Cover
Cover
ORIGINAL ARTICLE
Abstract
Objectives: Wilms tumour gene 1 (WT1) is overexpressed in leucocytes of most acute myeloid leukaemia
(AML) patients. However, the clinical relevance of WT1 gene expression as minimal residual disease (MRD)
marker in AML has been questioned. Methods: We determined the expression of WT1 gene in bone mar-
row (BM) mononuclear cells of 100 AML patients at diagnosis and compared it with other MRD markers
during follow up in 16 patients using quantitative reverse transcription-polymerase chain reaction. Results:
The median WT1 gene expression was 9.7% of K562 cell line WT1 expression (lower quartile 1.5%, upper
quartile 29.9%, n = 100) at diagnosis and, 0.053% (lower quartile 0.022%, upper quartile 0.125%, n = 87)
in molecular or immunophenotypic remission. Median WT1 expression in control BM was 0.029% (lower
quartile 0.013%, upper quartile 0.061%, n = 22). The upper 99% percentile of remission samples was
0.3%, which was regarded as the cut-off of increased WT1 gene expression in AML and was exceeded in
87% of all AML patients at diagnosis. WT1 and the other MRD markers showed only minor differences in
profiles during follow-up. WT1 expression at diagnosis with median value 9.7% as the cut-off level or as a
continuous variable had no prognostic significance for 2-yr survival. Conclusions: The sensitivity of WT1 as
a MRD marker was low due to the relatively high background WT1 gene expression in BM cells at remis-
sion and in subjects without haematological malignancies. Therefore, WT1 gene expression analysis would
be beneficial only in those patients who do not have a more specific and sensitive MRD marker.
Key words Wilms tumour gene 1; acute myeloid leukaemia; minimal residual disease
Correspondence Mauri Hämäläinen, PhD, Department of Clinical Chemistry, Turku University Central Hospital, FI-20520 Turku, Fin-
land. Tel: +358 2 3131899; Fax: +358 2 3133924; e-mail: mauham@utu.fi
Wilms tumour gene 1 (WT1) overexpression has been involved in the embryonic development of kidneys,
offered as a panleukaemic molecular marker because it spleen and sex organs. It is also expressed in early stages
has been demonstrated in many leukaemia types includ- of leukocyte differentiation (14, 15). Therefore, undiffer-
ing acute lymphatic and myeloid leukaemias (ALL and entiated leucocytes increase the background WT1 gene
AML), chronic myeloid leukaemia in blast crisis as well expression reducing its usefulness as a MRD marker.
as in myelodysplastic syndromes (1–8). The development The level of WT1 gene expression has also been offered
of quantitative reverse transcription-polymerase chain as a prognostic marker for survival in AML. Inoue et al.
reaction (RT-PCR) has given a tool for detection of low (3) reported significant correlation between low WT1
levels of WT1 mRNA as a marker of minimal residual expression and higher complete remission rate and longer
disease (MRD) in AML patients (9–13). WT1 gene is survival. Later, both supporting and conflicting data of
the prognostic significance of WT1 expression at diagno- RNA extraction and reverse transcription
sis have been presented (16–23). In the present study we Mononuclear leucocytes were isolated from BM and
determined the upper confidence limit for WT1 gene blood samples by Ficoll-PaqueTM (Amersham Bio-
expression in bone marrow (BM) mononuclear cells and sciences, Uppsala, Sweden) gradient centrifugation. The
evaluated the sensitivity of WT1 gene expression as a cells were frozen in liquid nitrogen in 107 cells aliquots
molecular marker for MRD detection in AML patients. and stored in )80C until used for analyses. Total RNA
We also studied whether the level of WT1 gene expression was extracted form the cells with RNeasy kit and DNA
at diagnosis predicts 2-yr survival in AML patients. were extracted with QiaAmp kit of Qiagen Inc (Valencia,
CA, USA). cDNA synthesis from 1 lg of RNA was car-
Material and methods ried out in 20 lL of buffer with 1 mm nucleotides, 5 mm
MgCl2, 25 lm random hexamers, 20 U RiboLockTM
Subjects RNase inhibitor (Fermentas Inc, Hanover, MD, USA)
One hundred AML patients (57 females, 43 males) and 100 U M-MLV reverse transcriptase (Promega
treated in Turku University Central Hospital for AML Corp., Madison, WI, USA).
participated this study. The numbers of patients in differ-
ent age groups were as follows: 1–16 yr, 8; 17–39 yr, 23; Minimal residual disease analyses
40–65 yr, 45; 66–81, 24. The subtypes of AML according
Multiparameter flow cytometry of BM total leucocytes
to the French-American-British classification were as
was carried out with Becton Dickinson FacsCaliburTM
follows: M0 in 6, M1 in 19, M2 in 18, M3 in 12, M4 in
(4-colour analysis) or FACSCantoTM (6-colour analysis)
12, M5 in 14, M7 in 3, respectively. In addition, 10
flow cytometer (San Jose, CA, USA) based on cross line-
patients with secondary AML, four patients AML with
age antigen expression, antigen overexpression or asyn-
multilineage dysplasia (according to WHO criteria), one
chronous antigen expression depending on the immune
biphenotypic and one triphenotypic AML cases were
phenotype at diagnosis. Sensitivity of the assay was gen-
included in the study. The karyotype of BM cells was
erally around 0.1%.
abnormal in 63 patients and normal in 37 patients.
Fusion gene analyses t(9;22) BCR-ABL, t(15;17)
In addition, 22 BM samples were taken from patients
PML-RARa, t(8;21) AML1-ETO and Inv(16)
suspected to have a haematological disease, but diag-
CBFB-MYH11 were analysed using RT-qPCR accord-
nosed later to suffer either of an infection, iron deficiency
ing to Europe Against Cancer Program protocol (27,
anaemia, mild hypoplacia or trombocytopenia. Blood
28). Nucleophosmin (NPM) mutation analysis was
samples were taken from 12 healthy controls.
carried out by multiplying the gene sequence with
The paediatric patients (1–16 yr) were treated accord-
PCR as described by Falini et al. (29) and sequencing
ing to the common Nordic AML protocol (NOPHO)
the PCR product with Abi Prism 310 genetic ana-
(24). There are two different protocols, NOPHO AML
lyzer (Applied Biosystems, Forster City, CA, USA).
1994 and 2004 with minor modifications. In NOPHO
FMS-like tyrosine kinase 3 (FLT3) point mutations
2004 the intensity of the treatments has increased and
were analysed according to Abu-Duhier et al. (30)
stratification is better defined. Induction consists of
and tandem mutations as described by Kottaridis
6-tioguanin, cytarabin, VP-16, idarubicin and it metho-
et al. (31).
trexate, second part of induction includes cytarabin and
Wilms tumour gene 1 gene expression was analysed
mitoxatrone and it methotrexate. Consolidation protocol
with RT-qPCR method of Ogawa et al. (12). ß-Glucu-
depends on the risk category. The patients aged 17–65 yrs
ronidase gene expression was used for normalizing
were treated according to the protocols of the Finnish
WT1 result with the sample. WT1 gene expression of
Leukaemia Group (25). The patients older than 65 yrs
K562 cells was kept as 100% standard. Accordingly to
were treated with less intensive treatments which typically
the study of Ogawa et al. (12) we found that the stan-
included low-dose cytarabine for 5–7 d and idarubicine
dard curve with K562 cDNA was linear in dilutions
+ ⁄ ) thioguanine. Few patients received per oral induc-
down to 10)5. Determinations were made in duplicates
tion treatment with etoposide + thioguanine + idarubi-
in ABI Prism 5700 Sequence Detector System (Applied
cine (26). After induction the patients received one or two
Biosystems). If the threshold cycle number (Ct) devi-
additional chemotherapy courses with cytarabine plus
ated more than one cycle between the duplicates
idarubicine or mitoxantrone + ⁄ ) thioguanine. Thereafter
the analysis was repeated. WT1 and a molecular or
few patients got also per oral maintenance therapy with
flow cytometric MRD marker were determined simul-
daily mercatopurine + ⁄ )weekly methotrexate. Eighty
taneously from 16 AML patients in the course of
per cent of all patients achieved cytological complete
disease.
remission during induction and consolidation treatments.
Figure 1 Wilms tumour gene 1 (WT1) expression of acute myeloid leukaemia patients at diagnosis. Abnormal karyotype group does not include
the patients with translocations t(15;17), t(8,21) or Inv(16). Control bone marrow samples were obtained from haematological patients with non-
malignant disease. Control peripheral blood leucocytes samples were obtained from healthy subjects. Bone marrow WT1 was significantly higher
in patients with t(15;17) than in the normal karyotype, abnormal karyotype or t(8,21) groups (P = 0.02, P = 0.003 or P = 0.016, respectively,
Mann–Whitney’s U-test).
0.4
WT1 (per cent of K562 expression)
0.2
0.1
0
0 20 40 60 80 100
AML remission samples
rapidly at the onset of chemotherapy whereas the specific cases where a more sensitive MRD marker is not
molecular marker shows the presence of residual leukae- available.
mic cells for longer time periods. Thus, WT1 may give
too optimistic view of MRD status. On the other hand,
References
rapid decrease of WT1 gene expression would be
regarded as an indication of effective therapy. The inhib- 1. Miwa H, Beran M, Sounders GF. Expression of the
itory effect of chemotherapy may also be the reason for Wilms’ tumor gene (WT1) in human leukemias. Leukemia
that WT1 did not respond to a rise in leukaemia cells 1992;6:405–9.
after the initial decline shown in the lowest diagram of 2. Miyagi T, Ahuja H, Kubota T, Kubonishi I, Koeffler HP,
(Fig. 4). The proposed relapse prediction by elevated Miyoshi I. Expression of the candidate Wilm’s tumor gene
WT1 expression needs more frequent sampling than is WT1, in human leukemia cells. Leukemia 1993;7:970–7.
3. Inoue K, Sugiyama H, Ogawa H, Nakagawa M,
the case in most treatment protocols and is only possible
Yamagami T, Miwa H, et al. WT1 as a new prognostic
with blood samples. However, in every case where a rise
factor and a new marker for the detection of minimal
in WT1 was found in initially successfully treated patient
residual disease in acute leukemia. Blood 1994;84:3071–9.
it was connected to a relapse. We conclude that, if a spe-
4. Tamaki H, Ogawa H, Ohyashiki K, et al. The Wilms’
cific molecular marker is available, it cannot be replaced tumor gene WT1 is a good marker for diagnosis of disease
by the WT1 gene expression analysis as a follow-up mar- progression of myelodysplastic syndromes. Leukemia
ker for MRD analysis in acute myeloid leukaemia. How- 1999;13:393–9.
ever, WT1 may offer additional information of the effect 5. Cilloni D, Gottardi G, Messa F, et al. Significant
of therapy thus supporting the specific MRD findings. If correlation between the degree of WT1 expression and the
a specific molecular marker is not available, WT1 expres- international prognostic scoring system score in patients
sion might be beneficial if the lower sensitivity and the with myelodysplastic syndromes. J Clin Oncol
possible lower values than other molecular markers due 2003;21:988–95.
to the therapy are kept in mind. 6. Rosenfeld C, Cheever MA, Gaiger A. WT1 in acute
As the early observations of WTI as a diagnostic tool, leukemia, chronic myelogenous leukemia and myelo-
a high level of WT1 expression at diagnosis has been dysplastic syndrome: therapeutic potential of WT1
regarded as a marker of poor prognosis for survival (3, targeted therapy. Leukemia 2003;17:1301–12.
12, 16, 21). However, contradictory views have been pre- 7. Cilloni D, Saglio G. WT1 as universal marker for minimal
sented (17, 19). In our study no significant difference residual disease detection and quantification in myeloid
existed in the survival between diagnostic high and low leukemias and in myelodysplastic syndrome. Acta
WT1 groups. WT1 as continuous variable had not a pre- Haematol 2004;112:79–84.
dictive value, either. The early studies (3, 16, 17) were 8. Keilholz U, Menssen HD, Gaiger A, Menke A, Oji Y,
based on semiquantitative PCR methods, which prevent Oka Y, Scheibenbogen C, Stauss H, Thiel E, Sugiyama H.
conclusive decisions. In two resent studies with quantita- Wilms’ tumour gene 1 (WT1) in human neoplasia.
Leukemia 2005;19:1318–23.
tive real-time RT-PCR technology, the cut-off value of
9. Kreuzer K-A, Saborowski A, Lupberger J, Applet C, Na
WT1 in outcome studies was stated around median as in
I-K, Le Coutre P, Schmidt A. Fluorescent 5’-exonuclease
our study but the discrepancy still persists (19, 21). In
assay for the absolute quantification of Wilms’ tumour
the studies of Lapillonne et al. (23) and Weisser et al.
gene (WT1) mRNA: implications for monitoring human
(22) WT1 expression at diagnosis had no predictive value leukaemias. Br J Haematol 2001;114:313–8.
whereas a high WT1 after induction therapy predicted 10. Siehl JM, Thiel E, Leben R, Reinwald M, Knauf W,
poor outcome. Our view is that high WT1 at presenta- Menssen HD. Quantitative real-time RT-PCR detects
tion has no prognostic significance whereas a rise in elevated Wilms tumor gene (WT1) expression in auto-
WT1 expression during treatment is a predictor for logous blood stem cell preparations (PBSCs) from acute
relapse. myeloid leukemia (AML) patients indicating contamina-
In this study our aim was to evaluate the usefulness of tion with leukemic blasts. Bone Marrow Transplant
WT1 gene expression in BM cells as a universal MRD 2002;29:379–81.
marker in 100 AML patients. Although WT1 kinetics 11. Cilloni D, Gottardi E, De Micheli D, et al. Quantitative
generally was in agreement with other MRD markers, assessment of WT1 expression by real time quantitative
we found that the baseline WT1 expression in remission PCR may be a useful tool for monitoring minimal residual
was so high that the sensitivity of MRD analysis based disease in acute leukaemia patients. Leukemia
on WT1 gene expression remained too low. We were not 2002;16:2115–21.
able to demonstrate a prognostic significance of high 12. Ogawa H, Tamaki H, Ikegame K, et al. The usefulness of
WT1 at presentation. Therefore, we suggest that BM monitoring WT1 gene transcripts for the prediction and
WT1 gene expression analysis is informative only in management of relapse following allogeneic stem cell
transplantation in acute type leukemia. Blood Oncology (NOPHO). Improved outcome after relapse in
2003;101:1698–704. children with acute myeloid leukaemia. Br J Haematol
13. Østergaard M, Olesen LH, Hasle H, Kjeldsen E, Hokland 2006;136:229–36.
P. WT1 gene expression: an excellent tool for monitoring 25. Koistinen P, Räty R, Itälä M, et al. Long-term outcome
minimal residual disease in 70% of acute myeloid of intensive chemotherapy for adults with de novo myeloid
leukaemia patients – results from a single-centre study. Br leukaemia (AML): the nationwide AML-92 study by the
J Haematol 2004;125:590–600. Finnish Leukaemia Group. Eur J Haematol 2007;78:477–
14. Baird PN, Simmon PJ. Expression of the Wilms’ tumor 86.
gene (WT1) in normal hematopoiesis. Exp Hematol 26. Ruutu T, Koivunen E, Nousiainen T, et al. Oral
1997;25:312–20. treatment of acute myeloid leukaemia with etoposide,
15. Pritchard-Jones K, King-Underwood L. The Wilms thioguanine, and idarubicin (ETI) in elderly patients: a
tumour gene WT1 in leukaemia. Leukemia Lymphoma prospective randomised comparison with intravenous
1997;27:207–20. cytarabine, idarubicin, and thioguanine in the second and
16. Bergmann L, Miething C, Maurer U, Brieger J, Karakas third treatment cycle. Eur J Haematol 2004;72:38–44.
T, Weidmann E, Hoelzer D. High levels of Wilms’ tumor 27. Gabert J, Beillard E, van der Velden VHJ, et al.
gene (wt1) mRNA in acute myeloid leukemias are Standardization and quality control studies of ‘real time’
associated with a worse long-term outcome. Blood quantitative reverse transcriptase polymerase chain
1997;90:1217–25. reaction of fusion gene transcripts for residual disease
17. Schmid D, Heinze G, Linnerth B, et al. Prognostic detection in leukemia–A Europe Against Cancer Program.
significance of WT1 gene expression at diagnosis in adult Leukemia 2003;17:2318–57.
de novo acute myeloid leukemia. Leukemia 1997;11:639–43. 28. Beillard E, Pallisgaard N, van der Velden VHJ, et al.
18. Garg M, Moore H, Tobal K, Liu Yin JA. Prognostic Evaluation of candidate control genes for diagnosis and
significance of quantitative analysis of WT1 gene tran- residual disease detection in leukemic patients using ‘real
scripts by competitive reverse transcription polymerase time’ quantitative reverse-transcriptase polymerase chain
chain reaction in acute leukaemia. Br J Haematol reaction (RQ_PCR) – a Europe against cancer program.
2003;123:49–59. Leukemia 2003;17:2474–86.
19. Yanada M, Terakura T, Yokozawa T, et al. Multiplex 29. Falini B, Mecucci C, Tiacci E, et al. Cytoplasmic
real-time RT-PCR for prospective evaluation of WT1 and nucleophosmin in acute myelogenous leukemia with a
fusion gene transcripts in newly diagnosed de novo acute normal karyotype. N Engl J Med 2005;352:254–66.
myeloid leukemia. Leukemia Lymphoma 2004;45:1803–8. 30. Abu-Duhier FM, Goodev AC, Wilson GA, Care RS,
20. Trka J, Kalinova M, Hrusak O, et al. Real-time quantita- Peake IR, Reilly JT. Genomic structure of human FLT3:
tive PCR detection of WT1 gene expression in children implications for mutational analysis. Br J Haematol
with AML: prognostic significance, correlation with 2001;113:1076–7.
disease status and residual disease detection by flow 31. Kottaridis PD, Gale RE, Frew ME, et al. The presence of
cytometry. Leukemia 2002;16:1381–9. FLT3 internal tandem duplication in patients with acute
21. Barragán E, Cervera J, Bolufer P, Ballester S, Martin G, myeloid leukemia (AML) adds important prognostic
Fernándes P, Collado R, Sayas MJ, Sanz MA. Prognostic information to cytogenetic risk group and response to the
implications of Wilms’ tumor gene (WT1) expression in first cycle of chemotherapy: analysis of 854 patients from
patients with de novo acute myeloid leukemia. Haemato- the United Kingdom Medical Research Council AML 10
logica 2004;89:926–33. and 12 trials. Blood 2001;98:1732–9.
22. Weisser M, Kern W, Rauhut S, Schoch C, Hiddemann W, 32. Grimwade D, Walker H, Oliver F, et al. The importance
Haferlach T, Schnittger S. Prognostic impact of RT-PCR- of diagnostic cytogenetics on outcome in AML: analysis
based quantification of WT1 gene expression during of 1,612 patients entered into the MRC AML 10 trial.
MRD monitoring of acute myeloid leukemia. Leukemia Blood 1998;92:2322–33.
2005;19:1416–23. 33. Grimwade D, Walker H, Harrison G, Oliver F, Chatters
23. Lapillonne H, Renneville A, Auvrignon A, et al. High S, Harrison CJ, Wheatley K, Burnett AK, Goldstone AH
WT1 expression after induction therapy predicts high risk on behalf of the Medical Research Council Adult Leukae-
of relapse and death in pediatric acute myeloid leukemia. mia Working Party. The predictive value of hierarchical
J Clin Oncol 2006;24:1507–15. cytogenetic classification in older adults with acute
24. Abrahamsson J, Clausen N, Gustafsson G, Hovi L, myeloid leukemia (AML): analysis of 1065 patients
Jonmundsson G, Zeller B, Forestier E, Heldrup J, Hasle entered into the United Kingdom Medical Council AML
H for the Nordic Society for Paediatric Haematology and 11 trial. Blood 2001;98:1312–20.