Quantitative assessment of WT1 expression by real time ... - Nature

Leukemia (2002) 16, 2115–2121  2002 Nature Publishing Group All rights reserved 0887-6924/02 $25.00 www.nature.com/leu

Quantitative assessment of WT1 expression by real time quantitative PCR may be a useful tool for monitoring minimal residual disease in acute leukemia patients D Cilloni1, E Gottardi1, D De Micheli1, A Serra1, G Volpe1, F Messa1, G Rege-Cambrin1, A Guerrasio1, M Divona2, F Lo Coco2 and G Saglio1 1

Division of Hematology and Internal Medicine, Dept of Clinical and Biological Sciences of the University of Turin, Turin, Italy; and 2Dept of Cellular Biotechnologies and Hematology, University La Sapienza, Rome, Italy

In order to verify if quantitative assessment of the WT1 transcript amount by the real time quantitative PCR (RQ-PCR) can be used as a marker for minimal residual disease detection, the WT1 transcript amount was determined in BM and PB samples of patients with myeloid and lymphoid acute leukemia, in normal controls, in regenerating bone marrow samples and in purified CD34-positive cells from normal subjects. In 10 patients bearing a fusion gene transcript suitable for minimal residual disease quantitative assessment, we performed a simultaneous analysis of the WT1 and of the fusion-gene transcript at sequential time intervals during follow-up. Sequential WT1 analysis was also performed in five AML patients lacking additional molecular markers. The data obtained show that normal and regenerating BM samples and purified CD34-positive cells consistently express minimal amounts of WT1 transcript and that this is extremely low and frequently undetectable in normal PB. By contrast, high levels of WT1 expression are present in the BM and PB samples of all acute leukemia (AL) cases at diagnosis. The WT1 levels during follow-up were found to follow the pattern of the other molecular markers (fusion gene transcripts) used for MRD monitoring and increased WT1 expression in the BM and/or PB during followup of AL patients was always found to be predictive of an impending hematological relapse. Leukemia (2002) 16, 2115–2121. doi:10.1038/sj.leu.2402675 Keywords: acute leukemia; minimal residual disease; real time quantitative PCR; WT1; fusion gene transcript

Introduction Recent studies have shown that assessment of minimal residual disease (MRD) may prove useful to better address treatment intensity in human leukemia.1–4However, so far applicability of this strategy has been limited to those leukemia subsets characterized by genetic markers amenable to sensitive detection by PCR. The latter include fusion genes derived from chromosome translocations, such as PML-RAR␣ in acute promyelocytic leukemia (APL) and BCR-ABL in chronic myelogenous leukemia (CML) or immunoglobulin and T cell receptor gene rearrangements in acute lymphoblastic leukemia (ALL).1–4 At present, more than 50% of the acute leukemias, and in particular most AMLs, lack known genetic lesions or clonality markers suitable for MRD monitoring. Thus, a number of studies have been performed in the attempt to identify cytogenetic and molecular abnormalities associated with leukemic transformation. In this setting the use of the Wilms’ tumor gene (WT1) marker would be of great interest. WT1 is a tumorsuppressor gene coding for a zinc-finger transcription factor located on chromosome 11p13, which was originally identified for its involvement in the pathogenesis of the Wilms’

Correspondence: G Saglio, Dept of Clinical and Biological Sciences of the University of Turin, San Luigi Hospital, Gonzole 10, 10043 Orbassano-Torino, Italy Received 8 February 2002; accepted 31 May 2002

tumor.5In normal bone marrow (BM) and peripheral blood (PB) WT1 expression is extremely low and often undetectable even by RT-PCR.6,7By contrast, high expression levels of WT1 have been reported in acute leukemia (AL).8–11 Although the significance of WT1 overexpression in acute leukemia remains unclear, the phenomenon might be exploited as a marker to establish the presence, the persistence or the reappearance of leukemic hematopoiesis, with the potential advantage of using a single marker suitable for the evaluation of most patients. At present, however, results obtained using qualitative RT-PCR have been controversial in terms of clinical significance, presumably due to the interference of the small amounts of WT1 transcript also expressed in normal hematopoiesis,12–14 and quantitative RT-PCR studies have provided only preliminary results in a limited number of cases.15,16 To try to study this problem, we therefore decided to set up a sensitive and reliable real time quantitative RTPCR (RQ-PCR) approach aimed at assessing the expression levels of WT1 and analyzed the PB and BM of normal subjects and of patients with cytogenetically defined groups of AML and ALL. In order to verify if this method could be conveniently applied to MRD monitoring, we evaluated sequential samples from 10 patients whose leukemias were characterized at diagnosis by a specific fusion gene transcript potentially suitable for quantititative assessment of MRD and from five other AML patients lacking additional molecular markers. Materials and methods

Patients and controls After informed consent, samples from 119 leukemia patients (71 AML and 48 ALL) were tested for WT1 expression. In all cases BM samples were studied. In 14 AML and 16 ALL cases we also tested the paired PB sample. In addition, we tested a control group of 53 normal samples (20 BM and 33 PB), and nine regenerating BM samples, obtained from AML patients in complete remission after recovery from chemotherapyinduced aplasia and samples of purified CD34-positive cells obtained from normal subjects (peripheral blood stem cell donors). All leukemia cases were classified according to FAB criteria, characterized at cytogenetic level and screened by RT-PCR for the presence of the most frequent fusion transcripts as previously described.17 To assess the significance of the WT1 expression as a marker for MRD detection, in 10 AML patients characterized by the presence of known fusion genes (six CBF␤-MYH11, four AML1-ETO) a simultaneous quantitative assessment of the WT1 transcript together with that of the present fusion transcript was performed by RQ-PCR. For this purpose, BM and/or PB sampling was performed at diagnosis and at sequential time intervals during follow-up. A similar sequential analysis

WT1 expression as MRD marker in acute leukemias D Cilloni et al

2116

for WT1 expression was also performed in five AML patients lacking other additional markers suitable for MRD monitoring.

CD34-positive cells enrichment CD34+ cells were enriched according to a magnetic cell sorting methodology (MACS; Miltenyi Biotec, Bergisch Gladbach, Germany).18 Briefly, mononuclear cells were labeled with a haptenized CD34 antibody (QBEND\10) that was magnetically labeled in a second step reaction with an anti-hapten antibody coupled to super paramagnetic microbeads. Labeled cells were then separated using a high gradient magnetic separator column placed in a strong magnetic field. The magnetically stained cells were retained in the column, and when this latter was removed from the magnetic field the CD34-positive cells were eluted. At the end of the procedure, the CD34positive cells represent more that 90% of the total as determined by flow cytometric analysis

Qualitative RT-PCR analysis Mononuclear cells were separated on a Ficoll–Hypaque density gradient. Total RNA was extracted using RNAwiz, following the manufacturer’s instructions (Ambion, Austin, TX, USA) The qualitative RT-PCR reactions were performed according to the protocols established by the European BIOMED 1 Concerted Action, as described.17For WT1 we used the following primers: Forward primer located on exon 7: 5⬘-GGCATCTGAGACCAGTGAGAA-3⬘; reverse primer located on exon 10: 5⬘GAGAGTCAGACTTGAAAGCAGT-3⬘.

Quantitative assessment of fusion-gene transcripts in acute leukemias The quantitative assessment of the CBF␤-MYH11 and AML1ETO transcripts was determined using the ABI PRISM 7700 Sequence Detector (PE Applied Biosystem, Foster City, CA, USA) according to the methods of the European SANCO Concerted Action of the Europe Against Cancer Program.19,20 The RQ-PCR primers and probes (designed with the Software Primer Express, Applied Biosystems, Foster City CA, USA) and the conditions at which the reactions were carried out can be obtained at the following web-site: http://194.214.97.12.scripts/defaultIE.asp. Following the indications of the European SANCO concerted action21 the fusion transcript values obtained were normalized with respect to the number of ABL transcripts and expressed as fusion-gene copy number every 104 copies of ABL (FG/ABL × 104). All experiments were carried out in triplicate with appropriate negative controls. Appropriate cell line serial dilutions were used as positive controls and for sensitivity tests. In dilution experiments, a sensitivity of at least 10−4 was observed. A dilution series of plasmids were employed for the generation of standard FG and ABL curves (see http://194.214.97.12/scripts/defaultIE.asp).

Quantitative assessment of WT1 transcripts The RT (reverse transcription) step was adapted from the BIOMED 1 protocol.17 Starting from 1 ␮g of total RNA, ranLeukemia

dom hexamers were used at a concentration of 25 ␮M and 100 U of the reverse transcriptase (Applera, Milan, Italy) were added to the reaction mixture, obtaining a significant enhancement of the assay sensitivity. RQ-PCR reactions and fluorescence measurements were made on the ABI PRISM 7700 Sequence Detection System (PE Applied Byosystems). The RQ-PCR primers and probe for WT1 were: forward primer (located on exon 7): 5⬘- CAGGCTGCAATAAGAGATATTTTAAGCT-3⬘; reverse primer (located on exon 8): 5⬘GAAGTCACACTGGTATGGTTTCTCA-3⬘; TaqMan probe (located on exon 7): 5⬘-CTTACAGATGCACAGCAGGAAGCACACTG-3⬘. The fluorescent probe was labeled with 6-carboxy-fluorescein phosphoramide (FAM) as reporter dye at the 5⬘-end and with the quencher dye carboxy-tetramethyl-rhodamine (TAMRA) at the 3⬘-terminus. For the PCR reaction, 5 ␮l of cDNA was added to 20 ␮l of PCR reaction mix containing 12.5 ␮l of TaqMan Universal PCR Master Mix (Applera) and 300 nM of each primer and 200 nM of the probe, respectively. Sterilized water was added to reach a final volume of 25 ␮l. PCR procedure started with a step of 2 min at 50°C to activate the UNG enzyme, followed by 10 min at 95° to inactivate the UGN enzyme and to provide a ‘hot start’ activating the AmpliTaq polymerase. Subsequently, 50 cycles of denaturation at 95°C for 15 s followed by annealing/extention at 62°C for 60 s were performed. All sample analysis was performed in triplicate and the results showing a discrepancy ⬎1 Ct in one of the wells were excluded and repeated. For low WT1 levels as those present in normal PB samples, even a single positive well out of three was considered positive and the WT1 copy number calculated dividing by three. For quantitative assessment of WT1, a calibration curve with a plasmid containing the WT1 target sequence was used (Amplimedical spa Bioline Division, Turin, Italy). Briefly, the standard curve was obtained by serial dilutions ranging from 106 to 10 molecules of a linearized plasmid obtained by cloning the target WT1 sequence into a PCR II TOPO vector (Invitrogen, Groningen, The Netherlands). The corresponding standard curve generated a mean slope of −3.45 and intercept of 39.8 ± 1 Ct. A threshold at 0.1 was selected in order to be in the exponential phase. The sensitivity of the RQ-PCR assay for WT1 was constantly able to detect 10 plasmid molecules and was further established by diluting the K562 cell line into normal lymphocytes obtained from the peripheral blood of a normal subject that scored repeatedly negative (Ct ⬎ 50) for WT1 expression. Under these experimental conditions, the RQ-PCR was able to detect a positive WT1 signal at a K562 cell line dilution of 1 × 10−6. The WT1 and fusion transcripts values obtained by RQ-PCR were normalized with respect to the number of ABL transcripts and expressed as WT1 or fusion-gene copy numbers every 104 copies of ABL. For the Philadelphia-positive ALLs (and for a limited number of samples representative of the other subgroups to show correspondence), normalization was performed using GUS (␤-glucuronidase gene) instead of ABL as control gene, as the RQ-PCR reaction for ABL also detects the BCR-ABL transcripts and gives misleading results. As shown in the cooperative work of the European SANCO Concerted Action of the Europe Against Cancer Program,21 the level of ABL and GUS are very similar and the expression of the two genes is highly correlated (P ⬍ 10−6), showing that both targets can represent very similar and good controls.


Table 1

WT1 quantitative assessment in control samples

Type of sample

Normal Regenerating BM CD34-positive cells

No. of samples tested

BM PBa BM CD34

20 33 9 7

WT1 copy number/104 ABL copies Median value

Range

78 4 74 50

3–180 1–22 18–112 40–61

a 14/33 peripheral blood samples scored completely negative at RQ-PCR and the median value is calculated only for the positive samples.

Statistical analysis The comparison between the WT1 values obtained in the different types of leukemia was performed with non-parametric tests (Mann–Whitney U test for unpaired samples). To compare the AML and the ALL groups we used the Student’s t-test. Results

WT1 expression in diagnostic leukemia samples The data obtained on the quantitative assessment of the WT1 transcript present in normal controls and in leukemia samples at diagnosis are summarized in Tables 1 and 2, respectively. The WT1 levels were extremely low in normal samples: 14 out of the 33 PB samples tested, scored negative and the Table 2

all cases t(15;17) t(8;21) inv(16) t(6;9) normal karyotype

ALL

2117

WT1 quantitative assessment in AML and ALL samples at diagnosis

Cytogenetic group

AML

median number of WT1 copies/104 ABL copies detected in the 19 positive was 4 (range 1–22). All normal BM were positive and the median number of WT1 copies was 78 (range 3– 180). Similarly, low copy numbers (median value 74; range 18–112) were found in nine regenerating BM samples obtained from AML patients in complete remission during recovery from chemotherapy-induced aplasia and in the seven samples of enriched CD34-positive cells obtained from normal peripheral blood stem cell donors (median value 50; range 40–61). These data are in agreement with the negative qualitative RT-PCR results generally found in normal and remission PB and BM samples, as in our hands single step qualitative RT-PCR analysis is less sensitive than RQ-PCR (10−3/10−4 for RT-PCR vs 10−6 for RQ-PCR as detection limit, as established by diluting K562 cells in normal lymphocytes). It must be underlined, however, that in our hands, the RTPCR results can be misleading as the presence of a weird primer–dimer artifact or of a small non-specific amplification band may sometimes simulate the presence of a faint band of WT1 amplification that is not confirmed by the RQ-PCR analysis. Conversely, all the samples from AML patients collected at diagnosis were positive for WT1 expression at qualitative RTPCR. The quantitative assessment carried out in 71 AML BM and 14 PB samples showed a median WT1 copy number of 27 669 (ranges: 1081–121 806) and 10 244 (range 758– 86 140), respectively. No statistical difference between cytogenetic groups and WT1 expression levels was found except for the t(15;17) APL cases, that expressed significantly higher amounts of WT1 with respect to all the other groups considered together (P = 0.009), whereas the four cases with a t(6;9) translocation showed low values of WT1, which, however, do not reach statistical significance with respect to the others.

all cases all cases t(4;11) t(9;22)a t(1;19) t(12;21) normal karyotype

Type of sample

No. of samples tested

WT1 copy number/104 ABL copies Median value

Range

BM PB BM PB BM PB BM PB BM BM PB

71 14 10 3 9 2 11 2 4 37 7

27 669 10 244 57 290 41 265 12 267 7 564 26 062 10 890 9158 34 720 8047

1081–12 1806 758–86 140 1081–121 806 758–86 140 7358–32 622 2687–12 441 18 632–58 510 4576–17 204 4073–21 107 2046–104 046 1513–24 974

BM PB BM PB BM PB BM PB BM BM PB

48 16 12 5 14 4 5 4 10 7 3

13 807 1388 14 612 1340 10 199 5262 490 259 2059 3546 1974

318–94 682 212–34 708 7237–81 725 905–34 708 5430–94 682 659–16548 318–2282 212–374 1213–3425 525–55 715 1573–2309

a The values of the t(9:22) cases have been calculated using GUS (␤-glucoronidase gene) instead of ABL as control gene as the RQ-PCR reaction for ABL also detects the BCR/ABL transcripts and gives false results.

Leukemia


2118

It is noteworthy that, at presentation, 30 out of 37 (81%) AML cases lacking a molecular marker suitable for MRD monitoring were showing levels of WT1 transcripts at least two logs higher than normal bone marrows and eight out of 37 (21%) were showing three logs higher, indicating that, at least in these cases, WT1 is expressed at a sufficiently high level to represent a reliable and sensitive marker for MRD detection. WT1 overexpression was also detected in all the ALL cases tested, but the median copy number was generally lower than in AML and the difference between the two groups was highly significant in the BM samples (P = 0.000028). A median of 1388 copies (range 212–34 708) in PB and 13807 copies (range 318–94 682) in BM was detected. The t(4;11) and the t(9;22) cases were the two groups expressing levels of WT1 significantly higher than all the other cases considered together (P = 0.0003 and P = 0.0015, respectively) whereas the t(1;19) translocation seems to define a subgroup which presents a low WT1 expression with respect to all the other groups (P = 0.0003).

Use of WT1 as a target for MRD detection To assess the significance of the WT1 expression as a marker for MRD detection in AML, the WT1 levels were determined during the follow-up of 10 AML patients characterized by the presence of known fusion-gene transcripts (six CBF␤-MYH11 and four AML1-ETO) and in five AML patients lacking additional molecular markers. For this purpose, BM and/or PB sampling was performed at diagnosis and during the subsequent follow-up. In all cases characterized by the presence of a fusion-gene transcript, the longitudinal pattern of the WT1 expression was always found to parallel that of the fusion gene. Three representative AML cases are illustrated in Figure 1. In the inv(16) AML subgroup, whereas the patient who remained in continuous complete remission (CCR) constantly showed WT1 values within the normal range (data not shown), the five patients who ultimately relapsed showed conversion to levels of WT1 above the normal range already during hematological remission in concomitance with an increase of the CBF␤-MYH11 transcript. In the case illustrated in Figure 1a, the WT1 values were found above the normal range in concomitance with a marked CBF␤-MYH11 increase in a BM sample taken 3 months before the hematological relapse, while the patient was still in hematological remission. In the case represented in Figure 1c, that reports the WT1 pattern in a patient subjected to alloBMT, the WT1 values were found above the normal range in a BM sample obtained already 1 year before the hematological relapse and in two subsequent samples. In this case a GVL effect hampering an overt hematological relapse for almost 1 year could probably have occurred, although we do not have direct evidence of this hypothesis. Similarly, in the t(8;21) group, three patients who were in CCR never showed levels of WT1 transcript above the normal range (data not shown), whereas in the patient who relapsed (Figure 1b), increased WT1 levels were detected in both BM and PB already 3 before hematological relapse, while the patient was still in hematological remission. Although the presented examples show a good degree of concordance of the curves representing WT1and FG levels, they are not completely parallel. As also noted in some cases not shown, it seems that WT1 expression has a clearance more rapid than that of the FG transcripts during induction of Leukemia

Figure 1 Parallelism between the WT1 levels determined by RQ-PCR and the levels of a fusion-gene transcript during the follow-up of three different AML patients. (a) Molecular follow-up of one inv(16) AML patient by RQ-PCRs for WT1 (blue line) and CBF␤-MYH11 levels (red line). The values obtained at definite intervals during therapy and subsequent followup (the dates are reported below the diagram) are expressed as transcript copy numbers every 104 ABL copies and are reported in absolute values in the diagram, which is in logarithmic scale. The WT1 values were found above the normal range (the dotted line corresponds to the upper limit found in normals) in concomitance with a marked CBF␤-MYH11 increase in a BM sample taken 3 months before the hematological relapse, while the patient was still in hematological remission. (b) Molecular follow-up of a t(8;21) AML patient by RQ-PCRs for WT1 (blue line) and AML1-ETO (red line). The diagram is in logarithmic scale. The WT1 values were found above the normal range both in a BM (the dotted line corresponds to the upper limit found in normals) and in a PB sample obtained 3 months before the hematological relapse, while the patient was still in hematological remission. (c) In a inv(16) AML patient subjected to alloBMT, the WT1 values were found increased above the normal range 1 year before relapse and in two subsequent samples. In this case, a GVL effect, controlling for approximately a year the appearance of an overt hematological relapse, could also be hypothesized, although no direct evidence of this hypothesis is available. Diagn., diagnosis; CR, complete hematological remission; Rel, relapse; ABMT, autologous bone marrow transplatation; BMT, allogeneic bone marrow transplantation.


remission, but it also seems that its elevation is more prompt than of the FG transcript in predicting relapse (see points 3 and 6 in Figure 1b). At the moment this cannot be easily explained due to our lack of knowledge on the kinetics of the WTI expression in AML. Finally, the sequential values found in five AML patients lacking additional molecular markers further suggest that an increase in WT1 level is predictive of hematological relapse. In the three patients who remain in first complete remission the WT1 level are constantly within the normal range both in BM and in PB (data not shown). By contrast, in the two patients who subsequently relapsed, the WT1 levels in the first patient rose above the normal range in BM and PB 4 months before the hematological relapse (Figure 2a) and in the second patient never reached values within the normal range (Figure 2b). Discussion WT1 is a tumor-suppressor gene coding for a zinc-finger transcription factor located on chromosome 11p13, which was originally identified for its involvement in the pathogenesis of the Wilms’ tumor.5 WT1 is expressed in a variety of tissues including ovary, testis and spleen.7 In normal peripheral blood and bone marrow WT1 expression was reported to be low and often undetectable even by qualitative RT-PCR.6,7 By contrast,

Figure 2 RQ-PCR analysis for WT1 transcript levels in the BM (dark blue profile) and PB (light blue profile) of two AML patients lacking additional molecular markers. (a) In this patient, the WT1 values were found above the normal range both in a BM (the dotted line corresponds to the upper limit found in normals) and in a PB (lighter dotted lines) sample obtained 4 months before the hematological relapse, while the patient was still in hematological remission. (b) In this patient, WT1 values below the normal range limits (dotted lines) were never found during the follow-up, although the patient was in apparent complete hematological remission.

WT1 expression has been found to be increased in most cases of acute leukemias,8–10 whereas little data are at present available concerning its expression in other types of hematological malignancies, such as chronic leukemias, non-Hodgkin lymphomas and myelomas.22 At present, the factors responsible for the increased expression of WT1 in acute leukemias, as well as its functional significance within the leukemogenesis process remain completely elusive. Using a real time quantitative PCR approach, in this paper we show that WT1 expression is detectable, although at low levels, in all normal bone marrows and that in normal PB the WT1 transcript level is extremely low and often even below the RQ-PCR detection limit (10−6). In the bone marrow and peripheral blood of all the myeloid and lymphoid acute leukemia patients tested, WT1 expression is always above the range observed in normal controls and, in most instances, very high. In the cases in which paired PB and BM samples are available, the WT1 expression levels appear generally higher in bone marrow than in peripheral blood. The WT1 expression levels, however, vary widely from case to case and tend to be higher in certain cytogenetic subgroups with respect to others. In particular, in the AML cases with the t(15;17) translocation and in the ALLs with the t(9;22) and t(4;11) translocations, very high values are observed, whereas in the ALLs with the t(1;19) translocation the mean values are generally only two to three times above the normal level. This suggests that the involvement and the role of the WT1 gene may be different in leukemias characterized by different molecular pathogenetic pathways. Moreover, regenerating BM samples obtained from patients who attained hematologic remission after chemotherapy, although rich in immature precursors, and normal samples highly enriched in CD34-positive cells express WT1 levels within the normal range. These data support the notion that increased levels of WT1 expression are indeed specific to leukemic blasts with respect to their normal counterpart (normal hematopoietic progenitors) and not a simple consequence of the differentiation degree. Since most myeloid and lymphoid acute leukemias show consistently increased WT1 expression levels, as already envisaged by Inoue et al23, this could represent a candidate marker suitable to discriminate between normal and leukemic hematopoiesis and useful to establish the presence, the persistence or the reappearance of leukemic blasts. In particular, our data show that 80% of the AML cases lacking other molecular markers suitable for MRD monitoring express at presentation WT1 values at least two logs higher than normal bone marrow samples and, therefore, WT1 is expressed at a sufficiently high level to represent a reliable and sensitive marker for MRD detection at least as a FISH analysis in the cases bearing a suitable cytogenetic marker. Moreover, in approximately 20% of these cases, the WT1 values are more than three logs higher than normal and the degree of sensitivity for MRD detection using WT1 as a marker becomes almost comparable to that achievable with the use of some fusion-gene transcripts expressed at low levels as PML/RAR␣. So far the data concerning the clinical significance of the detection of WT1 expression by RT-PCR for monitoring patients with acute leukemia during follow-up has been quite controversial in the literature.12–16 Different levels of sensitivity of the RT-PCR procedures used, different clinical settings in which the studies have been conducted and, as noted by us, the sporadic presence of small non-specific amplification bands mimicking the presence of WT1 expression, may account for the discrepancies observed and support the notion

2119

Leukemia


2120

that the RT-PCR results with the WT1 expression marker may be misleading for MRD detection. The data presented in this paper show that an accurate quantitative assessment of the WT1 transcript amount as that provided by the RQ-PCR method allows to distinguish clearly between normal and abnormal expression levels of WT1 and can overcome the problem represented by the minimal amount of WT1 transcript also expressed by the normal hemopoietic progenitors. In the cases in which a simultaneous quantitative assessment of the WT1 transcript and of a leukemia-specific fusion gene transcript was performed, we always observed a good parallelism between the behavior of the two markers. Indeed, minor discrepancies at low levels of expression of the two markers were observed. In particular, the decrease of the WT1 expression seems to be particularly rapid with respect to that of the FG transcript during induction of remission and its elevation more prompt than that of the FG transcript in predicting relapse. At the moment, our little knowledge on the kinetics of the WTI expression in AML does not allow us to reach conclusions on this point, but a possible explanation is that the FG transcript number could reflect the leukemia cell number, whereas WT1 copy levels could be related more to a functional state of the leukemia cells. If this point proves to be true, the WT1 sensitivity in predicting relapses could be even higher than expected. Therefore, even if at the moment the degree of sensitivity for MRD detection through analysis of WT1 expression does not seem as high as that achievable using FG markers or better than other methods (FISH, immunophenotyping), the results obtained show that an increase of WT1 expression above the normal levels can be of prognostic significance during the follow-up of patients with acute leukemia, being predictive of an imminent hematological relapse even some months before its occurrence. In this setting, the finding of extremely low and often undetectable WT1 levels in the peripheral blood of normal individuals and in leukemia patients in CCR, suggests that PB analysis could be even more sensitive than bone marrow in revealing impending relapses. Should this prove to be true in prospective studies, it would greatly simplify the approach to the molecular monitoring of AML patients.

4

5

6

7 8 9 10

11

12

13

14

Acknowledgements This work has been supported by grants from CNR (Progetto Finalizzato Biotecnologie), MURST-COFIN 2000, AIL (Associazione Italiana contro le Leucemie- Gimema-AML Project), AIRC (Associazione Italiana per la Ricerca sul Cancro) and Associazione Italiana Amici di Jose` Carreras.

15

16

References 1 Van Dongen JJ, Seriu T, Panzer-Grumayer ER, Biondi A, PongersWillemse MJ, Corral L, Stolz F, Schrappe M, Masera G, Kamps WA, Gadner H, van Wering ER, Ludwig WD, Basso G, de Bruijn MA, Cazzaniga G, Hettinger K, van der Does-van den Berg A, Hop WC, Riehm H, Bartram CR. Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood. Lancet 1998; 28: 1731–1738 2 Cave H, van der Werfften Bosch J, Suciu S, Guidal C, Waterkeeyn C, Otten J, Bakkus M, Thielemans K, Grandchamp B, Vilmer E. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer-Childhood Leukemia Cooperative Group. N Engl J Med 1998; 27: 591–598. 3 Lo Coco F, Diverio D, Falini B, Biondi A, Nervi C, Pelicci PG. Leukemia

17

18 19

Genetic diagnosis and molecular monitoring in the management of acute promyelocytic leukemia. Blood 1999; 94: 12–22. Diverio D, Rossi V, Avvisati G, De Santis S, Pistilli A, Pane F, Saglio G, Martinelli G, Petti MG, Santoro A, Pelicci PG, Mandelli F, Biondi A, Lo Coco F. Early detection of relapse by prospective reverse transcriptase-polymerase chain reaction analysis of the PML/RAR␣ fusion gene in patients with acute promyelocytic leukemia enrolled in the GIMEMA-AIEOP multicenter ‘AIDA’ trial. GIMEMA-AIEOP Multicenter ‘AIDA’ Trial. Blood 1998; 92: 784–789. Call KM, Glaser T, Ito CY,Bukler AJ, Pelletier J, Haber DA, Rose EA, Kral A, Yeger H, Lewis WH. Isolation and characterization of a zinc finger polypeptide gene at the human chromosome 11 Wilms’ tumor locus. Cell 1990; 60: 509–520. Park S, Schalling M, Bernard A, Maheswaran S, Shipley GC, Roberts D, Flecher J, Shipman R, Rheinwald J, Demetri G, Griffin J, Minden M, Housman DE, Harber DA. The Wilms tumour gene WT1 is expressed in murine mesoderm-derived tissues and mutated in a human mesothelioma. Nat Genet 1993; 4: 415–420. Baird PN, Simmons PJ. Expression of the Wilms’ tumor gene (WT1) in normal hemopoiesis. Exp Hematol 1997; 25: 312–320. Miwa H, Beran M, Saunders GF. Expression of the Wilms tumor gene (WT1) in human leukemias. Leukemia 1992; 6: 405–409. Miyagi T, Ahuja H, Kubota T, Kubonishi I, Koeffler HP, Miyoshi I. Expression of the candidate Wilm’s tumor gene, WT1, in human leukemia cells. Leukemia 1993; 7: 970–977. Inoue K, Ogawa H, Sonoda Y, Kimura T, Sakabe H, Oka Y, Miyake S, Tamaki H, Oji Y, Yamagami T, Tatekawa T, Soma T, Tishimoto T, Sugiyama H. Aberrant overexpression of the Wilms tumor gene (WT1) in human leukemia. Blood 1997; 89: 1405– 1412. Inoue K, Sugiyama H, Ogawa H, Nakagava M, Yamagami T, Miwa H, Kita K, Hiraoka A, Masaoka T, Nasu K, Kyo Y, Dohy H, Nakauchi H, Ishidate T, Akiyama T, Kishimoto T. WT1 as a new prognostic factor and a new marker for the detection of minimal residual disease in acute leukemia. Blood 1994; 84: 3071–3079. Elmaagacli AH, Beelen DW, Trenschel R, Schaefer UW. The detection of WT-1 transcripts is not associated with an increased leukemic relapse rate in patients with acute leukemia after allogeneic bone marrow or peripheral blood stem cell transplantation. Bone Marrow Transplant 2000; 25: 91–96. Schmid D, Heinze G, Linnerth B, Tisljar K, Kusec R, Geissler K, Sillaber C, Laczika K, Mitterbauer M, Zochbauer S, Mannhalter C, Haas OA, Lechner K, Jager U, Gaiger A. Prognostic significance of WT1 gene expression at diagnosis in adult de novo acute myeloid leukemia. Leukemia 1997; 11: 639–643. Gaiger A, Schmid D, Heinze G, Linnerth B, Greinix H, Kalhs P, Tisljiar K, Priglinger S, Laczika K, Mitterbauer M, Novak M, Mitterbauer G, Mannhalter C, Haas OA Lechner K, Jager U. Detection of WT1 transcript by RT-PCR in complete remission has no prognostic relevance in de novo acute myeloid leukemia. Leukemia 1998; 12: 1886–1894. Kreuzer KA, Saborowski A, Lupberger J, Appelt C, Na IK, le Coutre P, Schmidt A. Fluorescent 5⬘-exonuclease assay for the absolute quantification of Wilms’ tumour gene (WT1) mRNA: implications for monitoring human leukaemias. Br J Haematol 2001; 114: 313–318. Sakatani T, Shimazaki C, Hirai H, Okano A, Hatsuse M, Okamoto A, Takahashi R, Ashihara E, Inaba T, Yokota H, Nakahara K, Hirai H, Nakagawa M. Early relapse after high-dose chemotherapy rescued by tumor-free autologous peripheral blood stem cells in acute lymphoblastic leukemia: importance of monitoring for WT1 mRNA quantitatively. Leuk Lymphoma 2001; 42: 225–229. van Dongen JJ, Macintyre EA, Gabert JA, Delabesse E, Rossi V, Saglio G, Gottardi E, Rambaldi A, Dotti G, Griesinger F, Perreira A, Gameiro P, Diaz MG, Malec M, Langerak AW, San Miguel JF, Biondi A. 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. Miltenyi S, Muller W, Weichel W, Radbruch A. High gradient magnetic cell separation with MACS. Cytometry 1990; 11: 231– 238. Gabert J, Beillard E, Bi W, Pallisgaard N, Gottardi E, Cazzaniga


G, Barbany G, Cave` H, Cayuela JM, Grimwade D, Aerts JLE, van der Velden VHJ, Pane F, Saglio G, van Dongen JJM. European SANCO concerted action. European standardization and quality control program of Real Time quantitative RT-PCR analysis of fusion gene transcripts for minimal residual disease detection in leukemia patients. Blood 2000; 96: 1343 (Abstr.). 20 Gabert J, Beillard E, van der Velden VHJ, Bi W, Pallisgaard N, De Micheli D, Cazzaniga G, Barbany G, Cayuela JM, Pane F, Cave` H, Grimwade D, Aerts JLE, Thirion X, Pradel V, Gonza´lez M, Viehmann S, Malec M, Saglio G, van Dongen JJM. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction (RQ-PCR) of fusion gene transcripts for minimal residual disease detection in leukemia – A Europe Against Cancer Program. (In preparation).

21 Pallisgaard N, Hokland P, Bi W, Dee R, Van Der Schoot E, Delabesse E, Macintyre E, van der Velden VHJ,van Dongen JJM, Gottardi E, Saglio G, Watzinger F, Lion T, Beillard E, Gabert J. Control genes for Real-Time Quantitative RT-PCR in leukemia. A report from SANCO Concerted Action within the Europe Against Cancer Program. (In preparation). 22 Spinsanti P, de Grazia U, Faggioni A, Frati L, Calogero A, Ragona G. Wilms’ tumor gene expression by normal and malignant human B lymphocytes. Leuk Lymphoma 2000; 38: 611–619. 23 Inoue K, Ogawa H, Yamagami T, Soma T, Tani Y, Tatekawa T, Oji Y, Tamaki H Kyo T, Dohy H, Hiraoka A, Masaoka T, Kishimoto T, Sugiyama H. Long-term follow-up of minimal residual disease in leukemia patients by monitoring WT1 (Wilms tumor gene) expression levels. Blood 1996; 88: 2267–2278.

2121

Leukemia