Pediatric Acute Myeloid Leukemia Patients Research Articles

A hypoxia risk signature for the tumor immune microenvironment evaluation and prognosis prediction in acute myeloid leukemia

Acute myeloid leukemia (AML) is the most prevalent form of acute leukemia. Patients with AML often have poor clinical prognoses. Hypoxia can activate a series of immunosuppressive processes in tumors, resulting in diseases and poor clinical prognoses. However, how to evaluate the severity of hypoxia in tumor immune microenvironment remains unknown. In this study, we downloaded the profiles of RNA sequence and clinicopathological data of pediatric AML patients from Therapeutically Applicable Research to Generate Effective Treatments (TARGET) database, as well as those of AML patients from Gene Expression Omnibus (GEO). In order to explore the immune microenvironment in AML, we established a risk signature to predict clinical prognosis. Our data showed that patients with high hypoxia risk score had shorter overall survival, indicating that higher hypoxia risk scores was significantly linked to immunosuppressive microenvironment in AML. Further analysis showed that the hypoxia could be used to serve as an independent prognostic indicator for AML patients. Moreover, we found gene sets enriched in high-risk AML group participated in the carcinogenesis. In summary, the established hypoxia-related risk model could act as an independent predictor for the clinical prognosis of AML, and also reflect the response intensity of the immune microenvironment in AML.

Abstract 1: Novel pediatric AML patient risk stratification by inferred protein activity through integrative network analysis and machine learning

Abstract Despite recent advances, overall survival for children with acute myeloid leukemia (AML) remains relatively low at 70 to 80%. Further refinement of risk stratification is paramount to guiding optimal therapy selection and improving outcomes. In this study, we leverage machine learning and network-based systems biology approaches that infer tumor protein activity and define key disease drivers - master regulator (MR) proteins - in pediatric AML to further refine existing patient risk stratification and identify potential therapeutic targets. Using the ARACNe and VIPER algorithms1, we reverse-engineered an AML gene regulatory network and inferred protein activity from enrichment of the proteins' transcriptional targets in gene expression signatures from a novel 1080-patient cohort treated on COG AAML1031. We integrated ensemble clustering with feed-forward feature selection on VIPER-inferred protein activity to identify groups of patients with unique protein activity signatures and correlated them to survival by Cox proportional-hazards regression analysis. We identified features predictive of each group by Monte Carlo cross-validation and binary classification. We found that the activity of 2-4 MR proteins was sufficient to predict patient stratification for each group with areas under the receiver-operator characteristic curve above 0.9. We validated our findings in three external test datasets: St-Jude Children's Research Hospital (n=249), TARGET (n=682) and Erasmus MC Sophia's Children's Hospital (n=237). We identified 10 groups with significant survival differences (P < 0.0001, overall survival range 42-85%). This includes a new high-risk group that is not associated with any known cytogenomic features, but shows high activity of NOTCH signaling pathway regulators HES1 and HEY2. Other groups show aberrant activity for known transcription regulators; for example RUNX1, KDM5B and POU4F1. In conclusion, we propose novel patient risk stratification based on VIPER-inferred MR protein activity in pediatric AML. We define 10 groups with differential survival, including refinement of a classically-characterized intermediate-risk cytogenomically normal group as high-risk. We classify patients into therapeutically-relevant risk groups that can guide treatment decision-making. We define the MR proteins driving these groups to offer mechanistic insights into AML tumorigenesis as well as suggest potential molecular targets for therapies. Reference: 1. Califano A, Alvarez M, The recurrent architecture of tumour initiation, progression and drug sensitivity, Nat Rev Cancer (2017), 17(2): 116-130 Citation Format: Alexandre P. Alloy, Jovana Pavisic, Rhonda E. Ries, Tanja Gruber, C Michel Zwaan, Monique L. den Boer, James R. Downing, Adolfo Ferrando, Soheil Meshinchi, Andrea Califano. Novel pediatric AML patient risk stratification by inferred protein activity through integrative network analysis and machine learning [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1.

Cytogenetic risk groups for childhood acute myeloid leukemia based on survival analysis in a cancer referral hospital from Perú

ResumenIntroducción. La leucemia mieloide aguda es una neoplasia heterogénea caracterizada por la proliferación de células mieloides inmaduras. El análisis citogenético ha revelado la presencia de aberraciones cromosómicas de importancia en el pronóstico del paciente.Objetivo. Determinar los grupos de riesgo citogenético de pacientes pediátricos con leucemia mieloide aguda a partir de la supervivencia global.Materiales y métodos. Se hizo un estudio observacional de corte transversal. Se incluyeron los registros clínicos de los pacientes pediátricos con diagnóstico de leucemia mieloide aguda de novo admitidos en el Instituto Nacional de Enfermedades Neoplásicas entre el 2001 y el 2011 y sometidos a análisis citogenético de médula ósea. Los grupos de riesgo citogenético se establecieron según los criterios del Medical Research Council. Las curvas de supervivencia global se elaboraron con el método de Kaplan-Meier y se compararon mediante la prueba de Mantel-Cox y una regresión de Cox, utilizando el programa R, versión 3.3.2.Resultados. Se incluyeron 130 pacientes, 68 varones (52,3%) y 62 mujeres (47,7%), mayoritariamente del subtipo M2 (33%). La edad promedio fue de 7,7 (rango de 0 a 15 años). Se observaron aberraciones cromosómicas en el 60,8% y la más frecuente fue la traslocación t(8;21). Según el análisis de supervivencia global, se observaron dos grupos de riesgo citogenético: favorable y desfavorable.Conclusión. Se determinaron dos grupos de riesgo citogenético: alto (o desfavorable) y estándar (o favorable).

Mutation Profiling Reveals Novel Epigenetics and Splicing Mutations in Chemorefractory Pediatric AML

Despite substantial improvement of treatment outcome in pediatric acute myeloid leukemia (AML), chemotherapy resistance remains to be a major concern. While molecular basis behind therapeutic failure was intensely interrogated, genomic alteration of chemotherapy resistance remains poorly characterized in pediatric AML. Here, we performed whole exon sequencing (WES) to identify gene mutations associated with primary and secondary chemo-resistance in pediatric AML.Taking advantage of the available WES dataset from pediatric AML patients in two hospital centers, we compared mutation frequency of total 45 pediatric AML cases, including 24 cases of induction remission and 21 cases of chemo-resistance, to that of a cohort of 200 adult AML in TCGA dataset. We observed mutation frequency of signaling molecules were much higher in pediatric AML than in adult AML. The most frequent mutated genes are NRAS (27%, 12/45) and KIT (20%, 9/45) in pediatric AML, which has significant higher mutation frequency than in adult AML (NRAS, P=0.0002; KIT, P=0.0001). In consistent with previous findings, epigenetic regulators such as DNMT3A (0%, 0/45), TET2 (0%, 0/45), IDH1/2 (2%, 1/45) and ASXL1/2 (2%,1/45) are rarely mutated in pediatric AML patients, while they are commonly found mutated in adult AML. It's worth noting that we uncovered several novel epigenetic mutations which were not found in TCGA dataset, such as KDM5C, SIRT6, CHD4, SS18 and SETD8.We assessed whether any mutated genes are associated with primary chemo-resistance group as comparing to primary chemo-sensitive group in studies cohorts. The chemo-sensitive group contained 24 primary induction remission samples, and 3 primary chemo-sensitive diagnosis samples of secondary chemo-resistance patients. Besides high recurrence of FLT3-ITD, we observed none of signaling molecule mutations are significantly enriched in primary chemo-resistance group. It's been reported that KIT mutations are frequently mutated in pediatric AML, and associated with adverse outcome. In our studied cohort, KIT mutations are more frequently found in relapsed AML (57%, 4/7), than primary chemo-resistance (28%, 2/7). Thus, poor clinical outcome of KIT mutations is more likely attributed to disease recurrence but not primary chemo-resistance.Furthermore, epigenetic mutations were enriched at primary chemo-resistance group (7/18 vs 2/27, P=0.009), including histone modified enzyme KDM5C, SIRT6, SETD8, splisomal factors PRPF6, PRPF8 and other epigenetic regulators CHD4, and IDH2. We found high allele frequency of KDM5C point mutation (35%) and SIRT6 frameshift mutation (42%) in one chemorefractory t (8:21) translocated AML patient. In addition, we also identified splicing mutation PRPF6 and PRPF8 in chemorefractory AML patients. In light of recent discovery of epigenetic alteration mediating chemo-resistance in adult AML, we thought these epigenetic mutations found in our cohort may also play a critical role in chemo-refractory pediatric AML.The development of new treatment strategies for pediatric AML must be based on an improved understanding of the molecular mechanisms underlying chemotherapy resistance. Previously, Fiona et al, shown that SETBP1 and ASXL1 and RELN mutations were associated with primary chemo-resistance in adult and pediatric AML patients by utilizing targeted sequencing of 670 genes. We have, for the first time, performed whole exon sequencing into 21 pediatric chemo-resistance AML. We identified novel epigenetic and splicing alterations such as KDM5C, SIRT6, SS18, CHD4 and PRPF8, PRPF6, which may play critical role in leukemogenesis and chemo-resistance in pediatric AML patients. Taking together, our study shed light into genetic lesions and evolution patterns underlying chemotherapy resistance and failure in pediatric AML. Larger cohort of patients and functional studies are needed to further understand molecular spectrum underlying chemo-resistance in pediatric AML. DisclosuresNo relevant conflicts of interest to declare.

Global and Targeted Metabolic Profiling Identifies Metabolic Markers for FLT3-ITD in Pediatric AML Patients

Acute myeloid leukemia (AML) is a clinically challenging disease with high interpatient variability in response to chemotherapy. Despite continuing advances in treatment options, current 5-year survival rates for pediatric AML are suboptimal at ~60%. Variability in treatment response and survival outcomes are due in part to the heterogeneous nature of AML, with many genetic lesions and cytogenetic features contributing to disease progression. One of the most well known genetic lesions associated with AML involves Fms-Like Tyrosine Kinase-3 (FLT3), a receptor tyrosine kinase expressed in hematopoietic stem cells. Internal tandem duplication of the juxtamembrane domain coding sequence of FLT3 (FLT3/ITD) causes autonomous cellular proliferations leading to disease progression. Previous metabolomics studies have successfully identified significant metabolic alterations in hematological malignancies. However, no metabolomics studies on pediatric AML have been reported at this time. In this study, we propose to use global and targeted metabolomics to identify differential metabolite abundance associated with FLT3-ITD status in pediatric AML patients treated in the St Jude AML02 clinical trial.Serum metabolomics profiles were generated with samples obtained at diagnosis from patients treated in the St. Jude AML02 study. Patients were assigned to FLT3 Wild Type (n=59) and FLT3-ITD (n=13) groups. Global metabolomics profiling was performed on a Thermo Q-Exactive Orbitrap mass spectrometer with Dionex UHPLC and autosampler. Targeted metabolomics profiling was generated for a select group of organic acids and acylcarnitines. The organic acid panel included eight metabolites related to the tricarboxylic acid cycle and glycolysis. The acylcarnitine panel included 57 varieties of acylcarnitine. Metabolomics profiling was performed on an Agilent 6490 triple quadrupole with an Agilent 1290 HPLC. Absolute quantification was achieved through external comparison on standard curves generated for each targeted metabolite on the organic acid and acylcarnitine panels. Univariate and multivariate analyses were performed using MetaboAnalyst web based software.A total of 3218 features were detected in the global metabolome, with 137 known metabolites and 3081 unknown features. Only the known metabolites were used for association analysis. All subsequent data analyses were log transformed, auto-scaled, and normalized to the sum of metabolites for each sample. Statistical analysis on metabolomics data identified 17 known metabolites significantly associated with FLT3 status (p<0.05). Some of the top significant metabolites include nicotinamide, 5,6-dihydrouracil, L-proline, 4-imidazoleacetic acid, taurine, L-asparagine, L-kynurenine, L-serine, D-ribose, and L-histidine. Pathway enrichment analysis identified 22 metabolic pathways significantly impacted by difference in FLT3 status, which included nicotinate and nicotinamide metabolism, aminoacyl-tRNA biosynthesis, arginine and proline metabolism, and histidine metabolism.Organic acid targeted metabolomics generated the concentrations of eight metabolites in patient samples. Analysis identified four organic acids with significantly different abundance associated with FLT3 status (p<0.05), including succinate, pyruvate, lactate, and α-ketoglutarate. Pathway enrichment analysis identified 24 metabolic pathways significantly impacted by difference in FLT3 status (p<0.05), including pyruvate metabolism, TCA cycle, butanoate metabolism, and glycolysis/gluconeogenesis. Acylcarnitine targeted metabolomics generated the concentrations of 57 acylcarnitine metabolites. Analysis identified two acylcarnitines (octadecanoylcarnitine and hexanoylcarnitine) significantly associated with difference in FLT3 status.Overall, this study identifies several metabolites and metabolic pathways significantly associated with the FLT3-ITD status in pediatric AML patients. These results help expand on previously conducted pilot studies and further clarify the metabolic differences associated with the FLT3-ITD form of AML. Ideally, continued metabolic profiling of additional AML subtypes can reveal pathways and networks that can be used to improve the efficiency of AML diagnosis and risk evaluation. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Genome-Wide Association Study (GWAS) Identifies SNPs in GPR56 and IGF1R Predictive of Leukemic Cell In Vitro cytarabine Sensitivity in Pediatric AML

Introduction: Acute myeloid leukemia (AML) is the second most common pediatric leukemia with the worst prognosis of all major cancers in pediatrics. Resistance is the most common reason for treatment failure and adverse side effects also contribute to morbidity and mortality. Since cytarabine is the mainstay of AML treatment, the wide interpatient variability in response to cytarabine complicates chemotherapy further. This necessitates the identification of additional genetic variants important in the susceptibility to cytarabine cytotoxicity in order to allow for personalized treatment that maximizes efficacy and minimizes toxicity. Previous studies have utilized genome-wide gene expression signatures to identify markers associated with cytarabine response. Few studies have also utilized lymphoblastoid cell lines from normal individuals available through HapMap project to perform genome-wide association analysis (GWAS) to identify genetic-variants and gene-expression associated with in vitro cellular cytotoxicity to cytarabine. However, to best of our knowledge no GWAS study has so far been performed on in vitro cytarabine chemosensitivity of leukemic cells obtained from pediatric AML patients.Hypothesis: To identify SNPs predictive of cytarabine in vitro chemosensitivityof leukemic cells obtained from pediatric AML patients (St Jude AML02 clinical trial) at diagnosis.Methods: Leukemic cells from diagnosis obtained from N=65 patients were treated with varying concentrations of cytarabine and IC50 values were obtained.Samples were divided into resistant (n= 15; IC50 > 5ng/μl) and sensitive groups (n=50; IC50 < 5ng/μl), Figure 1A. Genome-wide SNPdata were generated using Illumina Omni 2.5M and Exome Beadchip SNP array at University of Miami, Hussman Institute for Human Genomics. We performed the standard GWAS QC procedure in order to remove SNPs with call rate < 95%, monomorphic SNPs, SNPs with MAF < 5% and samples with call rate < 95%. Following QC, GWAS was performed on the remaining high quality SNPs (n=1,317,146) to identify if any of these are significantly associated with resistant or sensitive groups. Second, genes to which SNPs passing the suggestive threshold (p < 1x10-4) map and genes within +/-500kb of these SNPs were tested for differential expression between sensitive and resistant groups using T-test. Third, SNPs mapping to genes or SNPs within 500kb from genes that were differentially expressed (nominal p<0.05) were then tested for their eQTL effect using T-test or ANOVA.Results: Although none of the SNPs reached genome-wide significance (p<5.0 × 10-8), 112 SNPs located within or near 68 unique genes reached the suggestive level (p<1x10-4,Figure 1B). Among the top candidate SNPs were GPR56 rs1376041 G>A and rs75400242 G>ASNP found within 500kb of IGF1R gene. For both these SNPs the minor allele A was significantly associated with cytarabine resistance (p=1.42x10-5 and p=2.19x10-5, respectively). Furthermore, the minor allele of rs1376041 and rs75400242 was associated with a significantly higher level of GPR56 and IGF1R gene expression level (p=0.019 and p=0.044). Results for GPR56 for association of rs1376041 SNP with cytarabine resistance and GPR56 gene expression are shown in Figure 2A and 2C. Consistent with above observation for both GPR56 (Figure 2B) and IGF1R higher leukemic cell gene expression was associated with resistant group as compared to sensitive cases (p=0.0313 and p=0.0270).Conclusions: These results suggest that these SNPs might contribute to cytarabine in vitro sensitivity through regulation of gene expression. This is biologically relevant since it has been documented that GPR56 expression is associated with cytarabine resistance and poor outcomes in AML. In fact, recent data suggests high expression of GPR56 to contribute to AML development identified it as a potential target for antibody targeted therapy. For IGF1R, consistent with our results it has been associated with promoting growth of AML cells and cytarabine resistance as well. Future functional pharmacogenomic studies will help in better understanding the mechanisms behind the SNP associations reported here.Acknowledgments: We are thankful for funding from NIH R01CA139246 and Dr. McDonough for helpful discussions regarding GWAS analysis. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

The Unfolded Protein Response As a Predictor of Chemosensitivity in Pediatric Acute Myeloid Leukemia

Background: Survival in newly diagnosed pediatric patients with acute myeloid leukemia (AML) is approximately 65% with a much lower survival in patients with relapsed disease. Leukemia cells are exposed to proteotoxic stress due to their rapid, inefficient metabolism and accumulation of reactive oxygen species. This proteotoxic stress results in the production of misfolded proteins, which leads to activation of cell stress pathways, such as the unfolded protein response (UPR). We hypothesized that UPR induction correlates with a better response to chemotherapy.Methods: Peripheral blood samples from pediatric patients with AML were collected prior to initiation of induction chemotherapy, then at ~6 hours (h) and 24 h post initiation of chemotherapy. Peripheral blood mononuclear cells were frozen as pellets after depletion of non-tumor cells using magnetic bead selection (Miltenyi Biotech). The expression of UPR proteins was determined by chemiluminescence using an automated capillary electrophoresis system, ProteinSimple WesTM. The expressions of 5 UPR proteins were examined including Grp78, IRE1, total and phosphorylated (active) eIF2-α, and ATF6-α. Protein expression was normalized to actin expression and fold differences were compared between samples and within each sample at different time points.Results: Within each group of UPR proteins, there was a wide variability in the level of baseline expression, with 6 of 38 samples expressing >4 fold increase in Grp78, compared to the median level in healthy pediatric and adult controls (Figure 1). Basal IRE1 levels were also significantly higher in pediatric controls vs. adult controls. A trend towards lower overall survival (OS) and event-free survival (EFS) was noted in patients who had a low baseline Grp78 and a low baseline IRE1 (low baseline IRE1 was defined as <2 x below the median level of healthy pediatric controls). Post chemotherapy, UPR induction (defined as >2 x above baseline level) was seen in a subset of patients, most notably for ATF6α. However, post chemotherapy UPR induction was not associated with differences in either event-free survival or overall survival.Conclusion: UPR has a variable expression at baseline in pediatric AML patients, with a trend for lower overall survival in patients with a low basal Grp78 and IRE1 expression, suggesting chemoresistance in this subset of patients. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

A Dynamic Epigenetic Landscape in Genetic and Immunophenotypic Stable Pediatric AML

The majority of pediatric acute myeloid leukemia (AML) patients achieve complete remission. However, after this initial response to therapy, relapse occur in around forty per cent. The failure of therapy has been thought to be due to drug resistance resulting either from accumulating mutations, or already a characteristic of leukemic stem cells present at diagnosis. These mechanisms are shown to be true in part of the patients, however in quite a number of these relapses other mechanisms must play a role. In adult leukemias high epigenetic heterogeneity is associated with worse clinical outcomes. Remarkably, the DNMT3, TET and IDH mutations often found in adults are almost non-existent in children, despite the fact the leukemias are phenotypically very alike. We tested the hypothesis that, apart from genetic heterogeneity, epigenetic heterogeneity contributes to therapy resistance and thereby initiation of relapse.Methods: We performed flow cytometric immunophenotypic analyses, whole exome sequencing and epigenome sequencing (by enhanced reduced representation bisulfite sequencing) on bone marrow (BM) samples taken at time of diagnoses, within two weeks after achievement of CR and at time of relapse of three selected patients (diagnosed between 2007-2014, treated according to the Dutch-Belgian DB-AML-01 protocol). Of each patient, apart from whole diagnosis sample and whole relapse sample (all with a minimum of 90% blast cells), the MRD sample was FACS sorted to select for blast cells.Results: Leukemia associated immunophenotypes phenotype found at time of diagnosis remained the same over time with no emerging new immunophenotypes coinciding with emergence of relapses. We further evaluated whether changes in variant allele frequency (VAF) of mutations between diagnosis and subsequent time points could be linked to the development of relapse. Generally, VAF numbers decreased or remained stable over time, with no new mutations arising that characterized relapse. None of the patients had a mutation in DNTM3A, TET or IDH genes, accounting for any epigenetic dysregulation.In contrast to the findings for immunophenotypes and genetics, extensive epigenetic changes were found at the different time points. In samples taken within two weeks after reaching complete remission, a median of 23,665 (range 19,303-28,808) differentially methylated regions (DMRs) was found compared with the diagnosis samples. When comparing the corresponding relapse sample with this CR/MRD sample, another set of DMRs was found: median 19,625 (range 14,680-30,410). In all three patients, the majority of exons, intergenic, intron and promoter regions are uniquely hyper or hypo methylated at time of relapse (median of 88%, range 81%-95%), with a small proportion already differentially methylated at time of remission (median 11%, range 4%-18%). We hypothesize that these changes in epigenetic status occurring after CR, possibly contributes to relapse. Pathway analysis will probably give clues for understanding the pathophysiology and hopefully new targets for treatment.Conclusion: Our preliminary data suggest that epigenetic regulation may play a vital role in development of relapse. Ultimately, a better understanding of the role of epigenetics could open new therapeutic approaches for pediatric AML. DisclosuresOssenkoppele:Roche: Honoraria; Novartis: Research Funding; Karyopharm: Consultancy, Research Funding; J&J: Consultancy, Honoraria; Celgene: Honoraria, Research Funding. Kaspers:Janssen-Cilag: Research Funding. Cloos:Takeda: Honoraria.

Mesothelin Is a Novel Disease Marker and Potential Therapeutic Target in Pediatric Acute Myeloid Leukemia

Acute myeloid leukemia (AML) remains a significant therapeutic challenge in children with only 50% cure rate with highly intensive cytotoxic chemotherapy. Further intensification is limited by acute and late effects, including severe infectious and cardiac toxicities; therefore, alternative therapeutic strategies are needed. As part of the NCI/COG TARGET AML Initiative, using transcriptome sequencing (RNA Seq) in 635 children and 178 adult AML, we demonstrated that mesothelin (MSLN) was highly-overexpressed in over one-third of pediatric AML. This transcript is not expressed in normal hematopoiesis, making it an ideal target for therapeutic intervention. MSLN is a cell-surface protein normally found in pleura and peritoneum and overexpressed in a variety of solid tumors, such as mesothelioma. The protein product of the MSLN gene exists in 4 forms in the body, including cell surface MSLN, soluble MSLN, and soluble megakaryocyte potentiating factor (MPF), which is the N-terminus of the MSLN pro-protein. Elevated serum soluble MSLN is an established diagnostic test for mesothelioma, with an FDA-approved ELISA-based assay (Mesomark™).Three different AML RNA Seq data sets totaling 813 cases, including TARGET validation cohort (n= 475), TARGET discovery phase (n= 160), and TCGA AML (n= 178), were interrogated for expression of MSLN, with normal marrow cases as a control. MSLN transcript expression at >5 transcripts per million (TPM) was observed in 268 patients (range 0-1225 TPM) with observed prevalence of 48% in TARGET discovery set, 33% in TARGET validation and 15% in adult TCGA set with a collective prevalence of 33% in the entire cohort. MSLN expression was enriched in patients with t(8;21), inv(16), and KMT2A fusions and rare in those with common AML-associated mutations (FLT3- ITD, NPM1, CEBPa mutations). MSLN was not expressed in 20 normal marrow samples (<1 TPM in all cases).We conducted orthogonal validation of the transcriptome data by evaluating diagnostic specimens from an independent cohort of 323 newly diagnosed pediatric AML patients for the presence of MSLN using quantitative RT-PCR and ELISA. RNA extracted from leukemic blasts or plasma obtained from diagnostic specimens were tested for MSLN expression by quantitative RT-PCR or sandwich ELISA, respectively. Optimized q-RT-PCR identified 36% of patients with MSLN expression. Similarly, sandwich ELISA (similar to Mesomark™) confirmed expression of soluble MSLN in 34% of patients tested, with strong concordance between the two tests (p <0.0001). We have also optimized diagnostic multi-dimensional flow cytometry (MDF) and MPF ELISA for evaluation of MSLN.Mesothelin is a cell surface protein highly expressed in AML blasts but absent in other hematopoietic cells, making it an attractive target for immunotherapy. We sought to test the principle of MSLN targeting in leukemia with antibody-drug conjugate (ADC) anetumab ravtansine (AR; BAY94-9343), comprised of an anti-MSLN antibody conjugated to tubulin inhibitor DM4. MSLN -overexpressing cells were created by lentivirally-transducing MSLN into K562 cells (K562+ MSLN). K562+ MSLN exhibited exquisite sensitivity to AR in vitro with an IC50 of 1.4 nM, compared to 360 nM in the non-transduced K562 and 250-260 nM in isotype control conditions (Fig. 1). In vitro cytotoxicity of AR in two primary pediatric MSLN + AML samples also demonstrated MSLN -dependent killing. Additionally, mice bearing K562+ MSLN xenografts were treated in vivo with AR, which demonstrated significant MSLN -dependent efficacy, resulting in median survival of 87 days compared to 32-41 days in control conditions, including chemotherapy with daunorubicin and cytarabine (Fig. 2).MSLN is a cell-surface protein uniquely overexpressed in one-third of pediatric AML, as well as a subset of adult AML. We have shown that MSLN can be detected through several sensitive diagnostic mechanisms including MDF, q-RT-PCR and serum soluble MSLN, making it an optimal marker for disease monitoring. The leukemic expression of MSLN and absence from normal hematopoietic tissue make it an ideal target for immunotherapeutic targeting, and our findings demonstrate that MSLN can be exploited for therapeutic benefit using ADCs. Further studies in MSLN + AML on optimal detection and immunotherapeutic strategies are needed, as MSLN shows great promise for enhanced disease monitoring and leukemia-specific therapeutic targeting. [Display omitted] DisclosuresLoken:Hematologics Inc: Employment, Equity Ownership. Pardo:Hematologics Inc: Employment.

CSF3R Mutations Synergize with CEBPA and SETBP1 Mutations in Acute Myeloid Leukemia and Chronic Neutrophilic Leukemia

CSF3R is the receptor for granulocyte colony stimulating factor and plays a critical role in the proliferation and maturation of granulocytes. Although rare in most hematologic malignancies, we have discovered CSF3R mutations in ~80% of patients with Chronic Neutrophilic Leukemia (CNL) (Maxson et al, NEJM 2013). These mutations render the receptor constitutively active, leading to unrestrained neutrophil production. Although initially felt to be specific to CNL, through a collaboration with the National Cancer Institute/Children's Oncology Group TARGET Initiative, we identified a small fraction of patients with pediatric Acute Myeloid Leukemia (AML) that also have CSF3R mutations (Maxson et al, Blood, 2016). To interrogate this observation, we looked at co-occurring genetic lesions for the two malignancies. Interestingly, we found that the cooperating lesions are very different between the CNL and pediatric AML patient cohorts. For instance, ~25% of patients with CNL also have a mutation in SETBP1, and nearly half of patients with CSF3R-mutated pediatric AML have mutations in CEBPA. Furthermore, Lavallee et al identified CSF3R mutations as being specifically enriched within the cohort of AML patients with bi-allelic CEBPA mutations (Blood, 2016).We therefore explored whether these co-occurring mutations might synergize with CSF3R mutations in primary mouse bone marrow cells. We found that when WT CEBPA is co-expressed with the CSF3R T618I mutation, it strongly suppressed mouse bone marrow colony formation, while a CEBPA mutant enhanced colony formation. In addition, mutated SETBP1, which leads to protein overexpression, was co-expressed with CSF3R T618I. Interesting, WT SETBP1 enhanced colony formation, with a further increase in colony number with the SETBP1 G870S mutant. These studies suggest that there is a synergistic relationship between the CSF3R T618I mutation and two commonly co-mutated genes in CNL and pediatric AML. Since the CSF3R T618I mutation is very strongly activating, we wanted to use a model in which combinations of mutations could be expressed at endogenous levels to further study their interactions at the transcriptional level. To this end we are using CRISPR-Cas9 to generate myeloid cellular models. Two CRISPR-Cas9 gene-editing techniques are being employed to generate our mutations of interest following a Cas9-generated double-stranded break: (1) homology directed repair to insert point mutations into the CSF3R and SETBP1 loci, and (2) non-homologous end joining which results in frameshift mutations in CEBPA. We have completed validation of guide RNAs and HDR templates in 293T17 cells, with successful generation of both frameshift and specific point mutations. We have applied these tools to HL-60 cells, a human myeloid leukemia cell line, with successful isolation of clones with frameshift mutations. Investigation of these combinations of mutations in cell lines will allow us to determine how they cooperate to regulate transcriptional programs and produce distinct cellular identities. Our hope is that the principles learned from these well-defined diseases will provide a conceptual framework that will help us understand other myeloid malignancies with more complex clonal architecture and mutation combinations. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

NUP98-KDM5A Fusion Gene Is Not Exclusive to Acute Megakaryoblastic Leukemia and Is an Independent Predictor of Poor Prognosis

NUP98 rearrangements are recurrent genetic events in pediatric acute myeloid leukemia (AML). We identified NPU98-NSD1 as a recurrent poor risk abnormality (Hollink et al, Blood 2011). Another recurrent fusion is NUP98-KDM5A, which was found in pediatric acute megakaryocytic leukemia (AMKL) and was associated with poor clinical outcome (de Rooij et al, Nature Genetics 2017). To date, it is unclear whether NUP98-KDM5A exclusively occurs in AMKL or also in other FAB-subtypes of childhood AML. The aim of this study was to determine the incidence of NUP98-KDM5A in large pediatric AML cohorts, and to define the biological and clinical characteristics of this subgroup.The study comprised a COG and a European cohort of patients, and included 997 patients from COG trials 03p1 and 0531 (Gamis et al, Blood 2013) and 354 European pediatric AML patients, previously described by Balgobind et al (Haematologica 2011), NUP98-KDM5A rearrangements were detected either by paired-end RNA sequencing or RT-PCR. Patients with FAB M3 AML were excluded from this study. For comparison of clinical characteristics between NUP98-KDM5A positive and negative patients, the χ-squared and the Fisher's exact test were used. Event-free survival (EFS) and overall survival (OS) were estimated using the Kaplan-Meier method; the Cox proportional-hazard model was used for multivariate analysis. For this analysis, age, white blood cell count (WBC), cytogenetic risk group, allogeneic hematopoietic stem cell transplant (HSCT) as time dependent variable and study group (COG vs European cohort) were used as co-variables.NUP98-KDM5A rearrangements were detected in 31 out of 1351 patients (2.3%). There was no significant difference in prevalence between the COG and European cohorts. 41% of the NUP98/KDM5A+ patients were male, compared to 53% of the NUP98-KDM5A- patients (p=0.24). The median age of NUP98-KDM5A patients was significantly lower as compared to patients without this fusion gene (5.8 vs 9.4 yrs; p=0.03). There was no difference in WBC between NUP98-KDM5A positive and negative patients (13.1*109/l vs.25.2*109/l, respectively, p=0.13). NUP98-KDM5A occurred in all FAB types, with FAB M7 being most frequent (33% vs 8% in NUP98-KDM5A - patients; p<0.001).The 5-year EFS of NUP98-KDM5A + patients was 36.8 ± 18.8%, compared to 48.1 ± 2.8% in the NUP98-KDM5A - patients (p=0.12). The 5-year OS of NUP98-KDM5A + patients was 41.1 ± 19.7%, a value significantly lower compared to the NUP98-KDM5A - patients (63.9 ± 2.8%; p=0.01), this finding suggesting that NUP98-KDM5A+ patients are difficult to rescue when they relapse. In multivariate analysis, NUP98-KDM5A was an independent risk factor for OS, but not for EFS, with hazard ratios of 1.76 (p=0.045) and 1.58 (p=0.083), respectively. Non-M7 (n=19) patients had lower survival rates with EFS of 26.4 ± 21.7% and OS of 30.0 ± 22.3% compared to M7 (n=9) patients (EFS 55.6 ± 33.1% and OS 64.8 ± 33.1%). These differences were not statistically significant.This study shows that NUP98-KDM5A is an aberration that occurs in all FAB types and not just in AMKL. NUP98-KDM5A is an independent risk factor for OS but not for EFS and identifies a poor prognostic subgroup. Although there was a big difference in survival between M7 and non-M7 NUP98-KDM5A rearranged patients, this was not statistically significant, probably because of the low number of patients in the two cohorts. Overall, these data suggest to screen for the presence of NUP98-KDM5A rearrangements all children with non-APL AML irrespectively of their AML FAB subtype, and to stratify these patients in the high risk group. DisclosuresNo relevant conflicts of interest to declare.

Distinct Outcome Based on Fusion Transcript in t(16;21) Rearranged Pediatric AML, a Collaborative Retrospective Analysis By the I-BFM SG

Despite the use of of intensive chemotherapy, and allogeneic stem cell transplant (HSCT) in selected cases, children with acute myeloid leukemia (AML) have 5-year overall survival (OS) rates of ~65%. Apart from early clinical response, cytogenetic and molecular aberrations are the most reliable prognostic factors for survival. Identifying prognostic subgroups to stratify patients for treatment in future trials is clinically important. The study of rare AML subtypes can be facilitated by international collaborations. AML with t(16;21) rearrangements includes 2 different translocations: t(16;21)(p11;q22) resulting in the FUS-ERG fusion, and t(16;21)(q24;q22) resulting in the RUNX1-CBFA2T3 fusion.To study outcomes and associated prognostic factors, clinical data from children with t(16;21) AML were collected from 14 study groups participating in the international Berlin-Frankfurt-Munster (I-BFM) network. Patients (age 0-18 years) diagnosed from January 1, 1995, to January 1, 2016, were included. The AML-BFM study group provided data from 1326 unselected patients diagnosed between 1997 and 2013 as a reference cohort. All karyotypes were centrally reviewed by 2 independent cytogeneticists according to the International System for Human Cytogenetic Nomenclature (ISCN 2005). Patients with inconclusive karyotypes were screened by reverse-transcription PCR. Chi-squared and Fisher exact tests were used to compare clinical characteristics. OS and event free survival (EFS) analysis was performed by the Kaplan-Meier method. Cumulative incidence of relapse (CIR) was calculated according to Kalbfleisch and Prentice. The Cox proportional hazards model was used for multivariate analysis, considering as covariables age, white blood cell (WBC) count at diagnosis, cytogenetic risk group and allogeneic hematopoietic stem cell transplantation (HSCT) as time-dependent variable.There were 54 patients with t(16;21): 32 with FUS-ERG and 22 with RUNX1-CBFA2T3 . There were no significant differences in sex and age between patients in the FUS-ERG- or RUNX1-CBFA2T3- groups and the reference cohort. Median WBC at diagnosis was significantly lower for patients with RUNX1-CBFA2T3 (12.5×109/L, range 0.01-185×109/L) than for the reference cohort (19.4×109/L, range 0.01-1900×109/L, P=0.030). Patients with FUS-ERG had no predominant French-American-British (FAB) type, whereas 76.4% of those with RUNX1-CBFA2T3 had a granulocytic FAB type (M1, M2) compared with 42.1% in the reference cohort (P=0.004).The 4-year EFS, OS, and CIR for the reference cohort were 51% [standard error (SE)=1%], 68% (SE=1%), and 32% (SE=1%), respectively. Median follow up for survivors in the t(16;21) cohort was 1.6 years for those with FUS-ERG and 5.0 years for RUNX1-CBFA2T3 . Patients with FUS-ERG had a 4-year EFS of 7% (SE=5%, P <0.0001), 4-year OS of 21% (SE=8%, P <0.0001), and CIR of 74% (SE=8%, P <0.0001). Patients with RUNX1-CBFA2T3 had a 4-year EFS of 77% (SE=9%, P=0.06), OS of 81% (SE=8%, P=0.34) and a CIR of 0%. When patients in the reference cohort were classified in SR patients comprising (inv(16)/t(8;21), and HR patients comprising other cytogenetic subtypes excluding t(16;21), EFS rates were 74% (SE=3%) and 45% (SE=2%), respectively (Fig. 1). Hence, patients with RUNX1-CBFA2T3 had a similar outcomes as SR patients (Figure 1). In multivariate analysis, presence of FUS-ERG and RUNX1-CBFA2T3 were independent risk factors for EFS, with hazard ratios of 1.93 (P <0.0001) and 0.325 (P=0.025), respectively. HSCT was performed in 14 (45.2%) patients with FUS-ERG in first complete remission (CR1). The 4-year EFS for transplanted patients was 15% (SE=15%) compared with 0% (SE=13%) for patients receiving chemotherapy only (p=NS).We identified 2 clinically relevant, distinct subtypes of pediatric AML patients with different t(16;21) rearrangements: FUS-ERG had poor outcomes, whereas RUNX1-CBFA2T3 had favorable outcomes. Patients with RUNX1-CBFA2T3 -rearranged AML might benefit from inclusion in the SR group and receiving less intensive treatment, whereas those with FUS-ERG rearranged AML should be stratified in the HR group. In this study, only patients with FUS-ERG receiving HSCT in CR1 did not relapse. Most current pediatric AML trials consider children with a 5-year OS below 25%-30% for HSCT in CR1. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Risk Score Generated from Ara-C Metabolic Pathway Predicts Leukemic Cell Intracellular Ara-CTP Levels in AML-97 Pediatric Patients

Background: Acute myeloid leukemia (AML) is the cancer of the myeloid precursor stem cells. AML was historically treated with combinations of cytarabine (ara-C) and anthracyclines. Previous studies showed that polymorphism in genes known to be involved in cytarabine pathway can affect patient's drug response and may explains part of the inter-patient variability in treatment response.Patients and methods: AML patients (n= 89) treated under St. Jude AML97 clinical trial were included in the study. Intracellular levels of ara-CTP the active metabolite of ara-C was quantitated in leukemic cells obtained 24hrs after initiation of cytarabine infusion. Morphological response to induction 1 therapy was assessed in bone marrow aspirate obtained on day 15. Genomic DNA was tested for SNPs in DCK and CMPK1 required for the drug activation, CDA, DCTD, NT5C2, NT5C3A, RRM1 and RRM2 involved in Cytarabine inactivation and the drug transporters SLC29A1, SLC28A1 and SLC28A3. Additionally, CTPS1, NME1-NME2 and NME3 that are important for the synthesis of nucleoside triphosphates other than ATP were also included. Overall 104 SNPs in genes indicated above were genotyped using Sequenom (CA, USA) based genotyping that uses MALDI-TOF-based chemistry at University of Minnesota, Biomedical Genomics Center.Results: Tweleve SNPs demonstrated a statistically significant association with ara-CTP level at day 1 post treatment (P<0.05). We found SNP rs4643786 in DCK gene significantly associated with lower ara-CTP levels and poor outcomes in white patients. CDA SNP rs12404655, was associated with high ara-CTP level at day 1 and better treatment outcomes (Fig 1B). CMPK1 SNP rs7543016, a missense cleavage site variant was associated with lower ara-CTP levels. In addition, SNP rs9916651 in NME1-NME2 was predictive of higher ara-CTP levels (Fig 1A).We observed 7 SNPs to be significantly associated with the treatment response after induction 1 this included a previously reported NT5C2 SNP rs11598702 and 4 SNPs in DCTD that were associated with poor response and occured in linkage disequilibrium with each other's (r2=1: rs3190314, rs1130902, rs7277 and rs851).Using the genotypes of SNPs associated with intracellular ara-CTP levels we created a risk score by combining the effect of SNPs reported in each patient. For this we took into consideration the direction of genotype association for each SNP. For example, if the SNP is associated with higher ara-CTP level, it takes a score ranging between 0 and +2 (0 if the patient has normal alleles, 1 if he/she is heterozygous and 2 if homozygous for the SNP). If the SNP is associated with lower ara-CTP level, it takes a score ranging between 0 and -2. Risk scores ranging between -7 and 5 were obtained after this analysis. We further compressed the risk groups into three categories (risk score ≥1, risk score 0, -1 and risk score ≤ -2 (Figure 2). Increased value of the risk score was associated with increased intracellular ara-CTP level suggesting a cumulative or synergistic effect of the reported SNPs.Conclusion: Our results suggest that SNPs in genes related to cytarabine pathway can significantly affect level of ara-CTP inside the leukemic cells. To comprehensively include the potentially significant SNPs we developed a risk score for each patient by taking into account the directional genotype for each SNP in each patient. We hope that this score although require validation in other cohorts will help in identification of patients with potentially lower levels of active form of ara-CTP in the leukemic cells. We would also like to point out that it is very challenging to perform this study as it requires obtaining bone marrow samples for determining intracellular levels of ara-CTP after day 1 of initiation of chemotherapy, which is not a standard procedure. To the best of our knowledge this is the only clinical cohort of pediatric AML patients where intracellular ara-CTP levels post ara-C treatment are available and we leveraged this to evaluate genetic variants. Our result would provide significant information in further understanding the pharmacogenomics of cytarabine in AML. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

A Multicenter Comparison of Rates of Central Venous Catheter-Associated Bloodstream Infections By Central Venous Catheter Type Among Pediatric Patients Receiving Treatment for Newly Diagnosed Acute Myeloid Leukemia

▪Therapy for pediatric acute myeloid leukemia (AML) requires long term central vascular access, which puts patients at risk for central line-associated bloodstream infections and their associated morbidity and mortality. Options for central venous catheter (CVC) access include tunneled external catheters, implanted ports, or peripherally inserted central catheters (PICC). Multiple studies provide conflicting reports on which CVC type is least likely to be associated with bloodstream infections in pediatric hematology-oncology patients. This multicenter cohort of pediatric AML patients is the first to investigate the course-specific incidence of positive CVC blood cultures by line type.The study included patients receiving frontline chemotherapy for AML between 1/1/2012 and 6/30/2017 at 12 US institutions. Trained personnel completed manual chart abstraction from 12/1/2015 to 07/15/2017 to obtain information on demographics, diagnosis, treatment, antibacterial prophylaxis, CVC type at the start of each course, neutropenia duration, and blood culture results obtained from CVC. Abstractors followed a standardized protocol and were required to achieve >95% concordance with trainer before starting formal chart review.Course-specific follow-up began the first day absolute neutrophil count (ANC) fell below 200 cells/uL and continued to the earliest of first positive CVC culture (overall and separately for gram-positive and gram-negative species), ANC recovery, start of conditioning for stem cell transplant, relapse, death or start of next course. Per National Healthcare Safety Network guidance, common commensals were not considered true events with the exception of viridans group streptococci (VGS). Nonbacterial pathogens were also excluded. Covariate distributions were compared by CVC type using chi-square tests. Variability in CVC type by site was examined graphically. Incidence rates for CVC positive blood cultures were reported by line type as per 1000 neutropenic days. Multivariate Poisson regression with log of neutropenia duration as offset was used to compute adjusted rate ratios (aIRR) comparing rates of positive CVC blood cultures by line type, controlling for covariates itemized in Table 1. General estimating equation methods were used to account for non-independence of observations from patients from the same institution.The study population included 357 patients who received 1,164 chemotherapy courses. Median course duration was 38 days (IQR: 34-44) with a median neutropenia duration of 19 days (IQR: 14-25). CVC were present in all patients at the start of each course, with the exception of 11 (3.3%) patients lacking a CVC at the start of Induction I. While tunneled CVC were most prevalent representing 59.1% of courses, PICC and implanted ports were utilized in 28.9% and 11.6% of courses, respectively. There was considerable variation in the distribution of CVC type by treating institution (Figure 1). Age, race, and ethnicity distributions varied by line type (Table 1). Antibacterial prophylaxis was utilized in 47% of courses.In total, there were 300 (25.8%) courses in 199 (55.7%) patients complicated by a positive CVC blood culture during the neutropenic period. The overall incidence rate of positive CVC blood cultures was 6.5 per 1000 neutropenic days. Rates increased over the course of frontline treatment from 3.1 per 1000 neutropenic days in Induction I to 9.6 per 1000 neutropenic days in Intensification II. The rate of gram-positive events (4.2 per 1000 neutropenic days) was higher than gram-negative events (2.0 per 1000 neutropenic days). The most common isolated organisms were VGS at 57.3% and less than 1% of events were polymicrobial infections. Overall rates of positive CVC blood cultures were higher for PICC relative to tunneled CVC (aIRR=1.19, 95% CI: 1.02, 1.38), while rates for implanted ports and tunneled CVC were similar (aIRR=1.06, 95% CI: 0.75, 1.51).While bacteremia is an expected complication during post-chemotherapy neutropenia and not always due to the presence of a CVC, the risk for a positive CVC blood culture is significantly increased for children with PICC compared with those with tunneled CVC or implanted port. Also, the risk was higher after Induction I and increased with age. Given the notable practice variation in line choice, these data highlight an opportunity for standardization with a goal to reduce infectious complications. [Display omitted] DisclosuresFisher:Pfizer: Research Funding; Merck: Research Funding. Gramatges:Bristol Meyer Squibb: Research Funding.

Incidence and Predictors of Adverse Respiratory Events during Induction Therapy in Children with Acute Myeloid Leukemia

Background: Mortality in pediatric acute myeloid leukemia (AML) has plateaued between 30-40%, with 5-15% of these deaths occurring during initial induction. While pulmonary complications contribute to the early morbidity and mortality in pediatric AML, our knowledge of the incidence, risk factors, and patterns of respiratory adverse events (AEs) is lacking. Hypothesized risk factors for respiratory AEs during induction include hyperleukocytosis, fluid overload, coagulopathy, and infection. Our objective was to characterize the incidence, categories, and risk factors of respiratory AEs in de novo pediatric AML patients during induction chemotherapy.Methods: We conducted a single institution retrospective longitudinal study from presentation to Day 42 or initiation of the next chemotherapy cycle in de novo pediatric AML patients (age 0-21 years) presenting between January 1, 2001 and December 31, 2016. Patients were identified through the institutional cancer registry. Exclusion criteria included: relapsed AML, induction at another institution, and failure to treat with conventional protocols. Outcomes of interest included development of an NCI CTCAE (v4.0) grade 2-5 respiratory AE or death from another cause. Data was abstracted in regards to demographic and disease characteristics at presentation. Supportive care measures included fluid status, with fluid overload defined as a ≥ 10% increase from body weight at presentation or the use of ≥ 3 doses of furosemide during the follow up period. Respiratory AE was attributed to infection in cases with a positive bacterial or fungal blood culture, respiratory viral panel, endotracheal tube aspirate, or a chest imaging report explicitly suggesting a pulmonary infectious process. A single rater assigned the most specific, best fitting CTCAE category and grade for each incident event at its maximal severity. Time to event distribution curves were created and descriptive statistics using two-sample T-test, Chi square test for independence, or Fisher's exact test as appropriate were calculated (SAS v X, Cary, NC).Results: To date we have reviewed the induction course on 63 of 245 eligible patients with de novo AML. Median age at diagnosis was 8.0 years (range, 0-21 years), 58.7% were female (n = 37), and 42.9% non-Hispanic White (n = 28). During the follow up period, 3 patients died (4.8% induction death at days 5, 25, 73). The deaths at days 25 and 73 were due to pulmonary fungal infection and the death at day 5 from an intracranial bleed. During the first 42 days, 52.4% (n = 33) patients experienced 34 discrete respiratory AEs: CTCAE grade 2 in 2.9% (n = 1), grade 3 in 64.7% (n = 22), grade 4 in 26.5% (n = 9) and grade 5 in 5.9% (n = 2). Among the respiratory AEs, 44.1% required BiPAP/CPAP (n = 4) or mechanical ventilation (n = 11). As shown in Table 1, those with and without respiratory AEs differed significantly in regards to median age at diagnosis (p = .003), coagulopathy at diagnosis (PT (p = .019), D-dimer (p = .012)), and fluid overload status (p < .0001). Those with respiratory AEs did not differ by sex (p = .08), development of infection (p = .65), or WBC at diagnosis (p = .08). A bimodal distribution of time to event was (Figure 1) with 67.7% (n = 23) of AEs occurring between presentation and day 7 and classified as “early.” As seen in Table 2, there were statistically significant differences between the early and late AE groups for WBC at diagnosis (p = .007), PT at diagnosis (p = .012), and fluid overload (p = .026). While a larger proportion of the late AEs experienced infection (45.5% vs 30.4%), this did not significantly differ from the early AEs (p = .39).Conclusion: We note a high incidence of grade 2-5 respiratory AEs in pediatric patients during AML initial induction [52.4%]. Fluid overload appears to be a significant factor in early AEs. Documented infection occurred in a higher proportion of late AE cases, although this did not reach statistical significance. Future analyses will examine the role of cytogenetic factors and morphology. Pulmonology and oncology raters will independently characterize and validate the category of each respiratory AE (e.g. atelectasis, pulmonary edema, pulmonary leukostasis, pneumonia) through a standardized process for the full eligible cohort. Identifying modifiable risk factors will enable design of supportive care interventions to decrease the risk of respiratory AEs and their associated morbidities during induction. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Differences in Infection Prophylaxis between Pediatric AML Study Groups within the International BFM Study Group

BackgroundPediatric patients with acute myeloid leukemia (AML) are at high risk of severe infections. Prophylaxis of infections is of obvious relevance, but recommendations are often non-specific and supported by low-quality evidence. Despite appeals by experts and working groups in the field to harmonize guidelines, many divergent regimens among hospitals and study groups seem to exist. Whereas previous surveys on this matter mainly focused on hospital-based recommendations, the aim of the current study was to gain insight in all nationally available infection prophylaxis guidelines for pediatric AML patients within the international Berlin-Frankfurt-Munster (I-BFM) study group.MethodsUsing a web-based survey, we inquired the infection prophylaxis guidelines of 22 pediatric AML study groups within the I-BFM consortium. The survey included questions on the use of both pharmacological measures, i.e. antibiotics, antifungal medication, antiviral prophylaxis, granulocyte colony stimulating factor (G-CSF), and non-pharmacological measures, e.g. the use of isolation rooms, out-patient clinic visits and food or social restrictions. In order to check for discordant answers or inconsistencies, a copy of the formal study group guidelines (if available) was requested.In order to evaluate differences in daily practice among hospitals within one study group, hospital representatives (n =27) from the NOPHO-DBH AML Study Group participated in a slightly modified survey.Descriptive statistics are presented.ResultsAlmost all participating study groups prescribe Pneumocystis Jiroveci pneumonia prophylaxis with sulfonamides (n =21, 95%). About 32% of the study groups (n =7) advise gram-negative antibiotic prophylaxis, mainly with fluoroquinolones (n =6, 86%). Gram-positive prophylaxis is prescribed by 36% of the study groups (n =8), of whom 2 study groups only prescribe this in specific situations, for example after high-dose cytarabine. The majority of the study groups do not recommend the use of G-CSF (n =10, 46%) or recommend it only in specific cases of severe infections and/or prolonged neutropenia (n =7, 32%).Over 60% of the study group prescribe food and social restrictions, but the specific topics and strictness differed largely between all groups. Most frequently reported food restrictions included consumption of raw meat and raw seafood (n =16, 76%), unpasteurized milk products (n =14, 67%) and dishes with undercooked eggs (n =13, 62%). Frequently reported social restrictions included visiting indoor crowded places (n =13, 62%), using public transportation (n =12, 57%) and visiting daycare/kindergarten/pre-school (n =12, 57%). Study groups who recommend antibiotic prophylaxis were comparable stringent with regard to social or food restrictions and outpatient management.According to the hospital-based survey, sites roughly complied with the common study group guidelines. However, the use of any gram-negative antibiotic prophylaxis, the specific prophylactic antifungal agent and the strictness of the food and social restrictions differed substantially between the hospitals.ConclusionThis study gives an overview of the currently applied international recommendations on infection prophylaxis for pediatric AML patients. Despite a long history of close collaboration within the I-BFM consortium, many differences are still present between the affiliated pediatric AML study groups. Differences between study groups, as well as the discrepancies between hospital practices, are probably explained by generally formulated guidelines based on conflicting results of previous studies, and the lack of solid clinical evidence. Fortunately major efforts are now being undertaken to answer several of the remaining issues on infection prophylaxis. Furthermore, an update of the international guidelines on management of pediatric oncology patients with fever and neutropenia was recently published. The results of this survey provide an appropriate baseline measure to study the impact of these and emerging guidelines in the future. DisclosuresKaspers:Janssen-Cilag: Research Funding.

Global Proteomic Profiling of Pediatric AML: A Pilot Study.

Simple SummaryThe second most common childhood leukemia, acute myeloid leukemia (AML), is a heterogeneous disease with a poor prognosis. In order to improve outcomes, efforts to understand the genomic/transcriptomic landscape of AML have been performed. However, there is a significant gap in our understanding of leukemic cell proteomics. In comparison to the proteome, the transcriptome alone cannot adequately represent the biological functions within cells and it is often not a target for immediate drug development. In the current study, we investigated cytogenetic differences at the proteomic level and sought potential predictive biomarkers and druggable proteins.Acute Myeloid Leukemia (AML) is a heterogeneous disease with several recurrent cytogenetic abnormalities. Despite genomics and transcriptomics profiling efforts to understand AML’s heterogeneity, studies focused on the proteomic profiles associated with pediatric AML cytogenetic features remain limited. Furthermore, the majority of biological functions within cells are operated by proteins (i.e., enzymes) and most drugs target the proteome rather than the genome or transcriptome, thus, highlighting the significance of studying proteomics. Here, we present our results from a pilot study investigating global proteomic profiles of leukemic cells obtained at diagnosis from 16 pediatric AML patients using a robust TMT-LC/LC-MS/MS platform. The proteome profiles were compared among patients with or without core binding factor (CBF) translocation indicated by a t(8;21) or inv(16) cytogenetic abnormality, minimal residual disease status at the end of the first cycle of chemotherapy (MRD1), and in vitro chemosensitivity of leukemic cells to cytarabine (Ara-C LC50). Our results established proteomic differences between CBF and non-CBF AML subtypes, providing insights to AML subtypes physiology, and identified potential druggable proteome targets such as THY1 (CD90), NEBL, CTSF, COL2A1, CAT, MGLL (MAGL), MACROH2A2, CLIP2 (isoform 1 and 2), ANPEP (CD13), MMP14, and AK5.

DNA Methylation Signatures Predict Cytogenetic Subtype and Outcome in Pediatric Acute Myeloid Leukemia (AML).

Pediatric acute myeloid leukemia (AML) is a heterogeneous disease composed of clinically relevant subtypes defined by recurrent cytogenetic aberrations. The majority of the aberrations used in risk grouping for treatment decisions are extensively studied, but still a large proportion of pediatric AML patients remain cytogenetically undefined and would therefore benefit from additional molecular investigation. As aberrant epigenetic regulation has been widely observed during leukemogenesis, we hypothesized that DNA methylation signatures could be used to predict molecular subtypes and identify signatures with prognostic impact in AML. To study genome-wide DNA methylation, we analyzed 123 diagnostic and 19 relapse AML samples on Illumina 450k DNA methylation arrays. We designed and validated DNA methylation-based classifiers for AML cytogenetic subtype, resulting in an overall test accuracy of 91%. Furthermore, we identified methylation signatures associated with outcome in t(8;21)/RUNX1-RUNX1T1, normal karyotype, and MLL/KMT2A-rearranged subgroups (p < 0.01). Overall, these results further underscore the clinical value of DNA methylation analysis in AML.

Correlation of ex vivo drug sensitivity with clinical response in pediatric AML.

10032 Background: Pediatric acute myeloid leukemia (AML) is a rare disease with roughly 500 cases diagnosed in the United States each year and has had minimal improvement in clinical outcomes over recent decades. Novel treatment development to improve outcomes may be enhanced with an accompanying test for predicting treatment response. We previously demonstrated that an ex vivo drug sensitivity assay (DSA) can predict clinical response in myelodysplastic syndrome. Here we investigated the use of the DSA in pediatric AML patients, including a subset participating in a clinical trial of atovaquone. Atovaquone is an FDA-approved anti-parasitic drug that was associated with lower relapse rates in adult AML patients. Adding atovaquone and other standard of care combination treatments into the DSA, we investigated whether the assay, performed on pre-induction samples, correlated with measures of clinical response. Methods: We assayed pre-induction blood or bone marrow samples from 22 de novo pediatric AML patients diagnosed at Texas Children’s between 5/2015 and 10/2020 who consented to research (82% enrolled in clinical trial identifier NCT03568994). We subsetted this analysis to patients who received ADE (Cytarabine, Daunorubicin, Etoposide) (n = 20) induction, with the majority additionally receiving atovaquone (n = 16). For the DSA, samples were incubated with up to 25 compounds, including the treatment drug combinations and each of the compounds individually. After incubation, changes in tumor blast populations were assessed by flow cytometry. For each drug condition, drug sensitivity was calculated based on the number of blasts remaining after treatment. After quality control, downstream analyses were limited to 13 samples. Clinical response data, including minimal residual disease (MRD) percentage by flow cytometry and one year relapse-free survival, were correlated with the drug sensitivity results. Results: For the de novo subset analysis, we observed correlations between ex vivo drug sensitivity with both MRD percentage (r = 0.63) and one year relapse-free survival (RFS1, AUC = 0.92). The 3 patients with lowest ex vivo sensitivity had the highest MRD percentages (mean 21%). 2 of the 3 patients who did not achieve one year relapse-free survival had the lowest ex vivo sensitivity. Drug combination sensitivity correlated more with MRD and RFS1 than the single agents alone (single agent mean MRD r = 0.39). Conclusions: In our cohort, ex vivo DSA for ADE and atovaquone in pediatric AML cases correlated with both MRD and one year relapse-free survival. This suggests that clinical response in pediatric AML may be assessed prior to treatment using a DSA. This study also suggests that the DSA can be used to test drug combinations, and thus may be used for investigating novel treatment combinations. Further development of the DSA may benefit treatment decisions and prioritization of drug development.

Exploiting Clonal Evolution to Improve the Diagnosis and Treatment Efficacy Prediction in Pediatric AML.

Simple SummaryThe use of innovative technologies has revolutionized cancer research in recent years, and in the field of pediatric oncohematology, great results have been achieved. However, within this context, acute myeloid leukemia still represents a considerable challenge for clinicians, as frequent cases of relapse or refractory disease, especially in specific subgroups, remain present. With this review, the authors aim to recapitulate the main features of this extremely heterogeneous malignancy, by highlighting the concept of clonal evolution and how, thanks also to the impact of new high throughput techniques, this could be exploited to deepen the current knowledge on molecular mechanisms driving this disease. Overall, this study will seek to pave the way for making new tools available to further improve diagnosis and treatment protocols in pediatric patients.Despite improvements in therapeutic protocols and in risk stratification, acute myeloid leukemia (AML) remains the leading cause of childhood leukemic mortality. Indeed, the overall survival accounts for ~70% but still ~30% of pediatric patients experience relapse, with poor response to conventional chemotherapy. Thus, there is an urgent need to improve diagnosis and treatment efficacy prediction in the context of this disease. Nowadays, in the era of high throughput techniques, AML has emerged as an extremely heterogeneous disease from a genetic point of view. Different subclones characterized by specific molecular profiles display different degrees of susceptibility to conventional treatments. In this review, we describe in detail this genetic heterogeneity of pediatric AML and how it is linked to relapse in terms of clonal evolution. We highlight some innovative tools to characterize minor subclones that could help to enhance diagnosis and a preclinical model suitable for drugs screening. The final ambition of research is represented by targeted therapy, which could improve the prognosis of pediatric AML patients, as well as to limit the side toxicity of current treatments.