27.12.2012 Views

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

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

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

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Ph + cells in <strong>the</strong> bone marrow). We also evaluated <strong>the</strong> possible influence<br />

<strong>of</strong> SVV levels on <strong>the</strong> clinical outcome (time to progression, response to<br />

<strong>the</strong>rapy and overall survival) <strong>of</strong> <strong>the</strong> pts. All CML pts underwent IM <strong>the</strong>rapy<br />

at 400 mg/die and <strong>the</strong>ir SVV levels were assessed by quantitative RT-<br />

PCR prior to <strong>the</strong> first IM treatment. Complete hematologic response<br />

(CHR), cytogenetic response (Cy-R) and molecular response (MolR) were<br />

scored as previously described (Baccarani et al. Blood 2006;108:1809).<br />

Results and Conclusions. All CML pts expressed <strong>the</strong> SVV transcript (median<br />

SVV=727 units). Nineteen out <strong>of</strong> 20 CML pts (95%) reached a CHR<br />

at 3 months; 10/20 pts (50%) obtained a Cy-R (complete + partial) at 12<br />

months; 7/20 pts (35%) achieved a MolR (complete + major) at 12<br />

months. Four pts (20%) displayed a failure to treatment within 12<br />

months. In our analysis, SVV transcripts produced a median time to progression<br />

and survival <strong>of</strong> 31 (p=0.04) and 53 months (p=0.6), respectively.<br />

These findings indicate that increased SVV expression reduces IM<br />

<strong>the</strong>rapeutic efficacy, in agreement with our previous in vitro observation<br />

that high SVV levels significantly reduce cell death in CML cells. Thus,<br />

our data suggest that strategies aimed at lowering SVV expression may<br />

improve <strong>the</strong> efficacy <strong>of</strong> IM, <strong>the</strong>refore representing a possible <strong>the</strong>rapeutic<br />

option for both IM-sensitive and IM-resistant CML patients.<br />

0534<br />

DETECTION OF SINGLE NUCLEOTIDE POLYMORPHISMS IN THE ABL KINASE DOMAIN OF<br />

CML PATIENTS<br />

T. Ernst, M.C. Müller, P. Erben, N. Härtel, C. Walz, R. Hehlmann,<br />

A. Hochhaus<br />

Universitätsklinikum Mannheim, MANNHEIM, Germany<br />

Background. BCR-ABL kinase domain mutations constitute <strong>the</strong> leading<br />

cause <strong>of</strong> resistance in CML patients treated with imatinib. Detection <strong>of</strong><br />

BCR-ABL mutations prior to and during <strong>the</strong> course <strong>of</strong> imatinib <strong>the</strong>rapy<br />

may aid in risk stratification as well as in determining optimal individual<br />

<strong>the</strong>rapeutic strategies. More than 50 different point mutations have<br />

been described so far. The K247R change has been identified as a rare<br />

gene polymorphism occurring also in normal control patients and o<strong>the</strong>r<br />

cell types. Despite its position near to <strong>the</strong> P-loop, biochemical and cellular<br />

assays <strong>of</strong> imatinib and dasatinib sensitivity showed no alteration<br />

compared to wild-type ABL. Aims. Since gene polymorphisms do not<br />

automatically necessitate a change in <strong>the</strong> <strong>the</strong>rapeutic strategy we sought<br />

to investigate if o<strong>the</strong>r BCR-ABL mutations are ra<strong>the</strong>r polymorphisms<br />

than mutations associated with resistance to tyrosine kinase inhibitors.<br />

Methods. In comparison to real BCR-ABL kinase domain mutations, single<br />

nucleotide polymorphisms (SNPs) also have to occur on <strong>the</strong> normal,<br />

untranslocated ABL allele. Therefore, we designed an allele-specific PCR<br />

for amplification <strong>of</strong> <strong>the</strong> normal ABL allele following mutation analysis<br />

by denaturing high-performance liquid chromatography (D-HPLC) and<br />

subsequent sequencing. A multiplex PCR was performed to ensure that<br />

both Ib and Ia ABL splice variants were amplified. Ninety-one imatinibresistant<br />

CML patients (chronic phase, n=88; accelerated phase, n=1;<br />

blast crisis, n=2) with 65 different BCR-ABL kinase domain mutations<br />

leading to 50 different amino acid changes were investigated in our<br />

study. 39 <strong>of</strong> 65 different mutations were evaluable in two different<br />

patients and 26 different mutations in one patient, respectively. Results.<br />

Amplification <strong>of</strong> <strong>the</strong> normal ABL allele succeeded in all cases. The D-<br />

HPLC analysis indicated a mutation in <strong>the</strong> normal ABL allele in eight cases.<br />

The nucleotide exchanges were confirmed by direct sequencing in<br />

both directions. In all cases <strong>the</strong> same mutation <strong>of</strong> <strong>the</strong> BCR-ABL allele was<br />

confirmed in <strong>the</strong> normal ABL allele (mutated ABL (%)/ mutated BCR-<br />

ABL (%)): T240T (50/100), F311V (50/100), T315T (40/100), Y320C<br />

(90/30), K247R (100/20; 40/100) and E499E (100/100; 40/100). In three<br />

cases <strong>the</strong> SNP did not lead to an amino acid change (silent mutation). Five<br />

patients had a wild-type as well as a mutated ABL allele leading to 50%<br />

mutated ABL allele. In <strong>the</strong>se patients <strong>the</strong> SNP including ABL allele seems<br />

to be translocated to BCR-ABL thus containing 100% mutated BCR-<br />

ABL. Ano<strong>the</strong>r patient harbouring <strong>the</strong> F311V mutation at <strong>the</strong> BCR-ABL<br />

allele did not show <strong>the</strong> mutation at <strong>the</strong> normal ABL allele. Conclusions.<br />

In addition to <strong>the</strong> previously described polymorphism K247R we uncovered<br />

five new polymorphisms, two <strong>of</strong> <strong>the</strong>se led to amino acid changes.<br />

SNPs should be taken into consideration for design <strong>of</strong> mutation-specific<br />

PCRs, particularly <strong>the</strong> T315T polymorphism for T315X amplifications.<br />

Newly identified mutations should be confirmed by amplifying<br />

<strong>the</strong> normal ABL allele to exclude polymorphisms. Clinicians should be<br />

aware that polymorphisms do not reflect mutations that automatically<br />

demand a change in <strong>the</strong>rapeutic strategy, unless <strong>the</strong>re are o<strong>the</strong>r signs <strong>of</strong><br />

inadequate response to treatment.<br />

12 th <strong>Congress</strong> <strong>of</strong> <strong>the</strong> <strong>European</strong> <strong>Hematology</strong> Association<br />

0535<br />

MAPK-ACTIVATION IN RESPONSE TO BCR-ABL INHIBITION IS CYTOKINE-DEPENDENT<br />

AND CAN BE OVERCOME BY DASATINIB<br />

P. La Rosée, N. Härtel, T. Klag, R. Hehlmann, A. Hochhaus<br />

University <strong>of</strong> Heidelberg, MANNHEIM, Germany<br />

Background. Inhibition <strong>of</strong> BCR-ABL by imatinib mesylate is current<br />

standard treatment for patients with chronic myelogenous leukaemia<br />

(CML). Despite favourable response rates, imatinib fails to eradicate<br />

minimal residual disease in <strong>the</strong> majority <strong>of</strong> patients, which is believed<br />

to be caused by stem-cell resistance. A proposed mechanism <strong>of</strong> human<br />

CML stem-cell survival is growth-factor- (GF) dependent activation<br />

(phosphorylation) <strong>of</strong> <strong>the</strong> MAPK Erk1/2 in response to imatinib (Chu et<br />

al., Blood 2004). Aims. We used IL3-dependent murine haematopoietic<br />

cell lines and <strong>the</strong>ir IL3-independent BCR-ABL-expressing progenies to<br />

study MAPK-signalling in response to imatinib and <strong>the</strong> high affinity<br />

BCR-ABL inhibitor dasatinib. Primary CD34 + enriched human CML cells<br />

were used, to validate cell-line results in a clinically relevant cellular context.<br />

Methods. Murine Baf3 and 32D cells transformed by p210Bcr-Abl<br />

were grown in <strong>the</strong> presence <strong>of</strong> 5 µM imatinib with and without WEHIconditioned<br />

media as source <strong>of</strong> IL3. Cytotoxicity was determined by<br />

trypan blue viability count. Cells were harvested after 3h, 6h, and 24h<br />

and lysed for Western blot analysis. Human CML mononuclear cells<br />

derived from 5 patients with first chronic-phase CML prior to treatment<br />

were enriched by magnetic separation for CD34 + . MAPK-signalling after<br />

treatment with BCR-ABL-inhibitors for 16h was studied in serum-free<br />

media in <strong>the</strong> presence <strong>of</strong> standard GF-mix. Results. Baf3p210 and<br />

32Dp210 cells treated with 5 µM imatinib for 24h were rescued by IL3<br />

as determined by trypan blue viability count. Western blot analysis after<br />

3h, 6h, and 24h <strong>of</strong> treatment demonstrated time-dependent significant<br />

activation <strong>of</strong> Erk1/Erk2 with peak phosphorylation after 6h in <strong>the</strong> presence<br />

<strong>of</strong> IL3. In <strong>the</strong> absence <strong>of</strong> IL3, Erk1/Erk2 phosphorylation was abolished<br />

at any time. Simultaneous detection <strong>of</strong> phosphotyrosine (pTyr)<br />

and BCR-ABL demonstrated complete inhibition <strong>of</strong> BCR-ABL with and<br />

without IL3. Similar results were obtained in both cell lines. Exposure <strong>of</strong><br />

Baf3p210 cells with 60nM dasatinib induced marked MAPK-inhibition<br />

independent <strong>of</strong> IL3 as early as 3h after start <strong>of</strong> treatment. Treatment <strong>of</strong><br />

CD34+ enriched CML cells confirmed imatinib-induced MAPK-activation.<br />

Exposure <strong>of</strong> patient (n=5) cells with dasatinib (12 nM, 62 nM) in<br />

contrast showed significant (p

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!