Cancer Genetics and Cytogenetics 162 (2005) 57–62

t

Quantitative molecular monitoring ofBCR-ABL andMDR1 transcriptsin patients with chronic myeloid leukemia during Imatinib treatmenSara Galimbertia,* , Giulia Cervettia, Francesca Guerrinia, Rossana Testia, Simone Pacinia,

Rita Fazzia, Paolo Simib, Mario PetriniaaDepartment of Oncology, Transplant and Advances in Medicine, Section of Hematology, Ospedale S. Chiara,

bU.O. Cytogenetic and Molecular Genetics, Department of Pediatrics, University of Pisa, Via Roma, Pisa 56-56100, Italy

Received 29 September 2004; received in revised form 13 January 2005; accepted 25 January 2005

Abstract Different mechanisms could sustain Imatinib resistance, including overexpression ofMDR1, a genealready known to be responsible for multidrug resistance in other hematologic malignancies. Insearch for a possible correlation,BCR-ABL and MDR1 expression were measured in 115 serialbone marrow samples from 33 CML patients during Imatinib treatment. All patients achievedcomplete hematologic responses, and 22 patients also achieved complete cytogenetic responses,with medianBCR-ABL mRNA values significantly lower than those observed in the group ofcases that were persistently Philadelphia positive. All three cases treated during the accelerated phaseshowed disease progression after an initial period of remission; all presented either increased levelsof BCR-ABL or MDR1 3 months before clinical progression. In the subgroup of cases treated duringthe chronic phase,BCR-ABL andMDR1 levels were significantly correlated after 3 and 6 months(88 and 80%, respectively) but not after 12 months of treatment (32%). Reported data maintainthat MDR1 expression would play an important role in Imatinib resistance when the disease is notfully controlled (e.g., progressive disease or during the first months of treatment).� 2005 ElsevierInc. All rights reserved.

th

,e

i

ea

s

g

-

e

,p)

o-

n-ted

er-

lp)

1. Introduction

Chronic myeloid leukemia (CML) is a malignant clonaldisorder characterized by the Philadelphia (Ph) chromosomtranslocation, which generates theBCR-ABL fusion gene.This rearrangement is detected in about 95% of patients wiCML and, because of its constitutively activated tyrosinekinase activity, it is considered to be the causative moleculaabnormality in CML[1].

Imatinib mesylate (STI571 or Gleevec; Novartis, BaselSwitzerland) is a potent and selective inhibitor of the tyrosinkinase activity of theBCR-ABL gene[2,3]. High responserates and durable remissions have been reported with thdrug in patients with CML who are in chronic (CP) andaccelerated phases (AP) or in blast crisis (BC)[4,5].

However, because some patients in CP fail to achieva major cytogenetic response or lose the initial one, andsignificant proportion of patients treated in AP/BC progres

* Corresponding author. Tel.:�39050992815; fax:�39050830162.E-mail address: s.galimberti@med.unipi.it(S. Galimberti).

0165-4608/05/$ – see front matter� 2005 Elsevier Inc. All rights reserved.doi:10.1016/j.cancergencyto.2005.01.015

e

r

s

or relapse[6], Imatinib resistance is becoming an interestinmatter to investigate.

Different mechanisms could be implicated in this phenomenon: amplification and overexpression of theBCR-ABLgene, point mutations in the ATP-binding site with kinasreactivation[7], or overexpression of theMDR1 gene[8].

TheMDR1 gene, which belongs to the ABC superfamilyencodes for a transmembrane glycoprotein (P-170 or Pgcapable of pumping several different drugs out of the neplastic cell[9]. An increased expression of theMDR1 or itsprotein has been detected in several hematologic malignacies, and its negative prognostic role has been documenfrequently, especially in acute leukemia[10].

Conversely, the role of theMDR1 gene has not beenunequivocally found in CML, where inhibitors of this gene[11] cannot modulate the resistance to anthracyclines. Nevtheless, Mahon et al.[8] showed that the retroviral-mediatedoverexpression of theMDR1 gene in the AR230 cell lineled to in vitro resistance to Imatinib. In this experimentasetting, the resistance was reversible by P-glycoprotein (Pginhibitors, thus sustaining the relevance of theMDR1 genein the resistance to Imatinib.

S. Galimberti et al. / Cancer Genetics and Cytogenetics 162 (2005) 57–6258

r

t

sr

rh

;L

To further clarify the role ofMDR1 gene expressionin vivo during Imatinib treatment, we adopted the real-timreverse-transcription polymerase chain reaction (RT-PCto serially measureBCR/ABL and MDR1 expression in 33CML patients. Expression of these two genes has been colated with hematologic and cytogenetic response.

2. Patients and methods

2.1. Patients, treatment, and response criteria

A total of 115 bone marrow samples from 33 patienaffected by CML (16 men and 17 women), with a mediaage at diagnosis of 56 years (range 24–78 years), were teby cytogenetic and molecular assays at baseline, afteweeks of treatment, and every 3 months during the enttreatment with Imatinib (for clinical characteristics, seTable 1), after obtaining the patients’ informed consent.

According to the EURO parameters[12], at the time ofdiagnosis, 18 patients scored low, 11 intermediate, andhigh-risk. The median number of white blood cells (WBCwas 43× 109/L and that of platelets was 326× 109/L. Mediantime from diagnosis to the start of Imatinib treatment was 5months (range 1–106 months). Before Imatinib treatment,patients were in the CP and three were in the AP of disea

With regard to previous treatments, 29 patients were ptreated with various doses of hydroxyurea, interferon-alp

Table 1Clinical characteristics of patients enrolled in the study

Characteristics Patients,n � 33

AgeRange 24–78Median 56

SexM 16F 17

WBC count (× 109/L)Range 3.9–360Median 43

PLT count (× 109/L)Range 99–1342Median 326

Blast median % (peripheral blood) 1Eosinophils 1Basophils 2

Splenomegaly (No. of patients) 13EURO score

Low 18Intermediate 11High 4

Previous treatments (No. of patients)HU 20IFN 7BUS 2PBSCT 2

Abbreviations: HU, hydroxyurea; IFN, interferon; BUS, busulphanPBSCT, autologous peripheral blood stem cell transplantation; Pplatelets.

eR)

re-

snted4

iree

4)

.630se.e-a

T,

(IFN-α), or busulfan. Two cases had also received previousautologous transplantations.

At time of diagnosis, all patients were Philadelphia posi-tive and expressed either the b2a2 or the b3a2BCR-ABLtranscript; only one patient had achieved cytogenetic remis-sion after IFN-α, but she had tested PCR-positive again justbefore starting Imatinib.

Patients in the CP received 400 mg/day of Imatinib; thosein AP received 600 mg/day.

Complete hematologic response was defined as the nor-malization of peripheral counts and WBC counts�10 ×109/L, platelets� 450 × 109/L, as well as myelocytes andmetamyelocytes�5%, without blasts or promyelocytes inthe peripheral blood and no extramedullary involvement.

Resistance was defined as the loss of the complete hema-tologic or cytogenetic remission lasting for at least 3 months,or the transformation to AP or BC after a period of CP.

2.2. Cytogenetic tests

Bone marrow cytogenetic analyses were assessed beforeand then every 3 months during Imatinib treatment. Cyto-genetic analysis was performed by standard binding tech-niques. At least 20 metaphases were analyzed for eachsample.

Cytogenetic responses were defined as follows: completeresponse (CCR), 100% Ph-negative metaphases; majorresponse (MCR), 65–99% Ph-negative metaphases; minorresponse (mCR), 10–64% Ph-negative metaphases; no re-sponse (NCR), more than 90% Ph-positive metaphases[13].

2.3. Molecular assays

All RT-PCR analyses were performed on the same bonemarrow samples that were used for the cytogenetic analysis.Total RNA was extracted using TriReagent (MolecularResearch Center, Cincinnati, OH) according to the manufac-turer’s instructions. Suitable aliquots were used for PCRtests after spectrophotometric quantitative evaluation.

cDNA synthesis reaction was performed with 1µg oftotal RNA in a total volume of 20µL containing 200 unitsof M-MLV reverse transcriptase (Invitrogen, Carlsbad, CA),1× first-strand buffer [50 mmol/L Tris-HCl (pH8.3), 75mmol/L KCl, 3 mmol/L MgCl2], 1 mmol/L dNTPs, 32 unitsof RNaseOUT (Invitrogen), 10 mmol/L DTT, and 5µmol/Lrandom primers. The synthesis programs included an initialincubation at 37�C for 10 minutes, followed by incubationat 42�C for 45 minutes. The reaction was inactivated byheating at 99�C for 3 minutes.

Qualitative PCR reactions were performed by using spe-cific primers forABL exon 3 andBCR exon 13, according torecommendations by European Concerted Action Biomed1[14]. Quantitative real-time RT-PCR assays were performedto quantify eitherBCR-ABL or MDR1 expression, and se-quences forBCR-ABL amplification were adopted accordingsuggestions made by the European Group Against Cancer

S. Galimberti et al. / Cancer Genetics and Cytogenetics 162 (2005) 57–62 59

redth

ing

n-S

anr’s

P,rs

re1

pnis

0rieCP

eenyoo

tiveem

iestaee

rd

-te

A

1

-

d

.

-

f

t

e

[15]. MDR1 expression was evaluated as published pviously [16]. The cycle threshold (Ct) value measurebefore the start of Imatinib was adopted as reference forcalculation of∆∆Ct (difference between∆Ct values derivingfrom difference between Ct of target and housekeepgenes) for subsequent samples[17].

2.4. Statistical analysis

All calculations were performed using the SPSS for Widows software, release 11-2002. Failure-free survival (FFand overall survival curves were estimated using the KaplMeier method. Comparisons were performed using Fisheexact test for binary variables and the Mann-Whitney tefor continuous variables. A value ofP � 0.05 was consid-ered significant.

3. Results

3.1. Hematologic and cytogenetic response

All patients, including the three cases treated in Aachieved a complete hematologic response within the fi4 weeks of therapy with Imatinib. Nevertheless, these thpatients showed disease progression after 8, 12, andmonths of treatment, respectively. With a median follow-uof 18 months (range 4.4–35.5 months), there was only odeath, which occurred at 68 months after diagnosis. Itrelevant that this patient started Imatinib in AP, 3months before death. The two-year FFS of the whole sewas 83%; none of the patients who were treated inshowed disease progression.

Twenty-two patients (66.7%) achieved a CCR and thr(9.1%) an MCR. Eight cases (24.2%) did not achieve aresponse, including patients treated in AP, despite their ghaematologic results.

Twenty-one of the 22 cases in CCR achieved Ph negaity within 3 months from the start of therapy, and thremaining one after 6 months. Cytogenetic responses windependent from sex, age, EURO risk score, interval frothe diagnosis, and Imatinib or previously adopted therap

Only one patient, who had reduced Ph-positive mephases to 70% after 6 months, showed a further increas90% after 9 months of treatment, concomitant with increasBCR-ABL andMDR1 expression.

3.2. Sensitivity of serial real-time RT-PCR

To test the sensitivity of the real-time RT-PCR, standacurves with serial dilutions of K562 (forBCR-ABL) andHL60/Dx (for MDR1) cell lines were performed and compared to those obtained with serial dilutions of appropriaplasmids.

Standards were purchased from Amplimedical S.p.(Turin, Italy); dilutions from 101 to 105 copies were used toperform these curves. These assays showed that 1 in5

cells was already detectable for both genes.

-

e

)-st

te7

e

s

d

-

re

.-tod

.

0

The sensitivity of the quantitative method was then testedin comparison to that of qualitative PCR: the two methodsshowed a concordance of 73%, with the quantitative methodshowing a significantly higher sensitivity.

3.3. Quantitative monitoring of BCR-ABLand MDR1 expression

After 3 months of treatment, theBCR-ABL transcript wasno longer detectable in 6 cases (18%); in the other 16cases (66.7%),BCR-ABL expression decreased by more than1.5 logs. This reduction did not correlate with any evaluatedclinical feature (sex, age, EURO risk score, previous therapeutic regimens, or time elapsed from diagnosis to beginningof Imatinib therapy).

After 6 months of Imatinib treatment, 20 patients (71.4%)presented a reduction ofBCR-ABL expression that exceededthan 2 logs.

Nevertheless, four patients (14.3%) showed increaseBCR/ABL transcript (median� 0.5 logs), a phenomenon thathad been not observed during the previous treatment period

After 12 months, 22.7% of patients were PCR negative,but in 36.4% of cases an increasedBCR-ABL expression(but always greater than 2 logs) was found.

After 18 months, 37.5% of patients were PCR negative,andBCR-ABL mRNA levels were comparable to those ob-served after one year of treatment.

Thus, quantitative molecular data showed that Imatinibwas able to significantly decreaseBCR-ABL expression inmore than two thirds of tested patients. The maximum spanof reduction was observed between the third and sixthmonths of therapy (Fig. 1).

This reduction ofBCR-ABL expression did significantlycorrelate with cytogenetic response: in the subgroup of patients in CCR,BCR-ABL expression values decreased by 3ormore logs in the 82.4% of cases,with median ratiovalues o0.02% after 3 months, 0.002% after 6 months, and 0.003%after 12 months. On the other hand, in cases that did noachieve the CCR, medianBCR-ABL expression values were0.81% after 3 months, 0.90% after 6 months, and 0.38%after 12 months of therapy; moreover, only 18% of these

Fig. 1. MedianBCR-ABL values; as represented in the figure,BCR-ABLexpression decreased especially in the first 3 months of treatment, with thminimum achieved after six months.

S. Galimberti et al. / Cancer Genetics and Cytogenetics 162 (2005) 57–6260

rt

l

s-n

yh-

n

dI

a

l

r

n-h

s;

-

d

o

.

.

tst,

patients presented a reduction ofBCR-ABL expressionhigher than 3 logs (Fig. 2).

Interestingly, patients who did not achieve a major ocomplete cytogenetic responses showed concomitanhigher levels (10-fold) ofMDR1 expression (P � 0.02).

Moreover, in the three cases in which a considerabchange in percentage of Ph-positive metaphase was detecduring treatment, levels ofMRD1 expression were concomi-tantly modified (Fig. 3).

As reported earlier, the reduction of theBCR-ABL mRNAwas not always persistent because these values increaduring follow-up in one third of the patients. This phenomenon was more evident after 6 months of therapy, passifrom 14.3 to 36.4% after one year of treatment (Fig. 4).

It is worthwhile to note that these increments were alwayless than 3 logs, and that they did not comport with anloss of hematologic or cytogenetic response. Supposing tthis increasedBCR-ABL expression could be a sign of resistance to Imatinib, we analyzedMDR1 gene expression indetail.

All tested samples wereMDR1 positive at diagnosis,probably because of the physiological expression of this gein hematopoietic stem cells as well[18].

After 3 months of Imatinib administration, decreaseMDR1 expression was measured in the 36% of patients.89% of cases there was a concomitant reduction ofBCR-ABL mRNA levels, and these percentages remained substtially the same until the follow-up at 6 months.

After 12 months of therapy,MDR1 mRNA levels werefurther decreased in 10 patients, but a concomitantdecreasedBCR-ABL expression was observed in only 32%of them. Finally, we measuredBCR-ABL andMDR1 expres-sion in the three patients who had showed disease pgression after an initial period of remission; all of thempresented either increased levels ofMDR1 or BCR-ABL,already 3 months before clinical progression.

4. Discussion

Sensitive molecular techniques, such as qualitative aquantitative RT-PCR, proved to be very useful for monitoring minimal residual disease in CML patients treated wit

Fig. 2. Comparison of medianBCR-ABL values between patients in CCRand cytogenetically non-responsive.P � 0.002.

ly

eted

ed

g

s

at

e

n

n-

y

o-

d

Fig. 3. Correlation between changes of cytogenetic pattern andMDR1expression in cases #2 (a), #15 (b), #22 (c). -----, positive methaphase______, MDR1 expression levels.

conventional therapies, including cases in complete cytogenetic remission[19].

Almost all patients remained PCR-positive after therapywith interferon but, interestingly, the median ratio ofBCR-ABL/ABL at maximal response was significantly higher inpatients who had relapsed than in those who had remainein CR [20].

The majority of allotransplanted patients are still positiveduring the first 6 months after grafting, and persistent PCRnegativity after one year is a marker for good prognosis[21].

The significance of PCR monitoring in patients treatedwith Imatinib has not been studied widely thus far, even ifthe ability to detect serially raisedBCR-ABL values could beuseful in managing further treatments (increasing Imatinibdosages, adding other agonist drugs, or proceeding tallotransplantation).

In this context, we used a quantitative RT-PCR to seriallymeasureBCR-ABL transcripts in 33 patients treated withImatinib, with the aim of investigating if eventually in-creasedBCR-ABL levels would correlate with raisedMDR1expression, as a sign of the emerging resistance to Imatinib

In our series, during follow-up, 12.1% of patients revertedto PCR positivity, and 36.4% showed increasedBCR-ABLlevels, even maintaining hematologic and cytogeneticresponses.

Other authors reported a significant prognostic role forBCR-ABL levels [22]: a reduction ofBCR-ABL/ABL ratioby more than 3 logs significantly correlated with better PFS

In our series,BCR-ABL expression decreased by 3 ormore logs in 82% of cases; nevertheless, 36.4% of patienshowed increased mRNA values after one year of treatmeneven if the transcripts increased by�3 logs.

To investigate if the observed increasedBCR-ABL valuescould be related to increasedMDR1 levels, MDR1 gene

S. Galimberti et al. / Cancer Genetics and Cytogenetics 162 (2005) 57–62 61

R

d

ed

n

d

s

nal

g.

fr

yleald

al

-

n

S,

,J,rtic

oid

ni

,illeicase2;

M,

e.

hpy1;

,rsd

eEB

,

tote

,aby

,nid

nte-

V,a,tsi-

n:ia

,F,

Fig. 4. Percentage of cases with decreased, stable, or increasedBCR-ABLvalues. Dark gray, stable values; dotted, decreased values; light gray, PCnegative patients; black, increased values.

expression was serially measured. Concomitantly increaseBCR-ABL and MDR1 mRNA values were observed in allthree patients who were treated in AP before loss of thinitial hematological response. This observation sustainethe hypothesis thatMDR1 is relevant in Imatinib resistance.

In the subgroup of cases treated in CP, the correlatiobetweenBCR-ABL and MDR1 expression was significantonly in the first 6 months of therapy. These results prompteus to suppose that the role of theMDR1 expression would bemore important when disease is not yet fully controlled, ain the case of the first Imatinib treatment. Moreover, oneshould bear in mind that only values increasing by more tha3 logs have been reported by other authors to be of reclinical prognostic significance[22].

On the other hand, data from the literature concerninMDR1-mediated resistance to Imatinib are still controversialIn selected cell lines resistant to Imatinib, Mahon et al.[8]showed that overexpression of theMDR1 gene or its proteinwas responsible for the resistance, and that inhibitors othe Pgp, such as verapamil and PSC833, were able to restothe in vitro sensitivity. When blast cells from six patientswho were in blast crisis and clinically resistant to Imatinibwere incubated with PSC833, they recovered their sensitivitbut, interestingly, no overexpression of Pgp was detectabin these samples. These ex vivo results could suggest thanother pump-mediated drug resistance mechanism coube effective in CML patients. As a result, other multidrugresistance-related genes and proteins, such as BCRP, a htransporter acting as a homo- or heterodimer in effluxingcytotoxic agents, very similar to Pgp[23], would be aninteresting field for future investigation.

Nevertheless, we cannot rule out other previously described mechanisms (e.g.,BCR-ABL amplification or muta-tions of the kinase site) because they may also be relevain the phenomenon of Imatinib resistance.

-

e

t

f-

t

References

[1] Deininger MWN, Goldman JM, Melo JV. The molecular biology ofchronic myeloid leukemia. Blood 2000;96:3343–56.

[2] Druker BJ, Tamura S, Buchdunger E, Onhno S, Segal GM, FanningZimmermann J, Lydon N. Effects of a selective inhibitor of ABLtyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med1996;2:561–6.

[3] Okuda K, Weisberg E, Gilliland DG, Griffin JD. ARG tyrosine kinaseactivity is inhibited by STI571. Blood 2001;97:2440–8.

[4] Kantarjian HM, Sawyers CL, Hochhaus A, Guilhot F, Schiffer CADeininger MW, Gambacorti-Passerini C, Stone RM, GoldmanFischer T, Rosamilia M, Zoellner U, Resta DJ, Capdeville R, DrukeBJ. Gleevec (imatinib mesylate) induces hematologic and cytogeneresponses confirmed and expanded in patients with chronic myelleukemia (CML) – a phase II study update. Blood 2001;98:845a.

[5] Talpaz M, Silver RT, Druker BJ, Goldman JM, Gambacorti-PasseriC, Guilhot F, Schiffer CA, Fischer T, Deininger MW, Lennard AL,Hochhaus A, Ottmann OG, Gratwohl A, Baccarani M, Stone RTura S, Mahon FX, Fernandes-Reese S, Gathmann I, CapdevR, Kantarjian HM, Sawyers CL. Imatinib induces durable hematologand cytogenetic responses in patients with accelerated phchronic myeloid leukemia: results of a phase 2 study. Blood 20099:1928–37.

[6] Druker BJ, Sawyers CL, Kantarjian HM, Resta DJ, Reese SF, Ford JCapdeville R, Talpaz M. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemiaand acute lymphoblastic leukemia with the Philadelphia chromosomNew Engl J Med 2001;344:1038–42.

[7] Gorre ME, Mohammed M, Ellwood K, Hsu N, Paquette R, NagesRao P, Sawyers CL. Clinical resistance to STI-571 cancer theracaused by BCR-ABL gene mutation or amplification. Science 200293:876–80.

[8] Mahon FX, Belloc F, Lagarde V, Chollet C, Moreau-Gaudry FReiffers J, Goldman JM, Melo JV. MDR1 gene overexpresion conferesistance to imatinib mesylate in leukemia cell line models. Bloo2003;101:2368–73.

[9] Juranka P, Zastawny R, Ling V. P-glycoprotein: multidrug-resistancand a superfamily of membrane-associated transport proteins. FASJ 1989;3:2582–5.

[10] Wuchter C, Leonid K, Ruppert V, Schrappe M, Buchner T, Schoch CHaferlach T, Harbott J, Ratei R, Dorken B, Ludwig WD. Clinicalsignificance of P-glycoprotein expression and function for responseinduction chemotherapy, relapse rate and overall survival in aculeukemia. Haematologica 2000;85:711–21.

[11] Carter A, Dann EJ, Katz T, Shechter Y, Oliven A, Regev R, Eytan ERowe JM, Eytan GD. Cells from chronic myelogenous leukemipatients at presentation exhibit multidrug resistance not mediatedeither MDR1 or MRP1. Br J Haematol 2001;114:581–90.

[12] Hasford J, Pfirrmann M, Hehlmann R, Baccarani M, Guilhot FMahon FX, Kluin-Nelemans HC, Ohnishi K, Thaler J, SteegmanJL. Prognosis and prognostic factors for patients with chronic myeloleukemia: nontransplant therapy. Semin Hematol 2003;40:4–12.

[13] Talpaz M, Kantarjian HM, McCredie K, Trujillo JM, Keating MJ,Gutterman JU. Hematologic remission and cytogenetic improvemeinduced by recombinant human interferon alpha in chronic myelognous leukemia. N Engl J Med 1986;314:1065–9.

[14] Van Dongen JJ, Macintyre EA, Gabert JA, Delabesse E, RossiSaglio G, Gottardi E, Rambaldi A, Dotti G, Griesinger F, ParreirA, Gameiro P, Diaz MG, Malec M, Langerak AW, San Miguel JFBiondi A. Standardized RT-PCR analysis of fusion gene transcripfrom chromosome aberrations in acute leukemia for detection of minmal residual disease. Report of the BIOMED-1 Concerted Actioinvestigation of minimal residual disease in acute leukemia. Leukem1989;13:1901–28.

[15] Gabert J, Beillard E, van der Velden VH, Bi W, Grimwade DPallisgaard N, Barl G, Cazzaniga G, Cayuela JM, Cave H, Pane

S. Galimberti et al. / Cancer Genetics and Cytogenetics 162 (2005) 57–6262

y

e

s

d

Aerts JL, De Micheli D, Thirion X, Pradel V, Gonzalez M, ViehmannS, Malec M, Saglio G, van Dongen JJ. Standardization and qualitcontrol studies of “real-time” quantitative reverse transcriptsfor residual disease detection in leukemia – A Europe Against Cancprogram. Leukemia 2003;17:2318–57.

[16] Galimberti S, Guerrini F, Palumbo GA, Consoli U, Fazzi R, Morabito F,Santini V, Petrini M. Evaluation of BCRP and MDR-1 co-expressionby quantitative molecular assessment in AML patients. Leuk Re2004;28:367–72.

[17] Livak KJ, Schmittgen TD. Analysis of relative gene expression datausing real-time quantitative PCR and the 2(-Delta Delta C(T)) methodMethods 2001;25:402–8.

[18] Drach J, Zhao S, Drach D, Korbling M, Engel H, Andreeff M. Expres-sion of MDR1 by normal bone marrow cells and its implication forleukemic hematopoiesis. Leuk Lymphoma 1995;16:419–24.

[19] Chomel JC, Brizard F, Veinstein A, Rivet J, Sadoun A, Kitzis A,Guilhot F, Brizard A. Persistence of BCR-ABL genomic re-arrangement in chronic myeloid leukemia patients in complete ansustained cytogenetic remission after interferon-α therapy or alloge-neic bone marrow transplantation. Blood 2000;95:404–9.

r

.

[20] Hochhaus A, Reiter A, Sauβele S, Reichert A, Emig M, Kaeda J,Schultheis B, Berger U, Shepherd PCA, Allan N, Hehlmann R,Goldman JM, Cross NCP, for the German CML study Group andthe UK MRC CML Study Group. Molecular heterogeneity in completecytogenetic responders after interferon-α therapy for chronic myeloidleukaemia: low levels of minimal residual disease are associated withcontinuing remission. Blood 2000;95:62–6.

[21] Elmaagacli AH, Freist A, Hahn M, Opalka B, Seeber S, Schaefer UW,Beelen DW. Estimating the relapse stage in chronic myeloid leukaemiapatients after allogeneic stem cell transplantation detected using anew real-time polymerase chain reaction method. Br J Haematol2001;13:1072–5.

[22] Hughes TP, KaedaJ, Brandford S, Rudzki Z, HochhausA, Hensley ML,Gathmann Bolton AE, van Hoomissen IC, Goldman JM, RadichJP. International Randomized Study of Interferon versus STI571 (IRIS)Study Group. Frequency of major molecular response to imatinib orinterferon alpha plus cytarabine in newly diagnosed chronic myeloidleukemia. N Engl J Med 2003;349:1399–401.

[23] Ross DD, Karp JE, Chen TT, Doyle LA. Expression of breast cancerresistance protein in blast cells from patients with acute leukemia.Blood 2000;96:365–8.