CA, USA). The numbers of copies of the plasmid containing fusion transcripts were determined. In addition, to control for the quality of RNA and the efficiency of reverse transcription, and also to avoid false negative results, each sample was evaluated for the presence of intact RNA and for the efficiency of cDNA synthesis by using a GAPDH control, which was provided as the TaqMan GAPDH Control Reagent (PE Biosystems) and used according to the manufacturer’s instructions. The instructions as well as the primers included in the kit are under the following, GAPDH plasmid construction. To construct the GAPDH plasmid, a 226-base pair fragment spanning exons 2 and 4 of the GAPDH genomic locus was amplified using the forward primer 5⬘-GAAGGTGAAGGTCGGAGTC-3⬘ and reverse primer 5⬘-GAAGATGGTGATGGGATTTC-3⬘. The GAPDH TaqMan probe was labeled at the 5⬘ end with 2,7dimethoxy-4,5-dichloro-6-carboxy-fluorescein (JOE) and at the 3⬘ end with TAMRA. JOE-CAAGCTTCCCGTTCTCAGCCTAMRA. The GAPDH control standard was constructed by the same method as was used for the e1a2 BCR-ABL standard. The results were normalized by calculating the ratio of the numbers of copies of e1a2 BCR-ABL transcript and of GAPDH transcript in each sample. All experiments were performed in duplicate, and the average value of both duplicates was used for quantification.

1169

Results

Standard curve and amplification profiles The concentration of the GAPDH standard ranged from 10 to 1 × 107 copies/well. A standard curve for the control GAPDH gene was constructed with seven standard dilutions and showed a good correlation efficiency between the number of copies of GAPDH and the threshold cycle (Ct) values (r = 0.997) (data not shown). The concentration of the e1a2 BCR-ABL standard ranged from 10 to 1 × 107 copies/well. A specific increase in reporter fluorescence could be detected in samples containing the e1a2 BCR-ABL standard plasmid but not in samples with Phnegative ALL patients and normal individuals. Linearity was observed from 10 to 1 × 107 copies/well. The standard curve showed a good correlation between the number of copies of e1a2 BCR-ABL and the Ct values (r = 0.998) (data not shown). The mean values of the slope of each standard curve, as obtained from three separate experiments, were −3.46 and −3.38 for e1a2 BCR-ABL and GAPDH, respectively.

Reproducibility of real-time RT-PCR assay The coefficients of variation (CV) obtained for the GAPDH control from low (103 copies/well), middle (105 copies/well) and high (107 copies/well) concentration samples in a test of intra-assay variation were 6.42%, 3.81% and 1.12%, respectively, with 10 replicates each. In a test of inter-assay variation, these values were 10.62%, 4.78% and 1.18%, respectively, with five replicates each. The CV obtained for the e1a2 BCRABL control from low (102 copies/well), middle (104 copies/well) and high (106 copies/well) concentration samples in the test of intra-assay variation were 10.51%, 2.17% and Leukemia


1170

4.56%, respectively, with 10 replicates each. In the test of inter-assay variation, these values were 10.62%, 2.48% and 5.06%, respectively, with five replicates each.

Sensitivity of real-time RT-PCR at the cellular level To examine the sensitivity of real-time RT-PCR, we varied the ratio of ALL/MIK to HL60 cells from 1 to 1:105 (total number of cells: 107 in each sample) by serial dilution. Total RNA was extracted from each cell mixture, and real-time PCR was performed. Real-time PCR was able to quantify the number of e1a2 BCR-ABL transcripts in cDNA prepared from samples covering this range, from pure ALL/MIK samples to a 105 dilution; a linear correlation between Ct and the log of ALL/MIK cell number could be observed over a five log range (Figure 1a). The sensitivity of real-time PCR was the same as that of conventional RT-nested-PCR (Figure 1b).

The e1a2 BCR-ABL/GAPDH ratios in clinical samples Figure 2 shows the e1a2 BCR-ABL/GAPDH ratios of 99 BM specimens from 13 e1a2 BCR-ABL-positive ALL patients at the time of initial diagnosis (n = 11), during hematological complete remission (n = 74), and at relapse (n = 14). At the time of initial diagnosis, this ratio ranged from 0.41 to 2.3 (n = 11, median = 1.44). The mean value at diagnosis was very close to that observed for the ALL/MIK cell line (n = 3, median = 1.3). On the other hand, 74 BM specimens from 13 patients in CR showed markedly low (⬍6.6 × 10−2) e1a2 BCRABL/GAPDH ratios. In contrast, these values were high at

Figure 2 The levels of e1a2 BCR-ABL/GAPDH ratios. The results were normalized by dividing the copy number of mBCR-ABL by that for GAPDH in each sample. Differences between mBCR-ABL/GAPDH ratio at different disease phases are significant (P ⬍ 0.001). (䊉), mBCR-ABL mRNA positive by RT-nested-PCR; (䊊), mBCR-ABL mRNA negative by RT-nested-PCR.

relapse, ranging from 0.2 to 1.8 (n = 14, median = 0.67). Differences of e1a2 BCR-ABL/GAPDH ratios among the different disease phases were statistically significant (P ⬍ 0.001). The results obtained by the real-time RT-PCR are similar to those obtained with conventional RT-nested-PCR, and in all cases for which a positive result could be obtained by conventional RT-nested-PCR, the e1a2 BCR-ABL/GAPDH ratio could be clearly quantified by real-time RT-PCR. In addition, all e1a2 BCR-ABL-negative cases, as assessed by real-time RT-PCR, were also negative when tested by RTnested-PCR, but the internal control GAPDH could be clearly detected by both methods. The GAPDH value for the e1a2 BCR-ABL negative cases ranged between 1.1 × 106 and 4.4 × 106 copies. The mean GAPDH value for all clinical samples was 3.4 × 106 copies, ranging from 1.1 × 106 to 5.7 × 106 copies. Furthermore, we evaluated PB samples from 50 normal individuals (25 men and 25 women). The mean GAPDH value for these samples was 2.1 × 106 copies, ranging from 1.5 × 106 copies to 3.2 × 106 copies. Consequently, no samples were excluded from the study based on their GAPDH value.

Figure 1 Sensitivity analysis of real-time PCR and RT-nested-PCR. (a) Quantification of e1a2 BCR-ABLmRNA in samples containing different ratios of ALL/MIK to HL-60 cells by real-time PCR. The solid line shows the threshold cycle Ct values for e1a2 BCR-ABL m RNA with respect to the ratio of ALL/MIK to HL-60 cells. There is a linear increase in the threshold cycle from 1 to 1:105. (b) Results of RTnested-PCR. Detection of e1a2 BCR-ABL m RNA in samples of HL60 cells containing ALL/MIK cells at ratios of 1 to 1:106. Leukemia

The e1a2 BCR-ABL/GAPDH ratios of patients who received chemotherapy alone Table 2 shows the e1a2 BCR-ABL/GAPDH ratios of BM samples at times after remission-induction chemotherapy, after the first and second consolidation chemotherapies of 11 patients that could be monitored at the molecular level since

16.5 59.1

16.2

50.6

30.1

87.2

33.1

23.1

4 5

7

8

10

11

12

13

84

90

92

87

98

94

65 94

94 71 96

Blast (%)

2.1

1.2

0.93

0.63

2.3

1.6

0.70 1.8

1.9 0.41 0.99

BCR/ABL/ GAPDH

CR1

CR1

CR1

CR1

CR1

CR1

CR1 CR1

CR1 CR1 CR1

Outcome

CR1 CR1 CR1 CR1 CR1 CR1 CR1 CR1 CR1 CR1 CR1

2.0 × 10−2 (−1.5) 9.9 × 10−3 (−2.3) 1.5 × 10−4 (−4.0) 9.3 × 10−3 (−2.4) 6.3 × 10−3 (−2.0) 9.5 × 10−3 (−2.0) 1.4 × 10−2 (−1.9) 1.4 × 10−2 (−2.2)

Outcome

3.3 × 10−2 (−1.8) Not detected Not detected

BCR-ABL/ GAPDH

Received BMT in 2nd consolidation CR1 5.4 × 10−3 (−2.1) Relapse in 3rd consolidation Relapse before 2nd consolidation CR1 6.8 × 10−3 (−2.2) Relapse in 2nd consolidation

2.1 × 10−3 (−3.0)

6.8 × 10−2 (−1.5)

9.0 × 10−3 (−2.1)

1.3 × 10−2 (−1.7)

5.8 × 10−3 (−2.0)

CR1

Not detected

Relapse in 2nd consolidation Relapse in 2nd consolidation

CR1 CR1 CR1

Outcome

After 2nd consolidation

Not detected

6.6 × 10−2 (−1.0) 8.7 × 10−2 (−1.3)

1.4 × 10−2 (−2.1) Not detected Not detected

BCR-ABL/ GAPDH

After 1st consolidation

8.8 × 10−3 (−2.3) Not detected 2.9 × 10−4 (−3.5)

BCR-ABL/ GAPDH

After induction

12.6+ (received BMT) 4

3.1

36.8+ (received BMT) 54+ (received BMT) 8.1

3.2 3.3

4.3 10.8 16.8

Duration of CR1 months

NCC, nucleated cell count (×1010/l); e1a2 BCR-ABL/GAPDH, numbers indicate the ratio of copy number of e1a2 BCR-ABL/GAPDH; in parentheses, the log-fold difference between post and pre-chemotherapy e1a2 BCR-ABL/GAPDH; CR1, first hematological complete remission; BMT, bone marrow transplantation; Not detected, the GAPDH was clearly detected in all not detected cases, and the GAPDH value ranged between 1.1 × 106 copies and 4.4 × 106 copies.

85.1 12.7 73.4

NCC

Cell count

Pretreatment

Effect of induction and consolidation chemotherapy on e1a2 BCR-ABL/GAPDH

1 2 3

Patient

Table 2


1171

Leukemia


1172

the time of initial diagnosis. In three patients (Nos 2, 3 and 7), in whom molecular remission was achieved by chemotherapy alone, e1a2 BCR-ABL/GAPDH ratios after the first consolidation chemotherapy were below the limit of detection for the method. In the other eight patients (Nos 1, 4, 5, 8, 10, 11, 12 and 13), molecular remission could not be achieved by chemotherapy alone. The e1a2 BCR-ABL/GAPDH ratios in these cases ranged from 0.63 to 2.3 (n = 8, mean = 1.44) at the time of initial diagnosis, and after remission-induction chemotherapy a 1.5–2.4 log (mean = 2.1 log) decrease was observed. In five of these patients (Nos 1, 4, 5, 11 and 13), although CR could be maintained after the first consolidation chemotherapy, a progressive increase in the ratio of e1a2 BCRABL/GAPDH was observed, and after periods as short as 3.1– 4.3 months, early relapse was observed. In these cases, first and second consolidation chemotherapies led to hematological CR with e1a2 BCRABL/GAPDH ratios reaching their lowest points after remission-induction chemotherapy. However, a progressive increase was observed thereafter, resulting in disease relapse. Figure 3a shows the typical kinetics of e1a2 BCR-ABL/GAPDH expression for early relapsed cases. On the other hand, in three patients (Nos 2, 3 and 7) who achieved molecular remission after the first consolidation chemotherapy, the e1a2 BCR-ABL/GAPDH ratios ranged from

0.41 to 2.3 (n = 3, mean = 1.00 ) at the time of initial diagnosis, showing a 3.5–4.6 log (mean = 4.0 log ) decrease after remission-induction chemotherapy. In two patients (Nos 2 and 3), chemotherapy was continued during CR1, and a long-term CR1 could be maintained. However, at 2 months and 7 months, respectively, prior to relapse 1, the e1a2 BCRABL/GAPDH ratios started to increase, and clinical relapse developed. The kinetics of expression of e1a2 BCRABL/GAPDH for patient 3 is shown in Figure 3b.

The e1a2 BCR-ABL/GAPDH ratios in patients who received chemotherapy followed by allogeneic BMT The kinetics of expression of e1a2 BCR-ABL/GAPDH for three of five patients who received allogeneic BMT are shown in Figure 4. Patient 6 (Figure 4a) is a typical case of early relapse. The patient developed repeated clinical relapses, and BMT was performed at the time of the third relapse, when the e1a2 BCR-ABL/GAPDH value reached 0.28. CR4 was achieved 30 days following BMT (18.8 months after the initial diagnosis), when a 2.6 log decrease in the e1a2 BCR-ABL/GAPDH ratio compared to that prior to BMT was observed, and this ratio reached values as low as 6.8 × 10−5 (4.2 log decrease) 22 months after the time of diagnosis. Although a favorable molecular evolution was thought to be obtained, molecular remission was not achieved after BMT, and the patient developed relapse 4. Patient 8 (Figure 4b) received BMT prior to diagnosis of molecular remission (2.5 months after the initial diagnosis). The mBCR-ABL/GAPDH ratio immediately prior to BMT was as low as 1.6 × 10−4. By 28 days after BMT, the e1a2 BCRABL/GAPDH ratio was below the limit of detection, and at this time molecular remission was also confirmed by RTnested-PCR. Patient 12 (Figure 4c) received BMT prior to diagnosis of molecular remission. At the time BMT was performed, the e1a2 BCR-ABL/GAPDH ratio was as high as 5.6 × 10−2, but 64 days after BMT, it was below the limit of detection, and the patient remained in CR. Patient 7 is a case in which molecular remission was achieved after chemotherapy alone, and this patient received BMT immediately afterwards. These patients are currently in CR. On the other hand, one patient (No. 9) received BMT during the third CR (12.2 months after diagnosis) but unfortunately developed acute graft-versus-host disease and died 1.8 months later. e1a2 BCR-ABL/GAPDH ratios for this patient could not be monitored after BMT. Discussion

Figure 3 Kinetics of e1a2 BCR-ABL/GAPDH expression in patients 1 and 3, who received chemotherapy alone. Circles represent the e1a2 BCR-ABL/GAPDH ratio in BM samples. CR, hematological complete remission; R, relapse; NCC, nucleated cell count in BM (× 104/␮l); Bl, blast; †, deceased; (䊉), e1a2 BCR-ABL mRNA positive by RT-nested-PCR; (䊊), e1a2 BCR-ABL mRNA negative by RTnested-PCR. Leukemia

In the present study, e1a2 BCR-ABL levels were measured at the time of diagnosis and during the course of treatment using real-time PCR, and the e1a2 BCR-ABL molecular diagnosis was compared with the hematological diagnosis and clinical outcome. First, to obtain a large dynamic range, we cloned the e1a2 BCR-ABL translocation and used it to establish standard curves for quantitation. To control for inter-sample variation of RNA, we normalized the e1a2 BCR-ABL values to those for the housekeeping gene GAPDH, as reported previously.31,32 A linearity in values from 10 to 107 copies (6.0 log) was observed for this method, and the CV values obtained in the intra- and inter-assay variations were satisfactory for clinical application. Next, the efficiency of the method for quantification of e1a2


Figure 4 Kinetics of e1a2 BCR-ABL/GAPDH expression in patients 6, 8 and 12 who received chemotherapy followed by allogeneic BMT. Circles represent the e1a2 BCR-ABL/GAPDH ratio in BM samples. CR, hematological complete remission; R, relapse; NCC, nucleated cell count in BM (× 104/␮l); Bl, blast; †, deceased; (䊉), e1a2 BCR-ABL mRNA positive by RT-nested-PCR; (䊊) e1a2 BCR-ABL mRNA negative by RT-nested-PCR.

BCR-ABL mRNA was evaluated using both the ALL/MIK cell line and clinical samples isolated from e1a2 BCR-ABL-positive ALL patients. The e1a2 BCR-ABL/GAPDH value for the ALL/MIK cell line was 1.3, and that for clinical samples obtained at the time of initial diagnosis and at relapse ranged from 0.04 to 2.3. In addition, the limit of detection of the method, as confirmed by serial dilutions of the ALL/MIK cell

line, was found to be 1 in 105 cells, similar to what is observed for the conventional RT-nested PCR method. The results obtained by both methods are very similar, confirming their sufficient dynamic range and high sensitivity (5.0 log) for clinical application. The decrease in the e1a2 BCR-ABL/GAPDH ratio at the time CR was achieved after remission-induction chemotherapy ranged from 1.5 to 4.6 log, and individual differences in the rate of decrease correlated well with the response to chemotherapy. In four early relapsed cases, the mean log decrease of e1a2 BCR-ABL/GAPDH was 2.4 log, lower than that for the three cases for which long-term hematological CR (more than 10 months) could be achieved (mean = 3.8 log). Compared to the early-relapsed cases, these three patients exhibited a better response to chemotherapy, in addition to their greater decrease in the level of e1a2 BCR-ABL/GAPDH. This suggests that the log decrease rate of e1a2 BCR-ABL/GAPDH levels is a good marker for predicting responses to chemotherapy. Mitterbauer et al27 reported a maximum of 2–3 log reduction in the BCR-ABL/ABL after conventional chemotherapy, but in the present study we were able to observe a more than 4 log decrease after remission induction chemotherapy in four patients, two of them after BMT. These latter two cases are still alive and are in CR, but the other two relapsed after long-term remission. Consequently, the rate of reduction in the e1a2 BCR-ABL/GAPDH ratio may be a useful index for the prediction of the period of disease remission in Ph-positive ALL patients. In early relapsed cases, the e1a2 BCR-ABL/GAPDH ratios did not undergo an additional decrease after reaching a minimal level following remission-induction chemotherapy, and they started to increase in spite of consolidation chemotherapy. Ph-positive ALL patients frequently relapse at an early phase, even when CR achieved by remission-induction chemotherapy is maintained by subsequent consolidation chemotherapy.4–7 The kinetics of e1a2 BCR-ABL/GAPDH in such cases probably change, as observed in the present cases. Therefore, monitoring the kinetics of e1a2 BCR-ABL/GAPDH ratios after the first consolidation chemotherapy may be highly effective for the early diagnosis of disease relapse. In addition, relapse occurred in two cases (Nos 2 and 3) for which molecular remission could be achieved, but 2 and 7 months prior to diagnosis of relapse, molecular relapse was indicated by an increase in the e1a2 BCR-ABL/GAPDH ratio. Consequently, relapse could be predicted at an early phase, even in patients in whom molecular remission was achieved, and in addition, the degree of molecular relapse could also be determined from the level of mRNA expression. At the time molecular relapse was diagnosed, these patients were in the course of maintenance chemotherapy according to the conventional protocol. In such cases, if remission-induction chemotherapy is initiated at an early phase of molecular relapse, hematological relapse can probably be prevented. In addition, quantitation of e1a2 BCR-ABL would be useful for the evaluation of new therapeutic strategies such as tyrosine kinase inhibitorbased STI,33,34 which is provided for maintenance in CR if BMT is not an option, or after allogeneic BMT in such patients. Among the patients who received allogeneic BMT after chemotherapy, patient 6 received it after the third relapse, and the e1a2 BCR-ABL/GAPDH ratio after BMT showed a maximal decrease of 4.2 log. The patient seemed to progress favorably at the hematological and molecular levels, but molecular remission could not be achieved after BMT, and hematological relapse developed. This case leads us to speculate that molecular remission is a necessary condition for long-lasting

1173

Leukemia


1174

CR in these patients but that is not sufficient to guarantee continuing CR. The kinetics of e1a2 BCR-ABL expression (Figure 4a) in this patient was dramatically fast, with a 4 log increase in just 2 months. This observation suggests that, in such patients in whom e1a2 BCR-ABL expression is detected in the BM sample even after BMT, a monthly follow-up is indicated, and additional chemotherapy should be introduced in the case of an increase. In contrast, patient 7 responded well to chemotherapy, and BMT was performed at the time molecular remission was achieved. At present, the patient is still well and remains in CR after 32 months. In patient 8, molecular remission could not be achieved by chemotherapy alone, but the subsequent BMT resulted in molecular remission which has been maintained for a long period of time (⬎48 months). In this patient, the e1a2 BCR-ABL/GAPDH ratio, although not achieving molecular remission levels, reached extremely low levels (1.6 × 10−4) immediately prior to BMT, and this may be the cause of the long-term remission obtained after BMT. Although the retrospective nature and the small sample size of the present study are limiting factors, an important finding is that quantitation of the e1a2 BCR-ABL may be a good marker for the evaluation of response to chemotherapy after remission induction in Ph-positive ALL patients. Cases with a less than 2 log reduction after chemotherapy had a higher risk of disease relapse. In addition, this methodology would be useful for the follow-up of patients after molecular remission is achieved, and it may be especially useful in evaluating the efficacy of new therapeutic strategies, such as STI treatment given to patients after CR is achieved, or provided to patients for whom a compatible donor is not yet available or to those with clinical complications preventing immediate BMT, or since STI may even maintain long-lasting CR. Even when molecular relapse develops but e1a2 BCR-ABL levels remain low, similar strategies can be introduced for the achievement of a higher quality disease remission (a more than 4 log reduction in the e1a2 BCR-ABL levels). Future large prospective studies involving patients in molecular remission are necessary to establish the prognostic value of e1a2 BCR-ABL mRNA quantification in Ph-positive ALL patients.

Acknowledgements We thank Mrs Emiko Takahashi, Mr Ryoichi Saito and Mr Junichi Hayakawa for their excellent technical assistance.

References 1 Hermans A, Heisterkamp N, von Lindern M, van Baal S, Meijer D, van der Plas D, Wiedemann LM, Groffen J, Bootsma D, Grosveld G. Unique fusion of bcr and c-abl genes in Philadelphia chromosome positive acute lymphoblastic leukemia. Cell 1987; 51: 33–40. 2 Clark SS, McLaughlin J, Timmons M, Pendergast AM, Ben-Neriah Y, Dow LW, Crist W, Rovera G, Smith SD, Witte ON. Expression of a distinctive BCR-ABL oncogene in Ph-positive acute lymphocytic leukemia (ALL). Science 1988; 239: 775–777. 3 Schaefer-Rego K, Arlin Z, Shapiro LG, Mears JG, Leibowitz D. Molecular heterogeneity of adult Philadelphia chromosome-positive acute lymphoblastic leukemia. Cancer Res 1988; 48: 866–869. 4 Bloomfield CD, Goldman AI, Alimena G, Berger R, Borgstrom GH, Brandt L, Catovsky D, de la Chapelle A, Dewald GW, Garson OM, Garwicz S, Golomb HM, Hossfeld DK, Lawler SD, Mitelman F, Nilsson P, Pierre RV, Philip P, Prigogina E, Rowely JD, Sakurai M, Sandberg AA, Secker-Walker LM, Tricot G, Van Den Berghe H, Van Orshoven A, Vuopio P, Whang-Peng J. Chromosomal abnormalities Leukemia

identify high-risk and low-risk patients with acute lymphoblastic leukemia. Blood 1986; 67: 415–420. 5 Fenaux P, Lai JL, Morel P, Nelken B, Taboureau O, Deminatti M, Bauters F. Cytogenetics and their prognostic value in childhood and adult acute lymphoblastic leukemia (ALL) excluding L3. Hematol Oncol 1989; 7: 307–317. 6 Secker-Walker LM, Craig JM, Hawkins JM, Hoffbrand AV. Philadelphia positive acute lymphoblastic leukemia in adults: age distribution, BCR breakpoint and prognostic significance. Leukemia 1991; 5: 196–199. 7 Westbrook CA, Hooberman AL, Spino C, Dodge RK, Larson RA, Davey F, Wurster-Hill DH, Sobol RE, Schiffer C, Bloomfield CD. Clinical significance of the BCR-ABL fusion gene in adult acute lymphoblastic leukemia: a Cancer and Leukemia Group B Study (8762). Blood 1992; 80: 2983–2990. 8 Maurer J, Janssen JW, Thiel E, van Denderen J, Ludwig WD, Aydemir U, Heinze B, Fonatsch C, Harbott J, Reiter A, Riehm H, Hoelzer D, Bartram CR. Detection of chimeric BCR-ABL genes in acute lymphoblastic leukaemia by the polymerase chain reaction. Lancet 1991; 337: 1055–1058. 9 Kurzrock R, Shtalrid M, Romero P, Kloetzer WS, Talpaz M, Trujillo JM, Blick M, Beran M, Gutterman JU. A novel c-abl protein product in Philadelphia-positive acute lymphoblastic leukaemia. Nature 1987; 325: 631–635. 10 Fainstein E, Marcelle C, Rosner A, Canaani E, Gale RP, Dreazen O, Smith SD, Croce CM. A new fused transcript in Philadelphia chromosome positive acute lymphocytic leukaemia. Nature 1987; 330: 386–388. 11 Kawasaki ES, Clark SS, Coyne MY, Smith SD, Champlin R, Witte ON, McCormick FP. Diagnosis of chronic myeloid and acute lymphocytic leukemias by detection of leukemia-specific mRNA sequences amplified in vitro. Proc Natl Acad Sci USA 1988;85: 5698–5702. 12 Kurzrock R, Gutterman JU, Talpaz M. The molecular genetics of Philadelphia chromosomepositive leukemias. N Engl J Med 1988; 319: 990–998. 13 Kantarjian HM, Talpaz M, Dhingra K, Estey E, Keating MJ, Ku S, Trujillo J, Huh Y, Stass S, Kurzrock R. Significance of the P210 versus P190 molecular abnormalities in adults with Philadelphia chromosome-positive acute leukemia. Blood 1991; 78: 2411– 2418. 14 Radich JP, Kopecky KJ, Boldt DH, Head D, Slovak ML, Babu R, Kirk J, Lee A, Kessler P, Appelbaum F, Gehly G. Detection of BCRABL fusion genes in adult acute lymphoblastic leukemia by the polymerase chain reaction. Leukemia 1994; 8: 1688–1695. 15 Schlieben S, Borkhardt A, Reinisch I, Ritterbach J, Janssen JW, Ratei R, Schrappe M, Repp R, Zimmermann M, Kabisch H, JankaSchaub G, Bartram CR, Ludwig WD, Riehm H, Lampert F, Harbott J. Incidence and clinical outcome of children with BCR/ABL-positive acute lymphoblastic leukemia (ALL). A prospective RT-PCR study based on 673 patients enrolled in the German pediatric multicenter therapy trials ALL-BFM-90 and CoALL-05–92. Leukemia 1996; 10: 957–963. 16 Hoelzer D, Thiel E, Loeffler H, Bodenstein H, Plaumann L, Buechner T, Urbanitz D, Koch P, Heimpel H, Engelhardt R, Mueller U, Wendt FC, Sodomann H, Ruehl H, Herrmann F, Kaboth W, Dietzfelbinger H, Pralle H, Lunscken C, Hellriegel KP, Spors S, Nowrousian RM, Fischer J, Fuelle H, Mitrou PS, Pfreundschuh M, Goerg C, Emmerich B, Queisser W, Meyer P, Labedzki L, Essers U, Koenig H, Mainzer K, Herrmann R, Messerer D, Zwingers T. Intensified therapy in acute lymphoblastic and acute undifferentiated leukemia in adults. Blood 1984; 64: 38–47. 17 Hoelzer D. Treatment of acute lymphoblastic leukemia. Semin Hematol 1994; 31: 1–15. 18 Miyamura K, Tanimoto M, Morishima Y, Horibe K, Yamamoto K, AkatsukaM, Kodera Y, Kojima S, Matsuyama K, Hirabayashi N, Yazaki M, Imai K, Onozawa Y, Kanamaru A, Mizutani S, Saito H. Detection of Philadelphia chromosome-positive acute lymphoblastic leukemia by polymerase chain reaction: possible eradication of minimal residual disease by marrow transplantation. Blood 1992; 79: 1366–1370. 19 Barrett AJ, Horowitz MM, Ash RC, Atkinson K, Gale RP, Goldman JM, Henslee-Downey PJ, Herzig RH, Speck B, Zwaan FE, Bortin MM. Bone marrow transplantation for Philadelphia chromosome-


20

21

22

23

24

25

26

27

positive acute lymphoblastic leukemia. Blood 1992; 79: 3067– 3070. Chao NJ, Blume KG, Forman SJ, Snyder DS. Long-term follow-up of allogeneic bone marrow recipients for Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood 1995; 85: 3353–3354. Gehly GB, Bryant EM, Lee AM, Kidd PG, Thomas ED. Chimeric BCR-abl messenger RNA as a marker for minimal residual disease in patients transplanted for Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood 1991; 78: 458–465. Mitterbauer G, Fodinger M, Scherrer R, Knobl P, Jager U, Laczika K, Schwarzinger I, Gaiger A, Geissler K, Greinix H, Kalhs P, Linkesch W, Lechner K. Mannhalter C. PCR-monitoring of minimal residual leukaemia after conventional chemotherapy and bone marrow transplantation in BCR-ABL-positive acute lymphoblastic leukaemia. Br J Haematol 1995; 89: 937–941. Preudhomme C, Henic N, Cazin B, Lai JL, Bertheas MF, Vanrumbeke M, Lemoine F, Jouet JP, Deconninck E, Nelken B, Cosson A, Fenaux P. Good correlation between RT-PCR analysis and relapse in Philadelphia (Ph1)-positive acute lymphoblastic leukemia (ALL). Leukemia 1997; 11: 294–298. Sierra J, Radich J, Hansen JA, Martin PJ., Petersdorf EW, Bjerke J, Bryant E, Nash RA, Sanders JE, Storb R, Sullivan KM, Appelbaum FR, Anasetti C. Marrow transplants from unrelated donors for treatment of Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood 1997; 90: 1410–1414. Radich J, Gehly G, Lee A, Avery R, Bryant E, Edmands S, Gooley T, Kessler P, Kirk J, Ladne P, Thomas ED, Appelbaum FR. Detection of bcr-abl transcripts in Philadelphia chromosome-positive acute lymphoblastic leukemia after bone marrow transplantation. Blood 1997; 89: 2602–2609. van Rhee F, Marks DI, Lin F, Szydlo RM, Hochhaus A, Treleaven J, Delord C, Cross NC, Goldman JM. Quantification of residual disease in Philadelphia-positive acute lymphoblastic leukemia: comparison of blood and bone marrow. Leukemia 1995; 9: 329–335. Mitterbauer G, Nemeth P, Wacha S, Cross NC, Schwarzinger I,

28

29 30

31

32

33

34

Jaeger U, Geissler K, Greinix HT, Kalhs P, Lechner K, Mannhalter C. Quantification of minimal residual disease in patients with BCR-ABL-positive acute lymphoblastic leukaemia using quantitative competitive polymerase chain reaction. Br J Haematol 1999; 106: 634–643. Higa T, Okabe M, Kunieda Y, Kodama S, Itaya T, Kurosawa M, Sakurada K, Maekawa I, Shoji M, Kasai M, Miyazaki T. Establishment and characterization of a new Ph1-positive ALL cell line (ALL/MIK) presenting bcr gene rearrangement on bcr-2 and ALLtype bcr/abl transcript: suggestion of in vitro differentiation to monocytoid lineage. Leuk Lymphoma 1994; 12: 287–96. Koeffler HP, Golde DW. Human myeloid leukemia cell lines: a review. Blood 1980; 56: 344–350. Martin H, Atta J, Bruecher J, Elsner S, Schardt C, Stadler M, von Melchner H, Hoelzer D. In patients with BCR-ABL-positive ALL in CR peripheral blood contains less residual disease than bone marrow: implications for autologous BMT. Ann Hematol 1994; 68: 85–87. Elmaagacli AH, Beelen DW, Opalka B, Seeber S, Schaefer UW. The amount of BCR-ABL fusion transcripts detected by the realtime quantitative polymerase chain reaction method in patients with Philadelphia chromosome positive chronic myeloid leukemia correlates with the disease stage. Ann Hematol 2000; 79: 424– 431. Barbany G, Hagberg A, Olsson-Stromberg U, Simonsson B, Syvanen AC, Landegren U. Manifold-assisted reverse transcription-PCR with real-time detection for measurement of the BCRABL fusion transcript in chronic myeloid leukemia patients. Clin Chem 2000; 46: 913–920. Druker BJ, Tamura S, Buchdunger E, Ohno S, Segal GM, Fanning S, Zimmermann J, Lydon NB. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 1996; 2: 561–566. Kawaguchi Y, Jinnai I, Nagai K, Yagasaki F, Yakata Y, Matsuo T, Kuriyama K, Tomonaga M. Effect of a selective Abl tyrosine kinase inhibitor, STI571, on in vitro growth of BCR-ABL-positive acute lymphoblastic leukemia cells. Leukemia 2001;15: 590–594.

1175

Leukemia