Search

Introduction: Hematopoietic stem cell (HSC) gene therapy/editing is a viable treatment option for various hematological diseases and disorders including hemoglobinopathies and HIV/AIDS. Most if not all currently available approaches target CD34-enriched cell fractions, a heterogeneous mix of mostly committed progenitor cells and only very few true HSCs with long-term multilineage engraftment potential. As a consequence, gene therapy/editing approaches are currently limited in their HSC targeting efficiency, very expensive consuming huge quantities of modifying reagents, and can lead to unwanted side-effects in non-target cells. We recently described a novel HSC-enriched CD34 subset (CD90+CD45RA-) that is exclusively responsible for rapid recovery onset, robust long-term multilineage engraftment, as well as entire reconstitution of the bone marrow stem cell compartment in the nonhuman primate (NHP) stem cell transplantation and gene therapy model (Radtke et al. 2017, STM). Most importantly, we demonstrate that this CD34 subset reduces the number of target cells, modifying reagents and costs by more than 10-fold without compromising the long-term efficiency of gene-modification in the NHP (Humbert and Radtke et al. 2019, STM). Here, we aimed to develop a clinical protocol to reliably purify and efficiently gene-modify human HSC-enriched CD90+ cell fractions. Methods: Large-scale enrichment of CD34+ cells from GCSF-mobilized leukapheresis products was initially performed on the Miltenyi CliniMACS Prodigy according to previously established protocols (Adair et al. 2017, Nat. Comm.). Yield, purity, quality, and feasibility of CD90 sorting was then comprehensively tested on two different commercially available cell sorting systems comparing the jet-in-air sorter FX500 from Sony and the cartridge-based closed-system sorter MACSQuant Tyto from Miltenyi Biotech with our clinically approved gold-standard CD34-mediated gene therapy approach. Sorted CD90+ and bulk CD34+ cells were transduced with a clinical-grade lentivirus encoding for GFP and the multilineage differentiation as well as engraftment potential tested using in vitro assays and the NSG mouse xenograft model, respectively. Results: Flow-cytometric sort-purification of CD90+ cells was similarly efficient in purity and yield using either the FX500 or Tyto (Figure A,B). Both approaches reliably reduced the overall target cell count by 10 to 15-fold without impacting the cells viability and in vitro colony-forming cell potential. Unexpectedly, the transduction efficiency of sort-purified CD90+ cells was significantly improved compared to bulk-transduced CD34+ cells and especially the CD34+CD90+ subset (Figure C). All cell fractions demonstrated robust mouse xenograft potential (Figure D). Most importantly, significantly higher levels of GFP+ expression in the peripheral blood, bone marrow, spleen and thymus were observed after transplantation of gene-modified CD90+ compared to bulk CD34+ cells in NSG mice (Figure E). Conclusion: Here, we show that sort-purification of our HSC-enriched CD34+CD90+ cell subset is technically feasible and highly reproducible in two different systems. Purification of human CD90+ cell fractions significantly increased the gene-modification efficiency of primitive human HSCs with multilineage mouse engraftment potential. These findings should have important implications for currently available as well as future HSC gene therapy and gene editing protocols. Isolation of an HSC-enriched phenotype will allow more targeted gene modification and thus likely reduce unwanted off target effects. Our approach further reduced the overall costs for gene modifying reagents, can be combined with a closed transduction system, increase the portability and ultimately make HSC gene therapy GMP-facility independent and affordable. Finally, this stem cell selection strategy may also allow efficient and effective depletion of donor T cells in the setting of allogeneic stem cell or organ transplantation. Figure: A) Purity and B) yield of CD90+ cells after sort-purification. C) Transduction efficiency of bulk-transduced CD34+CD90+ cells and sort-purified CD90+ cells. Frequency of D) human chimerism and E) GFP+ human CD45+ cells in the peripheral blood (PB), bone marrow, spleen and thymus after transplantation of gene-modified bulk CD34+ or sort-purified CD90+ cells. Figure Disclosures Kiem: CSL Behring: Consultancy; Rocket Pharma: Consultancy, Equity Ownership; Homology Medicines: Consultancy, Equity Ownership; Magenta Therapeutics: Consultancy.

Current autologous stem cell transplantation and gene therapy strategies are limited by the inability to identify a true hematopoietic stem cell (HSC) propulation that reliably predicts engraftment. The ability to enrich and target a specific HSC pool and predict engraftment levels of the gene-modified cells would be a major advance, reducing manufacturing costs and off-target effects. Here we describe a distinct HSC phenotype, conserved between humans and nonhuman primates (NHP) which overcomes this limitation. We used our NHP stem cell transplantation and gene therapy model to study the engraftment potential of phenotypically distinct hematopoietic stem and progenitor cell (HSPC) subpopulations. We were able to identify an exclusive HSPC subpopulation capable of multi-lineage engraftment (Figure 1A and 1B). This HSPC subpopulation (denoted "I") accounts for ~3-5% of the entire CD34+ cell population in primed bone marrow, reducing the number of cells targeted for cell and gene therapy approaches by 20 to 30-fold. For autologous transplants, only 300-400K HSPCs/kg body weight were required to achieve rapid neutrophil and plateletet recovery within 9-10 and 19-20 days, respectively. Stable 30-35% gene-marking was obeserved in all blood lineages including T cell, B cell, NK cells, granulocyte, monocytes/macrophages, erythrocytes and platelets. Complete reconstitution of the bone marrow compartment was achieved within Most importantly, and in support of the above described engraftment studies, the use of this HSPC population I phenotype allowed us to establish a reliable flow cytometry based analysis to assess and predict engraftment after HSC transplantion. This method predicted the onset of neutrophil and platelet recovery (Figure 1C and 1D). This assay was also able to predict the onset of NHP hematopoietic recovery for transplanted HSCs from different stem cell sources including steady state (unprimed) bone marrow, as well as gene-modified/transduced and ex vivo expanded HSCs. From this data, we were able to define the minimum cell number of HSCs/kg body weight required for sustained multi-lineage long-term engraftment with recovery of all blood cell lineages in the NHP transplant setting as 110,000 cells/kg. Importantly, we demonstrated that this cell population and phenotype is conserved between NHP and human hematopoiesis (Radtke et al. 2016, submitted). In summary, we identified a distinct and evolutionarily conserved HSC phenotype that is associated with multi-lineage engraftment in nonhuman primates and also reliably predicted the engraftment kinetics after transplant. This will greatly facilitate evaluation and "potency" of autolgous stem cell products especially after stem cell expansion or gene therapy. In addition, it will allow for an improved enrichment of the target cells for gene therapy and gene editing protocols and thus likely improve efficacy and safety. Figure 1 HSCs are exclusively driving multi-lineage engraftment and allow prediction of neutrophil and platelet engraftment in the NHP transplant model. (A) Hierarchical organization of HSPCs in the NHP. HSPC of group I, II, and III were sort-purified and transduced with LV vectors expressing either green fluorescent protein (GFP; population I), mCherry (population II) or mCerulean (population III), respectively. (B) Long-term follow-up of gene modified white blood cell (WBC) levels in vivo after myeloablative transplantation of these cell populations. (C and D) Correlation of transplanted HSCs/kg body weight with (C) neutrophil and (D) platelet recovery. Animals demonstrating engraftment failure (red squares) were excluded from this correlation. Figure 1. HSCs are exclusively driving multi-lineage engraftment and allow prediction of neutrophil and platelet engraftment in the NHP transplant model. / (A) Hierarchical organization of HSPCs in the NHP. HSPC of group I, II, and III were sort-purified and transduced with LV vectors expressing either green fluorescent protein (GFP; population I), mCherry (population II) or mCerulean (population III), respectively. (B) Long-term follow-up of gene modified white blood cell (WBC) levels in vivo after myeloablative transplantation of these cell populations. (C and D) Correlation of transplanted HSCs/kg body weight with (C) neutrophil and (D) platelet recovery. Animals demonstrating engraftment failure (red squares) were excluded from this correlation. Disclosures Adair: Rocket Pharmaceuticals: Consultancy, Equity Ownership. Kiem:Rocket Pharmaceuticals: Consultancy, Equity Ownership, Research Funding.

Pediatric ALL is the most common childhood tumor and the leading cause of childhood cancer deaths. To gain a better understanding of the landscape of somatic mutations in ALL, we performed whole exome and targeted sequencing of 240 pediatric B-ALL patients with their matched remission samples. The significantly mutated genes fall into several common categories: RAS/receptor tyrosine kinases, epigenetic regulators, transcription factors involved in lineage commitment and p53/cell cycle pathway. RAS/receptor tyrosine kinases: the most frequently mutated genes were members of RAS signaling (NRAS, KRAS, FLT3, PTPN11). Besides the well know hotspot mutations [G12D/V/C (NRAS 13 cases, KRAS 13 cases), G13D (NRAS 14 cases, KRAS 11 cases) and Q61H/L/R/K (NRAS 15 cases, KRAS 1 case)], novel mutational sites were also identified for KRAS: A146T/P (3 cases), K117N/T (4 cases) and V14I (1 case). High frequency missense mutations of PTPN11 clustered in SH2 domain (included the canonical hotspot A72T (5 cases) and E76K/V (4 cases)) and tyrosine-phosphatase catalytic domain (G503R/V). For FLT3, well-appreciated activating hotspot mutations in the kinase domain (D835Y/Y842C) and several novel recurrent mutationswere identified. Epigenetic regulators: hotspot mutations were identified in histone H3K36 methyltransferase WHSC1. Mutation E1099K located in the SET domain, was identified in 10 patients as well as two of the 5 ALL cell lines that we sequenced (RS4;11, SEM). Stable silencing of E1099K mutant WHSC1 in RS4;11 cells by either lentiviral shRNA or CRISPR guide RNA (sgRNA) markedly reduced clonogenic growth both in vitro and in vivo, underscoring the critical role of WHSC1 in lymphoid malignancies. Two highly-related histone/non-histone acetyltransferases, CREBBP and EP300, were also prominently mutated in our cohort. Mutations of CREBBP predominantly occurred in the acetyltransferase domain, particularly in the hotspot R1446C/H. Mutations of chromatin remodeling genes (ARID1A and ARID2) have been identified in a number of cases. Silencing of ARID1A in ALL cell lines by lentiviral shRNA resulted in upregulation of the pro-growth regulator c-MYC, while forced expression of ARID1A reduced c-MYC luciferase reporter activity. In addition, silencing of ARID1A by either shRNA or CRISPR-sgRNA resulted in enhanced clonogenic growth, suggesting that ARID1A may be involved in the c-MYC pathway and modulates the ALL cell proliferation. Mutations of epigenetic regulators were also found in the polycomb complex (EZH2, EED, SUZ12), chromatin/nucleosome structure modifying proteins (CHD2, CHD3, CHD4), TET family proteins [TET1 (2 cases), TET2 (5 cases)] and histone modification proteins (HDAC1, SIRT1, BCOR, BRD8, lysine demethylase PHF2/KDM6A, histone acetyltransferase KAT6B). Transcription factors and p53/cell cycle pathway: a number of alterations of transcription factors essential for hematopoietic and lymphoid differentiation were noted including the lineage regulator PAX5 (5 missense, 3 indels) and ETV6 (6 cases, 3 were frameshift indel and 1 was a splice-site mutations). In addition, mutations were also found in other lineage transcription factors (IKZF2, IKZF3, EBF1), WT1 (6 cases, including 3 indels and 1 stop-gain mutations), RUNX family member [RUNX2 (7 cases), RUNX1 (1 case)], ERG1 (3 cases), GATA1/3 (1 case each) and CTCF. Somatic mutations of genes involved in the p53 pathway occurred in 18 patients, including TP53, ATM and the kinases that regulate p53 activities (HIPK1, HIPK2). Germline TP53 pathogenic variants were found in these 2 patients. Taken together, we extensively interrogated the mutational landscape of a large cohort of pediatric ALL samples by exome and targeted resequencing. This study provides a detailed mutational portrait of pediatric ALL and gives new insights into the molecular pathogenesis of this disease. Disclosures Kantarjian: Amgen: Research Funding; ARIAD: Research Funding; Bristol-Myers Squibb: Research Funding; Pfizer Inc: Research Funding; Delta-Fly Pharma: Research Funding; Novartis: Research Funding. Ogawa:Sumitomo Dainippon Pharma: Research Funding; Kan research institute: Consultancy, Research Funding; Takeda Pharmaceuticals: Consultancy, Research Funding.

For more than 30 years, hematopoietic stem cell (HSC) research has been performed according to the classical model of human hematopoiesis suggesting an early segregation of lymphoid and erythro-myeloid potentials. However, several studies have recently proposed a great variety of models for the human blood hierarchy showing alternative lineage relationships and read-outs for multipotent HSCs. While the debate about hematopoietic lineage relationships is still ongoing, consequences of these findings and the challenges they pose for the development of treatment strategies for hematological diseases and malignancies are rarely discussed. A critical factor for the development of stem cell therapies is the availability of a reliable and robust read-out for multipotent HSCs/MPPs (multipotent progenitor cells) supporting the development of all blood cell lineages. Unfortunately, the current gold standard, NOD/SCID mouse xenograft repopulation assay does not support erythrocyte and megakaryocyte development and, thus, does not fully read-out multi-lineage potential of human HSCs/MPP. In addition, development of therapeutic approaches in the mouse model is not possible due to differences in cell surface marker expression, physiology, life span, and the demand on stem cell self-renewal and differentiation compared to humans. The pigtail macaque (PM; Macaca nemestrina) and the rhesus macaque (RM: Macaca mulatta) share a close evolutionary relationship with humans and have been used as a pre-clinical model system to study basic HSC biology or to develop specific HSC gene therapy approaches. Surprisingly, however, a comparison of hematopoietic subpopulations and the hierarchical organization of defined lineages has not been performed between NHPs and humans. This will be a critical factor for a better understanding of the newly defined blood lineage associations and hierarchies, as well as the development of treatment approaches based on these lineages. Here, we comprehensively analyzed all known markers of human hematopoiesis in the NHP to identify subpopulations of candidate NHP hematopoietic stem and progenitor cells (HSPCs) and then validated HSPC phenotypes of these fractions with functional in vitro read outs. We further evaluated and compared lineage relationships between these subpopulations to recently proposed models of human hematopoiesis to determine whether conservation of hematopoiesis exists with the goal of informing studies evaluating treatments for hematological diseases in the NHP model. We show for the first time a phenotypic mapping strategy in NHP hematopoietic cells predicting a revised model of hematopoiesis. Similar to humans, NHP HSCs give rise to multipotent progenitors (MPPs), followed by a segregation of lympho-myeloid, erythro-myeloid, and megakaryocytic lineages (see figure). Conservation of hematopoietic lineage relationships was confirmed by RNA expression analysis of corresponding subpopulations. In summary, we identified corresponding human and NHP hematopoietic subpopulations, which share phenotypical, functional and transcriptional properties in both species, validating the NHP as an excellent pre-clinical model system for HSC biology and the development of novel HSC-based treatment approaches. Figure Figure. Disclosures Adair: Rocket Pharmaceuticals: Consultancy, Equity Ownership. Kiem:Rocket Pharmaceuticals: Consultancy, Equity Ownership, Research Funding.

Abstract 448 The c-myb proto-oncogene was first identified as the cellular homologue of the v-myb oncogene carried by the avian leukemia viruses AMV, and E26. c-myb encodes a transcription factor, c-Myb, that is highly expressed in immature hematopoietic cells. In such primitive cells, c-Myb has been found to exert an important role in lineage fate selection, cell cycle progression, and differentiation of both myeloid, B, and T lymphoid progenitor cells. c-Myb is also highly expressed in many leukemia cells and on this basis has been implicated in leukemic transformation. Despite intensive study, a mechanisms based understanding for c-Myb's myriad effects on blood cell development has yet to be fully achieved though c-Myb's ability to interact with a variety of transcriptionally active co-factors, such as p300, CBP, and FLASH, as well as to modulate its own expression, have all been reported to contribute to its activities. Therefore, we undertook a series of biochemical, molecular, and clinical studies to further address c-Myb's role in leukemic hematopoiesis. Using in vitro translated proteins and nuclear extracts from leukemic cells in immunoprecipitation (IP) assays, we found that c-Myb is associated with MLL1, the SET1 proteins WDR5, RbBp5, and Ash2L, and menin, all of which form a complex with histone methyltransferase (HMT) activity. c-Myb associated with the MLL1 and SET1 proteins through menin, which served as an adapter protein by interacting (as previously shown) with the extreme amino terminus of the MLL1 protein, and, as we show, with a region around the c-Myb transactivation domain (aa 194-325). We demonstrated in vitro with purified proteins and an H3 peptide, that c-Myb contributed to the HMT activity of the MLL1 complex. In leukemia patients being treated with a c-myb targeted antisense oligodeoxynucleotide (ASODN), and in leukemic cell lines, silencing c-myb evoked a significant decrease in H3K4 methylation demonstrating biological relevance of this observation. The decrease in H3K4 methylation is the direct result of silencing c-myb and is not due to changes in cell proliferation, and could not be reproduced by silencing B-myb. Also, we confirmed that c-Myb is a downstream target of HoxA9, and Meis 1, but showed unexpectedly that leukemic blasts derived from the c-myb ASODN treated patients, and c-myb siRNA treated cell lines, decrease c-myb expression also led to a decrease in Hoxa9 and Meis1 expression. This suggested the presence of an autoregulatory feedback loop between c-Myb and HoxA9. This finding too was specific for c-myb and not associated with a block in proliferation or silencing B-myb. Finally, disrupting the c-Myb-MLL1 interaction impairs localization of MLL1 and menin on the Hoxa9 gene promoter, as well as the MLL-ENL induced transformation of normal murine bone marrow cells. In summary, our results bring new insights regarding c-Myb function in human hematopoietic cells, suggest new mechanisms whereby c-Myb may contribute to cell transformation, and suggest new therapeutic targets for the treatment of acute leukemia. Disclosures: No relevant conflicts of interest to declare.