Search

The phosphoinositide 3-kinase (PI3K/Akt) pathway is activated in acute myelogenous leukemia (AML) and is promising for targeted inhibition. Ninety-two patients with primary AML were analyzed for PI3K/Akt constitutive activation. Fifty percent of the patients presented with constitutive PI3K activation (PI3K +). No difference was observed between PI3K + and PI3K − groups concerning age, sex, white blood cell count, lactate dehydrogenase (LDH) level, bone marrow blast cells, French-American-British (FAB) classification, cytogenetics, RAS or nucleophosmin (NPM) mutations. Slightly more FLT3-ITD was detected in the PI3K − group (P = .048). The complete remission rate was similar between the 2 groups. With a median follow-up of 26 months, we observed for PI3K + and PI3K − patients, respectively, 56% and 33% overall survival (P = .001) and 72% and 41% relapse-free survival (P = .001). Constitutive PI3K/Akt activity is a favorable prognosis factor in AML, even after adjustment for FLT3-ITD, and may confer a particular sensitivity to chemotherapy. A better understanding of the downstream effectors of the PI3K/Akt pathway is needed before targeting in AML.

In the present study, we address the issue of the discontinuation of imatinib mesylate (Gleevec) in chronic myelogenous leukemia with undetectable residual disease for more than 2 years. Twelve patients were included. The median duration of real-time quantitative–polymerase chain reaction (RTQ-PCR) negativity and imatinib therapy were, respectively, 32 months (range, 24-46 months) and 45 months (range, 32-56 months) before imatinib interruption. Six patients displayed a molecular relapse with a detectable BCR-ABL transcript at 1, 1, 2, 3, 4, and 5 months. Imatinib was then reintroduced and led to a novel molecular response in most patients. Six other patients (50%) still have an undetectable level of BCR-ABL transcript after a median follow-up of 18 months (range, 9-24 months). We hypothesize that relapses observed within 6 months reflect the kinetics of undetectable dividing chronic myelogenous leukemia (CML) cells. Those cells may be eradicated or controlled in long-term nonrelapsing patients, as described in our study.

Not all patients with core binding factor acute myeloid leukemia (CBF-AML) display a good outcome. Modern risk factors include KIT and/or FLT3 gene mutations and minimal residual disease (MRD) levels, but their respective values have never been prospectively assessed. A total of 198 CBF-AML patients were randomized between a reinforced and a standard induction course, followed by 3 high-dose cytarabine consolidation courses. MRD levels were monitored prospectively. Gene mutations were screened at diagnosis. Despite a more rapid MRD decrease after reinforced induction, induction arm did not influence relapse-free survival (RFS) (64% in both arms; P = .91). Higher WBC, KIT, and/or FLT3-ITD/TKD gene mutations, and a less than 3-log MRD reduction after first consolidation, were associated with a higher specific hazard of relapse, but MRD remained the sole prognostic factor in multivariate analysis. At 36 months, cumulative incidence of relapse and RFS were 22% vs 54% (P < .001) and 73% vs 44% (P < .001) in patients who achieved 3-log MRD reduction vs the others. These results suggest that MRD, rather than gene mutations, should be used for future treatment stratifications in CBF-AML patients. This trial was registered at EudraCT as #2006-005163-26 and at www.clinicaltrials.gov as #NCT 00428558.

The Sokal, Hasford and EUTOS scoring systems have been successively proposed to predict the outcome of CML patients in chronic phase (CP-CML), both in terms of disease response and survival. The EUTOS score, which is the only one developed for patients treated with tyrosine kinase inhibitors, considers 2 risk groups based on spleen size and peripheral blood basophil percentage. No clear link has been identified between the covariates used to calculate those scores and CML molecular pathophysiology. To better delineate the molecular mechanisms underlying those scoring systems, a gene expression profiling study was performed on 48 CP-CML patients included in the Novartis ENEST1st study, for whom good quality peripheral blood total RNA was available at diagnosis. First, patients with low-risk (n=21) and high-risk (n=12) Sokal scores were compared, leading to identify 13 genes differentially expressed between those groups (q Disclosures Gardembas: Novartis: Speakers Bureau. Spentchian:Novartis: Research Funding. Giles:Novartis: Consultancy, Honoraria, Research Funding. Hochhaus:Pfizer: Honoraria, Research Funding; ARIAD: Honoraria, Research Funding; Bristol-Myers Squibb: Honoraria, Research Funding; Novartis: Honoraria, Research Funding. Radich:Gilliad: Consultancy; Incyte: Consultancy; Novartis: Consultancy, Research Funding; Ariad: Consultancy. Rousselot:Pfizer: Consultancy; BMS: Consultancy, Speakers Bureau; Novartis: Speakers Bureau.

Cytogenetically normal acute myeloid leukemia (CN-AML) compose between 40% and 50% of all adult acute myeloid leukemia (AML) cases. In this clinically diverse group, molecular aberrations, such as FLT3-ITD, NPM1, and CEBPA mutations, recently have added to the prognostic accuracy. Aberrant DNA methylation is a hallmark of cancer, including AML. We investigated in total 118 CN-AML samples in a test and a validation cohort for genome-wide promoter DNA methylation with Illumina Methylation Bead arrays and compared them with normal myeloid precursors and global gene expression. IDH and NPM1 mutations were associated with different methylation patterns (P = .0004 and .04, respectively). Genome-wide methylation levels were elevated in IDH-mutated samples (P = .006). We observed a negative impact of DNA methylation on transcription. Genes targeted by Polycomb group (PcG) proteins and genes associated with bivalent histone marks in stem cells showed increased aberrant methylation in AML (P < .0001). Furthermore, high methylation levels of PcG target genes were independently associated with better progression-free survival (odds ratio = 0.47, P = .01) and overall survival (odds ratio = 0.36, P = .001). In summary, genome-wide methylation patterns show preferential methylation of PcG targets with prognostic impact in CN-AML.

Various transplantation strategies have been designed to improve the poor prognosis of adult (ages 15 to 60 years) acute lymphoblastic leukemia (ALL). The GOELAL02 trial evaluated the impact of early allogeneic bone marrow transplantation (alloBMT) or delayed unpurged autologous stem cell transplantation (ASCT) for patients who had no human leukocyte antigen (HLA)-matched sibling donor or who were older than 50 years. Inclusion criteria included at least one of the following: age older than 35 years; non-T-ALL; leukocytosis greater than 30 x 10(9)/L; t(9;22), t(4;11), or t(1; 19); or failure to achieve complete remission (CR) after one induction course. Among 198 patients, the median age was 33 years. The CR rate was 80% with vincristine, idarubicine, L-asparaginase, and randomized intravenous injection or oral steroids (P = nonsignificant [ns]). AlloBMT was performed after 2 consolidation courses while ASCT was delayed after 1 additional reinduction. Intensified conditioning regimen before transplantation included etoposide, cyclophosphamide, and total body irradiation (TBI). Median follow-up was 5.1 years. The median overall survival (OS) was 29 months, with a 6-year OS of 41%. On an intent-to-treat analysis for patients younger than 50 years, alloBMT significantly improved the 6-year OS (75% versus 40% after ASCT; P = .0027). Randomized interferon-alpha maintenance had no effect on relapse or survival after ASCT. In conclusion, the outcome of adult ALL is better after early alloBMT than after delayed ASCT.

Monoclonal gammopathy of undetermined significance (MGUS) is a frequent condition in patients over 50 years old, that ultimately leads to multiple myeloma (MM) in 20% of patients after 20 to 35 years of follow-up. Little is known about cytogenetic changes associated with this condition. We studied 19 MGUS patients both at diagnosis and after 12 to 35 months of follow-up (mean = 26), using DNA content measurement of bone marrow plasma cells (BMPC), and a new interphase fluorescence in situ hybridization technique (FISH) allowing the simultaneous identification of monotypic BMPC (fluorescent anti light-chain antibodies) and the determination of the number of copies for two different chromosomes within the same PC nucleus (one biotin-labeled probe coupled next to texas red avidin and one FITC-labeled probe). At diagnosis of the MGUS, single interphase FISH showed at least one numeric chromosome change in 13 of 19 patients, after the use of centromeric probes directed against chromosomes no. 3, no. 7, no. 9, and no. 11. At follow-up, abnormalities found at diagnosis in 13 patients were still shown. Moreover, abnormalities occurred in three of the last six patients (trisomy for one to three different chromosomes), although no patient evolved into MM. Dual interphase FISH showed that some BMPC bore numeric changes with both probes tested whereas other BMPC bore abnormality with only one of the probes tested. In patients who showed trisomy for at least three different chromosomes, distribution of numeric changes within BMPC defined significant numbers of up to seven different BMPC clones. All these various clones were shown both at diagnosis and at follow-up. In every patient, these various clones differed only for the number of abnormalities they exhibited, and could be related to each other in a model of gradual acquisition of chromosome changes. Eventually, data reported here show that MGUS patients acquire slowly, gradually, but ineluctably chromosome changes, distributed within several related subclones. However, these changes are not related to transformation into MM: among the various clones coexisting within the same patient, a peculiar change, still to demonstrate, might develop and lead to overt MM.

Abstract 79 The prognosis of acute myeloblastic leukemia (AML) with intermediate-risk cytogenetics is refined further by testing the mutational status of NPM1,FLT3 and CEBPa genes. Patients with a NPM1mut\FLT3wt and CEBPa double mutation are now classified in a separate favorable group of the new ELN classification. Conversely, all other molecular profiles including these 3 genes, considered as carrying a worse prognosis, are included in ELN intermediate groups 1 and 2. Nevetheless the therapy of AML has remained unchanged since many years and needs improvement even in this subset of patients. The GOELAMS has tested the association of gemtuzumab ozozgamycin (GO), a monoclonal anti-CD33 antibody conjugated with chalicheamycin to standard chemotherapy in a phase III prospective randomized trial focusing on such patients with intermediate cytogenetics. Methods: Between 2007 and 2010, 254 patients aged between 18 to 60 years with de novo AML and intermediate karyotype have been included. The molecular status for NPM1, FLT3 and CEBPa mutations was determined at diagnosis. After 1/1 randomization, GO 6m/m2 was added to standard 3+7 induction and to a first MidAc intensive consolidation course (Mitoxantrone and intermediate doses of cytarabine). Patients in the ELN favorable molecular group received a second MidAc course followed by an autologous bone marrow transplant (BMT). Patients in the ELN intermediate 1 or 2 groups were considered for geno or phenoidentical allogeneic transplant. Patients 50 years of age or lower, received either standard allo-BMT preceded by a single course of chemotherapy or a reduced intensity chemotherapy (RIC) regimen after the two courses of intensive consolidation. In this sub-group, patients with no donor received autologous BMT after the two randomized courses of consolidation. Results: Two hundred and thirty-eight patients were analyzed with a median follow-up of 20 months. Their median age was 50 years old (18–60). There was no significant difference between the two arms according to age, WBC, % of circulating and % of marrow blasts, FAB subtypes, cytogenetics (normal vs abnormal) and molecular subgroups (favorable versus intermediate 1 and 2 ELN groups). In the GO group, the complete remission (CR) rate was 91.6% versus 86.5% (P=NS), early death(ED) was 10% versus 4.5% (P=NS). Major toxicity was observed in the GO arm during induction phase with 4 cases of veno-occlusive liver disease (VOD) and overall more toxic grade III/IV hepatic toxicities (23% versus 13%, p=0.031). No significant difference was observed between the two arms for grade III/IV hematological toxicity. No difference was observed either at 3 years for event free survival (EFS) at 51% in the GO arm and 33% in the other arm, nor for overall survival (OS) at respectively 53% and 46%. In the subset of patients who could not receive an allogeneic transplant, EFS was significantly higher in the GO group (53.7% vs 27%, p=0.0308) while there was no difference for OS. This better outcome was mostly observed in the group of patients classified ELN intermediate 1 or 2 (p=0.0126). Conclusion: In the subset of patients with intermediate cytogenetics AML, the adjunction of GO to standard chemotherapy failed to improved OS, yet a better EFS was observed with the addition of GO for patients who could not receive an allogeneic SCT. More detailed results will be presented. Disclosures: No relevant conflicts of interest to declare.

Abstract 3556 IDH1/2, TET2 and DNMT3A mutations have been reported in myeloid malignancies including de novo AML. In this study, we have analyzed the frequency and prognostic impact of these mutations in a large retrospective cohort of patients (pts) with secondary AML (SA) which encompass myelodysplasia-related changes (MRC) AML and therapy-related (TR) AML according to the WHO classification. Bone marrow samples were collected from 247 pts at diagnosis with SA and the mutational status of IDH1/2, TET2 and DNMT3A genes together with other genes frequently mutated in AML (NPM1, FLT-3, N and K-RAS, WT1) was determined by Sanger sequencing or high resolution melting analysis. The cohort of 247 pts consisted in 201 MRC AML and in 46 TR AML, 39.5% of which with a normal karyotype (NK). The frequency of IDH1/2, TET2 and DNMT3A mutations was 12.6, 19.8 and 4.5%, respectively. Two pts had both TET2 and IDH1/2 mutations, 2 pts had TET2 and DNMT3A mutations and 5 pts had both IDH1/2 and DNMT3A mutations showing that these mutations were not mutually exclusive in SA. IDH1/2 and TET2 mutations were significantly more frequent in MRC AML (14.1 and 22.3%) than in TR AML (6.4 and 8.7%) (P =0.04 and P =0.03) while the frequency of DNMT3A mutations was identical in the two subgroups. The SA pts harbouring at least one IDH1/2 or TET2 or DNMT3A mutation were significantly older (P Complete remission rate and overall survival were evaluated in a group of 158 pts which had received intensive chemotherapy at diagnosis, and were identical in the IDH1/2 or TET2 or DNMT3A mutated and unmutated groups. These mutations did significantly influence survival neither in the subgroup of pts with normal karyotype, nor in the subgroup of MRC-AML, or TR-AML which were of very poor prognosis. These data show that IDH1/2, TET2 or DNMT3A mutations could modify the clinical presentation without impact on prognosis. Disclosures: No relevant conflicts of interest to declare.

Abstract 410 Background. CBF-AML is a favorable AML subset cytogenetically defined by t(8;21) or inv(16)/t(16;16) rearrangements, respectively responsible for RUNX1-RUNX1T1 (CBFa) or CBFB-MYH11 (CBFb) gene fusion. Nonetheless, relapse incidence may reach 30–40% in these patients. Even if not yet prospectively validated, early MRD monitoring based on chimeric fusion gene transcript quantification by RQ-PCR may help to define patients at higher risk of relapse. Detection of activating KIT and FLT3 gene mutations has also been retrospectively associated with higher relapse rate in these patients. The French CBF-2006 trial (NCT00428558) aimed to prospectively evaluate both markers as predictors of remission duration in homogeneously treated patients. Methods. Patients (18–60y) with newly diagnosed CBF-AML were eligible. All were studied for KIT exon 8/17, FLT3-ITD, FLT3-D835, N-RAS and K-RAS mutations. They were randomized to receive timed-sequential or standard induction. Timed-sequential induction was cytarabine (AraC) 500 mg/m2 CI D1-3 and daunorubicine (DNR) 60 mg/m2 D1-3, followed by AraC 1 g/m2/12h D8-10 and DNR 35 mg/m2 D8-9. Standard arm was AraC 200 mg/m2 CI D1-7 and DNR 60 mg/m2 D1-3, followed by the second sequence at D16-18 only in case of persistent bone marrow (BM) blasts at D15. Patients in first complete remission (CR1) received 3 HDAC consolidation cycles with AraC 3 g/m2/12h D1/3/5. Fusion gene expression was quantified at baseline and before each HDAC cycle (MRD1, MRD2, MRD3) by the 100 × fusion gene/cABL ratio. Patients not reaching a 3-Log ratio reduction at MRD2 were offered allogeneic stem cell transplantation (SCT) in CR1 if they had an HLA-identical donor or could enter a single-agent dasatinib sub-study (see Boissel et al., this meeting). Results. Among the planned 200 patients randomized, 198 met eligibility criteria (96 CBFa and 102 CBFb, compared in Table 1). Overall, incidences of KIT, RAS, FLT3-D835 and FLT3-ITD mutations were 21%, 29%, 10% and 6%, respectively. Patients with KIT and FLT3-D835 mutations had higher WBC (P=0.04 and 0.02) and BM blast percentage (P=0.0004 and 0.05). Those with RAS mutations had lower BM blast percentage (P=0.01) and baseline transcript ratio (P=0.001). Overall, 196 patients reached CR1 and 55 relapsed. At 3 years, probabilities of durable remission, disease-free (DFS) and overall survival from CR were 64%, 61% and 84%, respectively, without differences between the CBF subsets or induction arms. Fifty-five patients did not reach the 3-Log MRD2 reduction (18 CBFa and 37 CBFb, P=0.006), 9 receiving SCT in CR1. Poor MRD2 response correlated with age (P=0.01), WBC (P=0.04), KIT mutations (P=0.04), and initial transcript ratio (P=0.004). Among the characteristics listed in Table 1, shorter remission duration was associated with high WBC (P=0.011), high BM blast percentage (P=0.062), del(9q) (P=0.041), KIT mutation (P=0.048), FLT3-D835 mutation (P=0.077), high initial transcript ratio (P=0.053), and Conclusions. This prospective study of 198 CBF-AML patients homogeneously treated with HDAC revealed significant correlations between initial characteristics and response to therapy. Prognostic analysis did not find that KIT mutations had independent impact on outcome, which was rather influenced by baseline fusion transcript ratio and post-consolidation MRD reduction (reminiscent of Schnittger et al. 2003), and FLT3-D835 mutation. Ongoing pan-genomic analyses might help to further dissect CBF-AML heterogeneity. Disclosures: No relevant conflicts of interest to declare.

Abstract 2608 Aim. CBF-AML is a favorable AML subset cytogenetically defined by t(8;21) or inv(16)/t(16;16) rearrangements, respectively responsible for RUNX1-RUNX1T1 (CBFα) or CBFB-MYH11 (CBFβ) gene fusion. Nonetheless, relapse incidence may reach 30–40% in these patients. The tyrosine kinase (TK) receptor KIT is expressed in the vast majority of CBF-AML and activating KIT gene mutations, including exon 8 del/ins and exon 17 point mutations, have been reported in both CBF-AML subtypes. These mutations have been retrospectively associated with a higher risk of relapse. Dasatinib is a TK inhibitor active on both wild-type and mutant KIT isoforms. These observations led several groups to initiate prospective trials testing dasatinib/chemotherapy combinations in newly diagnosed CBF-AML patients. The aim of the present Phase 2 DASA-CBF trial (EudracCT 2006-00655-12) was to search for a positive signal by treating CBF-AML patients in first complete remission (CR), but with persistent or re-appearing molecular minimal residual disease (MRD), by the single-agent dasatinib. Prevention of further hematological relapse was the primary endpoint. Methods. Eligible patients (18–60y) had to have been previously enrolled and treated in the CBF-2006 study (Jourdan et al. this meeting). MRD was quantified by RQ-PCR and results were expressed as 100 × CBF fusion gene/cABL ratios. Patients in first CR were eligible if: 1) MRD ratio reduction less than 3-Log before the second consolidation cycle (refractory Mol-REF patients); or 2) MRD ratio re-increase more than 1-Log on two successive evaluations (relapsing Mol-REL patients). Patients with a sibling or unrelated donor and no contra-indication to allogeneic stem cell transplantation (SCT) were not eligible. Dasatinib was administered orally at 140 mg once daily for a total duration of 12 months. In case of grade 3 adverse event (AE), treatment was discontinuated until AE resolution to grade 0–1. In case of grade 4 AE or AE reappearance, dose reduction to 100 mg/d was allowed. Dose escalation to 90 mg bid was allowed in case of stable or increasing MRD after 2 months of therapy without toxicity. Results. Between June 2008 and June 2011, 26 CBF-AML patients (median age, 44y) were included. They were 12 CBFα patients (6 Mol-REF, 6 Mol-REL) and 14 CBFβ patients (12 Mol-REF, 2 Mol-REL). Seven patients had a KIT mutation (4 exon 8 and 3 exon 17; 3 CBFα and 4 CBFβ). The median time between CR achievement and DASA-CBF enrollment was 161 days (106–406) in Mol-REF patients and 413 days (167–530) in Mol-REL patients. Overall, dasatinib was well tolerated. Two grade 3 AEs (hypertension, headache) led to dasatinib dose reduction without further reoccurrence. Dose escalation was performed in two patients. With the exception of one patient (without KIT mutation) still alive in CR1 at 929 days, 7 of the 8 Mol-REL patients had rapid hematological relapse under dasatinib treatment after a median time of 60 days (52–120). In the 18 Mol-REF patients, the probability of persistent hematological remission was 65% (95% CI, 38–82) at 12 months and 45% (95% CI, 20–67) at 24 months, with a trend for shorter remission duration in the 6 patients with KIT mutation (P=0.07). These 18 Mol-REF patients were compared to the 37 other Mol-REF patients from the CBF-2006 trial not enrolled in the DASA-CBF trial (9 SCT in CR1, 28 without further therapy). Both series were comparable in terms of age, WBC, baseline fusion transcript ratio, MRD ratio Log-reduction after induction and first HDAC consolidation, as well as additional cytogenetic anomalies and KIT, FLT3 and RAS gene mutations. Despite dasatinib treatment, the probability of persistent hematological remission was not higher in the 18 DASA-CBF patients than in the 28 patients who received no further therapy (45% versus 53% at 24 months, P=0.91). Conversely, only one hematological relapse was observed in the 9 patients who received SCT in CR1, the remaining 8 patients being alive in continuous CR. Conclusion. This Phase 2 trial failed to show a significant effect of single-agent dasatinib in the prevention of hematological relapse in CBF-AML patients in first CR after standard therapy, but at high risk of relapse because of persistent or re-appearing MRD. Moreover, no trend towards a preferential effect was observed in KIT mutated patients. This observation does not preclude any dasatinib activity in CBF-AML patients when combined with conventional chemotherapy. Disclosures: Off Label Use: Dasatinib is not approved for AML patients. Mohamed:Bristol-Myers Squibb: Employment.

Abstract 1658 Prognostic evaluation is a critical step in newly diagnosed pts with AML. With samples containing a high leukemic blast load, microarray-based gene expression profiling (GEP) and class prediction analyses have demonstrated their ability to assign AML samples to one of the three well characterized favorable-risk AML subtypes, ie, APL, CBFA-AMLs, CBFB-AMLs, with high accuracy and low error rates. One major issue regarding the use of microarray-based GEP in a routine prognostic workflow, concerns its ability to perform accurate class prediction analyses with samples containing as low as 20% blasts. So far, the minimum leukemic cell load that would still lead to correct class assignment has not been extensively studied. Another critical issue for a routine use of microarray-based GEP concerns samples that do not fulfill all quality control criteria along the process. The accuracy of class prediction analyses using data derived from poor quality AML samples also deserves further evaluations. In this study, using the Illumina BeadChip technology, GEP was first assessed for its ability to correctly classify APLs, CBFA-AMLs, CBFB-AMLs within a training set of 72 samples containing at least 60% of leukemic cells (20 APL samples from 14 pts, 12 CBFA-AML samples - n=6 pts, 11 CBFB-AML samples - n=10 pts). This training set also included normal bone marrow samples as well as a more heterogeneous group of 29 NK-AML samples (28 pts). The classifiers derived from this supervised analysis were then evaluated for their ability to assign to the correct class 93 test samples with either low leukemic blast load (as low as 5% of blasts – n=79 - 22 samples that originally contained less than 60% of leukemic cells, 57 samples with high blast load that were artificially diluted within a pool of normal bone marrows at 50 and 75%, ie, final leukemic blast load after dilution ranging from 10 to 50%), poor quality control criteria (n=10), or with atypical characteristics (n=4). The Class Prediction Module of ArrayMiner 5.3.3 software, which is a supervised method based on a proprietary optimization technique called grouping genetic algorithms (GGA), was used to build the classifiers associated with each class of the Training Set and assess the predictive capacity of these classifiers on Test Set samples. Using the Training Set and the GGA-base supervised method, the best model fitness was achieved with classifiers including 14 markers per class. With these classifiers, all Training Set samples were assigned to the correct class. Considering Test Set samples, all APL and CBFA-AML samples were assigned to the correct class, including the ones with the lowest leukemic blast load, i.e., 7% for APLs and 15% for CBFA-AMLs. One of the 23 CBFB-AML Test Set samples, which contained 5% of blasts was misclassified. All other CBFB-AML test samples were correctly classified, even those contained as low as 10% of leukemic cells. All but one of the NK-AML test samples were assigned to the correct class, even though some contained as low as 19% of leukemic cells. Concerning the 10 poor quality control samples characterized by either low RIN values and degraded total RNAs, low amounts of labeled cRNAs or both low RIN value and low labeled cRNA, all were assigned to the expected class. All atypical AMLs were assigned, as expected, to the NK-AML group. In conclusion, favorable cytogenetic risk AMLs with low leukemic cell load or poor quality control criteria can be correctly assign to the appropriate group using GEP. Transposability of this Illumina technology-based model and its efficiency in term of predictive capacity, will be presented using an independent series of 245 AML samples processed at a different institution, on Affymetrix GeneChips. Disclosures: No relevant conflicts of interest to declare.

Abstract 176 TET2 mutations were recently involved in myeloid malignancies as in 20–26% of MDS, in 12% of MPN and in 10–17% of AML (Delhommeau et al, 2009, Langemeijer et al, 2009, Tefferi et al 2009). A link between TET2 and NPM mutations was recently highlighted in de novo AML with normal karyotype (Nibourel et al, 2009). Actually, the impact of TET2 mutations is not fully examined in secondary AML (SA). In the new OMS classification of hematological malignancies, SA are related to myelodysplasia-related changes (MRC) AML and therapy-related (TR) AML where the impact of TET2 mutations is not fully examined. In our retrospective study, we have collected bone marrow samples from 247 patients at diagnosis identified as SA and classified according WHO. On genomic DNA, we have determined the status of TET2 as described before (Delhommeau et al, 2009) but also the status of other genes as NPM1, FLT-3, N and K-RAS, c-KIT. According to the WHO classification, the 247 samples were subdivided in 201 MRC AML where the most of the patients had a clear previous history of haematological malignancies (MDS, MPN or MDS/MPN, n=114) and in 46 TR AML. In this cohort of 247 SA, we found 70 patients (39.5%) with a normal karyotype (NK). 69 abnormalities of the TET2 sequence were found and dispatched into 49 patients leading to 19.8% of patients with at least one TET2 mutation (29 Frameshift all inducing a premature stop codon, 24 Nonsenses, 12 Missenses in conserved regions, 2 insertions of one AA and 2 mutations of a splicing site). The SA patients harbouring at least one TET2 mutation are older than others (64.6 years vs 71.4 years, p Cytogenetically, the patients without a TET2 mutation (n=198) present a NK in 45 cases (22.7%), an abnormal KT in 153 cases and in 57.5% (n=88) a complex KT with at least 3 anomalies. In the group of TET2 mutated patients, there is 51% of NK indicating that TET2 mutations are strongly associated with NK (p When we have restricted the study of biological parameters to patients presenting a NK (n=70 with 25 TET2 mutated patients), the patients with a TET2 mutation are still older (60.5 vs 67.8 years, p Among the others molecular abnormalities tested in these 247 SA, no statistical association could be found between TET2 mutations and NPM1 mutations tested on 148 patients (p=0.071), or FLT3 abnormalities tested on 218 patients (p=0.09) or Ras mutations tested on 151 patients (N Ras and K Ras, p=0.886). However, there is a statistical association between KIT 816 mutations (determined on 156 samples) and TET2 mutations (p Finally and in a very interesting way, we can observe that the frequency of TET2 mutations in MRC AML is higher (22.3%) compared to the frequency established in TR AML (8.7%). In conclusion, the SA associated with at least one TET2 mutation present a specific biological pattern, not related to a NK, and actually not understood. Moreover, considering the fact that TET2 mutations are strongly associated with NK and that the TR AML present less TET2 mutations than the others, TET2 mutations are, in these cohort, not associated with markers of poor prognosis which are complex karyotype and a previous intensive chemotherapy. Disclosures: No relevant conflicts of interest to declare.

We have identified FOXC1 gene as being over-expressed in more than 50% of AML patients (pts) with normal karyotype when compared to normal hematopoietic cells in microarray analysis. Using RQ-PCR, we validated this finding and assessed the prognostic value, at diagnosis, of FOXC1 expression in a series of 142 adult AML pts (AML0 to AML6) followed at our institution. At diagnosis, pts median age was 62 years. Cytogenetics were available for 136 pts (MRC low-risk n=25 [AML3 n=12, CBF-AML n=13], intermediate-risk n=79, high-risk n=32). NPM1/Flt3ITD status was available for 129 pts (NPM1+/Flt3ITD– n=26, Flt3ITD+ n=18, NPM1–/Flt3ITD– n=85), and WT1 expression for all pts. Ninety-three pts received induction chemotherapy. Allogeneic SCT was performed in 21 pts. PCR results were compared to FOXC1 expression in K562 (100%). Results: FOXC1 expression level was higher in AML pts (median signal 152%; 25–75th percentiles 32–1036%; range 3–25280%) than in normal blood mononuclear cells (n=10; median signal 16%; 25–75th percentiles 14–26%), marrow cells (n=21; median signal 24%; 25–75th percentiles 16–42%)(p In pts with intermediate-risk cytogenetics receiving induction chemotherapy (n=57), those within Q1 had a worse 5-year survival (0% vs 58%; p=.038 - RR, 3.05; 95%CI, 1.03–9.04) and EFS (0% vs 44%, see Figure), with a higher relapse incidence (66% vs 32%, p Conclusion: FOXC1 expression level appears to be a useful new prognostic marker in adult AML pts with intermediate-risk cytogenetics. Main covariates associated with FOXC1expression Covariates % of patients in Q1 (S1036%) AML3 75% 16% 0% Other FAB AML 20% 205% 26% Low R KaryoT 44% 48% 0% Interm R KaryoT 19% 183% 25% High R KaryoT 22% 295% 37% Monosomy 7 18% 1258% 55% Normal KaryoT 10% 225% 31% NPM1+/Flt3ITD– 8% 917% 44% Other genotypes 28% 97% 19% Figure Figure

Chronic Myelogenous Leukemia (CML) originates in the chromosome (Ph1), a reciprocal translocation, corresponding to the BCR-ABL fusion oncogene. A small proportion (1–2%) of CML patients show breakpoints falling outside of the M-BCR gene on chromosome 22, leading to the synthesis of a variety of atypical BCR-ABL transcripts [either shortened: e1a2 (m-BCR), e6a2, e8a2, b2a3 (e13a3), b3a3 (e14a3), or elongated transcripts: e19a2 (m-BCR)] and to the synthesis of different BCR-ABL proteins. In this study, we retrospectively analysed the clinical characteristics and outcomes of cohorts of CML patients harbouring atypical transcripts in and treated with imatinib (IM). Twenty-two patients were analysed: 9 e1a2 [Group 1 (G1)], 4 e6a2 [Group 2 (G2)], 5 e19a2 [Group 3 (G3)], and 4 e8a2 [Group 4 (G4)] BCR-ABL transcripts. Two patients were in myeloid blastic phase at onset (1 in G1 and 1 in G2) and others in chronic phase. Age at diagnosis was significantly younger for e19a2 patients (39.5 years versus 64 for G1, 58.5 for G2, 72 for G3, p=0.005). Female patients were predominant for G1 (5F/3M), but not for other groups. All patients presented a classical Ph1 at karyotyping analysis at diagnosis, but 1 had a -7 (G1), 1 an additional t(11;16) with the Ph1 (G2), 1 a +8 (G3) and 1 a -Y (G3). The majority of patients presented typical CML features at diagnosis, however number of differences could be found: WBC counts were higher for e1a2 and e8a2 patients (74.2 109/l and 62.7 respectively vs 20.9 for G2, and 37.8 for G3, p