Littérature scientifique sur le sujet « CML, NGS, aCML, SETBP1 »

Abstract Atypical Chronic Myeloid Leukemia (aCML) is a clonal disorder belonging to the myelodisplastic-myeloproliferative neoplasms, according to the WHO-2008 classification. From a clinical point of view it closely resembles the classical Chronic Myeloid Leukemia (CML), however it lacks the presence of the Philadelphia chromosome and of the BCR-ABL1 fusion gene. In recent works, we and others characterized the somatic lesions present in the aCML genome, mainly by using Next Generation Sequencing (NGS) technologies, demonstrating the presence of a large set of recurrent somatic mutations involving, among the others, SETBP1, ETNK1, ASXL1, EZH2, CBL, TET2, NRAS and U2AF1 genes. The identification of somatic variants occurring in a large number of genes clearly indicates that the genetic bases of aCML are very heterogeneous, in striking contrast with classical CML. This heterogeneity poses a great challenge to the dissection of the molecular steps required for aCML leukemogenesis. The hierarchical reconstruction of the different mutations occurring in a clonal disorder can have important biological, prognostic and therapeutic repercussions; therefore we started a project focused on the dissection of the aCML clonal evolution steps through the analysis of individual leukemic clones by methylcellulose assays in samples whose mutational status has been previously characterized by matched whole-exome sequencing. Patient CMLPh-019 was characterized by the presence of a complex mutational status, with somatic variants occurring in SETBP1, ETNK1, ASXL1 and CBL genes (Fig. 1a). Targeted resequencing analysis of individual clones revealed the presence of all the 4 variants in 44/60 (73.3%) clones; in 15/60 (25%) we detected the presence of mutated ETNK1, ASXL1 and CBL and wild-type (WT) SETBP1. Of these 15 clones, 33% carried heterozygous and 67% homozygous CBL mutations. In one clone (1.7%) we detected heterozygous ETNK1, homozygous CBL and WT sequences for ASXL1 and SETBP1, suggesting a strong selective pressure towards the acquisition of homozygous CBL mutations. Identification of homozygous CBL mutations in all the main clonal phases suggests that a significant positive selective pressure is associated with this event. Allelic imbalance analysis of CMLPh-019 exome using CEQer revealed that CBL homozygosity is caused by a somatic uniparental disomy event occurring in the telomeric region of the long arm of chromosome 11. Patient CMLPh-005 (Fig. 1b) was mutated in ASXL1, CBL and SETBP1. Targeted analysis done on 68 clones revealed a complex, branching evolution, with 63 clones carrying all the 3 variants. Of them, 47 (74.6%) had a heterozygous and 16 (25.4%) a homozygous CBL variant. Four clones (4.2%) carried ASXL1 and SETBP1 but not CBL mutations, while 1 clone was mutated in ASXL1 and CBL in absence of SETBP1 mutations, which suggests that CBL mutations occurred independently in two different subclones. Also in this case, allelic imbalance analysis of exome data revealed that CBL homozygosity was caused by a telomeric somatic uniparental disomy event. According to exome sequencing, patient CMLPh-003 carried SETBP1 mutation G870S and NRAS variant G12R. Clonal analysis confirmed the presence of SETBP1 G870S in all the clones analyzed, while heterozygous NRAS G12R mutation was detected in 67% (Fig. 1c). Notably in the remaining 33% another heterozygous NRAS variant, G12D, was detected. Retrospective reanalysis of exome data confirmed the presence of the newly identified variant, which had been previously filtered-out from exome data because of the low frequency. Patient CMLPh-013 was mutated in ASXL1, ETNK1, NRAS and SETBP1. Of the 39 clones analyzed, 34 (82.9%) showed the coexistence of ASXL1, ETNK1, NRAS and SETBP1, 4 were mutated in ASXL1, ETNK1 and NRAS and 1 in ETNK1 and NRAS, suggesting that ETNK1 and NRAS were early events, ASXL1 an intermediate one and SETBP1 a late variant (Fig. 1d). Taken globally, these data indicate that ETNK1 variants occur very early in the clonal evolution history of aCML, while ASXL1 represents an early/intermediate event and SETBP1 is often a late event. They also suggest that, in the context of aCML, there is a strong selective pressure towards the accumulation of homozygous CBL variants, as already shown in other leukemias. Figure 1. Clonal analysis of four aCML cases. The asterisks indicate hypothetical clones. Figure 1. Clonal analysis of four aCML cases. The asterisks indicate hypothetical clones. Disclosures Rea: Novartis: Honoraria; Bristol-Myers Squibb: Honoraria; Pfizer: Honoraria; Ariad: Honoraria.

Abstract Introduction: Recently, next-generation sequencing (NGS) has revolutionized the molecular characterization and understanding of several hematologic entities, including myeloproliferative neoplasms (MPN) and myelodysplastic syndrome (MDS)/MPN overlap syndromes. Nevertheless, the frequency and clinical impact of the mutations detected by NGS, remain largely unclear, especially in rare MPN which were analyzed in this study. Methods: Thus, we established a novel amplicon-based NGS panel, comprising genes that are known to be recurrently mutated in MPN and/or MDS/MPN. Hot spot regions or all exons of the following 32 genes were chosen: ABL, ASXL1, BARD, CALR, CBL, CEBPA, CHEK2, CSF3R, DNMT3A, ETNK1, ETV6, EZH2, IDH1, IDH2, JAK2, KIT, KRAS, MPL, NFE2, NRAS, PDGFRA, PTPN11, RUNX1, SETBP1, SF3A1, SF3B1, SH3B2 (LNK), SRSF2, TCF12, TET2, TP53, U2AF1. After establishing this panel, peripheral blood samples of 19 patients, which were diagnosed with CMML(10), aCML(2), MPNu(1), MDS/MPNu or other MPN(6), were analyzed on a MiSeq® illumina sequencer. Variants were only analyzed if the absolute coverage at each SNV site was >50 reads, and the absolute coverage of the mutant allele was 10 or more reads and its relative coverage exceeded 10%. Results: Mean coverage of the run was 1516 reads with good Phred-score quality parameters (>84% of called bases with Q-score >= 30). In 300 bidirectional cycles, a yield of nearly seven gigabases of sequencing data was reached. One out of 19 analyzed patients was excluded from analysis due to insufficient DNA quality. In 89% of the samples(16/18), mutations were detected which had not previously been known to be present in these patients. TET2 (50%, 9/18) and SETBP1 mutations were the most common (44%, 8/18). As expected, TET2 mutations were spread over the entire gene and SETBP1 mutations were restricted to the known hot spot region (exon 4, c.2602-c.2620). Additionally, CSF3R mutations were detected in 22% (4/18) of patients. Epigenetic regulator genes were also affected as EZH2 mutations were detected in 17% (3/18), ASXL1 mutations in 39% (7/18) and IDH1/2 mutations were found in 6% (1/18) of all samples, whereas DNMT3A mutations were not present. Further mutations were found in the following genes: CBL (11%), ETV6 (6%), JAK2V617F (6%), KRAS (11%), NRAS (11%), PTPN11 (6%), SH2B3 (6%) and SRSF2 (11%). Besides previously known mutations, several novel variants could be detected. All but one patient harbored more than one of these mutations. Furthermore, clinical correlates and morphologic and cytogenetic subtypes of each patient were available to associate with the NGS data of individual patients. For example, the one patient with a solitary NRAS c.35G>A (amino acid: p.G12D) mutation showed the most aggressive clinical course in our cohort with transformation to AML only 7 months after first diagnosis of CMML. Moreover, CSF3R mutations have been shown to confer sensitivity to ruxolitinib and may thus open up new avenues of treatment for our patients. Conclusion: In a cohort of unclassified MPN and rare MDS/MPN subtypes, NGS is a powerful tool to characterize samples more extensively. Our data suggests that a more comprehensive understanding of the mutational spectrum may have important clinical impact in individual patients, including diagnosis, prognosis, and more specific treatment. Disclosures Isfort: Pfizer: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel; Ariad: Honoraria, Membership on an entity's Board of Directors or advisory committees; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel; BMS: Honoraria; Mundipharma: Other: Travel; Amgen: Other: Travel; Hexal: Other: Travel. Brümmendorf:Novartis: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria; Bristol-Myers Squibb: Consultancy, Honoraria; Ariad: Consultancy, Honoraria; Patent on the use of imatinib and hypusination inhibitors: Patents & Royalties. Koschmieder:Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.

Abstract Atypical Chronic Myeloid Leukemia (aCML) is a clonal disorder belonging to the group of myelodysplastic/myeloproliferative (MDS/MPN) syndromes. In aCML many clinical features suggest the diagnosis of CML, however the lack of the BCR-ABL1 fusion point to a different pathogenetic process. Recently, we identified the presence of clonal somatic mutations occurring in the SETBP1 gene in approximately 25% of aCML samples (Piazza R. et al., Nat Genet. 2013 Jan;45(1):18-24). A subsequent study (Maxson J. et al., N Engl J Med. 2013 May 9;368(19):1781-90) demonstrated the presence of somatic mutations of the CSF3R gene in Chronic Neutrophilic Leukemia (CNL) and, with lower frequency, in aCML. In a recent follow-up of the first study (Gotlib J. et al., Blood. 2013 Jul 29), the presence of both CSF3R and SETBP1 variants was tested in a cohort of 9 CNL and 20 aCML cases, demonstrating the presence of CSF3R somatic mutations in 40% of the aCML patients. Of these mutations, 20% were membrane, 5% truncating and 15% compound variants. Interestingly, 5% of the aCML patients showed coexistence of CSF3R and SETBP1 mutations, suggesting that variants occurring in these genes are not mutually exclusive. To gain further insight into the relationship between CSF3R and SETBP1 in aCML, we extended our initial study by analyzing an expanded cohort of 65 aCML plus a total of 230 AML, ALL, CLL, CML, PV, TE, MMM, CMML and MDS cases for the presence of CSF3R and SETBP1 mutations. In line with previous findings (Piazza R. et al., Nat Genet. 2013 Jan;45(1):18-24; Maxson J. et al., N Engl J Med. 2013 May 9;368(19):1781-90), we found evidence of SETBP1 and/or CSF3R mutations only in MDS/MPN disorders. In aCML we identified a total of 18 (27.7%) mutations occurring in SETBP1 and 8 (12.3%) in the CSF3R gene. A large fraction (94.4 %) of the SETBP1 mutations was clustered in a 14 amino acid stretch that is also mutated in the Schinzel-Giedion syndrome, as previously reported (Piazza R. et al., Nat Genet. 2013 Jan;45(1):18-24). Of the 8 CSF3R mutations 5 were membrane proximal (4 T618I and 1 T615A) and 3 were truncating (2 Q776X and 1 Q781X). In 2 aCML samples we detected the coexistence of CSF3R and SETBP1 mutations. In both cases the CSF3R variant was a membrane proximal mutation; CSF3R and SETBP1 mutations were at comparable levels at the time of detection, therefore no conclusion can be drawn about the timing of the two mutational events. Taken globally these data indicate that somatic mutations occurring in SETBP1 and CSF3R are present in aCML and can coexist. Interestingly, the frequency of the CSF3R mutations in our aCML cohort is largely different from that of Maxson and colleagues (12.3 vs 40%), although the frequency of the combined CSF3R/SETBP1 mutations is similar (3.1 vs 5%): the reasons for this discrepancy are actually unclear. Previously we demonstrated that the presence of SETBP1 mutations in aCML is an independent negative prognostic factor (Piazza R. et al., Nat Genet. 2013 Jan;45(1):18-24). Further studies with larger cohorts will be required to assess the prognostic impact of concurrent SETBP1 and CSF3R mutations. Disclosures: Cross: Novartis: Honoraria, Research Funding; Bristol Myers Squibb: Honoraria. Gambacorti-Passerini:Pfizer, BMS: Consultancy, Consultancy Other; Pfizer: Research Funding.

Abstract Introduction Chronic neutrophilic leukemia (CNL) and atypical chronic myeloid leukemia (aCML) are rare myeloproliferative and myelodysplastic/myeloproliferative neoplasms. So far, the diagnosis of CNL and aCML has been based on cytomorphology and the absence of JAK2V617F and PDGFR rearrangements. Recently, mutations in CSF3R and SETBP1 were identified and associated with CNL and aCML, respectively. Chronic myelomonocytic leukemia (CMML) and aCML also share several characteristics and need to be discriminated especially by the absolute number of monocytes in the peripheral blood. Aim To determine the frequency of CSF3R mutations (CSF3Rmut) in CNL, aCML, and CMML and to investigate a mutation pattern, cytogenetics and clinical data in all three entities. Patients and Methods To first delineate patients with potential CNL, we investigated blood and bone marrow smears and depicted patients with a white blood cell count >25x109/L, neutrophils >80%, immature granulocytes <10%, <1% myeloblasts and hypercellular bone marrow (according to WHO 2008). BCR-ABL1 fusion transcript, JAK2 and MPL mutations were excluded in all cases by RT-PCR and melting curve analyses. Indication for PDGFR rearrangements was precluded by over-expression analyses of PDGFRA and PDGFRB by quantitative real-time PCR, resulting in a final cohort of 20 cases declared as CNL patients. Additional 60 aCML and 252 CMML patients were included. Cytogenetics was available in 330/332 cases. Mutations in CSF3R exons 14 and 17 (n=332), in ASXL1 exon 13 (n=321), and the mutational hot spots in SETBP1 (n=331) and SRSF2 (n=320) were analyzed by Sanger sequencing. Results In the total cohort of 332 patients we detected CSF3R mutations in 11 cases (3.3%). 8/11 cases showed a p.Thr618Ile mutation in exon 14, four of them carried an additional nonsense/frame-shift mutation in exon 17. One additional patient was mutated in p.Thr615Ala and showed a nonsense mutation in exon 17. Two cases showed a mutation in exon 17 only, one a nonsense the other a frame-shift mutation, respectively. Analyzing the mutation frequencies within the different entities revealed a clustering of CSF3Rmut within CNL cases with 7 of 20 (35%) mutated cases in contrast to 2 of 60 (3.3%; p=0.001) aCML and 2 of 252 (0.8%; p<0.001) CMML cases. Cytogenetics in CSF3Rmut cases showed that 9/11 cases had a normal karyotype and only one aCML patient harbored a del(3q) and one CMML patient a complex karyotype. Mutations in the three other analyzed genes ASXL1, SETBP1 and SRSF2 were detected in the total cohort in 156/321 (49%), 34/331 (10%), and 149/330 (45%) patients, respectively. Analyses regarding concomitant mutations of CSF3R with ASXL1, SETBP1 or SRSF2 revealed no additional mutation in two cases. In 8 of 11 parallel analyzed CSF3Rmut patients an ASXL1mut was identified, SETBP1 as well as SRSF2 were mutated in 3 of the 11 cases. Notably, the 7 CSF3Rmut within the CNL group had no mutation in SETBP1. Analysis of mutational loads in CNL showed that 6/7 CSF3Rmut had a higher mutational load than the second mutated gene (range: 25-50% vs. 10-30%). In one case both mutated genes had equal mutational loads (40%). In contrast, in CMML and aCML 3/4 patients had lower mutational loads in CSF3Rmut than in the additional mutated genes (20-50% vs. 40-50%), while also one case showed equal mutational loads in the mutated genes (50%). Combining the mutational results of these four genes indicate a specific and individual molecular pattern for these three different entities. While ASXL1 is frequently mutated in all entities (CNL: 8/11 (73%); aCML: 38/59 (64%); CMML: 110/251 (44%)), SRSF2 shows the highest mutation frequency in CMML cases (121/251; 48%), followed by aCML (24/60; 40%) and CNL (4/19; 21%). In contrast, SETBP1 is often mutated in aCML (19/60; 32%) and rarely in CMML (13/252; 5%) and CNL (2/19; 10.5%) patients. In addition, CSF3R is much more associated with the CNL cases (35%) and less frequently found in aCML (2%) and CMML (1%). Conclusion 1) CNL, aCML and CMML are related diseases and difficult to distinguish by cytomorphology alone and therefore require additional diagnostic criteria, i.e. molecular mutations. 2) ASXL1 is the most frequently mutated gene in these entities and thus can help to prove clonality. 3) SETBP1 much more closely relates to aCML and SRSF2 to CMML. 4) Mutations in the novel marker CSF3R are closely related to CNL and thus qualify as a new molecular marker for diagnosis of CNL. Disclosures: Meggendorfer: MLL Munich Leukemia Laboratory: Employment. Alpermann:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Schrauder:MLL Munich Leukemia Laboratory: Employment. Konietschke:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

Abstract Abstract LBA-2 The SETBP1 gene codes for a predominantly nuclear protein with a predicted MW of 170 kD. Germline mutations of SETBP1 were described in patients affected by the Schinzel-Giedion syndrome (SGS), a rare disease characterized by bone, muscle and cardiac abnormalities, and presenting neuroepithelial neoplasms. In an effort to investigate the molecular pathogenesis of myeloid malignancies we applied a HTS strategy, including both exome sequencing and RNA-SEQ, to atypical Chronic Myeloid Leukemia (aCML), as defined by WHO criteria, with the aim of identifying novel recurrent driver mutations. aCML shares clinical and laboratory features with CML, but it lacks the pathognomonic BCR-ABL1fusion. Since no specific recurrent genomic or karyotypic abnormalities have been identified in aCML, the molecular pathogenesis of this disease has remained elusive and the outcome dismal (median survival 37 months) with no improvement over the last 20 years. This sharply contrasts with the outcome for CML, for which the prognosis was dramatically improved by the development of imatinib as a specific inhibitor of the BCR/ABL protein. Whole-exome sequencing of 9 aCML patients revealed the presence of 62 unique mutations (range 5–14 per patient), including a recurrent alteration of SETBP1 (G870S and D868N) in three cases. Targeted resequencing performed in 70 aCMLs, 574 patients with different hematological malignancies and 344 cell lines, identified SETBP1 mutations in 17 of 70 aCML patients (24.3%; 95% CI: 16–35%), 4 of 30 (13%) MDS/MPN-u and 3 of 82 (3.6%) CMML patients. Patients with mutations had higher white blood cell counts (p=0.008) and worse prognosis (p=0.01) when tested in multivariate analysis. TF1 cells transfected with SETBP1G870S showed increased SET levels, decreased PP2A activity and increased proliferation rates. The vast majority of mutations (85%) was located between residues 858 and 871, in the SKI homologous region of SETBP1, and were identical to germline changes seen in patients with SGS. This region may be critical for ubiquitin binding and for subsequent protein degradation, since the Eukaryotic Linear Motif (ELM) identified with high probability score a putative functional site (aa. 868–873) for beta-TrCP, the substrate recognition subunit of the E3 ubiquitin ligase. This prediction was experimentally validated using biotinylated, phosphorylated peptides encompassing this region (aa 859–879): while the wild type peptide could efficiently bind beta-TrCP as predicted, a peptide presenting the G870S mutation was incapable of binding this E3 ligase subunit, indicating a possible alteration in SETBP1 protein stability caused by this mutation. In agreement with these findings, cells transfected with SETBP1G870Sshowed increased levels of SETBP1 protein when compared to cells with similar expression levels of the wild type gene. Finally, RNA-SEQ yielded gene expression profiles with overrepresentation of genes under the control of Transforming Growth Factor Beta 1 (TGFβ1) among genes differentially expressed between SETBP1-mutated and unmutated aCML patients. Mutated SETBP1 represents a novel type of oncogene which is specifically present in aCML and closely related diseases. These data allow for a better understanding of the molecular pathogenesis of this disease; they provide evidence that SETBP1 mutations might be a new biomarker for future diagnosis and classification of aCML and related diseases, and indicate a potential strategy to develop new treatment modalities for malignancies caused by mutated SETBP1. Disclosures: Schnittger: MLL Munich Leukemia Laboratory: Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership.

Abstract Atypical chronic myeloid leukemia (aCML) and chronic neutrophilic leukemia (CNL) are rare myeloid neoplasms defined largely by morphologic criteria. The discovery of CSF3R mutations in aCML and CNL have prompted a more comprehensive genetic profiling of these disorders. These studies have revealed aCML to be a genetically more heterogeneous disease than CNL, however, several groups have reported that SETBP1 and ASXL1 mutations occur at a high frequency and carry prognostic value in both diseases. We also report a novel finding—our study reveals a high frequency of U2AF1 mutations at codon Q157 associated with CSF3R mutant myeloid neoplasms. Collectively, these findings will refine the WHO diagnostic criteria of aCML and CNL and help us understand the genetic lesions and dysregulated signaling pathways contributing to disease development. Novel therapies that emerge from these genetic findings will need to be investigated in the setting of a clinical trial to determine the safety and efficacy of targeting various oncogenic drivers, such as JAK1/2 inhibition in CSF3R-T618I–positive aCML and CNL. In summary, recent advances in the genetic characterization of CNL and aCML are instrumental toward the development of new lines of therapy for these rare leukemias that lack an established standard of care and are historically associated with a poor prognosis.

Abstract INTRODUCTION: Atypical Chronic Myeloid Leukemia (aCML) is a heterogeneous disorder belonging to the group of myelodysplastic/myeloproliferative syndromes, characterized by a poor prognosis with a median survival time of 37 months. In 2013, by applying Next Generation Sequencing (NGS) technologies on 8 aCML cases, we demonstrated the presence of a recurrent somatic mutations in the SETBP1 gene (Piazza et al, Nat Gen 2013). SETBP1 mutations were identified in approximately 30% of aCML cases. AIM: To further characterize the molecular pathogenesis of aCML and to possibly identify other recurrent lesions responsible for SETBP1 unmutated cases, we extended our initial NGS effort: we applied whole-exome and transcriptome sequencing to a total of 16 matched samples taken at onset of the disease. MATERIAL and METHODS: Whole-exome and transcriptome sequencing data were generated using an Illumina Genome Analyzer IIx following standard library-preparation protocols. Alignment to the reference GRCh37/hg19 genome was performed using BWA. Alignment data were processed using Samtools. Single nucleotide and small indel detection was performed using in-house software. Copy number analyses from whole-exome data were generated using CEQer (Piazza et al, PLoS One 2013) and gene fusions transcriptome data were screened using FusionAnalyser (Piazza et al, Nucleic Acids Res. 2012). RESULTS: The application of NGS techniques to the cohort of aCML cases led to the identification of a somatic, non-synonymous single-nucleotide mutation (chr4:g.55599321A>T) in the KIT gene in 1/16 (6%) cases. At protein level this mutation translated into the D816V variant that has been already described in several clonal disorders, such as systemic mastocytosis, gastrointestinal stromal tumors and acute myeloid leukemia. To assess whether the mutation identified by NGS was recurrent, we extended our analysis by targeted resequencing on a larger cohort of 68 aCML cases. This analysis revealed the presence of KIT mutations in 3 additional patients, thus confirming the recurrence of KIT variants in aCML. All the KIT mutations identified correspond to the D816V that is responsible for the constitutive activation of the tyrosine kinase. This finding suggests that the activation of the KIT tyrosine kinase signaling may play an important role in this subset of aCML patients. It is known from the literature that KIT D816V is highly sensitive to the tyrosine kinase inhibitor dasatinib (Schittenhelm MM, Cancer Res 2006). To test whether dasatinib is able to affect the growth of the leukemic clone in KIT mutated aCML cases, we performed ex vivo tritiated thymidine proliferation assays on bone marrow (BM) cells from one of the KIT D816V positive aCML patients in presence of either dasatinib, imatinib or vehicle alone: the proliferation assay showed that dasatinib was able to inhibit the proliferation of the leukemic clone with an IC50 of 1nM, while, as expected, neither imatinib nor vehicle alone were able to significantly impair cell growth. In line with these data, western blot with an anti- Phospho-KIT antibody on KIT+ lysates after treatment with increasing concentration of dasatinib showed that the drug was highly effective in inhibiting KIT autophosphorylation. To further confirm the inhibitory activity of dasatinib, we performed a colony assay on peripheral blood cells from a KIT D816V positive aCML patient grown in presence of increasing concentrations of the drug: treatment with 100nM dasatinib was able to completely inhibit cell growth, leading to a virtually complete absence of colonies in the D816V-positive plates. CONCLUSION: These data indicate that KIT D816V is a pro-oncogenic lesion recurrently present in aCML, albeit with low frequency (5/84, 6%) and that aCML cells bearing this mutation are highly sensitive to dasatinib, at least ex vivo. Given the very poor prognosis of this disorder, these findings suggest a new, highly efficient targeted treatment for a subset of aCML patients. Disclosures Schnittger: MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Meggendorfer:MLL Munich Leukemia Laboratory: Employment.

Abstract BCR-ABL1 negative myeloproliferative neoplasms not only include atypical chronic myeloid leukemia (aCML), but also chronic myelomonocytic leukemia (CMML), chronic neutrophilic leukemia (CNL), and myelodysplastic/myeloproliferative neoplasm, unclassifiable (MDS/MPN, U). Despite the recent advances in characterizing aCML more specifically, based on next generation sequencing data, the differential diagnosis and subsequent treatment decisions remain difficult. Therefore, we analyzed the transcriptome and performed whole genome sequencing (WGS) in a cohort of morphologically defined 231 patients (pts): 49 aCML, 30 CNL, 50 MDS/MPN, U, and 102 CMML all diagnosed according to WHO classification. WGS libraries were prepared with the TruSeq PCR free library prep kit and sequenced on a NovaSeq 6000 or HiSeqX instrument with 100x coverage (Illumina, San Diego, CA). The Illumina tumor/unmatched normal workflow was used for variant calling. To remove potential germline variants, each variant was queried against the gnomAD database, variants with global population frequencies >1% where excluded. For transcriptome analysis total RNA was sequenced on the NovaSeq 6000 with a median of 50 mio. reads per sample. The obtained estimated gene counts were normalized and the resulting log2 counts per million (CPMs) were used as a proxy of gene expression. Unsupervised exploratory analysis techniques, such as principal component analysis (PCA) and hierarchical clustering (HC) were used to identify groups of samples with similar expression profiles. We observed a high similarity between the different entities with CMML being the most distant entity, followed by CNL. MDS/MPN, U and aCML were the most similar entities. Due to high within-group heterogeneity, we found that it was impossible to identify a gene expression signature that separated aCMLs reliably from the other MDS/MPN overlap entities. Surprisingly, PCA as well as HC indicated the existence of two subgroups within the aCMLs. Therefore, we searched for genes with a bimodal-like expression profile. We found that FOS expression levels strongly separated aCMLs into two groups of 16 pts (FOSlow) and 33 pts (FOShi), respectively. Interestingly, FOShi correlated with mutations in SETBP1 (12/33, 36% vs. 3/16, 19%), a known marker typically mutated in aCML (Figure 1a). Addressing the mutational landscape of these two groups (FOShivs.FOSlow) we found that ASXL1 (88% vs. 100%), TET2 (33% vs. 50%), SRSF2 (45% vs. 56%), EZH2 (27% vs. 31%), and NRAS (21% vs. 25%) showed rather similar mutation frequencies. GATA2 (15% vs. 31%) and RUNX1 (18% vs. 38%) mutations were less frequent in FOShi, whereas SETBP1 and CBL (18% vs. 6%) were more frequent in this group. Consistent with known features of SETBP1 mutation this group showed a higher white blood cell count (78 x109/L vs. 52 x109/L) and platelet count (158x109/L vs. 90x109/L), although none of these differences were significant. The two groups were further analyzed for gene expression differences and we found 16 genes with synchronized upregulation within the FOShi group that were differentially expressed (FDR < 0.05, absolute logFC > 1.5) compared to FOSlow. Functional enrichment analysis linked those genes with regulation of cell proliferation (p<0.001), negative regulation of cell death (p<0.001), and the AP-1 complex (p<0.001). Those 16 genes included the transcription factors JUN, FOSB, EGR3, and KLF4, the cancer-related genes DUSP1, RHOB, OSM, TNFRSF10C, and CXCR2, and the FDA approved drug targets JUN, COX-2, and FCGR3B (Figure 1b). JUN/FOS are the main components of the AP-1 complex, a regulator of cell life and death. The upregulation of these genes results in increased proliferation as clinically observed in aCML pts. Furthermore, for these pts a treatment with INFα might result in an anti-proliferative effect by modulation of FOS transcript levels. Further, COX-2 inhibitors might also suppress proliferation and differentiation of leukemia cells. However, for the FOSlow pts these treatments might not be as effective due to the already low expression levels of the respective genes. Since the expression levels of FOShi equal those of MDS/MPN overlap, whereas FOSlow levels are closer to the ones of a healthy control cohort, SETBP1 mutation might be a marker and indicator for pts with high FOS expression and therefore providing further treatment options by targeting specifically the FOS mediated pathways. Disclosures Meggendorfer: MLL Munich Leukemia Laboratory: Employment. Walter:MLL Munich Leukemia Laboratory: Employment. Hutter:MLL Munich Leukemia Laboratory: Employment. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

Abstract Background: Atypical chronic myeloid leukemia (aCML) is a rare disorder classified as one of the MPN/MDS overlap syndromes. aCML usually presents like CML but lacks the pathognomonic BCR-ABL fusion found in CML. Most patients progress to acute myeloid leukemia (AML) with a median time to AML of 11.2 months and have a median overall survival of only 12.4 months (Wang et al. Blood 2014). Recurrently mutated genes found in aCML patients include SETBP1 , CSF3R, NRAS, EZH2, ASXL1, ETNK1, and U2AF1. The pathogenesis of aCML is poorly understood and neither specific nor effective treatments besides hematopoietic stem cell transplantation are available. We therefore aimed at developing a patient derived xenotransplantation model that allows serial transplantation and expansion of human leukemic cells and evaluation of novel treatments and drugs in vivo. Patient and Methods: Bone marrow cells were harvested from a patient diagnosed with atypical CML based on persistent leukocytosis, immature circulating myeloid precursors (16% metamyelocytes, 8% myelocytes, 9% blasts), marked dysgranulopoiesis, minimal monocytosis and basophilia, hypercellular bone marrow with high myeloid/erythroid ratio and 6% myeloid blasts, dysplasia in megakaryocytes and erythroid progenitors, and absence of BCR-ABL and mutated JAK2. The patient had moderate anemia and normal platelet counts and cytogenetic analysis showed a normal karyotype. Eight hundred thousand bone marrow cells were injected intravenously in NOD/SCID IL-2 receptor γ (NSG) deficient mice. We monitored these mice for human cell engraftment by regular eye bleeds every 4 weeks. Bone marrow and spleen cells from engrafted mice were retransplanted in secondary and tertiary mice of the NSG strain transgenic for SCF, IL3 and GM-CSF (NSGS). Patient cells were analyzed for mutations in fifty four genes by next generation sequencing and mutations were confirmed by Sanger sequencing in primary patient cells and cells from tertiary mice. Results: Human CD45+ cells from the aCML patient showed increasing engraftment over time in the NSG mouse reaching 16% in peripheral blood and 35% in spleen at 26 weeks after transplantation. In secondary (n=2) and tertiary (n=4) mice we used NSGS recipient mice and observed considerably accelerated engraftment kinetics leading to 19, 21 and 73% human cells in peripheral blood, spleen and bone marrow, respectively, between 12 and 15 weeks after transplantation. The myeloid marker CD33 was expressed in 86% of human bone marrow cells, while lymphoid markers CD3 and CD19 were absent. The stem and progenitor phenotype CD34+CD38- was found in 11% of human cells. The progenitor marker CD123 was expressed in 42% of cells, while the myeloid marker CD14 was expressed in 6% of cells. Hemoglobin levels and platelet counts were considerably lower in secondary and tertiary recipients of aCML cells compared to control animals. Spleens were enlarged at time of sacrifice with an average spleen weight of 150 mg. Morphological evaluation of bone marrow cells in tertiary recipients revealed a characteristic picture for aCML with 39% neutrophils, 8% blasts and 53% myeloid progenitors and monocytes. Molecular analysis identified mutations in ASXL1, RUNX1 and EZH2 with variant allele frequencies of 49, 48 and 46 percent, respectively that were confirmed in human cells from tertiary recipient mice. Thus, we show that primary aCML cells can be expanded and serially transplanted in immunodeficient mice and suggest clonal stability of this model. Conclusion: We provide the first patient derived xenotransplantation model for atypical CML, which preserves the phenotypic and molecular characteristics of the primary disease and allows serial transplantation and expansion of aCML cells. This model will serve to better understand the pathogenesis of aCML and to test urgently needed novel treatment approaches. Disclosures No relevant conflicts of interest to declare.

Abstract Introduction The identification of mutations (mut) in SETBP1 recently shed light on a molecular marker in atypical chronic myeloid leukemia (aCML), a disease previously defined by exclusion criteria. SETBP1mut have been identified in different myeloid malignancies. We previously reported mutation frequencies in the range of 5-10% in MPN and MDS/MPN overlap, 32% in aCML, while we found SETBP1 less frequently mutated in AML (3%). SETBP1mut has been shown to associate with ASXL1, CBL and SRSF2 mutations, as well as the cytogenetic abnormalities -7 and i(17)(q10). Aim To investigate the mutation frequency of ASXL1, SETBP1, and SRSF2 in different myeloid entities in correlation to the cytogenetic abnormalities -7 and i(17)(q10). Patients and Methods A cohort of 451 patients (pts) with different myeloid entities was analyzed. Diagnoses according to cytomorphology followed the WHO classification from 2008 (n=439, for n=12 cases no cytomorphology was available): AML (n=29), aCML (n=62), MDS/MPN overlap (n=16), CMML (n=283), MDS (n=5), MPN (n=43), CML (n=1). The cohort consisted of 303 males and 148 females; cytogenetics was available in 445 cases. Patients were grouped by normal karyotype (n=291), i(17)(q10) (n=16), -7 (n=22), and other cytogenetic aberrations (n=117); one case carried both a i(17)(q10) and a -7. ASXL1 exon 13, the mutational hotspot regions of SETBP1 and SRSF2 were analyzed by Sanger sequencing in all cases. Results In the total cohort ASXL1 was mutated in 222/451 (49%), SETBP1 in 61/451 (14%), and SRSF2 in 209/451 (46%) cases. 137 pts showed no mutation in any of these three genes. 171 pts carried one mutation, thereof 84 a sole ASXL1mut, 82 a sole SRSF2mut and only 5 cases showed sole SETBP1mut. In 108 pts two, and in 35 pts all three analyzed genes were mutated. The most frequent combination within the group with two mutations was ASXL1 and SRSF2 (n=78), followed by ASXL1 and SETBP1 (n=16), only 5 cases were mutated in SRSF2 and SETBP1. Addressing the association with cytogenetics revealed that in cases with only one mutation SRSF2mut associated as sole mutation with a normal karyotype (68/124 (55%) SRSF2mut in the normal karyotype group vs. 12/42 (28%) SRSF2mut in all other karyotypes; p=0.003). In contrast, ASXL1mut and SETBP1mut as sole mutations showed no correlation to any addressed karyotype. However, addressing the cases with two mutations the combination of SRSF2mut and ASXL1mut correlated with a normal karyotype (67/291 (23%) SRSF2mut/ASXL1mut in the normal karyotype group vs. 19/154 (12%) SRSF2mut/ASXL1mut in all other karyotypes; p=0.008), while SRSF2mut and SETBP1mut occurred more frequently in i(17)(q10) pts (2/16 (13%) SRSF2mut/SETBP1mut in i(17)(q10) vs. 2/429 (1%) SRSF2mut/SETBP1mut in all other karyotypes; p=0.007). Remarkably, cases with mutations in all three analyzed genes (ASXL1mut, SETBP1mut, and SRSF2mut) highly associated with i(17)(q10) and -7. 11 of 16 cases with i(17)(q10) (69%) showed all three mutations (vs. 24/429 (6%) in all other karyotypes; p<0.001). Furthermore, 6 of 22 cases with -7 (27%) showed mutations in all three genes (vs. 29/423 (7%); p=0.005). Therefore, 15 pts carried all three mutated genes as well as i(17)(q10) or -7. Interestingly, there was no case with only i(17)(q10) and no additional mutation, and only one case with i(17)(q10) and only one additional molecular mutation, 4 cases with two additional molecular mutations and 11 cases carrying all three mutations, possibly indicating that i(17)(q10) appear during clonal evolution. Therefore one might assume that this represents a specific genetic phenotype that is driven by the accumulation of molecular events, since addition of SETBP1mut shifts the association from a normal karyotype to i(17)(q10) or -7. Analyzing the distribution of these cases for mutations in all three analyzed genes +/- additional cytogenetic aberration i(17)(q10) or -7 in the different myeloid entities showed that AML, aCML, CMML, MDS as well as MPN showed this genetic phenotype (AML: n=7 (24%); atypical CML: n=12 (19%); CMML: n=8 (3%); MDS: n=1 (20%); MPN: n=6 (14%)). Conclusions Mutations in SETBP1 associate with ASXL1mut and SRSF2mut and are frequently found in patients with i(17)(q10) or -7. This combination of genetic lesions occurs in different myeloid entities and might therefore define a specific genetically defined subtype of myeloid malignancy. Disclosures: Meggendorfer: MLL Munich Leukemia Laboratory: Employment. Alpermann:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Sirch:MLL Munich Leukemia Laboratory: Employment. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

In hematological myeloid malignancies the accumulation of oncogenic events plays a significant role in disease progression. Therefore, in this work, we studied (i) the mutational landscape in typical chronic myeloid leukemia (CML) patients and (ii) the neoplastic role of Setbp1 mutation in atypical chronic myeloid leukemia (aCML). 1) We evaluated somatic variants in CML patients by Next Generation Sequencing, to study the molecular pathogenesis of cancer. We conducted a mutational analysis on 23 chronic phase BCR-ABL1+ CML patients through exome and RNA sequencing performed on diagnosis samples. A total of 107 non-synonymous variants (range 0-11 per patient) were identified by setting a threshold of mutation frequency >25%, which corresponds to the presence of a heterozygous mutation in >50% of cells, assuming a pure tumoral sample. A positive correlation was observed between number of mutations and patient age, indicating that several events were passenger mutations, being immortalized by the neoplastic transformation. However, when using a newly in-house developed tool (Oncoscore) to weigh the oncogenic potential of each mutation, a significant correlation was observed between the Sokal score and Oncoscore by using linear model statistical analysis. In long term follow-up (>2 years), 21 CML patients achieved complete cytogenetic responses (CCyR) and 2 failed to achieve any cytogenetic response with tyrosine kinase inhibitors. These two patients showed an Oncoscore value of (165.4 ±27.2) which was significantly higher than the one (80.6 ±12.7) in the 21 responding patients. No fusions (other than BCR/ABL1) were identified by RNA Seq, and no chromosomal alterations were observed by using the CEQer software. In conclusion, CML patients at diagnosis carry genetic alterations additional to the BCR/ABL1 fusion, which could be relevant for response to treatment and progression of the disease. 2) We aimed to gain insights into the biological role of Setbp1 mutations found in aCML patients, by invivo genetic manipulation techniques. Recently, by NGS approach, we identified a recurrent SETBP1 missense mutation in aCML patients, associated with poor overall survival. The most frequent SETBP1 mutations identified in various MDS/MPN neoplasms were positioned at D868N, S869G, G870S and I871T. The same mutations identified in myeloid malignancies had previously been observed as de novo recurrent germline mutations responsible for Schinzel-Giedion syndrome. Unfortunately, the biological role of Setbp1 and its activity in leukemic transformation is not exactly known. Therefore, an improved understanding of the molecular mechanism is imperative. So, we applied genetic engineering to construct a conditional knock-in model for dissecting the role of leukemia transforming factors in heterozygous Setbp1G870S mice. For construction, 3 genomic fragments of Setbp1 intron 3 and exons 4 through 6 were subcloned into the conditional replacement vector pDELBOY-3X. The linearised vector was then transfected into murine ES cells. We are currently screening ES cells to identify a correctly targeted clone for blastocyst injection and transplantion into pseudo-pregnant mice. The 1st generation Setbp1WT/floxed mice will express wild type Setbp1 under the control of its endogenous promoter. Thereafter, the expression of Setbp1G870S would be induced in a conditional manner with Cre-mediated recombination. Depending on the type of promoter driving Cre recombinase expression, the mutant allele will be expressed either constitutively (germline) or somatically, and it will be possible to study the oncogenic effects of Setbp1G870S in specific tissues, or in all tissue/cells. Additionally, the molecular interactions and physiological pathways accountable for tumorigenesis and the clonal evolution pattern will be examined by implementing the molecular and functional genomic techniques, which help in better understanding of developing targeted therapies.

In the past years, the improvements in sequencing technology led to the development of “Next Generation Sequencing” (NGS) technologies. Several NGS approaches exist. Whole genome sequencing (WGS) and whole exome sequencing (WES) allow the identification of genomic alterations such as small insertions/deletions, point mutations and structural variants. Whole transcriptome sequencing (RNA-Seq) permits to quantify gene expression profiles and to detect alternative splicing and fusion transcripts. Recently, by using WES on atypical chronic myeloid leukemia (aCML) samples, our group identified recurrent mutations in SETBP1 gene; also, by using RNA-Seq on acute myeloid leukemia (AML), we identified a new fusion gene: ETS2-ERG. In aCML, SETBP1 mutations disrupt a degron binding site, leading to a decreased protein degradation. This leads to an increased amount of SETBP1 protein interacting with its natural ligand SET, which in turn acts inhibiting the protein phosphatase 2A (PP2A) oncosuppressor. Interestingly, the SETBP1 mutational cluster affected in aCML is highly conserved and the same mutations were also observed in the Schinzel-Giedion syndrome (SGS). However, the inhibition of the PP2A by SET, the only known interactor of SETBP1, does not explain the phenotype of SGS. To further characterize the role of SETBP1 protein, 293 Flp-In isogenic cellular models expressing the empty vector or the wild type (WT) or mutated (G870S) form of SETBP1 were established. In these models SETBP1 was fused with a V5 tag. Chromatin Immunoprecipitation sequencing experiments (Chip-Seq) performed against V5 confirmed the binding of SETBP1 to DNA, both for the WT and G870S forms. In addition, RNA-Seq experiments were performed. The comparison between Chip-Seq and RNA-Seq data has allowed us to identify 130 genes presenting both the binding of SETBP1 to their promoter region and transcriptional upregulation. Together these data suggest a role for SETBP1 as a transcriptional activator. Co-immunoprecipitation (Co-IP) experiments in transiently transfected HEK293T cells coupled with mass spectrometry (MS) analysis were performed to identify potential interactors of SETBP1. MS analysis led to the identification of the host cell factor 1 (HCF1), a component of the SET1/KMT2A COMPASS-like complex. Independent validation by western blot and fluorescence resonance energy transfer (FRET) confirmed the direct binding of HCF1 to SETBP1. Further independent experiments confirmed the Co-Ip of SET1/KMT2A and PHF8 with SETBP1. SET1/KMT2A is a core component of COMPASS-like complex and possesses H3K4 methyltransferase activity, whereas PHF8 possesses H4K20 demethylase activity. Both marks are associated with actively transcribed genes. Taken together, we have shown that SETBP1 protein is able to act as a transcriptional activator recruiting the HCF1/KMT2A/PHF8 complex. In a previous study, comparing cytogenetic analysis and RNA-Seq to detect chromosomal abnormalities on AML patient samples, a new fusion between the ETS2 and ERG genes was reported. The patient carrying this fusion was affected by acute promyelocytic leukemia (APL) and did not respond to therapy with retinoic acid. The role of the ETS2-ERG fusion is not known. To gain insight about the functional role of ETS2-ERG fusion in APL two cellular models were established. HL-60 and NB-4 cells were stable transfected with retroviral empty vector or with a vector carrying the fusion gene. This vector also carries the GFP as a positive selection marker. HL-60 cells carrying the ETS2-ERG fusion treated with retinoic acid showed a decrease in the expression at membrane level of the differentiation marker CD11b. This suggests that the ETS2-ERG fusion is able to impair the differentiation of APL cells upon retinoic acid treatment. Further experiments are ongoing to confirm the data in the NB4 cellular model.