Bone Marrow Transplantation, (1998) 21, 159–166  1998 Stockton Press All rights reserved 0268–3369/98 $12.00

Detection of CBF␤/MYH11 fusion transcripts in patients with inv(16) acute myeloid leukemia after allogeneic bone marrow or peripheral blood progenitor cell transplantation

AH Elmaagacli1, DW Beelen1, M Kroll1, S Trzensky1, C Stein2 and UW Schaefer1

Departments of 1Bone Marrow Transplantation, and 2Department of Forensic Medicine, University Hospital of Essen, Essen, Germany

Summary: Keywords: AML; inv(16); CBF␤/MYH11; MRD; BMT

We evaluated the occurrence of the CBF␤/MYH11 fusion transcripts by PCR analysis in 10 patients with inv(16)(p13;q22) acute myeloid leukemia (AML) who Acute myelomonocytic leukemia with bone marrow eosino- underwent allogeneic bone marrow transplantation philia (AML-M4Eo), according to the French–American– (BMT) (n = 5), peripheral blood progenitor cell trans- British (FAB) classification, is a distinct subtype of AML. plantation (PBPCT) (n = 3), or autologous transplan- AML-M4Eo is often associated with rearrangements of tation (n = 2). In addition to the analysis of minimal chromosome 16, mostly involving 16p13 and 16q22, lead- residual disease (MRD), the chimerism status of patients ing either to a pericentric inversion, inv(16)(p13q22) or, after allogeneic transplant was studied by PCR. The less commonly, to a translocation between the homologous CBF␤/MYH11 fusion trancript was not detectable in six chromosomes, t(16;16)(p13;q22). The pericentric inversion of seven patients who remained in remission after allo- inv(16)(p13q22) is one of the most frequently occurring geneic BMT or PBPCT. Two of these patients in chromosomal rearrangements detected in this neoplasm, remission were monitored for 50 months and 64 months which has been reported to account for approximately 16% post-BMT. One patient in remission was PCR-positive of all AML.1–3 for CBF␤ 3 months post-BMT in a single BM sample, The breakpoints involved in the inv(16) have recently but not in a simultaneously examined blood sample, been cloned and shown to involve the core binding factor suggesting that analyses from BM samples are more ␤-gene (CBF␤) on 16q22 and the smooth muscle heavy sensitive than those from blood samples. Sequential chain (MYH11) on 16p13. The formation of the chimeric PCR assays performed 6 and 12 months post-BMT CBF␤/MYH11 fusion gene results in the disruption of the obtained from the same patient were negative. Another normal interaction of the transcription factor complex ␣, ␤, patient with a positive PCR assay 3 months post-allo- and ETS.3 geneic PBPCT, remained PCR positive for the Patients with AML-M4Eo with inv(16)(p13q22) appear CBF␤/MYH11 fusion transcript when tested 6 months to have a high response rate to induction chemotherapy and post-PBPCT. A chimerism analysis by PCR revealed a a favorable prognosis, but relapse of AML after chemo- mixed chimerism status in this patient. He relapsed 7 therapy remains a major cause of treatment failure. Allo- months post-transplant. Before transplant, in all nine geneic bone marrow transplantation (BMT) has become an patients who were in complete remission of AML (eight effective treatment modality for patients with hematological patients in 1CR, one patient in 2CR), the CBF␤/MYH11 malignancies. Patients with AML in remission who transcript was detectable. In one patient in relapse, the undergo allogeneic BMT have a significantly lower risk of fusion transcript was not only detectable in blood and relapse than their counterparts treated with chemotherapy bone marrow, but also in a cerebrospinal fluid sample alone. On the other hand allogeneic BMT is associated with prior to transplant. Two patients who received autolog- more treatment-related mortality than autologous BMT or ous BMT were monitored for CBF␤/MYH11 transcripts chemotherapy alone. 3 months after BMT. The CBF␤/MYH11 was detected Several reports suggest that the CBF␤/MYH11 fusion in these patients. Both patients subsequently relapsed 3 mRNA can often be detected in patients in long-term months and 23 months post-autologous BMT. The remission after chemotherapy.4–7 To date, other than single results study show that analysis of the CBF␤/MYH11 case reports, studies of the detection of the CBF␤/MYH11 fusion transcript by PCR seems to be a suitable method fusion transcripts after allogeneic BMT do not exist.7,8 for monitoring minimal residual disease in AML The aim of the present study was to determine if patients with inv (16). CBF␤/MYH11 fusion transcripts can be detected in patients treated with allogeneic BMT (n = 5), allogeneic peripheral blood progenitor cell transplantation (PBPCT) (n = 3), or Correspondence: Dr AH Elmaagacli, Department of Bone Marrow Trans- = plantation, University Hospital of Essen, Hufelandstr. 55, 45122 Essen, autologous BMT (n 2) for inv(16) positive AML at our Germany institution. In addition to the analysis of minimal residual Received 3 July 1997; accepted 13 August 1997 disease (MRD), the chimerism status of patients after allo- CBF␤/MYH11 fusion transcripts in patients with AML AH Elmaagacli et al 160 geneic BMT or PBPCT was studied by PCR of minisatellite one busulfan (4 mg/kg/day for 4 days), thiotepa (5 gene regions (VNTR-PCR). mg/kg/day for 2 days) and cyclophosphamide (60 mg/kg/day × 2). All transplants were performed without prior ex vivo removal of donor lymphocytes from the graft. Materials and methods Irradiated leukocyte-depleted blood products were used exclusively for blood component substitution throughout Patients the post-transplant course. Prophylaxis of acute GVHD of patients treated by allo- Ten patients with inv(16) leukemia who underwent allo- geneic transplantation consisted of short-course methotrex- geneic transplantation with bone marrow (n = 5) or periph- ate and cyclosporin A (CsA) in five patients, of CsA alone eral blood progenitor cells (n = 3), or autologous BMT (n in three patients, as shown in Table 1. = 2) at our institution were studied (Table 1). Six patients were diagnosed with de novo FAB M4Eo, and four patients were diagnosed with AML M4. The median age at diag- nosis was 31 years (range 17–40). Seven patients were male Morphologic and cytogenetics and three female. Patients were transplanted in first remission (1CR, eight AML was diagnosed according to standard FAB morpho- 9 patients), second remission (2CR, one patient), and first logic and cytochemical criteria. Chromosome analyses relapse (1rel, one patient). Two patients were transplanted were performed on bone marrow cells (24 and/or 48 h in with autologous bone marrow, and eight patients received vitro culture with GTG (G-bands obtained by trypsin and allogeneic grafts. Seven donor/recipient pairs were matched stained with Giemsa)). In all patients a pericentric inversion at the HLA-A, B, C and DR loci and were nonreactive in 16 could be demonstrated prior to transplant. mixed lymphocyte culture (MLC), and one patient (UPN 936) received a graft from an antigen-mismatched unrelated donor (one mismatch of six). All recipients of allogeneic Isolation of RNA and cDNA synthesis and reverse transplantation except one received grafts from family transcription donors. BM or blood samples were studied 1 month before transplantation and 1 to 64 months after transplant. The RNA was prepared from fresh or cyropreserved peripheral median time to last follow-up for all patients was 6 months. blood cells or BM buffy coat cells, Kasumi-1 and K562 Approval for this study was obtained from the Institutional cells provided negative controls for RT-PCR. Blood from Review Board on Medical Ethics at the Essen University a patient (UPN 799) with a CBF␤/MYH11 fusion transcript Hospital. All patients gave informed consent before was used as a positive control. RNA was extracted by the material was obtained. acid guanidium/phenol/chloroform method.10 For each PCR, cDNA was synthesized from 2 ␮g of total RNA with 200 U Moloney murine leukemia virus (MoMuLV) reverse Conditioning regimens and GVHD prophylaxis transcriptase (Gibco-BRL, Gaithersburg, MD, USA) in The conditioning regimen consisted in five patients of PCR buffer (50 mmol/l KCl, 10 mmol/l Tris-HCl pH 8.3, cyclophosphamide (60 mg/kg/day × 2) and fractioned total 1.5 mmol/l MgCl2, 0.001% gelatin (Perkin Elmer Cetus, body irradiation (TBI) delivered from a cobalt source in Weiterstadt, Germany), using random hexamers (10 four daily fractions of 2.5 Gy to a total dose of 10 Gy. mmol/l) (Boehringer, Mannheim, Germany), deoxynucleo- Four patients received busulfan (4 mg/kg/day for 4 days) tide triphosphates 8dNTP; 1 mmol/l each), 20 U RNAsin followed by cyclophosphamide (60 mg/kg/day × 2), and at a final volume of 20 ␮lat37°C for 1 h.

Table 1 Clinical characteristics of the AML patients with inv(16)

No. UPN Sex Age Trans Disease GVHD Condit GVHD HLA Present status R/D stage proph regimen ac/chr constel (months)

1 176 — 17 autoBMT 1CR — Bu + CY — — Relapse/death (3) 2 189 — 40 autoBMT 1CR — Bu + CY — — Relapse/death (23) 3 469 M/F 34 BMT 1CR CsA + MTX TBI + CY —/lim de novo si/id DSF (69) 4 554 M/F 34 BMT 1CR CsA + MTX Bu + CY II/lim si/id DFS (54) 5 799 M/F 32 PBPCT 1CR CsA + MTX TBI + CY II/lim si/id Relapse/death (7) 6 856 M/M 30 BMT 1CR CsA Bu,Th,CY II/lim bro/id Death through infection (12) 7 868 F/F 25 BMT 1CR CsA TBI + CY I/lim si/id DFS (12) 8 876 F/M 36 PBPCT 2CR CsA + MTX Bu,Th,CY II/lim bro/id DFS (10) 9 882 F/F 25 BMT 1CR CsA TBI + CY IV/ext si/id DFS (10) 10 936 M/M 35 PBPCT Relap CsA + MTX TBI + CY II/— unr/MM Death through infection (3)

UPN = unique patient number; R = recipient; D = donor; GVHD = graft-versus-host disease; 1CR/2CR, first/second remission; PBPCT = peripheral blood progenitor cell transplantation; BMT = bone marrow transplantation; MTX = methotrexate; CsA = cyclosporin A; TBI = total body irradiation; CY = cyclophosphamide; Bu = busulfan; Th = thiotepa; bro = brother; id = identical; unr = unrelated; si = sister; lim = limited; ext = extensive; DFS = disease-free survivor. CBF␤/MYH11 fusion transcripts in patients with AML AH Elmaagacli et al 161 Detection of CBF␤/MYH11 fusion product ␮g of total cellular RNA per reaction; (2) be successfully coamplified for the GAPDH or ␤-actin transcript in each Ten microliters of cDNA was used in a total volume of 50 × reaction; (3) be successfully coamplified for a positive con- ␮l containing 1 PCR buffer 50 mmol/l KCl, 10 mmol/l trol in both reactions. Tris-HCl pH 8.3, 1.5 mmol/l MgCl2, 0.001% gelatin Oligonucleotide primers were synthesized on an Applied (Perkin Elmer Cetus), 0.2 mmol/l each of NTP, 2.5 U Taq Biosystems 380 B DNA synthesizer (Foster City, CA, polymerase (Perkin Elmer Cetus) and 0.3 ␮mol/l of each USA). of the primers C1, 5Ј GCAGGCAAGGTATATTTGAAGG 3Ј and M1, 5Ј CTCTTCTCCTCATTCTGCTC 3Ј.4 After an initial denaturation step of 3 min at 95°C, 35 cycles of 30 Isolation of genomic DNA sat95°C, 30 s at 57°C and 60 s at 72°C were performed For chimerism studies, 1 ␮g of DNA was prepared from followed by a final extension of 8 min at 72°C. peripheral blood mononuclear cells obtained from the In the second round of PCR, 5 ␮l of the first PCR pro- donor and recipient before transplantation using a standard ducts were added in a 50 ␮l total volume containing 1 × phenol/chloroform method.13 Post-transplant, DNA was PCR buffer, 50 mmol/l KCl, 10 mmol/l Tris-HCl pH 8.3, extracted exclusively from the peripheral blood to allow 1.5 mmol/l MgCl2, 0.001% gelatin, 200 ␮mol/l of each determination of chimerism. NTP. 0.3 ␮l of inner primers C2, 5Ј ACACGC GAATTTGAAGATAGAG 3Ј, and M7, 5Ј TTC- Minisatellite PCR TCCAGCTCATGGACCTCC 3Ј,4 and 2.5 U Taq DNA polymerase. Amplification was performed with an initial For PCR amplification we used specific primers designed denaturation step of 3 min at 95°C, followed by 25 cycles to flank repeat units of the following human minisatellite of denaturation at 95°C for 30 s and annealing at 57°C for regions: D1S80 as previously described.14 These loci were 30 s. selected because they exhibit a high degree of polymor- The C1M1 and C2M7 product of type A was amplified phism and are suitable for PCR amplification. as a band at 420 bp (C1M1) or 65 bp (C2M7) on a 4% PCR of minisatellite region D1S80: PCR reaction was agarose gel stained with ethidium bromide (data of C1M1 performed in a volume of 50 ␮l with 10 ng of blood DNA product are not shown). template using a commercially purchased PCR amplifi- cation kit from Perkin Elmer. DNA was denatured for 4 min at 95°C, followed by 29 cycles at 95°C (15 s), 66°C RNA quality control (15 s), and 72°C (40 s). After the last cycle, the samples The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were held at 72°C for 10 min and finally at 4°C. messenger RNA or ␤-actin mRNA were coamplified as a control in all RT-PCR.4,11 From all samples, the cDNA was PCR primers for analysis of Y chromosome-specific divided into two tubes. One tube was used as an RNA qual- nucleotide sequences ity control containing a final volume of 25 ␮l with 1 × PCR buffer with 50 mmol/l KCl, 10 mmol/l Tris-HCl pH 8.3, Chimerism analysis of male patients with female donor were perfomed by PCR amplification of Y-chromosome- 1.5 mmol/l MgCl2, 0.001% gelatin (Perkin Elmer Cetus), 15 0.2 mmol/l each of NTP, 2.5 U Taq polymerase (Perkin specific nucleotide sequences as previously described. A Elmer Cetus) and 12 pmol/l of each GAPDH specific pri- 150 bp sequence was amplified with the primers H3 and H4.15 The ZP3 gene located on chromosome 7 was used mers or ␤-actin specific primers, which results in a positive 16 200 bp band or 420 bp band from all human RNA, and as a control gene with the primers ZP1 and ZP2. therefore, controls for the quality of RNA and successful Briefly, reagents were transferred to 0.5 ml PCR-reaction PCR amplification.4,11 tubes. The reaction volume was 25 ␮l and contained 50 mmol/l KCl, 10 mmol/l Tris pH 8.3 at 20°C, 0.001% gela- tin, 0.2 mmol/l of each dNTP, 20 pmol of each primer, 2 PCR controls U of AmpliTaq DNA polymerase (Perkin Elmer Cetus), 1.5 mmol/l MgCl2 and 250 ng of blood DNA template. DNA Elaborate measures were taken to minimize contamination ° 12 was denatured for 4 min at 94 C, followed by 35 cycles at following the recommendations of Kwok and Higuchi. 94°C (1 min), 65°C (1 min), and 72°C (2 min). RNA and DNA extractions were performed at separated work places using dedicated PCR reagents and pipettes. All PCR products were kept in separate laboratory rooms from Analysis of PCR products patients samples, RNA and PCR reagents. Amplified DNA of minisatellite region D1S80 was run on Blank controls were included in all RNA extraction pro- a high resolution gel according to the manufacturer’s cedures as negative controls to assess quality and cross- instructions (Gene Amp Detection Gel; Perkin Elmer contamination between samples. Cells from Kasumi-1 and Cetus). Following electrophoresis, the gels were silver K562 cell lines were used as a negative control, and cells stained and thereafter photographed. of a CBF␤/MYH11-positive patient were included as a positive control. Sensitivity of PCR assays For a patient sample to be considered negative for ampli- fication of the CBF␤/MYH11 fusion transcript, it had to: Chimerism studies: Using the minisatellite primer D1S80 (1) be amplified in two independent experiments using 2.0 a sensitivity of 1:103 was obtained as previously pub- CBF␤/MYH11 fusion transcripts in patients with AML AH Elmaagacli et al 162 lished.14 The sensitivity of the Y-chromosome-specific PCR was 1:105.15 UPN

MRD studies: To assess the sensitivity of the RT-PCR 936 Infection amplification of the CBF␤/MYH11 fusion transcripts, 1 ␮g of total RNA isolated from BM of patient UPN 936 at pres- entation prior to transplantation was serially diluted in RNA 882 isolated from mononuclear peripheral blood cells of a heal- thy donor. The resulting cDNA was then submitted to 35 876 cycles of amplification with primers C1 and M1, followed by 25 cycles of amplification with nested primers C2 and M7. 868

856 Infection Results 799 R Detection of CBF␤/MYH11 fusion transcript in patients with inv(16) AML before transplant 554 In all 10 patients, diagnostic or relapse BM or blood samples obtained before allogeneic or autologous transplant 469 were available for amplification by RT-PCR. Eight patients were in first remission, one was in second remission (UPN 876) and another was in first relapse (UPN 936) (Figure 1). 189 R In all patients analyzed prior to transplant, a 65 bp nested 176 R RT-PCR product consistent with a type A CBF␤/MYH11 fusion transcript could be detected in bone marrow or blood 1 3 6 9 12 24 36 48 60 samples. In patient UPN 936 a 420 bp RT-PCR product Months from the first PCR step consistent with type A CBF β /MYH11 negative complete chimerism CBF␤/MYH11 fusion transcript could be demonstrated in CBF β /MYH11 positive mixed chimerism blood and bone marrow (data not shown). A cerebrospinal Figure 1 MRD and chimerism post-transplant. fluid sample obtained from this patient prior to transplant was also found to be positive in the second step of RT-PCR analysis by detection of a 65 bp RT-PCR product (Figures 2 and 3). In all patients who were in complete remission of AML, a 65 bp nested RT-PCR product consistent with a type A MV positiveUPN control 176negativeno control RNA CBF␤/MYH11 fusion transcript was detected in the bone marrow sample.

Detection of CBF␤/MYH11 fusion transcript post- autologous transplant

Two patients (UPN 176 and 189) were transplanted with 65 bp> autologous marrow. Bone marrrow samples of one of these patients (UPN 176) obtained 3 months post-autologous transplant showed the presence of type A CBF␤/MYH11 Figure 2 RT-PCR results of patient (UPN 176) with AML after autolog- ous BMT. Type A of CBF␤/MYH11 fusion transcript after two-step PCR fusion transcript (Figure 2). This patient relapsed a short at 65 bp. MV, DNA molecular weight marker PhiX174 DNA-HaeIII time after the PCR analysis 3 months post-transplant and digest (Boehringer Mannheim). died. Another patient was analyzed for CBF␤/MYH11 fusion Detection of CBF␤/MYH11 fusion transcript post transcripts 3 months and 23 months post autologous trans- allogeneic transplant plant. Three months post-transplant, this patient was in a complete remission demonstrated by bone marrow cyto- Five patients received an allogeneic bone marrow trans- genetic and histological analysis. However, RT-PCR-analy- plant, and three patients were transplanted with PBPCs. All sis was positive for the CBF␤/MYH11 transcript at this eight patients initially achieved a complete remission observation time. Twenty-three months post-transplant, the demonstrated by routine cytogenetic and histological PCR test remained positive for the CBF␤/MYH11 tran- examination. script and this patient subsequently relapsed. All eight patients were assayed for the CBF␤/MYH11 CBF␤/MYH11 fusion transcripts in patients with AML AH Elmaagacli et al 163 analyses. These three patients remained in complete remission according to morphologic and cytogenetic bone marrow analyses.

MV positiveUPN control 936negativeno control RNA One patient (UPN 936) was found to be positive for CBF␤/MYH11 in blood and cerebrospinal fluid prior to PBPCT, but had a negative PCR assay 1 and 3 months post-transplant in BM, blood and cerebrospinal fluid. This patient achieved a complete remission after PBPCT but developed a severe infection and died 3 months post-allo- geneic PBPCT (Figures 3 and 4). 65 bp> Two patients (UPN 469 and 554) could be monitored 64 and 50 months after allogeneic BMT for CBF␤/MYH11 fusion transcripts. Both patients were found to be negative for CBF␤/MYH11 in BM samples which were obtained 3 Figure 3 RT-PCR of a cerebrospinal fluid sample (UPN 936) prior to months post-BMT. PCR analysis 50 and 64 months post- allogeneic PBPCT. BMT from these two patients remained negative for the fusion transcript. Morphological and cytogenetic BM analysis revealed a remaining complete remission for both patients.

MV positiveUPN control 936negativeno control RNA

Chimerism in AML-M4Eo patient

The chimerism status was studied in all eight AML-M4Eo patients using peripheral blood samples. In three male patients with a female donor the chimerism analysis was performed by PCR amplification of Y-chromosome-specific sequences. In five patients PCR amplification of the minisa- tellite region D1S80 was used to analyze their chimerism Figure 4 RT-PCR of same patient (UPN 936) 1 month after allo- status. geneic PBPCT. One month post-transplant chimerism analyses were per- formed in six of eight patients. In none of the six patients fusion transcript 3 months post-transplant (Figure 1). Two studied was a mixed chimerism detected. of these eight patients (UPN 799, 856) showed a positive Three months post-transplant chimerism analyses were PCR assay, whereas the remaining six patients were performed in all eight patients who received an allogeneic CBF␤/MYH11 negative. One patient of the two patients transplant. All patients showed a complete chimerism when who was positive for CBF␤/MYH11 (UPN 799) 3 months analyzed at that time. Patient UPN 799 revealed a mixed post-transplant remained PCR-positive for the fusion tran- chimerism status 6 months post-transplant demonstrated by scripts when tested 6 months post-transplant. This patient performing a PCR of Y-chromosome-specific gene was transplanted with allogeneic PBPCs. He relapsed 7 sequences. An MRD analysis performed at the same time months post-transplant by developing an extramedullary showed him to be positive for CBF␤/MYH11. This patient tumor. A histological examination of a biopsy sample of developed an extramedullary relapse of the AML-M4Eo 7 the extramedullary tumor showed a relapse of the AML- months post-transplant. M4Eo leukemia. After surgery was done and chemotherapy The remaining four patients who were tested to have a was given this patient achieved a transient remission again, complete chimerism 1 month or 3 months post-transplant, but remained PCR-positive for CBF␤/MYH11 fusion tran- remained with complete chimerism when analyzed at a later script when monitored up to 16 months post-transplant. time period, including two patients who were analyzed 64 This patient died 18 months post-PBPCT. and 50 months post-transplant. Another patient (UPN 856) was PCR positive for CBF␤/MYH11 in a BM sample, but not in a simultaneously examined blood sample, suggesting that analyses from BM samples are more sensitive than blood samples. A BM sam- ple obtained 6 months post-BMT from the same patient was Sensitivity of the RT-PCR for the detection of the found to be negative by PCR. This patient died 12 months CBF␤/MYH11 fusion transcript post-BMT from a severe CMV infection. Sequential PCR tests were performed 1, 3, 6, and 12 Using an amount of 1 ␮g RNA in all RT-PCR, the sensi- months post-transplant in three patients (UPN 890, 876, tivity analysis showed that the RT-PCR technique could 868). All three patients were found to be negative for the reliably detect 100 pg of rearranged RNA in 1 ␮g of RNA CBF␤/MYH11 fusion transcript in these sequential PCR of normal cells (data not shown). CBF␤/MYH11 fusion transcripts in patients with AML AH Elmaagacli et al 164 Discussion results were found in four of five patients, using a PCR assay with a low sensitivity of 10Ϫ4. In three other reports, 11 patients with inv(16) were CBF␤/MYH11 transcripts are found in the great majority monitored for CBF␤/MYH11 fusion transcripts after of inv(16) AML and also in approximately 10% of patients achieving a complete remission by chemotherapy. Claxton with AML-M4.14 Four types (A–D) of fusion transcripts et al5 studied three AML patients with inv(16) who had for this gene have been described, representing different hematological remissions; one patient was RT-PCR-posi- breakpoints within the MYH11 gene, with type A being tive with morphologic remission at day 21 after therapy for predominant.4–8 relapse, and two patients in remission for more than 1 year The common A-type fusion transcript was found in all were RT-PCR-negative. Poirel et al6 reported RT-PCR pos- AML patients with inv(16) studied here prior to BMT or itivity in three AML/inv(16) patients studied at 1, 3 and 6 PBPCT. months after diagnosis. Herbert et al4 studied five AML- In the present report, we provide the largest series to date M4Eo patients with inv(16): four patients in remission for of leukemia patients with inv(16) studied for the evidence 4–22 months had detectable CBF␤/MYH11 fusion tran- of the CBF␤/MYH11 fusion transcript following allogeneic scripts while one patient in complete remission for 60 BMT. Before this series, analyses of the CBF␤/MYH11 months was RT-PCR negative. were reported in only three cases following allogeneic The different results which were reported in several stud- BMT. In one of these three cases, 22 months post-allog- ies above might be caused primarily by using PCR tech- eneic BMT, CBF␤/MYH11 fusion transcripts were niques with different sensitivities and different RNA detected only in bone marrow, but not in peripheral blood. amounts. According to Jurlander et al18 the sensitivity of a In two cases CBF␤/MYH11 could not be detected by RT- RT-PCR assay is influenced by the amount of RNA which PCR when analyzed 3 months post-transplant. RT-PCR was is used in the PCR. In our study we therefore used 2 ␮g performed in one case from bone marrow, and in the other RNA in all PCR to ensure a sensitivy of 10Ϫ5 in all PCRs. case only from peripheral blood.7,8 With regard to monitoring minimal residual disease, The results presented in this study demonstrate that the there might be some similarities found between the detec- CBF␤/MYH11 fusion trancript was no longer detectable in tion of CBF␤/MYH11 fusion transcripts and AML1/ETO six of the seven patients who were in remission after allo- transcripts in patients with AML. As described above for geneic BMT or PBPCT. Two of these patients in remission the detection of CBF␤/MYH11 transcripts in patients with were monitored up to 50 and 64 months post-BMT. inv(16), the AML1/ETO transcript is also detectable in One patient who remained in complete remission after patients in remission after chemotherapy.19–24 Therefore, allogeneic BMT, was PCR positive 3 months post-BMT in studies after chemotherapy or autologous BMT on a BM sample, but not in a simultaneously examined blood CBF␤/MYH11 or AML1/ETO may not be very predictive sample, suggesting that analyses from BM samples are in regard to leukemic relapse when a persistence of these more sensitive than blood samples. This observation was fusion transcripts was reported. In these cases, a reliable also described by Tobal et al.7 quantitative PCR assay may be helpful. However, the Another patient with a positive PCR test 3 months post- reliability of semiquantitative competitive PCR assays as PBPCT, remained PCR positive for the CBF␤/MYH11 developed by Evans et al17 has not yet been sufficiently fusion transcript when tested 6 months post-PBPCT. This determined and needs to be proven by further studies from patient suffered from leukemic relapse 7 months post-trans- other investigators. plant. Patients in remission after allogeneic BMT or PBPCT Before transplant, in all nine patients who were in com- did not show a persistence of CBF␤/MYH11 fusion tran- plete remission (eight patients in 1CR, one patient in 2CR), scripts in this report. A negative PCR assay for AML1/ETO the CBF␤/MYH11 transcript was detectable. In patient is also found in many patients in remission after allogeneic UPN 936, who was transplanted in relapse, the fusion tran- BMT or PBPCT studied for AML1/ETO.19 This shows that script was not only detectable in blood and bone marrow, patients who were positive for CBF␤/MYH11 prior to allo- but also in a cerebrospinal fluid sample. geneic BMT or PBPCT, seem to be able to obtain a molecu- Two patients who received an autologous BMT were lar remission after transplant. Therefore, it may confirm on monitored for CBF␤/MYH11 transcripts 3 months after a molecular level the superior antileukemic effect of an transplant. The CBF␤/MYH11 transcript was detected in allogeneic BMT or PBPCT towards that of chemotherapy both patients. Both patients subsequently relapsed 3 and 23 and autologous BMT in patients with AML. months post-autologous BMT. Similarities between AML1/ETO and CBF␤/MYH11 are Studies of the detection of CBF␤/MYH11 transcripts in not only seen in their use as MRD markers. Moreover, the patients who were in remission after chemotherapy or auto- CBF␤ protein and the AML-1 protein are able to interact logous transplantation, show a persistence of the fusion with each other. So, CBF␤ is a human homologue of the transcript in a number of patients.4–7 Using a competitive ␤-subunit of the heterodimeric polyomavirus enhancer core PCR assay, Evans et al17 found in five of eight patients binding factor protein, PEBP2␣␤, a murine transcription positive PCR results for CBF␤/MYH11 transcripts after factor, whereas the protein AML-1 has been identified and chemotherapy alone. Costello et al8 reported PCR positivity fulfills the requirements for the DNA-binding ␣-subunit. for CBF␤/MYH11 fusion trancripts in five of six patients Therefore, AML-1 directly interacts with CBF␤. It has been within 2 months after the induction therapy, whereas after shown that in the majority of cases, the chimeric protein autologous transplantation with PBPCs negative PCR CBF␤/MYH11 retains its ability to interact strongly with CBF␤/MYH11 fusion transcripts in patients with AML AH Elmaagacli et al 165 AML-1, the DNA-binding component of the transcription 6 Poirel H, Radford-Weiss I, Rack K et al. Detection of the factor complex. Binding sites for this transcription factor chromosome 16 CBF␤-MYH11 fusion transcript in myelo- have been identified in a number of viral enhancers and monocytic leukemias. Blood 1995; 85: 1313–1322. mammalian cellular genes, including the myeloid-specific 7 Tobal K, Johnson PRE, Saunders MJ et al. Detection of CD13 and myeloperoxidase genes and the T cell-specific CBF␤/MYH11 transcripts in patients with inversion 16 and other abnormalities of chromosome 16 at presentation and genes interleukin-3, Ick, as well as the T cell antigen 2,25 remission. Br J Haematol 1995; 91: 104–108. receptor. 8 Costello R, Sainty D, Blaise D et al. Prognosis value of Chimerism analyses using sensitive PCR techniques residual disease monitoring by polymerase chain reaction in might help to increase the possible predictive value of PCR patients with CBF beta/MYH11-positive acute myeloblastic assays for CBF␤/MYH11 fusion transcripts post-transplant leukemia. Blood 1997; 89: 2222–2223 (letter). with regard to relapse. By having a lower sensitivity usually 9 Bennett JM, Catovsky D, Daniel MT et al. Proposed revised by a factor of 101 to 102, chimerism analyses by VNTR- criteria for the classification of acute myeloid leukemia. A PCR may detect a larger number of cells of host-type hema- report of the French–American–British cooperative group. topoiesis. Thus, it has been demonstrated that in CML Ann Intern Med 1985; 103: 620–624. patients after BMT, mixed chimerism is associated with a 10 Chomczynski P, Sacchi N. Single step method of RNA iso- 15,26 lation by acid guanidium thiocyanate phenol chloroform higher incidence of MRD and risk of leukemic relapse. extraction. Anal Biochem 1987; 162: 156–159. In this study, we reported that in a single patient who suf- 11 Dveksler GS, Basile AA, Dieffenbach CW. Analysis of gene fered from a leukemic relapse 7 months post-allogeneic expression: use of oligonucleotide primers for glyceraldehyde- PBPCT the positive PCR assay for CBF␤/MYH11 was also 3-phosphate dehydrogenase. PCR Meth Appl 1992, pp 283– associated with a mixed chimeric status 6 months post- 284. transplant, whereas all other patients with a complete chim- 12 Kwok S, Higuchi R. Avoiding false positives with PCR. erism in the VNTR-PCR assay and negative PCR assays for Nature 1989; 339: 237–238. CBF␤/MYH11 transcripts remained in complete remission. 13 Maniatis T, Fritsch EF, Sambrook J. Molecular Cloning. A The performance of chimerism analyses by PCR Laboratory Manual. Cold Spring Harbor Laboratory Press: methods may help to identify patients with a high risk of Cold Spring Harbor, NY, 1982. 14 Elmaagacli AH, Beelen DW et al. Molecular studies of chim- leukemic relapse after allogeneic BM or PBPC transplan- erism and minimal residual disease after allogeneic peripheral tation. blood progenitor cell or bone marrow transplantation. Bone Marrow Transplant 1996; 18: 397–403. 15 Elmaagacli AH, Becks HW, Beelen DW et al. Detection of minimal residual disease and persistence of host-type hemato- poiesis: a study in 28 patients after sex-mismatched, non-T Acknowledgements cell-depleted allogeneic bone marrow transplantation for Phil- adelphia-chromosome positive chronic myelogenous leuke- mia. Bone Marrow Transplant 1995; 16: 823–829. This work was supported by grants of ‘Aktion Kampf dem Krebs’ 16 Cui K-H, Warnes GM, Jeffrey R Colin DM. Sex determination der Deutschen Krebsgesellschaft and ‘Deutsche Krebshilfe’, grant of pre-implantation embryos by human testis-determining- No. 106430–4. gene amplification. Lancet 1994; 343: 79–82. 17 Evans PAS, Short MA, Jack AS et al. Detection and quantit- ation of the CBF␤/MYH11 transcripts associated with inv(16) in presentation and follow-up samples from patients with AML. Leukemia 1997; 11: 364–369. References 18 Jurlander J, Caligiuri MA, Ruutu T et al. Persistence of the AML1/ETO fusion transcript in patients treated with allog- 1 Le Beau MM, Larson RA, Bitter MA et al. Association of an eneic bone marrow transplantation for t(8;21) leukemia. Blood inversion of chromosome 16 with abnormal marrow eosino- 1996; 88: 2183–2191. phils in acute myelomonocytic leukemia: a unique cyto- 19 Miyamoto T, Nagafuji K, Akashi K et al. Persistence of multi- genetic-clinical pathological association. New Engl J Med potent progenitors expressing AML1/ETO transcripts in long- 1983; 309: 630–636. term remission patients with t(8;21) acute myelogenous leuke- 2 Shurtleff SA, Meyers S, Hierbert SW et al. Heterogeneity in mia. Blood 1996; 87: 4789–4796. CBF␤/MYH11 fusion messages encoded by the 20 Kusec R, Laczika K, Knobl P et al. AML1/ETO fusion mRNA inv(16)(p13q22) and the t(16;16)(p13;q22) in acute myelogen- can be detected in remission blood samples of all patients with ous leukemia. Blood 1995; 85: 3696–3703. t(8;21) acute myeloid leukemia after chemotherapy or autolog- 3 Okuda T, van Deursen J, Hiebert SW et al. AML1, the target ous bone marrow transplantation. Leukemia 1994; 8: 735–739. of multiple chromosomal translocations in human leukemia, 21 Downing JR, Head DR, Curcio-Brint AM et al.An is essential for normal fetal liver hematopoiesis. Cell 1996; AML1/ETO fusion transcript is consistently detected by RNA- 84: 321–324. based polymerase chain reaction in acute myelogenous leuke- 4 Herbert J, Cayuela JM, Danie MT et al. Detection of minimal mia containing the (8;21)(q22;q22) translocation. Blood 1993; residual disease in acute myelomonocytic leukemia with 81: 2860–2865. abnormal marrow eosinophils by nested polymerase chain 22 Saunders MJ, Tobal K, Yin JA. Detection of t(8;21) by reverse reaction with allele specific amplification. Blood 1994; 84: transcriptase polymerase chain reaction in patients in 2291–2296. remission of acute myeloid leukaemia type M2 after chemo- 5 Claxton DF, Liu P, Hsu HB et al. Detection of fusion tran- therapy or bone marrow transplantation. Leukemia Res 1994; scripts generated by the inversion 16 chromosome in acute 18: 891–895. myelogenous leukemia. Blood 1994; 83: 1750–1756. 23 Nucifora G, Larson RA, Rowley JD. Persistence of the 8;21 CBF␤/MYH11 fusion transcripts in patients with AML AH Elmaagacli et al 166 translocation in patients with acute myeloid leukemia type M2 ation of subunits of the T cell receptor and murine leukemia in long-term remission. Blood 1993; 82: 712–715. virus enhancer core-binding factor. EMBO J 1993; 13: 24 Guerrasio A, Rosso C, Martinelli G et al. Polyclonal haemo- 3324–3329. poiesis associated with long-term persistence of the 26 Mackinnon S, Barnett L, Heller G, O’Reilly RJ. Minimal AML1/ETO transcript in patients with FAB M2 acute myeloid residual disease is more common in patients who have mixed leukemia in continous clinical remission. Br J Haematol 1995; T cell chimerism after bone marrow transplantation for 90: 364–373. chronic myelogenous leukemia. Blood 1994; 83: 3409–3416. 25 Wang S, Wang Q, Crute BE et al. Cloning and characteriz-