Search

Key Points Patients with AML who received IC were compared with those who received ven to investigate predictors of outcomes.Presence of RUNX1 mutations is associated with better outcomes for ven/aza compared with IC., Visual Abstract, Venetoclax (ven) plus azacitidine (aza) is the standard of care for patients with newly diagnosed acute myeloid leukemia (AML) who are not candidates for intensive chemotherapy (IC). Some patients who are IC candidates instead receive ven/aza. We retrospectively analyzed patients with newly diagnosed AML who received ven/aza (n = 143) or IC (n = 149) to compare outcomes, seek variables that could predict response to 1 therapy or the other, and ascertain whether treatment recommendations could be refined. The response rates were 76.9% for ven/aza and 70.5% for IC. The median overall survival (OS) was 884 days for IC compared with 483 days for ven/aza (P = .0020). A propensity-matched cohort was used to compare outcomes in the setting of equivalent baseline variables, and when matched for age, biological risk, and transplantation, the median OS was 705 days for IC compared with not reached for ven/aza (P = .0667). Variables that favored response to ven/aza over IC included older age, secondary AML, and RUNX1 mutations. AML M5 favored response to IC over ven/aza. In the propensity-matched cohort analyzing OS, older age, adverse risk, and RUNX1 mutations favored ven/aza over IC, whereas intermediate risk favored IC over ven/aza. In conclusion, patients receiving IC have improved OS compared with those receiving ven/aza. However, in a propensity-matched cohort of patients with equivalent baseline factors, there was a trend toward favorable OS for ven/aza. Specific variables, such as RUNX1 mutations, reported here for the first time, can be identified that favor ven/aza or IC, helping to guide treatment decisions for patients who may be eligible candidates for either therapy.

Venetoclax is approved for older untreated acute myeloid leukemia (AML) patients. Venetoclax was available prior to approval off-label. We assessed our single-institution off-label experience with venetoclax/azacitidine, comparing outcomes with a clinical trial cohort that administered this regimen at the same institution. Thirty-three untreated AML patients unfit or unwilling to receive induction chemotherapy and prescribed venetoclax/azacitidine off-trial were retrospectively analyzed and compared with 33 patients who received the same therapy on trial. Outcomes were compared, and comparisons were made to a theoretical scenario in which off-trial patients received induction. Digital droplet polymerase chain reaction evaluated measurable residual disease (MRD). Off-trial venetoclax was attainable in nearly all patients for whom this was desired. The complete remission (CR)/CR with incomplete blood count recovery rate was 63.3% for off-trial patients who received treatment and 84.9% for trial patients (P = .081). The median overall survival for off-trial patients who received treatment was 381 days (95% confidence interval [CI], 174, not reached) vs 880 days (95% CI, 384, not reached) for trial patients (P = .041). Prior exposure to hypomethylating agents was associated with worse outcomes. Response rates with venetoclax/azacitidine were not inferior to a theoretical scenario in which patients received induction, and early death rates were less than expected with induction. MRD negativity was achievable. Newly diagnosed AML patients treated in a “real-world” scenario with off-trial venetoclax/azacitidine had inferior outcomes compared with patients treated in the setting of a clinical trial. Additionally, this therapy may be as effective, and less toxic, when compared with induction chemotherapy.

Background: The mechanisms that cause the progression of myelodysplastic syndrome (MDS) are poorly understood. Little is known about major signaling networks and energy metabolism in MDS cells as patients progress from low risk (LR) to high risk (HR) disease and from high risk to secondary acute myelogenous leukemia (sAML). As many as 30% of HR MDS patients progress to sAML and a portion of LR MDS patients progress to HR. The goal of this project is preventing progression by identifying MDS-specific targets for therapy. A deeper understanding of the metabolic properties of leukemia stem cells (LSCs) in AML has shown these cells are uniquely vulnerable to venetoclax and azacitidine (Ven/Aza) (Pollyea et al, Nat. Medicine, 2018) and metabolic changes cause resistance to Ven/Aza (Stevens et al, Nat. Cancer, 2021). Little however is known about the contribution of metabolism to the pathogenesis of MDS. The contributing factors to progression including metabolic properties, transcriptional programs, and immunophenotype are examined in this study. Methods: Bone marrow specimens from MDS patients at various disease stages, including serial samples during progression, were obtained. Single cell techniques including mass cytometry, antibody based single cell RNA sequencing (CITE-Seq) and transcriptional profiling with RNA sequencing were used to elucidate novel mechanisms of progression. Selective targeting of primitive MDS cells was tested using several agents. Results: Our previous work characterizing MDS stem cells (MDSC) showed significant similarities between MDSCs and AML LSCs (Stevens et al, Nat. Communications, 2018). However, little is known about lower risk disease. In order to understand transcriptional changes and their relationship to metabolism across pathogenesis, the transcriptome of blasts from patients with LR, intermediate (INT), or HR IPSS scores was investigated. The first major transcriptional difference identified was enrichment of glycolysis pathway at LR and INT stage. In contrast, HR MDS demonstrated enrichment of oxidative phosphorylation. Furthermore, comparison of intermediate to HR MDS showed increased RNA polymerase and Ribosome pathways at the HR stage. These changes demonstrate the progressive alteration of metabolic properties during MDS pathogenesis with cells first relying on mechanisms associated with normal stem cells (i.e. glycolysis) and later transitioning to a state associated with AML stem cells (i.e. reliance on oxidative phosphorylation). Using serial specimens of patients of who progressed from LR to HR MDS we performed CITE-Seq and mass cytometry. CITE-seq in serial specimens showed up-regulation of protein translation and oxidative phosphorylation in a subset of MDS stem and progenitor cells (CD34+ at transcript and antibody level) present at LR stage and conserved at HR stage (Fig 1A-C). MDSCs also acquired surface antigens including CD99 and CD52 upon progression from LR to HR. Analysis of the mass cytometry data showed significant overlap with CITE-Seq data including increased CD123+ and MCL1 expression in MDS stem cells upon progression. In order to understand therapeutic vulnerabilities as they relate to progression, we investigated ex vivo drug response in LR and HR specimens. MDS samples were challenged with two regimens, Ven/Aza, a regimen known to inhibit OXPHOS; and omacetaxine and azacitidine (Oma/Aza), which inhibits translation. CITE-seq showed that MDSC were selectively sensitive to these agents (Fig 1D). Importantly, addition of either drug regimen caused ablation of MDSC at LR and HR stages and these changes were most profound in cells with LSC properties. Based on preclinical findings, we are investigating MDS patients treated with Ven/Aza or Oma/Aza via CITE-seq and metabolomics for correlation of clinical response with properties of MDSC. Preliminary studies show that patients that respond to Oma/Aza present with a population of MDSC with transcriptional signatures of protein translation and LSCs (Fig. 1E). Studies are underway to investigate overlapping properties of ven/aza resistance in AML to resistance in MDS specifically investigating fatty acid metabolism in MDSC. Conclusions: Analysis of MDS patient bone marrow reveals acquisition of aberrant metabolic properties at both low and high risk stages of disease. These distinct aspects of MDSC biology create unique and targetable features. Figure 1 Figure 1. Disclosures Pollyea: Genentech: Consultancy; Novartis: Consultancy; Pfizer: Consultancy; Janssen: Consultancy; Karyopharm: Consultancy; Syndax: Consultancy; Takeda: Consultancy; Daiichi Sankyo: Consultancy; Celgene/BMS: Consultancy; Amgen: Consultancy; AbbVie: Consultancy, Research Funding; Agios: Consultancy; Glycomimetics: Other.

Current treatment options for relapsed acute myeloid leukemia (AML) are limited and ineffective for the majority of patients. In AML, primitive leukemia stem cells (LSCs) and pre-leukemic populations are able to maintain the disease and drive relapse. Thus, therapies targeting LSC populations may increase the overall survival of AML patients. In this study, we aim to identify the drivers favoring LSC expansion following treatment and relapse and develop potential therapies for AML. The transcriptome analyses of 12 pairs of functionally defined LSC fractions at diagnosis and relapse revealed significant changes of IL-1 signaling in AML patients. We demonstrated that the protein expression levels of interleukin-1 receptor type I (IL1R1) and its complex member interleukin-1 receptor accessory protein (IL1RAP) were both up-regulated in human leukemia stem and progenitor cells (LSPCs) at diagnosis or in relapse compared to normal hematopoietic stem and progenitor cells (HSPCs). Knockdown of IL1R1 and IL1RAP suppressed the clonogenicity and engraftment growth of primary human AML cells but showed low impacts on HSPCs in the normal bone marrow. Additionally, knockout of IL1R1 in leukemia MLL-AF9 mice significantly reduced the LSC frequency and prolonged the overall survival rate. To target IL-1/TLR signaling in LSCs, we performed iterative structure-activity relationship (SAR) guided medicinal chemistry, in silico modeling and leukemia cell line reporter assays to screen and identify a novel interleukin-1 receptor-associated kinase 1/4 (IRAK1/4) inhibitor (termed UR241-2). UR241-2 robustly inhibits IL-1/TLR signaling in AML cells including the activation of NF-κB following IL-1 stimulation. UR241-2 repressed LSPC function as assessed by colony-forming unit assays in primary human AML cells at diagnosis and in relapse while minimally impacting normal HSPC function. Taken together, our findings demonstrate the important role of IL-1/TLR signaling in supporting AML LSC expansion following treatment and relapse and suggest that targeting IL-1/TLR signaling using the novel IRAK1/4 inhibitor, UR241-2, can target LSC function to improve patient outcomes in AML. Disclosures No relevant conflicts of interest to declare.

Background: Venetoclax-based therapy regimens are now FDA-approved for treatment of acute myeloid leukemia (AML) in older patients or those unfit to tolerate intensive chemotherapy (IC). Remission rates are high at 60-70% but relapses do frequently occur. Outcomes for newly-diagnosed patients who receive venetoclax-based therapies and proceed to a potentially curative allogeneic stem cell transplant (SCT) have largely been unreported. In the current study we compare outcomes of patients who received SCT following either IC or venetoclax + azacitidine (ven/aza) at the University of Colorado Hospital. Methods: Patients 18 years or older who received SCT in first remission of AML between 2010-2020 were included in the analysis. Patients were stratified into the IC arm if they initially received a backbone of cytarabine and an anthracycline; some patients in this cohort received IC in combination with other targeted agents. Patients who received ven/aza as first-line therapy followed by SCT were grouped in a separate cohort. Demographic and clinical information - including flow cytometry-based (MCF) MRD - was extracted from the electronic medical record. Comparisons of demographic and clinical variables between IC and ven/aza groups were made with t-test, Chi-squared, or Fisher's exact test depending on the nature of the variable. Relapse-free (RFS) and overall (OS) survival were calculated from the day of SCT to the respective endpoint or last documented follow-up using log-rank statistics. Finally, a Cox proportional hazards model was used to assess the interplay between variables pre- and post-SCT. P-values Results: We identified 179 patients who received SCT for AML in first remission. Of these patients, 151 received IC and 28 received ven/aza prior to SCT. Patients in the ven/aza group had higher median age than those in the IC group, as well as a higher proportion with adverse ELN genetic risk scoring. Patients in the ven/aza group received less intensive conditioning regimens. Sex, rates of MCF MRD negativity pre-BMT, incidence of severe acute or chronic GVHD, and causes of death were not significantly different between the two groups. There was no difference between the two groups in post-transplant RFS or OS (Figure 1). In a multivariate Cox model of pre-transplant variables predicting OS, the only factor that achieved significance was pre-SCT MCF MRD; the induction regimen was not a multivariate factor. Negative MCF MRD going into SCT was associated with decreased likelihood of relapse, GVHD, and death, respectively. Conclusions: In our cohort of AML patients receiving SCT, we found that ven/aza as a pre-transplant therapy yielded equivalent post-transplant outcomes compared to IC, in a population of older age and with higher ELN genetic risk. MCF MRD pre-SCT was confirmed as a key prognostic factor for post-SCT outcome. These findings support ongoing use of ven/aza as a first line therapy for elderly patients with AML as well as its exploration as a candidate therapy for younger patients. Figure 1 Figure 1. Disclosures Pollyea: Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other; Aprea: Honoraria, Membership on an entity's Board of Directors or advisory committees; Kiadis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Syndax: Honoraria, Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria; Agios: Other, Research Funding; Karyopharm: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Teva: Research Funding; AbbVie: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Curis, Servier: Other; Pfizer: Research Funding; Syros: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Astellas: Honoraria, Membership on an entity's Board of Directors or advisory committees; Bristol Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Honoraria, Other: advisory board; Foghorn: Honoraria, Membership on an entity's Board of Directors or advisory committees; Genentech: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy, Honoraria, Other: advisory board; Jazz: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: advisory board.

Background Myelodysplastic syndrome (MDS) arises from and is perpetuated by malignant stem cells. Curative therapy requires eradication of this population, but this is not achieved with currently available medical therapies. MDS stem cells possess unique molecular properties that can be therapeutically exploited, but in general these strategies have not been explored clinically. Our recent in vitro studies showed that combination treatment with the protein synthesis inhibitor omacetaxine (oma) and the hypomethylating agent (HMA) azacitidine (aza) efficiently targeted MDS stem cells with minimal impact on normal stem cells. Based on these findings we conducted a Phase 1 clinical trial for MDS patients with concomitant oma + aza (NCT03564873). Methods Patients were eligible if they had newly diagnosed MDS with ≥5% blasts. Aza was given at 75 mg/m 2 IV days 1-7 of a 28-day cycle; oma was given subcutaneously twice daily on days 1-7 of a 28-day cycle with 4 dose cohorts (0.75, 1.0 and 1.25 mg/m 2, and a de-escalation cohort of 0.5 mg/m 2). Dose escalation was based on a 3+3 design. Bone marrow biopsies occurred at baseline and day 8 and 28 of cycle 1, and at the conclusion of odd-numbered cycles. The primary endpoint was the maximum tolerated dose (MTD) of oma with aza. Responses were assessed according to 2006 IWG criteria with hematologic improvement (HI). Using pre- and post-treatment patient samples, advanced single cell techniques including mass cytometry and antibody-based single-cell RNA sequencing (CITE-Seq) were used to elucidate mechanisms of sensitivity and resistance. High sensitivity DNA sequencing techniques were used to understand clonal evolution and detect measurable residual disease (MRD). Results Nine patients were required to complete phase 1. See Table 1 for baseline characteristics. Two patients in cohort 1 experienced dose limiting toxicity (DLT) (grade 3 hypoxia and grade 3 respiratory failure). De-escalation to cohort 0 resulted in one DLT event (grade 4 GI bleed) among the six patients treated in this cohort. The MTD was established as 0.5 mg/m 2 oma days 1-7. Adverse events are detailed in Table 2. Median number of cycles was 2 (1-3). 8/9 patients achieved a marrow CR; 5 had HI. 6 patients proceeded to transplantation and none have relapsed with median time from transplant of 772 (183-1158) days. Median response duration is 587 days (29-938). 5 remain alive. MRD negativity, as measured with droplet digital PCR (sensitivity up to 0.1%), was achieved in 2 out of 4 patients measured. Both MRD negative patients had complete clearance of mutated splice factor genes; we previously showed no ability to clear these mutations (0/13 patients) using venetoclax+aza in AML (Nature Medicine (2018) 24:1859-1866). Baseline analysis of patient specimens by single cell transcriptomics demonstrates a distinct population of MDS stem cells (Figure 1A). This subset has overlapping transcriptional properties with known AML stem cell profiles as highlighted by increased geneset expression. Within the stem cell compartment, we observed up-regulation of protein synthesis pathways (Figure 1B) (e.g. KEGG ribosome pathway in MDS stem cells), indicating that oma/aza treatment may specifically target the malignant stem cell population. Subsequent studies are investigating the use of CITEseq as a means to evaluate drug response, which will provide very high resolution analyses of transcriptional signatures in MDS patients. By specifically monitoring eradication of MDS stem cells during the course of treatment, it should be possible to predict therapeutic efficacy and response duration while elucidating genes that correspond to oma/aza sensitivity and resistance. Conclusions Phase 2 of this study is ongoing; 23 newly-diagnosed high risk MDS patients will be enrolled with a primary endpoint to determine the overall response rate. In addition a ten-patient HMA failure expansion cohort will be enrolled to determine preliminary toxicity and efficacy assessments. Correlative studies will seek to determine the impact of this regimen on the MDS stem cell population and the mechanism of this impact. Figure 1 Figure 1. Disclosures McMahon: Takeda: Membership on an entity's Board of Directors or advisory committees. Smith: Syros: Research Funding; Kura: Research Funding; Argenx: Research Funding. Pollyea: Foghorn: Honoraria; Gilead: Consultancy, Honoraria; Celgene: Honoraria; Bristol Myers Squibb: Honoraria; Genentech: Consultancy, Honoraria; Jazz: Honoraria; AbbVie: Consultancy, Honoraria, Research Funding; Amgen: Honoraria; Astellas: Honoraria; Aprea: Honoraria; Karyopharm: Consultancy, Honoraria; Kiadis: Honoraria; Novartis: Consultancy, Honoraria; Syndax: Honoraria; Syros: Consultancy, Honoraria; Takeda: Honoraria; Teva: Research Funding. OffLabel Disclosure: Omacetaxine for MDS

Chronic myeloid leukemia (CML) is a heterogeneous disease, initiated by reciprocal translocation of chromosome 9 and 22, resulting in the generation of a BCR-ABL fusion protein and constitutive activation of the ABL kinase. ABL tyrosine kinase inhibitors (TKIs) have been very successful in suppressing CML disease. However, TKIs may not eliminate leukemia stem cells (LSCs), as evidenced by the frequent re-emergence of the disease upon TKI discontinuation. Moreover, blast phase CML (bpCML) remains a formidable challenge in disease management. Recent clinical evidence suggests that the BCL2 inhibitor venetoclax (Ven) in combination with ABL-targeting tyrosine kinase inhibitors (TKI) can eradicate bpCML LSCs. However, the exact mechanism by which this combination may targets LSCs is not known. In this report, we confirm the efficacy and LSC-targeting capacity of Ven/TKI combination in preclinical models of bpCML and we further identify that inhibition of free fatty acid (FFA) mobilization pathways may provide enhanced efficacy against LSCs. To establish the efficacy of Ven/TKI combination, we treated bpCML samples with Ven+Dasatinib (Das) combination for 24h, this resulted in a significant decrease in the viability of bulk and primitive populations (CD34+, CD38+). Patient-derived xenografts of bpCML samples in NSGS-mice, were treated with Ven/Das as well as single agents. The result showed a significant decrease in leukemia burden in the combination treated group, compared to either drug alone, albeit, some resistant cells survived in the combo treated group. Furthermore, using a syngeneic mouse model of bpCML, co-expressing Bcr-Abl and Nup98-HOXA9 translocations, the mouse leukemic cells treated with Ven/Dasatinib combination demonstrated a significant loss of viability of the bulk as well as phenotypically defined LSCs (Lin-/Sca1+). Treatment of leukemic mice with Ven/Das had a significant survival benefit and remained disease free at 80 days post treatment. We also showed significant survival benefits of Ven/ponatinib in NSGS-mice harboring syngeneic bpCML cells with the T315I gatekeeper mutation. Treatment of normal mice with Ven/Das combo did not affect the colony forming ability of LSK cells from the bone marrow, indicating a leukemia-specific response. Based on these results, we conclude that Ven/TKI combination effects were due to direct targeting of the LSC population. To investigate the potential mechanism of Ven/TKI activity in LSC targeting, we performed gene expression studies using RNA-seq based methods after short term treatment. Our findings indicated that the LSC population from Ven/TKI-treated mice showed enrichment of a gene signature associated with lysosome biology. Pre-treatment of mouse leukemia cells with bafilomycin, an inhibitor of lysosome function, resulted in increased sensitivity to Ven/TKI combo. Intriguingly, we also found significant induction of lysosomal acid lipase (LIPA), an enzyme involved in the generation of free fatty acids for energy needs. Metabolomic analysis of LSCs isolated after short term treatment with Ven/TKI, showed that a number of fatty acids were up-regulated in the Ven/Das treated group compared to control. Knocking down Lipa using CRISPR technology resulted in increased sensitivity to Ven/TKI combination, whereas overexpression of Lipa resulted in decreased sensitivity to the Ven/TKI combination, implicating Lipa upregulation and a resultant increase in free fatty acids as a protective response to Ven/TKI treatment. Furthermore, knocking down CPT1A, an important free fatty acid mitochondrial transporter, resulted in increased sensitivity to Ven/TKI combination both in mouse and primary human leukemic cells, leading to the hypothesis that activation of fatty acid processing through enhanced Lipa activity may represent a compensatory response to venetoclax based therapies in bpCML. In summary, we demonstrate the preclinical efficacy of Ven/TKI combination therapies for targeting of bpCML LSCs. Furthermore, our data suggest that blocking upregulation of free fatty acids through mechanisms such as inhibition of LIPA activity, might synergize with Ven/TKI combinations to eradicate LSCs, allowing for more durable response. Our findings provide a therapeutic rationale for blocking pathways involved in free fatty acids generation, as a potential strategy for increasing remission duration. Disclosures Pollyea: Amgen: Consultancy; Janssen: Consultancy; Genentech: Consultancy; AbbVie: Consultancy, Research Funding; Syndax: Consultancy; Daiichi Sankyo: Consultancy; Takeda: Consultancy; Pfizer: Consultancy; Celgene/BMS: Consultancy; Agios: Consultancy; Karyopharm: Consultancy; Novartis: Consultancy; Glycomimetics: Other. Smith: Syros: Research Funding; Kura: Research Funding; Argenx: Research Funding.

Background: Three CD19 directed CAR-T products have gained FDA approval for systemic large B-cell lymphoma. Due to heightened concerns of immune cell associated neurotoxicity syndrome (ICANS), patients with primary CNS lymphoma (PCNSL) were excluded from all pivotal CAR-T studies. Consequently, all three products carry a limitation of use in the PCNSL patient population per their FDA labels . Due to these exclusions, little is known about the treatment-related toxicities and therapeutic potential of the currently available CD19 directed CAR-T products in this challenging patient population with significant unmet need. Methods: Based on our prior experience of tisagenlecleucel in secondary CNS lymphoma (PMID: 31320380) we conducted a pilot study with expansion of tisagenlecleucel in adults with relapsed or refractory PCNSL (NCT04134117). Patients had to be 18 years of age or older and have had progression or relapse following methotrexate-based therapy. All patients needed a confirmed diagnosis of PCNSL without evidence of systemic disease. Patients received a standard regimen of fludarabine (25 mg/m 2) and cyclophosphamide (250 mg/m 2) daily on days -5, -4, and -3 of infusion and a dose of 0.6-6.0 x 10 8 tisagenlecleucel CAR+ T-cells. Patients who had progressed on prior BTKi were allowed to continue given its beneficial effect on CAR-T expansion and function with cessation by month 3. The primary endpoint of this study was tolerability and toxicity including the rate and grade of CRS and ICANS per the 2019 ASTCT (American Society of Transplantation and Cellular Therapy) consensus criteria. Secondary endpoints included overall response rate and complete response rate to tisagenlecleucel per the international PCNSL Collaborative Group (IPCG) criteria which included MRI and CSF assessments. Exploratory endpoints included long-term efficacy, expansion, persistence and phenotype of tisagenlecleucel, cytokine profiling of the blood and CSF and CNS trafficking of CAR-T cells. Results: As of April 1, 2021, 10 subjects (ages 35-70 years) were enrolled and 9 were infused with a median age of 67 years (range, 34-81). Of the 9 infused patients, the median time from leukapheresis to infusion was 30 days (range, 27-37). Patients were heavily pretreated prior to study enrollment and received a median of 4 prior lines of anti-neoplastic therapy. All patients had progressed or failed first line high-dose methotrexate (HD-MTX); two had a history of prior thiotepa based autologous stem cell transplant (ASCT). Eight out of 9 patients had progressed following a prior BTKi and/or an immunomodulatory drug (IMiD) as part of TEDDI-R (temozolomide, etoposide, doxil, dexamethasone, ibrutinib and rituximab, n = 3), ViPOR (venetoclax, ibrutinib, prednisone, obinutuzumab and Revlimid, n = 3), or as monotherapy (n = 5) and 2 patients had received prior stereotactic radiotherapy. All patients had measurable disease at time of lymphodepletion (pre-infusion). Grade 1 CRS was observed in 6 patients with a median onset of 4 days (range, 1-5) following tisagenlecleucel and no patients required intervention for CRS. ICANS developed in 5 out of the 9 patients, only a single case of grade 3 ICANS, with a median time of onset was 5 days (range, 3-11). With a median follow-up of 7.43 months for survivors, 6/9 patients were alive with 4/9 showing ongoing responses (Figure 1). Expansion of tisagenlecleucel was demonstrated in the peripheral blood and CSF. Nanostring and RNA pathway analysis of CSF infiltrates demonstrated higher degrees of CNS CAR-T penetration in responding patients and increased T-cell and macrophage gene signatures. Peripheral and CSF cytokines were assessed. Conclusion: Tisagenlecleucel in r/r PCNSL was safe and efficacious in a highly refractory group of patients with significant unmet need. The majority of patients demonstrated a response per IPCG, including responses beyond 12 months. Tisagenlecleucel was found to expand in the peripheral blood and CNS with CSF gene signatures suggestive of higher CAR-T cell infiltrates in responding patients. Full trial safety data as well as additional follow-up and correlative studies will be presented. Figure 1 Figure 1. Disclosures Frigault: Editas: Consultancy; Iovance: Consultancy; Arcellx: Consultancy; Kite: Consultancy, Research Funding; Takeda: Consultancy; Novartis: Consultancy, Research Funding; BMS: Consultancy. Dietrich: Unum: Consultancy; Blue Earth Diagnostics: Consultancy; Magnolia: Consultancy; Gamaka Bio: Consultancy; Beacon Biosignals: Research Funding; Boehringer Ingelheim: Research Funding; BMS: Research Funding; Medimmune: Research Funding; Acerta: Research Funding; Orbus: Research Funding. Jordan: CereXis: Consultancy; Recursion: Consultancy; Navio Theragnostics: Consultancy. Forst: Eli Lilly: Current holder of individual stocks in a privately-held company. Plotkin: AstraZeneca: Consultancy; Akuous: Consultancy; NFlection Therapeutics: Other: Co-founder; NF2 Therapeutics: Other: Co-founder. Spitzer: Qihan Bio: Consultancy; Bluebird Bio: Consultancy; Jazz Pharmaceuticals: Consultancy; Syneos Health: Consultancy. Defilipp: Omeros, Corp.: Consultancy; Incyte Corp.: Research Funding; Regimmune Corp.: Research Funding; Syndax Pharmaceuticals, Inc: Consultancy. Maus: Tmunity: Consultancy; Novartis: Consultancy; Micromedicine: Consultancy, Current holder of stock options in a privately-held company; Kite Pharma: Consultancy, Research Funding; GSK: Consultancy; Intellia: Consultancy; In8bio (SAB): Consultancy; CRISPR therapeutics: Consultancy; Cabaletta Bio (SAB): Consultancy; BMS: Consultancy; Bayer: Consultancy; Atara: Consultancy; AstraZeneca: Consultancy; Astellas: Consultancy; Arcellx: Consultancy; Agenus: Consultancy; Adaptimmune: Consultancy; tcr2: Consultancy, Divested equity in a private or publicly-traded company in the past 24 months; century: Current equity holder in publicly-traded company; ichnos biosciences: Consultancy, Current holder of stock options in a privately-held company; Torque: Consultancy, Current holder of stock options in a privately-held company; WindMIL: Consultancy. Chen: Gamida: Consultancy; Incyte: Consultancy.

For more than 50 years, investigators have considered a malignant stem cell as the potential origin of and a key therapeutic target for acute myeloid leukemia (AML) and other forms of cancer.1-4 The nature and existence of tumor-initiating cells for leukemia and other malignancies have long been the subject of intense and rigorous study; indeed, the promise of the potential to eradicate such cells is clear. However, until recently, deficiencies in our understanding of the nature of these cell populations, coupled with a limited ability to therapeutically exploit their weaknesses, have been limiting factors in realizing the goal of targeting leukemic stem cells (LSCs). Exciting new insights into the fundamental underpinnings of LSCs are now being made in an era in which drug development pipelines offer the potential to specifically target pathways of significance. Therefore, the focus in this new era, characterized by the confluence of understanding LSCs and the ability to target them, is shifting from “if it can be done” to “how it will be done.” Moving from a theoretical stage to this hopeful era of possibilities, new challenges expectedly arise, and our focus now must shift to determining the best strategy by which to target LSCs, with their well-documented heterogeneity and readily evident intra- and interpatient variability. The purpose of this review is therefore both to summarize the key scientific findings pertinent to AML LSC targeting and to consider methods of clinical evaluation that will be most effective for identifying successful LSC-directed therapies.

Relapse is the most common cause of mortality in patients undergoing allogeneic stem cell transplantation (ASCT) for acute myeloid leukemia (AML). The presence of measurable residual disease (MRD) at the time of transplant is associated with very high relapse rates, and novel strategies to address relapse risk are needed for these patients. Venetoclax has demonstrated single agent activity in AML, and venetoclax combined with azacitidine, decitabine, or low dose cytarabine is FDA approved to treat newly diagnosed patients >75yrs or older and those unfit for induction. Given the ability of venetoclax to target leukemic stem cells combined with its favorable toxicity profile and ease of administration, we have initiated off label post-transplant venetoclax maintenance for AML patients with MRD at the time of transplant. We report outcomes of our experience to date. Venetoclax therapy is planned to be initiated approximately 40-80 days post-ASCT upon count recovery (ANC>1000 109/L and platelets>100 109/L) and recovery from early ASCT toxicity. Dose is initiated at 100 mg daily for one week and titrated up by 100 mg weekly to a final maintenance dose of 400 mg daily. Therapy is planned until one year post-ASCT. Since February 2019, we have administered venetoclax to 23 post-ASCT patients (22 AML (6 with prior diagnosis of MDS) and 1 MDS), median age 65 (range 19-73). Venetoclax therapy was initiated beginning median 67 days (range 36-146) after ASCT. 16 patients underwent transplant following one line of therapy (twelve of whom received venetoclax azacitadine), 5 patients received 2 lines of therapy, and 2 patients received 3 lines of therapy. Three patients were second transplants. Donor sources were cord (n=14), haplo cord (n=5), and matched related donor (n=4). 7 patients underwent myeloablative, 14 reduced-intensity, and 2 non-myeloablative conditioning. All patients had MRD at time of transplant (15 by cytogenetics/FISH, 10 by flow cytometry, and 6 by digital droplet PCR). At time of transplant, 1 patient was aplastic, 6 were in morphologic leukemia free state (MLFS), one was in complete remission with incomplete count recovery (CRi), 15 were in complete remission (CR), and 1 had MDS with ringed sideroblasts with multilineage dysplasia. Patient data is summarized in Table 1. Median follow up among survivors is 219 days (range 92-439). Details of individual patients' courses on venetoclax are summarized in Figure 1. In total, 11 patients were still on venetoclax at the time of this analysis, and 4 had completed the planned 1 year of treatment. 3/23 (11%) had the drug permanently discontinued due to potential adverse effects in the setting of additional post-ASCT complications. Four patients relapsed 191, 288, 325, and 367 days post-transplant and all died, and 3 patients experienced transplant related mortality 146, 164, and 307 days post-ASCT. 6 month overall survival (OS) and relapse free survival (RFS) were both 87%. Venetoclax was temporarily held or dose-reduced in 11/23 (47.8%) patients due to adverse events (AEs). 14/23 (61%) patients experienced at least one AE that could be potentially attributed to venetoclax. Most common AEs were cytopenias (7/23, 30%) and diarrhea (7/23, 30%). 12 patients experienced grade 2-4 acute GVHD (3 grade 3). 10 of these patients developed GVHD symptoms prior to starting venetoclax. 5/7 patients who held venetoclax due to diarrhea had acute GVHD prior to initiation of ventoclax, and whether the diarrhea was GVHD or venetoclax related was unclear. Three patients have developed chronic GVHD (2 mild and 1 severe). Our preliminary data suggests venetoclax is tolerable in the post-ASCT maintenance setting without unexpected side effects. In this small cohort, GVHD rates are comparable to historical controls. Ongoing follow-up will continue to examine the safety and efficacy of this approach. Disclosures Pollyea: Novartis: Consultancy; Karyopharm: Consultancy; Takeda: Consultancy; Janssen: Consultancy; Amgen: Consultancy; Pfizer: Consultancy; Agios: Consultancy; 47: Consultancy, Research Funding; Genentech: Consultancy; Glycomimetics: Other; Celgene/BMS: Consultancy; Syndax: Consultancy; Syros: Consultancy; Abbvie: Consultancy, Research Funding; Daiichi Sankyo: Consultancy. OffLabel Disclosure: off label venetoclax post-transplant maintenance

Most cancers evolve over time as patients initially responsive to therapy acquire resistance to the same drugs at relapse. Cancer stem cells have been postulated to represent a therapy-refractory reservoir for relapse, but formal proof of this model is lacking. We prospectively characterized leukemia stem cell populations (LSCs) from a well-defined cohort of patients with acute myelogenous leukemia (AML) at diagnosis and relapse to assess the effect of the disease course on these critical populations. Leukemic samples were collected from patients with newly diagnosed AML before therapy and after relapse, and LSC frequency was assessed by limiting dilution analyses. LSC populations were identified using fluorescent-labeled cell sorting and transplantation into immunodeficient NOD/SCID/interleukin 2 receptor γ chain null mice. The surface antigen expression profiles of pretherapy and postrelapse LSCs were determined for published LSC markers. We demonstrate a 9- to 90-fold increase in LSC frequency between diagnosis and relapse. LSC activity at relapse was identified in populations of leukemic blasts that did not demonstrate this activity before treatment and relapse. In addition, we describe genetic instability and exceptional phenotypic changes that accompany the evolution of these new LSC populations. This study is the first to characterize the evolution of LSCs in vivo after chemotherapy, identifying a dramatic change in the physiology of primitive AML cells when the disease progresses. Taken together, these findings provide a new frame of reference by which to evaluate candidate AML therapies in which both disease control and the induction of more advanced forms of disease should be considered.

Venetoclax and azacitidine (ven/aza) is our institutional preference for initial treatment of non-core binding factor AML patients age ≥60. Given potential durability of response in these patients, the role of transplant (SCT) in patients who achieve response to ven/aza is uncertain. To examine the potential contribution of SCT, we identified 119 patients age ≥60 who received ven/aza as initial AML therapy at our center between 1/2015 and 1/2020. We then compared outcomes for these patients based on whether they were deemed eligible or ineligible for SCT; for those deemed eligible, we compared outcomes between patients who received SCT and those who did not due to personal preference. SCT consults were generally conducted within 2 cycles of initiation of ven/aza but were delayed in some patients achieving CR until evidence of persistent or recurrent measurable residual disease (MRD). Among the 119 patients, 61 (median age 76 (61-90)) were not SCT candidates due to age or comorbidities (n=44) or early death due to primary refractory disease/acute toxicity (n=17). 60 patients were potential SCT candidates. 21 patients median age 65 (60-73) underwent SCT. 2 of these patients had MDS prior to developing AML and underwent SCT consult before developing AML and initiating ven/aza. For the remaining 19 patients, consult occurred median of 50 days (20-287) from diagnosis. 2 patients relapsed after initiating ven/aza prior to transplant, underwent induction, and then went on to SCT. SCT occurred median 176 (69-643) days from diagnosis and median 133 (49-615) days from SCT consult. 37 patients median age 72 (64-78) had SCT consults but did not undergo SCT. 5 of these patients had refractory disease and never achieved SCT eligibility, one patient was deemed not an SCT candidate after consult due to social issues, and one patient was lost to follow up. The remaining 30 patients median age 72 (64-78) achieved adequate response for SCT but elected to defer. 2 of these 30 patients had MDS prior to developing AML and underwent SCT consult before developing AML and initiating ven/aza. Among the 28 patients who had SCT consults after initiating ven/aza and deferred SCT, SCT consult occurred median of 64 days (14-222) from diagnosis. Details of ELN risk status and disease status at time of SCT consult, time of SCT, and best disease response for non-SCT patients are provided in Table 1. MRD positivity included disease detected by cytogenetics, FISH, flow cytometry, or digital droplet PCR (ddPCR). Donor sources were cord (n=11), haplocord (n=3), and matched related donor (n=7). Conditioning regimens included 7 non-myeloablative, 12 reduced-intensity, and 2 myeloablative. Overall survival (OS) was significantly greater among SCT patients (median survival not reached) versus patients undergoing SCT consult and deferring SCT (median 588 days) (p=0.01) versus patients not considered SCT candidates due to age or comorbidities (median 291 days) (p= For SCT eligible patients, SCT appears to improve outcomes among patients who respond to ven/aza compared to patients who defer SCT. Preliminary data suggests that outcomes may be improved for patients going to transplant with MRD following initial therapy with ven/aza as compared to traditional chemotherapy induction. Ongoing data to further refine the significance of MRD, molecular and cytogenetic prognostic factors, and optimal timing of SCT, as well as standardized tools to assess MRD, are necessary. Disclosures Pollyea: Syndax: Consultancy; Glycomimetics: Other; Syros: Consultancy; Abbvie: Consultancy, Research Funding; Karyopharm: Consultancy; Takeda: Consultancy; Daiichi Sankyo: Consultancy; Agios: Consultancy; Celgene/BMS: Consultancy; Pfizer: Consultancy; 47: Consultancy, Research Funding; Janssen: Consultancy; Amgen: Consultancy; Genentech: Consultancy; Novartis: Consultancy.

Background: Venetoclax (ven)+azacitidine (aza) is the standard of care for untreated acute myeloid leukemia (AML) patients (pts) ≥75 years or unfit for intensive induction chemotherapy (IC) due to comorbidities. While the standard of care for younger newly diagnosed pts is IC, long term outcomes for all but favorable risk pts are suboptimal. Specifically, pts with adverse risk disease, per the European Leukemia Network (ELN), are more likely to be refractory to IC and have low rates of long term survival. Ven based regimens have a ~70% response rate in the upfront setting; in contrast to IC, those with adverse risk do not have lower response rates. This study explores the feasibility and outcomes of ven+aza for newly diagnosed younger AML pts who are not necessarily unfit for IC, but carry adverse risk features that make them less likely to respond to this therapy. Methods: This is an interim analysis of an ongoing phase 2 study (NCT03573024) of a planned 36 untreated adverse risk AML pts 18-59. Pts have adequate organ function and white blood cell counts Results: 8 have enrolled between November 2018 and July 2020; median age is 33 (22-52). All have adverse risk disease (Table 1). There were 179 adverse events (AEs); 35 related. The ≥grade 3 events were: fatigue (N=1), leukopenia (N=2), neutropenia (N=2), anemia (N=2), thrombocytopenia (N=1). There were no deaths in the first 60 days. In cycle 1 pts spent a median 10 days (4-19) inpatient and required a median of 2 (0-10) red blood cell transfusions and 3 (0-15) platelet transfusions. Six of 8 (75%) pts responded; all 6 were CRs. 4/6 responses were cytogenetic remissions, 5/6 were MRD negative by MFC and 1/6 was MRD negative by ddPCR. Two pts relapsed, 35 and 84 days after remission; one was successfully salvaged with IC and proceeded to transplant while the other was refractory to IC and died from AML. Two pts were refractory to ven+aza; both were successfully salvaged with IC and have proceeded to transplant. 7/8 proceeded to transplant; 4 required only ven+aza for this outcome and proceeded to transplant a median of 89 days (56-93) after diagnosis. No pts who have proceeded to transplant have relapsed, a median of 340 days (136-444) after transplant. 7/8 remain alive, 345 days (91-586) after enrolling. Conclusions: At the stage 1 interim analysis, the CR rate for ven+aza remains >60% in adverse risk younger patients. With small numbers, up-front treatment with ven+aza allows some younger pts with adverse risk disease a non IC opportunity to achieve a response and proceed to transplant; most of those who relapsed or were refractory were salvaged with IC. Disclosures Pollyea: Syndax: Consultancy; Syros: Consultancy; Abbvie: Consultancy, Research Funding; Daiichi Sankyo: Consultancy; Karyopharm: Consultancy; Novartis: Consultancy; Genentech: Consultancy; Amgen: Consultancy; 47: Consultancy, Research Funding; Janssen: Consultancy; Agios: Consultancy; Glycomimetics: Other; Celgene/BMS: Consultancy; Pfizer: Consultancy; Takeda: Consultancy. OffLabel Disclosure: Venetoclax for younger patients fit for induction chemotherapy

PTPN11 mutations occur in roughly 5-10% of AML patients. There is no defined impact on prognosis in the literature, although a single center analysis from our group suggested an association with poor outcomes. The pathogenesis of activating PTPN11 mutations is hypothesized to result in increased signaling through RAS resulting in uncontrolled cellular proliferation. PU-H71 is an epichaperome inhibitor that has shown pre-clinical promise in ex-vivo and PDX models of AML (Cell Reports 2015). The efficacy of the drug is dependent on the tumor possessing epichaperomes, hyperconnected networks of chaperones and co-chaperones facilitating tumor survival (Nature 2016, Nature Medicine 2018). Prior studies have found an association between epichaperome abundance and overactivity of the transcription factor MYC. Activating PTPN11 mutations have also been shown to result in MYC overactivity through increased SHP2 signaling (protein product of PTPN11 gene) and thus we hypothesized that through increased activity of MYC, activating PTPN11 mutations in AML lead to an abundant epichaperome network, and in turn render these tumors sensitive to the epichaperome inhibitor, PU-H71. 15 PTPN11 mutated AML peripheral blood and bone marrow samples from two academic medical centers (Weill-Cornell Medical Center and University of Colorado) were collected to be analyzed for epichaperome abundance and sensitivity to the epichaperome inhibitor PU-H71. To detect epichaperome abundance, PTPN11 mutated AML samples were incubated with FITC-bound PU-H71 (labeled F2) as well as a chemically altered FITC- bound PU-H71 "control" (labeled F9) which does not bind the epichaperome (Bioorg Med Chem Lett 2011). The AML cell line MV411 was used as a positive control (given known epichaperome abundance and sensitivity to PU-H71) and intra-sample T-cells were used as negative controls. These samples were analyzed using multi-parameter flow cytometry including markers to distinguish blasts, progenitors and lymphocytes. For analysis of sensitivity to PU-H71, samples were incubated with two different concentrations of PU-H71, 1uM and 0.5uM, for 48 hours and subsequently evaluated for viability using multi-parameter flow cytometry (including markers to distinguish blasts, progenitors and lymphocytes). Leukemic blasts and T cells were gated based on immunophenotype, and the ratio of blast F2 to F9 was combined with the F2-F9 blast to T-cell ratio to calculate epichaperome abundance, using a cut-off value of 4.5 to determine medium to high or low levels of the epichaperome (medium to high, >/=4.5, low, Of the 15 samples analyzed for epichaperome abundance. Seven out of 15 (46.7%) samples showed an abundant epichaperome. Nine of these samples had complete viability data, of which, five out of nine (55.5%) showed >60% cell death after 48 hour incubation with PU-H71. For these nine samples with data on sensitivity to PU-H71, the presence or absence of epichaperome abundance was predictive of ex vivo sensitivity (5 sensitive, 4 resistant). We have established PDX-mouse models to evaluate in vivo sensitivity to PU-H71, using epichaperome abundant samples. We have confirmed the epichaperome levels in AML blast cells from the peripheral blood of the animals. Currently the treatment and analysis is ongoing. These data suggest that PTPN11 mutated AML represents a subset of AML that could be particularly sensitive to epichaperome inhibition. Additionally, the findings from our studies suggest this assay has promise as a means to accurately predict sensitivity to PU-H71, a useful precision medicine tool for a difficult to treat disease. All samples are currently being analyzed for MYC transcriptional activity and other signaling pathways. Disclosures Chiosis: Samus Therapeutics: Equity Ownership, Patents & Royalties: Intellectual rights to the PU-FITC assay. Ritchie:agios: Other: Advisory board; Pfizer: Other: Advisory board, travel support; Celgene: Other: Advisory board; Celgene, Novartis: Other: travel support; Celgene, Incyte, Novartis, Pfizer: Consultancy; Ariad, Celgene, Incyte, Novartis: Speakers Bureau; AStella, Bristol-Myers Squibb, Novartis, NS Pharma, Pfizer: Research Funding; Tolero: Other: Advisory board; Genentech: Other: Advisory board; Jazz Pharmaceuticals: Research Funding. Desai:Astellas: Honoraria; Celgene: Consultancy; Cellerant: Consultancy; Astex: Research Funding; Sanofi: Consultancy. Lee:Ai Therapeutics: Research Funding; Karyopharm Therapeutics: Consultancy; AstraZeneca Pharmaceuticals: Consultancy; Roche Molecular Systems: Consultancy; Helsinn: Consultancy; Jazz Pharmaceuticals, Inc: Consultancy. Roboz:Astex: Consultancy, Membership on an entity's Board of Directors or advisory committees; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amphivena: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees; Actinium: Consultancy, Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Membership on an entity's Board of Directors or advisory committees; Trovagene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Sandoz: Consultancy, Membership on an entity's Board of Directors or advisory committees; Roche/Genentech: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees; Otsuka: Consultancy, Membership on an entity's Board of Directors or advisory committees; Orsenix: Consultancy, Membership on an entity's Board of Directors or advisory committees; MEI Pharma: 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; Jazz: Consultancy, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees; Eisai: Consultancy, Membership on an entity's Board of Directors or advisory committees; Daiichi Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celltrion: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Bayer: Consultancy, Membership on an entity's Board of Directors or advisory committees; Astellas: Consultancy, Membership on an entity's Board of Directors or advisory committees. Guzman:Cellectis: Research Funding; SeqRx: Consultancy; Samus Therapeutics: Patents & Royalties: intellectual rights to the PU-FITC assay.

Background: Venetoclax (ven) was approved for older untreated acute myeloid leukemia (AML) patients due to high response rates and durable remissions. As a participating site in the dose escalation study, we observed deeper/more durable responses in some who received >400mg ven. We also noted 16/33 discontinued azacitidine (aza) after achieving a response; 9 relapsed and 7 remained in long term remission on ven only. Based on these observations, we designed a study that hypothesized: A)Higher initial doses of ven would allow deeper/more durable responses, and B)Multi modality high sensitivity measurable residual disease (MRD) testing could identify patients able to discontinue aza and remain on maintenance ven. Methods: This is an ongoing phase 2 study (NCT03466294) of 42 untreated AML patients ≥60 who decline/are ineligible for induction. Patients have adequate organ function and white blood cell counts Results: 30 patients enrolled between May 2018 and July 2019; median age is 71 (60-88), 10% evolved from MDS and 10% and 73% had intermediate and unfavorable risk disease by ELN, respectively (Table 1). 732 adverse events (AEs) occurred; 46 (6%) were serious, the most common were neutropenic fever (37%) and pneumonia (13%). The most common >grade 2 related AEs were leukopenia (53%), thrombocytopenia (44%) and neutropenia (35%); there were no related grade 5 AEs. The overall response rate was 70% (21/30; CR=19, MLFS=2). Median number of cycles to achieve best response was 1. Significant blast reductions were seen on day 8; of the 28 with interpretable day 8 BMBXs, 10 achieved MLFS on day 8. 4 completed ≥1 cycle and were refractory. An additional 4 did not complete cycle 1: 1 died of disease and 3 elected to come off therapy (all subsequently died of disease). Four (19%) responders relapsed, after a median 180 days (27-279). With median follow up of 214 days, median response duration has not been reached. 10 patients died, after a median 65 days (29-256); 1/30 died within 30 days. Median overall survival has not been reached. Of the 26 who completed ≥1 cycle, 19 were MRD- by MPFC, including 18/19 who achieved CR. Of these 26, 3 were not monitored by ddPCR: for 2 patients this was due to the absence of detectable baseline mutations and for 1 patient it was due to refractory disease. The remaining 23 had ddPCR monitoring; 3 became MRD- by this modality (Fig 2). All 3 were also MRD- by MPFC and per protocol discontinued aza and initiated ven maintenance (Fig 1). MRD negativity by both parameters occurred after cycles 1, 2 and 3, respectively. One MRD- patient relapsed after 216 days; two remain in remission after 301 and 124 days. An additional 4 who achieved MRD+ responses discontinued aza at their insistence (and in violation of the protocol); 1 relapsed after 279 days, and 3 remain in ongoing remission. Univariate predictors of refractory disease were FAB M0/M1 (OR 0.070, p=0.02) and RAS pathway mutations (OR 14.25, p=0.02). Conclusions: Higher initial doses of ven are tolerated in this population. Blast reduction occurs quickly in many patients (day 8), for this low intensity regimen. Response rates are consistent with lower doses of ven. Very deep responses, as measured by highly sensitive MRD methods (MPFC and ddPCR are capable of sensitivity up to 0.02%), are attainable. Longer follow up time will determine if higher ven doses and MRD-driven decisions related to continuation of aza result in more durable responses. Increased maturation of blasts and RAS pathway mutations are predictors for refractory disease. Disclosures Lyle: Pfizer: Membership on an entity's Board of Directors or advisory committees; Daiichi Sankyo Incyte: Membership on an entity's Board of Directors or advisory committees; Agios: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees. Pollyea:Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Astellas: Consultancy, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Diachii Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees; Forty-Seven: Consultancy, Membership on an entity's Board of Directors or advisory committees.

Introduction: The prognosis of patients with multiple myeloma (MM) who are resistant to proteasome inhibitors (PIs) and immunomodulatory drugs (IMiDs) is extremely poor. Even daratumumab and BCMA CAR-T cell therapies appear to provide only temporary benefit before patients eventually succumb to their disease. We show that protein translation is upregulated in MM cells. We also demonstrate that translational inhibition with omacetaxine (Oma) is specifically cytotoxic in a manner that is preserved in MM cells from PI/IMiD resistant patients. Furthermore, protein translation inhibition is synergistic with IMiDs, even in IMiD refractory patients. We hypothesize these findings stem from the origin of MM cells as professional antibody secretory cells, with highly active protein translation fundamental to their biology. Here, we show these findings and provide support to test the translation inhibitor omacetaxine with a clinical trial in multiple myeloma patients. Methods: Primary MM cells in mononuclear cell (MNC) cultures were assayed ex vivo for Oma effect at 48-hrs by high-throughput flow cytometry using our myeloma drug sensitivity testing (My-DST) platform. Protein synthesis rates were measured via flow cytometry using the Cayman OP-puromycin assay. Initial drug combination matrices were performed in MM cell lines via MTS assay after 96-hr incubation. Cell line xenograft study of the Oma and IMiD combination in vivo were performed using an orthometastatic MM model with MM1.S luciferase-expressing cells injected IV in NSG mice. Vehicle control, Oma, pomalidomide (Pom) and the combination were dosed 5 days/week by IP injection beginning on day 30. Disease burden was monitored with imaging of luciferase bioluminescence. Results: Oma had specific cytotoxicity across 5 MM cell lines (EC50 = 15-35 nM, data not shown). Oma also showed MM cell specific cytotoxicity in patient samples with EC50 ranging from 25-100 nM (Fig 1A). Ex vivo treatment with 50 nM Oma for 48 hrs caused >50% reduction in viable MM cells in 39/50 (78%) patient samples. Importantly, 50 nM Oma retained activity in PI/IMiD-refractory patients compared to sensitive patients (64.2% vs 71.6% mean decrease in MM cell viability, respectively, P = 0.44, Fig 1B). Primary MM cells exhibited significantly higher levels of baseline protein translation than other bone marrow MNCs with mean 3.9-fold increase (n = 14, P = Conclusion: Protein translation inhibitors represent a potential new drug class for clinical myeloma treatment. We have shown that the protein synthesis inhibitor omacetaxine possesses a potent and specific killing effect against MM cells. Importantly, this anti-myeloma cytotoxicity was equally active in biopsy samples obtained from patients at diagnosis and patients with disease advanced to the late stage of PI/IMiD resistance. The pretreatment level of MM cell protein translation may serve as a biomarker for clinical response. In further exploration of the optimized treatment approach for clinical trials, we found synergy between omacetaxine and pomalidomide in vitro, in vivo and using patient samples ex vivo. A phase I clinical trial of omacetaxine single agent and in combination with pomalidomide is currently in development at the University of Colorado. Disclosures Mark: Janssen: Honoraria; Takeda: Honoraria; Amgen: Honoraria; Celgene: Honoraria.

Introduction: Acute myeloid leukemia (AML) is an aggressive disease with a dismal prognosis. This is largely due to high relapse rates, which stem from our inability to eliminate leukemia stem cells (LSCs) with conventional chemotherapy. As we have gained a deeper understanding of the biology of LSCs, new targets against these chemo-resistant cells have surfaced. One such example is treatment using venetoclax/azacititine, a therapy that has significantly improved the outcome for these patients. Notwithstanding, some patients show resistance to all treatments, and developing a larger repertoire of agents that target LSCs remains an unmet need in this disease. One key vulnerability of LSCs is their dependence on oxidative phosphorylation (OXPHOS). Although signal transducer and activator of transcription 3 (STAT3) has been classically studied as a transcription factor that regulates self-renewal and proliferation, it has also been shown to play an essential role in OXPHOS via regulation of the electron transport chain (ETC). Given that STAT3 is commonly overexpressed in AML, and LSCs are dependent on OXPHOS, we hypothesized that STAT3 may be an effective target for eradication of LSCs. Methods: We have developed a novel small molecule inhibitor of STAT3, SF25. This compound, as well as genetic knockdown of STAT3, was employed to test the functional role of STAT3 in primary AML specimens. Flow cytometry, colony-forming potential, and engraftment of primary samples in PDX mouse models were performed to assess therapeutic efficacy. RNA-seq, seahorse assays, and metabolomics experiments were also performed to determine molecular mechanisms linked to targeting STAT3. Results: Our data shows that inhibition of STAT3 in primary AML samples leads to decreased cell viability, colony-formation and engraftment potential in xenograft models, while not affecting normal hematopoietic stem cells. This effect appears to be a result of mitochondrial dysfunction in LSCs, as seen by a significant decrease in oxygen consumption rate of STAT3 depleted cells. The mitochondrial dysfunction and reduction in OXPHOS is mediated by the downregulation of several mitochondrial and nuclear encoded genes that are important for oxidative phosphorylation, including several electron transport chain complex genes. Inhibition of STAT3 also affects glutaminolysis, as noted by metabolomics analysis of leukemia stem cells treated with STAT3 inhibitor. We suspect this effect is mediated by down-regulation of Myc upon STAT3 inhibition, which blocks glutamine conversion to glutamate, and leads to further decrease in TCA cycle intermediates. Conclusions: Acute myeloid leukemia is an aggressive disease, largely due to the presence of a chemo-resistant population of leukemia stem cells. LSCs highly depend on proper mitochondrial function and OXPHOS, a process that is partly regulated by STAT3 via multiple mechanisms. We propose that inhibition of STAT3 is therefore an effective way of eliminating this population, making this a promising new target in the treatment of AML. Disclosures No relevant conflicts of interest to declare.

The combination of venetoclax with hypomethylating agents has resulted in highly promising clinical outcomes for acute myeloid leukemia (AML) patients. However, a subset of patients are refractory or develop resistance to venetoclax based regimens, resulting in disease recurrence. The goal of this project was to determine a mechanism to re-sensitize resistant leukemia stem cells (LSCs) to venetoclax with azacitidine (ven/aza) treatment. LSCs are the population of leukemia cells that initiate disease and are not fully eradicated by conventional treatments resulting in disease recurrence. We have previously reported that ven/aza targets LSCs in de novo AML patients by perturbing amino acid uptake resulting in decreased oxidative phosphorylation (OXPHOS). To investigate how some AML patients, develop resistance to ven/aza, we first determined if ven/aza reduced amino acid uptake in primary human AML ven/aza resistant LSCs by stable isotope labeled metabolic flux and mass spectroscopy analysis. Amino acid uptake was significantly reduced in both ven/aza sensitive and resistant LSCs upon ven/aza treatment, indicating that ven/aza is still biologically active in resistant LSCs. Next, we performed gene expression analysis from LSCs isolated from AML patients who were treated with ven/aza, responded, and then either remained in remission or progressed on ven/aza therapy. Gene set enrichment analysis revealed that fatty acid transport was enriched in LSCs isolated from patients who eventually progressed on ven/aza therapy (FDR = 0.0088) (Figure A). We then determined differences in overall fatty acid levels by lipidomics mass spectroscopy analysis in ven/aza sensitive and resistant LSCs. We observed a significant increase in abundance of 20% (6/29) of fatty acids detected in resistant LSCs. To determine if targeting fatty acid transport could re-sensitize resistant LSCs to ven/aza we knocked down genes involved in fatty acid transport including CD36, CPT1A and CPT1C in 4 ven/aza resistant AML specimens and then measured viability and colony-forming potential upon ven/aza treatment (Figure B and C). Knockdown of CD36, CPT1A, or CPT1C in combination with ven/aza treatment significantly decreased both viability and colony forming ability in each of the AML specimens. In addition, knockdown of CPT1A or CPT1C in combination with ven/aza reduced OXPHOS, a known metabolic requirement of LSCs. To perturb fatty acid transport in a therapeutically relevant manner, we treated LSCs isolated from ven/aza resistant patient specimens with a CPT1 inhibitor, etomoxir, as a single agent or in combination with ven/aza, and then measured viability and OXPHOS. The combination but not single agents reduced viability and OXPHOS, consistent with our genetic studies. To determine if ven/aza with etomoxir targeted functional LSCs we treated a primary AML specimen with etomoxir, ven/aza or the combination and measured engraftment into immune deficient mice. Combination therapy significantly reduced engraftment potential compared to ven/aza or etomoxir alone indicating that the combination of ven/aza with etomoxir decreased LSC function (Figure D). To determine if ven/aza with etomoxir could target AML cells in vivo, we treated a primary patient derived xenograft model with ven/aza, etomoxir, or the combination for 2 weeks and measured leukemic burden in the bone marrow (Figure E). The combination reduced leukemic burden more significantly than ven/aza or etomoxir alone. Finally, we measured the consequences of ven/aza, etomoxir, or the combination on normal hematopoietic stem and progenitor cells. Neither single agents nor combination therapy decreased CD34+ cell viability or colony forming ability, indicating that there may be a therapeutic window to targeting these metabolic pathways in AML without harming normal stem cells. Gene expression analysis revealed that CD36, CPT1A, and CPT1C are expressed at significantly higher levels in AML compared to HSCs, which may contribute to this therapeutic window. In conclusion, these data indicate that ven/aza resistance can be overcome by targeting fatty acid transport in LSCs. Furthermore, combining ven/aza with a CPT1 inhibitor such as etomoxir may be a clinically relevant strategy to overcoming ven/aza resistance. Figure Disclosures Pollyea: Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees; Astellas: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Forty-Seven: Consultancy, Membership on an entity's Board of Directors or advisory committees; Diachii Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees.

Hematopoietic stem cell (HSC) quiescence is crucial for maintaining lifelong blood production and preventing potentially toxic overproduction of blood cells. HSC are capable of re-entering quiescence following exposure to pro-inflammatory stimuli such as interleukin (IL)-1, implying the existence of one or more 'braking' mechanisms that limit and/or overcome the mitogenic properties of these signals to limit HSC cell cycle entry. However, mechanism(s) regulating HSC quiescence during chronic inflammation have yet to be fully elucidated. In the present study, we find that the master myeloid transcription factor PU.1 represses Myc-regulated cell cycle and protein synthesis pathways in HSC during chronic inflammation, thereby guarding HSC quiescence in this context. To gain insight into HSC cell cycle regulation in the context of chronic inflammatory signaling, we performed cell cycle and RNA-seq analysis on purified HSC from mice injected daily for 20 days with IL-1β. Strikingly, HSC from IL-1-treated mice remained in a quiescent state, and our RNA-seq analyses uncovered a significant decrease in mRNA transcripts from pathways related to cellular proliferation, translation, and ribosome biogenesis. Also, flow cytometry analyses revealed that IL-1 significantly reduced Myc protein expression and ribosomal protein S6 phosphorylation in HSC. Altogether, these data identify an inflammation-induced cell cycle restriction gene program that limits HSC proliferation in response to inflammation. Strikingly, ingenuity pathway analysis (IPA) predicted PU.1as a possible driver of the cell cycle restriction gene program. Indeed, PU.1 mRNA and protein levels in HSC increased significantly in PU.1-eYFP knockin reporter mice treated with IL-1. Furthermore, fractionating HSC based on reporter levels showed that IL-1-exposed HSC expressing high PU.1 levels activated the cell cycle restriction program more robustly than HSC with lower PU.1 levels or from untreated control mice. Along these lines, cell cycle analyses showed that PU.1-high HSC were quiescent following IL-1 exposure. Surprisingly, functional assays revealed that the PU.1-high fraction became enriched for functional HSC following IL-1 exposure, in contrast to a primarily PU.1-low phenotype under control conditions. Collectively, these data show that low Myc levels, a quiescent cell cycle state and functional HSC identity are all associated with high PU.1 levels during chronic inflammation. To address the requirement for PU.1, we compared HSC following IL-1 exposure from wild-type (WT) and PU.1-knockin (KI) mice, which express ~30% of normal PU.1 levels due to a point mutation in the PU.1 upstream regulatory element (URE). Strikingly, IL-1 induced aberrant HSC expansion in the BM of PU.1 KI mice, underwritten by IL-1-induced quiescence loss in PU.1 KI HSC. Excess cell cycle activity in HSC from IL-1-treated PU.1 KI mice was associated with Myc overexpression, hyperinduction of cycle and protein synthesis genes, and an increased protein synthesis rate. Interestingly, IL-1-treated PU.1 KI mice also exhibited exuberant platelet production and aberrant activation of megakaryocyte (Mk) lineage programs in HSC. Taken together, these data show PU.1 is required to limit HSC proliferation and cell cycle activity, and that failure to upregulate PU.1 during chronic inflammation leads to aberrant HSC expansion and dysregulated Mk lineage output. Altogether, our data identify PU.1 as a key enforcer of HSC dormancy, and as a regulator of HSC proliferation and proteostasis during chronic inflammation. As PU.1 has been shown to slow cell cycle entry to promote myeloid differentiation in actively proliferating cells, our results suggest a mechanistic conservation in which inflammation induces PU.1 expression and engages a similar cell cycle restriction program that preserves HSC quiescence. Such a mechanism can prevent excessive Mk lineage output leading to thrombosis, and/or HSC exhaustion. Importantly, these findings re-cast inflammation-induced PU.1 expression as a guardian of blood system function, rather than a pathogenic by-product. As PU.1 serves a tumor suppressor role, these findings also may provide insight into the link between inflammation and leukemogenesis, particularly in contexts where PU.1 function is impaired in leukemic stem cells (LSC) due to mutation or indirect mechanism(s). Disclosures No relevant conflicts of interest to declare.

Recent clinical trials have reported that in combination with hypomethylating agents, the BCL-2 inhibitor venetoclax can induce responses in over 70% of older previously untreated AML patients who are unfit for conventional chemotherapy. These findings led to the recent United States Food and Drug Administration approval of this regimen for this population, and it is now considered to be the standard care. However, a significant minority of patients do not achieve a remission and are refractory. In addition, the majority of patients who do achieve a remission ultimately relapse. It is therefore critical to identify AML patients who are likely to be resistant to venetoclax-based therapy. To initially address this question, we retrospectively reviewed 75 newly diagnosed AML patients who received venetoclax + azacitidine (VEN+AZA) at our institution and analyzed several baseline clinical features to determine the ability of each to predict disease that was refractory to treatment (defined as a lack of complete remission [CR], CR with incomplete recovery of peripheral blood counts [CRi], partial remission, or morphological leukemia free state [MLFS]). Both univariate and multivariate analyses revealed the presence of FAB-M5 to be associated with disease that was refractory to VEN+AZA (Table 1). Given that FAB-M5 represents AML with monocytic differentiation, these findings indicate a strong correlation between myeloid differentiation status and resistance to venetoclax. Using multicolor flow cytometry, we show bone marrow specimens of typical FAB-M5 patients who were refractory to VEN+AZA presented dominant monocytic disease that has an immunophenotype of CD45-bright/SSC-high/CD117-/CD11b+/CD68+ (Figure A). In contrast, bone marrow specimens of typical FAB-M0/M1/M2 patients who achieved CR with VEN+AZA presented as a single dominant disease population that is CD45-med/SSC-low/CD117+/CD11b-/CD68- (Figure B). In a subset of AML patients, we observed the co-existence of both phenotypically primitive and monocytic populations, which we term "MPM" AML (for Mixed Primitive/Monocytic). We observe that after attaining CR with VEN+AZA treatment and subsequent relapse, MPM-AML showed almost complete loss of the primitive subpopulation, and evolved to a dominant monocytic disease (Figure C). These data indicate that VEN+AZA treatment induces strong selection of the monocytic phenotype. Importantly, when we compared the immunophenotype of six pairs of diagnostic/relapse specimens from AML patients treated with conventional intensive induction chemotherapy, we observed selection of a more primitive phenotype, suggesting the drive toward a monocytic phenotype observed at relapse appears to be a unique consequence of VEN+AZA therapy. To our knowledge, selection of a monocytic phenotype at relapse has never been previously observed in AML, suggesting the relapse after VEN+AZA may represent a new clinical entity. Mechanistically, using RNA-seq we show the global transcriptome of monocytic AMLs are distinct from primitive AMLs, suggesting they represent two broad classes of AML with likely differential responses to therapy. Indeed, we demonstrate that AML with a primitive immunophenotype is dependent on BCL-2 activity as a means to drive oxidative phosphorylation, a critical requirement for survival of leukemia stem cells. Conversely, AML with a more differentiated monocytic phenotype is no longer dependent on BCL-2, but rather switches to MCL-1 as a mediator of oxidative phosphorylation. Using colony-forming and xenograft assays, we show the stem and progenitor potential of monocytic AMLs are selectively more sensitive to MCL-1 inhibition comparing to BCL-2 inhibition. Together, our study suggests a significantly higher refractory/relapse risk for monocytic AML patients treated with VEN+AZA (Figure D). Further, for those AML patients who do respond to initial VEN+AZA treatment, the therapy drives a powerful selective process resulting in emergence of more differentiated monocytic disease in some patients. Based on these findings, we propose that AML exists on a developmental spectrum that is inherently fluid, where with appropriate selective pressure the disease can acquire characteristics of a more differentiated state. Further, our data indicate that optimal AML therapy will require strategies designed to target both primitive and myeloid phenotypes. Disclosures Pollyea: Forty-Seven: Consultancy, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Membership on an entity's Board of Directors or advisory committees; Astellas: Consultancy, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Diachii Sankyo: Consultancy, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Savona:TG Therapeutics: Membership on an entity's Board of Directors or advisory committees, Research Funding; Takeda: Membership on an entity's Board of Directors or advisory committees, Research Funding; Selvita: Membership on an entity's Board of Directors or advisory committees; Karyopharm Therapeutics: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees; Incyte Corporation: Membership on an entity's Board of Directors or advisory committees, Research Funding; Sunesis: Research Funding; Celgene Corporation: Membership on an entity's Board of Directors or advisory committees; Boehringer Ingelheim: Patents & Royalties; AbbVie: Membership on an entity's Board of Directors or advisory committees.

Abstract 243 Disruption of normal hematopoietic development is a major problem in acute myeloid leukemia (AML). Osteoblastic cells have been shown to support hematopoiesis. We hypothesized that myeloid leukemia inhibits osteoblastic cells contributing to the loss of normal hematopoiesis. To define effects of leukemia on osteoblastic cells, we employed a previously described syngeneic murine model of AML where transplant of leukemic cells does not require irradiation of recipients (Neering et al. Blood, 2007). Leukemic mice had decreased serum levels of the bone formation marker osteocalcin (OC) (92.6 ± 11.6 vs 25.9 ± 4.4 ng/ml Normal (N) vs Leukemic (L) p Disclosures: No relevant conflicts of interest to declare.

We have previously shown that the plant-derived compound parthenolide (PTL) can impair the survival and leukemogenic activity of primary human acute myeloid leukemia (AML) stem cells. However, despite the activity of this agent, PTL also induces cellular protective responses that likely function to reduce its overall cytotoxicity. Thus, we sought to identify pharmacologic agents that enhance the antileukemic potential of PTL. Toward this goal, we used the gene expression signature of PTL to identify compounds that inhibit cytoprotective responses by performing chemical genomic screening of the Connectivity Map database. This screen identified compounds acting along the phosphatidylinositol 3-kinase and mammalian target of rapamycin pathways. Compared with single agent treatment, exposure of AML cells to the combination of PTL and phosphatidylinositol 3-kinase/mammalian target of rapamycin inhibitors significantly decreased viability of AML cells and reduced tumor burden in vitro and in murine xenotransplantation models. Taken together, our data show that rational drug combinations can be identified using chemical genomic screening strategies and that inhibition of cytoprotective functions can enhance the eradication of primary human AML cells.

Previous studies have demonstrated the importance of energy metabolism as it relates to numerous aspects of leukemia stem cell (LSC) biology. Specifically, in acute myelogenous leukemia (AML), it has been shown that LSCs have a unique reliance on oxidative phosphorylation (OXPHOS), and that inhibition of B-cell lymphoma 2 (BCL-2) acts to down-regulate OXPHOS and eradicate LSCs in pre-clinical models. In the clinical setting, when BCL-2 inhibitor venetoclax is combined with azacitidine, high response rates (~80%) in elderly de novo AML patients have been observed (PMID: 29339097). Nonetheless, a significant portion of these patients ultimately experience disease progression. The mechanism of resistance and characteristics of these patients is poorly understood. Our preliminary data shows that the venetoclax + azacitidine (ven/aza) regimen targets LSCs through alteration of energy metabolism. Specifically, the regimen disrupts the TCA cycle leading to decreases in ATP production and inhibition of OXPHOS. This metabolic targeting is central to ven/aza efficacy and we hypothesize that resistance and progression of AML patients is due to compensatory mechanisms that restore sufficient levels of OXPHOS (PMID:3333149). Investigation of such mechanisms led us to explore the potential activity of MCL1. Previous studies have shown MCL1 can influence venetoclax resistance, however little is known about MCL1's role in metabolism, although a report in breast cancer cells suggests MCL1 modulates OXPHOS (PMID: 28978427). In leukemia, MCL1 expression has been shown to be partially upregulated through mutations in protein tyrosine phosphatase non- receptor type 11 (PTPN11), and PTPN11 mutations have been shown to increase LSC frequency. Thus, we hypothesized that PTPN11 mutations may confer resistance to venetoclax-based regimens at least partially by up-regulation of MCL1. To test this hypothesis, we investigated the relationship between PTPN11 mutations, MCL1, and the metabolic phenotype. In comparison to specimens with a wild type allele, LSCs isolated from PTPN11 mutant patient specimens showed increased levels of OXPHOS as well as glycolysis, amino acids, and fatty acids, suggesting an ability to utilize multiple energy sources for survival. PTPN11 mutant specimens also show decreased sensitivity to venetoclax, suggesting OXPHOS is not affected by venetoclax to the same degree as PTPN11 wild type specimens (fig. 1). Furthermore, when we introduced a mutated allele of PTPN11 into a primary AML specimen we observed increased oxidative phosphorylation and glycolysis, which correlated with decreased in vitro sensitivity to venetoclax (fig. 2). To test the potential role of MCL1 in PTPN11 mutant specimens, we employed a small molecule MCL-1 inhibitor. Metabolic analysis of specimens treated with the MCL-1 inhibitor showed decreased OXPHOS in PTPN11 mutant specimens (fig 3). Further, PTPN11 mutant specimens exhibit increased sensitivity to the MCL-1 inhibitor (fig. 4). To investigate a potential mechanistic link to clinical observations, we next examined 45 older, previously untreated AML patients from our institution who received ven/aza, both in the context of the multi-institutional study NCT02203773 (N=33) and with off-label use (N=12). Of 12 variables examined, only the presence of PTPN11 predicted shorter response duration (table 1). In addition, of the 9 patients who progressed ven/aza, 2 (22%) acquired PTPN11 mutations upon progression, further suggesting PTPN11 may represent a resistance mechanism to this regimen. Notably, PTPN11 is not preferentially detected in patients who progress after regimens other than ven/aza. In conclusion, AML containing PTPN11 mutations exhibit a unique energy metabolism profile. These specimens also appear to have increased sensitivity to MCL-1 inhibitors. The presence of PTPN11 mutations represents both a novel method for predicting response to ven/aza and a potential strategy for targeting patients who progress. We propose that addition of an MCL-1 inhibitor for treatment of AML patients bearing PTPN11 or related mutations may increase therapeutic responses. Disclosures Savona: Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Incyte: Membership on an entity's Board of Directors or advisory committees, Research Funding; Boehringer Ingelheim: Consultancy. Fesik:Boehringer Ingelheim: Consultancy. Pollyea:Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Curis: Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Research Funding; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy.

The goal of this project was to identify and target metabolic vulnerabilities of leukemia stem cells (LSCs) to improve therapeutic outcomes for patients with AML. We have previously shown that primary human LSCs reside in a unique metabolic condition characterized by a relatively low oxidative state (termed "ROS-low") and increased levels of glutathione (Lagadinou et al. Cell Stem Cell, 2013). Cells in this condition are highly dependent on oxidative phosphorylation for survival, in striking contrast to many tumor cells which often rely on glycolysis; indicating that LSCs are governed by distinct metabolic properties. To further elucidate key metabolic properties of LSCs, we measured differences in the global metabolome of ROS-Low LSCs in comparison to ROS-high AML blasts. Our preliminary data demonstrated that ROS-low LSCs have higher levels of amino acids and require amino acid catabolism for survival. We hypothesized that certain individual amino acids may be more important for LSC survival. If true, then targeting specific amino acids may be an avenue towards improved AML therapy. To determine if any individual amino acid is essential for LSC survival, we analyzed AML cells from five patients that were systematically cultured in media lacking one of the twenty amino acids. Cysteine depletion was consistently the most cytotoxic, showing decreased cell viability and colony forming potential of LSCs. We next determined the effect of an engineered human enzyme that selectively degrades cysteine and cystine (AEB3103, Aeglea BioTherapeutics, Inc.) on LSC viability and colony forming potential. We found that AEB3103 treatment decreased viability of LSCs in all AML specimens tested and significantly decreased colony formation (p To determine how AEB3103 decreased LSC viability we investigated the metabolic changes that occur upon AEB3103 treatment by mass spectroscopy. We found that AEB3103 treatment decreased the abundance of metabolites involved in glutathione synthesis (cysteine, cystine, glutathione, taurine, and Cys-Gly) in ROS-low LSCs. Tracing of cysteine13C315N in LSCs demonstrated that all detectable heavy cysteine was metabolized to cystine and glutathione. Furthermore, pretreatment with cell permeable glutathione rescued cell viability and colony-forming potential upon AEB3103 treatment. These data suggest that modulation of glutathione is central to the mechanism by which AEB3103 kills LSCs. Glutathione is a well-characterized antioxidant, therefore, we measured ROS levels and the expression of genes known to be expressed upon ROS induction in LSCs. Surprisingly, we did not observe changes in ROS induction or the expression of ROS induced genes upon AEB3103 treatment. Next, we interrogated cellular functions of glutathione that are independent from the role of glutathione as an antioxidant. Glutathione has previously been shown to mediate electron transport chain complex II function via posttranslational glutathione modification, glutathionylation. Therefore, we hypothesized that decreased glutathione levels upon AEB3103 treatment could result in decreased complex II activity and therefore decreased levels of OXPHOS and ATP production. To test this hypothesis, we measured complex II activity, complex II glutathionylation, oxidative phosphorylation, and ATP levels upon AEB3103 treatment. We observed AEB3103 treatment significantly decreased glutathionylation, complex II activity, oxidative phosphorylation, and ATP levels in AML cells (p Disclosures Pollyea: AbbVie: Consultancy, Research Funding; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Consultancy; Curis: Membership on an entity's Board of Directors or advisory committees.

Introduction: Allogeneic hematopoietic stem cell transplantation (HSCT) remains one of the only curative therapies for acute myeloid leukemia (AML) in adult patients. However, a high proportion of patients will relapse post-HSCT. Measurable residual disease (MRD) has emerged as highly predictive of relapse after chemotherapy and multiple studies have demonstrated inferior outcomes after HSCT for patients with pre-HSCT MRD. Fewer studies have evaluated post-HSCT MRD, but those have suggested that MRD is a powerful predictive tool in this setting as well. We have previously utilized droplet digital PCR (ddPCR) as an MRD platform to track 33 unique AML-associated mutations in a cohort of patients receiving either cytotoxic chemotherapy or targeted agents, and our preliminary data suggest that persistence or recurrence of these mutations during therapy is predictive of relapse and diminished overall survival. In the present study we sought to evaluate the utility of ddPCR specifically for post-HSCT MRD monitoring. Methods: Patients 18 years of age or older with AML who had received allogeneic HSCT were identified via retrospective chart review. Thirty-seven patients were identified who had sufficient diagnostic testing to identify AML-associated point mutations and had available DNA from serial bone marrow samples. Thirty-six of these patients had at least one pre-HSCT sample analyzed; post-HSCT day 28, day 80, and day 365 samples were available for 26, 25, and 11 patients, respectively. DdPCR was utilized to track 21 different AML-associated mutations in this patient cohort. MRD negativity was defined as variant allelic frequency (VAF) below the limit of detection for all assayed mutations (1-4 mutations per patient). Relapse-free (RFS) and overall (OS) survival were defined as time from date of transplant to relapse or death. Survival statistics were analyzed as Kaplan-Meyer curves, with p-values determined by the log-rank (Mantel-Cox) test. P-values less than 0.05 were considered statistically significant. Outcome data for all patients were censored July 6, 2018. Results: Median follow-up from time of HSCT was 13.3 months. Of the 37 patients included in this cohort, the majority (73%) received regimens containing cytotoxic chemotherapy prior to HSCT; the remainder (27%) received targeted therapies +/- low-intensity chemotherapy prior to HSCT. Only 6 of 36 patients (17%) were MRD negative pre-HSCT, all of whom had received cytotoxic chemotherapy. All 37 patients had one or more post-HSCT samples available for MRD assessment. Of these, 30 achieved MRD negativity after HSCT (81%). Post-HSCT MRD status was predictive of both RFS and OS (Figure 1). This replicates previously published data with respect to ddPCR-based MRD, but is based on evaluation of both more patients and more mutations. In agreement with previously published data showing high predictive value of the 1-month post-HSCT MRD, we found this timepoint specifically to be predictive of RFS and OS in the 26 patients for whom a sample was assayable (Figure 2). Neither outcomes nor MRD status by ddPCR varied based on conditioning regimen (myeloablative versus reduced-intensity) in our cohort (data not shown). We also compared ddPCR MRD status with whole bone marrow chimerism measured by short tandem repeat (STR) analysis. Concordance between MRD negativity and >95% donor chimerism was 69% at 1 month post-HSCT and 93% at 80 days post-HSCT. Of 8 discordant samples at the 1 month post-HSCT time point, 3 had donor chimerism Conclusions: ddPCR-based molecular MRD assessment is predictive of relapse-free and overall survival post-HSCT in patients with AML, and may be a powerful adjunct to surveillance measures already in clinical use, such as chimerism. Presence of molecular MRD specifically at 1 month post-transplant showed significant correlation with relapse and with overall mortality. Detection of AML-associated mutations at this and later time points in a prospective manner could in many cases allow salvage of patients prior to clinical relapse, thus improving survival for this high-risk population. Disclosures Pollyea: Gilead: Consultancy; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Curis: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Research Funding; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding.

Most AML patients who receive intensive chemotherapy achieve a significant clinical response; however, the majority will relapse and succumb to their disease, indicating that leukemia stem cells (LSCs) are not effectively targeted. Further, it has recently been shown that LSC frequency and phenotypic diversity are increased at relapse (Ho et al. Blood, 2016), thereby creating an even more challenging clinical scenario. Thus, novel therapies specifically designed to target LSCs in relapsed AML patients are urgently needed. Previously, we have shown that LSCs can be targeted by perturbing energy metabolism (Lagadinou et al. Cell Stem Cell, 2013). Therefore, the goal of the current study was to identify and target metabolic dependencies of relapsed LSCs, with the hope that this would allow improved efficacy for AML patients with relapsed disease. To achieve this objective we first measured metabolic differences in LSCs isolated from de novo and relapsed patients. This analysis revealed that relapsed LSCs have significantly increased levels of nicotinamide compared to de novo LSCs (Figure 1A). Nicotinamide is a precursor of NAD+, an essential coenzyme in energy metabolism. We hypothesized that relapsed LSCs are dependent on nicotinamide metabolism to maintain energy metabolism. To test this hypothesis, we targeted nicotinamide metabolism with the small molecule APO866, an inhibitor of Nampt, the rate-limiting enzyme for conversion of nicotinamide to NAD+. This resulted in a significant decrease in NAD+ in LSCs isolated from both de novo and relapsed AML specimens (data not shown). However, strikingly, inhibition of nicotinamide metabolism only decreased viability and colony-forming ability of LSCs isolated from relapsed AML patients, not LSCs from untreated patients (Figure 1B). To verify that inhibition of Nampt was targeting functional LSCs, we treated a relapsed AML patient specimen with APO866 for 24 hours and measured the ability of the leukemia cells to engraft into immune deficient mice. We observed a significant reduction in leukemia engraftment upon APO866 treatment (data not shown). Importantly, inhibition of nicotinamide metabolism did not affect normal hematopoietic stem cell frequency or colony forming ability (data not shown). Altogether, these data suggest that inhibition of nicotinamide metabolism specifically targets relapsed LSCs. We next sought to understand the mechanism by which inhibiting nicotinamide metabolism targets relapsed LSCs. To this end we measured changes in the major energy metabolism pathways (oxidative phosphorylation [OXPHOS] and glycolysis) in LSCs isolated de novo and relapsed AML patient specimens. Upon APO866 treatment, we observed a significant decrease in OXPHOS and OXPHOS capacity in relapsed LSCs but not de novo LSCs (Figure 1C). Furthermore, no change in glycolysis was observed (data not shown). These data demonstrate that inhibition of nicotinamide metabolism targets OXPHOS specifically in relapsed LSCs. To determine how APO866 reduced OXPHOS, we measured stable isotope metabolic flux of amino acids, the fatty acid palmitate, and glucose into the TCA cycle after APO866 treatment. We observed an increased accumulation of citrate, malate, and α-ketoglutarate from amino acids and palmitate, consistent with decreased activity of the NAD+ dependent enzymes isocitrate dehydrogenase, α-ketoglutarate dehydrogenase and malate dehydrogenase (data not shown). Through direct measurement of enzyme activity, we confirmed that isocitrate dehydrogenase, α-ketoglutarate dehydrogenase and malate dehydrogenase activity were each significantly decreased upon APO866 treatment (Figure 1D). Consistent with our previous findings we did not observe any changes in glycolysis or glucose contribution to the TCA cycle (data not shown). Overall, these data suggest that inhibition of nicotinamide metabolism through Nampt inhibition results in decreased OXPHOS through decreased TCA cycle activity. In conclusion, we have shown that relapsed LSCs have distinct metabolic properties including increased levels of nicotinamide, which can be selectively targeted to eradicate relapsed LSCs. We propose that therapeutic strategies designed to target nicotinamide metabolism may be useful for relapsed AML patients and may allow for broad efficacy such as that observed when LSCs are targeted in the up-front treatment setting. Disclosures Nemkov: Omix Technologies inc: Equity Ownership. Pollyea:Curis: Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; AbbVie: Consultancy, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy; Karyopharm: Membership on an entity's Board of Directors or advisory committees.

Epidemiological studies indicate that obese populations have an increased risk for leukemia. However, the mechanism underlying this phenomenon remains unclear. In the current study, utilizing both murine AML models and human AML samples, we observe that AML pathogenesis leads to aberrancies in adipose tissue, pancreatic function, and gut/microbiome, all of which contribute to an insulin resistant phenotype. We demonstrate that through induction of insulin resistance, leukemic disease alters systemic metabolism and thereby redirects systemic glucose to be preferentially available to malignant cells. We found that insulin effect on leukemic mice was significantly impaired as shown by insulin tolerance tests (ITT) (Fig.1) as well as by reduced glucose utilization in both adipose and muscle tissues. Interestingly, glucose utilization in leukemia cells was not affected by insulin. Several mechanisms were found to underlie this insulin resistant phenotype. First, leukemia induced a high-level production of IGFBP1 from adipose tissue and led to a 100-fold increase in circulating IGFBP1, which impaired insulin sensitivity (Fig.1). Blocking IGFBP1 partially restored insulin sensitivity and reduced leukemic burden, whereas pretreating with IGFBP1 facilitated leukemic progression. Second, a 95% reduction in the gut-derived circulating serotonin led to a significant decrease in insulin secretion from pancreas in leukemic mice (Fig.2). Serotonin supplementation partially restored serum insulin levels and decreased leukemic burden by 50% (Fig.2). Finally, the profile of gut microbiota in leukemic mice was distinct from normal mice. Leukemia-associated microbiota functionally contributed to the insulin resistant phenotype and therefore promoted disease progression. Mechanistically, decreased productions of microbiota-derived short chain fatty acids (SCFAs) i.e. butyrate and propionate, were found in leukemic fecal materials and butyrate supplementation reduced leukemic burden by 50% (Fig.2). Of note, there was interplay between these mechanisms. For example, agents increasing insulin levels resulted in a decreased IGFBP1 production in leukemic mice. Further, leukemia-associated microbiota also contributed to the elevated IGFBP1 level and the reduced insulin level in leukemic mice. To test the therapeutic relevance of our findings, we combined serotonin supplementation with butyrate, and found this combination (hereafter termed 'Ser-Bu') reduced leukemic burden by 80% (Fig.3) and provided survival benefits. More importantly, multiple assays including PET-CT scanning (Fig.3), 3H-2-DG labeling, and 13C-NMR labeling were employed to demonstrate that Ser-Bu treatment increased glucose uptake/utilization in adipose and muscle tissues by 200% and 50% respectively and reduced glucose uptake/utilization in leukemia cells by 70% (Fig.3). Therefore, leukemia progression can be significantly attenuated solely by modulation of systemic glucose metabolism. To examine the human relevance of our findings, we analyzed serum samples from normal controls, MDS and AML patients and found that a 6-fold and a 14-fold increase of IGFBP1 in MDS and AML serum respectively compared to normal controls (Fig.4). Additionally, other insulin resistance indicators such as serum free fatty acids, inflammatory cytokines and serum Leptin and Resistin levels were all elevated in AML serum. An insulin resistant phenotype and elevated serum IGFBP1 were also observed in a primary human leukemia specimen xenograft model. Further, a 65% reduction in serotonin was found in AML serum compared to normal controls (Fig.4). Impressively, analyses of paired diagnostic, remission and relapsed BM specimens showed that IGFBP1 was decreased in remission samples and rebounded in the relapsed state, whereas serotonin level showed the opposite pattern (Fig.5). Together, these data support that an insulin resistant phenotype is also evident in AML patients. Collectively, our studies suggest that leukemic tumors gain a competitive advantage by co-opting multiple mechanisms to induce a diabetes-like physiological condition and thereby subvert systemic glucose metabolism to facilitate disease progression. Our studies demonstrate that restoration of normal glucose regulation may be a feasible strategy to suppress systemic growth of malignant cell types. Disclosures Pollyea: Gilead: Consultancy; Curis: Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Karyopharm: Membership on an entity's Board of Directors or advisory committees; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; AbbVie: Consultancy, Research Funding; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees.

Rheumatoid arthritis (RA) is a debilitating autoimmune disease resulting from autoantibodies that cause damage to synovial joints. Joint damage causes increased systemic inflammatory cytokines which may lead to aberrant hematopoiesis. Indeed, RA is accompanied by many hematological complications including anemia, cytopenias, and suppressed bone marrow function. Hematopoietic stem cells (HSC) at root of the blood system can respond to inflammatory signals by activating the cell cycle and preferentially generating myeloid cells. However, chronic inflammation can also lead to HSC dysfunction. Previous studies using genetic mouse models of RA have identified myeloid overproduction in this context; however, HSC long-term reconstitution activity was maintained. To better understand hematopoietic alterations in RA, our group used the collagen induced arthritis (CIA) mouse model, which is inducible in adult mice and recapitulates many immunological features of the human disease, including elevated inflammatory cytokine levels in the bone marrow (BM) and peripheral blood (PB). Confirming prior reports, we found increased numbers of myeloid lineage cells in the PB and BM of CIA mice. We also found reduced erythroid and lymphoid lineage progenitor cell numbers in the BM, consistent with anemia and immunosenescence phenotypes in RA patients. Interestingly, these features were accompanied by a significant increase in the number of myeloid-biased multipotent progenitor-3 (MPP3) cells, suggesting increased activation of myeloid differentiation pathways. However, we found no changes to the number of activated (MPP1), short-term HSC (ST-HSC), or long-term HSC (LT-HSC) in CIA mice. Likewise, and in line with previous reports, long-term HSC potential was not reduced in CIA mice, as assessed by transplantation of either purified HSC or with unfractionated BM into irradiated recipient mice. While HSC from control and CIA donor mice displayed similar blood chimerism and lineage distribution over a 16-week period, we did observe increased proportions of donor-derived MPP3 in CIA recipient animals, further supporting an activation of myeloid HSC differentiation pathways. Overall, these results reveal underlying changes in the BM driving aberrant hematopoiesis, with the HSC pool remaining intact despite activation of a myeloid differentiation pathway. To better understand how HSC are impacted by arthritic inflammation, we assessed the molecular state of control and CIA HSC using RNA-seq. We found 292 genes significantly upregulated and 237 genes significantly downregulated in CIA HSC. Analysis of these genes using Ingenuity, GSEA, and DAVID tools revealed broad downregulation of inflammatory and proliferation signaling pathways including IL-1β, NFκB, MYC, and ERK. Genes for G1/S cell cycle transition, transcription, protein translation, and proliferation pathways were also significantly downregulated in CIA HSC. On the other hand, genes corresponding to cell cycle arrest and negative regulation of transcription were significantly upregulated. Notably, we find that IL-1β, which is produced in the BM of CIA mice, is sufficient to induce this molecular program. We find that HSC in CIA mice have a global downregulation of transcripts required for activation and proliferation, and a global upregulation of transcripts that would promote quiescence. Hence, HSC are forced back into a quiescent state by broad downregulation of cell growth and proliferation genes even during chronic inflammation caused by RA. Altogether, our data show that a mouse model of rheumatoid arthritis causes hematopoietic lineage skewing towards the myeloid lineage with simultaneous loss of lymphoid and erythroid lineage potential. Interestingly, in this chronic inflammatory setting HSC downregulate pathways involved in cytokine signaling, cell cycle activation, and translation. This mechanism can be triggered by chronic exposure to pro-inflammatory cytokines, and may serve to limit HSC proliferation and potential for damage in disease settings. These results may explain the relative rarity of outright bone marrow failure in autoimmune disease patients, while providing insight into mechanisms driving aberrant hematopoiesis in these individuals. Lastly, they provide functional evidence for cytokine blockade to normalize HSC function in the setting of RA and other chronic inflammatory diseases. Disclosures No relevant conflicts of interest to declare.

Outcomes for AML patients remain poor because of the inability to fully eliminate leukemia stem cells (LSCs). We have previously shown that primary human LSCs reside in a unique metabolic condition characterized by a relatively low oxidative state (termed "ROS-low") and increased levels of glutathione (Lagadinou et al. Cell Stem Cell, 2013). Cells in this condition are highly dependent on oxidative phosphorylation (OXPHOS) for survival, in striking contrast to many tumor cells which often rely on glycolysis, suggesting that LSCs are governed by distinct metabolic properties. Therefore, the goal of this project was to identify and target metabolic vulnerabilities of LSCs. To achieve this objective, we used mass spectroscopy to interrogate the metabolome of leukemia stem cells (LSCs) isolated from primary human AML specimens. We observed significant increases in the levels, uptake, and metabolism of amino acids in LSCs compared to bulk AML cells. These data suggest that LSCs may preferentially rely on amino acids for survival. To investigate this hypothesis, we cultured LSCs and bulk leukemia cells isolated from primary leukemia specimens in media lacking amino acids and measured cell viability and colony forming potential. We found that LSCs were uniquely sensitive to amino acid loss. In addition, LSCs formed significantly fewer colonies upon amino acid depletion compared to LSCs cultured in media containing amino acids. To confirm that amino acid depletion was targeting functionally-defined LSCs, we employed engraftment assays in immune incompetent mice. Culturing primary AML cells without amino acids for 24 hours resulted in significantly reduced levels of leukemia cell engraftment. Next, we interrogated whether amino acid depletion impaired normal HSC survival and function by culturing mobilized peripheral blood without amino acids and measuring the frequency of CD34+ cells, colony forming ability, and engraftment into immune deficient mice. HSC frequency, colony forming ability, and engraftment potential were not changed by amino acid depletion. Altogether, these data demonstrate the LSCs are selectively dependent on amino acids for survival. We next determined how amino acids modulate LSC biology by measuring the consequences of amino acid loss on LSC metabolism. We observed that amino acid depletion decreased OXPHOS specifically in LSCs and not in bulk leukemia cells. We have previously shown that BCL-2 inhibition decreases OXPHOS in LSCs (Lagadinou et al. Cell Stem Cell, 2013). Importantly, recent studies have shown that inhibition of BCL-2 using the BCL-2 inhibitor venetoclax in combination with azacitidine has resulted in superior outcomes for AML patients (Dinardo et al. Lancet Oncology, 2018). Furthermore, our preliminary data demonstrates that venetoclax with azacitidine targets LSCs in AML patients. Therefore, we hypothesized that venetoclax with azacitidine may be targeting LSCs by modulating OXPHOS via amino acid metabolism. To test this hypothesis, we isolated LSCs from AML patients undergoing treatment with venetoclax and azacitidine. LSC specimens obtained pre and 24 hours after initiation of therapy were analyzed for changes in OXPHOS, gene expression, and the metabolome. We observed that venetoclax with azacitidine treatment decreased OXPHOS and reduced amino acid levels. In addition, expression of amino acid transporters was down-regulated. Finally, we sought to determine if culturing LSCs in high levels of amino acids before venetoclax and azacitidine treatment could rescue LSC viability and OXHPOS. We found that culturing LSCs with increased levels of amino acids rescued LSCs survival and OXPHOS, demonstrating that venetoclax with azacitidine targets LSCs by decreasing amino acid levels. Taken together, our data indicate that LSCs are selectively reliant on amino acid metabolism to fuel OXPHOS. Furthermore, amino acid metabolism can be targeted in AML patients by venetoclax with azacitidine treatment. These studies are the first to characterize metabolic targeting of LSCs in AML patients. Disclosures Nemkov: Omix Technologies inc: Equity Ownership. Pollyea:Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Research Funding; Celgene: Membership on an entity's Board of Directors or advisory committees; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Gilead: Consultancy; Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Curis: Membership on an entity's Board of Directors or advisory committees.

Background: The oncogenic drivers and progression factors in multiple myeloma (MM) are heterogeneous and difficult to target therapeutically. As a result, personalized medicine approaches have not yet been realized. However, clinical availability of numerous anti-myeloma drugs and readily obtainable bone marrow (BM) aspirates raises the possibility to benefit patients by profiling the drug sensitivity of their MM. Despite newly available drugs, resistance inevitably develops and there is no defined rationale for sequencing therapy. Treatments now consist of >15 agents, including IMiDs, protease inhibitors (PIs), alkylator chemotherapies, steroids, monoclonal antibodies, and options are still expanding. To profile sensitivity to the wide array of options, we developed a rapid functional assay of primary MM cells, distinct from prior assays by forgoing CD138+ cell selection to optimize MM cell viability and allow measurement of monoclonal antibody activity. Methods: Myeloma drug sensitivity testing (My-DST) was performed on patient BM aspirates to measure the MM sensitivity in heterogeneous mononuclear cell (MNC) mixtures. My-DST is performed by culturing samples with a 7-drug panel consisting of bortezomib (Bor), carfilzomib (Car), lenalidomide (Len), pomalidomide (Pom), dexamethasone (Dex), cyclophosphamide (Cy, metabolite 4HC), and daratumumab (Dara) for 48h, followed by multiparameter, high-throughput flow cytometry to measure the surviving MM population. All conditions are performed in triplicate and in parallel, clonality verified by kappa/lambda staining. Results: MM cell viability was substantially better in unselected MNC cultures than CD138-selected cultures (data not shown). Dose-response was evaluated for each agent in MNC cultures to select single active concentrations for high throughput screening (data not shown). Single concentration profiling was completed on 55 patient samples from diagnosis (n = 24), first relapse (n = 11) and after multiple relapses (n = 20) with 2.5 nM PIs, 10 µM IMiDs, 1 µM Dex, 3.75 µM Cy and 20 nM Dara. Each was scored as sensitive or resistant using a cutoff of 80% surviving MM cells normalized to untreated controls (Fig 1A). Mild inherent resistance to varied tested agents was evident at diagnosis and progressively increased with lines of therapy (LOT) until multidrug resistance predominated (Fig 1B). My-DST displayed significant correlation for LOT with increasing resistance to Bor (r = 0.27, P = 0.046), Car (r = 0.27, P = 0.044), Len (r = 0.44, P = 0.0006) and Pom (r = 0.43, P = 0.0009), but not Dara (r = 0.27, P = 0.10), Cy (r = 0.17, P = 0.24) and Dex (r = 0.032, P = 0.82) (Fig 1C). The ex vivo sensitivity for Bor and Car were highly correlated (r = 0.67, P < 0.0001) from sensitive to both to resistant to both, but still differential results favoring Bor or Car were observed (Fig 1D). The ex vivo results for the tested IMiDs were even more highly correlated (r = 0.79, P < 0.0001), with fewer samples showing greater sensitivity differentially to Len or Pom (data not shown). In patients at diagnosis, the ex vivo sensitivity to all drugs given during treatment led to significantly deeper clinical responses than those with resistance to ≥1 drugs received (Fig 1E). As an example of potential clinical utility of My-DST, the test for #847 at diagnosis identified sensitivity to PIs and resistance to IMiDs, which correlated with inadequate response to Len/Dex, followed by complete response (CR) after change to Bor/Dex (Fig 1F). Dara had ex vivo activity in most naïve patients, but patients already exposed showed significant acquired resistance (Fig 1G). We also found that CD38 expression was significantly lower in patients that showed ex vivo Dara resistance (Fig 1H). The ex vivo Dara activity was partially blocked by inhibiting macrophages with clodronate-liposomes (CL) (Fig 1I). Conclusion: My-DST on unselected MM BM aspirates with minimal perturbations ex vivo was clinically predictive of resistance to IMiDs and PIs. For the first time, this approach is able to measure the antibody-mediated cytotoxicity of daratumumab in clinical MM samples. Our data indicate that personalized regimens based on rapidly obtained ex vivo sensitivity may lead to optimized depth of clinical response. We next plan to translate this approach into clinical use in an effort to improve patient outcomes and avoid wasted time, money and complications that may arise from the use of ineffective drugs. Disclosures Mark: Janssen, Takeda, Celgene, Amgen: Consultancy; BMS, Celgene: Research Funding; Celgene: Honoraria.

Effective targeting of the acute myeloid leukemia (AML) leukemia stem cell (LSC) population may allow for deep, durable remissions and curative potential. In older, newly diagnosed AML patients who are not candidates for induction, venetoclax + azacitidine (aza) targets specific metabolic vulnerabilities of LSCs, resulting in very promising clinical outcomes. In our single institution experience treating 45 previously untreated AML patients with venetoclax + aza both in the context of the multi-institutional study NCT02203773 (N=33) and with off-label use (N=12), 36/45 (80%) achieved a complete remission (CR) or CR with incomplete count recovery (CRi). In the relapsed/refractory (R/R) setting, the efficacy of venetoclax + aza has been reported to be significantly worse. In our single-institution off-label experience (N=7), only 1/7 (14%) R/R patients had a CR/CRi (p=0.005 compared to the untreated group). R/R and untreated patients had similar baseline characteristics, although more R/R patients had an antecedent hematological disorder (Table 1). Multivariate analysis showed cytogenetic risk and R/R disease as the sole predictors of response to venetoclax + aza (Table 2). In light of existing data regarding biological changes that occur in LSCs after treatment and subsequent relapse, we aimed to determine whether laboratory analysis of LSCs from patients treated with venetoclax + aza would show differential sensitivity to this therapy in the up-front vs R/R setting that could help to explain the different clinical activity. We have previously shown that LSCs from untreated patients are uniquely reliant on oxidative phosphorylation (OXPHOS), and that venetoclax + aza targets LSCs by decreasing OXPHOS. Therefore, we tested the hypothesis that inferior responses of R/R patients to venetoclax + aza are due to changes in OXPHOS regulation in relapsed LSCs. LSCs were defined as cells bearing relatively low levels of reactive oxygen species (ROS-low), an effective means of enriching primary human LSCs. We found that in contrast with untreated patients, venetoclax + aza does not decrease viability or OXPHOS in LSCs from R/R patients (Fig1). Furthermore, R/R LSCs had altered fatty acid metabolism that contributed to these OXPHOS differences, with increased flux of fatty acids into the TCA cycle (Fig 2). In addition, R/R samples compensated for the metabolic perturbations that occurred upon exposure to venetoclax + aza through upregulation of fatty acid uptake and metabolism into the TCA cycle (Fig 3). Fatty acid metabolism is controlled by multiple genes and pathways. Integral to its activity is the gene Carnitine Palmitolytransferase 1 (CPT1), due to its pivotal role in the beta-oxidation of long chain fatty acids. Investigation of the Cancer Genome Atlas AML dataset reveals higher expression of CPT1 leads to significantly worse overall survival, suggesting increased fatty acid metabolism may drive a more resistant LSC population in R/R AML patients. We also found elevated baseline levels of CPT1 in patients who progressed on venetoclax + aza compared to those that had long term remissions (not shown). Therefore we utilized the CPT1 inhibitor etomoxir to block fatty acid metabolism. We found addition of etomoxir to cultures of R/R LSCs rescued the ability of venetoclax + aza to decrease OXPHOS and re-sensitized R/R LSCs to venetoclax + aza (Fig 4). To prove that this novel regimen targets functionally-defined R/R LSCs we performed ex vivo treatment followed by xenotransplantation of R/R patient specimens, which showed that upon etomoxir addition, engraftment potential is significantly decreased over venetoclax + aza alone (not shown). Therefore we propose a novel mechanism for the increased resistance of R/R AML patients to venetoclax + aza involving altered energy metabolism. We find increased fatty acid metabolism in R/R patient specimens, and targeting this pathway using the CPT1 inhibitor etomoxir leads to sensitization to venetoclax + aza and rescued targeting of OXPHOS, allowing for LSC eradication. Disclosures Pollyea: Pfizer: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Argenx: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; AbbVie: Consultancy, Research Funding; Agios: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Gilead: Consultancy; Celyad: Consultancy, Membership on an entity's Board of Directors or advisory committees; Curis: Membership on an entity's Board of Directors or advisory committees; Karyopharm: Membership on an entity's Board of Directors or advisory committees.

Increasing evidence indicates that malignant stem cells are important for the pathogenesis of acute myelogenous leukemia (AML) and represent a reservoir of cells that drive the development of AML and relapse. Therefore, new treatment regimens are necessary to prevent relapse and improve therapeutic outcomes. Previous studies have shown that the sesquiterpene lactone, parthenolide (PTL), ablates bulk, progenitor, and stem AML cells while causing no appreciable toxicity to normal hematopoietic cells. Thus, PTL must evoke cellular responses capable of mediating AML selective cell death. Given recent advances in chemical genomics such as gene expression-based high-throughput screening (GE-HTS) and the Connectivity Map, we hypothesized that the gene expression signature resulting from treatment of primary AML with PTL could be used to search for similar signatures in publicly available gene expression profiles deposited into the Gene Expression Omnibus (GEO). We therefore devised a broad in silico screen of the GEO database using the PTL gene expression signature as a template and discovered 2 new agents, celastrol and 4-hydroxy-2-nonenal, that effectively eradicate AML at the bulk, progenitor, and stem cell level. These findings suggest the use of multicenter collections of high-throughput data to facilitate discovery of leukemia drugs and drug targets.

In this study, we characterized nuclear factor κB (NF-κB) subunit DNA binding in chronic lymphocytic leukemia (CLL) samples and demonstrated heterogeneity in basal and inducible NF-κB. However, all cases showed higher basal NF-κB than normal B cells. Subunit analysis revealed DNA binding of p50, Rel A, and c-Rel in primary CLL cells, and Rel A DNA binding was associated with in vitro survival (P = .01) with high white cell count (P = .01) and shorter lymphocyte doubling time (P = .01). NF-κB induction after in vitro stimulation with anti-IgM was associated with increased in vitro survival (P < .001) and expression of the signaling molecule ZAP-70 (P = .003). Prompted by these data, we evaluated the novel parthenolide analog, LC-1, in 54 CLL patient samples. LC-1 induced apoptosis in all the samples tested with a mean LD50 of 2.8 μM after 24 hours; normal B and T cells were significantly more resistant to its apoptotic effects (P < .001). Apoptosis was preceded by a marked loss of NF-κB DNA binding and sensitivity to LC-1 correlated with basal Rel A DNA binding (P = .03, r2 = 0.15). Furthermore, Rel A DNA binding was inversely correlated with sensitivity to fludarabine (P = .001, r2 = 0.3), implicating Rel A in fludarabine resistance. Taken together, these data indicate that Rel A represents an excellent therapeutic target for this incurable disease.

Diffuse large B-cell lymphoma (DLBCL) is an aggressive and the most common type of non-Hodgkin lymphoma. Despite recent advances in treatment, less than 50% of the patients are cured with current multiagent chemotherapy. Abnormal NF-κB activity not only contributes to tumor development but also renders cancer cells resistant to chemotherapeutic agents. Identifying and targeting signaling molecules that control NF-κB activation in cancer cells may thus yield more effective therapy for DLBCL. Here, we show that while overexpression of protein kinase C–associated kinase (PKK) activates NF-κB signaling in DLBCL cells, suppression of PKK expression inhibits NF-κB activity in these cells. In addition, we show that NF-κB activation induced by B cell–activating factor of tumor necrosis factor family (BAFF) in DLBCL cells requires PKK. Importantly, we show that knockdown of PKK impairs the survival of DLBCL cells in vitro and inhibits tumor growth of xenografted DLBCL cells in mice. Suppression of PKK expression also sensitizes DLBCL cells to treatment with chemotherapeutic agents. Together, these results indicate that PKK plays a pivotal role in the survival of human DLBCL cells and represents a potential target for DLBCL therapy.

Leukemia stem cells (LSCs) are thought to play a central role in the pathogenesis of acute leukemia and likely contribute to both disease initiation and relapse. Therefore, identification of agents that target LSCs is an important consideration for the development of new therapies. To this end, we have previously demonstrated that the naturally occurring compound parthenolide (PTL) can induce death of human LSCs in vitro while sparing normal hematopoietic cells. However, PTL has relatively poor pharmacologic properties that limit its potential clinical use. Consequently, we generated a family of PTL analogs designed to improve solubility and bioavailability. These studies identified an analog, dimethylamino-parthenolide (DMAPT), which induces rapid death of primary human LSCs from both myeloid and lymphoid leukemias, and is also highly cytotoxic to bulk leukemic cell populations. Molecular studies indicate the prevalent activities of DMAPT include induction of oxidative stress responses, inhibition of NF-κB, and activation of p53. The compound has approximately 70% oral bioavailability, and pharmacologic studies using both mouse xenograft models and spontaneous acute canine leukemias demonstrate in vivo bioactivity as determined by functional assays and multiple biomarkers. Therefore, based on the collective preclinical data, we propose that the novel compound DMAPT has the potential to target human LSCs in vivo.

Recent studies have described malignant stem cells as central to the initiation, growth, and potential relapse of acute and chronic myelogenous leukemia (AML and CML). Because of their important role in pathogenesis, rare and biologically distinct leukemia stem cells (LSCs) represent a critical target for therapeutic intervention. However, to date, very few agents have been shown to directly target the LSC population. The present studies demonstrate that parthenolide (PTL), a naturally occurring small molecule, induces robust apoptosis in primary human AML cells and blast crisis CML (bcCML) cells while sparing normal hematopoietic cells. Furthermore, analysis of progenitor cells using in vitro colony assays, as well as stem cells using the nonobese diabetic/severe combined immunodeficient (NOD/SCID) xenograft model, show that PTL also preferentially targets AML progenitor and stem cell populations. Notably, in comparison to the standard chemotherapy drug cytosine arabinoside (Ara-C), PTL is much more specific to leukemia cells. The molecular mechanism of PTL-mediated apoptosis is strongly associated with inhibition of nuclear factor κ B (NF-κB), proapoptotic activation of p53, and increased reactive oxygen species (ROS). On the basis of these findings, we propose that the activity of PTL triggers LSC-specific apoptosis and as such represents a potentially important new class of drugs for LSC-targeted therapy.

Major strides have been made in our understanding of the molecular basis of adult and pediatric leukemias. More than one hundred disease alleles have been identified and characterized in cell culture and murine models of leukemia. In some instances, molecularly targeted therapies have been developed based on these insights that are currently in clinical trials, such as small molecule inhibitors of FLT3. In addition, it has recently been appreciated that, as with normal hematopoiesis, there is a hierarchical organization among leukemic cells that includes a rare population of leukemic stem cells that have properties of self-renewal. Understanding the characteristics of these leukemic stem cells may provide new insights into leukemia therapies that target self-renewal pathways. In Section I, Dr. Craig Jordan reviews the data that supports the existence of a “leukemia stem cell.” He provides an overview of the functional properties of leukemic stem cells, their relationship to hematopoietic stem cells, and the relevance of leukemic stem cells in other human malignancies including solid tumors. He briefly discusses what is known of the pathways that regulate properties of self-renewal. Dr. Gary Gilliland provides an overview of the genetics of adult leukemias in Section II and ongoing genome-wide strategies for discovery of new disease alleles. He describes the clinical and therapeutic implications of these findings and provides examples of bench-to-bedside translation of molecularly targeted therapies for AML, including the use of FLT3 inhibitors. In Section III, Dr. Carolyn Felix reviews recent advances in our understanding of the genetics and therapy of pediatric leukemias. She provides an overview of leukemias that are common in pediatric malignancies but rarely observed in adults, including the TEL-AML1 (ETV6-RUNX1) fusion associated with pediatric B-cell ALL, the OTT-MAL fusion associated with infant megakaryoblastic leukemia, PTPN11 mutations in juvenile myelomonocytic leukemia, and MLL fusion genes in leukemogenesis, among others.

Human acute myelogenous leukemia (AML) is thought to arise from a rare population of malignant stem cells. Cells of this nature, herein referred to as leukemic stem cells (LSCs), have been documented for nearly all AML subtypes and appear to fulfill the criteria for stem cells in that they are self-renewing and give rise to the cells found in many leukemic populations. Because these cells are likely to be critical for the genesis and perpetuation of leukemic disease, the present studies sought to characterize unique molecular properties of the LSC population, with particular emphasis on the transcription factor, nuclear factor-κB (NF-κB). Previous experiments have shown that unstimulated human CD34+ progenitor cells do not express NF-κB. In contrast, primary AML CD34+ cells display readily detectable NF-κB activity as assessed by electrophoretic mobility shift assay and gene expression studies. Furthermore, detailed analyses of enriched AML stem cells (CD34+/CD38−/CD123+) indicate that NF-κB is also active in the LSC population. Given the expression of NF-κB in leukemic, but not normal primitive cells, the hypothesis that inhibition of NF-κB might induce leukemia-specific apoptosis was tested by treating primary cells with the proteasome inhibitor MG-132, a well-known inhibitor of NF-κB. Leukemic CD34+/CD38− cells displayed a rapid induction of cell death in response to MG-132, whereas normal CD34+/CD38− cells showed little if any effect. Taken together, these data indicate that primitive AML cells aberrantly express NF-κB and that the presence of this factor may provide unique opportunities to preferentially ablate LSCs.

Optimization of mobilization, harvest, and transduction of hematopoietic stem cells is critical to successful stem cell gene therapy. We evaluated the utility of a novel protocol involving Flt3-ligand (Flt3-L) and granulocyte colony-stimulating factor (G-CSF) mobilization of peripheral blood stem cells and retrovirus transduction using hematopoietic growth factors to introduce a reporter gene, murine CD24 (mCD24), into hematopoietic stem cells in nonhuman primates. Rhesus macaques were treated with Flt3-L (200 microgram/kg) and G-CSF (20 microgram/kg) for 7 days and autologous CD34(+) peripheral blood stem cells harvested by leukapheresis. CD34(+) cells were transduced with an MFGS-based retrovirus vector encoding mCD24 using 4 daily transductions with centrifugations in the presence of Flt3-L (100 ng/mL), human stem cell factor (50 ng/mL), and PIXY321 (50 ng/mL) in serum-free medium. An important and novel feature of this study is that enhanced in vivo engraftment of transduced stem cells was achieved by conditioning the animals with a low-morbidity regimen of sublethal irradiation (320 to 400 cGy) on the day of transplantation. Engraftment was monitored sequentially in the bone marrow and blood using both multiparameter flow cytometry and semi-quantitative DNA polymerase chain reaction (PCR). Our data show successful and persistent engraftment of transduced primitive progenitors capable of giving rise to marked cells of multiple hematopoietic lineages, including granulocytes, monocytes, and B and T lymphocytes. At 4 to 6 weeks posttransplantation, 47% +/- 32% (n = 4) of granulocytes expressed mCD24 antigen at the cell surface. Peak in vivo levels of genetically modified peripheral blood lymphocytes approached 35% +/- 22% (n = 4) as assessed both by flow cytometry and PCR 6 to 10 weeks posttransplantation. In addition, naive (CD45RA(+) and CD62L(+)) CD4(+) and CD8(+) cells were the predominant phenotype of the marked CD3(+) T cells detected at early time points. A high level of marking persisted at between 10% and 15% of peripheral blood leukocytes for 4 months and at lower levels past 6 months in some animals. A cytotoxic T-lymphocyte response against mCD24 was detected in only 1 animal. This degree of persistent long-lived, high-level gene marking of multiple hematopoietic lineages, including naive T cells, using a nonablative marrow conditioning regimen represents an important step toward the ultimate goal of high-level permanent transduced gene expression in stem cells.

We have examined the ability of recombinant adenoviral vectors to transduce human hematopoietic cells. Our findings indicate that adenovirus readily infects a large proportion of CD34+ cells. Using adenovirus vectors that transduce either a lacZ or an alkaline phosphatase reporter gene, we observed up to 45% of total CD34+ cells infected. Upon more detailed analysis, we observed comparable levels of transduction for CD34+/CD38- cells and for CD34+ cells in G(zero) phase of the cell cycle. Importantly, exposure to adenovirus resulted in negligible levels of toxicity as assayed by propidium iodide staining and colony-forming ability. Using adenovirus vectors, we also describe a model system for regulated gene expression in early hematopoietic tissues. CD34+ cells were simultaneously infected with two viruses, one carrying a TetR/VP16 transactivator (tTA) and the second carrying a tTA- dependent lacZ reporter gene. Using this approach, beta-gal expression was only observed upon coinfection with the transactivator vector. In addition, as shown previously (Gossen and Bujard, Proc Natl Acad Sci USA 89:5547, 1992), tetracycline was able to inhibit tTA mediated induction, thereby providing an effective means to regulate expression of the reporter gene. We conclude that recombinant adenovirus is an effective vehicle for transiently expressing genes in primitive human hematopoietic cells.

Background : Post Transplant Lymphoproliferative disorder (PTLD) represents a distinct and rare complication following solid organ transplantation (SOT). Insight into the biology of this disorder is limited to retrospective reviews and case series. In one of the first reports for post-transplant Diffuse Large B cell Lymphoma (PT-DLBCL) cases, we demonstrated a higher incidence and improved outcomes in PT-DLBCL non-germinal center B-cell (non-GCB) subtype compared to PT-DLBCL germinal center B-cell (GCB) Subtype. Published data suggests immunocompetent DLBCL non-GCB subtypes are less common and fare worse than immunocompetent GCB DLBCL. The reason for this unexpected finding in our PT-DLBCL pts is not fully understood. Recently Kiyasu and colleagues demonstrated that PD-L1 overexpression was significantly associated with non-GCB DLBCL, EBV virus positivity and poor prognosis in immunocompetent DLBCL samples. Therefore based on this we hypothesized that PT-DLBCL non-GCB subtype may have negative PD-L1 overexpression thus possibly accounting for improved outcomes compared to their immunocompetent counterparts. Hence we sought to test PD-L1 expression in our samples with PT-DLBCL. Methods: With IRB approval, we retrospectively identified PT- DLBCL patients treated at the University of Colorado between Jan 1989 to April 2015. We retrieved formalin fixed paraffin embedded PT-DLBCL tissue specimens and determined cell of origin by the Hans Algorithm. We assessed PD-L1 expression by immunohistochemistry. PD-L1 positive PT-DLBCL was defined as 30% of more of the lymphoma cells showing distinct membranous and or cytoplasmic staining. In addition EBER-ISH was performed to assess EBV status. Results: 86 adult SOT pts with PTLD were treated at our institution. 75 of 86 pts (87%) had monomorphic histology. Among monomorphic PTLD, 64% (48 of 75) had DLBCL. The median age at transplantation was 49.5 yrs (5-74 yrs). Median time from SOT to PTLD was 37 mos (1.4-499). The most common transplanted organ included kidney (40%), liver (38%), lung (13%) and heart (9%). 31% had early PTLD (12mos of SOT). 60% were EBV positive. 77% with early PTLD and 49% with late PTLD were EBV positive. Due to a paucity of archived tissue blocks, IHC staining was applied to 32/48 samples with DLBCL. Non-GCB subtype was identified in 75% (24/32) samples and GCB subtype in 25% (8/32) samples. Of the 48 pts with PT-DLBCL histology, PD-L1 stain was performed on 18 samples. Of the 18 PT-DLBCL samples, 77% (14/18) had non-GCB subtype and 16% (3/18) had GCB subtype. PD-L1 expression was negative in 78% (11/14) and positive in 22% (3/14) of non-GCB DLBCL samples. PD-L1 expression was negative in 100% (3/3) of GCB DLBCL samples. The sample size was too small to effectively describe the survival experience of pt subsets. Using Fisher's exact test we found no evidence to support an association between EBV Status and PDL1 expression (p-value 0.316). Conclusions: We previously reported in our consecutive series of PTLD after SOT an increased incidence and improved survival in pts with PT-DLBCL non-GCB subtype (ASH 2015) compared to PT-DLBCL GCB subtype. The reason for this is not fully understood. However, our limited series reveals that a majority of pts with PT- DLBCL non-GCB subtype was negative for PD-L1 overexpression. This might explain the improved outcomes in the PT-DLBCL non-GCB population. Despite a small sample size it is also interesting to note that 100% pts with PT-DLBCL GCB subtype were negative for PD-L1 overexpression. In the era of immunotherapy further studies in larger patient cohorts are warranted in order to understand the unique biology and outcomes of PT-DLBCL since it may have therapeutic implications. Disclosures Pollyea: Celgene: Other: advisory board, Research Funding; Ariad: Other: advisory board; Alexion: Other: advisory board; Pfizer: Other: advisory board, Research Funding; Glycomimetics: Other: DSMB member. Kamdar:Seattle Genetics: Speakers Bureau.

We recently reported that adipose tissue functions as a reservoir for leukemia stem cells (LSCs) using a murine model of leukemia (Ye et al., Cell Stem Cell, 2016). Intriguingly, the presence of leukemic cells in adipose tissue induces increased lipolysis and the release of free fatty acids (FFA) which in turn fuels the growth of LSCs. Further, adipose tissue protects resident LSCs from chemotherapy. These findings indicate that the unique characteristics of adipose tissue may provide key insights on the growth and survival of leukemic cells. Thus, in the present study we focus on the endocrine function of adipose tissue and explore its role in leukemia development. First, to evaluate secreted factors, we applied adipokine arrays which detect 38 adipokines to leukemia serum collected at different stages of leukemia pathogenesis. Interestingly, we observed a significantly elevated level of serum IGFBP1 as soon as leukemic disease became evident at low levels (~0.5%) in marrow. IGFBP1 increased to levels approximate 200X normal at peak stages of disease burden. ELISA analyses further confirmed these observations. IGFBP1 is normally considered as a liver-derived protein. However, we did not find any changes of IGFBP1 expression in leukemic liver. Rather, we observed a significant increase in the expression of IGFBP1 in adipose tissue. IGFBP1 in conditioned medium (CM) from leukemic gonadal adipose tissue (GAT) is approximate 100X higher than naive GAT. Together, these findings suggest adipose tissue-derived IGFBP1 contributes to the increased serum IGFBP1 we detected in leukemic animals. We next examined the role of IGFBP1 in leukemia development using a neutralizing antibody. Treatment of experimental animals with anti-IGFBP1 antibody significantly decreased leukemic burden in GAT (~40% IgG treated VS. 20% Anti-IGFBP1 treated) while bone marrow (BM) and spleen leukemic engraftment remained unchanged (~40%). Further, less atrophy of adipose tissue as well as less body weight loss was seen in the IGFBP1 neutralizing antibody treated group. Consistent with this observation, serum FFA level was also reduced, suggesting less lipolysis in the IGFBP1 antagonized group. Together, these results indicate that IGFBP1 is involved in the regulation of leukemic cells homing to adipose tissue and consequently leukemia-induced lipolysis. Next we explored the mechanisms for the increased expression of IGFBP1 in adipose tissue. Studies have shown that both FGF21 and hypoxia induce IGFBP1 expression. We did not find any changes of FGF21 expression in adipose tissue or in liver, suggesting FGF21 was not involved in IGFBP1 regulation in our system. In contrast, we observed a five-fold elevation in IGFBP1 mRNA in primary adipose tissue cultured under hypoxic conditions. Studies have shown that both BM and spleen in leukemia mice are already hypoxic in early stages of leukemic development (Benito et al., Plos One, 2011). Thus, we hypothesize that tissue hypoxia may at least partially regulate IGFBP1 in leukemia. Ongoing experiments are testing this hypothesis. Additionally, inflammatory cytokines have been reported to increase IGFBP1 expression. We previously reported significantly increased levels of inflammatory cytokines in the adipose tissue of leukemic mice including TNF-α, IL1 and CSF2 (1.5X normal adipose tissue). Therefore, we hypothesize the local inflammatory cytokine production may also contribute to increased expression of IGFBP1, a theory that is also currently under investigation. IGFBP1 has recently been reported to activate osteoclasts and thus promotes bone metabolism (Wang et al., Cell Metabolism, 2015). Studies have found an increased number of osteoclasts and thus bone loss in our leukemic model (Frisch et al., Blood, 2012). We hypothesize that adipose tissue-derived IGFBP1 contributes to bone loss in leukemic mice. Ongoing experiments are testing whether antagonization of IGFBP1 in leukemic mice will rescue bone loss. Collectively, these findings suggest that adipose tissue-derived IGFBP1 facilitates progression of leukemia through regulation of adipose tissue lipolysis and may promote bone marrow colonization by leukemia cells through activation of osteoclasts. Disclosures No relevant conflicts of interest to declare.

IL-1R3 is the co-receptor required for signaling of IL-1β, IL-1α, IL-33 and IL-36α, β and γ. The naturally occurring IL-1 receptor antagonist (anakinra), is used to block IL-1β, and IL-1α and clinical trials have demonstrated a reduction in progression to multiple myeloma as well as increased overall survival in stage IV pancreatic cancer using anakinra. In addition, a monoclonal antibody that targets IL-1α increases overall survival in metastatic colorectal cancer. However, blocking IL-1R3 would reduce not only IL-1β and IL-1α but also IL-33 as well as IL-36α, β and γ. The data described below reveal the broad efficacy of anti-IL-1R3. In the present studies, monoclonal humanized anti-IL-1R3 antibodies were studied in the mixed leukocyte reaction (MLR), in peripheral mononuclear cells (PBMC) stimulated with LPS, heat-killed Candida albicans or anti-CD3/antiCD28 as well as in THP-1 cells, a cell line derived from a patient with acute myeloid leukemia (AML). In the MLR, anti-IL-1R3 at 5 µg/mL reduced IFNγ by 81% and IL-6 by 48% compared to 36% by anakinra. In LPS-stimulated PBMCs, IL-6 was reduced by 40% and 58% with anakinra and anti-IL-1R3 respectively. Using heat-killed Candida, the suppression of IL-6 production by anti-IL-1R3 was up to 70%, comparable to anakinra. Immuno-stimulation using anti-CD3/CD28 resulted in a similar inhibition capacity for anakinra (42%) and anti-IL-1R3 (45%) on IL-6 production also. Since neutralizing anti-IL-1α antibodies have increased overall survival in metastatic colorectal cancer, we also measured the effect of anti-IL-1R3 on intracellular levels of IL-1α. In PBMCs stimulated with LPS, the levels of IL-1α were reduced by 35% in cells cultured with anakinra at 10 µg/mL and by 66% in cells exposed to anti-IL-1R3 at same concentration. Activating PBMCs with anti-CD3/anti-CD28 led to a reduction in IL-1α by anakinra of 17%, whereas anti-IL-1R3 suppressed by 29%. We next examined the effect on THP-1 cells as a model for blocking IL-1 family members in AML. THP-1 were differentiated with phorbol myristate acetate (PMA) for 3 hours, washed and subsequently rested for 3 days. The cells were then stimulated with LPS for 3 hours and anti-IL-1R3 was added for 1 hour followed by inflammasome activation with ATP. The reduction in secretion of IL-1β was here 30% at 5 µg/mL of anti-IL-1R3. Collectively, these studies indicate that antibody blockade of IL-1R3 is effective in reducing cytokines from primary cells using in-vitro models of organ rejection, infection and immunostimulation. In THP-1 cells, the reduction in inflammasome-dependent IL-1β suggests that anti-IL-1R3 can be used to treat acute myeloid leukemia, or progression to multiple myeloma. Since anti-IL-1R3 inhibits the signaling through IL-1R1, both IL-1α and IL-1β are targeted by this antibody. Suggesting a future role for this antibody in not only AML, but also other cancer types dominated by IL-1 mediated inflammation, such as metastatic colorectal and pancreatic cancer. Disclosures Fischer: MAB Discovery GmbH: Employment. Pollyea:Ariad: Other: advisory board; Pfizer: Other: advisory board, Research Funding; Glycomimetics: Other: DSMB member; Alexion: Other: advisory board; Celgene: Other: advisory board, Research Funding.

NADPH dependent oxidase 2 (NOX2) is the founding member of a family of multimeric, oxido-reductase enzymes that catalyze the production of superoxides by transferring a single electron from the cofactor NADPH to molecular oxygen. It is primarily utilized in neutrophils and macrophages to generate copious amount of reactive oxygen species (ROS) to facilitate the neutralization of engulfed particulates during phagocytosis. In sharp contrast to this specialized function however, recent evidence implies a non-phagocytic role for NADPH oxidases in which physiologic levels of ROS generated by these enzymes modulate key signaling proteins and transcription factors to exert profound biological effects. Based on this information we decided to investigate the potential role of NOX2 in normal and leukemic stem cells. Using transgenic NOX2 knock out mice, genetically defined murine models of myeloid leukemia and primary human acute myeloid leukemia (AML) specimens, we show that NOX2 is critical for the proper function of normal and malignant hematopoietic stem cells. In silico analysis using published transcriptional profiles of hematopoietic populations revealed that multiple subunits of the NOX2 complex are expressed at low levels in hematopoietic stem cells (HSCs) and at relatively higher levels in multipotent progenitors (MPPs). Next, we characterized the different hematopoietic compartments from age and sex matched wild type (WT) and transgenic NOX2 knock out (KO) mice. Our studies revealed that in the bone marrow of KO mice, a subset of multipotent progenitor populations (MPP2 & MPP3), which often have biased myelo-erythroid output are markedly expanded relative to their wild type counterparts. Consistently, we found increased levels of granulocytes and monocytes in the peripheral circulation of NOX2 KO mice. To test whether NOX2 has a functional, biological role in the self-renewal of HSCs, we performed competitive transplantation assays using equal numbers of whole BM cells from WT and KO mice to co-repopulate lethally irradiated hosts. Analysis of engrafted mice showed that the contribution from NOX2 KO HSCs was severely compromised in all lineages and developmental stages of hematopoiesis examined. Collectively, these results suggest a critical biological role for NOX2 in maintaining the quiescence and long term self-renewal of HSCs. Similar to normal hematopoiesis, we found out that NOX2 is also widely expressed by functionally defined leukemic stem cells in a murine model of myeloid leukemia generated by expressing the oncogenic translocations BCR-ABL and NUP98-HOXA9. To evaluate the role of NOX2 in leukemogenesis, we established the BCR-ABL/NUP98-HOXA9 model using primitive cells derived from either WT or KO. Intriguingly, NOX2 KO leukemic cells generated a much less aggressive disease upon transplantation into primary and subsequently into secondary recipients. Furthermore, leukemic cells in which NOX2 is suppressed displayed aberrant mitotic activity and altered developmental potential marked by loss of quiescence, enhanced entry into cycle and terminal differentiation. To gain mechanistic insight into the observed phenotype, we isolated leukemic stem cells and performed whole genome expression analysis. The data showed that deficiency of NOX2 leads to downregulation of the cell cycle inhibitor CDKN2C (p18) and robust activation of the granulocyte fate determining transcription factor CEBPε. Thus we conclude that loss of NOX2 impacts leukemogenesis through rewiring of the cell cycle machinery and developmental programs in leukemic stem cells. Finally, we found that in CD34+ primary human AML cells, NOX2 and the other subunits of the complex are abundantly expressed. Furthermore, pharmacologic inhibition of NOX2 with VAS2870, a selective NADPH oxidase inhibitor, reduced the level of ROS and limited the in vitro proliferation and survival of leukemic cells. Next we genetically suppressed the expression of NOX2 in primary human AML cells using sh-RNAs and transplanted these cells into immune compromised mice. Consistent with the murine leukemia, NOX2 knocked down AML cells failed to engraft and expand in vivo. Taken together, our results firmly establish a hitherto unrecognized, prominent regulatory role for NOX2 in the biology of normal and malignant hematopoietic stem cells and imply a potential therapeutic opportunity that can get exploited to treat AML. Disclosures Pollyea: Celgene: Other: advisory board, Research Funding; Ariad: Other: advisory board; Pfizer: Other: advisory board, Research Funding; Glycomimetics: Other: DSMB member; Alexion: Other: advisory board.