Long-Template DNA Polymerase Chain Reaction for the Detection of the bcr/abl Translocation in Patients with Chronic Myelogenous Leukemia1Cornelius F.Waller,Gereon Dennebaum, Christoph Feldmann,and Winand Lange2 Department of Internal Medicine I,Hematology/Oncology,Albert Ludwigs University Freiburg Medical Center,D-79106Freiburg, GermanyABSTRACTIn most patients with chronic myelogenous leukemia (CML)primitive hematopoietic progenitors carry the ac-quired reciprocal bcr/abl gene rearrangement t(9;22)(q34.1; q11.21).However,not all of the progenitor cells express the bcr/abl hybrid mRNA or the p210fusion protein.These cells, therefore,might escape detection by techniques that are based on expression of the fusion gene.To circumvent this problem,we established a new detection method for the rearrangement at the DNA level.Because breakpoints might occur in a very large genomic region(>200kb),we devel-oped a long-template DNA-PCR(LT-DNA-PCR).In22of59 CML patients,fragments of up to19kb could be amplified. Furthermore,6of7leukapheresis products of three bcr/abl-positive patients which were collected after mobilization chemotherapy and had been shown to be negative for the bcr/abl rearrangement by FISH and by RT-PCR were clearly positive by ing a specific pair of primers,it is possible to detect the presence of,and to characterize,the individual gene rearrangement.This ap-proach could serve for diagnostic purposes as well as detec-tion of minimal residual disease under cytotoxic therapy or after purging regimens,being independent of expression of the bcr/abl hybrid mRNA or the fusion protein. INTRODUCTIONCML3is a clonal hematological stem cell malignancy. Usually,a typical peripheral blood count leads to the diagnosis.In95%of patients the Philadelphia chromosome(Ph)is present (1,2).This is the result of an acquired reciprocal translocation t(9;22)(q34.1;q11.21)of the c-abl oncogene from chromosome 9to the bcr on chromosome22(3).Due to this positioning effect a patient-specific chimeric fusion gene ually, fusion gene products of two slightly different sized mRNAs of approximately8.5kb are expressed,b2a2and b3a2.Both of them encode for a Mr210,000fusion protein(p210bcr/abl),a protein with increased tyrosine kinase activity(4,5).In the bcr,the translocation occurs somewhere in an intron within a5.8-kb genomic region(6,7).In abl,the chromosomal breakpoints are dispersed over an intron region as large as about 200kb(8).Complete intron sequence is available only for the bcr introns,whereas in abl,approximately95kb of the200kb are yet unknown(Fig.1A).Two research groups independently reported that primitive CML progenitors carry the rearrangement but do not necessarily express the bcr/abl hybrid mRNA or the p210fusion protein(9, 10).Because of this,these cells might escape detection by techniques that are based on expression of the fusion gene. Therefore,we intended to establish a method for detection of patient-specific rearrangements at the DNA level.Unfortu-nately,the breakpoints in CML are dispersed over such a large genomic region that at present,it is not possible to cover them by conventional PCR methods.In1994,Cheng et al.(11) showed effective amplification of long targets from human genomic DNA using a recombinant Thermus thermophilus DNA polymerase.In accordance with their approach,we de-signed a set of LT-PCRs with one constant bcr primer and10abl primers at a distance of approximately15kb each.Depending on the translocation site the first abl primer3Јof the transloca-tion together with the constant5Јbcr primer should be able to amplify the fusion region(Fig.1B).MATERIALS AND METHODSPatient Samples.Peripheral blood from59patients with CML,4healthy volunteers,3control patients and three Phϩcell-lines(BV173,K-562,and LAMA84)were analyzed (Table1).Furthermore,samples from leukapheresis products of three patients with CP-CML collected after mobilization chem-otherapy consisting of idarubicin,cytosine arabinoside,and etoposide were analyzed(Table3).Samples were drawn after informed consent was obtained according to institutional guide-lines.RNA Isolation.Total cellular RNA from1ϫ105to1ϫ106cells from peripheral blood or thawed leukapheresis samples was extracted using a RNeasy Total Kit(Quiagen,Hilden, Germany)according to the manufacturer’s instructions.RT-PCR.RT-PCR amplification using nested primers was performed according to standard protocols as described previously(12).Received5/6/99;revised8/24/99;accepted9/10/99.The costs of publication of this article were defrayed in part by thepayment of page charges.This article must therefore be hereby markedadvertisement in accordance with18U.S.C.Section1734solely toindicate this fact.1Supported in part by the W.Sander-Stiftung(C.F.W.,W.L.),theDeutsche Krebshilfe(W61/93/La1),and the Zentrum Klinische For-schung I,Albert Ludwigs University Freiburg(C.F.W.,W.L.).2To whom requests for reprints should be addressed,at Department ofHematology/Oncology,University of Freiburg,Hugstetter Strasse55,D-79106Freiburg,Germany.Phone:49-761-270-3852;Fax:49-761-270-3418;E-mail:Lange@l.uni-freiburg.de.3The abbreviations used are:CML,chronic myelogenous leukemia;CP,chronic phase;Ph,Philadelphia chromosome;bcr,breakpoint clusterregion;LT-PCR,long template-PCR;RT-PCR,reverse transcription-PCR;FISH,fluorescence in situ hybridization.4146Vol.5,4146–4151,December1999Clinical Cancer ResearchDNA Isolation.DNA was isolated using 1–10ml of EDTA-blood or samples from leukapheresis products with a Quiagen Blood and Cell Culture DNA Midi Kit (Quiagen)according to the manufacturer’s instructions.Three ϫ107cells yielded a median of 97.6␮g genomic DNA,which was dis-solved in Tris-EDTA buffer (pH 8.0)and stored at Ϫ20°C.LT-DNA-PCR Primer Design.A constant bcr -primer and 10abl -primers at a distance of approximately 15kb each were designed (Fig.1B ).All of the 22-mer primers where chosen with an annealing temperature of 68.0–68.7°C and a GC content of 50–55%.Primers were adjusted to 20pmol per reaction.abl genomic localization (Fig.1B ;Table 2A )refers to GenBank sequences hsablgr1(access code (ac):UO7561),hsablgr2(ac:UO7562),hsablgr3(ac:UO7563)and bcr to hsuo7000(ac:UO7000).The sequence of used primers is shown in table 2A .LT-PCR.Initially,a set of ten PCR reactions per patient was performed.After identification of a patient characteristic amplification product the specific primer combination was used for subsequent analysis.A Perkin-Elmer GeneAmp XL PCR Kit (Perkin-Elmer,Foster City,CA)was used.Briefly,the reaction mix (100␮l)is divided into a lower reagent mix,the upperreagent mix and the sample.The lower reagent mix (40␮l)was composed of 12␮l of 3.3XL Buffer II;10m M dNTPs,20pmol of each abl primer and the bcr B2primer and 25m M Mg(OAc)2.The volume was adjusted to 40␮l with dH 20.To this mix,an AmpliWax PCR Gem 100(Perkin-Elmer)was added and melted at 75°C for 5min.Thereafter,as soon as the wax hardened,the upper reagent mix was placed above the solid wax layer.The upper reagent mix (20␮l)consisted of 18.5␮l of 3.3XL Buffer II and 3units of rTth (recombinant T.thermophilus )polymerase.The sample was added to the upper reagent mix.It was composed of 0.5␮g of DNA (in 0.5␮l)and 39.5␮l of dH 20.Division of the lower and upper reagent mix allowed a hot-start technique and ensured that all of the tubes had a synchronized start time.Cycle conditions were as follows:de-naturation time for 10s at 93°C;annealing and extension time,starting at 10min and increased by 10s per cycle both at 68°C.Thirty-six cycles were performed with a final hold at 72°C for 10min.All of the steps were performed using a Perkin-Elmer DNA Thermal Cycler.Restriction Enzyme Digestion.Two ␮l of 10ϫrestric-tion endonuclease buffer,7␮l of dH 2O,and 1␮l of restriction endonuclease (4–5units)were added to 10␮l of PCR products and then incubated at 37°C for 60min.The reaction was stopped by adding 0.5␮l of 5M EDTA buffer (pH 8.0).Agarose Gel Electrophoresis.For the analysis of ampli-fied products,agarose gel electrophoresis was performed to visualize PCR products.Fifteen ␮l of the amplification product were loaded per lane and run at a constant voltage of 80V.A 0,8%SeaKemGold Agarose gel (FMC BioProducts,Rockland,ME)was used and stained with ethidium bromide.Hybridization with Specific bcr Probes and DNA Se-quencing.Aliquots of amplified products were dot-blotted and hybridized with eight digoxigenin probes about 400bp apart spanning the bcr region between exons 13and 15(Fig.1C ).The DIG easy Hyb system (Boehringer,Mannheim,Germany)to detect complementary sequences was used according totheFig.1A ,exon-intron distribution .Map of bcr and abl exon-intron region.The bcr gene with the two exons (b2and b3)including the 5.8-kb bcr;abl with exons (1b ,1a ,and 2)and the 200-kb breakpoint region.hsablgr refers to GenBank sequences hsablgr1(access code:UO7561),hsablgr2(access code:UO7562),hsablgr3(access code:UO7563);bcr,refers to hsuo7000(access code:UO7000);dashed line ,unknown sequence.B ,primer-location,restriction enzyme recognition sites.Primers A-2and 3–1to 3–9are complementary to abl sequences;primer B2represents the constant bcr primer.C ,localization of hybridization oligonucleotides within the bcr gene.Table 1Patient and LT-DNA-PCR data Samples n LT-DNA-PCR-positiveCML,1st CP 4521CML,2nd CP 40CML,AP a 61CML,BC40CML cell lines 31Healthy volunteers 40AML 20Ph ϩALL1aAP,accelerated phase;BC,blast crisis;AML,acute myeloid leukemia;Ph ϩALL,Philadelphia chromosome-positive acute lympho-cytic leukemia.4147Clinical Cancer Researchmanufacturer’s instructions.The most3Ј-located probe that gave a positive signal was subsequently used as sequencing primer for sequence analysis of the breakpoints at the3Јend of the bcr gene and the5Јend of the abl gene.Sequence analysis was performed by PCR cycle sequencing using the Big Dye Terminator Cycle Sequencing Ready Reaction(PE Applied Biosystems,Foster City,CA)with an automated sequencer (DNA Sequencer,model373A,PE Applied Biosystems).FISH.Cells from leukapheresis products were analyzed by interphase FISH for quantitation of Phϩcells as described previously(13).The threshold of detection was determined at Ͼ6%to classify any specimen as Ph-positive.FISH was per-formed using a DNA probe mixture for the M bcr/abl translo-cation(Oncor,Gaithersburg,MD)according to the manufactur-er’s instructions.RESULTSIn22(37.3%)of59CML patients,fragments of up to19 kb could be amplified by LT-PCR showing an individual patient characteristic PCR product(Table1).The breakpoints were dispersed throughout intron1b without any obvious clustering. In one patient,the breakpoint was in intron1a.With the excep-tion of primer3–3,3–7,and3–8,every abl primer generated fragments in selected patients.To prove the identity of the amplified fragments,restriction endonuclease digestion was used in four patients.Fragments matching exactly the calculated size according to the available sequence information could be generated(Fig.2).In addition,nine of the amplified products were further characterized by sequence analysis.In eight samples,the break-points on the abl gene could be located in published sequences, whereas in one sample a localization was impossible(Fig.3).In DNA dilution experiments,the sensitivity of our LT-PCR method was determined to be25pg,the DNA content of approximately four primary CML cells(Fig.4A)or the equiv-alent of four BV173cells(data not shown;a fragment ofϽ1kb was amplified).RT-PCR was able to detect the RNA equivalent of about one cell(Fig.4B).Six of seven leukapheresis products from three patients that repeatedly had been found to be negative for the bcr/abl rear-rangement by RT-PCR and interphase FISH were clearly posi-tive by LT-DNA-PCR(Fig.5).Only one of seven apheresis products was constantly negative by all three of the methods (Table3).All three of the patients were positive by RT-PCR for bcr/abl before mobilization chemotherapy. DISCUSSIONA point of controversy is the question whether all of the CML cells actively express the hybrid fusion gene or not(14). Primitive CML progenitor cells were shown to be RT-PCR-negative despite carrying the bcr/abl translocation.In these cells,bcr/abl mRNA might be either absent or minimally ex-pressed.Antisense oligonucleotides targeted against the hybrid mRNA might,therefore,not be effective as eradication tech-niques(9).A second research group reported that23%of Phϩmyeloid colonies were found to be bcr/abl mRNA-negative (10).Both of the studies have been criticized because of their inadequate controls for cDNA synthesis(15,16).Now that we are able to use PCR at the DNA level,the presence of gene expression as mRNA or protein is no prereq-Table2Primer and sequence dataPrimer Location a Sequence5Ј–3ЈPosition(GenBank)A.Primer locations,sequences,and genomic positionsB2bcr exon b2CAGAAGCTTCTCCCTGACATCC hsuo7000;123599–123620A2abl exon a2ATTATAGCCTAAGACCCGGAGC hsablgr3;50667–506463-1abl intron1b CTGGATCTGAAGCTCCCAGATG hsablgr3;34346–343253-2abl intron1b TTTCCCCTGATCAGTCTGGGTA hsablgr3;16817–167963-3abl intron1b GCCATGGGAGAAGAATATCCCG hsablgr3;228–2073-4abl intron1b CCATGCCCACATTCAGGACTTG hsablgr2;58991–589703-5abl intron1b CTACCAACTTCACACCAGCCTG hsablgr2;44891–448703-6abl intron1b GGGAGAGAGAAAAGACCATGGG hsablgr2;29633–296123-7N abl intron1b TGGAACAAGGGGTCAGAGTGAT hsablgr2;15276–152553-8abl intron1b TGGTCTCCACTATTCAAGGGACA hsablgr2;318–2973-9N abl intron1b CACCCCCATATCCCACATCTAA hsablgr1;35766–35745B.Hybridization primer locations and sequenceshyb0bcr intron13TCAGATGCTC TGTGCCTT hsuo7000;123606–123623hyb1bcr intron13ACACAGTGTC CACCGGAT hsuo7000;123955–123972hyb2bcr intron13CCTGCAGGTG GATCGAGT hsuo7000;124377–124394hyb3bcr intron14TGAGTTGCAC TGTGTAAG hsuo7000;124618–124635hyb4bcr intron14TTAGGTGAGA GCAGTGTC hsuo7000;125047–125064hyb5bcr intron14CATGGCCAAG CCAGAAAC hsuo7000;125474–125491hyb6bcr intron14TACACCTCTC TGTCCCCA hsuo7000;125930–125945hyb7bcr intron14GCGTCTACAG GGACACAG hsuo7000;126319–126336a abl genomic localization refers to GenBank sequences hsablgr1(access code(ac):UO7561),hsablgr2(ac:UO7562),hsablgr3(ac:UO7563);bcr genomic location refers to hsUO7000(ac:UO7000).4148LT-DNA-PCR for bcr/abl in Patients with CMLuisite for positive testing.There have been earlier attempts to establish detection methods at the DNA level.A bubble PCR technique was described to facilitate cloning of bcr/abl break-points (14).DNA from patients with CP-CML was diluted into normal DNA and subjected to two-step PCR.It is necessary to clone and sequence the breakpoint from each patient and to design patient specific oligonucleotide primers.Therefore,this technique involves an appreciable amount of time and effort and,in addition,has a lower sensitivity than RT-PCR (17).On the other hand,the new LT-PCR is an easy way to perform a one-step PCR,once a patient-specific primer pair has been identified.In comparison with RT-PCR,the DNA LT-PCR offers certain advantages.Because of the independence of gene ex-pression,genomic DNA could be more informative for the detection of residual disease or for very immature selected progenitors (14).A technical advantage of LT-PCR is the han-dling of stable DNA instead of RNase sensitive RNA and using patient-specific primers would help to minimize the problem of PCR contamination.In this first study,we used a set of indi-vidual LT-PCR reactions to show the feasibility of the method-ology.In future studies,however,multiplex LT-PCR with all of the different primer pairs in the same reaction tube might facilitate the practicability of this approach.To our knowledge,only about one-half of the sequence of the involved part of the abl gene is known (Fig.1).Nevertheless,it was possible to amplify the breakpoint region in more than one-third of the examined samples.In addition,further characterization of the amplified products by endo-nuclease restriction digest in four patients and sequence analysis at the breakpoint were successful in eight of nine analyzed samples.Once the complete intron sequence of abl will be available,every breakpoint should be representable after appropriate primer design.With reference to the break-point site in the abl gene,our results do not yet support data published by Jiang et al.(18).In our findings,the breakpoints were equally dispersed over intron 1b and did not cluster,whereas Jiang et al.suggested three cluster regions—30Ϯ5,Fig.2Agarose gel electrophoresis of one representative patient sam-ple.A characteristic fragment of approximately 6000bp could be amplified with the constant bcr primer and the abl primer 3–2(Lane 2).Restriction enzyme digestion produced fragments of approximately 5400bp and 600bp,respectively.On the basis of sequence information,the calculated size of the smaller fragment is nes 1and 4,␭HindIII.Fig.3Breakpoint localization in the abl gene .Individual breakpoints were dispersed throughout the entire abl intron sequence without clus-tering.Fig.4A ,sensitivity of LT-DNA-PCR.An approximately 4400-bp fragment could be generated with the constant bcr primer and abl primer 3–5in one representative patient with 100%Ph ϩhematopoiesis at the time of analysis.On the basis of sequence analysis,the calculated size is 4341bp.For amplification,a dilution series of template DNA was used as follows:Lane 2:0.25␮g;Lane 3:25ng;Lane 4:2.5ng;Lane 5:250pg;Lane 6:25pg;Lane 7:2.5pg;Lane 8:250fg;Lane 9:25fg;Lanes 10and 11:not loaded;Lanes 1and 12:␭Hin dIII.B ,sensitivity of RT-PCR .For amplification,a dilution series of BV173mRNA was used as follows:Lane 2:5␮g;Lane 3:500ng;Lane 4:50ng;Lane 5:5ng;Lane 6:500pg;Lane 7:50pg;Lane 8:5pg;Lane 9:0.5fg;Lanes 1and 10:␾X174/HAE III.4149Clinical Cancer Research100Ϯ13,and135Ϯ8kb—downstream from exon1b.In one case,the breakpoint was identified in intron1a,a finding previously reported in three other patients(7,17).Neverthe-less,because of the relatively small number of cases exam-ined,these findings should not be overrated.A potential disadvantage of LT-DNA-PCR could be single-copy target amplification,whereas,in RT-PCR,high expression per sin-gle cell might allow potentially more sensitive detection. However,our results comparing RT-versus LT-DNA-PCR showed comparable sensitivity no matter whether large or small fragments were amplified from DNA templates.LT-PCR analysis of seven leukapheresis products of three CP-CML patients after mobilization chemotherapy showed re-producibly the presence of bcr/abl at the DNA level in six of them whereas analysis by RT-PCR and FISH failed to detect the fusion gene.The sensitivity of FISH does not allow detection of the rearranged bcr/abl gene unlessϾ5%of rearranged cells are present within the analyzed cell population,a percentage also comparable to classical cytogenetic analysis.This fact is well known from patients receiving␣-IFN therapy who become negative by cytogenetics or FISH,but are still positive by RT-PCR. Considering the sensitivity of RT-PCR and LT-DNA-PCR about equal,our results imply either that some rearranged cells do not express the bcr/abl fusion gene or that the expression is rather low, possibly just below the detection limit of RT-PCR.The data that is shown support the usefulness of our method as a diagnostic tool and as a means to detect residual disease even in the absence of gene expression in CML patients. It will probably improve the evaluation of purging strategies in CML.In addition,a modification of the methodology could also be used to detect breakpoints associated with Ph-positive acute lymphocytic leukemia.ACKNOWLEDGMENTSWe thank J.Fischer and A.Rosenstiel for excellent technical assistance.Also we would like to acknowledge Dr.M.Follo for se-quence analysis.REFERENCES1.Carella,A.M.,Frassoni,F.,Melo,J.,Sawyers,C.,Eaves,C.,Eaves,A.,Apperley,J.,Tura,S.,Hehlmann,R.,Reiffers,J.,Lerma,E.,and Goldman,J.New insights in biology and current therapeutic options for patients with chronic myelogenous leukemia.Haematologica,82:478–495,1997.2.Gale,R.P.,Grosveld,G.,Canaani,E.,and Goldman,J.M.Chronic myelogenous leukemia:biology and therapy.Leukemia(Baltimore),7: 653–658,1993.3.Bartram,C.R.,de Klein,A.,Hagemeijer,A.,van Agthoven,T., Geurts van Kessel,A.,Bootsma,D.,Grosveld,G.,Ferguson-Smith, M.A.,Davies,T.,Stone,M.,Heisterkamp,N.,Stephenson,J.R.,and Groffen,J.Translocation of the c-abl oncogene adjacent to a translo-cation break point in chronic myelocytic leukemia.Nature(Lond.),306: 277–280,1983.4.Konopka,J.B.,and Witte,O.N.Detection of c-abl tyrosine kinase activity in vitro permits direct comparison of normal and altered abl gene products.Mol.Cell.Biol.,5:3116–3123,1985.5.Shtivelman,E.,Lifshitz,B.,Gale,R.P.,and Canaani,E.Fused transcript of abl and bcr genes in chronic myelogenous leukemia. Nature(Lond.),315:550–554,1985.6.Groffen,J.,Stephenson,J.R.,Heisterkamp,N.,de Klein,A.,Bar-tram,C.R.,and Grosveld,G.Philadelphia chromosomal breakpoints are clustered within a limited region,bcr,on chromosome22.Cell,36: 93–99,1984.7.Heisterkamp,N.,Stam,K.,Groffen,J.,de Klein,A.,and Grosveld,G.Structural organization of the bcr gene and its role in the PhЈtranslocation.Nature(Lond.),315:758–761,1985.8.Bernards,A.,Rubin,C.M.,Westbrook,C.A.,Paskind,M.,and Baltimore,D.The first intron in the human c-abl gene is at least200 kilobases long and is a target for translocations in chronic myelogenous leukemia.Mol.Cell.Biol.,7:3231–3236,1987.9.Bedi,A.,Zehnbauer,B.A.,Collector,M.I.,Barber,J.P.,Zicha, M.S.,Sharkis,S.J.,and Jones,R.J.BCR-ABL gene rearrangement and expression of primitive hematopoietic progenitors in chronic myeloid leukemia.Blood,81:2898–2902,1993.10.Keating,A.,Wang,X.H.,and Laraya,P.Variable transcription of BCR-ABL by Phϩcells arising from hematopoietic progenitors in chronic myeloid leukemia.Blood,83:1744–1749,1994.Table3Analysis of leukapheresis productsPatientno.Apheresisno.FISH RT-mRNA-PCR LT-DNA-PCR 11Negative Negative Positive2Negative Negative Positive3Negative Negative Negative 21Negative Negative Positive2Negative Negative Positive 31Negative Negative Positive2Negative Negative Positive Fig.5Apheresis sample analysis.Analysis of two apheresis sampleseach of patients S.I.(Lanes2and3)and H.G.(Lanes4and5).Molecular weight marker in Lane1.a,abl mRNA RT-PCR to controlintegrity of the RNA preparation.All of the samples are positive;bandsat106bp.b,bcr/abl mRNA RT-PCR to detect the b2a2or b3a2mRNA.All of the samples are negative;no bands at234bp.c,bcr/abl LT-DNA-PCR;for patient S.I.,primers abl3–2and bcr B2and,for patientH.G.,primers abl3–4and bcr B2were used to generate amplificationproducts of approximately6.0kb(Lanes2and3and Lanes4and5,respectively.4150LT-DNA-PCR for bcr/abl in Patients with CML11.Cheng,S.,Fockler,C.,Barnes,W.M.,and Higuchi,R.Effective amplification of long targets from cloned inserts and human genomic A,91:5695–5699,1994.12.Maurer,J.,Kinzel,H.,Nentwig,T.,and Thiel,R.Molecular diag-nosis of the Philadelphia chromosome in chronic myelogenous and acute lymphoblastic leukemia by PCR.Disease Markers,8:211–218, 1990.13.Heinzinger,M.,Waller,C.F.,Rosenstiel,A.,Scheid,S.,Burger, K.J.,and Lange,W.Quality of IL-3and G-CSF-mobilized peripheral blood stem cells in patients with early chronic phase CML.Leukemia (Baltimore),12:333–339,1998.14.Zhang,J.G.,Goldman,J.M.,and Cross,N.C.Characterization of genomic BCR-ABL breakpoints in chronic myeloid leukemia by PCR. Br.J.Haematol.,90:138–146,1995.15.Cross,N.C.,Lin,F.,and Goldman,J.M.Appropriate controls for reverse transcription polymerase chain reaction(RT-PCR).Br.J. Haematol.,87:218,1994.16.Melo,J.V.,Kent,N.S.,Yan,X.H.,and Goldman,J.M.“Controlsfor reverse transcriptase-polymerase chain reaction amplification of BCR-ABL transcripts”(Letter).Blood,84:3984–3986,1994.17.Zhang,J.G.,Lin,F.,Chase,A.,Goldman,J.M.,and Cross,N.C. Comparison of genomic DNA and cDNA for detection of residual disease after treatment of chronic myeloid leukemia with allogeneic bone marrow transplantation.Blood,87:2588–2593,1996.18.Jiang,X.Y.,Trujillo,J.M.,and Liang,J.C.Chromosomal break-points within the first intron of the ABL gene are nonrandom in patientswith chronic myelogenous leukemia.Blood,76:597–601,1990.4151Clinical Cancer Research。