Infant ALL Research Articles

Enhanced Efficacy of the SYK Inhibitor Entospletinib and Chemotherapy in Infant Acute Lymphoblastic Leukemia

Background: Chemotherapy resistance and subsequent relapse remain a major cause of childhood cancer mortality, particularly for infants (<12 months of age) with KMT2A -rearranged (R) B-cell acute lymphoblastic leukemia (B-ALL), who have extremely poor event-free survival (EFS). Addition of kinase inhibitors to chemotherapy has markedly improved outcomes for patients with some genetic subsets of B-ALL, such as BCR-ABL1 -R (Ph+) ALL. Unfortunately, addition of the FMS-like tyrosine kinase 3 (FLT3) inhibitor lestaurtinib did not improve 3-year EFS for infants with KMT2A -R and FLT3-overexpressing B-ALL in the Children's Oncology Group trial AALL0631 (Brown et al, SIOP 2016), indicating continued need for alternative therapeutic strategies. We recently observed constitutive phosphorylation (p) of spleen tyrosine kinase (SYK) at tyrosine 323 in infant ALL specimens (Figure 1A) and hypothesized that activated signaling is targetable with the selective SYK inhibitor entospletinib (ENTO). We further hypothesized that combining ENTO with chemotherapy would have superior activity in comparison to single-agent treatments.Methods: To test these hypotheses, we established patient-derived xenograft (PDX) models of KMT2A -R (n=7) and non- KMT2A -R (n=4) infant ALL in NSG mice using diagnostic and relapse bone marrow specimens from patients treated on or as per the AALL0631 trial protocol. Total and phosphorylated (p) SYK and other phosphoprotein levels were assessed by Simple WesternTM analysis of splenic lysates from ALL-engrafted PDX mice to identify leukemias with constitutive SYK signaling activation. Infant ALL PDX models were then treated with control mouse chow or chow containing 0.07% ENTO ad libitum, 0.25 mg/kg vincristine (VCR) dosed intraperitoneally once weekly, or a combination of ENTO and VCR for up to 28 days to determine treatment efficacy. Cohorts of additional infant ALL PDX mice were treated for 72 hours to assess plasma levels and pharmacodynamic effects of ENTO administration.Results: Constitutive SYK signaling was observed in several models of de novo KMT2A -R and non- KMT2A -R B-ALL. ENTO treatment of KMT2A -R ALL PDX models ALL3103 with t(9;11) and ALL135 with t(11;19) inhibited leukemic cell burden in vivo versus control animals by flow cytometric and immunohistochemical (IHC) analyses of murine spleens (representative data in Figure 1B). Importantly, combined ENTO and VCR treatment resulted in a greater reduction in ALL cell number, which was confirmed by human CD19+ IHC analysis of murine spleens and bone marrow. ENTO concentrations were maintained throughout study duration with terminal free fraction plasma levels of 61 and 230 nM (ALL3103) and 89 and 214 nM (ALL135) for ENTO and ENTO/VCR treatment arms, respectively. Evaluation of SYK pathway modulation in vivo demonstrated inhibition of pSYKY323 in human ALL cells within murine spleens that correlated with the level of ENTO exposure of treated mice (ALL3103 PDX; Spearman correlation of -0.63, p<0.05). No alterations in total SYK protein were observed. RNA sequencing of ALL3103 PDX tissues showed that ENTO treatment was associated with downregulation of MYC and of cell cycle and DNA damage gene expression signatures.Conclusions: Constitutive activation of SYK signaling occurs in infant KMT2A -R B-ALL. Entospletinib treatment of infant B-ALL PDX models potently inhibited the SYK pathway and decreased leukemic cell burden in vivo . Superior anti-ALL activity was observed with combined ENTO and VCR compared to ENTO alone. Additional studies are ongoing to determine the activity of ENTO in non- KMT2A -R and other ALL PDX models. [Display omitted] DisclosuresYahiaoui:Gilead Sciences, Inc.: Employment, Equity Ownership. Lopez:Gilead Sciences, Inc.: Employment, Equity Ownership. Mikaelian:Gilead Sciences, Inc.: Employment, Equity Ownership. Tannheimer:Gilead Sciences, Inc.: Employment, Equity Ownership. Forslund:Gilead Sciences, Inc.: Employment, Equity Ownership. Tasian:Gilead Sciences, Inc.: Research Funding; Incyte Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding.

The Age of the Hematopoietic Microenvironment Specifies Lineage Commitment in MLL-Rearranged Leukemia

Leukemias are a heterogenous group of hematologic malignancies classified by their pathobiologic characteristics including differentiation lineage (myeloid, B-lymphoid, T-lymphoid), oncogenic mutations, karyotypic abnormalities, and clinical characteristics, such as presenting white blood cell count, in order to guide therapy. It is well known that many such characteristics of leukemia are associated with a typical age of disease onset. For example, in children, B-acute lymphoblastic leukemia (B-ALL) occurs most commonly, while acute myeloid leukemia (AML) is more frequent in older adults. Furthermore, B-ALL of infancy, typically diagnosed before 12 months of age, is a unique clinical entity with a poor prognosis compared to B-ALL of later childhood. Infant B-ALL often aberrantly expresses markers of myeloid differentiation, is commonly driven by rearrangements involving the MLL gene, and is treated with hybrid chemotherapy regimens incorporating drugs directed against both myeloid and lymphoid differentiation. The mechanistic underpinnings of the age specificity of leukemia remain unknown, but enhanced understanding could improve disease models and lead to new therapies.The hematopoietic system undergoes a highly conserved process of temporal maturation on a regimented schedule in concert with development of the host. Postnatally, during aging of the hematopoietic system, hematopoietic stem and progenitor cells (HSPCs) alter their self-renewal and lineage biases. Normal development and aging of the blood forming system are associated with changes in the hematopoietic microenvironment. We therefore hypothesized that the normal process of development and aging of blood regulates leukemia phenotypes due to intrinsic changes in HSPCs and/or the hematopoietic niche over time.To test this, we used heterochronic transplantation in a model of MLL translocation-driven leukemogenesis. We first isolated normal HSPCs from midgestation murine fetal liver and adult bone marrow, and transformed these cells with the MLL-AF9 oncogene. Upon transplantation into congenic adult recipients, both cell sources gave rise to pure AML with identical latency and leukemia initiating cell (LIC) content. Next, we transplanted MLL-AF9 -transformed bone marrow HSPCs into congenic adult or neonatal recipients. We found that leukemia developing in neonates contained a small (3.5 ± 1%) population of cells expressing the B-cell marker B220, which were not present in pure AML developing in adults. Serial transplantation of this neonatal leukemia through neonatal recipients decreased the latency of disease onset to as short as 20 days, coincident with expansion of the B220+ component (to 45 ± 4% in tertiary recipients) and infiltration of thymus, lymph nodes, and testes. This mixed B-lymphoid/myeloid leukemia developing in infant mice bore a similar immunophenotype and transcriptional profile by gene set enrichment analysis (GSEA) to human MLL-AF9 -driven infant ALL. These results extend previous observations that MLL -driven leukemia is sensitive to signals from the microenvironment by demonstrating the importance of the age of the niche in dictating lineage differentiation (Cancer Cell 13: 483 [2008] and Science 316:600 [2007]).The sustainment of mixed lineage leukemia through serial transplantation in neonates near limiting dilution suggested the existence of a multipotent progenitor-like LIC. By sorting neonatal leukemia cells using markers of lymphoid commitment, we identified Flk2+ cells within the L-GMP compartment (Lin- kit-lo CD16/32+ CD34+) primed to produce both B-lymphoid and myeloid cells in single cell clonogenicity assays, while Flk2- LGMPs were strongly myeloid-biased (P < 0.05). By screening for differentially expressed cytokines between adult and neonatal bone marrow stroma, we implicated the chemokine Ccl5 as a factor restraining B-lymphoid commitment in leukemia in pro-myeloid adult niches. Taken together, our data show that the neonatal hematopoietic niche supports the development infant-like mixed B/myeloid leukemia, while an adult niche promotes pure AML from identical cells of origin. We attribute this to inhibition of B-lymphoid output from multipotent progenitor-like LICs by Ccl5 from adult bone marrow stroma. These findings connect normal maturation and aging of the hematopoietic system and its microenvironment to the age specificity of leukemia. DisclosuresNo relevant conflicts of interest to declare.

CCG 1953: Acute Lymphoblastic Leukemia in Infants: Analysis of Prognostic Factors: A Report from the Children's Oncology Group.

Background: Infant ALL has a poor therapeutic outcome. In contrast to other subtypes of ALL, infants have not benefited from risk-adapted therapy and intensification of therapy. This is the first cooperative group trial characterizing all infants at the molecular level for MLL/11q23 rearrangements for better stratification in future trials.Procedure: All infants enrolled on CCG 1953 were tested for MLL rearrangement by Southern Blot, and the 11q23 translocation partner was identified (4;11, 9;11, 11;19, or “other”) by reverse-transcriptase PCR.Results: 115 infants were enrolled (July 1996 to August 2000). Treatment for patients consisted of 4 courses of induction-consolidation-intensification followed by 3 cycles of intensified maintenance and 4 cycles of routine maintenance. Infants with 11q23/MLL abnormalities and a protocol-defined 5-6/6 related or unrelated donor identified within 4 months were assigned to BMT. Central nervous system prophylaxis consisted of triple intrathecal therapy during the first 10 weeks of therapy. Infants not undergoing BMT received additional CNS prophylaxis with very high dose methotrexate. Patients then received one year of intensified maintenance, followed by a year of standard maintenance. Overall event-free survival (EFS) was 41.7% (SD = 9.2%) and overall survival (OS) was 44.8% at 5 years. Five-year EFS for MLL rearranged cases was 33.6%, and for MLL non-rearranged cases was 60.3% (p=.006). 5-year EFS for MLL/11q23 subgroups was as follows: t(4;11): 29.0%; t(11;19): 30.0%; t(9;11): 22.2%; and “other 11q23” 53.5% (p=.122). Multivariate Cox regression analyses showed a rank order of significance for negative impact on prognosis for CD10 negativity, age less than 6 months, and MLL rearrangement, in that order. We did not observe a statistically significant difference in outcome between different MLL translocation partners, though we did observe a non-significant trend toward improved outcome in patients with “other” MLL translocation partners as compared to those with t(4;11), t(9;11) or t(11;19). Toxicity was the most frequent cause of death. Relapse as a first event in 1953 was later (median 295 days) compared to the predecessor1883 infant ALL trial (median 207 days).Conclusions: MLL/11q23 rearrangement, CD10 expression, and age are important prognostic factors in infant ALL, but individual molecular 11q23 translocation partners do not predict outcome. These results indicate that certain subgroups have an unusually poor outcome and are therefore candidates for novel therapeutic approaches in first remission. Given the high mortality, further dose intensification strategies are unlikely to result in better long-term survival.

NUTM1-Rearranged Infant and Pediatric B Cell Precursor Acute Lymphoblastic Leukemia: A Good Prognostic Subtype Identified in a Collaborative International Study

Background and Aims A novel genetic subtype of B cell precursor acute lymphoblastic leukemia (B-ALL) is characterized by rearrangement of NUTM1 (NUTM1r) on 15q14 resulting in fusion of NUTM1 to one of several partner genes such as CUX1, ACIN1, BRD9, and IKZF1. The downstream effects of NUTM1r include upregulation of the proto-oncogene BMI1 and specific fusions also induce transcription of the HOXA gene cluster (Hormann et al. Haematologica 2019; Li et al. PNAS 2018). This novel subtype is rare in children, but appears to be more prevalent among infants negative for KMT2A rearrangement (KMT2Ar) based on the frequency of karyotypic 15q aberrations (De Lorenzo et al. Leukemia 2014). This international collaborative study aimed to determine the frequency of NUTM1r in infant and pediatric cohorts, and to characterize the demographic, clinical and molecular features of NUTM1r-positive B-ALL. Patients and Methods Interfant-related study groups provided NUTM1 screening results for KMT2Ar-negative Interfant-99 and -06 cases with karyotypic 15q aberration, normal karyotype, or missing karyotype. Additionally, NUTM1r-positive cases of any age were collected from the study groups united in the Ponte di Legno consortium. The identified NUTM1r-positive children were diagnosed between 1995-2019, infants (≤365 days of age) included in the Interfant-99 or 06 trials were diagnosed between 2000-2016, and remaining infants between 1986-2019. The techniques used for the detection of NUTM1r were break-apart FISH, RNA sequencing, and RT-PCR. Event-free survival (EFS) and overall survival (OS) were estimated according to Kaplan-Meier, standard error according to Greenwood, and the curves were compared by log-rank test. Results We identified 81 NUTM1r cases, including 35 Interfant-enrolled infants, 10 other infants and 36 children. NUTM1r was reported to be rare among pediatric B-ALL with an estimated frequency range of 0.28-0.86%. The median age among NUTM1r-positive children was 4.5 years (range 1-15). Among KMT2Ar-negative infants the frequency of NUTM1r was 21.7%. Of NUTM1r-positive infants, 54% were &amp;lt;6 months at diagnosis (median 5.6, range 0.4-11.0 months) compared with 16% in the remaining KMT2Ar-negative infants (median 9.3, range 0.1-11.9; p&amp;lt;0.0001). Other baseline characteristics (WBC, gender, prednisone poor response) were similar between NUTM1r-positive and -negative infants. Of the NUTM1r-positive cases, all achieved complete remission, 82% had minimal residual disease &amp;lt;10e-4 at the end of induction, and no patient received stem cell transplant in first remission. The 4-year EFS was 100% in Interfant-enrolled NUTM1r-positive patients versus 74% (95% CI 65.1-81.0, p=0.001) in the remaining KMT2Ar-negative cases (n=126). The better outcome was confirmed also after adjusting for WBC, gender and prednisone response (p=0.0001). The 4-year OS were 100% and 88.0% (95% CI 80.5-92.7) for NUTM1r-positive and other KMT2Ar-negative infant cases, respectively (p-value=0.04). Children and non-Interfant-enrolled infants treated on different treatment protocols showed 89.4% (95% CI 78.6-1) 4-year EFS and 100% 4-year OS. In order of frequency, NUTM1 fusion partners were ACIN1 (30.4%), CUX1 (21.7%), BRD9 (17.4%), ZNF618 (13%), AFF1 (4.3%), SLC12A6 (4.3%), IKZF1 (2.9%), and three novel partners: ATAD5 (2.9%), CHD4 (1.4%) and RUNX1 (1.4%). Infants mainly showed fusions with ACIN1, CUX1, BRD9 and AFF1, associated with HOXA9 upregulation. Older infants and children showed both HOXA-upregulating and non-HOXA-upregulating fusions. Epigenetic profiling showed a distinct pattern of DNA methylation and histone modification of the HOXA gene cluster region in leukemic cells of an ACIN1-NUTM1 pediatric case compared with KMT2Ar-positive and KMT2Ar/NUTM1r-negative pediatric cases. Conclusions NUTM1r ALL was identified as the second largest subtype in infants, found in 21.7% of KMT2Ar-negative infant B-ALL, representing 5-7% of total infant ALL, and associated with excellent outcome on Interfant standard risk protocols. The favorable outcome was confirmed in the Ponte di Legno cohort of infant and pediatric NUTM1r-positive patients enrolled on different treatment protocols over more than two decades. We conclude that NUTM1r ALL is a favorable genetic subtype in infants and children and possibly eligible for treatment reduction. Disclosures No relevant conflicts of interest to declare.

Abstract IA13: Molecular targeted therapies and precision medicine for children with neuroblastoma and other refractory malignancies

Abstract Despite the increase in overall child cancer survival rates, pediatric malignancies such as high-risk neuroblastoma, high-risk leukemias (including MLL-translocated infant ALL), and aggressive brain tumors (including DIPG) remain refractory to current multimodal therapies. We have been developing new treatment approaches for these aggressive childhood cancers by (i) utilizing novel targeted therapies either alone or combined with other new agents or established chemotherapeutic drugs, and (ii) by developing new drugs that target key pathways in these child cancers. In neuroblastoma, we have targeted polyamines, showing that combined inhibition of polyamine synthesis by the ODC1 inhibitor DFMO, and of polyamine uptake using the small-molecule drug AMXT 1501, is highly effective at inhibiting tumor growth in Th-MYCN transgenic mice. This combination also shows great efficacy in preclinical models of DIPG, and clinical trials for these diseases are now being planned. We are also targeting metabolism of arginine, the precursor of ornithine, using the pegylated-recombinant arginase BCT-100, which significantly delays tumor development and prolongs survival of neuroblastoma-prone Th-MYCN mice. We have further shown that combining BCT-100 with either DFMO or conventional chemotherapy results in increased survival benefit. CBL0137 is a nontoxic novel anticancer drug currently in phase I trial for adult refractory and relapsed cancers. CBL0137 destabilizes nucleosomes and traps histone chaperone FACT into chromatin, thereby modulating several anticancer mechanisms. We have shown that CBL0137 is effective in mouse models of neuroblastoma, MLL-rearranged leukemia, and DIPG, and that its action is potentiated by the HDAC inhibitor, panobinostat. Moreover, we have developed OT-82, a novel nontoxic NAMPT inhibitor with impressive anticancer activity against mouse models of high-risk childhood ALL, potentiating standard-of-care drugs, and showing similar efficacy as the three-drug induction-type treatment used for pediatric ALL. In addition, for all Australian children with high-risk malignancies, we have developed the Zero Childhood Cancer national precision medicine program. ZERO utilizes whole-genome and whole-transcriptome sequencing, methylation profiling, and where possible, in vitro and in vivo drug testing. To date (July 2019), 74% of 207 patients on the national clinical trial have received a Multidisciplinary Tumor Board recommendation (therapy, germline referral, or change of diagnosis), and of 25 patients with evaluable response data thus far who have received the ZERO recommended therapy, a significant proportion have had a complete response, partial response, or maintained stable disease. Moreover, early experience with drug efficacy studies suggests these data may corroborate genomic therapeutic recommendations and may also identify unanticipated active therapeutics. Citation Format: Michelle Haber, Laura Gamble, Lin Xiao, Ruby Pandher, Klaartje Somers, Jayne Murray, Aaminah Khan, Denise Yu, Laura Franshaw, Mark R. Burns, Maria Tsoli, Anahid Ehteda, Anthony Cesare, Aisling O’Connor, Francis Mussai, Carmela de Santo, Paul Cheng, Lioubov Korotchkina, Katerina Gurova, Vanessa Tyrrell, Emily Mould, Loretta Lau, Dong Anh Khuong Quang, Chelsea Mayoh, Greg Arndt, Paulette Barahona, Tim Failes, Jamie Fletcher, Noemi Fuentes- Bolanos, Marie-Emilie Gauthier, Andrew Gifford, Dylan Grebert-Wade, Alvin Kamili, Amit Kumar, Sumanth Nagabushan, Tracey O’Brien, Patrick Strong, Alexandra Sherstyuk, David Thomas, Toby Trahair, Katherine Tucker, Meera Warby, Marie Wong, Jinhan Xie, Kathryn Evans, Richard Lock, Olga B. Chernova, Michelle Henderson, Andrei V Gudkov, Paul Ekert, Mark J. Cowley, Glenn M. Marshall, David S. Ziegler, Murray D. Norris. Molecular targeted therapies and precision medicine for children with neuroblastoma and other refractory malignancies [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr IA13.

Dissecting the early steps of MLL induced leukaemogenic transformation using a mouse model of AML

Leukaemogenic mutations commonly disrupt cellular differentiation and/or enhance proliferation, thus perturbing the regulatory programs that control self-renewal and differentiation of stem and progenitor cells. Translocations involving the Mll1 (Kmt2a) gene generate powerful oncogenic fusion proteins, predominantly affecting infant and paediatric AML and ALL patients. The early stages of leukaemogenic transformation are typically inaccessible from human patients and conventional mouse models. Here, we take advantage of cells conditionally blocked at the multipotent haematopoietic progenitor stage to develop a MLL-r model capturing early cellular and molecular consequences of MLL-ENL expression based on a clear clonal relationship between parental and leukaemic cells. Through a combination of scRNA-seq, ATAC-seq and genome-scale CRISPR-Cas9 screening, we identify pathways and genes likely to drive the early phases of leukaemogenesis. Finally, we demonstrate the broad utility of using matched parental and transformed cells for small molecule inhibitor studies by validating both previously known and other potential therapeutic targets.

Single-Cell Multi-Omics Reveals Elevated Plasticity and Stem-Cell-Like Blasts Relevant to the Poor Prognosis of &lt;i&gt;KMT2A&lt;/i&gt;-Rearranged Leukemia

Infant ALL is a devastating malignancy caused by rearrangements of the KMT2A gene (KMT2A-r) in approximately 70% of patients. The outcome is dismal and younger age at diagnosis is associated with increased risk of relapse. To discover age-specific differences and critical drivers that mediate the poor outcome in KMT2A-r ALL, we subjected KMT2A-r leukemias and normal hematopoietic cells from patients of different ages to multi-omic single cell analysis using scRNA-Seq, scATAC-Seq and snmC-Seq2. We uncovered the following critical new insights: Leukemia cells from infants younger than 6 months have a greatly increased lineage plasticity and contain a hematopoietic stem and progenitor-like (HSPC-like) population compared to older infants. We identified an immunosuppressive signaling circuit between the HSPC-like blasts and cytotoxic lymphocytes in younger patients. Both observations offer a compelling explanation for the ability of leukemias in young infants to evade chemotherapy and immune mediated control. Our analysis also revealed pre-existing lymphomyeloid primed progenitor and myeloid blasts at initial diagnosis of B-ALL. Tracking of leukemic clones in two patients whose leukemia underwent a lineage switch documented the evolution of such clones into frank AML. These findings provide critical insights into KMT2A-r ALL and have potential clinical implications for targeted inhibitors or multitarget immunotherapy approaches. Beyond infant ALL, our study demonstrates the power of single cell multi-omics to detect tumor intrinsic and extrinsic factors affecting rare but critical subpopulations within a malignant population that ultimately determines patient outcome.

Clinical Experience of CAR T Cell Immunotherapy for Relapsed and Refractory Infant ALL Demonstrates Feasibility and Favorable Responses

Background: The youngest patients referred for CAR T cell therapy are those with relapsed or refractory (R/R) KMT2A-rearranged infant B-ALL. Infants with relapsed ALL following Interfant-99 therapy have a dismal reported 3-yr OS of 20.9%, indicating the need for novel therapies. Smaller patient size, heavily pre-treated disease and high leukemia burden are often characteristics of this subgroup of patients that pose unique challenges to apheresis and manufacture of a T cell product. Additionally, reports of KMT2A-rearranged leukemia undergoing lineage switch following CD19-targeting pressure raises concern for an increased risk of myeloid leukemia relapses after B-lineage targeted CAR T cell therapy in this population. Here we report our experience using CAR T cell immunotherapy for patients with R/R infant ALL enrolled on clinical trials PLAT-02 (NCT02028455) and PLAT-05 (NCT03330691). Methods: PLAT-02 is a phase 1/2 trial of CD19-specific (FMC63scFv:IgG4hinge:CD28tm:4-1BB:ζ) CAR T cells. PLAT-05 is a phase 1 trial of CD19xCD22 dual specific CAR T cells, transduced with two separate lentiviral vectors to direct the co-expression of the CD19-specific CAR above and a CD22-specific CAR (m971scFv:IgG4hinge-CH2(L235D)-CH3-CD28tm:4-1BB:ζ). Eligible subjects on both studies have R/R B-ALL, an absolute lymphocyte count ≥100 cells/µL, and were at least 1 year of age. In addition, subjects on PLAT-02 were ≥ 10kg, and ≥ 8kg on PLAT-05. For cell manufacture, apheresis products were immuno-magnetically selected for CD4 and CD8 cells. Selected T cells were activated with anti-CD3/CD28 beads, transduced, and grown in culture with homeostatic cytokines to numbers suitable for clinical use. Infant ALL subjects received a range of 5x105 to 10x106 CAR+ T cells/kg following lymphodepleting chemotherapy. Disease response assessments were required at Day 21 and Day 63 following CAR T cell infusion. Adverse events were graded according to CTCAEv4, except CRS which was graded according to 2014 Lee criteria. Results: Eighteen subjects with R/R infant ALL have enrolled on PLAT-02 (n=14) or PLAT-05 (n=4), with a median age of 22.5 months at enrollment (range: 14.5 - 40.1 months). Of these, 2 (11.1%) had primary refractory disease, 8 (44.4%) were in 1st relapse, 7 (38.9%) were in 2nd relapse and 1 (5.6%) was in 3rd or greater relapse. Ten subjects (55.6%) had an M2 marrow or greater at enrollment prior to apheresis, and 9/18 had a history of hematopoietic cell transplant (HCT). The mean ALC was 1309 cells/µL (range 253-6944). Successful CAR T cell products were manufactured in 17/18 subjects, including in 9/9 subjects with no prior history of HCT. Of these, 16/17 subjects with available products were infused, with a median follow up of 26.9 months. One subject died of disease complications prior to CAR T cell infusion. Of the 16 treated subjects, 1 is pending disease and toxicity assessments. The maximum grade of CRS was 3 and occurred in two of 15 evaluable subjects (13%) and neurotoxicity was limited to a maximum grade of 2. Fourteen of 15 (93.3%) achieved an MRD negative complete remission (MRD-CR) by Day 21. Of the 14 subjects with an MRD-CR, 6 went on to HCT with 1 subsequent CD19 negative relapse. Of the 8 subjects who did not proceed to HCT, 1 developed lineage switch at one month following CAR T cells, and 1 died of infectious complications with aplasia. A "wait and watch" approach was taken for the remaining 6 subjects, and 2 developed CD19+ relapse. The incidence of lineage switch among the infant ALL group was 1/15 (6.7%). The estimated 1-year LFS was 66.7% and 1-year OS was 71.4%. Conclusion: This is the largest reported cohort to date of R/R infant B-ALL subjects treated with CAR T cell therapy. We report successful manufacture and administration of a CAR T cell product in the significant majority of infant subjects. Toxicity and MRD-CR rates are comparable to that of non-infant ALL subjects. In our experience, subjects with R/R infant ALL are not at increased risk for lineage switch relapse compared with the entire study populations following B-antigen targeting CAR T cell immunotherapy. Numbers in this report are too small to make definitive conclusions about the value of consolidative HCT. However, the LFS of this cohort is remarkably higher when compared with historical controls. Future work is focused on overcoming feasibility issues for the smallest of subjects, to enable a larger number of these cases to access CAR T cell therapy. Disclosures Pulsipher: Amgen: Other: Lecture; Bellicum: Consultancy; Miltenyi: Research Funding; Medac: Honoraria; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Jazz: Other: Education for employees; Adaptive: Membership on an entity's Board of Directors or advisory committees, Research Funding; CSL Behring: Membership on an entity's Board of Directors or advisory committees. Wayne:AbbVie: Consultancy; Spectrum Pharmaceuticals: Consultancy, Research Funding; Servier: Consultancy; Kite, a Gilead Company: Consultancy, Research Funding. Jensen:Bluebird Bio: Research Funding; Juno Therapeutics, a Celgene Company: Research Funding. Gardner:Novartis: Honoraria.

Frequency and Prognostic Impact of IKZF1 and Co-Existed Gene Alterations in Childhood B-ALL in Taiwan

Introduction: IKZF1 deletion was firstly reported to be very frequent in Philadelphia (Ph) positive B-cell acute lymphoblastic leukemia (B-ALL), and continuously found in other B-ALL subtypes. IKZF1plus defined as the presence of any of CDKN2B, CDKN2A,PAX5 or PAR1 in the absence of ERG deletion and was reported to describe a new minimal residual disease (MRD)-dependent poor prognostic profile in B-ALL.1 We aimed to analyze the frequency and prognostic relevance of IKZF1 deletion with or without co-occurring gene alterations in Taiwanese children with B-ALL treated with TPOG-ALL-2002 protocol or MRD-directed TPOG-ALL-2013 protocol. Methods: Bone marrow samples at diagnosis from 561 children excluding Ph+ and infant ALL were analyzed. Detection of gene deletion was carried out with multiplex ligation dependent probe amplification (MLPA) kit (SALSA MLPA P335 and P327) and mutations of RAS pathway genes (NRAS, KRAS, and PTPN11) were assessed by Sanger sequencing. The outcome was analyzed on 259 patients treated with TPOG-ALL-2002 protocol and 158 patients treated with TPOG-ALL-2013 protocol. Results: IKZF1 deletions were present in 11.2% of 561 patients. DEL 4-7, DEL 1-8, and other variants were present in 50.8%, 14.3%, and 34.9%, respectively of 63 IKZF1-deleted patients. Co-existence of additional deletions with IKZF1 deletion was detected in 34.0% for CDKN2B, 37.3% for CDKN2A, 27.7% for PAX5, 2.1% forPAR1, and 8.0% for ERG. IKZF1plus comprised of 4.4% of 546 patients who had all the genes examined. Co-existed mutations of RAS pathway genes were detected in 5.0% for PTPN11, 13.3% for NRAS, and 10.0% for KRASin the IKZF1-deleted patients. The 5-year event-free survival (EFS) of IKZF1-undeleted patients was significantly better compared with IKZF1-deleted patients in TPOG-2002 cohort (82.5 ± 2.6% vs. 56.9 ± 9.1%, P = 0.002) as did TPOG- 2013 cohort (89.4 ± 3.9% vs. 59.1 ± 19.8%, P = 0.012). The 5-year overall survival (OS) of IKZF1-deleted patients was worse than that of patients with undeleted IKZF1 in TPOG-2013 cohort (58.3 ± 19.8% vs. 89.2 ± 4.2%, P = 0.004) but not in TPOG-2002 cohort (90.9 ± 5.0% vs. 86.9 ± 2.3%, P = 0.846). DEL 1-8 patients had an inferior 5-year EFS (33.3 ± 19.2%) compared with DEL 4-7 (63.3 ± 12.0%,) or DEL-others (60.0 ± 18.4 %) in TPOG- 2002 cohort (P = 0.003). No significant difference in OS was observed between patients among different IKZF1-deleted subtypes in TPOG-2002 cohort (P = 0.283) (Fig. 1A) but there was different in TPOG-2013 cohort (P = 0.003) (Fig. 1B). Patients with IKZF1 deletion alone had comparable 5-year EFS and 5-year OS compared with patients with IKZF1plus in TPOG-2002 cohort whereas patients with IKZF1 deletion alone had worse 5-year EFS (33.3 ± 27.2% vs. 80.0 ± 17.9%, P = 0.284) and 5-year OS (33.3 ± 27.2% vs. 80.0 ± 17.9%, P = 0.415) than that of KZF1plus in TPOG-2013 cohort. The prognostic impact of IKZF1 deletion with co-existed gene alterations in TPOG-2002 cohort was further analyzed as follows: IKZF1plus with ≧ 2 co-existed deletions had an inferior 10-year EFS (23.8 ± 20.3%) compared with patients with IKZF1 deletion alone (36.7 ± 17.5%) or IKZF1-undeleted patients (81.5 ± 2.7%) (P = 0.005) (Fig. 1C). Three patients carried both IKZF1 and ERG deletions had a superior 5-year EFS (100%) compared with IKZF1-deleted alone or IKZF1-undeleted patients (P = 0.005) (Fig. 1D). The 10-year EFS of patients with any gene mutation of RAS pathway was worse than that of patients with RAS wild-type genes (66.6% ± 6.4% vs. 83.7% ± 2.6%, P = 0.011). In multivariate analysis, in addition to the initial white blood count &gt; 50 x 109/L and KMT2A-rearranged, IKZF1 deletion (HR=2.609, 95% CI: 1.363-5.034; P = 0.004) and RAS pathway gene mutations (HR=2.360, 95% CI: 1.336-4.168; P = 0.003) were independent genetic predictors for inferior EFS. Co-existence of IKZF1 deletion with RAS pathway mutations had worst 5-year EFS (20.0% ± 12.6%, P &lt; 0.0001, Fig. 1E), and 10-year OS (53.3% ± 17.6%, P = 0.042, Fig. 1F). Conclusions: Our results showed that IKZF1 deletion alone was associated with an inferior OS in MRD-directed TPOG-2013 treated cohort but not in TPOG-2002 cohort in which it became significant when co-existed with RAS pathway mutations. Reference: 1. IKZF1plus defines a new minimal residual disease-dependent very-poor prognostic profile in pediatric B-cell precursor acute lymphoblastic leukemia. J Clin Oncol. 2018, 1240-1249. Figure 1 Disclosures No relevant conflicts of interest to declare.

Outcome of Infants Younger Than 1 Year with Acute Lymphoblastic Leukemia Treated with the Interfant-06 Protocol; Results from an International Randomised Study

INTRODUCTIONInfant acute lymphoblastic leukemia is a high-risk subtype characterized by a very high incidence of KMT2A (MLL) gene rearrangements. In 1999, a large international consortium (Interfant) commenced; the first Interfant-99 protocol (Pieters et al, Lancet 2007) was used as baseline for the current Interfant-06 study, which run in 18 groups worldwide.AIMSTo determine:In a randomised study whether two AML-type courses ADE (araC, daunorubicin, etoposide) and MAE (mitoxantrone, araC, etoposide) after induction improved outcome compared to the consolidation IB course in KMT2A-rearranged cases (MR and HR risk group). With a target of 320 randomised cases, the study had 80% power to detect a DFS difference of 16% at 3 years (41% DFS in the control arm, α=0.05).The prognostic relevance of clinical and biological parameters.The outcome of Interfant-06 versus Interfant-99.The role of SCT in HR patients and in MR patients with MRD ≥ 10e-4 at the start of OCTADAD.As Interfant-99 showed outcome differences between the West-European/North-American groups who started Interfant (original countries) versus other countries that joined later, analyses were also performed separately in these groups.RISK GROUPS AND TREATMENTThree risk groups were defined: low risk (LR) - KMT2A germline; high risk (HR) - KMT2A rearrangement plus age < 6 months plus white blood cell count (WBC) >300 x 10e9/L or a poor prednisone response; medium risk (MR) - comprised all other KMT2A-rearranged cases.Treatment consisted of induction (prednisone prephase, dexamethasone, VCR, asparaginase, DNR), consolidation phase, either IB (araC, 6-MP, cyclofosfamide) or ADE/MAE in the experimental arm, MARMA (HD-araC, HD-MTX, 6-MP, asparaginase), OCTADAD (VCR, dexamethasone, asparaginase, DNR, 6-TG, araC, cyclofosfamide) and maintenance (6-MP, MTX) with intrathecal therapy in all courses. All HR patients were allocated to receive allogeneic stem cell transplantation (SCT). After June 2009, SCT was also undertaken in MR cases with MRD ≥ 10 e-4 at start of OCTADAD as the Interfant-99 update showed a very poor outcome for these patients.RESULTS651 infants were included (167 LR, 320 MR, 164 HR). In the original countries, 6-yr EFS (SE) and survival rates were 49.4 (2.5) and 62.1 (2.4) respectively, which were each 6% (not significantly) higher than in Interfant-99 study. In the other countries, the 6-yr EFS and survival rates were 39.0 (3.6) and 49.7 (3.7). The outcome of patients treated in the original countries was significantly better than in other countries due to lower rates of toxic death and resistant disease. The 6-year EFS and survival of all 651 cases were 46.1 (2.1) and 58.2 (2.0) respectively.330/407 eligible cases (81%) were randomised. The 6-yr cumulative incidence (CI) of relapse with ADE/MAE was 47.5 (4.0) which was not significantly (p = 0.11) lower than the 54.9 (4.1) with IB. The 6-yr CI of death in remission was 10.2 (2.4) with ADE/MAE versus 8.3 (2.2) with IB (p = 0.51). This resulted in no significant differences in 6-yr DFS rates of 39.3 (4.0) and 36.8 (3.9) respectively (the 3-yr DFS were 45.3 [3.9] versus 38.6 [3.9]).The 6-yr EFS rate of 164 HR patients was 20.9 (3.4) with the intention to use SCT in all HR patients; 76 of them received SCT as many HR patients had very early events. Out of 23 MR patients with MRD ≥ 10e-4 before OCTADAD, 16 received SCT; 4/23 are in CCR. In total, 14.4% of MR/HR patients receiving SCT died of transplant-related mortality.T(4;11) and t(11;19) translocated cases had a significantly lower EFS than other KMT2A-rearranged cases (p = 0.005). Multivariate analyses identified KMT2A rearrangement, age < 6 months, WBC ≥ 300 x 10e9/L, and poor prednisone response but not CD10, type of MLL translocation and sex as significant predictors of adverse DFS.CONCLUSIONSThe Interfant-06 study is by far the largest study performed in infant ALL. The 6-yr survival in West-European/North-American countries is 62%, which is 12% higher than in other participating countries. Two postinduction AML-type chemotherapy courses versus course IB did not lead to a significant better DFS. 6-yr survival rates are 6% higher than on Interfant-99, which is not a statistically significant improvement. Pilot studies by COG and Interfant are currently exploring the feasibility of adding azacytidine and blinatumomab to the Interfant backbone, which may lead to a worldwide randomised trial with these drugs in infant ALL. DisclosuresLocatelli:Miltenyi: Honoraria; Bellicum: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees; bluebird bio: Consultancy; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees. Vora:Jazz: Other: Advisory board; Medac: Other: Advisory board; Pfizer: Other: Advisory board; Amgen: Other: Advisory board; Novartis: Other: Advisory board.

Inhibition of PRMT1 Mediated FLT3 Arginine Methylation As a Potent Therapeutic Strategy for MLL-r ALL

Mixed-lineage leukemia-rearranged (MLL-r) ALL, seen in 70% of infant ALL, has a dismal prognosis compared to those with wild type MLL1 gene. Transcriptional profiling has identified Fms-like receptor tyrosine kinase 3 (FLT3) as one of the most significantly upregulated genes in MLL-r ALL. The highly expressed FLT3 protein is activated by the autocrine ligand, making the kinase a therapeutic target. FLT3 tyrosine kinase inhibitors (TKIs) such as PKC412, although effective in kinase inhibition, partially impair survival of MLL-r ALL cells and clinical trial results are not promising, promoting us to ask whether FLT3 regulates the ALL cells survival also through a kinase-independent mechanism.Herein, we report the finding of dimethylated arginines on FLT3, detected through mass spectrometry analysis of a MLL-r ALL specimen and a MLL-r ALL line SEM. The most conserved and enriched of dimethylated arginines are residues R972/R973. Using home-made arginine methylation (R-Me) antibody, we found that PRMT1, which is responsible for most type I arginine methyltransferases activity, catalyzes FLT3 methylation. Immunoblot (IB) analysis validated the expression of FLT3 R-Me in MLL-r ALL samples (6 out of 6) and MLL-r ALL lines (4 out of 4). Analysis of the GEO dataset (GSE13204) revealed that PRMT1 mRNA levels are increased in MLL-r ALL relative to normal cells (MLL-r, n=70 vs. normal, n=73, p<0.0001). We studied FLT3 R-Me biological function using two approaches that specifically blocked FLT3 methylation levels: cells expressing FLT3 methylation deficient construct (R972/973K, arginine [R] to lysine [K]) exhibited reduced survival (BaF3: FLT3-WT 98.5±0.11% vs. R972/973K 71.5±0.53%, p=0.0004); knockdown of PRMT1 in SEM cells also had an inhibitory effect (siCtrl 95.1±0.1% vs. siPRMT1 74.7±0.5%, p=0.0007). Moreover, the type I arginine methyltransferase inhibitor MS023 (5 µM) treatment markedly induced apoptosis of primary ALL cells but spared normal counterparts from healthy donors (ALL: vehicle 10.4±0.4% vs. MS023 23.7±0.8%, n=4; p<0.0001; normal CD19+: 8.3±0.3% vs. 8.2±0.1%, n=3, p=0.86). Interestingly, inhibition of FLT3 methylation decreased FLT3 phosphorylation at tyrosine 969 (Y969) but not Y589/591 or Y842. Expression of R972/973K decreased FLT3 downstream signaling like phospho-STAT5 and -AKT to a greater extent than that of Y969F mutant (Y to phenylalanine [F] substitution, mimics loss of Y phosphorylation). Next, FLT3 WT, R972/973K or Y969F transduced primary MLL-r ALL cells were transplanted into NSGS mice for analysis of leukemia development (n=6/group). Mice transplanted with FLT3 Y969F MLL-r ALL had longer survival relative to FLT-WT injected animals (p=0.0031), and the median survival was further extended in mice injected with R972/973K mutant compared with FLT3 Y969F MLL-r ALL (p=0.0007). Additionally, PKC412 treatment alone did not alter FLT3 R-Me, and high FLT3 methylation level in SEM cells was not affected by FLT3 ligand stimulation, confirming that the function of R-Me is independent of FLT3 phosphorylation. Importantly, we observed that the combination of MS023 with PKC412 significantly induced a higher rate of apoptosis in primary MLL-r ALL cells compared with each drug alone (control, 10±0.43%, MS023, 21.1±1.2%, PKC412, 21.5±0.11%, combination, 39.8±2.9%, PKC412 vs combination, p<0.01, n=4). We further tested the effects of in vivo administration of MS023 plus PKC412 on primary MLL-r ALL cells xenografted in NSGS mice. Following engraftment >1% in peripheral blood, mice were subdivided into four groups and treated with vehicle, PKC412 (100 mg/kg, i.g.), MS023 (80 mg/kg, i.p, bid), or the combination (n=7/group) for 4 weeks. The BM tumor burden of CD45+ CD19+ cells was reduced in single drug-treated mice cohorts, with further reduction after combination treatment (vehicle, 94.4±0.5%, PKC412, 50.2±6.3%, MS023, 55.6±4.5%, combination, 30.7±4.9%, PKC412 vs. combination, p<0.001). Secondary transplantation of BM cells from mice receiving combination treatment resulted in significantly reduced BM engraftment at 16 weeks compared to PKC412 treatment alone (PKC412, 62.2±4.9%, combination, 8.4±5.1%, n=5, p<0.0001), indicating reduced leukemia initiating capacity.Our results support further exploring the molecular function of FLT3 R-Me. We will determine whether PRMT1 and FLT3 methylation are potential druggable targets in MLL-r ALL. DisclosuresKonopleva:Stemline Therapeutics: Research Funding.

NUTM1 Is a Rare but Recurrent Fusion Gene Partner in B Cell Precursor Acute Lymphoblastic Leukemia Associated with Increased Expression of Genes on Cytoband 10p12.31

▪INTRODUCTION In 20-25% of the pediatric B cell precursor acute lymphoblastic leukemia (BCP-ALL) patients, the driving cytogenetic aberration is unknown. It is important to identify more primary lesions in this remaining B-other group to provide better risk stratification and identify possible treatment options. In this study, we aimed to identify novel recurrent fusion genes in BCP-ALL through RNA sequencing.METHODS We used paired-end total RNA Illumina sequencing to detect fusion genes with STAR-fusion and FusionCatcher in a population-based ALL cohort (n=71). We used Affymetrix U133 Plus2 expression arrays in a larger population-based ALL cohort (n=661) and an infant ALL cohort (n=70) to compare gene expression levels. Fluorescent in situ hybridization (FISH) was performed using Cytocell NUTM1 break-apart probe set MPH4800.RESULTS We identified an in-frame SLC12A6-NUTM1 fusion transcript composed of exons 1-2 of SLC12A6 fused to exons 3 to 8 of NUTM1 by RNA sequencing. Both genes are located on 15q14 within 5.3 Kb distance on opposite strands, and the fusion could result from an inversion. The fusion transcript is predicted to encode almost the total NUTM1 protein including the acidic binding domain for the histone acetyltransferase EP300. The SLC12A6-NUTM1 fusion case showed high NUTM1 expression, while NUTM1 expression was absent in the remaining cases.Using gene expression profiling, we identified four additional pediatric and two non-KMT2A-rearranged infant BCP-ALL cases with high NUTM1 expression. In the population-based cohort reflecting all different cytogenetic subtypes, these cases were restricted to the B-other group without known sentinel cytogenetic abnormalities. FISH showed a NUTM1 break apart pattern in all four tested NUTM1-positive cases indicative of a balanced translocation. RNA sequencing confirmed an ACIN1-NUTM1 fusion in one of the infant cases. We conclude that NUTM1 is normally not expressed in leukemic lymphoblasts, and that its expression can be induced by a gene fusion. The karyotypes of the predicted NUTM1 fusion cases combined with RNA sequencing data suggest that different chromosomal rearrangements are involved, likely resulting in different NUTM1 fusion partners. In literature, BRD9-NUTM1, IKZF1-NUTM1, and CUX1-NUTM1 fusions were reported in pediatric B-other cases, and BRD9-NUTM1 and ACIN1-NUTM1 fusions were reported in non-KMT2A-rearranged infants. Our combined aberrant gene expression and FISH results indicate that NUTM1 fusions occur in 2.4% (5/210) of pediatric and in 28% (2/7) of infant BCP-ALL cases without a sentinel cytogenetic aberration.The recurrence of NUTM1 aberrations in BCP-ALL cases without a known driver and the resulting expression of NUTM1 suggests that this fusion could be a new oncogenic driver in leukemia. All seven patients with a NUTM1 fusion achieved continuous complete remission with a median follow-up time of 8.3 years (range 4.8-13.8 years), suggesting that NUTM1 fusions in BCP-ALL have a favorable prognosis.To get an insight in the underlying biology, we compared gene expression between NUTM1-positive and NUTM1-negative pediatric B-other cases. We identified 130 differentially expressed probe sets (FDR ≤0.01) with a peculiar enrichment of those located on chromosome band 10p12.31 (Bonferroni adjusted p=4.05E-04). The genes in cytoband 10p12.31, including BMI1, were variably upregulated in 6/7 NUTM1-positive cases and positively correlated to NUTM1 expression levels. The NUTM1 protein is capable of binding and hereby stimulating the histone acetyltransferase activity of the EP300 protein. The EP300 protein preferentially binds a risk allele of BMI1 associated with increased risk for BCP-ALL. The BMI1 protein has been shown to convert BCR-ABL1-positive progenitor cells into BCR-ABL1-positive BCP-ALL cells. Hence, we postulate that NUTM1 fusion proteins contribute to leukemogenesis by stimulating EP300, leading to upregulation of BMI1 and other 10p12.31 genes in BCP-ALL.CONCLUSIONNUTM1 fusions are a rare but recurrent event in BCP-ALL that seems to have a good prognosis. The NUTM1 fusions result in expression of the normally silent NUTM1 gene and are associated with upregulation of a cluster of genes on 10p12.31 including the leukemogenic BMI1 gene. DisclosuresNo relevant conflicts of interest to declare.

3108 - Blockage of NG2 Antigen Synergizes in Vivo with VXL Treatment Resulting in a Delayed Relapse in a Nsg Mice-Based Preclinical Model for MLL-Rearranged BCP-ALL

MLL-rearranged (MLLr) infant B-cell precursor ALL (iMLLr-BCP-ALL) has a dismal prognosis and is associated with a pro-B/mixed phenotype, therapy refractoriness and frequent central nervous system (CNS) disease/relapse. Neuron-glial antigen 2 (NG2) is specifically expressed in MLLr leukemias and is clinically used because of its predictive value. However, its role in MLL-mediated leukemogenesis remains elusive. We previously reported in a large patient cohort that high expression of NG2 associates with poor outcome. Furthermore, we also showed that although NG2 is a malleable marker not enriching for leukemia-initiating cells, it is consistently upregulated in blasts infiltrating extramedullar hematopoietic sites and the CNS, suggesting a key role in migration. We here hypothesize that NG2 may be a therapeutic target for iMLLr-BCP-ALL. Here, we demonstrate that blocking of NG2 in vivo promotes the exit of MLLr BCP-ALL blasts from BM to PB, suggesting a role for the extracellular domain of NG2 in retaining MLLr blasts within the BM niche. When combined with standard ALL chemotherapy (Vincristine, Prednisone and L-Asp, VxL), anti-NG2 resulted in 3-fold decreased levels of minimal residual disease. Importantly, follow-up of mice in (near)-complete remission evidenced that anti-NG2 when combined with VXL significantly delayed mice relapse. Preliminary studies suggest that NG2+ blasts are more resistant than NG2- blast to Glucocorticoids. In addition, NG2 promoter has binding sites for GC receptors, and treatment with dexamethasone results in increased levels of NG2, likely reflecting a NG2-mediated mechanism to escape GC toxicity. Collectively, our data propose NG2 antigen as a therapeutic target in MLLr-B-ALL using either antibody-based or CAR T-cell-based immunotherapy.

Maternal fetal loss history and increased acute leukemia subtype risk in subsequent offspring: a systematic review and meta-analysis.

History of fetal loss including miscarriage and stillbirth has been inconsistently associated with childhood (0-14 years) leukemia in subsequent offspring. A quantitative synthesis of the inconclusive literature by leukemia subtype was therefore conducted. Eligible studies (N = 32) were identified through the screening of over 3500 publications. Random-effects meta-analyses were conducted on the association of miscarriage/stillbirth history with overall (AL; 18,868 cases/35,685 controls), acute lymphoblastic (ALL; 16,150 cases/38,655 controls), and myeloid (AML; 3042 cases/32,997 controls) leukemia. Sensitivity and subgroup analyses by age and ALL subtype, as well as meta-regression were undertaken. Fetal loss history was associated with increased AL risk [Odds Ratio (OR) 1.10, 95% Confidence Intervals (CI) 1.04-1.18]. The positive association was seen for ALL (OR 1.12, 95%CI 1.05-1.19) and for AML (OR 1.13, 95%CI 0.91-1.41); for the latter the OR increased in sensitivity analyses. Notably, stillbirth history was significantly linked to ALL risk (OR 1.33, 95%CI 1.02-1.74), but not AML. By contrast, the association of ALL and AML with previous miscarriage reached marginal significance. The association of miscarriage history was strongest in infant ALL (OR 2.34, 95%CI 1.19-4.60). In this meta-analysis involving >50,000 children, we found noteworthy associations by indices of fetal loss, age at diagnosis, and leukemia type; namely, of stillbirth with ALL and miscarriage history with infant ALL. Elucidation of plausible underlying mechanisms may provide insight into leukemia pathogenesis and indicate monitoring interventions prior to and during pregnancy.

Infant Lymphoblastic Leukemia: A Single Center Experience

Infantacute lymphoblastic leukemia (ALL) is a rare disease and consist of 4-5% of all childhood ALL. Despite improved survival rates of childhood ALL, infants with ALL have a worse prognosis . There were 206 patients diagnosed with ALL at Hacettepe University, Pediatric Hematology Department between 1 January 2008 and 31 December 2016, 11 out of 206 (5.3%) were diagnosed as infant ALL. We aimed to evaluate clinical findings, laboratory results and treatment outcome of our infant ALL patients retrospectively. There were 9 boys and two girls with a median age of 8 months (range 1-12 months) and nine out of 11 (82%) were older than 6 months of age at the initial diagnosis. Interestingly 2 of 3 siblings born from spontaneous triplet pregnancy diagnosed at the same time when they were 7 months old while the other sibling diagnosed 4 months later when he was 11 months old. Physical examination revealed hepatosplenomegaly in 10 out of 11 patients and 1 patient had extramedullary involvement with subcutaneous nodules on the forehead. On admission median WBC count was found 202x10-9/L (range 37.4-739x10-9/L) with a median hemoglobin level 9 g/dL (range 2.6-12.9 g/dL) and median platelet count 81x10-9/L (range10- 314x10-9/L). Immunophenotyping revealed CALLA+ B cell leukemia in 3 patients, CALLA- B cell in 6 patients and T cell leukemia in 2 patients. Cytogenetic analysis showed t(4;11) positivity in 4 patients and 3 of them were spontaneous triplets. Initial CNS involvement were positive in 2 of the patients. Eight patients (73%) were treated with Interfant protocols (Interfant 99 (n=4), Interfant 06 (n=4)) and 3 patients received Modified St. Jude Total XV protocol. Relapse were observed in 2 patients (one had both CNS and bone marrow relapse one year after the completion of chemotherapy and the other experienced an early bone marrow relapse at the 7thmonth of the treatment). Hematopoetic stem cell transplantation (HSCT) was performed in five patients from matched related donor (n=2), unmatched related donor(n=1), and matched unrelated donor(n=2). Moreover, relapse after HSCT was observed in 2 patients 3 and 4 months later; one received second transplantation and survived while the other died because of progressive disease. Four of the five patients who had HSCT survived 74 months, 45, 18 and 13 months after the initial diagnosis, respectively. After a median of 18 months (range 4- 74 months ) follow-up period 6 patients are alive (45 %) and on remission 4 after HSCT and 2 with only chemotherapy. Infant leukemia usually presents with high hyperleucocytosis and extramedullary involvement. The prognosis is poor and international colloborative studies reprorted that 4 year -event free survival rates were between 28-54%. Our results suggest that bone marrow transplantation seems to be a good and efficient choice of treatment. However there is still a big issue to decide which patient should go under transplantation and more studies are needed to reevaluate the eligibility criteria for HSCT in this group of patients. Moreover to the best of our knowledge infant leukemia in triplets without monozygocity were first time reported. DisclosuresNo relevant conflicts of interest to declare.

Molecular Profiling of High-Risk Pediatric Acute Myeloid Leukemia

Background: Several landmark genomic profiling studies have dramatically advanced our understanding of the origin, progression and clonal evolution of adult acute myeloid leukemia (AML) and directly impacted clinical care. However, very little is known about the mutational landscape of pediatric AML, a distinct entity that shares few genetic and clinical characteristics with adult AML. To investigate potential drivers of high-risk pediatric AML, comprehensive genomic profiling was performed on high-risk AML samples as part of a prospective clinical next-generation sequencing program.Methods:Samples obtained from patients with known high-risk features at diagnosis or with refractory or relapsed AML were selected for molecular profiling. Comprehensive testing included whole-exome sequencing (WES) of matched tumor (bone marrow or chloroma tissue) and normal tissue (peripheral blood or buccal swab) samples and transcriptome analysis (RNAseq). Targeted sequencing of 467 cancer-associated genes was used when tumor tissue was limited. Sequencing was performed on Illumina's HiSeq 2500 with 150X and 500X average coverage for WES and targeted sequencing, respectively. Variants were filtered to select alterations in cancer-related genes or genes relevant for patient care.Results:Fifteen patients with AML (mean age 7.7 yrs; range 0.75-19 yrs) met high-risk criteria (high-risk features at diagnosis = 4, relapsed disease = 8, refractory disease = 3) and were selected for profiling. WES and RNAseq were performed on 11 samples, WES only on 3 samples and targeted DNA sequencing on 1 sample. The median number of variants was 60 (range 14- 5950) per case. After filtering, 54 mutations were identified in 35 genes with a mean of 3.6 mutated genes per patient sample (range 0-14); two samples only carried a fusion gene with no other genetic alterations. At least one driver genetic alteration was detected in each patient sample. Thirteen samples carried mutations in at least one gene known to be altered in AML (e.g. IDH1, WT1, TP53, NRAS) (mean, 2; range, 1-6) and 5 samples carried novel mutations in 15 genes not previously implicated in AML (e.g. CARD9, CHD9, Axin1). Mutations in 11 AML related genes were detected in more than one sample including NRAS in 4, TP53 in 3 and KRAS, PTPN11 PHF6, JAK3 in 2 samples each; genes not previously implicated in AML were only mutated in single patients. Of note, mutations in genes encoding members of the RAS pathway occurred in 60% of cases (9/15 samples). RNAseq identified gene fusions in 7/11 samples (63%). Four fusions involving KMT2A and core binding factor genes were also detected by FISH while three fusions were detected by RNAseq only: NUP98-NSD1 in two patients and CBFA2T3-GLIS2 in one patient. Samples carrying driver gene fusions had the lowest number of mutated genes (0-1) compared to samples lacking a gene fusion (1-5 mutated genes), with one exception of a patient with history of infant ALL who later developed KMT2A-AFF1AML with the highest number of mutated genes (n=14). There was no correlation between the number of mutated genes and age, clinical characteristics, initial risk classification at diagnosis or intensity of therapy prior to sequencing.Conclusion:Our study provides an initial overview of the genetic alterations that characterize high-risk, chemo-resistant pediatric AML. Analysis of the data highlights the overall low genetic complexity of high-risk AML despite the aggressive clinical behavior and exposure to intense chemotherapy, including stem cell transplant. Of interest, similar to adult AML, we found that mutations leading to aberrant activation of the RAS pathway were also very frequent in our cohort of pediatric high-risk AML, while genes typically mutated early in the process of leukemogenesis in adult AML, such as NPM1, DNMT3A, FLT3, IDH1, IDH2 were not affected. Such findings suggest that distinct, age-specific mechanisms of leukemogenesis might exist. Furthermore, our data also highlights the important role of RNA sequencing in complementing current standard diagnostic tools, allowing the identification of driver fusion genes in samples for which no other driver event is detected. Larger studies, preferably including diagnostic samples and utilizing broader approaches, are needed to better understand the mechanisms responsible for the initiation and progression of childhood AML. DisclosuresNo relevant conflicts of interest to declare.

Interaction with the DNA Binding Domain of MLL and a Small Molecular Drug Induces Proteolysis of MLL-Fusion Proteins and Ablates Leukemic Stem Cell Activity

IntroductionLeukemias harbouring 11q23 abnormalities causing mixed-lineage leukaemia gene (MLL) rearrangements are associated with poor clinical outcomes. Despite being an aggressive leukaemia, the MLL rearranged infant ALL has among the lowest mutation rates reported for any cancer. This means that to improve survival for patients with this aggressive leukaemia we need drugs that target the abnormal proteins produced by the MLL fusion gene or that interact with the abnormal MLL fusion protein to shut down the cellular machinery that drives these leukemias. Indeed, targeting MLL fusion dependent gene pathways has become a major focus.Our previous studies have shown that inhibition of MLL-fusions, in a conditional mouse model of MLL-ENL driven acute myeloid leukaemia, resulted in a block in self-renewal of the leukemic cells and ablated the leukaemia in the mice. This led us to hypothesise that if, we could achieve pharmacological inactivation of the MLL fusion proteins, we could achieve improved clinical outcomes. To achieve this, we set out a drug screening programme in acute leukaemia with the aim to discover drugs that can inactivate MLL-fusion oncoproteins.ResultsOur drug discovery pipeline screened clinical approved drugs for their ability to inhibit the function of the MLL fusion protein. This lead to the discovery of a drug that interacts with the DNA binding domain within the MLL fusion protein. This interaction destabilises the MLL fusion protein so that the fusion protein gets degraded within 24 hours of addition of the drug. So far, we have shown that we can inhibit and induce the degradation of MLL-AF9, MLL-AF6 and MLL-AF4 (and WT MLL) in the human MLL rearranged cell lines (THP-1, SHI-I and MV4-11), in primary immortalized cells in which the MLL-AF9 is overexpressed from a lentiviral backbone and in patient derived leukemic samples.Inactivation/degradation of the MLL fusion protein should shut down the cellular machinery that drives these leukemias. It is well established that MLL-fusions lead to abhorrent upregulation of its target genes HOXA9, MEIS1 and c-MYB. Treatment of MLL rearranged cells resulted in the downregulation of these MLL-fusion target genes within 24hrs of addition of the drug. Furthermore, Gene Set Enrichment Analysis of drug treated MLL-AF9 cells showed strong negative enrichment to various published MLL fusion target gene sets. Inactivation of MLL fusion protein should also result in block in self-renewal as we have previously shown in our conditional mouse model. Indeed, Gene Set Enrichment Analysis showed negative enrichment to published Leukemic Stem Cell gene set. To analyse the impact of drug treatment on self-renewal, we used a well-established self-renewal assay, whereby self-renewal is assessed by their ability to form colonies derived from single cells in methylcellulose. While treatment had no significant impact on the colony formation of CD34 positive cord blood progenitors, the drug was able to block the colony formation ability of MLL rearranged cell lines while only slowing a slight reduction in in the colony numbers of non MLL rearranged cell lines.ConclusionOverall, the data indicates that we may have discovered a new targeted treatment for MLL rearranged leukemia, which shows excellent clinical properties. We have successfully generated Patient Derived Xenografts (PDX) models and we are currently testing this drug to verify its effectiveness in the treatment on PDX. We will include this data in our presentation. DisclosuresNo relevant conflicts of interest to declare.

Differential Regulation of c-Myc/Lin28 Discriminates Subclasses of Rearranged MLL Leukemia

Rearrangements of the 11q23 locus account for ~70% of infant ALL and ~50% of infant AML1 and about 10% of leukemia overall. The prognosis for 11q23 patients is generally poor, however, outcomes vary depending on the fusion partner2. Rearrangements fuse the N-terminus of MLL with one of >70 different partner genes that includes both nuclear and cytoplasmic proteins. Despite the different intracellular localization of these partner proteins, to date, all studied MLL fusion proteins (MLL-FPs) localize to chromatin in the nucleus and drive aberrant transcriptional activation. Recent seminal work by a number of groups has revealed that several of the most common nuclear translocation partner genes (including AF9, ENL, AFF1 (AF4), AFF4 (AF5q31), AF10, AF17 and ELL) assemble into a transcriptional activation complex that includes p-TEFb and/or the histone H3K79 methyltransferase DOT1l. Translocation of MLL with members of this complex results in deregulated transcriptional activation of target genes. Conversely, oligomerization motifs are necessary for transformation following 11q23 translocation with a cytoplasmic partner. However, these mechanisms fail to explain different survival outcomes observed in patients. Further, the transcriptional programs induced in these diverse 11q23 leukemias are currently not well understood. In this study, we examined the genome wide expression profiles in leukemic cells transformed by several MLL-FPs representative of nuclear translocations [t(9;11) (MLL-AF9), t(10;11) (MLL-AF10) and t(11;19) (MLL-ENL)] or cytoplasmic translocations [t(1;11) (MLL-AF1p), t(6;11) (MLL-AF6) and t(11;17) (MLL-Gas7)]. Leukemia cell lines established from mouse bone marrow cells expressing these MLL-FPs proliferated at different rates and mice transplanted with the transformed cells develop leukemia with different latencies remarkably consistent with differences observed in patients harboring different MLL translocations. To elucidate differences in the gene programs induced by different MLL-FPs we performed genome wide expression profiling by RNA-sequencing. These data demonstrated that while the MLL-AF9 and MLL-ENL fusion proteins induce very similar gene programs, the cytoplasmic fusion proteins (MLL-AF6, MLL-AF1p and MLL-GAS7) all possess unique gene signatures. We then performed a pathway analysis comparing nuclear fusion proteins and cytoplasmic fusion proteins and discovered the Myc transcription factor program as one of the top distinguishing features. Myc overexpression significantly increased the growth rate of slow-growing cells that also had low intrinsic Myc, while the growth rate change of more highly proliferative cells was minimal. While all leukemic cell lines were sensitive to the BET inhibitor JQ1 which regulates c-Myc expression, greater sensitivity was observed in those with low c-Myc expression demonstrating the universal importance of this gene program. The Myc target and micro RNA binding protein Lin28B is also differentially expressed between nuclear and cytoplasmic fusions. Negative regulation of miR-150 by Lin28B was observed in all MLL-FP cell lines, which is necessarily downregulated in 11q23 leukemias. We then investigated another Lin28B microRNA target let-7. Interestingly, let-7g expression was significantly increased in MLL-FP transformed cells associated with the longest disease latency. These data demonstrate that differential activation of the c-Myc/Lin28 program accounts for changes in let-7g expression and is associated with MLL-FP disease latency. These data also suggest that patients harboring different 11q23 rearrangements will respond differentially to therapeutic targeting of c-Myc expression dependent on fusion partner.

Clustering Adult ACUTE Lymphoblastic Leukemia (ALL) Philadelphia Negative (Ph-) By Whole Exome Sequencing (WES) Analysis

Introduction: Adult ALL represents a biologically and clinically heterogeneous group. Incidence and cure rates differ among children and adults. In adults, ALL is less common and generally carries a worse prognosis with shorter long-term survival probability. Although the remarkable progress made in the treatment of ALL in children and, with less efficacy, in adults, several ALL subtypes continue to have a poor prognosis.Aims: focus our attention on adult Ph-negative ALL pts using whole exome experiments to discover novel insights into the mechanisms involved in leukemogenesis and to develop genetic models that accurately define novel ALL subtypes based on the genomic profile of individual patients (pts).Patients and Methods: we performed the WES analysis of 72 samples of B-cell precursor ALL acute lymphoblastic leukemia (B-ALL) cases using the Illumina Hiseq2000 platform. All were adult patients (18-79 years) and were negative for Philadelphia chromosome (BCR-ABL) translocation and negative for the recurrent known molecular rearrangements (E2A-PBX, TEL, AML1-MLL-AF4). Peripheral blood and/or bone marrow samples were collected from adult B-ALL at the time of diagnosis and/or at the time of relapse. Matched samples of primary tumour (peripheral blood or bone marrow) and germline DNA from buccal swab or peripheral blood at the remission time were analyzed.MuTect and GATK tools to call mutations (Single Nucleotide Variants=SNVs and/or INDELs) were used and we selected variants with a minor allele frequency (MAF) lower than 0.05 and filtered using dbSNP142.Results: The WES analysis of the 41 Ph negative cases identified 735 point mutations and 25 mutations that occur in splicing sites in 651 genes. The average number of somatic coding mutations was 17 per case (range 1-47). 38 genes were recurrently mutated with 11 genes mutated in at least 3 cases: PAX5, PRDM12, JAK2, TTN, TP53, PTPN11, PKHD1L1, CUL3, PIEZO2, TACC2, RBBP6. The first two genes present more point mutations, in 5 pts and in 4 pts respectively. Some mutations in genes like PAX5, JAK2, TP53, PTPN11 were deeply described in acute lymphoblastic leukemic; PKHD1L1 was described mutated in one case of T-cell large granular lymphocyte leukemia; PRDM12 disruption was described in an aggressive CML case;TACC2 expression in infant ALL was described as predictor of outcome andtranscription factor and RBBP6 expression was differentially expressed in leukemic cells that overexpressed Gfi-1B gene. The alterations in the remaining genes were not previously described in ALL and/or leukemia.Using KEGG database we mapped the 651 mutated genes to detect the mostly represented pathways. The following resulted significantly enriched (p=0.0004 to p=0.006): Jak-STAT signaling pathway (11 genes), Cell Cycle (13), Dilated Cardiomyophaty (9), Hypertrophic cardiomyophaty (8), Axon Guidance (9), Calcium Signaling pathways (10), Huntington's disease (10), Wnt signaling pathway (9), Metabolic pathways (30), Pacreat Secretion (7). Preliminar analysis lead considering both SNVs and INDELs, detected totally 956 gene variations. Again the pathways mainly significantly (p=8.05e-05 to p=0.0076) affected are the Jak-STAT signaling pathway (14) and the Cell Cycle (13). Also Huntington's disease (14), Dilated Cardiomyophaty (10), Hypertrophic cardiomyophaty (9), Wnt signaling pathway (12), Metabolic pathways (40 genes), Calcium Signaling pathways (12), Metabolic pathways (40), TGF-Beta signaling pathway (8), Pacreat Secretion (8) were alterated.Prediction of protein interactions, using STRING database, generated a network with the genes mutated in more than 5 patients. Then, the nodes were clustered with K-means identifying 4 groups that contain several of our analysis variations (Fig.1).Conclusions: Point mutations are the prevalent mechanism identified in our pts cohort (75.5%). INDELs are less represented (21.5%). Altogether the identified mutations may help cluster Ph- ALL pts. Analysis of SNVs confirmed mutations in important genes known to be involved in leukemogenesis. Relevant alterations affect crucial pathways as cell cycle and Jak-STAT signaling which may be effectively targeted by currently available JAK inhibitors. Supported by: ELN, AIL, AIRC, PRIN, progetto Regione-Università 2010-12 (L. Bolondi), FP7 NGS-PTL project. DisclosuresHaferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Martinelli:AMGEN: Consultancy; Ariad: Consultancy; Pfizer: Consultancy; ROCHE: Consultancy; MSD: Consultancy; BMS: Consultancy, Speakers Bureau; Novartis: Consultancy, Speakers Bureau.

First Clinical Application of Talen Engineered Universal CAR19 T Cells in B-ALL

Chimeric antigen receptor (CAR)19 T-cells exhibit powerful anti-leukemic effects in patients with B cell malignancies. However, the complexity of production of patient bespoke T cell products is a major barrier to the broader application of this approach. We are investigating a novel strategy to enable "off-the-shelf"' therapy with mismatched donor CAR19 T cells. Transcription activator-like effector nucleases (TALEN)s can be used to overcome HLA barriers by eliminating the risk of graft-versus-host disease (GvHD) through disruption of T cell receptor expression, and by simultaneously targeting CD52, cells can be rendered insensitive to the lymphodepleting agent Alemtuzumab. Administration of Alemtuzumab can then be exploited to prevent host-mediated rejection of HLA mismatched CAR19 T cells. We manufactured a bank of such cells from volunteer donor T cells under GMP conditions on behalf of Cellectis S.A for final stage validation studies using a third generation self inactivating lentiviral vector encoding a 4g7 CAR19 (CD19 scFv- 41BB- CD3ζ) linked to RQR8, an abbreviated sort/suicide gene encoding both CD34 and CD20 epitopes. Cells were then electroporated with two pairs of TALEN mRNA for multiplex targeting of both the T cell receptor alpha constant chain locus, and the CD52 gene locus. Following ex-vivo expansion, cells still expressing TCR were depleted using CliniMacs alpha/beta TCR depletion, yielding a T cell product with <1% TCR expression, 85% of which expressed CAR19, and 64% becoming CD52 negative. This universal CAR19 (UCART19) cell bank has been characterized in detail, including sterility, molecular and cytometric analyses and human/murine functional studies ahead of submissions for regulatory approvals and Phase 1 testing in trials for relapsed B cell leukaemia. In the interim we received a request for therapy on a compassionate basis for an infant with refractory relapsed B-ALL, and with the agreement of Cellectis, we treated this first patient under UK special therapy regulations.An 11 month girl with high risk CD19+infant ALL (t(11;19) rearrangement) relapsed in bone marrow 3 months after a myeloablative 8/10 mismatched unrelated donor transplant. Leukaemic blasts expressed CD19 but were CD52negative. Her disease progressed despite treatment with Blinatumomab (70% blasts in marrow) and we were unable to generate donor-derived CAR19 T cells on an existing study. Following institutional ethics review, detailed counseling, and parental consent, the patient received cytoreduction with Vincristine, Dexamethasone and Asparaginase followed by lymphodepleting conditioning with Fludarabine 90mg/m2, Cyclophosphamide 1.5g/m2 and Alemtuzumab 1mg/kg. Immediately prior to infusion of UCART19 cells, the bone marrow showed persisting disease (0.5% FISH positive). She received a single dose (4.5x106/kg) of UCART19 T cells without any significant toxicity. To date there has been no significant perturbation of cytokine levels in peripheral blood, and no indication of cytokine release syndrome. Although profoundly lymphopenic, UCART19 T cells were detectable by qPCR in the circulation by day 14 and at increased levels in both blood (VCN 0.35) and marrow (VCN 0.22) on day 28. The patient exhibited signs of count recovery and the bone marrow, while hypoplastic, was in cytogenetic and molecular remission. Chimerism was 90% donor, and a clearly demarcated population (7%) of third party cells indicated persistence of UCART19. A residual persistence of 3% recipient cells in the marrow suggests that leukemic clearance was not mediated by transplant mediated alloreactivity.Within the short period of follow up available, our intervention comprising lymphodepletion and infusion of UCART19 T cells has induced molecular remission where all other treatments had failed. This first-in-man application of TALEN engineered cells provides early proof of concept evidence for a ready-made T cell strategy that will now be tested in early phase clinical trials. DisclosuresQasim:CATAPULT: Research Funding; CELLMEDICA: Research Funding; CALIMMUNE: Research Funding; MILTENYI: Research Funding; AUTOLUS: Consultancy, Equity Ownership, Research Funding; CELLECTIS: Research Funding. Off Label Use: UCART19 T Cells are an unlicensed investigational medicinal product and in this case were used under MHRA special licence arrangements. Stafford:CELLECTIS: Research Funding. Peggs:Cellectis: Research Funding; Autolus: Consultancy, Equity Ownership. Thrasher:CATAPULT: Patents & Royalties, Research Funding; MILTENYI: Research Funding; AUTOLUS: Consultancy, Equity Ownership, Research Funding. Pule:AUTOLUS: Employment, Equity Ownership, Research Funding; CELLECTIS: Research Funding; AMGEN: Honoraria; UCLB: Patents & Royalties.