ALBERT

Description: Key Points AML patients with isolated trisomy 13 have a very poor clinical outcome Isolated trisomy 13 in AML is associated with a high frequency of mutations in SRSF2 (81%) and RUNX1 (75%)

Description: It is increasingly recognized that the tumor microenvironment plays a pivotal role in cancer initiation and progression. In mouse models it was shown that a genetically altered bone marrow (BM) micro milieu was sufficient to induce leukemia (Raaijmakers, Nature 2010); however, the pathogenic role and contribution of the BM stroma in leukemia initiation and during disease progression warrants further investigation. To address this, we have performed gene expression, methylation, RNAseq, whole exome sequencing (WES) in BM mesenchymal stroma cells (BM-MSC) and leukemic cells from AML patients (pts) to unravel underlying molecular alterations. We collected BM hematopoietic cells (BM-HC) as well as plastic-adherent BM-MSC from aspirates from AML pts and healthy donors (HD). BM-MSC were expanded to passage 4 and defined as CD73+/CD105+/CD271+/low/CD45-/CD33-. We investigated gene expression profiles (Affymetrix) of BM-MSC from newly diagnosed AML pts (n=20) and compared these to BM-MSC from HD (n=4). BM-MSC from AML pts displayed an altered expression signature with 191and 175genesbeingsignificantly 2-fold over- and under-expressed. KEGG analysis of differentially expressed genes in BM-MSC from AML pts exhibited enrichment for TGF-ß signalling, whereas downregulated genes were enriched for cytokine receptor interactions. Several of these candidates were validated in a larger set of BM-MSC samples by RT-PCR. One putative stroma-leukemia interaction molecule, lumican (LUM) was highly overexpressed in BM-MSC (n=60) from AML pts compared to HD (n=5; p value =0.019) indicating that LUM may affect the BM niche in AML. To explore the altered expression pattern in AML BM-MSC compared to HD BM-MSC, global methylation analyses (Illumina Infinium HumanMethylation 450 bead chip arrays) were performed in 5 AML pts where we had collected BM-HC and BM-MSC at 3 sequential time points [initial diagnosis (ID), remission (CR), relapse (REL); n=30] as well as in BM-HC and BM-MSC from HDs (n=6). A significantly different methylation profile was evident comparing AML BM-HC to the corresponding AML BM-MSC samples, the latter showing a homogenous pattern during the course of disease. When AML BM-MSC were compared to a set of HD BM-MSC, we identified 2416 differentially methylated CpG sites (p value

Description: In acute myeloid leukemia (AML), DNA methylation is frequently altered and epigenetic regulators are commonly mutated. Here, we describe the effects of mutations in commonly mutated genes (NPM1, FLT3-ITD, DNMT3A, IDH1, IDH2, TET2, and WT1), including epigenetic regulators, on DNA methylation profiles and differential gene expression. Moreover, we show that subgroups of epigenetically homogeneous AML patients differ significantly in clinical outcome. We have characterized 212 AML patients treated on consecutive trials of the AMLCG study group using Illumina Infinium MethylationEPIC BeadChips, gene panel sequencing, and transcriptome sequencing. In order to detect differentially methylated CpG sites (dmCpGs) according to a particular mutation, we have selected sets of mutated and control samples for each gene of interest individually. We excluded samples with subclonal variants only and selected control samples based on matching karyotype and matching pattern of co-mutations. Mutations at the R882 hotspot of DNMT3A result in global hypomethylation while alterations of IDH1, IDH2, TET2, or WT1 lead to global hypermethylation. Still, subsets of dmCpGs are hypermethylated in DNMT3A-R882+ AML as well as hypomethylated in IDH1+, IDH2+, TET2+, and WT1+ AML. NPM1 mutations result both in hypo- and hypermethylation, while based on FLT3-ITD status, we could not detect significant changes in DNA methylation. Of note, we observed wildtype samples with a methylation profile highly matching that of mutated samples for most comparisons, suggesting alternative mechanisms. Moreover, mutations in IDH1, IDH2, TET2, and WT1 show substantial overlaps in dmCpGs, which is in line with their reported function, while the overlap with DNMT3A-R882 is rather small. Of note, we also detected overlaps in gene expression profiles by comparing test and control samples, in particular between AML with IDH1, IDH2, or WT1 mutations. The proto-oncogenes FOSB, FOSL2, and JUN are differentially expressed in IDH1+ AML, while in IDH2+ and WT1+ AML, members of the RPL and RPS gene families of ribosomal proteins are deregulated, known to alter FOS and JUN function. Unsupervised hierarchical clustering of all samples in our cohort results in two highly distinct epigenetic subgroups, each with three subclusters (Figure 1A). Of note, clusters are associated with distinct mutations. Most AML samples in clusters 1, 2, or 3 are mutated in NPM1, while clusters 4, 5, and 6 are mostly NPM1 wildtype. Still, the genetic profiles of subclusters differ based on the presence of mutations in IDH1/IDH2/TET2 (clusters 1 and 4), DNMT3A (cluster 2), and DNMT3A-R882/WT1 (cluster 3). Clusters 5 and 6 show only few mutations in DNMT3A, IDH1, IDH2, TET2, or WT1. Mutations in FLT3 are not associated with any cluster. Of note, epigenetic subgroups are also associated with differences in overall survival (OS) and event free survival (EFS) (Figure 1B). Clusters 1, 3, and 5 show significantly better outcome (median OS: 1113, 1046, and 1054 days; median EFS 513, 374, and 305 days) as compared to clusters 4 and 6 (median OS: 378 and 296 days; median EFS 103 and 70 days; p

Description: A complex karyotype, detected in approximately 10%-15% of patients with myelodysplastic syndrome (MDS), is associated with a very short median survival and a high risk of transformation into AML. The most frequent chromosome aberrations in complex karyotypes are deletion of 5q (del(5q)) and deletion of 17p (del(17p)) harboring the tumor suppressor gene TP53. It is still unclear, how complex karyotypes develop. We have identified an unbalanced translocation der(5)t(5;17)(q11∼q13;q11∼q13) in 199 patients with MDS or secondary AML after MDS. The cohort consists of 111 male (56 %) and 88 (44 %) female patients between 4 and 91 years of age (median age 68 years). In order to better understand the underlying pathomechanism of this aberration, we have investigated these cases in greater detail. For all patients, we performed cytogenetic banding analysis, fluorescence in-situ hybridization (FISH) and, in about one third, multicolor-FISH. In all patients, a complex aberrant karyotype with a median of 7 aberrations was observed, indicating high chromosomal instability. In 30 patients, clonal evolution was identified. To identify the breakpoints in 5q and 17q more precisely, array-CGH was performed in 7 patients. The breakpoints on 5q and 17p were located between the centromere of chromosome 5 and 5p15 and between the centromere of chromosome 17 and 17q22, respectively. The breakpoints were in gene-poor regions, suggesting that no fusion genes would result from these rearrangements. Notably, the breakpoints were all very close to the centromeric region and heterochromatin. It is known that an altered methylation of heterochromatic regions plays an important role in tumor development. Therefore, alterations of DNA methylation or histone modifications may be involved in the generation of the unbalanced translocation t(5;17). Using whole exome sequencing, we sought to define the mutational spectrum of complex karyotypes with t(5;17). In one patient we were able to analyse bone marrow cells from different time points: complex karyotype at diagnosis, complete remission and relapse with complex karyotype again. As possible candidate genes for driver mutations we identified mutations in the genes NF1, ETV6 (TEL) and KMT2C (MLL3). Of note, in this patient the allele frequencies of mutations affecting NF1 and KMT2C (MLL3) increased during the course of the disease, whereas the ETV6 (TEL) mutation found at diagnosis was lost at relapse indicating clonal evolution. Especially the identification of a mutation in NF1, a negative regulator of the RAS pathway, is of great significance. NF1 is encoded on 17q11.2. A possible underlying mechanism could be a downregulation of NF1 by a mutation of one allele and by a deletion evolved from the unbalanced translocation t(5;17) of the second allele. These data provide further evidence that the inactivation of NF1 seems to play an important role in clonal evolution and leukemic progression. Disclosures: No relevant conflicts of interest to declare.

Description: Acute lymphoblastic leukemia (ALL) consists of genetically heterogeneous cell subpopulations, but little is known about how genetic differences lead to functional differences between the clones. Of major clinical importance, aggressive, treatment-resistant and putatively relapse-inducing subclones need to be identified and require effective eradication by treatment. The most aggressive subpopulation likely determines prognosis and outcome in each patient. We aimed at characterizing on a functional as well as on a genetic level single stem cell clones derived from patients' samples growing in mice and to combine the results of both levels in order to learn which genetic characteristics are associated with adverse functional behavior. We transplanted primary tumor cells from a 5-year old girl with hyperdiploid ALL involving a trisomic X chromosome at first relapse into severely immune-compromised mice and lentivirally modified them to express the fluorochromes red, blue and green at different amounts and combinations (RGB marking, Weber et al., 2012). Eight single stem cell clones were generated by limiting dilution transplantation and their uniqueness was verified by ligation-mediated (LM) PCR. We functionally compared the single stem cell clones between each other by re-mixing them in a single mouse for in vivo assays; analysis was performed one-by-one for each clone by flow cytometry where they could be distinguished from each other using their unique color codes. Clones showed clear differences in proliferation rate with faster and slower growing clones, independently whether 2 or 5 clones were mixed. When mice harboring clone mixtures were treated with conventional chemotherapy, clonal composition changed markedly and resistant clones overgrew sensitive clones indicating selective clonal responses and clonal advantage. A clone which showed especially slow growth in vivo was most resistant to in vivo treatment with Glucocorticoids. The slowly proliferating, Glucocorticoid-resistant clone had lost the additional X chromosome, which was present in all other clones and the bulk and showed a distinct DNA-methylation pattern analyzed by 450K arrays (illumina). In exome analysis, the clone showed 11 unique alterations including a single nucleotide variant in the oncogene USP6. We are currently performing RNA sequencing analysis to assess the differential gene expression in the clones. Taken together, genetic multicolor marking PDX ALL cells in the individualized xenograft mouse model allowed generating viable single cell clones for genetic functional characterization in vivo. Within the heterogeneous tumor bulk, an subclone existed which showed slow tumor growth and drug resistance which was associated with distinct genetic characteristics. Our studies allow the challenging functional characterization of subclones in vivo in order to develop efficient novel treatment approaches to eliminate aggressive stem cell clones in ALL. Weber K, Thomaschewski M, Benten D, Fehse B., RGB marking with lentiviral vectors for multicolor clonal cell tracking. Nat Protoc. 2012 Apr 5;7(5):839-49. doi: 10.1038/nprot.2012.026. Disclosures No relevant conflicts of interest to declare.

Description: Cytogenetically normal acute myeloid leukemia (CN-AML) is a heterogeneous disease with regard to genetic alterations and clinical outcome. Recent sequencing studies categorized the growing number of recurrently mutated genes into different functional groups, e.g. myeloid transcription factors, tumor suppressors, signal transducers, chromatin modifiers, cohesin-complex and spliceosome-complex. We set out to characterize mutations in genes linked to epigenetic regulation during the progression of CN-AML. Besides genes directly involved in chromatin modification (i.e. DNMT3A, TET2, MLL, ASXL1, KDM6A, KDM2A, NSD1 and EZH2), we also studied mutations in WT1 and IDH1/2 since they are known to inhibit TET2 function. Targeted sequencing of 46 genes related to leukemia (mean coverage 〉500x) was performed on matched diagnostic, remission and relapse samples of 50 patients with CN-AML (median age: 66, range: 21-89). We called somatic variants at diagnosis or at relapse and filtered for mutations with translational consequences, excluding known error-prone genes and common germline polymorphisms (dbSNP 138; MAF〉=1%). At diagnosis, 36/50 patients (72%) carried a total of 48 mutations in epigenetic regulators (Figure 1). The majority of patients harbored a single mutation affecting this functional group, while 2 or 3 mutations were observed in 9 and 1 patient(s), respectively. The median variant allele frequency (VAF) of the mutations was 42% (range: 22-98%), indicating that mutations in epigenetic regulators are early events and are present in the founding clone. Of the 48 mutations detected at diagnosis, only 2 were lost at relapse, highlighting the stability of these lesions during disease progression. Moreover, in 12/50 patients (24%), mutations in epigenetic regulators were acquired at relapse. All but one of these patients already had a mutation in another epigenetic regulator at diagnosis. We did not identify patients who acquired DNMT3A, TET2 or ASXL1 mutations during disease progression. However, mutations in WT1, IDH1, and KDM6A were gained in several patients at relapse. In 4/13 cases, the gained mutations were already detectable at low levels at diagnosis (median VAF: 2.9%, range: 0.3-6%, mean coverage at the investigated sites: 629x, range: 85-1625x). We also evaluated the presence of these mutations in remission: In 18 out of 36 (50%) patients, some of the mutations affecting DNMT3A (n=14), TET2 (n=3) or IDH2 (n=2) were present at a VAF 〉5% (median: 22%, range: 9-75%) in cytomorphologically defined complete remission, suggesting the persistence of pre-leukemic clones with limited response to chemotherapy. Longer relapse-free survival was observed in patients with DNMT3A mutations that did not persist at remission (np-DNMT3A) in comparison to patients with persisting DNMT3A mutations (p-DNMT3A). Remarkably, the latter group was enriched for patients that also harbored FLT3 internal tandem duplications (ITDs) (10/14 versus 1/8; Fisher's exact test, p=0.02). The vast majority of p-DNMT3A showed alterations of R882, whereas mutations at other positions of DNMT3A tended to be undetectable in remission. When including the NPM1 status, only 1/8 patient with np-DNMT3A was triple mutated, compared to 11/14 patients with p-DNMT3A, suggesting that co-occurrence of DNMT3A, FLT3- ITD and NPM1 c is associated with p-DNMT3A (p=0.006). In summary, we show that a high proportion of patients (72%) with relapsing CN-AML is affected by mutations in genes linked to epigenetic regulation. The stability of these mutations between diagnosis and relapse in combination with their acquisition during disease progression, as well as the frequent persistence of DNMT3A, TET2 and IDH2 mutations during remission underscore the necessity for new therapeutic approaches. The striking association of DNMT3A R882 mutations with NPM1 c and FLT3 -ITD suggest a unique mechanism of oncogenic collaboration. Persistence of DNMT3A R882 mutations may indicate a fertile ground for relapse. Further studies will be required to clarify whether the actual relapse arises from a preleukemic clone harboring only the founder mutation or from residual leukemia cells containing several genetic lesions. Disclosures No relevant conflicts of interest to declare.

Description: Acute lymphoblastic leukemia (ALL) is known to consist of several clones that might have different chromosomal, genetic or epigenetic aberrations. However, little is known about functional diversity in these different clones. In some patients, cells cannot be eradicated by standard therapy regimens, and aggressive or otherwise unfavorable clones might survive, eventually resulting in relapse and a poor prognosis of the patients. Here, we asked whether genetically distinct clones of ALL from a single patient would show a functionally distinct response towards drug treatment in vivo. As technical approach, we genetically engineered primary patients' ALL cells growing in immuno-compromized NSG mice as patient derived xenograft (PDX) cells by lentiviral transduction. ALL PDX cells were red-green-blue (RGB) color marked in order to discriminate several differently colored cell populations in the same mouse in functional in vivo experiments. ALL PDX cells further expressed luciferase for bioluminescence in vivo imaging (BLI) for sensitive and reliable monitoring of disease burden. Limiting dilution transplantation of RGB marked PDX cells transplanted into groups of mice allowed generating individually marked single cell clones which were discriminated by flow cytometry. Populations expressing a distinct color were sorted and analyzed by ligation mediated PCR to verify distinct integration of lentiviral inserts to prove single cell clone (SCC) origin of the population. In sum, eight distinct SCCs could be generated and were used for functional and -OMICs approaches. Targeted resequencing of the eight SCCs and the bulk cells revealed that all samples had mutations in CSMD1 and HERC1 with variant allele frequencies (VAF) of 0.5, indicating that these mutations represent the founding clone. However, we also found mutations that were only present in single samples: FAT1 and STAG2 mutations were found in SCC 3, whereas CSMD1 and USP6 mutations were found in SCC 6. Whole exome sequencing revealed SCC specific patterns, identifying SCC 6 being the clone furthest away from the bulk population. As the patient showed a high hyperdiploidy (+6,+13,+14,+17,+18,+21,+22,+X), we tested SCC and bulk cells by fluorescence in situ hybridization (FISH) and found that both the bulk sample and the SCCs consisted mainly of cells harboring three X chromosomes and to a minor proportion (between 2% and 20%) of cells harboring two X chromosomes. Only SCC 6 consisted exclusively of cells harboring two X chromosomes. Additionally, this SCC showed a distinct DNA-methylation pattern analyzed by 450K arrays (illumina). To analyze if the chromosomal, genetic and epigenetic differences also resulted in functional diversity, we first performed a competitive transplantation assay, injecting a mixture of five SCCs in the same ratio (20% each) into single mice. After 42 days when overt leukemia had established in the mice, cells were re-isolated and proportion of SCCs reanalyzed according to their specific color. Interestingly, SCC 5 (25%) and 7 (36%) had a clear growth advantage over SCCs 1 (14%), 6 (13%) and 8 (12%). The same pattern could be overserved if only SCC 5 (50% in, 92% out) and SCC 6 (50% in, 8% out) were transplanted. Next, response towards chemotherapeutic drugs was assessed. In vitro, SCC 6 was much more resistant towards the glucocorticoids prednisolon and dexamethasone (Dexa) compared to all other SCCs and bulk cells. Cells of SCC 5 and SCC 6 were mixed in equal amounts and transplanted into mice. Four days after transplantation, mice were randomized and treated with PBS or Dexa (2 or 8 mg/kg i.p., 5 days a week, 5 weeks). BLI showed a clear response towards therapy of the entire tumor. After 61 days, control treated mice showed again an outgrowth of SCC 5 (83% vs. 17% SCC 6), while Dexa treated animals showed the opposite pattern (Dexa 2 mg/kg: SCC 6 35%; Dexa 8 mg/kg: SCC 6 59%) indicating that SCC 6 was more resistant towards Dexa treatment in vivo. Taken together, our results clearly show that within a single ALL patient, genetically and functionally distinct subpopulations exist. Combining PDX model with genetic marking of the cells enables us to in-depth analyze SCCs of a single patient sample and eventually identify adverse prognostic markers. Disclosures No relevant conflicts of interest to declare.

Description: Exome sequencing is widely used and established to detect tumor-specific sequence variants such as point mutations and small insertions/deletions. Beyond single nucleotide resolution, sequencing data can also be used to identify changes in sequence coverage between samples enabling the detection of copy number alterations (CNAs). Somatic CNAs represent gain or loss of genomic material in tumor cells like aneuploidies (e.g. monosomies and trisomies), duplications, or deletions. In order to test the feasibility of somatic CNA detection from exome data, we analyzed 13 acute myeloid leukemia (AML) patients with known cytogenetic alterations detected at diagnosis (n=8) and/or at relapse (n=11). Corresponding remission exomes from all patients were available as germline controls resulting in 19 comparisons of paired leukemia and remission exome data sets. Exome sequencing was performed on a HiSeq 2500 instrument (Illumina) with mean target coverage of 〉100x. Exons with divergent coverage were detected using a linear regression model on mean exon coverage, and CNAs were called by an exact segmentation algorithm (Rigaill et al. 2012, Bioinformatics). For all samples, cytogenetic information was available either form routine chromosomal analysis or fluorescent in situ hybridization (FISH). Blast count were known for all but one AML sample (n=19). Copy number-neutral cytogenetic alterations such as balanced translocations were excluded from the comparative analysis. By CNA-analysis of exomes we were able to detect chromosomal aberrations consistent with routine cytogenetics in 18 out of 19 (95%) AML samples. In particular, we confirmed 2 out of 2 monosomies (both -7), and 9 out of 10 trisomies (+4, n=1; +8, n=8; +21, n=1), e.g. trisomy 8 in figure 1A. Partial amplifications or deletions of chromosomes were confirmed in 10 out of 10 AML samples (dup(1q), n=3; dup(8q), n=1; del(5q), n=3; del(17p), n=1; del(20q), n=2), e.g. del(5q) in figure 1B. In the one case with inconsistent findings of chromosomal aberrations between exome and cytogenetic data there was a small subclone harboring the alteration described in only 4 out of 21 metaphases (19%). To assess the specificity of our CNA approach, we analyzed the exomes of 44 cytogenetically normal (CN) AML samples. Here we did not detect any CNAs larger than 5 Mb in the vast majority of these samples (43/44, 98%), only one large CNA was detected indicating a trisomy 8. Estimates of the clone size were highly correlated between CNA-analysis of exomes and the parameters from cytogenetics and cytomorphology (p=0.0076, Fisher's exact test, Figure 1C). In CNA-analysis of exomes, we defined the clone size based on the coverage ratio: . Clone size estimation by cytogenetics and cytomorphology was performed by calculating the mean of blast count and abnormal metaphase/interphase count. Of note, clones estimated by CNA-analysis of exomes tended to be slightly larger. This may result from purification by Ficoll gradient centrifugation prior to DNA extraction for sequencing and/or the fact that the fraction of cells analyzed by cytogenetics does not represent the true size of the malignant clone accurately because of differences in the mitotic index between normal and malignant cells. Overall, there was a high correlation between our CNA analysis of exome sequencing data and routine cytogenetics including limitations in the detection of small subclones. Our results confirm that high throughput sequencing is a versatile, valuable, and robust method to detect chromosomal changes resulting in copy number alterations in AML with high specificity and sensitivity (98% and 95%, respectively). Figure 1. (A) Detection of trisomy 8 with an estimated clone size of 100% (B) Detection of deletion on chromosome 5q with an estimated clone size of 90% (C) Correlation of clone size estimation by routine diagnostics and exome sequencing (p=0.0076) Figure 1. (A) Detection of trisomy 8 with an estimated clone size of 100%. / (B) Detection of deletion on chromosome 5q with an estimated clone size of 90%. / (C) Correlation of clone size estimation by routine diagnostics and exome sequencing (p=0.0076) Figure 2. Figure 2. Disclosures No relevant conflicts of interest to declare.

Description: Introduction: Over the last years, genome and exome sequencing approaches have increased our knowledge of molecular alterations in acute myeloid leukemia (AML). However, some important limitations still need to be addressed. First, insights into the spectrum of molecular alterations of patients with refractory AML are rare, partly due to the lack of remission samples as germline control. As these patients have a dismal prognosis, there remains an unmet need to improve therapeutic options and to identify druggable molecular lesions. Secondly, in AML patients achieving a complete remission (CR), preleukemic alterations may persist in CR and are underestimated in frequency and relevance. In this work, we investigated mesenchymal stromal cells (MSC) as germline control to decipher the spectrum of molecular alterations in refractory patients with induction failure and to disclose preleukemic hits in patients achieving CR. Patients and methods : Bone marrow (BM) aspirates at initial diagnosis (ID) were obtained from 18 AML patients (9 pts with subsequent induction failure and 9 pts that achieved CR after first induction). MSC were expanded to passage 4 and defined as CD73+/CD105+/CD271+/low/CD45-/CD33- plastic-adherent cells. For all patients, BM hematopoietic cells (BM-HC; n=18) as well as MSC (n=18) were analysed at the time of first diagnosis. All samples (n=45) were analysed by exome sequencing on a HiSeq2500 (100 bp paired end) with four samples per lane. For variant calling, MSC were used as germline control. We demanded a variant allele frequency (VAF) of 〉20%, coverage of 〉30 reads and translational consequences. In germline samples, the VAF had to be 〈 5%. For patients achieving CR, BM-HC at CR were also studied (n=9). We repeated the analysis with CR BM-HC as germline control and compared the two results. For all patients, clinical as well as molecular characteristics were available. Results: We obtained an average coverage of 96 reads per base for the protein coding regions. 96% of the target region was covered at least 10-fold. The use of MSC as germline control allowed us to detect somatic mutations at initial diagnosis of refractory AML. In 9 refractory AML samples, we found 90 single nucleotide variants (SNV) and indels, which resulted in a median of 11 alterations per sample (range: 3-17). The spectrum of mutations showed an unexpectedly high rate of mutations in the spliceosome gene SRSF2 (3/9). Other recurrent mutations affected TET2 (2/9) and WT1 (2/9). Genes frequently mutated in non-selected AML were only present in one refractory patient (DNMT3A, RUNX1, IDH2, ASXL1, TP53, NRAS) or not found mutated (IDH1, KRAS). To uncover preleukemic alterations in AML patients achieving CR (n=9), we compared MSC and BM-HC at CR as germline controls. Using MSC as germline, we called 97 SNVs and indels (median: 11 per sample; range: 4-18) in the leukemic cells at ID. Thirty-three additional SNVs were called in the leukemic BM by using MSC as germline, whereas these would have been missed using BM-HC at CR as germline (median: 3 SNVs per sample, range: 0-7). These represent preleukemic hits persistent in CR with a VAF between 5% (lower bound) and 75%. Recurrently mutated genes included genes recently associated with clonal haematopoiesis in the elderly population: DNMT3A (3/9; VAF: 18%, 24%, 75%) and TET2 (2/9; VAF: 13%, 23%). In addition, mutations in ASXL1 (VAF: 14%), SRSF2 (VAF: 15%), and RUNX1 (VAF: 5%) persisted in at least one patient in CR. This unbiased approach also allowed us to identify lesions, which have not yet been associated with AML, but account for clonal events in remission. Candidates included genes linked to cancer like PROX1 (VAF: 5%), or ERBB2 (VAF: 35%), but also genes involved in NF-kB activation such as CARD8 (VAF: 30%), or NLRC3 (VAF: 10%). Conclusion: The use of MSC allows to unravel molecular lesions in refractory AML by exome sequencing. Refractory AML patients showed a high rate of mutations in the spliceosome gene SRSF2 that needs further investigations as potential therapeutic target for patients with treatment failure. Moreover, the comparison of two different germline controls (MSC and BM-HC in CR) allowed detecting persistent preleukemic alterations. In addition to known hits like in DNMT3A, TET2, or ASXL1, we systematically identified a broader spectrum of premalignant events that indicate clonal hematopoietic expansion and thereby may provide insights into leukemic transformation. Disclosures No relevant conflicts of interest to declare.

Description: The evolution of acute myeloid leukemia (AML) has been previously described either in studies of large patient cohorts with focus on only a restricted number of AML-associated genes or in smaller series of relapsed patients studied by genome-wide techniques. We set out to comprehensively characterize the genetic evolution in a large AML cohort in order to understand molecular mechanisms of relapse and therapy-resistance. We performed exome-sequencing of matched bone marrow or peripheral blood samples taken at diagnosis, complete remission and relapse from 47 patients with cytogenetically normal AML (CN-AML). Samples were collected within the German Cancer Consortium (DKTK) at the partner sites in Berlin and Munich. The median age at diagnosis was 65y (range: 21-89y). FLT3 internal tandem duplication (ITD) and NPM1 mutation status at diagnosis was available for all but one patient (FLT3-ITD-/NPM1-, n=5; FLT3-ITD+/NPM1-, n=9; FLT3-ITD-/NPM1+, n=16; FLT3-ITD+/NPM1+, n=16). On average, 96% of the target sequence was covered at least 10-fold (minimum coverage defined for variant calling). The following criteria were applied for identification of somatic mutations: Variant allele frequency (VAF) ³20% either at diagnosis or at relapse and VAF20%, likely due to partial eradication/expansion of leukemic or pre-leukemic clones. Persistent mutations in DNMT3A, TET2, RUNX1 and IDH2 were observed in 28%, 11%, 6% and 4% of patients in our cohort, respectively (Figure 1 B). Among patients with DNMT3A mutation at diagnosis, those with persistent mutations tended to relapse earlier (n=13; median time to relapse 270 days; range: 81-586) than patients without detectable DNMT3A mutations at remission (n=7; median time to relapse 508 days; range: 235-1697; p=0.111). Our findings provide insights into the genetic evolution during the course of disease in a large cohort of relapsed CN-AML. Information about the dynamics of genetic lesions (e.g. persistent or relapse-specific mutations) may have prognostic significance and allow for tailored approaches to treat or to prevent relapse of AML. Figure 1 Figure 1. Figure 2 Figure 2. Disclosures No relevant conflicts of interest to declare.