Search

BACKGROUND: AML in elderly patients, secondary AML and/or AML with complex karyotype respond poorly to conventional intensive chemotherapy (CT) and have a poor prognosis. In those patients, however, AZA can bring some benefit, especially in the absence of adverse cytogenetics, if WBC < 15 × 10⁹/L and performance status (PS) < 2 (Thepot, Am J Hemat, 2014; Pleyer Ann Hematol, 2014). Whether molecular findings can predict outcome in AML receiving first line AZA (as they do in AML receiving CT) is however unknown. PATIENTS AND METHODS: 96 AML patients who received between 2007 and 2012 first line treatment with AZA (75mg/m2/d x 7 d/4 weeks SC for a median of 7 cycles, range 1-33) in 8 French centers were retrospectively analyzed. Initial response was evaluated based on blood and marrow examination after 4 -6 cycles of AZA. We analyzed the impact on overall survival (OS) of conventional clinico-biological parameters, and molecular biology parameters including EVI1 expression, MLL-PTD, and the mutational status of 16 genes analyzed by next-generation sequencing (NGS) and Sanger sequencing: ASXL1, CEBPA, DNMT3A, EZH2, FLT3, IDH1, IDH2, NPM1, NRAS, KRAS, RUNX1, SF3B1, SRSF2, U2AF1, TET2, and TP53. RESULTS: Median age was 73 (range, 44-88). Median PS was 1 (range, 0-3). 55% patients had secondary AML, including post MDS or MPN AML (N=39), and therapy related AML (N=14). 64% of patients had high risk cytogenetics and only 14 normal karyotype. Median WBC was 3.87 G/L and 28 patients had WBC >10 G/L. Median platelets and Hb were 61 G/L and 9.25 g/dL respectively. 41 and 55 patients had marrow blasts < 30% and ≥ 30% respectively, and median circulating blast % was 15% (range, 0-95). The median number of somatic mutations per patient was 2 (range, 0-5). The most frequently mutated genes were TP53 (40%), TET2 (22%) and SRSF2 (19%), then DNMT3A (14%), ASXL1 (13.5%), IDH2 (11%) and RUNX1 (11%). EVI1 overexpression (defined stringently in this study as DD CT ≤ 8 EVI1/ABL) was found in 12 patients. Overall response rate was 29% (14% CR, 8% PR and 7% Cri), and median response duration was 9.4 months. Median OS was 10.9 months. In univariate analysis, the following factors were significantly associated with shorter OS: platelets and Hb below median (p OS was significantly worse in patients with TP53 gene mutation (median 8.9 vs 12.3 months in unmutated cases, p=0.01). It tended to be worse for SRSF2 gene mutation (median 8.4 vs 11.5 months, p=0.16) and TET2 gene mutation (median 8.8 vs 11.7 months, p=0.20) and tended to be better in patients with EVI1 overexpression (median 16.1 vs 9.4 months in other patients, p=0.17), while the mutational status of ASXL1, DNMT3A, IDH2 and RUNX1 genes had no impact on OS. Median survival of patients with at least one mutation of TP53, TET2, SRSF2 genes, was 9.1 vs 14.8 months if the 3 mutations were absent (p=0.0008, figure A). In multivariate analysis (logistic regression according to the Wald method), higher PS (p=0.0061), lower Hb (p=0.0014), lower platelets (p=0.0139) higher % circulating blasts (p=0.0145) and presence of TP53, TET2 or SRSF2 mutation (p=0.0035) were associated with poorer OS. Due to the low number of patients with EVI1 overexpression, this parameter was not included in the multivariate analysis. However, the 7 patients with EVI1 overexpression without any poor prognostic mutation had a median OS of 25.1 months ,versus 8.8 in patients carrying TP53, TET2 and/or SRSF2 mutations and 12.6 months in remaining patients without EVI overexpression (p=0.022) (figure B). Figure 1 Figure 1. Figure 2 Figure 2. CONCLUSION: In this first analysis (to our knowledge) of the prognostic value of molecular factors in poor risk AML (55% secondary cases, 64% unfavorable karyotype) receiving first line treatment with AZA, mutations of the TP53 and to a lesser extent TET2 or SRSF2 genes were adverse prognostic factors, and added prognostic value to conventional clinical and hematological parameters. For TET2, the effect seemed different from the favorable prognosis observed in MDS treated with AZA (Itzykson, Leukemia, 2011; Sekeres, Blood, 2012). Surprisingly also, EVI1 overexpression, a poor prognostic factor in AML, seemed associated with better outcome with AZA treatment in the absence of any poor prognostic mutation Those findings will require confirmation in larger prospective series. Disclosures No relevant conflicts of interest to declare.

Abstract 3932 Poster Board III-868 Background Waldenstrom's macroglobulinemia (WM) is a rare lymphoproliferative disorder characterized by bone marrow (BM) infiltration of lymphoplasmacytic cells that secrete monoclonal IgM antibody. Approximately 50% of patients (pts) with WM exhibit a normal karyotype using either conventional chromosome banding analysis (CBA) or FISH approach. However, CBA is a low resolution method and FISH only target previously described abnormalities. Comparative genomic hybridization (CGH) array delineated the minimal deleted region on 6q deletion, the most frequent aberration in WM, and pointed out the role of NFKB pathway key regulators genes. However, tumour suppressor gene or oncogene involved in WM physiopathology is not known, to date. Partial uniparental disomy (UPD) induced by copy neutral loss of heterozygosity (LOH) are important mechanisms for tumour suppressor gene inactivation or oncogene activation in cancer. Recently, Single Nucleotide Polymorphism (SNP) based arrays had been described as a powerful high resolution method allowing both the detection of LOH and copy number alteration (CNA) analysis in the same experiment, a major advantage over CGH array. Our aim was to identify new CNA and LOH involved in WM pathogenesis using SNP arrays analysis. Material and Methods. BM samples of 12 pts with WM (7 males, mean age: 67 years, 7 symptomatic pts) were analysed. DNA was extracted following CD19 B cells selection. Genome-Wide Human SNP Array 6.0 (Affymetrix chips) was used to detect both LOH and CNA. In 6 pts, paired samples (tumor / normal T lymphocytes) were used as an intra-individual reference to identify germline polymorphisms. In the 6 left pts, CN polymorphisms were excluded using copy-number variants database (http://projects.tcag.ca/variation) and the reference genotyping data from the HapMap project. Size, position and location of genes were identified with UCSC Genome Browser HG18 assembly, LOH and CNA using genotyping console 3.02 software (Affymetrix) and Partek genomic suite. FISH analysis was performed to detect deletion 6q; 13q14, 11q22, TP53, trisomy 4 and 12 chromosomal aberrations using Vysis probes. Results SNP array detected 35 CNA (23 gains, 12 losses) in 9/12 pts, with a mean of 2.9 abnormalities per patient (range 0 to 10). 91% of CNA were observed in symptomatic pts. 50% of CNA were < 5 mb in size, the lower limit of detection by CBA. 15 cryptic aberrations were identified among the somatic CNA spread over 7 pts. These 11 gains and 4 losses are identified by *. Deletions were observed on chromosomal segments 1p36*, 3p21.*, 6q16-q27 (2pts), 8p21*, 8p23-q24, 9p21*, 11q22 (2pts), 16p13*, 19p13*. One cryptic homozygous deletion in 13q14 was identified. Gains were observed on chromosomal segments 3p26, 4p16-q35 (3pts), 4q24*, 5p15 (2pts)*, 5p14, 7p22, 12p13-q24, 16p13*, 17q11*, 18p11-q23, 19p13-q13, Xq21* (5pts), Xq22*, Xq25*, Xp22-q28, Xq27-q28*(2pts). We also identified 42 LOH in 9/12 pts (mean of 3.5 per genome, range 0 to 7) dispersed on 17/23 chromosomes. The LOH observed in the absence of CNA loss are consistent with UPD in 35 cases (83%). 59% of LOH were observed in symptomatic pts. The interstitial or telomeric UPD regions varied in size, from 0.4 to 154mb. Two recurrent regions were identified on chromosome 4 (1.2mb involving genes DCLK2 and LRBA) and 13 (0.8mb involving genes KPNA3, ARL11 and SETDB2), probably subsequent to mitotic recombination. No UPD was observed on chromosome 6q. SNP arrays detected all FISH and CBA findings, except in one pt with subclonal deletion of TP53. Of the 5 pts with normal CBA or FISH, SNP array detected a CNA and 4 UPD in one patient and 7 UPD in an other patient. The remaining 3 cases (25%) had no detected CNA or LOH. Overall, SNP array detected a total of 77 genetic aberrations (CNA + LOH) (6.4 per genome, range 0 -16) in this cohort allowing the selection of 976 relevant genes (159 implicated in cellular growth and proliferation, 165 in cancer, 69 in cell cycle). Conclusion New cryptic clonal chromosomal lesions were detected using high resolution SNP array in this study. Several abnormalities were recurrent. We described a high frequency of UPD in WM, that might contribute to the inactivation of tumour suppressor genes by mutations or epigenetic alterations and subsequently to the regulation of tumor progression in WM. Further confirmation of the role of these candidate genes is underway. Disclosures: Leleu: Celgene: Research Funding; Janssen Cilag: Research Funding; Chugai: Research Funding.

Background. In AML, the frequency of TET2 mutations increases with age ranging from 1-5% in children to 6-27% in adults. These mutations have both been reported to exert unfavorable or neutral effects on patient outcomes. Besides mutations, fluctuation in TET2 expression has been evidenced in AML while we recently found that TET2 exon 2 skipping (TET2E2S) is associated with favorable outcome in adult AML (Mint-Mohamed et al., ASH 2014 and submitted). Here we investigated the effect TET2 expression and TET2E2S in homogeneously treated adult and pediatric AML patients with intensive chemotherapy (IC). Methods. Samples derived from 341 patients with available high-quality RNA and enrolled in the ELAM2 protocol (n=120) and the ALFA-0701 protocol (n=221).qRTPCR amplification of 2 conserved TET2 mRNA sequences was performed with GUS and ABL as reference genes. TET2 exon 2 skipping was quantified throughexon(E)-specific qRTPCR (qESPCR) amplification of E1E3 (spliced) and E2E3 (unspliced) TET2 isoforms. TET2E2S was quantified by calculating the ratio E1E3/E2E3. Patients were classified according to their standardized cytogenetic and molecular (NPM1 mutations, FLT3-ITD, CEBPA double mutations) risk subgroups. In children, treatment consisted in 1 induction course (AraC and mitoxantrone) and 3 consolidation courses (course 1 and 3 with high dose AraC); children with either intermediate or high-risk disease were candidates for hematopoietic stem cell transplant (HSCT) in complete remission (CR) after 1 to 2 consolidation courses. Adult patients were given a 3+7 induction course without (control group, n=110) or with fractionated intravenousgemetuzumab ozogamicin (GO) (n=111) as already described (Castaigne S, Lancet 2012; 379:1508-1516). Results. In pediatric AML, median age, median WBC, CR rate, relapse rate, median follow-up, 3-years EFS, DFS, and OS were 9.4 years, 19 G/L, 95%, 29%, 48 months, 60±5%, 63±6%, and 72±5%, respectively. Lower TET2 expression was associated with inferior EFS (3-years: 44±10%vs 69±6%, log-rank p = 0.01), DFS (51±8%vs 64±5% p=0.02), and OS (60±8%vs 83±3%, p=0,0003) while TET2E2S was associated with inferior EFS (44±8%vs 70±7%, p = 0.006) and OS (61±7%vs 81±4%, p=0.01). For TET2 expression, multivariate analysis identified WBC, age, risk group, and TET2 expression as independent prognostic factors for EFS and OS, and age, WBC, and TET2 expression for DFS (Table 1). For TET2 splicing, WBC and TET2E2S independently predicted EFS while TET2E2S alone independently predicted DFS (Table 1). In the low- (n=29) and intermediate-risk (n=62) groups, lower TET2 expression remained associated with shorter 3-years EFS. There was no significant correlation between TET2 expression and TET2E2S and TET2E2S retained its negative prognostic impact in the 85 patients with higher TET2 expression (3-years EFS 48±8vs 84±6%, p=0.004). In adult AML, median age, median WBC, CR rate, relapse rate, median follow-up, 3-years EFS, DFS, and OS were 62 years, 8 G/L, 76%, 59%, 47 months, 26±3%, 30±4%, and 44±4%, respectively. In the entire population, TET2 expression and splicing had no significant effect on outcome. In the 110 patients treated with IC alone, TET2 expression was associated with inferior OS (25±9vs 29±9%, p=0.02) while TET2E2S was associated with better EFS (17±5vs 7±6%, p=0.008), DFS (28±8vs 8±6%, p=0.02), and OS (46±8vs 29±8%, p=0.05). TET2E2S, age, and disease risk represented independent prognostic factors for EFS (Table 1). In the GO arm, TET2 expression had no significant effect on outcome whereas TET2E2S was associated with inferior EFS (27±5vs 60±12%, p=0.01), DFS (31±6vs 63±12%, p=0.05), and OS (39vs 72±11%, p=0.04). Conclusion. Inchildren, higher TET2 expression confers a favorable outcome whereas TET2E2S represents an independent unfavorable prognostic factor. In adult treated with IC alone, present results confirm the favorable prognostic impact of TET2E2S and the weak prognostic effect of TET2 expression. In contrast, in patients fulfilling the same inclusion criteria, TET2E2S loses its favorable prognostic effect and is even associated with unfavorable outcome when GO is associated to IC. While TET2 expression and splicing might represent useful biomarkers, the data highlight the strong difference in tumor biology between adult and pediatric AML and suggest distinct relationships between TET2E2S and the cytotoxic effect of GO plus IC versus IC alone. Disclosures Thomas: Pfizer: Consultancy.

The MILE (Microarray innovations in Leukemia) project has recently reported on the efficiency and accuracy of micro-array use in the diagnosis and sub-classification of leukemia. The MILE algorithm classifier of AML recognized 6 subgroups according to major cytogenetic findings:t(15;17), t(8;21),inv(16), MLL rearrangement, complex karyotype and a last group including normal cytogenetics and other patients. In the latter group, beside age, leucocytosis and antecedent of previous hemopathy, molecular alterations, especially of CEBPa, NPM1 and FLT3 mutations have been shown to be frequent and to have a major prognosis value. In this work, we have attempted to enhance the MILE classification of AML by examining data for a gene signature specific of each of these mutations. The study was made using the data generated from HG-U133 plus 2.0 Affymetrix arrays from the 79 AML samples included from the French study centers of MILE phase 1 trial. Among these patients with AML, according to the gold standard classification used in the MILE project were 2 t(8;21) AML, 2 inv(16) AML and the others belonging to intermediate cytogenetic group class MILE 13 (mainly patients with normal karyotype). Into the cohort, 7 samples harbored CEBPa mutations, 33 samples have NPM1 mutation and 26 samples have FLT3-ITD. Data generated from GEP profiles were combined to these molecular data to obtain predictor of mutation status. Predictors were based on differential gene expression signature using SVM methods. Differentially expressed genes sets were obtained using volcano plot with p value of 0.05 and fold change of 2. The CEBPa predictor shows a sensitivity of 100% in the detection of CEBPa mutation with a specificity of 81%, only six sample were misclassified among them 3 were non informative and 3 were called as CEBPa mutated (2 t(8;21) samples and one AML with MLL alteration), the efficiency of the predictor was 83%. The NPM1 predictor shows a sensitivity of 87.9% in the detection of NPM1 mutation and a specificity of 75%, 7 samples were misclassified, among them 3 NPM1 mutated samples (2 were called as wt and one assigned as non informative) and at the other hand 4 NPM1 wt samples were called as mutated), efficiency of this predictor was 81%. To note using the set of genes used for this predictor, non supervised hierarchical tree clearly shows the dichotomy of NPM1 mutated samples into two sub-clusters according to the FLT3 status. Results were weaker for the FLT3-ITD predictor with a sensitivity of 57% and a specificity of 70% with 29 samples misclassified (9 miscalled and 20 without answer).These results shows that the use of the data generated by the MILE in the diagnosis of leukemia can be used in the detection of CEBPa, NPM1 and FLT3 mutations in AML that have a major value in the prognosis of patients. The use of this 3 new predictors, in second line after the use of MILE algorithm to discriminate AML belonging to the intermediate cytogenetic groups clearly improve the use of microarrays in the diagnosis of AML.

Abstract 2701 The Wilms tumor 1 (WT1) gene, located at chromosome band 11p13, encodes a transcriptional regulator involved in normal hematopoietic development. The role of WT1 in acute myeloid leukemia (AML) has been underlined by the finding of WT1 overexpression in most AML cases and WT1 gene mutations in approximately 10% of AML patients. Recently, the minor allele of the silent Arg301 single nucleotide polymorphism (SNP) (rs16754), which is a 903A>G substitution in WT1 exon 7, was suggested to predict a favorable prognosis in adult patients with cytogenetically normal (CN) acute myeloid leukemia (AML). In contrast, no prognostic impact of this SNP was found in another study performed in pediatric AML. Our aim was to evaluate the frequency, the main associated features and the prognostic significance of WT1 SNP rs16754 in elderly patients with AML. Diagnostic bone marrow or peripheral blood samples were analyzed from 266 patients (age 50 to 70 years) with previously untreated de novo AML included in the French ALFA-9801 trial. WT1 SNP rs16754 and gene mutations of NPM1, CEBPA, FLT3 (internal tandem duplication, ITD), WT1 (exon 7 and 9), IDH1 and IDH2 (exon 4) were screened on genomic DNA by direct sequencing and fragment-length analysis for the detection of small insertion/deletion. The minor allele of WT1 SNP rs16754 was found in 85 (32%) out of 266 cases. Patients heterozygous or homozygous for minor allele (WT1AG or WT1GG) and patients homozygous for the major allele (WT1AA) did not significantly differ in terms of age, gender, white blood cell (WBC) count, FAB subtypes, cytogenetic categories, and gene mutations when those mutations are considered separately. However, among CN-AML, the frequency of the favorable genotype defined by the presence of NPM1 mutation or CEBPA mutation without neither FLT3-ITD nor IDH1 mutation was significantly lower in patients with at least one minor allele than in patients with two major alleles (5% vs 23% of CN-AML, p=0.02). In univariate analysis, complete remission rate was found similar between patients with at least one minor allele and patients with two major alleles (76% vs 75%, p=0.88). Patients with at least one minor allele tended to have a shorter median delay to relapse (7 vs 12.2 months, p=0.06) and had a significantly shorter overall survival compared to patients with the two major alleles (5-year OS, 14% vs 26%, p=0.02). In the subset of CN-AML (n=113), the presence of at least one minor allele was also associated with a shorter median delay to relapse (6.9 vs 12.5 months, p=0.02) but only a trend regarding overall survival was observed (5-year OS, 17% vs 30%, p=0.1). In multivariate analysis considering age, WBC count, cytogenetics (favorable, intermediate and unfavorable categories) as covariates, WT1 SNP rs16754 status was found to be an independent prognostic factor for relapse risk (HR 1.56, 95% CI 1.06 to 2.30, p=0.03) and overall survival (HR 1.54, 95% CI 1.08 to 2.20, p=0.02). Thus, in our cohort of older AML patients, the minor allele of WT1 SNP rs16754 appears to confer a relatively poor prognosis, which is in contradiction to what has been reported so far. Overall, our data suggest that WT1 SNP rs16754 is an interesting marker that may contribute to refine prognosis in AML, at least in this age group of patients. Further investigations are needed to clarify the relationship between WT1 SNP rs16754 and treatment outcome in AML and elucidate the biological effects of this SNP. Disclosures: No relevant conflicts of interest to declare.

Background: Relapse remains the main cause of treatment failure in patients with acute myeloid leukemia (AML) after allogeneic stem cell transplantation (allo-SCT). The Wilms’ tumor 1 gene (WT1) is reportedly over-expressed in 90% of patients with AML and thus could be useful for minimal residual disease (MRD). However, the clinical utility of WT1 monitoring is still controversial, especially after allo-SCT. The aim of this study was to evaluate the usefulness of WT1 expression as a relapse predictor marker before and after allo-SCT in 160 patients with AML treated in our unit between May 2005 and April 2014. Patients and methods: The study included 78 males and 82 females with a median age of 50 years (range, 18-68), who received an allo-SCT for AML. Hematologic status before transplant was complete remission (CR, n=127) and progressive disease (PD, n=33). Donors were HLA-identical sibling (n=57) and unrelated (n=103). Source of stem cells was bone marrow (BM; n=79), peripheral blood stem cell (n=39) and cord blood (n=9). Conditioning was myeloablative in 102 (64%) patients. WT1 expression was assessed by using quantitative real-time PCR and ELN criteria were used to define cut-offs for positive MRD in BM and peripheral blood. Data were retrospectively collected at diagnosis, after the first induction chemotherapy, within a month before transplant, at day 100 post-transplant and at any time point after allo-SCT. A total of 1793 PCR were performed (BM: n=736; peripheral blood: n=1057), including 741 after allo-SCT. Results: After a median follow-up of 35 months (4-111), the overall survival (OS), event-free survival (EFS), relapse and non-relapse mortality (NRM) were 58,8%, 55%, 30% and 13,1%, respectively. Thirty-six (22,5%) patients developed acute grade II-IV graft-versus-host disease (GVHD) and 55 (34,4%) chronic GVHD. MRD remained positive after induction chemotherapy in 47/103 (45,6%) evaluable patients, of which 24 reached CR1 before allo-SCT and 23 relapsed. Among MRD negative patients after induction, 25 (44,6%) relapsed before transplant. Post-induction and pre-transplant positive MRD were significantly associated with a progressive disease at transplant (p=0,27 and p Conclusion: WT1 expression, even after allo-SCT, can serve as a reliable marker for MRD in AML. Monitoring WT1 expression after allo-SCT could allow early detection of relapse and the subsequent selection of patients who may benefit from modulation of immunosuppression or donor lymphocyte infusion (DLI). Disclosures Quesnel: Oncoethix SA: Research Funding.

Among B-CLPD, characterized primarily by morphology and expression of cell surface markers, is is important to identify patients with CD5+ atypical B-CLL that is regarded as clinically more aggressive than typical B-CLL. However these cases are not well defined. To better characterize the genetic lesion observed in atypical B-CLL we have analysed a cohort of 40 patients by CGH microarrays.Study was made on 5 typical B-CLL and on 35 atypical B-CLL patients with either CD20dim or bright expression that do not express cyclin D1. All the patients with atypical B-CLL will be defined as the presence of an absolute B-CD5-positive lymphocytosis > 4 x 109/l and a RMH score < 4. DNA was extracted using QuiAmp kit according to manufacturer recommendations. 2μg of DNA was used in each hybridization experiment. DNA labeling was performed using Cy3dUTP and Cy5 dUTP respectively for control and tumoral sample respectively. CGH-arrays was performed using the 1 Mb Human genome micro-array from “arraygenomics” that contains 3400 BAC clones fully FISH mapped and end sequenced all printed in triplicate. Each experiment was made using two slides in dye swap method. Cy5 and Cy3 fluorescence intensities spot were quantified using Axon Scanner 4100 and Acuity Software. Data were imported into SpectralWare 2.0 software and Normalise Suite, version 2.0, Profiler from Squire lab. Results: Identification of known and previously uncharacterized copy number alterations (CNAs) in the a-B-CLL cells genome was made in all the cases. The CGH profiles revealed that a-B-CLL genome is highly rearranged harbouring large numbers of distinct copy-number aberrations (75 CNAs among 31 chromosomal regions were found). Some of these CNAs are recurrent across different samples, allowing the definition of minimal common regions (MCR) of amplification or deletion. The size of the CNAs was extremely variable from one Bac probe to complete chromosome gains or losses. Specifically, our dataset included the known gains of chromosome 12 (14 cases), and the known deletion at 11q23, 13q14.3, 17p region but also new region of interest as +3p, 3q22 to 3qter, 4pter to 4q35.2, 5p15, 6p25.3 to 6p22, 8q22 to 8q24, 15q15.3 to 15q26, 17q11 to 17qter, +18 and +19 for gains and 1p35,1, 1p33, 2q22.3, 3p26.3 to 3p21.3, 5q34, 6p25.3, 6q16, 6q25.3 to 6q27, 7q31.3 to 7q32.2, 8p23.3 to 8p12, 10q11.2 to 10q21.1, 10q21.3, 10q23.2 to 10q24, 11q22.3 to 11q24.2, 15q14 to 15q21, 16p11.2 to 16q21, 21q22.1 for the deletions.To further corroborate the above finding, we had performed conventional FISH analysis using known probes for del 13q, del 11q, trisomy 12, del 17p and correlate our results with conventional cytogenetic findings when they were available. In all the cases CGH-arrays findings were confirmed by fish analysis or karyotype. The above findings were confirmed also in few cases using the Agilent’s Human Genome CGH Microarray 44K that contains over 40,000 probes. Our first results confirmed BAC arrays results. CGH-arrays appears to be very informative to detect lesion in B-CLPD and show the high frequency of genetic lesions in a-CLL. The biological impact of this lesions by transcriptome analysis on the same sample and the prognosis impact is in progress.

Introduction AML is a heterogeneous group of neoplasm according to leukemogenic mechanisms and has variable response to treatment. In addition to age and WBC count at diagnosis, cytogenetic and molecular abnormalities are key factors to assess prognosis. In this work, we evaluated the number of adult AML cases diagnosed between 2009 and 2012 in the French Nord Pas de Calais and Picardie regions (North of France). The aim of the study was the collection of epidemiology data, treatments proposed and their results in real life. Matérials and Methods The Nord Pas de Calais and Picardie observatory contains data of all adult patients (>18 years) with AML (>20% marrow blasts) diagnosed in the oncology and/or hematology departments of the following participating Hospitals: Amiens (A), Arras (Ar), Boulogne (B), Dunkerque (D), Lens (L), Lille (CHRU et CH St Vincent (Vi)), Roubaix (R) et Valenciennes (V), whose overall catchment population is close to 6 million inhabitants , allowing a population based registry. A disadvantage of registry data is that investigators cannot control data collection and quality. In our observatory, data were obtained from the declaration of all the participants centralized with a web access reporting data base. The declaration of patients in the database was made both by clinicians, cytogeneticists and molecular biologists. The following data were entered:, disease course, WBC counts, bone marrow blast %, cytogenetics, immunophenotype, molecular biology, treatment received, outcome. To optimize data quality, each kind of data was entered respectively by clinicians, cytogeneticists and molecular biologists. Data were cross validated and the follow up data were updated every years during follow-up of the patients. Data was collected as of October 2013 Results Between January 2009 and December 2012, 899 patients were enrolled in the observatory, respectively 185 (A), 33 (Ar), 43 (B), 66 (D), 85 (L), 197 (CHRU), 57 (Vi), 98 (R) and 135 (V). Median age was 67 years (range 17 to 97), 54% of patients were older than 65 years, M/F was 1.12 (425 female (47%) et 474 male (53%)). Cytogenetic studies were performed in 87% of the patients, immunophenotype in 57%, molecular biology screening for FLT3 and NPM1 mutation in 61%. Intent of treatment was conventional chemotherapy (CTi) 50%, demethylating agent (DMT) 18% and best supportive care (BSC) 32% + Low dose cytarabine (LDAC). The overall median survival was 269 days (624d < 65 y and 115d for > 65 y) (fig1). 15% of the patients (132pts) were included in a clinical trial (90 % 65 y). Median overall survival of the patients included in a clinical trial and aged Of the 482 patients aged >65 y, MRC cytogenetic classification was unfavourable (139 pts; 29%), intermediate (209 pts; 43%), favourable (18 pts; 4%) and not done 116 pts (24%). Overall, in pts aged > 65, treatment was CTI 93 pts (19%), DMT 130 pts (27%), BSC 210 pts (44%), LDAC) 38 pts (8%), others 11 pts (2%). Among pts with unfavourable karyotype treatment was: Cti (14%), DMT (45%), LDAC (2%), BSC (38%) and others (1%). overall median survival in this unfavourable group was 222d with DMT vs 104d with LDAC vs 227d with CTI. In intermediate cytogenetics cases, treatment was Cti (24%), DMT (19%), LDAC (14%), BSC (42%) and other (1%), and median OS was 270d with DMT vs 189d with ARAC-LD vs 378d with Cti. Demethylating agent treatment results were different between MRC groups. In the intermediate MRC group of pts >65y, OS was better with CTI vs DMT (p:0.03), whereas in the unfavorable pts group aged >65y/o, DMT treatment seemed to be as effective than Cti (fig 2 and 3). Conclusion This registry study gives a real life idea of the therapeutic intent in AML patients, stressing in particular the small proportion of patients who can be included in clinical trials due to comorbidity and also the small number of available clinical trials in elderly patients, due to the current lack of promising drug apart from IC (among pts>65y/o only 13 pts have been included). Our observatory shows that in pts >65 y/o the intent of treatment should take account MRC classification, CTI being more effective that DMT or LDAC only in intermediate cases. Disclosures No relevant conflicts of interest to declare.

Background CEBPA mutations are found in approximately 10% of acute myeloid leukemia (AML). Recent studies revealed an association between CEBPA and GATA2 mutations. GATA2 gene encodes a transcription factor involved in hematopoiesis. In two recent studies (Grossmann et al., BJH 2013; Fasan et al., Leukemia 2013), GATA2 mutations appeared to be associated specifically with double CEBPA mutations and improved overall survival (OS). In contrast, another study failed to show any impact on OS (Green et al., BJH 2013). The aim of our study was to investigate the prognostic significance of GATA2 mutations in a large cohort of patients with CEBPA-mutated AML. Patients and methods We studied a cohort of 128 patients with CEBPA-mutated AML treated with intensive chemotherapy. The entire coding region of CEBPA, NPM1 exon 12, and GATA2 exons 4 to 6 (that encode the two multifunctional zinc finger domains of the protein) were screened on genomic DNA by PCR and direct Sanger sequencing. FLT3 internal tandem duplications (FLT3-ITD) were screened on genomic DNA by PCR and fragment analysis. Additionally, transcriptome analysis was performed with Affymetrix HG U133 Plus 2.0 array in pre-treatment samples from 72 patients for which RNA was available. Results Median age at AML diagnosis in the whole cohort was 50 years (range, 3-80). Almost all patients belonged to the intermediate cytogenetic risk-group (n=117), of which 90 had a normal karyotype. The remaining patients had favorable (n=1) or adverse cytogenetics (n=4). CEBPA mutations were distributed as follows: 29 single-mutated (sm) cases with N-terminal (N-ter) mutations, 12 sm cases with C-terminal (C-ter) mutations, 80 double-mutated (dm) cases with both N-ter and C-ter mutations, 2 cases with homozygous N-ter and 5 cases with homozygous C-ter mutations. GATA2 mutations were found in 29/128 patients (23%) and were significantly associated with CEBPA dm cases (4/41 sm vs 25/87 dm: Fisher's exact test p=0.022). NPM1 mutations were detected in 12 patients. They were specifically associated with CEBPA sm cases (12/40 sm vs 0/78 dm, p Conclusions This study confirmed the strong association between GATA2 mutations and CEBPA double mutations, in line with previous studies. However, we did not find any prognostic impact of GATA2 mutations in our cohort of CEBPA-mutated AML. Disclosures No relevant conflicts of interest to declare.

Background Acute myeloid leukemia (AML) with t(8;21) chromosomal translocation, leading to the RUNX1-RUNX1T1 fusion, belong to the favorable risk AML subset. However, relapse incidence may reach 30-40% in these patients. Minimal residual disease monitoring (MRD) based on the quantification of RUNX1-RUNX1T1 fusion transcript by real-time quantitative PCR (RQ-PCR) has been reported to be an independent prognostic factor for the risk of relapse. The specificity of the RUNX1-RUNX1T1 fusion and the high sensitivity of RQ-PCR techniques have made RUNX1-RUNX1T1 an ideal marker to assess treatment response in t(8;21) AML. Undetectable MRD could mean either that tumor cells persist in a latent state without RNA expression or that MRD level is below the sensitivity threshold. Studies in chronic myeloid leukemia showed that BCR-ABL DNA was still detectable in patients in long-term complete molecular response with undetectable BCR-ABL fusion transcript. Using a similar approach, we investigated the use of RUNX1-RUNX1T1 DNA as a MRD marker in t(8;21) AML, instead of RUNX1-RUNX1T1 mRNA. This approach allows linking results directly to the amount of leukemic cells, since each leukemic cell contains one copy of the RUNX1-RUNX1T1 sequence, while the level of RUNX1-RUNX1T1 mRNA may vary from a patient to another. Methods This study focuses on 17 patients with t(8;21) AML included in the CBF-2006 trial and for whom frozen material was available for further molecular analysis. Bone marrow and blood samples were collected at AML diagnosis and during follow-up, as defined in the CBF-2006 trial. Eight patients relapsed during follow-up and 9 were still in complete remission at the end of the study. Interestingly, 3 patients relapsed with a previously undetectable MRD (in blood and bone marrow samples). First, we identified the breakpoints in the RUNX1 and RUNX1T1 genes for each patient using long-range PCR approaches, coupled with next-generation sequencing (NGS) on Personal Genome Machine™ (PGM). The stability of the RUNX1-RUNX1T1 rearrangement at relapse was checked by Sanger sequencing. Then, we performed quantification of RUNX1-RUNX1T1 DNA by RQ-PCR using Taqman technology. For each patient, a primer pair and a probe were designed using the patient's unique RUNX1-RUNX1T1 breakpoint sequence. The forward and reverse primers were located in RUNX1 and RUNX1T1 genes, respectively, and the probe was located at the RUNX1-RUNX1T1 junction. Calibration curves were established using 10-fold dilutions of the diagnostic DNA of each patient in normal control DNA. Results were given as a ratio of chimeric DNA amount in the follow-up sample to chimeric DNA amount at diagnosis. Results Chromosomal breakpoints were located in RUNX1 intron 5 for all patients. RUNX1T1 breakpoints were located in intron 1b for 15 patients, and in intron 1a for 2 patients (Fig. 1). Quantification failed for 1 patient which was further leave up. Between 2 and 7 follow-up samples were studied for the other patients (median 4.5). DNA monitoring was strongly correlated with RNA monitoring (Fig. 2). Sensitivity threshold, determined by the lowest diagnostic sample dilution which gives a signal, was 10-5 for 7 patients, 10-4 for 6 patients, and only 5.10-4 for 3 patients. MRD was detectable in 31 samples and undetectable in 30 samples by both methods, whereas MRD was detectable only on RNA in 7 samples, probably because of a lack of sensitivity of the RQ-PCR assay. Interestingly, RUNX1T1-RUNX1 DNA was detected in 3 samples from 2 patients who relapsed and for whom RUNX1T1-RUNX1 transcript was undetectable, despite a good RNA quality. Conclusions Overall, RUNX1-RUNX1T1 MRD levels on DNA and RNA were quite similar. The level of mRNA expression did not seem to change during follow-up when compared with the amount of DNA. MRD monitoring on genomic DNA is a useful method, but with sensitivity variations depending on the patient's breakpoint sequence and the efficiency of the RQ-PCR assay. DNA has potential advantages: it is more stable than RNA and a best substrate for collection, processing, transport and storage. Additionally, interpretation of the results is easier because it is closely related to the number of leukemic cells. However, this method greatly increases complexity, time of implementation, and cost of monitoring MRD, which limits its interest in routine practice. Disclosures: No relevant conflicts of interest to declare.

Introduction: The impact of gene mutations, i.e. poor-prognosis FLT3 internal tandem duplications (ITDs) and good-prognosis NPM1 or CEBPA mutations, has been welldocumented in several recent reports dealing with younger patients with acute myeloid leukemia (AML). As these mutations were associated with cytogenetically normal (CN) AML, most of these reports focused on CN-AML patients. Both FLT3-ITD and NPM1 mutations were also associated with higher WBC. The objective of the present study was to evaluate the incidence, correlations, and prognostic value of these mutations in older patients with the disease. Methods: The French ALFA group has screened a total of 583 patients, including 333 younger patients (15–50 years) treated in the ALFA-9802 trial and 250 older patients (50–70 years) treated in the ALFA-9801 trial. The older ALFA-9801 trial included 468 patients with previously untreated de novo AML and studied the role of idarubicin (IDA) as compared to high-dose daunorubicin (DNR) as well as interleukine-2 as a maintenance therapy (C. Pautas et al. ASH 2007, abstract #162). Comparison between the 250 patients tested for mutations in that trial and the 218 patients not tested showed no differences in age, sex ratio, FAB classification, bone marrow blasts percentage, randomization arm, and performance status at entry in the study. There was, however, a higher rate of patients with intermediate cytogenetics (p=.01) or increased WBC (p=.01) in the former subgroup. Results: Median age of the 250 patients tested was 60 years. Cytogenetics was studied in 232 patients (12 favorable, 174 intermediate, 46 unfavorable). One hundred twenty-two patients (49%) had CN-AML. CR rate was 67.5% and estimated 4-year OS was 26% (95% CI, 20–33). Incidences of FLT3-ITD, NPM1, and CEBPA mutations were 37/250 (15%), 64/249 (26%), and 20/249 (8%), respectively. These incidences were very similar than in the younger ALFA-9802 population [50/329 (15%), 76/321 (24%), and 24/316 (8%), respectively]. In these older AML patients, the correlation between increased WBC and FLT3-ITD (p Conclusion: This study conducted in a large cohort of patients aged 50 to 70 years and prospectively treated in the same trial showed that gene mutational status still affect the outcome of older patients with AML. Mutation incidences are in the same range than in younger patients. Nevertheless, their impact on OS appeared to be less marked than in younger patients, probably due to the worse general outcome observed in these older patients.

Introduction Therapy decision in young adults with acute myeloid leukemia (AML) is mostly guided by pre-therapeutic risk assessment based on cytogenetical and molecular characterizations. Mutations in the nucleophosmin 1 (NPM1 m) gene represent one of the most common gene mutations, found in around one third of AML and associated with normal karyotype in 85% of cases. Because of their homogeneous mutation pattern and their clonal stability, NPM1 mutations are an effective tool for monitoring MRD. The aim of this study was to assess the prognostic impact of post-induction NPM1 MRD in homogeneously treated patients and to address the question of whether NPM1 MRD may be used as a predictive factor of allogeneic stem cell transplantation (SCT) benefit in this subgroup of patients. Materiel and methods Among 196 NPM1 m patients treated in the ALFA-0702 trial, 172 achieved complete remission (CR). A MRD was available in 152 patients, on peripheral blood (PB) in 135 patients and on bone marrow (BM) in 135 patients. MRD levels were reported as the normalized values of NPM1 m copy number/ABL copy number x 100 (%). Patients were also screened for FLT3-TKD, FLT3-ITD, CEBPA, DMT3A, IDH1, IDH2, WT1 and TET2 mutations. Patients that did not belong to ELN-favorable group were eligible for allogeneic SCT (N=71). Results The median follow-up was 3.5 years (95%CI: 3.1-3.9). The karyotype was normal in 122 patients (80.3%), and a FLT3-ITD was found in 59/150 (39.3%) with a median FLT3-ITD allelic ratio of 0.385 (range: 0.02-8.0). Fifty out of the 71 patients eligible for SCT were allografted in first CR. The 3-year CIR and OS censored at SCT were 29.3% (95%CI: 21.5-39.4) and 78.6% (95%CI: 69.3-85.5) respectively. At diagnosis, NPM1 baseline levels did not differ between PB and BM (p=.20). However, after induction, the median MRD log reduction was 4.5 in PB (range: 1.7-5.8) and 3.8 in BM (range: 0.01-5.5) (p Patients with a high WBC (>50 G/L) or FLT3-ITD had higher NPM1 baseline but also PB-MRD levels. However, no correlation between PB-MRD log reduction and age, WBC, karyotype, or any of the gene mutations was observed. Patients who did not achieve a 4-log reduction in NPM1 PB-MRD were exposed to a higher risk of relapse and had a shorter survival (3-year CIR censored at ASCT: 65.8% vs. 20.5%, SHR 5.83, 95%CI: 2.78-12.23, p Among the 59 patients with FLT3-ITD, a PB-MRD reduction < 4 log was also predictive of a higher risk of relapse and a shorter survival (SHR 3.74, 95%CI: 1.47-9.53, p=.006; HR 3.26, 95%CI: 1.43-7.44, p=.005) along with age, logWBC, and FLT3-ITD allelic ratio. In multivariable analysis of CIR and OS, only age, logWBC, and PB-MRD reduction < 4 log were significantly associated with a higher risk of relapse or death. Considering allogeneic SCT as a time-dependent variable, DFS but also OS were significantly improved by SCT in non-favorable ELN patients with PB-MRD reduction Conclusion Our study supports the prognostic significance of early NPM1 PB-MRD, independently of cytogenetical and molecular context. Moreover, NPM1 PB-MRD may be used as a predictive factor of allogeneic SCT benefit in non-favorable ELN population. Figure 1. DFS (A) and OS (B) according to PB-MRD log reduction and post-remission therapy (SCT) in ELN non-favorable patients (Simon Makuch plots). Figure 1. DFS (A) and OS (B) according to PB-MRD log reduction and post-remission therapy (SCT) in ELN non-favorable patients (Simon Makuch plots). Disclosures De Botton: Agios Pharmaceuticals: Research Funding. Dombret:Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.

ATP binding cassette (ABC) transporters are a superfamily of highly conserved membrane proteins that transport a wide variety of substrates across cell membranes and confer drug resistance against a wide range of chemotherapeutic agents. We recently found that WT1, which is regularly overexpressed in AML and interact with the splicing machinery, modifies the splicing of ABC transporters A2, A3, A5, and C2. For ABCA3, WT1 knock-down in three AML cell line coupled with Affymetrix HTA2 exon arrays analysis confirmed by exon-specific PCR revealed that WT1 influences the skipping of exon 19. ABCA3 belongs in the ABC subclass and induces a significant reduction in cytotoxicity observed following exposure to DNR, mitoxantrone, etoposide, Ara-C and vincristine. The ABCA3 domain encoded by exon 19 (amino acid 805-847) is localized at the junction of the first nucleotide-binding domain and the second transmembrane domain, and is involved in ATP hydrolysis. In silico, skipping of exon 19 deletes a sequence of 32 amino acids rich in positively charged residues and is thereby assumed to increase drug efflux through increased ATP hydrolysis. The effects of the skipping of exon 19 on chemoresistance and DNR efflux are currently investigated while for the present study, we hypothesized that skipping of exon 19 of ABCA3 might negatively influence outcome in AML patients. Analyzing 132 bone marrow AML samples harvested at diagnostic confirmed the statistically significant correlation between WT1 expression and ABCA3 splicing in vivo (p In order to confirm these results we investigated the prognostic impact of ABC A3 exon 19 skipping in a validation cohort of 108 additional AML case with normal cytogenetics. FLT3 internal tandem duplication (FLT3-ITD) and nucleophosmin (NPM1) exon-12 gene mutation were identified in 37 (34.3%) and 66 cases (61.1%). In the 86 patients treated with IC, the CR rate was 89,5% without significant difference between patients with or without exon 19 skipping. The relapse rate was higher in cases with exon 19 skipping (47,1% vs 23.5%) but the difference was not statistically significant. The 29 allografted patients were censored at the time of allograft. Median follow-up was 12 months. By univariate analysis ABC-A3 exon skipping of exon 19 significantly affected DFS (HR=2.03 (95% confidence interval: 1.22-3.39), p=0.007, and EFS (HR 1.62 (95% confidence interval: 1.06-2.48), p=0.027); higher the level of exon 19 skipping, poorer the outcome. In multivariate analysis, age, FLT3-ITD, ABC A3 exon 19 skipping but not NPM1 mutation were identified to be independent prognostic factors for EFS while FLT3-ITD and ABC A3 exon 19 skipping were independent prognostic factors for DFS. In conclusion ABC-A3 missplicing possesses a strong prognostic impact in AML indicating that besides whole gene transcription, quantitative analysis of alternative splicing might represent a promising tool for assessing AML aggressiveness at the time of diagnostic in patients with normal or abnormal karyotype. Disclosures Nicolini: Novartis: Consultancy.

Abstract 4171 NF1 acts as a tumor-suppressor gene by encoding neurofibromin1, a GTPase-activating protein (GAP) inhibiting Ras signaling pathway. Germline mutations or microdeletions of NF1 are responsible for neurofibromatosis type 1, and the somatic loss of the remained wild-type allele lead to malignant tumors or juvenile myelomonocytic leukemia (JMML). Furthermore, several studies revealed heterozygous somatic deletions of the 17q11.2 region including NF1 in adult myeloid malignancies. The reported frequencies of this abnormality varied between 2.6% and 11% in AML and this variation can be attributable to heterogeneity or size of the analysed cohorts. Previously, we analyzed 131 de novo AML cases (AML3 excluded) by Agilent™ 105K microarrays. 6/131 cases (4.6%) showed somatic deletions in 17q11.2, including a small minimal deleted region of 300 kb comprising the entire NF1 gene. To further investigate the incidence of NF1 deletion in de novo AML, 354 additional patients were therefore screened for the deletion by quantitative real-time PCR (Primers and TaqMan-based probe Hs 01778367_cn from Applied Biosystems), and FISH (NF1/MPO probe KBI-40144 from Kreatech) was performed to confirm the loss of NF1 copy number. Altogether, heterozygous NF1 deletion was observed in 17/485 (3.5%) de novo AML. Clinico-biological data were available from 14 NF1 deleted patients and 380 non-deleted patients included in the ALFA-9801 and 9802 French Trials. There were no significant differences between the 2 groups in age, sex ratio, leukocytosis, FAB classification of AML, mutational status of FLT3, NPM1, CEBPα and IDH. Interestingly, NF1 deletion was significantly correlated with unfavourable cytogenetic (50% vs 18%, p=0.008) and especially with monosomal karyotype (29% vs 9%, p=0.03). However, no statistical significant differences were observed for complete remission rate, relapse risk 3 years after diagnosis and 3-years overall survival. Screening for bi-allelic inactivation by sequencing the remained allele in NF1 deleted patients is in progress. We next evaluated NF1 gene expression for 93 patients of our cohort (3 with NF1 deletion and 90 without) by Affymetrix U133 Plus 2.0 microarrays. The 3 NF1 deleted patients revealed a significant reduced mean of NF1 expression level. Interestingly, about 10% of the NF1 non-deleted patients presented a similar decrease in NF1 expression rate. This suggests that mechanisms for transcriptional regulation (such as mutations or epigenetic silencing of NF1) may also contribute to AML pathogenesis. In conclusion, NF1 deletions occur in only 3.5% of de novo AML and are associated with unfavourable cytogenetic. This relatively low frequency of NF1 deletion can however be counterbalanced by others alterations acting at the transcriptional level and this remains to be investigated. Disclosures: No relevant conflicts of interest to declare.

Abstract 2506 Introduction: The development of massively parallel sequencing technologies has led to the identification of somatic DNA methyltransferase 3A (DNMT3A) gene mutations in acute myeloid leukemia (AML), with the highest frequency being found in cytogenetically normal (CN) AML. DNMT3A mutations have been suggested to predict poor clinical outcome in AML, but only few data are available on their prognostic significance within CN-AML. The aim of this study was to determine the frequency, the main associated features, and the prognostic significance of DNMT3A mutations in CN-AML. Patients and methods: This retrospective study was performed in 123 young adult patients (16–60 years) with previously untreated primary CN-AML and enrolled on two concomitant protocols of the Acute French Leukemia Association (ALFA), the ALFA-9801 and ALFA-9802 trials. DNMT3A mutations were screened on genomic DNA by PCR and direct Sanger sequencing. We focused our screening on the 3 conserved domains of DNMT3A (the proline-tryptophane-tryptophane-proline (PWWP) domain, the ADD-type zinc finger domain, and the C5-methyltransferase domain), corresponding to exons 8–9 and 11–23. The patients were also assessed for the presence of FLT3 internal tandem duplication (FLT3-ITD), FLT3 tyrosine kinase domain (FLT3-TKD), NPM1, CEBPA, WT1, IDH1, and IDH2 mutations. Results: Thirty-eight DNMT3A mutations were identified in 36 of the 123 (29%) patients. These alterations consisted of 36 nucleotide substitutions and 2 frameshift deletions. Thirty out of 36 (83%) nucleotide substitutions affected the amino acid residue R882 (R882H, n = 21; R882C, n = 7; R882P, n = 2), 5 represented other missense alterations, and 1 was a nonsense mutation. Two patients exhibited 2 heterozygous missense mutations in different exons, and one patient had a homozygous missense mutation. DNMT3A mutated and wild-type cases did not differ in terms of age, gender, and white blood cell (WBC) count at presentation. DNMT3A mutations were strongly associated with the French-American-British (FAB) subtypes M4/M5 (P =.0002) and the presence of NPM1 mutations (P =.0006), and tended to often co-occur with IDH1R132 mutations (P = .09). In the entire cohort, complete remission rate was found lower in DNMT3A mutated patients than in DNMT3A wild-type patients, but without reaching statistical significance (80% vs 90%, P =.24). DNMT3A mutated patients had a shorter event-free survival (5-year EFS: 13% vs 32%, P =.02) and overall survival (5-year OS: 23% vs 45%, P =.02) compared to DNMT3A wild-type patients. We next performed subgroup analysis according to the NPM1/FLT3-ITD genotypes. In patients with the non-favorable genotypes, that is NPM1 mutated/FLT3-ITD positive, NPM1 wild-type/FLT3-ITD positive, NPM1 wild-type/FLT3-ITD negative (n = 86), 18 (21%) had a concomitant DNMT3A mutation. In this high-risk subgroup of CN-AML, DNMT3A mutations conferred a worse clinical outcome (5-year EFS: 0% vs 23%, P =.02; 5-year OS: 14% vs 37%, P =.06). In patients with the favorable genotype NPM1 mutated/FLT3-ITD negative (n = 37), 18 (49%) were found to display a concomitant DNMT3A mutation. Within this favorable subgroup, patients carrying a DNMT3A mutation had a significantly inferior EFS and OS compared to DNMT3A wild-type patients (5-year EFS: 25% vs 65%, P =.01; 5-year OS: 29% vs 76%, P =.02). Furthermore, in multivariate analysis including age, WBC count, NPM1/FLT3-ITD genotypes, and DNMT3A mutational status, the presence of a DNMT3A mutation remained an independent adverse prognostic factor for EFS (hazard ratio = 2.29; 95% CI, 1.42 to 3.70; P =.0007) and OS (hazard ratio = 2.34; 95% CI, 1.37 to 4.00; P =.002). Conclusion: DNMT3A mutations are one of the most common gene mutations in CN-AML and independently predict poor clinical outcome. Testing for DNMT3A mutations could help further improve risk stratification in CN-AML. Disclosures: No relevant conflicts of interest to declare.

Background: Acute myeloid leukemia (AML) is a heterogeneous disease even within the same genetic subgroup, e.g. within the NPM1 mutated subgroup, some patients have additional mutations in FLT3, IDH1/2, DNMT3A or TET2. Recent reports have shown that minimal residual disease (MRD) in AML, during or after treatment has prognostic impact. However, the clinical assessment of MRD in AML faces many questions. Firstly, which of the potential molecular and/or cellular markers should be included? Secondly, what type of biological sample should be analyzed? Thirdly, what is the sensitivity threshold to set and what are the relevant time frames for the MRD evaluation. The current technology development next-generation sequencing (NGS), opens new perspectives for monitoring MRD in AML, with the opportunity to expand and multiplex various markers. In this context, we tested the NGS technology for MRD monitoring IDH1/2 and DNMT3A in AML cases with mutated NPM1. We validated our results against MRD evaluated by RTqPCR. Patients and methods: In this study, we included 94 patients (pts) from the ALFA 0701 trial. NPM1 monitoring by RTqPCR was performed as previously described (Lambert et al, Oncotarget july 2014). IDH1/2 and DNMT3A monitoring was done by NGS using the Ion Torrent Proton instrument. In order to obtain very high coverage (approximately 2 Millions reads by sample), 24 samples were analyzed per run. Bioinformatic analysis was performed according to our previous work (Boyer et al, Am JHematol 2013). Results. 94 samples from 31 NPM1 mutated pts were analyzed. (17pts were positive for IDH1/2 and 15 were positive for DNMT3A). Sequencing data showed an excellent depth with a median of 2012459 reads (range 102,657 to 5,160,118 reads) and 966,298 reads (range 565,152 to 2,700,349 reads) for IDH1/2 and DNMT3A, respectively. Unfortunately, despite of this excellent coverage, a median of 520 reads were found positive in the negative controls due to a multi steps cross contaminations (OT2, Clonal amplification, purification). Thus, limiting the sensitivity to 0.07% (0.001-0.097%, p The correlation with NPM1 MRD was 78% (r = 0.68183, p For the 17 IDH1/2 positive patients, we found concordant MRD results between IDH1/2 mutation and NPM1 RTqPCR in 13 cases: (4 pos/pos), which was associated to disease relapse, (9 neg/neg), which was associated to maintenance of CR1. For the four remaining patients, we have observed a discrepancy between some NPM1 results and IDH1/2 results. Indeed, these 4 patients presented one or more MRD negative points using the NPM1 marker, while rates of the IDH1/2 marker ranged from 0.1% to 47%. Three of them, have relapsed after 504,395,158 days, and all of them with the same NPM1 mutation detected at diagnosis. The one patient, who has not relapsed, did evolve to a myelodysplastic syndrome which was NPM1 negative. For the 15 DNMT3A positive patients, we found co-occurrence of the NPM1 mutation and IDH1/2 mutations in 9 cases (8 pos/pos), which was associated to disease relapse and we found concordant MRD results between DNMT3A mutation and NPM1 RTqPCR, (1 neg/neg), which was associated to maintenance of CR1. For the six remaining patients, we have observed a discrepancy between NPM1 results and DNMT3A results. Indeed, those 6 patients presented a great majority of negative results using MRD monitoring by NPM1, while rates of MRD by DNMT3A ranged from 5% to 45%. All of these 6 pts are still in first CR after a median follow up of 4 years. Bone marrow and PBMC were analyzed for 3 of them. A positive result was observed in all cell fractions except for the T cells compartment. Conclusion. Monitoring of MRD by NGS mutation detection in two genes IDH1/2 is feasible and valuable, because it enabled us to predict relapse in additional patients, with an area under the curve of 0.7971 (95% CI: 0.6693 - 0.9250), and 100% if we include the patient who evolved to MDS. Furthermore, this category of patients may benefit from new targeted therapies such as inhibitors of mutated IDH proteins. On the contrary, DNMT3A is not a good marker for MRD monitoring, because we detected persistence of a preleukemic clone in 40% of patients remaining in CR, reflecting the molecular heterogeneity clonal hematopoiesis. Disclosures No relevant conflicts of interest to declare.