12th Congress of the European Hematology ... - Haematologica

to set up and optimize a D-HPLC-based screening method for mutations
in critical regions of Kit; to assess the sensitivity and reliability of
our D-HPLC assay as compared to RFLP analysis; to characterize additional
mutations. Methods. Kit mutation analysis was performed on 77
bone marrow and/or peripheral blood samples obtained from 51
patients. For each sample a PCR product of 287 bp, spanning codons 763-
858 corresponding to the catalytic loop and to the activation loop, was
screened in parallel by D-HPLC assay and by RFLP assay. In case of a positive
D-HPLC signal, direct sequencing was performed to confirm the
presence of a mutation. The PCR product was digested with the restriction
enzime HinfI to detect a GAC-to-GTC nucleotide change at codon
816, leading to a Asp-to-Val amino acid substitution (D816V). For each
sample scored as wild-type by D-HPLC and by RFLP analysis, a PCR
product of 350 bp, spanning codons 510-626 corresponding to the transmembrane
domain and to the juxtamembrane domain, was screened
by D-HPLC combined with direct sequencing. Results. By RFLP analysis
34/51 pts were positive for the D816V. By D-HPLC analysis, an abnormal
eluition profile was seen in 36/51 pts - all the 34 RFLP-positive cases
as well as two additional pts. Direct sequencing confirmed the presence
of the D816V in all the 34 RFLP-positive cases and showed that in
two of these cases a I798I polymorphism was also present. The two pts
scored positive by D-HPLC but negative by RFLP were found to have the
I798I polymorphism. The 15 pts who did not harbour ES type mutations
were further investigated by D-HPLC analysis of a RT-PCR product
spanning the transmembrane and juxtamembrane domains. D-HPLC
showed an abnormal elution profile in 5 pts. By direct sequencing one
patient showed the K546K mutation and 4 pts showed the M541L mutation
in the TM domain. Conclusions. Our D-HPLC'based assay proved a
straightforward, reliable and sensitive method for Kit mutation analysis.
Furthermore our D-HPLC-based screening method highlighted the
importance of screening for mutations other than the D816V, mainly
because the function of Kit regions, such as TM domain, is still unclear.
Supported by: European LeukemiaNet, COFIN 2003 (M. Baccarani), Ateneo
Grant (GM), AIL, AIRC, Fondazione Del Monte di Bologna e Ravenna.
0570
PREPARING FOR JAK2 V617F TARGETED THERAPY: DEVELOPMENT OF A HIGHLY SENSI-
TIVE AND SIMPLE REAL-TIME RT-PCR METHOD FOR JAK2 V617F TRANSCRIPT QUANTIFI-
CATION
B. Maes, R. Smets, G. Bries, V. Madoe, V. Peeters, J.L. Rummens
Virga Jesse Hospital, HASSELT, Belgium
Background. The identification of the JAK2 V617F mutation, occurring
in almost all cases of polycythemia vera (PV) as well as in approximately
60% of cases of essential thrombocythemia (ET), has strongly simplified
the diagnosis and classification of these diseases. Recently, it has
been suggested that the JAK2 V617F burden may have prognostic significance.
In addition, new promising JAK2 V617F targeted drugs are
under development. Aims. We developed a novel real-time RT-PCR
method for quantification of JAK2 V617F transcripts that would allow
determination of JAK2 V617F burden at diagnosis as well as the evaluation
of the response to newly developed therapeutic agents. Methods.
RT-PCR reactions were performed on a Rotor-Gene 3000 (Westburg) in
single tubes with 1 set of primers and two differently labelled, allele-specific
TaqMan probes, directed to respectively wild-type (WT) and mutant
(Mut) JAK2 sequences (1 mismatch). Probes were adapted by Locked
Nucleic Acid (LNA) modification for increased hybridization specificity
and enhanced allelic discrimination. Standard curves were constructed
with JAK2 V617F WT and Mut plasmids. Results are expressed as percentage
of JAK2 V617F of total JAK2. Whole peripheral blood or bone
marrow samples of a total of 54 JAK2 V617F positive cases, including 23
untreated and 7 conventionally treated PV cases and 19 untreated and 5
conventionally treated ET cases, were analysed. In addition, also 30
peripheral blood samples of normal individuals were analysed. Results.
Reaction efficiencies of this single tube assay for JAK2 Mut and JAK2 WT
were equal (97%). Quantities down to 10 copies of JAK2 Mut plasmid
amongst WT cDNA and patient JAK2 V617F cDNA diluted down to
0,09% into WT cDNA could be reliably detected. Low intra- and interassay
variabilities ensure good reproducibility of the assay. None of the
negative control samples showed any increase of the fluorescent signal
derived from the Mut probe, demonstrating the high specificity of the
assay and no requirement for defining a cut-off value. For PV patient
samples, the assay showed mean JAK2 V617F quantities of 82% for
untreated cases versus 56% for treated cases. Untreated ET cases showed
a significantly lower mean JAK2 V617F% compared to untreated PV
cases (55% versus 82%). Conclusions. We have developed a robust and
simple method for quantification of JAK2 V617F transcripts that is more
sensitive than all previously described methods. It provides the potential
to evaluate the prognostic significance of the JAK2 V617F burden at
diagnosis as well as the response to JAK2 V617F targeted therapy that
will become available in the near future.
0571
THE VALUE OF CLOSE MONITORING OF WT1 EXPRESSION LEVEL DURING THERAPY AND
POST-THERAPY FOLLOW-UP OF ACUTE MYELOID LEUKEMIA: AN INDEPENDENT
PROGNOSTIC FACTOR AND A VALUABLE PREDICTOR OF RELAPSE
H.B. Ommen, 1 C.G. Nyvold, 1 K. Brændstrup, 1 B.L. Andersen, 1
H. Hasle, 2 P. Hokland, 1 M. Østergaard1 1 2 Laboratory of Immunohaematology, ÅRHUS C; Department of Paediatrics,
AARHUS, Denmark
Background. The Wilms’ tumour gene 1 (WT1) is expressed during normal
fetal development, but only to a low extent in adult tissues. However,
since it is highly expressed in virtually all AML patients, it has been
suggested as a tool for minimal residual disease (MRD) detection and for
individualized therapy decisions. 1-2 Aims. a) To determine whether high
residual WT1 expression in first complete remission (CR1), established
by morphology and immunophenotyping, is an adverse prognostic factor,
and b) to evaluate the value of WT1 expression levels in peripheral
blood (PB) and bone marrow (BM) for prediction of disease relapse.
Methods. WT1 levels were quantified using real-time quantitative RT-
PCR by normalization to the control genes β2M and ABL, and in followup
samples expressed as a fraction of the BM diagnostic level. (described
in detail in (1)) Normal BM and PB WT1 levels were defined as described
in. 1 185 patients (160 adults, 25 children) treated at the Departments of
Haematology and Paediatrics, Aarhus University Hospital, were analyzed
at diagnosis. A cohort of 89 patients (73 adults, 16 children) were
selected for follow-up based on high WT1 expression and lack of fusions
transcripts. These patients were sampled at every visit during therapy
and follow-up (median number of WT1 determinations per patient: 11,
range 2-38). Prognostic difference between groups was determined using
the Cox Proportional Hazards statistical model including age, sex, cytogenetics,
de novo secondary leukemia and FLT3-ITD. When comparing
the predictive value of rising WT1 expression levels in PB and BM
Wilcoxon’s ranksum test was employed. Results. When we analyzed the
WT1 expression of patients achieving CR1 we found that the disease free
survival (DFS) in the group with BM WT1 expression above normal levels
at CR1 was significantly shorter than in the BM WT1-normal group
(Hazard ratio (HR) = 6,88 (95% Confdence interval (CI) 2,07-22,9),
p=0,002). Similarly, the DFS was significantly shorter in the PB WT1
high group vs. the PB WT1 normal group (HR=8,60, CI 1,76-41,9,
p=0,008). Of even greater importance, we were able to address relapse
kinetics in 29/32 relapses observed in the 89 patient cohort. We were able
to detect WT1 above normal levels in 100% of the BM samples that
available 3 months before relapse. (Range 1-8, median 4 months). 33%
of PB samples were positive 3 months before relapse (Range 0-8, median
1 month). (p=0,0086) Conclusions. CR1 WT1 expression levels in both
BM and PB are independent prognostic factors in AML. Relapse was
seen significantly earlier in BM than PB, but WT1 levels above normal
can still be seen in PB in 33% of patients 3 months prior to relapse.
References
12 th Congress of the European Hematology Association
1. Østergaard, M., et al., 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.
2. Weisser, M., et al., Prognostic impact of RT-PCR-based quantification of
WT1 gene expression during MRD monitoring of acute myeloid
leukemia. Leukemia 2005;19:1416-23
0572
A MOLECULAR CYTOGENETICS STUDY OF THE ROLE OF CHROMOSOME 9P IN CML CELL
LINES AND CMPD PATIENT SAMPLES
R. Vaughan, D. Brazma, C. Grace, J.D. Howard, E.P. Nacheva
Royal Free Hospital, LONDON, United Kingdom
Chronic Myeloproiferative Disorders (CMPDs) are a spectrum of
haematological malignancies of which the molecular pathogenesis
remains unknown. Chronic Myeloid Leukaemia (CML) being the only
disorder with a recognisable pathogenomic abnormality, the Ph chromosome.
Recent published data has supported the role of tyrosine kinases
haematologica/the hematology journal | 2007; 92(s1) | 213