Subset Of AML Research Articles

Serine and Threonine Phosphorylation Marks Proteins for Degradation By Clpxp

The mitochondrial serine protease, ClpXP, regulates the integrity of the respiratory chain by degrading damaged and/or misfolded proteins. This protease is over-expressed in a subset of AML and inhibiting or hyperactivating it kills leukemic cells and stem cells in vitro and in vivo. Yet, it is unknown how the mitochondrial ClpXP recognizes proteins for degradation.In Bacillus subtilis, the bacterial ClpXP homologue recognizes proteins tagged with phospho-arginine for degradation. To determine if phosphorylated amino acids influence mitochondrial ClpXP-mediated protein degradation, we incubated recombinant ClpXP with its unnatural substrate FITC-casein and increasing concentrations of phospho-serine (pSer), phospho-threonine (pThr), phospho-arginine (pArg), or phospho-tyrosine (pTyr) in a cell-free assay and measured release of fluorogenic FITC. In a dose-dependent manner, pSer and pThr free amino acids inhibited casein cleavage by ClpXP while pTyr, pArg, and the dephosphorylated amino acids had no effect on ClpXP activity. Likewise, ApSA, RRApSVA, and ApTA peptides inhibited ClpXP enzyme activity, while the non-phosphorylated version (ASA, RRASVA, and ATA) had no effect.Next, we tested whether the phosphorylation state of full length proteins would influence their degradation by ClpXP. Using gel-based cell-free assays, the phosphorylation enriched α-casein and β-casein were degraded by recombinant ClpXP. In contrast, κ-casein with low levels of phosphorylation and dephosphorylated α-casein were not cleaved by ClpXP.As ClpX is an AAA ATPase, we asked if pSer and pThr acted on the ATPase of the enzyme. pSer and pThr did not inhibit the ATPase activity of ClpX. We also measured the effect of pSer and pThr on the peptidase activity of ClpP alone without its regulatory subunit ClpX. Neither pSer or pThr inhibited ClpP peptidase activity.We investigated if pSer and pThr could bind to ClpX. Using thermal shift binding assays, we demonstrated that pSer and pThr but not pTyr and pArg bound ClpX and none of the phosphorylated amino acids bound ClpP. Similarly, ApSA, RRApSVA, and ApTA peptides bound ClpX, while the non-phosphorylated ASA, RRASVA, and ATA did not bind ClpX.Previously we showed that ClpP interacted with respiratory chain complex II subunit SDHA, and ClpP knockdown in AML cells impaired respiratory chain complex II activity and reactive oxygen species increased. Therefore, we tested how ClpXP knockdown impacts levels of phospho-serine SDHA. Using shRNA, we knocked down ClpP and ClpX individually in OCI-AML2 cells. After target knockdown, we pulled down pSer proteins and then probed for SDHA. Knockdown of both ClpP and ClpX increased the abundance of serine phosphorylated SDHA comparing to control, suggesting that ClpXP degrades serine phosphorylated SDHA.Finally, as a chemical approach, we generated small molecules that mimic pSer and demonstrated that they inhibited ClpXP-mediated degradation of FITC-casein and bound ClpX with a potency similar to pSer.In summary, we discovered that ClpX binds pSer and pThr and phosphorylation of these amino acids mark proteins for degradation by the ClpXP mitochondrial protease. This work highlights a new strategy to develop inhibitors of ClpXP for the treatment of AML. DisclosuresSchimmer: Takeda Pharmaceuticals: Consultancy, Research Funding; Medivir AB: Research Funding; Otsuka Pharmaceuticals: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Jazz: Consultancy, Honoraria; UHN: Patents & Royalties.

Multicenter Analysis of Treatment and Outcomes for Patient with TP53 Mutated AML in the Era of Novel Therapies; Significant Impact of Allogeneic Stem Cell Transplantation on Survival

Multicenter Analysis of Treatment and Outcomes for Patient with TP53 Mutated AML in the Era of Novel Therapies; Significant Impact of Allogeneic Stem Cell Transplantation on Survival

A Novel Transcriptomic Classifier for AML Is Highly Associated with Drug Sensitivity

Introduction: AML is a heterogeneous disease with a wide array of common genetic aberrations. Traditional classification of AML leverages both classical cytogenetics and mutational profiling to stratify patients into four distinct risk groups (ELN). However, tumor gene expression profiles can play an important role in response to therapy, and are potentially useful for unravelling the heterogeneity of AML. In this study, we hypothesized that clinical outcomes and variable responses to therapeutic modalities in AML may be driven by patterns of gene expression, and sought to identify clinically actionable molecular subtypes using the available RNAseq data from the BEAT AML functional genomics study.Methods: Unsupervised machine learning approach based on consensus non-negative matrix factorization (cNMF) was applied to VOOM normalized BEAT-AML RNAseq data from patient samples with ≥50% blasts (N=389) to identify transcriptomic-based molecular subtypes. The subtypes were then compared to the genomic based subtypes for their association with clinical outcome (log-rank test) and ex-vivo drug sensitivity (Kruskal Wallis test). Subtypes were also biologically characterized by gene signature scoring using well curated pathway signatures (GSVA analysis using Hallmark pathways), cell type enrichment (xCell enrichment) and AML differentiation state (scRNAseq signature based on Van Galen et. al). Finally, a random forest classifier was defined based on samples from BEAT AML to predict the NMF subtypes in an independent data set (TCGA AML cohort).Results: Our cNMF based analysis identified six clusters of patients based on the 5,060 (top 10%) most variable genes. These novel subtypes were strongly prognostic (Figure 1A, log rank p=2.79e-08), and were independent of ELN genomic based subtypes (anova p=4.45e-07). Comparison to other genomic based classification is ongoing. The prognostic value of the transcriptomic subtypes was further validated by predicting the subtypes in an independent cohort (TCGA LAML, N=200). We observed a significant association with outcome (Figure 1B, p=0.00013), with clusters 5 and 1 showing markedly better prognosis, similar to BEATAML. These subtypes also displayed unique biological profiles, including significant association with scRNAseq-derived AML differentiation state cell types, Hallmark pathways and cellularity signatures. Notably, clusters 1 and 3 showed a mature phenotype, while clusters 2, 4, and 5 were more progenitor-like (table 1).Importantly, the transcriptomic subtypes were highly predictive of ex-vivo drug sensitivity, with sensitivity to 70 compounds significantly associated with cNMF subtype (Kruskal Wallis p>0.01), compared with 4 in the ELN subtypes.Of the tested molecules, single agent Venetoclax was the most strongly associated with subtype (p=1.7e-13); two subtypes were strongly resistant (median IC50 of 10uM) and four were sensitive, with IC50s in the sub-micromolar range (Table 1). No association was seen between the ELN subtypes and venetoclax sensitivity (p=.35).Conclusions: Unsupervised machine learning-based clustering analysis of transcriptomic data identified six novel subtypes which are similarly prognostic as the ELN genomic based subtype and provide a novel avenue for identifying clinically actionable subsets of AML. [Display omitted] DisclosuresHamidi: Genentech: Current Employment, Current equity holder in publicly-traded company. Bolen: Genentech: Current Employment; F. Hoffmann-La Roche: Current equity holder in publicly-traded company. Lasater: Genentech: Current Employment, Current equity holder in publicly-traded company. Dunshee: Genentech/Roche: Current Employment, Current equity holder in publicly-traded company. Punnoose: Genentech: Current Employment, Current equity holder in publicly-traded company. Dail: Genentech/Roche: Current Employment, Current equity holder in publicly-traded company.

Mutation Profiling and Associated Clinical Features in Chinese Cytogenetically Normal Acute Myeloid Leukemia Patients

AML without abnormal karyotypes (CN-AML), which accounted for nearly half of total AML patients, is a highly heterogeneous subset of AML. Here we used deep sequencing technology to describe the spectrum of mutations in 39 genes and clinical features of CN-AML in 152 patients.We identified 503 mutations in 145 (95.39%) patients, with the median number of 3 mutations per case. Nine genes (NPM1, CEBPAbi, DNMT3A, GATA2, NRAS, TET2, FLT3-ITD, IDH2, WT1) mutated in more than 10% patients. Function groups of myeloid transcription factors, activated signaling and DNA methylation mutated most. Three sets of co-occurrence and seven sets of mutual exclusivities existed in genes, three sets of overlapping and incompatibility in function groups were observed in genetic interaction analysis. The distribution of variant allele frequencies (VAF) of recurrent genes was different among functional groups, and obvious diversities of VAF existed in same genes among different mutation points were also observed.Compared with the western studies, the mutation frequencies of NPM1 and DNMT3A genes were lower in Asian CN-AML patients. High mutation rates of CEBPA and GATA2 together with the low frequency of FLT3-ITD mutation seemed to be the distinct characteristics of Chinese patients. Obviously, the mutation distribution differed significantly among different regions and ethnicities.Correlation were shown between genotype and clinical features. There were significant distinctions in the distribution of genotypes in FAB subtypes. NPM1 and DNMT3A mutated most in M5 while CEBPAbi and GATA2 mutated most in M2. In M4, leukemia cells are composed of both myeloid and monocytic blasts, the mutation pattern was uniformly distributed.Prognosis analysis identified CEBPAmo mutation as an inferior factor. Combined with NCCN guidelines, the FLT3-ITD, TP53, DNMT3A, CEBPAmo and WT1 mutations were selected as high-risk markers, which could further differentiate the patients with poor prognosis of CN-AML, suggesting the potential application of guidance in clinical treatment.In conclusion, we reported here a comprehensive study of gene mutations in 152 CN-AML patients, gained insights into unique genetic profile and better risk consultation in CN-AML patients, providing valuable information for clinical practice. DisclosuresNo relevant conflicts of interest to declare.

Molecular Genetic Classification of Brazilian Acute Myeloid Leukemia and Myelodysplastic Syndromes Patients

Introduction:Until the last decade, AML diagnoses relied on cell morphology and immunophenotyping by flow cytometry. Patients were stratified into good risk, intermediate or unfavorable according to karyotype. Treatment consisted of induction, consolidation and intensification or hematopoietic stem cell transplantation. Despite, rigorous attempt to follow standard therapeutic protocols and overcome adversities, results among Brazilian patients were dismal. In the last ten years, with the introduction of molecular tests for FLT3, NPM1, CEBPA and C-Kit, some of the good and intermediate risk patients moved to unfavorable, justifying the adverse response to therapy, especially in the favorable group. However, more recently, the advent of next generation sequencing (NGS) can provide a still better stratification in order to the optimization of therapy. MDS patients follow a somewhat similar pathway in respect to diagnosis, classification, prognosis and therapy decisions.Objective:The objective of this study was to reclassify AML/MDS Brazilian patients according to molecular genetic aberrations.Methods:14 AML (excluding promyelocytic leukemia) and 16 MDS patients were diagnosed according to WHO, 2016, after morphological analysis, immunophenotyping (with a wide panel of MoAb) and G-banding karyotype (ISCN, 2016). AML patients mean age was of 49,8 years and M:F ratio was 0.75 while MDS mean age was 62.5 yrs and M:F ratio = 1.3. One AML patient was classified as favorable risk, 11 as intermediary and two unfavorable. According to IPSS-R, among MDS, 14 patients were favorable and two intermediary risk. NGS and Sanger sequencing were performed in peripheral blood or bone marrow samples collected in EDTA tubes, for the following mutations: in RNA-splicing machinery (SF3B1, SRSF2, ZRSR2, U2AF1, U2AF2); DNA methylation (TET2, DNMT3A, IDH1/2); chromatin modification (ASXL1, EZH2); transcription factor (TP53, RUNX1); signal transduction/kinases (FLT3, JAK2); RAS pathway (KRAS, NRAS, CBL, NF1, PTPN11); cohesin complex (STAG2, CTCF, SMC1A, RAD21); and DNA repair (ATM, BRCC3, DLRE1C, FANCL)Results: All patients presented at least one or more molecular mutations with a mean of 2.43 mutations per AML case and 1.87 per MDS. Nine out of 11 AML patients classified as intermediary risk group (karyotype with +8 or normal) were moved into unfavorable prognosis after the molecular analyses, since they presented SRSF2, RUNX1, FLT3, TP53 and U2AF1 mutations. Besides that, one initially good prognosis patient (presence of inv16) showed FLT3 -ITD mutation, and was then considered as unfavorable risk. Only one patient classified as intermediary risk (normal karyotype) was reclassified into favorable (presence of NPM1 mutation). The genetics classification ended in only one favorable, one intermediary and 12 unfavorable patients. Referring to MDS, four patients with good prognosis (normal karyotypes) were moved to unfavorable due to the presence of ASXL1 mutations. Still, one MDS patient with 7% of ring sideroblasts and normal karyotype presented SF3B1 mutation and remained in the favorable risk.Conclusion and discussion:Genetic classification allowed a better allocation of AML patients and showed that the majority of them (85%) were stratified to unfavorable risk. This scenario explains the dismal treatment results obtained when patients were classified by karyotyping, FLT3, NPM1, CEBPa and CKIT only. Furthermore, approximately 40% to 50% of patients with AML have a normal karyotype at diagnosis, and this represents the largest subset of AML (Mitelman et al, 2010). All such cases are categorized in the intermediate-risk group; yet, this group is quite heterogeneous, and not all patients present the same response to treatment. This is likely a result of the large variability in gene mutations and gene expression in this population. Similarly, 50% of patients with MDS have normal karyotype, a heterogeneous group as well, and molecular alterations allowed classifying them differently. With the genetic classification approach precision therapy is being introduced aiming for better results. DisclosuresNo relevant conflicts of interest to declare.

DNA Repair Expression Profiling to Identify High-Risk Cytogenetically Normal Acute Myeloid Leukemia and Define New Therapeutic Targets

Cytogenetically normal acute myeloid leukemias (CN-AML) represent about 50% of total adult AML. Despite the well-known prognosis role of gene mutations such as NPM1 mutations or FLT3 internal tandem duplication (FLT3-ITD), clinical outcomes remain heterogeneous in this subset of AML. Given the role of genomic instability in leukemogenesis, expression analysis of DNA repair genes might be relevant to sharpen prognosis evaluation in CN-AML. Publicly available gene expression profile dataset from two independent cohorts of patients with CN-AML were analyzed (GSE12417). A list set of 175 genes involved in six major DNA repair pathways (base excision repair (BER), NER, mismatch repair (MMR), homologous recombination repair (HRR), non-homologous end joining (NHEJ) and FANC pathways) was defined using the REPAIRtoire database (http://repairtoire. genesilico.pl) and review of the literature. We investigated the prognostic value of these 175 genes involved in DNA repair. Among these genes, 23 were associated with a prognostic value, using the MaxStat R function. To further corroborate gene expression data on a functional level, CRISPR or RNAi screening publicly available data were used (Dependency Map data, Broad Institute, www.depmap.org). Among the 19 genes associated with a poor outcome, APEX (BER), RTEL1 (HRR) and COPS6 (NER) were identified as significant essential AML genes (p = 7.9e-05, 3.4e-04 and 2.8e-04 respectively). The prognostic information provided by these 23 genes was summed (sum of the beta coefficients of the Cox model for each prognostic gene, weighted by +1 or -1 according to the patient signal ≥ the probe set MaxStat value) in a DNA repair score to consider connection of DNA repair pathways. In the CN-AML training cohort (n=162), DNA repair score allowed to define a group of patients (n=87; 53,7%) with poor median overall survival (OS) of 233 days (95% CI: 184-260). These results were confirmed in the validation cohort (n=78) (median OS: 120 days; 95% CI: 36-303). In multivariate Cox analysis, the DNA repair score, NPM1and FLT3-ITD mutational status remained independent prognosis factors in CN-AML. Therefore, we investigated the interest of combining DNA repair score and NPM1/FLT3 mutational status to predict CN-AML outcome. Patients were classified according to prognosis value of DNA repair score (0 point for group I; 1 for group II; 2 for group III), and NPM1/FLT3 mutational status (0 point if NPM1 mutation without FLT3-ITD; 2 points if FLT3-ITD without NPM1 mutation; 1 point in other situations). The sum of the prognostic information was computed for all patients, allowing to separate patients in three new prognostic groups: group A including patients with 0 or 1 point, group B for patients with 2 points and group C for patients with 3 or 4 points. Combining these parameters allowed the identification of three risk groups with different clinical outcomes in both training and validation cohorts (Figure 1). In the training cohort, median OS was not reached (95% CI: NR-NR), 326 days (95% CI: 127-NR) and 236 days (95% CI: 190-263) respectively for patients in groups A, B and C. One-year OS was 90.3% (95% CI: 80.5-100) in group A, 49.3% (95% CI: 37.1-65.7) in group B, and 24.2% (95% CI: 16.2-36.2) in group C.These results were confirmed in the validation cohort where median OS was not reached (95% CI: 1278-NR), 516 days (95% CI: 308-NR) and 253 days (95% CI: 52-403) for patients respectively in groups A, B and C. One-year OS was 92.6% (95% CI: 83.2-100) in group A, 54.9% (95% CI: 39.8-75.7) in group B, and 26.5% (95% CI: 12.4-55.8) in group C. OS was statistically different between groups A, B and C in both training and validation cohorts. Combined with NPM1 and FLT3 mutational status, our GE-based DNA repair score might be used as a biomarker to predict outcomes for patients with CN-AML. DNA repair score has the potential to identify CN-AML patients whose tumor cells are dependent on specific DNA repair pathways to design new therapeutic avenues. Disclosures Cartron: Celgene: Consultancy, Honoraria; F. Hoffmann-La Roche: Consultancy, Honoraria; Sanofi: Honoraria; Gilead: Honoraria; Jansen: Honoraria; Abbvie: Honoraria. Moreaux:Diag2Tec: Consultancy.

Initial Results from a Biomarker-Directed Phase 2 Trial of SY-1425, a Potent and Selective RARα Agonist, in Combination with Azacitidine in Relapsed/Refractory Acute Myeloid Leukemia

Initial Results from a Biomarker-Directed Phase 2 Trial of SY-1425, a Potent and Selective RARα Agonist, in Combination with Azacitidine in Relapsed/Refractory Acute Myeloid Leukemia

Aberrant Telomere Length and Composition Are Recurrent Features of Myeloid Disorders

Background: In cancer evolution, development of telomere maintenance mechanisms represents a critical feature to achieve replicative immortality and prevent cell death after shortening below a critical threshold. Several studies have indicated that excessive telomere shortening or aberrant telomeric repeats are present in cancer genomes, however, their precise role in myeloid malignancies remains largely unexplored. We stipulated that immature leukemia-initiating cells (LIC) would display higher telomerase activity and thus longer telomeres while in mature LIC lower telomerase activity would be associated with shorter telomeres due to excessive division rate. Additional lesions may also lead to disturbed differentiation programs resulting in altered expression of telomerase. Taking advantage of a vast cohort (N=1648) of patients with myeloid neoplasia analyzed by whole genome sequencing (WGS, mean. coverage 100x) we determined telomere length (TL). Aim: We set out to hypothesize that WGS data can be used to reliably measure the TL and that if systematically applied to well defined patient cohorts, we could test whether preservation of TL or its excessive shortening are prognostically or biologically relevant to clinical outcome. Methods: Our cohort included patients with AML (N=732), MDS (N=701) and MPN (N=102 incl. CML, ET, PMF, PV). Healthy subjects (N=11) and patients with the clonal non-malignant HSC disorder PNH (N=102) served as controls. To retrieve TL and telomere repeat heterogeneity via singleton analysis from WGS data we used two independent bioinformatic tools TelomereHunter and Telseq (1A). Results: First, using bidirectional cross-confirmation with independent bioinformatics pipelines, reliability of WGS-based TL measurements (1B) was confirmed and, as anticipated, age-dependency of TL in healthy controls was reproduced (1C). Next we investigated age as a possible confounder of TL shortening in patients. While in PNH age had an expected impact, patients with AML or MDS exhibited TL decoupled from its inherent age-dependence (r=.00034, P=.72). In general, we noted TL shortening as an overarching feature of myeloid proliferation irrespective of entity (1D). Furthermore, mutations in the promotor region of TERT were not correlated with TL. In AML or MDS, neither morphological nor genetic subclassification where associated with differences in TL (1E), with the exception of a small subset of AML and MDS patients (n=103, n=97), harboring abnormally long telomeres compared to healthy subjects. These were significantly associated with mutations in TP53 (AML P=1.91x10-6, mean TL 558.4; MDS P=.55x10-7, mean TL 414.5) but not with any other common mutation. This is surprising, since TP53 mutations are thought to protect cells from telomeric catastrophe following excessive shortening. Moreover, comparing the VAF of TP53mut AML samples with the corresponding measured TL, revealed a positive correlation (r=.2281, P=.021), irrespective of the blast count by cytomorphology (1F). Singleton analysis from all 1648 patients revealed that in AML and MDS patients, alternative telomere repeat- variants (chiefly TTTGGG) were increased pointing towards a possible implication of alternative telomere extension (1G). Lastly, outcome analysis revealed that in AML, the presence of longer telomeres (compared to controls) was associated with worse prognosis (P=.0035) (1H). Multivariate analysis, which included age, karyotype, TP53 status and gender, revealed that this was due to the strong association with TP53 mutations (n=630 vs n=103). The implications of our study extend beyond the realm of telomere biology and leukemia pathophysiology. Therapies targeting the telomere machinery, e.g. imetelstat, may offer an opportunity for application beyond MPN, its current indication. Ongoing studies are currently investigating its use in MDS. Based on our findings we argue that treatment may be rationally evaluated in TP53mut AML. Conclusion: Our findings provide a rationale for (1) the use of WGS to investigate TL, (2) the fundamental discovery that TL shortening is a common principle in myeloid neoplasms, (3) TP53 mutations being accompanied with increased TL and (4) the exploitation of TL to identify subgroups eligible for treatment with imetelstat. Moreover, WGS-based TL analysis in leukemias can contribute to our understanding of TL maintenance mechanisms in leukemic genome. Disclosures Maciejewski: Novartis, Roche: Consultancy, Honoraria; Alexion, BMS: Speakers Bureau.

Preliminary Data on a Phase 1/2A First in Human Study of the Menin-KMT2A (MLL) Inhibitor KO-539 in Patients with Relapsed or Refractory Acute Myeloid Leukemia

Preliminary Data on a Phase 1/2A First in Human Study of the Menin-KMT2A (MLL) Inhibitor KO-539 in Patients with Relapsed or Refractory Acute Myeloid Leukemia

DNA Repair Expression Profiling to Identify High-Risk Cytogenetically Normal Acute Myeloid Leukemia and Define New Therapeutic Targets.

Simple SummaryAcute myeloid leukemia (AML) is the second most frequent type of adult leukemias. Cytogenetically normal acute myeloid leukemias represent about 50% of total adult AML, exhibit no chromosomal abnormalities, and present high heterogeneity regarding the clinical outcome. Deregulation of DNA repair mechanisms is involved in the adaptation of cancer cells to replicative stress and resistance to genotoxic agents. We investigate the prognostic value of genes related to the major DNA repair pathways. The data reveals specific patterns of gene expression in CN-AML that have prognostic value. Combined with NPM1 and FLT3 mutational status, our gene expression-based DNA repair score might be used as a biomarker to predict outcomes for patients with CN-AML. DNA repair score has the potential to identify CN-AML patients whose tumor cells are dependent on specific DNA repair pathways to design new targeted therapies.Cytogenetically normal acute myeloid leukemias (CN-AML) represent about 50% of total adult AML. Despite the well-known prognosis role of gene mutations such as NPM1 mutations of FLT3 internal tandem duplication (FLT3-ITD), clinical outcomes remain heterogeneous in this subset of AML. Given the role of genomic instability in leukemogenesis, expression analysis of DNA repair genes might be relevant to sharpen prognosis evaluation in CN-AML. A publicly available gene expression profile dataset from two independent cohorts of patients with CN-AML were analyzed (GSE12417). We investigated the prognostic value of 175 genes involved in DNA repair. Among these genes, 23 were associated with a prognostic value. The prognostic information provided by these genes was summed in a DNA repair score, allowing to define a group of patients (n = 87; 53.7%) with poor median overall survival (OS) of 233 days (95% CI: 184–260). These results were confirmed in two validation cohorts. In multivariate Cox analysis, the DNA repair score, NPM1, and FLT3-ITD mutational status remained independent prognosis factors in CN-AML. Combining these parameters allowed the identification of three risk groups with different clinical outcomes in both training and validation cohorts. Combined with NPM1 and FLT3 mutational status, our GE-based DNA repair score might be used as a biomarker to predict outcomes for patients with CN-AML. DNA repair score has the potential to identify CN-AML patients whose tumor cells are dependent on specific DNA repair pathways to design new therapeutic avenues.

Abstract 1756: JBI-802, novel dual inhibitor of LSD1-HDAC6 for treatment of cancer

Abstract Introduction: Lysine Specific Demethylase 1 (LSD1) is over-expressed in many cancers and down-regulation of LSD1 has been shown to effectively treat cancers by inducing re-expression of aberrantly silenced genes. Studies have shown that LSD1 may contribute to acute myelogenous leukaemia pathogenesis by inhibiting the normal pro-differentiative function of ATRA, paving the way for new combinatorial therapies for AML. In addition, LSD1 inhibition also leads to convertion of ‘cold' tumors to ‘hot' tumors by activating type 1 interferon response. Similarly, HDAC6 inhibition enhances immune response by over-coming immune suppression. Further, combined inhibition of LSD1 and HDAC has been shown to be more efficacious in inhibiting multiple cancers. Here, we show that JBI-802, a dual inhibitor targeting both LSD1 and HDAC6/8 shows stronger efficacy in several cancers, including AML, lymphoma and others and well tolerated at efficacious doses. Methods: Computational chemistry approaches were used to design LSD1 specific and LSD1-HDAC dual inhibitors. To assess in vitro LSD1 potency, TR-FRET assay was used. For assessing in vitro HDAC activity fluorescence based HDAC6 activity assay was performed. Western blotting and quantitative PCR were used to assess biomarkers of LSD1 and HDAC inhibition. Alamar blue cytotoxicity assay was used to assess cell proliferation. Xenograft and syngeneic models were used to assess in vivo efficacy. Results: Several compounds from this series show strong in vitro potency against LSD1 with excellent selectivity against MAOs. JBI-802 our lead dual molecule, showed an IC50 of 0.05 μM against LSD1 and isoform selective HDAC6/8 activity (IC50 of 0.011 and 0.098 µM on HDAC6 and HDAC8, respectively), with about 100 fold selectivity against other HDAC isoforms. When assessed in a panel of 25 cell lines, JBI-802 showed strong anti-proliferative activity on select AML, CLL, SCLC, sarcoma and multiple myeloma cell lines. In cell based and in vivo target engagement studies there was significant dose-dependent increase in CD11b, CD86 and GFI1b and tubulin acetylation levels. JBI-802 had a much stronger effect in inhibiting the growth of HEL92.1.7 erythroleukemia xenograft by oral administration when compared to ACY-1215, a HDAC6 selective inhibitor or ORY-1001 an LSD1 selective inhibitor. ED50 of JBI-802 in this model was ~6.25 mg/kg BID. Further, dose-dependent modulation of above mentioned biomarkers and inhibition of c-Myc could be observed in tumors. Stronger tumor growth inhibition was also observed in other haematological cancers models such as CMK-1 (leukemia) and Z-138 (lymphoma). In addition, JBI-802 showed comparable single agent activity to anti-PD-1/PD-L1 mAbs in a syngeneic murine colon cancer model CT26 and resulted in stronger tumor growth inhibition when combined with these antibodies. 14-day dose ranging finding toxicology studies clearly show that the molecule was well tolerated up to 300 mg/kg. Conclusion: The data obtained to date demonstrate that dual LSD1-HDAC6/8 inhibitors could serve as novel, effective therapeutic agents for treatment of select subset of AML and other cancers. IND-enabling studies are in progress with this inhibitor to be developed as a clinical candidate Citation Format: Dhanalakshmi Sivanandhan, Sridharan Rajagopal, Sreekala Nair, Basavaprabhu B, Reshma Dhkar, Santosh Viswakarma, Amir Siddiqui, Mohd Zainuddin, Rudresh G, Prashanthi Daram, Sunil Mohire, Krishnakumar V. JBI-802, novel dual inhibitor of LSD1-HDAC6 for treatment of cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1756.

Interrogation of molecular profiles can help in differentiating between MDS and AML with MDS-related changes

A subset of AML with myelodysplastic syndrome (MDS)-related changes (MRCs) occurs without a documented MDS phase. We studied genomic profile of 646 patients: 310 with MDS, 167 with AML without (w/o) MRC, 99 with primary (p) AML-MRC, and 70 with secondary (s) AML-MRC and sought to find differences in mutational patterns. Among the 32-myeloid associated genes studied, SF3B1 (p ≤ .001) was significantly mutated in higher proportion of patients with MDS, compared to other categories. NPM1 (p < .001), FLT3 ITD (p = .08), and NRAS (p = .02) mutations showed trend toward significance for AML w/o MRC, compared to other categories. In pAML-MRC, TP53 (p < .001) was significantly mutated in higher proportion of patients. Similarly, SETBP1 (p = .001), RUNX1 (p = .004), and SRSF2 (p = .04) mutations were more commonly seen in sAML-MRC. While these signatures may not be diagnostically discriminatory, they may help in disease categorization when other data are absent or in challenging cases.

Predictors of Outcomes in Adult Patients with Therapy Related Acute Myeloid Leukemia Undergoing Allogeneic Hematopoietic Stem Cell Transplant - Twenty Year Experience from a Tertiary Care Centre

<h3>Background</h3> Therapy related acute myeloid leukemia (t-AML) constitutes a subset of AML that has an increased proportion of high risk cytogenetic and molecular features, poorer response to therapy, higher relapse and decreased overall survival. Allogeneic hematopoietic stem cell transplantation (HSCT) offers a potential for cure in t-AML, with survival up to 40% at 2 years. Most studies on HSCT in t-AML have limited number of patients and are confounded by the inclusion of patients with secondary AML and t-MDS. We aim to report our 20-year experience of HSCT in t-AML. <h3>Patients and Methods</h3> All cases of t-AML who underwent HSCT at our centre from June 1999 to July 2019 were reviewed. Demographic characteristics, prior malignancy, cytogenetic and molecular characteristics of AML, induction treatment received, transplant details as well as post-transplant complications were collected. Primary outcome evaluated was overall survival and secondary outcomes were relapse rate and relapse free survival (RFS). Predictors of survival was identified using Cox-proportional hazards regression model. <h3>Results</h3> Fifty-two patients underwent HSCT for t-AML. 58% were male (n=30) and median age at HSCT was 55.5 years. Performance status prior to transplant was ECOG score 0/1 in 71.2% (n=37) and score 2 in 28.8% (n=15) patients. Myeloablative conditioning was given for 30.8% (n=16) and reduced intensity conditioning was given for 69.2% (n=36) patients. Transplant related mortality before day+100 was 21.1% (n=11). Acute and chronic GVHD occurred in 61.5% (n = 32) and 28.8% (n=15) patients respectively. Cumulative incidence of relapse at 2 years was 17.3%. OS and RFS at 2 years were 44.1% and 41.3% respectively (Figure 1 a,b). Significant predictors of reduced OS (Figure 2 a,b,c) after day+100 of HSCT by Cox regression were ECOG performance status 2 (Hazard ratio(HR) Ð 6.1), GVHD prophylaxis CSA/methotrexate (HR Ð 4.5) and haploidentical donor transplantation (HR Ð 34). Cytogenetic risk stratification (ELN 2017) was a predictor of OS in univariate analysis but not in multivariate regression analysis. <h3>Conclusions</h3> Patients with favourable cytogenetic profile, good performance status and having HLA matched donor may have better outcomes after allogeneic HSCT in t-AML. However, overall outcomes of patients with t-AML remain suboptimal and more research is needed in this sub-group of AML.

Ectopia associated MN1 fusions and aberrant activation in myeloid neoplasms with t(12;22)(p13;q12)

Chromosome translocation t(12;22)(p13;q12)/MN1-ETV6 and MN1 overexpression confer a subset of adverse prognostic AML but so far lack in-depth research. We focused on the clinical course and comprehensive genetic analysis of eight cases with t(12;22)(p13;q12) and one with t(12;17;22) (p13;q21;q13) to elucidate their molecular etiology and outcomes of allogeneic hemopoietic stem cell transplantation (allo-HSCT). The total incidence of t(12;22)(p13;q12) and related translocations was 0.32% in myeloid neoplasms. These patients were confirmed to have dismal prognosis when treated only with chemotherapy, and we firstly provided evidence that they can significantly benefit from timely allo-HSCT. Five cases were MN1-ETV6 positive, and a novel MN1-STAT3 fusion was identified in the patient with triadic translocation. Significant MN1 overexpression was observed in all three MN1-fusion-negative cases. Genetic analysis highlighted the evidence of an ectopic super-enhancer associated orchestrated mechanism of MN1 overexpression and ETV6 haploinsufficiency in t(12;22)(p13;q12) myeloid neoplasms, rather than the conventional thought of MN1-ETV6 fusion formation. We also disclosed the high concomitance of trisomy 8 and 531 Kbps focal 8q duplication in t(12;22)(p13;q12) cases. The new perspective about this entity of disease will enlighten further research to define the mechanism of tumorigenesis and discover effective treatments for MN1-driven malignancies.

Gilteritinib and Venetoclax Synergize to Eliminate FLT3/ITD+ Leukemia Cells through BIM

Gilteritinib and Venetoclax synergize to eliminate FLT3/ITD+ leukemia cells through BIM Abstract Acute myeloid leukemia (AML) is characterized by a clonal proliferation of immature myeloid cells in the bone marrow or other tissues. The most commonly mutated gene in AML is FMS-like tyrosine kinase (FLT3). FLT3 downstream signaling pathways include PI3K/AKT, STAT5 and MAPK, which affect apoptosis, differentiation and cell proliferation. An internal tandem duplication mutation in FLT3 (FLT3/ITD) is identified in approximately 25% of patients with AML, and this mutation is associated with a particularly poor prognosis. A subset of AML also has a point mutation in the tyrosine kinase domain of FLT3 (FLT3/TKD); however, this mutation does not have the same pronounced impact on prognosis as the FLT3/ITD mutation. Since tyrosine kinases are an attractive drug target, tyrosine kinase inhibitors (TKIs) that target FLT3 have been developed, including recent agents that show enhanced specificity, such as Gilteritinib. Despite these advances, TKI monotherapy continues to show limited success, indicating that combination therapy is likely necessary for the effective treatment of FLT3/ITD AML. As FLT3 signaling pathway activity is known to be anti-apoptotic, in this study we investigated the combinatorial effect of a FLT3-selective TKI and the BCL-2 inhibitor Venetoclax. The BCL-2 protein plays a key role in apoptosis, with anti-apoptotic/prosurvival effects. Venetoclax is a selective BCL-2 inhibitor, and is used clinically in the treatment of chronic lymphocytic leukemia and relapsed/refractory AML. We first investigated the combinatorial effect of treatment with Gilteritinb and Venetoclax in a FLT3/ITD+ leukemia cell line (Molm14). Combined treatment with Gilteritinib (20nM) and Venetoclax (80nM) reduced cell proliferation by 83.7%, as compared to Gilteritinib (62.2%, P&lt;0.05) or Venetoclax (37.3%, P&lt;0.05) alone. The drug combination demonstrated synergy (CI=0.42). Combined treatment shows that these agents act synergistically to enhance apoptosis (88.2%), as compared with Gilteritinib (52.1%, P&lt;0.05) or Venetoclax (12.1%, P&lt;0.05) alone. Moreover, the combined treatment also significantly reduced cell proliferation in patient samples with FLT3/ITD+ and FLT3/TKD mutations. Further experiments with cell line (Molm14 cells resistant to 60nM CEP-701) showed that Venetoclax can re-sensitize FLT3 TKI-resistant cell lines to TKI treatment. Western blot analysis indicates that this effect is mediated by inhibiting MAKP pathway-whose reactivation is an important reason for TKI resistance in FLT3/ITD+ patients. These data demonstrate that combined treatment with Gilteritinb and Venetoclax reduces cell proliferation and enhances apoptosis in a FLT3/ITD leukemia cell line. We next investigated the mechanism of action of this drug combination. BCL-2 and other antiapoptotic proteins in this subfamily (i.e. MCL-1, BCL-XL) exert their pro-survival effects by sequestering BIM. BIM is a proapoptotic protein, that, when released by BCL-2, can activate cell death mediators. These cell death mediators include BAX and BAK, which perforate the mitochondrial membrane, resulting in apoptosis. Venetoclax acts by binding to BCL-2 and displacing BIM, freeing BIM to associate with and activate cell death mediators, resulting in apoptosis. We found that in Molm14 cells, Venetoclax not only dissociates BIM from BCL2, but also decreases expression of BIM and enhances the binding of BIM and MCL-1. In contrast, we found that Gilteritinib increases expression of BIM, reduces the binding of BIM with MCL-1 via a reduction of MCL1 expression, and shows enhanced binding of BIM with BCL2. Cells treated with both Venetoclax and Gilteritinib show dissociation of BIM from both BCL2 and MCL-1, though the interaction between BIM and BCL-XL is not affected. Combination treatment also showed increased binding between BIM and the cell death mediator BAX, leading to increased apoptosis. These studies provide evidence that the addition of Venetoclax may enhance TKI therapy in the treatment of FLT3/ITD leukemia. Additionally, these findings suggest that enhanced cell death in FLT3/ITD AML cells treated with combination therapy occurs because Venetoclax mitigates unintentded pro-survival effects of the TKI, including an increase in BIM expression and increased association between BIM and BCL-2. Disclosures Duffield: MedImmune: Consultancy; Boston Biomedical/Sumitomo Dainippon Pharma Co., Ltd.: Consultancy, Membership on an entity's Board of Directors or advisory committees. Small:Pharos I, B &amp; T: Consultancy, Research Funding; InSilico Medicine: Membership on an entity's Board of Directors or advisory committees.

A Novel Machine Learning Derived Genomics-Based Scoring System Is Highly Predictive of Outcome in Core Binding Factor Acute Myeloid Leukemia

Introduction: Core binding factor acute myeloid leukemia (CBF-AML) is one of the commonest subtypes of AML characterized presence of t(8;21)(q22;q22) or inv(16)(p13q22)/t(16;16)(p13;q22). It is characterised by a high frequency of somatic mutations especially in RAS and tyrosine kinase signalling pathways. Here we investigated the feasibility of improving risk prediction of CBF-AML using machine learning algorithms. Methods: We developed a next generation sequencing panel that targeted 50 genes implicated in the pathogenesis of myeloid malignancies using single molecule molecular inversion probes. This panel was used to sequence 106 patients of CBF-AML accrued over a six year period (March 2012 - December 2018) treated with conventional "3 + 7" chemotherapy. Post data analysis, we devised a supervised machine learning (ML) approach for identification of mutations most likely to predict for favorable outcome in CBF-AML. We included somatic mutations in genes occurring in CBF-AML at a frequency of &gt;5%. A total of 11 variables were included for feature selection to predict for favorable outcome (including mutations in ASXL2, CSF3R,FLT3, KIT, NF1, NRAS, RAD21, TET2 and WT1 genes as well as mutation burden). Approaches for supervised ML were naïve bayes, generalized linear model, logistic regression, deep learning and random forest methods. Based on the ML results top 6 selected variables were allotted an individual score. A final score for that case was devised as a sum total of the individual scores. These sum were used to generate a genetic risk for a patient. Overall survival (OS) was calculated from date of diagnosis to time of last follow up or death. Relapse free survival (RFS) was calculated from date of CR till time to relapse or death or last follow up if in CR. Results of the genetic risk were analyzed for their impact on OS and RFS using log rank test. Multivariate analysis was performed using cox proportional hazards regression model. Results: The median follow up of the cohort was 27.6 months. A total of 181 somatic mutations were identified in this subset of AML with 86.7% harbouring at least one somatic mutation (median = 2). Based on ML data, a genetic score was formulated that incorporated mutations in RAD21, FLT3, KIT D816, ASXL2, NRAS genes as well as high mutation burden (≥2) into two genetic risk classes (favorable risk and poor ML derived genetic genetic risk). Patients classified as poor genetic risk had a significantly lower OS [median OS: 34.8 months; 95% confidence interval (CI) (14.2-34.8); p=0.0086] and RFS [median RFS: 17.9 months; 95%CI (12.7-33.6); p=0.0043] as compared to patients with favorable genetic risk (median OS and RFS not reached). These results can be seen in Figure 1. On multivariate analysis poor genetic risk was the most important independent risk factor that predicted for inferior OS [hazard ratio(HR), 2.7; 95% CI 1.3 to 5.7] and RFS (HR, 2.6; 95% CI:1.3 to 5.1). Conclusions In a proof of concept, we describe a novel ML derived genomics scoring model that provides a mechanism to risk stratify CBF-AML, a seemingly homogeneous disease entity. This study, to the best of our knowledge represents a novel application of ML to CBF mutated AML. Our data indicates that this scoring system will be useful in identifying CBF mutated AML patients who are at higher risk of relapse and distinguishes them from patients who are truly good risk. Figure 1 Disclosures No relevant conflicts of interest to declare.

Predictors of Outcomes in Adult Patients with Therapy Related Acute Myeloid Leukemia Undergoing Allogeneic Hematopoietic Stem Cell Transplantation - Twenty Year Experience from a Tertiary Care Centre

Background Therapy related acute myeloid leukemia (t-AML) constitutes a subset of AML that has an increased proportion of high risk cytogenetic and molecular features, poor response to therapy, higher relapse, and decreased overall survival. The incidence ranges from 10-20% across various studies. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) offers a potential cure in t-AML, with disease free survival reported up to 30% at 2 years. Most studies on HSCT in t-AML have limited number of patients and are confounded by the inclusion of patients with secondary AML and therapy related myelodysplastic syndromes (t-MDS). We aim to report our 20-year experience of allo-HSCT in t-AML and identify predictors of survival. Patients and Methods We retrospectively reviewed all cases of t-AML who underwent allo-HSCT at our centre from June 1999 to July 2019. We collected data for demographic characteristics, prior malignancy and treatment, latent period before AML, cytogenetic and molecular characteristics of AML, induction treatment received, transplant details (donor details, conditioning regimens, GVHD prophylaxis) as well as post-transplant complications (transplant related mortality, occurrence and severity of acute and chronic GVHD, CMV and EBV reactivations). Primary outcome evaluated was overall survival and secondary outcomes were relapse rate and relapse free survival (RFS). Cox-proportional hazards regression model was used to identify predictors of survival. Results Fifty-two patients underwent allo-HSCT for t-AML during the study period. 58% were male (n=30). Median age at HSCT was 55.5 years (Range 18-70). Baseline characteristics are summarized in Table 1. Complex cytogenetics were present in 23.2% (n=12) patients, while 11q23 rearrangement (MLL-KMT2A), monosomy 7, monosomy 5, and 17p deletion were present in 15.4% (n = 8), 15.4% (n= 8), 7.7% (n = 4) and 7.7% (n=4) patients respectively. Based on the ELN 2017 risk stratification schema, 13.5% (n=7) patients could be considered favorable risk based on cytogenetic and molecular profiles, 36.5% (n=19) intermediate risk, and 57.7% (n=30) in poor risk category. Performance status prior to transplant was ECOG 0/1 in 71.2% (n=37) patients, and 2 in 28.8% (n=15) patients. Myeloablative conditioning was used in 30.8% (n=16) patients, and reduced intensity conditioning in 69.2% (n=36). GVHD prophylaxis was CyclosporineA(CSA)/Methotrexate in 23.1% (n=12), Alemtuzumab/CSA in 21.2% (n=11), ATG/CSA/PTCy in 28.8% (n=15), and other regimens in 26.9% (n=14) patients. Transplant related mortality (death before day+100) was 21.1% (n=11). Acute and chronic GVHD (any grade) occurred in 61.5% (n = 32) and 28.8% (n=15) patients respectively. Eleven patients (21.2%) relapsed with a median RFS of 8 months (Range 0.17-158). Median OS of the whole cohort was 8.9 months (0.17-158 months). No patient had a relapse of their primary malignancy during follow up. RFS at 12 and 24 months were 46% and 28% while OS at 1 and 2 years was 46% and 30%, respectively. Significant predictors of reduced OS (Figure 1a,b,c) after day+100 of HSCT by Cox-regression were ECOG performance status 2 (Hazard ratio - 6.1; p value 0.003), GVHD prophylaxis with CSA/methotrexate (HR - 4.5; p value 0.01), and haploidentical donor transplantation (HR - 34; p value - 0.002). Cytogenetics were a predictor of OS in univariate analysis but not in multivariate regression analysis. No difference in OS was found between patients who underwent HSCT 2010 or prior versus after 2010. Conclusions Patients with favorable cytogenetic profile, better performance status, and an HLA matched donor may have better outcomes after allo-HSCT in therapy related AML. Patients with unfavorable cytogenetic risk profiles may require more intense therapy or post-transplant maintenance therapy to prevent relapse. Disclosures Michelis: CSL Behring: Other: Financial Support. Mattsson:Gilead: Honoraria; Celgene: Honoraria; Therakos: Honoraria.

Abstract 11: Novel dual inhibitors of LSD1-HDAC6/8 for treatment of cancer

Abstract Introduction: Lysine Specific Demethylase 1 (LSD1) is a flavin adenine dinucleotide (FAD)-dependent amine oxidase that has been reported to be over-expressed in many malignant tumors. Down-regulation of LSD1 has been shown to effectively treat cancers by inducing re-expression of aberrantly silenced genes. Studies have shown that LSD1 may contribute to acute myelogenous leukemia pathogenesis by inhibiting the normal pro-differentiative function of ATRA, paving the way for new combinatorial therapies for AML. Similarly, HDAC isoform selective inhibitors are beginning to be explored as less toxic alternatives to panHDAC inhibitors in select cancers. Further, combined inhibition of LSD1 and HDAC has been shown to be more efficacious in inhibiting multiple cancers. Here, we show that JBI-295, a dual inhibitor targeting both LSD1 and HDAC6/8 shows stronger efficacy without enhancing systemic toxicity, in a subset of AML and JAK-dependent myeloproliferative cancer. Methods: Computational chemistry approaches were used to design LSD1 specific and LSD1-HDAC dual inhibitors. To assess in vitro LSD1 potency, TR-FRET assay was used. For assessing in vitro HDAC activity fluorescence based HDAC6 activity assay was performed. Western blotting was used to assess biomarkers of LSD1 and HDAC inhibition. Alamar blue cytotoxicity assay was used to assess cell proliferation. Results: Several compounds from this series show strong in vitro potency against LSD1 with more than excellent selectivity against MAOs. JBI-295 one of the lead dual molecules, showed strong LSD1 potency (IC50 of 0.07 μM) and isoform selective HDAC6/8 activity (IC50 of 0.006 and 0.08 µM on HDAC6 and HDAC8, respectively), with about 100 fold selectivity against other HDAC isoforms. JBI-295 showed strong anti-proliferative activity on leukaemia and multiple myeloma cell lines. In cell based and in vivo target engagement studies there was a concomitant increase in CD11b, CD86 and GFI1b and tubulin acetylation levels. JBI-295 was more efficacious in inhibiting the growth of leukemia HEL92.1.7 xenograft by oral administration when compared to IP administration of ACY-1215, a HDAC6 selective inhibitor. Stronger tumor growth inhibition was also observed in melanoma A375 xenograft model as compared to inhibitors with single activity. In addition, JBI-295 showed single agent activity in a syngeneic murine colon cancer model CT26 and also resulted in stronger tumor growth inhibition when combined with anti-PDL1 antibody. Conclusion: The data obtained demonstrate that dual LSD1-HDAC6/8 inhibitors could serve as novel, effective therapeutic agents for treatment of select subset of cancer. Advanced efficacy and toxicology studies with JBI-295 are in progress. Citation Format: Dhanalakshmi Sivanandhan, Sridharan Rajagopal, Sreekala Nair, Chandru Gajendran, Dimpy Ghosh, P Nagaraj, Subramanyam J. Tantry, Purushottam Dewang, Mahanandeesha S. Hallur, Kannan Murugan, Srinatha K. C, Damodara Kuntrapaku, M Dilipkumar, R Sharma, S Meghashree, Durga Prasanna Kumar, Suraj P. Ingle, Mohd Zainuddin, A B. Vinod, Sriram Rajagopal. Novel dual inhibitors of LSD1-HDAC6/8 for treatment of cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 11.

PF290 INCIDENCE AND PROGNOSTIC IMPACT OF A BLASTIC PLASMACYTOID DENDRITIC CELL NEOPLASM‐LIKE PHENOTYPE IN AML PATIENTS

Background:Blastic plasmacytoid dendritic cell neoplasms (BPDCN) are rare disorders included by WHO 2016 classification among the myeloid neoplasms and characterized by aggressive clinical course and dismal outcome. BPDCN display a recurrent phenotype (CD45low/CD34‐/CD56+/CD4+/CD123+), in the absence of lineage differentiation. Our group recently reported a negative prognostic impact of CD123, CD56 and CD4 co‐expression (“BPDCN‐like” phenotype) in AML patients with NPM1‐mutation.Aims:The aim of the present study was to evaluate the incidence and the prognostic impact of BPDCN‐like phenotype in a wider cohort of cytogenetically normal AML patients, irrespectively of NPM1‐mutational status.Methods:We retrospectively evaluated a cohort of 83 younger (age &lt;60 yrs), consecutive AMLpatientswith normal karyotype, who have been intensively treated in our institution between 2006 and 2016. All patients were treated with the same fludarabine‐idarubicin‐Ara‐C containing induction. In all patients, four (until 2012) or eight color immunophenotypic analysis has been performed at diagnosis. We defined a BPDCN‐like signature as the combination of at least two among CD56+/CD4+/CD123+ and evaluated its prognostic impact as well as the correlation with biological, molecular and cytogenetic features.Results:Fifteen patients (18%) showed a BPDCN‐like signature. Neither the presence of NPM1‐mutation nor FLT3‐ITD or biallelic CEBPA mutation had a significant correlation with BPDCN‐like signature. BPDCN‐like patients had significantly higher WBC count at diagnosis (p &lt; 0.05). No clear correlation with sex or age at diagnosis was observed. In the whole cohort of patients, only the presence of NPM1‐mutation correlated with the achievement of complete response (CR).Median follow‐up was 63 months, 3‐year Overall Survival (OS) was 52% in the whole cohort (median 38 months). Patient with NPM1 mutation or biallelic CEBPA mutation had a better outcome (p &lt; 0.03). OS was not significantly influenced by the FLT3‐ITD mutation or by the presence of BPDCN‐like signature.However, as we previously reported, in the subgroup of 30 patients with NPM‐1 mut AML, the presence of BPDCN‐like features conferred a dismal prognosis (3‐year OS 25% vs 77% for BPDCN positive and negative patients, respectively, p &lt; 0.001), irrespectively of mutational status for FLT3‐ITD or other clinical features. Even if CR rate was not affected, all NPM1‐mut patient with BPDCN‐like phenotype failed to achieve minimal residual disease (MRD)‐negative CR (p &lt; 0.05).On the contrary the negative prognostic influence of BPDCN‐like phenotype was not observed in the 53 NPM‐1 wt patients (3‐year OS 71% vs 35% for BPDCN‐like positive and negative NPM‐wt patients, respectively, p = 0.156). No difference in MRD clearance probability was observed in this subgroup.Summary/Conclusion:Our extended analysis confirms that a peculiar BPDCN‐like immunophenotype among NPM‐1 mut AML patients is associated with significantly worse outcome in this otherwise favorable subset of AML. Patient with BPDCN‐like features, even if achieving morphological CR showed high level of minimal residual disease after induction, suggesting an intrinsic chemo‐resistance. Interestingly, this observation was strictly restricted to NPM‐1 mutated AML, suggesting that peculiar alterations in this distinct entity contribute to the prognostic impact of BPDCN‐like features in this setting.imageThe biological explanation of this finding is not clear; further gene‐expression profiling studies are ongoing in order to explain our findings.

Clinical Heterogeneity of AML Is Associated with Mutational Heterogeneity

BACKGROUND: AML is a group of clinically heterogeneous diseases. We hypothesized that heterogeneous presentation of AML is a reflection of equally heterogeneous genetic process during the leukemogenesis.METHODS: 536 AML patients (pts) bone marrow samples were analyzed by targeted capture exome sequencing of 295 genes. Extensive clinical-genotype correlation was performed using well annotated clinical data.RESULTS: The median age of the cohort was 62 years (IQR: 51-72) including 297 (55%) elderly (age ≥60), and 239 (45%) young (age <60) pts. Of the 536 pts, 308 (57%) pts had de novo AML (dnAML), and 103 (19%) had secondary or therapy-related AML (stAML). DNA sequencing revealed 1,586 high-confidence somatic mutations (922 SNVs and 664 indels) in 79 genes in 493 (92%) pts [median 3 (IQR 2-4) mutations/patient]. Cytogenetics were favorable in 10 (2%), intermediate in 326 (61%), and adverse in 177 (33%) (All defined by ELN 2017criteria); 23 (4%) pts had no cytogenetics data.Elderly pts and young pts had distinct mutational landscape. (1.3-9.6), p = 0.0079] were significantly more enriched in elderly AML, whereas young AML pts were enriched with mutations in FLT3 [OR 0.6 (0.4-0.9), p = 0.0089], NPM1 [OR 0.5 (0.3-0.9), p = 0.0113], PTPN11 [OR 0.2 (0.2-0.7), p = 0.0033], and WT1 [OR 0.4 (0.2-0.7), p = 0.0033]. Some of the mutations enriched in elderly pts are frequently observed in pts with clonal hematopoiesis with indeterminate potential. Based on the ontogeny of AML, PTPN11 [OR 7.6 (1-57.2), p=0.0210], NPM1 [OR 3.0 (1.5-6.1), p = 0.0007], WT1 [OR 2.9 (1.1-7.4), p=0.0279] mutations were significantly enriched in dnAML, while SF3B1 [OR 0.4 (0.18-0.89), p=0.0376], SRSF2 [OR 0.5 (0.3-0.85), p = 0.0109], TP53 [OR 0.5 (0.3-0.8), p = 0.0131], ASXL1 [OR 0.6 (0.36-0.95), p=0.0451] mutations were more enriched in stAML (Figure A).We then correlated mutation data with clinical and immunological parameters that are routinely tested in AML. Mutations in NPM1, FLT3, PTPN11 and NRAS were associated with significantly higher white blood cell (WBC) counts, bone marrow blast and LDH, which is consistent with their hyperproliferative activity as class 1 genes. In contrast, pts with mutations in TP53, STAG2 and ASXL1 presented with significantly low bone marrow blast, circulating blast, and WBC. Mutations in BCOR and ASXL1 was associated with significantly low LDH. Interestingly, pts with IDH2 mutations presented with significantly higher platelet, which is consistent with anecdotal report (DiNardo et al. Am J Hematology). Not surprisingly, TP53 mutations were associated with complex cytogenetics, whereas SRSF2, NPM1, IDH2, FLT3, and CEBPA mutations were associated with good and intermediate cytogenetics by ELN classification (Figure B). Pts with NPM1, IDH2, and IDH1 mutations were associated with less HLA-DR and CD34 expression in blast by flow cytometry. This is consistent with the frequent presentation of these AML sub-types with cuplike nuclei (Rakheja et al. BJH).DNA sequencing of a large cohort also allowed us to detect mutations that have not been as commonly reported in AML. We detected hot-spot mutations in exon 2 of MYC and MYCN genes in 9 (2%) AML pts. Additionally, internal tandem duplication (ITD) in MYC was also detected in one patient. Immunohistochemical staining showed that MYC expression was significantly elevated in patients with MYC mutations than in patients without the mutations (median H score 22 vs. 15 in MYC mutated vs. normal karyotype control, p < 0.001, 22 vs. 13.5 in MYC mutated vs. trisomy 8 control). These data suggest that a subset of AML is driven by the strong MYC signaling, consistent with a prior study (Ohanian et al. Leuk Lymphoma).CONCLUSION: Heterogeneous clinical presentation of AML has significant association with genetic heterogeneity, which suggest that distinct genetic basis of leukemogenic process has strong role in defining clinical presentation of AML. These data also help stratifying the patients for the likely target of precision medicine. DisclosuresDiNardo:Medimmune: Honoraria; Celgene: Honoraria; Agios: Consultancy; Abbvie: Honoraria; Karyopharm: Honoraria; Bayer: Honoraria. Kadia:Celgene: Research Funding; Pfizer: Consultancy, Research Funding; BMS: Research Funding; Jazz: Consultancy, Research Funding; Abbvie: Consultancy; Abbvie: Consultancy; Amgen: Consultancy, Research Funding; BMS: Research Funding; Pfizer: Consultancy, Research Funding; Jazz: Consultancy, Research Funding; Takeda: Consultancy; Amgen: Consultancy, Research Funding; Takeda: Consultancy; Novartis: Consultancy; Celgene: Research Funding; Novartis: Consultancy. Cortes:novartis: Research Funding. Daver:Daiichi-Sankyo: Research Funding; Pfizer: Consultancy; Alexion: Consultancy; ARIAD: Research Funding; Karyopharm: Consultancy; ImmunoGen: Consultancy; Kiromic: Research Funding; Otsuka: Consultancy; Sunesis: Consultancy; Novartis: Research Funding; BMS: Research Funding; Incyte: Consultancy; Novartis: Consultancy; Sunesis: Research Funding; Karyopharm: Research Funding; Pfizer: Research Funding; Incyte: Research Funding. Pemmaraju:SagerStrong Foundation: Research Funding; celgene: Consultancy, Honoraria; cellectis: Research Funding; samus: Research Funding; daiichi sankyo: Research Funding; Affymetrix: Research Funding; stemline: Consultancy, Honoraria, Research Funding; plexxikon: Research Funding; novartis: Research Funding; abbvie: Research Funding.