Letters to the Editor
treatment, marrow blasts increased to 40%. FISH for BCR-ABL remained negative (also negative with the sorted CD34 þ / CD38 þ and CD34 þ /CD38- population) whereas 7 was present in 82% of the marrow cells. Quantitative PCR analysis was not performed. However, with nested RT-PCR using sets of BIOMED primers2 the e1a2 transcript was still detectable suggesting presence of a very small and quiescent population of imatinib-resistant BCR-ABL-positive leukemic stem cells. There are only a few reports on the use of imatinib in Ph þ AML.3–7 Most cases reporting molecular remission with imatinib had short follow-up and long-term clinical and molecular data is not available. Our case of AML with 45,XX,inv(3)(q21q26),7, t(9;22)(q34;q11.2) shows two recognized secondary cytogenetic anomalies in advanced CML, that is, monosomy 7 and chromosome aberrations involving band 3q21or 3q26,8 which initially would have suggested a case of CML blast crisis. However, simultaneous presence of 7 and inv(3)(q21q26) in blastic CML is rare (http://cgap.nci.nih.gov/Chromosomes). We propose that owing to the secondary nature of the Ph chromosome, the rarity in which these three cytogenetic changes are found together in blast crisis, the lack of p210BCR-ABL protein transcript and the absence of splenomegaly, our case is more in keeping with de novo Ph þ AML over myeloid blastic CML. This is confirmed by FISH studies, which revealed the Ph chromosome is acquired as a late event during clonal evolution. It is salutary to note that combination of FISH and RT-PCR analysis was needed to avoid the erroneous conclusion of imatinib resistance as the size of the BCR-ABL clone went down which is not compatible with BCR-ABL-dependent imatinib resistance.
SF Yip1, TSK Wan1, HSY Liu2, MLG Wong3, C-C So1 and LC Chan1 1 Department of Pathology, Queen Mary Hospital, The University of Hong Kong, Hong Kong SAR, China;
2
Department of Medicine, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR, China and 3 Department of Pathology, Tuen Mun Hospital, Hong Kong SAR, China E-mail:
[email protected]
2051
References 1 Yip SF, Wan TSK, Lie AKW, Liu HSY, Chan LC. Monitoring of chronic myeloid leukemia (CML) imatinib response by fluorescence-in-situ hybridization (FISH). Blood 2005; 106 (Part 2): 289b. 2 van Dongen JJ, Macintyre EA, Gabert JA, Delabesse E, Rossi V, Saglio G et al. Standardized RT-PCR analysis of fusion gene transcripts from chromosome aberrations in acute leukemia for detection of minimal residual disease. Report of the BIOMED-1 concerted action: investigation of minimal residual disease in acute leukemia. Leukemia 1999; 13: 1901–1928. 3 Cividin M, Brizard F, Sorel N, Renaud M, Guilhot F, Brizard A. p190(BCR-ABL) rearrangement as a secondary change in a case of acute myelo-monocytic leukemia with inv(16)(p13q22). Leuk Res 2004; 28: 97–99. 4 Jentsch-Ullrich K, Pelz AF, Braun H, Koenigsmann M, Mohren M, Wieacker P et al. A Complete molecular remission in a patient with Philadelphia-chromosome positive acute myeloid leukemia after conventional therapy and imatinib. Haematologica 2004; 89: ECR15. 5 Viniou NA, Vassilakopoulos TP, Giakoumi X, Mantzouranis M, Pangalis GA. Ida-FLAG plus imatinib mesylate-induced molecular remission in a patient with chemoresistant Ph1+ acute myeloid leukemia. Eur J Haematol 2004; 72: 58–60. 6 Ito K, Tominaga K, Suzuki T, Jinnai I, Bessho M. Successful treatment with imatinib mesylate in a case of minor BCR-ABL-positive acute myelogenous leukemia. Int J Hematol 2005; 81: 242–245. 7 Yamaguchi M, Konishi I. Successful treatment with imatinib mesylate for Philadelphia chromosome-positive refractory acute myeloid leukemia. Rinsho Ketsueki 2003; 44: 254–256. 8 Johansson B, Fioretos T, Mitelman F. Cytogenetic and molecular genetic evolution of chronic myeloid leukemia. Acta Haematol 2002; 107: 76–94.
Wilms’ tumor 1 mutation accumulated during therapy in acute myeloid leukemia: biological and clinical implications
Leukemia (2006) 20, 2051–2054. doi:10.1038/sj.leu.2404389; published online 21 September 2006
Molecular biology is a powerful tool for investigating acute myeloid leukemia (AML). Given that concerted efforts have documented the reliability and clinical utility of the real-time quantitative polymerase chain reaction (RQ-PCR) for detecting fusion transcripts in AML,1 it is not surprising that similar assays have been sought for in the sizeable fraction of AML cases without balanced translocations. The Wilms’ tumor gene 1 (WT1) has received close attention in this respect. Initially described in childhood genitourinary cancers, the gene has subsequently been shown to be overexpressed in a number of different cancers. Although its exact function in the evolvement of a malignant clone is still a matter of debate, its role as an minimal residual disease (MRD) target in myelodysplastic syndrome and AML is now amply documented.2,3 Indeed, recent data from our laboratory suggest that the assay is able to predict relapse months ahead of morphological relapse.2,4
In de novo AML relapses are often thought to emerge from exactly the same clone as at diagnosis. This is in contrast to the situation in, for example, pre-B acute lymphoid leukemia, where IgH gene rearrangements have disclosed clonal instability. In this report, we present evidence that clonal heterogeneity with regard to WT1 may be present at diagnosis and that chemotherapy-resistant subclones can be selected for during chemotherapy. A 5-year-old boy was diagnosed with AML-M6 in April 2003. The immunophenotype was CD13 þ , CD14, CD33 þ , CD117 þ , and HLA class II þ . Karyotypic analysis did not demonstrate abnormalities (46,XY[25]). In addition, no recurrent balanced translocations could be found by multiplex PCR5 and the patient was negative for internal tandem duplication of the FLT3 gene. Two courses of induction chemotherapy were necessary to obtain complete remission (CR). He was treated according to NOPHO-AML93 protocol and completed therapy in September 2003. He remained in CR1 until March 2004, at which time relapse with an identical immunophenotype, morphology and karyotype was diagnosed. He received reinduction Leukemia
Letters to the Editor
2052 chemotherapy and obtained CR2. A stem cell transplantation (SCT) with unrelated donor was performed in June 2004, but a second relapse was diagnosed in November 2004 with identical immunophenotype and morphology whereas cytogenetics displayed 46,XY,t(6;12)(p12;p11)[35]. This second relapse proved refractory to salvage therapy and the patient died in January 2005. All biologic material from the patient was derived from samplings performed as part of the diagnostic process. This was
a
done in accordance to protocols approved by the Ethical Committee for the County of Aarhus, Denmark. Mononuclear cells (MNC) were regularly obtained from peripheral blood (PB) and bone marrow (BM) by Lymphoprep density centrifugation (Axis-Shield PoC AS, Oslo, Norway). Total RNA was prepared on a MagNa-Pure LC robot for automated nucleic acid purification (Roche Diagnostics, Basel, CH). Standard analyses for WT1 quantification and MRD calculations were done as previously described by RQ-PCR using the
WT1 expression, standard vs. alternative primer/probe set 1 CR1
Relapse
SCT
Relapse
0.1
0.01
0.001
0.0001
0.00001 Mut. DNA/ 5%/ Mut. cDNA 3% Apr-03
0%/ no WT1 Jun-03
Aug-03
Oct-03
NA/ 3% Dec-03
Feb-04
NA/ 100%
Apr-04
Jun-04
60%/ 95%/ 100% 100%
Aug-04
Oct-04
Dec-04
b
wt
del
del
wt
del
wt
c MW
250
98 64 50 36 30
1
2
3
4
MW 1
2
3
4
250
98 64 50 36 30
Figure 1 (a) WT1 expression in PB over time from diagnosis to second relapse. Y axis: log reduction from diagnostic sample. The clinical state at a given time point is indicated on the top of the graph. The blue solid line represents our standard primer/probe set, whereas the red solid line corresponds to our alternative set. The dashed lines symbolize the sensitivities of the RQ-PCR at the given day of sampling calculated as described previously.2 The normal background level of WT1 is illustrated with a green dashed line. The mutational status of WT1 in respectively genomic DNA and cDNA is indicated on the bottom of the graph. These results are generated from the fragment analysis in which fractions of the 8 bp deletion in WT1 are calculated. NA: samples are not analyzed owing to lack of material, no WT1: no PCR product was amplified owing to the low level of the fusion transcript in the remission samples. (b) Fragment analysis of genomic DNA showing the shift from wild-type (wt) DNA at diagnosis to the 8 bp deletion (del) at second relapse. Red spikes represent a molecular weight marker. (c) Silver-stained gel and Western blot. Lane1: diagnostic sample, 2: sample at second relapse, 3: negative control (a sample from an AML patient in remission) and 4: positive control (cell line K562) (lanes not relevant for this study have been omitted). Leukemia
Letters to the Editor
2053 specific WT1 primer and probe sequences: Forward primer (exon 6): 50 -AGAATACACACGCACGGTGTCT-30 , reverse primer (exon 7): 50 -GATGCCGACCGTACAAGAGTC-30 , probe (exon 6/7): FAM-50 -CTCCAGGCACACGTCGCACATCCTG-30 TAMRA.2 An alternative primer/probe set was designed using human LNA (locked nucleic acid) probe no. 4 (exon 7) 50 -GCAGGAAG-30 (Exiqon, Vedbaek, Denmark) in combination with a forward primer (exon 7) 50 -AGCTGTCCCACTTACA GATGC-30 and a reverse primer (exon 8) 50 -CACACTGGTA TGGTTTCTCACC-30 . For PCR and sequencing of complementary DNA (cDNA) the primer sequences were: 50 -ATGGACAG AAGGGCAGAGCA-30 (exon 5) and 50 -AAAACCTTCGTTCACA GTCCTTGA-30 (exon 8). Primer sequences for fragment analysis of cDNA were: 50 -ATGGACAGAAGGGCAGAGCA-30 (exon 5) and 50 -GGGAGAACTTTCGCTGACAAG-30 (exon 9) and for genomic DNA: 50 -ATGGGGATCTGGAGTGTGAAT-30 and 50 -ACAGCGGGCACACTTACC-30 . Forward primers were 50 -labeled with 6-FAM. The PCR conditions were: 35 cycles of 30 s at 951C, 60 s at 581C, 60 s at 721C, initiated by 15 min of denaturation at 951C, and completed by 5 min of extension at 721C. Quantification of the amount of mutated cDNA and DNA was done using the formula: Sm/(Sw þ Sm), where Sm is the spike intensity representing the PCR product harboring the mutation and Sw is the spike intensity representing the wild-type PCR product. 20 ml of MNC stored in messenger RNA (mRNA) lysis buffer (Roche Diagnostics, Basel, CH) was loaded directly on a sodium dodecylsulfate polyacrylamide gel. Protein was detected by Western blot using a monoclonal antibody (ab3236, Abcam, Cambridge, UK) corresponding to amino acids 1–318 of WT1 and an alkaline phosphate-conjugated secondary antibody (DAKO A/S, Copenhagen, DK). A silver-stained gel was performed by loading 2 ml of MNC in mRNA lysis buffer (Roche Diagnostics) in each slot. The patient was sampled for PB and BM at multiple time points during the course of disease. As demonstrated earlier,2 RQ-PCR data from PB and BM were similar. The WT1 gene expression measured by our standard primer/probe set2 in PB mirrored the clinical course up to and including the first relapse, comprising an increase of expression heralding the clinical relapse (Figure 1a). Surprisingly, the expression of WT1 was lost up to and at the second relapse in clear contradiction to immunophenotype and morphology. This prompted us to construct a differently located primer/probe set, and employing this, the WT1 expression was found to be consistent with clinical data displaying gradual increases up to the time of second relapse (Figure 1a). We hypothesized that the reason for this discrepancy could be a mutation in WT1 located within the annealing sites of our standard primer/probe set, which spans exon 6 and 7 in contrast to our alternative set, which cover exon 7 and 8 (Figure 2a). Indeed, an 8 bp deletion was demonstrated by PCR and sequencing (Figure 2b and c). To study whether the deletion was present at diagnosis we performed fragment analysis of the deleted region using both genomic DNA and cDNA as targets. Interestingly, we found minor spikes (3–5%) representing the deletion already at the time of diagnosis (Figure 1b). Not shown in the Figure, which only depicts measurement of PB is that at first relapse (March 2004) BM revealed 5% of the truncated clone. By computer analysis the 8 bp deletion resulted in a truncated protein terminating before the zinc-finger domain. Corroborating the molecular data, Western blot demonstrated WT1 expressed at diagnosis but not at second relapse (Figure 1c). Mutations within WT1 in cancer have been known for some time, varying from disease to disease. They are rare in AML
a
exon 6
exon 7
exon 8
Standard primer/probeset LNA primer/probeset
b Diagnosis
c Relapse 8 bp deletion Figure 2 (a) Illustration of part of WT1 showing the location of primer/probe sets used for RQ-PCR in the present study. Gray arrows/ bar indicates our standard primer/probe set whereas green arrows/bar represents our alternative primer/probe sets. (b) The sequence of the wild-type WT1 obtained from our pediatric patient at the time of diagnosis. (c) The sequence of the mutated WT1 at the time of second relapse. The dashed line illustrates the location of the 8 bp deletion.
though there have been reports of alterations in exon 7 leading to a truncated protein terminating before the zinc-finger region.6,7 Importantly, none of these mutations have to our knowledge been investigated more than once throughout the disease course. A single patient harboring a point mutation in exon 9 was investigated both at diagnosis and at relapse showing the mutation to occur only at relapse.7 Although the observation in this patient might cast doubt on WT1 as an MRD target, we would like to stress that this falsenegative result is a rare event, as WT1 expression remained high at subsequent relapses in 24 other AML patients we have tested longitudinally. The present case story demonstrates for the first time in AML that a minor subclone, characterized by a mutation in WT1 and present already at diagnosis, develops to be the dominant clone after repeated courses of chemotherapy including SCT. A plausible explanation for these observations may be that the mutant clone is less sensitive to the therapy given in second treatment course than the primary clone. The mechanism for this resistance development and its possible direct or indirect relation to WT1 is not clear. Indeed, the contribution of WT1 overexpression in malignant blood diseases is not resolved, even though a very recent report has shown that the gene can be pivotal in leukemogenesis in WT1 transgenic mice, in which AML/ETO-transduced BM cells developed rapidly into AML in contrast to the situation in non-transgenic animals, which only developed anemia and dysplasia.8 Even though it cannot be formally excluded, we feel that the t(6;12) aberration could only contribute marginally to the course of disease, as it first appeared at the final relapse. Nevertheless, our data are supported by previous reports which have speculated that mutations in WT1 could lead to progression of the leukemia by conferring drug resistance.6,7 These findings have wide-ranging implications for our knowledge of the biology of AML and for its clinical follow-up.
Acknowledgements This work was supported by grants from the Danish Cancer Society, the Danish Research Agency, and the Karen Elise Jensen Foundation. We thank Mette Østergaard for helpful suggestions, Gitte Biller for excellent technical assistance, and David Grimwade, Guy’s King’s and St Thomas’ School of Medicine, for careful review of this paper. Leukemia
Letters to the Editor
2054
CG Nyvold1, J Stentoft1, K Brændstrup1, D Melsvik1, SK Moestrup2, C Juhl-Christensen1, H Hasle3 and P Hokland1 1 Department of Hematology, Aarhus University Hospital, Aarhus, Denmark; 2 Department of Medical Biochemistry, Aarhus University, Aarhus, Denmark and 3 Department of Pediatrics, Skejby Hospital, Aarhus, Denmark E-mail:
[email protected] References 1 Gabert J, Beillard E, van der Velden VH, Bi W, Grimwade D, Pallisgaard N et al. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia – a Europe Against Cancer program. Leukemia 2003; 17: 2318–2357. 2 Ostergaard M, Olesen LH, Hasle H, Kjeldsen E, Hokland P. WT1 gene expression: an excellent tool for monitoring minimal residual disease in 70% of acute myeloid leukaemia patients – results from a single-centre study. Br J Haematol 2004; 125: 590–600.
3 Cilloni D, Gottardi E, De Micheli D, Serra A, Volpe G, Messa F et al. Quantitative assessment of WT1 expression by real time quantitative PCR may be a useful tool for monitoring minimal residual disease in acute leukemia patients. Leukemia 2002; 16: 2115–2121. 4 Stentoft J, Hokland P, Ostergaard M, Hasle H, Nyvold CG. Minimal residual core binding factor AMLs by real time quantitative PCR – initial response to chemotherapy predicts event free survival and close monitoring of peripheral blood unravels the kinetics of relapse. Leuk Res 2006; 30: 389–395. 5 Pallisgaard N, Hokland P, Riishoj DC, Pedersen B, Jorgensen P. Multiplex reverse transcription-polymerase chain reaction for simultaneous screening of 29 translocations and chromosomal aberrations in acute leukemia. Blood 1998; 92: 574–588. 6 King-Underwood L, Pritchard-Jones K. Wilms’ tumor (WT1) gene mutations occur mainly in acute myeloid leukemia and may confer drug resistance. Blood 1998; 91: 2961–2968. 7 Miyagawa K, Hayashi Y, Fukuda T, Mitani K, Hirai H, Kamiya K. Mutations of the WT1 gene in childhood nonlymphoid hematological malignancies. Genes Chromosomes Cancer 1999; 25: 176–183. 8 Nishida S, Hosen N, Shirakata T, Kanato K, Yanagihara M, Nakatsuka S et al. AML1-ETO rapidly induces acute myeloblastic leukemia in cooperation with Wilms’ tumor gene, WT1. Blood 2006; 107: 3303–3312.
Human neutrophil elastase is not a target for therapy in chronic myeloid leukaemia
Leukemia (2006) 20, 2054–2055. doi:10.1038/sj.leu.2404411; published online 28 September 2006
The primary granule protein (PGP), human neutrophil elastase (HNE), is often associated with inflammatory-type reactions but also has a role in granulopoiesis.1 PGP proteases, including HNE, normally provide negative feedback of granulopoiesis by controlling the availability of granulocyte colony-stimulating factor (G-CSF) proportionate to the relative abundance of mature granulocytes that are greatly elevated in number as a clinical feature of chronic myeloid leukaemia (CML). HNE added to normal CD34 þ stem cells in culture blocks granulocyte macrophage (GM) colony formation and induces apoptosis.1,2 The G-CSF receptor is not digested; instead HNE targets G-CSF protein exerting a negative feedback on cytokine activity.1 High plasma levels of HNE2 and low G-CSF3 have been reported in CML. Moreover, it is known that Ph þ cells can invoke autocrine G-CSF cytokine production.4 Taken together, these data suggest that HNE may contribute to the CML phenotype of Ph þ clonal dominance as Ph normal cells die without growth factors. Indeed, it has been shown that CML CD34 þ cells are insensitive to exogenously added HNE, whereas normal CD34 þ that are not cytokine-independent undergo apoptosis when cultured in vitro in serum-free medium supplemented with stem cell factor, G- and GM-CSF to which HNE is added.2 Elafin is a naturally occurring inhibitor of HNE, and can restore normal proliferation when added to mixed CML/normal cell cultures (El-Ouriaghli et al. Blood 2003; 102: 230a; abstract). We sought to determine whether HNE levels are elevated in CML bone marrow (BM) and peripheral blood (PB) samples taken pre- and post-treatment with the tyrosine kinase inhibitor, imatinib mesylate (IM) and to determine any correlation with disease stage. Consequently, it may be possible to target HNE with an inhibitor modelled on the naturally occurring anti-protease, elafin or elastase-specific inhibitor.5 Leukemia
PB or BM samples were collected into ethylenediaminetetraacetic acid tubes from CML and non-CML patients; PB was also drawn from healthy volunteers, all with informed consent. Non-CML and healthy individuals had white blood cell counts (WBC) in the normal range (o10 109/l). Samples from CML patients were collected at various time intervals before and during the course of their treatment with IM. The plasma supernatant from all samples was separated from cellular components by centrifugation and stored frozen (201C) until required. A total of 105 samples were analysed (Table 1). HNE protein was measured using a highly sensitive, specific enzyme-linked immunosorbent assay (ELISA) kit (BioVendor, The Netherlands) for the quantitative measurement of the complex of HNE with alpha-1 proteinase inhibitor in plasma. Serum was not suitable as during clotting, HNE can be released in vitro. Measurement of plasma HNE by ELISA showed that at diagnosis (chronic phase (CP) CML, n ¼ 16), PB HNE was significantly elevated (1934795 ng/ml, mean7s.e.m.) with respect to normal controls (6473 ng/ml, n ¼ 9; P ¼ 0.002) Table 1
Sample details
Sample details Normal ‘non-CML’ plasma Non-CML marrow CP diagnosis plasma CP plasma pre-IM CP plasma post-IM CP marrow pre-IM CP marrow post-IM AP/BC plasma pre-IM AP/BC plasma post-IM AP/BC marrow pre-IM AP/BC marrow post-IM Total
Number of samples 10 3 16 29 13 6 12 5 3 3 5 105
Abbreviations: AP, accelerated phase; BC, blast crisis; CML, chronic myeloid leukaemia; CP, chronic phase; IM, imatinib mesylate.
