ALBERT

Description: Key Points Jarid1b knockdown promotes enhanced HSC activity.

Description: Background: Venetoclax (VEN) is a potent small molecule BH3-mimetic drug that selectively targets the prosurvival protein BCL-2 and has an emerging role for treatment in Acute Myeloid Leukaemia (AML). This ongoing Phase 1b study aims to evaluate the optimal dose, safety and efficacy of VEN in combination with modified intensive chemotherapy in fit, elderly patients with AML who have not received prior induction chemotherapy. Here we present updated data including molecular correlates of response and treatment outcomes. Methods: Eligibility: Patients were enrolled with AML (excluding APL), age ≥65 years (≥60 years if monosomal karyotype), ECOG 0-1 and adequate organ function. Prior HMA/LDAC was permitted after a 14-day washout. Treatment: 1) Dose escalation cohorts comprised VEN 50mg (Cohort A), 100mg (B), 200mg (C), 400mg (D) or 600mg (E); 2) a 7-day VEN pre-phase incorporating dose ramp-up to minimise tumour lysis risk followed by a 7-day overlap with chemotherapy (day -6 to +7); 3) attenuated induction chemotherapy with cytarabine 100mg/m2/day IVI d1-5 staggered with idarubicin 12mg/m2 IV d2-3. For patients in remission, 4 cycles of consolidation were administered, comprising 14 days of VEN (day -7 to +7) with bolus cytarabine (100mg/m2/day IV d+1-2) and idarubicin (12mg/m2 IV d+1) followed by 7 cycles of VEN monotherapy maintenance; planned to commence q28d. Antifungal azoles were avoided during VEN exposure. The first patient was enrolled 17JUL2016. Molecular studies- Next generation sequencing using a 54-gene TruSight myeloid panel was performed on baseline bone marrow samples. FLT3-ITD testing was performed by capillary electrophoresis. RT-qPCR was used to detect minimal residual disease (MRD) in NPM1-mutated cases (assay sensitivity: 10-5-10-6). Results: The data cut-off date was 13JUL2018. 44 patients were enrolled with 3 patients still in the induction cycle. Median age was 72 years (range 63-80 years; 23% ≥75 years), 66% of patients were male, 41% had secondary AML and 38% prior hypomethylating agent (HMA) exposure. Main grade ≥3 non-haematological adverse events during induction were febrile neutropenia (56%), sepsis (32%), rapid atrial fibrillation (15%), diarrhoea (12%), nausea (10%) and localised infection (10%). No clinical TLS was observed. One hematologic dose-limiting toxicity was reported in cohort E (VEN 600mg), but a maximum tolerated dose has not been reached. We observed progressive delay in platelet count recovery as patients proceeded with each cycle of therapy (median time to platelets ≥50x109/L was 25d in induction, 37d in consolidation 1 and 57d in consolidation 4). Treatment-related mortality was 7% in induction. 18/38 (47%) had ≥30% relative reduction in bone marrow (BM) blasts after 7 days of pre-phase VEN monotherapy, with no correlation between VEN dosage and the degree of BM blast reduction. Overall CR/CRi rate was 71%. CR/CRi was achieved in 95% of patients with de novo AML (vs 42% in secondary/therapy-related AML). Response rates were 88% in intermediate vs 46% in adverse cytogenetic risk AML. Responses in patients with prior HMA therapy were 43% and 84% for previously untreated patients. Molecular data were available for 41 patients. 98% had ≥1 mutation detected. (Figure 1) Responses in response-evaluable patients were most common in NPM1 mutant AML (100%; n=7), followed by RUNX1 (90%; n=11), RAS (90%; n=10) and IDH (89%; n=9). Lowest responses were observed in TP53 mutant AML (33%; n=9). NPM1 MRD assessment was performed on 6 patients, of which 5 (83%) achieved NPM1 MRD negativity. (Figure 2) Updated response, response duration and survival data will be presented, along with molecular characteristics of relapse. Conclusion: To date, VEN up to 600mg in combination with 5+2 induction chemotherapy is tolerable in fit elderly patients with AML. Initial response rates 〉80% were observed in de novo AML, as well as NPM1, RUNX1, RAS and IDH mutant AML, whereas responses were lower in patients with prior HMA exposure, adverse karyotype, secondary and TP53 mutant AML. Molecular determinants of relapse require further study. Disclosures Wei: Servier: Consultancy, Honoraria, Other: Advisory committee, Research Funding; Amgen: Honoraria, Other: Advisory committee, Research Funding; Novartis: Honoraria, Other: Advisory committee, Research Funding, Speakers Bureau; Pfizer: Honoraria, Other: Advisory committee; Celgene: Honoraria, Other: Advisory committee, Research Funding; Abbvie: Honoraria, Other: Advisory board, Research Funding, Speakers Bureau. Fong:Celgene: Other: Advisory board; Novartis: Other: Advisory board; Amgen: Research Funding; Pfizer: Other: Speaker fees. Reynolds:Novartis: Equity Ownership, Other: former employee of Novartis AG and holds stock in the company. . Roberts:Janssen: Research Funding; AbbVie: Research Funding; Genentech: Research Funding; Walter and Eliza Hall: Employment, Patents & Royalties: Employee of Walter and Eliza Hall Institute of Medical Research which receives milestone and royalty payments related to venetoclax.

Description: Background: Dihydroorotate dehydrogenase (DHODH), catalyzing the ubiquinone-mediated oxidation of dihydroorotate to orotate, is the rate-limiting enzyme in the de novo synthesis of pyrimidines, and as such may control the rate of cell division. The enzyme is localised on the outer side of the inner mitochondrial membrane, and links both the electron transport chain to pyrimidine production and thus to the maintenance of cell viability. DHODH inhibitors were identified in a myeloid differentiation screen using an ER-HoxA9 GMP cell line, and ASLAN003, a novel small molecule DHODH oral inhibitor, has been found to be a potent inducer of myeloid differentiation in AML cell lines. ASLAN003 also demonstrated the ability to reduce leukemic burden and extend survival in AML xenograft models. ASLAN003 has previously been shown to exhibit a safe and tolerable profile in prior phase I studies in healthy volunteers. Methods: A multicenter, single arm phase IIA study was initiated to evaluate ASLAN003 monotherapy administered as a 28-day cycle in patients with AML who are ineligible for standard therapy. The primary objective is to determine the optimum dose of ASLAN003 in this AML cohort based on efficacy (Overall Complete Remission Rate, % of complete remission [CR] + CR with incomplete hematologic recovery [CRi]), tolerability and safety. The secondary objective is to assess the pharmacokinetics (PK) of ASLAN003 and its metabolites and to further assess the efficacy based on relapse-free survival and clinical benefit rate (CBR, % of partial remission + CR + CRi). Exploratory objectives are to examine the myeloid differentiation effects of ASLAN003 ex vivo and explore possible relationships between the clinical response and molecular profile of leukemic cells. The study contains 3 cohorts for the optimum dose determination (100 mg, 200 mg, and 300 mg once daily [QD], with planned enrollment for 6 patients for each cohort), and an additional expansion cohort with the selected optimum dose (20 patients). Results: Enrollment started in December 2017. As of 12 July 2018, 10 patients were enrolled and treated (6 in the 100 mg QD cohort and 4 in the 200 mg QD cohort). Although the data are immature, two patients to date have exhibited some evidence of clinical activity (one per cohort). A patient in the 100 mg QD cohort, with a baseline peripheral blood blast of 27%, experienced a reduction to 6% on Cycle 3 Day 10 (C3D10), coupled with an upward trend in the percentage of neutrophils (16% at baseline, 64% on C3D10). The patient had underlying chronic lung infection, but also showed suspected symptoms of differentiation syndrome (dyspnea, leukocytosis). After 126 days of ASLAN003 initiation, the patient expired due to AML related clinical deterioration. The second patient in the 200 mg QD cohort experienced a decrease in bone marrow blasts from 26% at baseline, to 12% on C2D1. Concomitantly, there was a slight decrease in peripheral blood blasts with an increase in neutrophil percentages from 7% to 51%. The patient remains on study with further efficacy data awaited. To date, the safety profile is consistent with the previous healthy volunteer studies. Treatment-related adverse events (AEs) are listed in Table 1. The most frequent treatment-related AEs include nausea (n=2, 20%) and leukocytosis (n=2, 20%). Two treatment-related Grade ≥ 3 AEs were observed (Grade 3 anaemia and Grade 3 leukocytosis). The study continues and is currently enrolling to the 200 mg cohort. Conclusion: At the date of submission, this is the first report of clinical activity of a DHODH inhibitor in AML patients. Monotherapy with ASLAN003 was well tolerated and has showed encouraging signs of clinical activity in AML patients. This study is on-going and the optimal dose for AML has not yet been determined. Safety, PK and efficacy data will be updated at the time of presentation. Clinical trial information: NCT03451084 Disclosures Chng: ASLAN Pharmaceuticals: Research Funding. McHale:ASLAN Pharmaceuticals: Employment, Equity Ownership. Hsieh:ASLAN Pharmaceuticals: Employment, Equity Ownership. Shih:ASLAN Pharmaceuticals: Employment; AstraZeneca Taiwan: Employment. McIntyre:ASLAN Pharmaceuticals: Consultancy. Kwek:ASLAN Pharmaceuticals: Employment. Chang:ASLAN Pharmaceuticals: Employment. Lindmark:ASLAN Pharmaceuticals: Employment, Equity Ownership.

Description: The stem cell–intrinsic model of self-renewal via asymmetric cell division (ACD) posits that fate determinants be partitioned unequally between daughter cells to either activate or suppress the stemness state. ACD is a purported mechanism by which hematopoietic stem cells (HSCs) self-renew, but definitive evidence for this cellular process remains open to conjecture. To address this issue, we chose 73 candidate genes that function within the cell polarity network to identify potential determinants that may concomitantly alter HSC fate while also exhibiting asymmetric segregation at cell division. Initial gene-expression profiles of polarity candidates showed high and differential expression in both HSCs and leukemia stem cells. Altered HSC fate was assessed by our established in vitro to in vivo screen on a subcohort of candidate polarity genes, which revealed 6 novel positive regulators of HSC function: Ap2a2, Gpsm2, Tmod1, Kif3a, Racgap1, and Ccnb1. Interestingly, live-cell videomicroscopy of the endocytic protein AP2A2 shows instances of asymmetric segregation during HSC/progenitor cell cytokinesis. These results contribute further evidence that ACD is functional in HSC self-renewal, suggest a role for Ap2a2 in HSC activity, and provide a unique opportunity to prospectively analyze progeny from HSC asymmetric divisions.

Description: Introduction: Whilst relatively common in lymphoid malignancies, hypercalcaemia is an extremely rare complication of acute myeloid leukaemia (AML). Previous case reports have described ectopic parathyroid hormone secretion, leukaemic bony invasion and the release of boneresorptivemediators as causes of hypercalcaemia in AML (GewirtzAM et al. Br JHaematology1983;31(12):1590,ZidarBL, et al.NEJM.1976;295:692). We describe a case of severe hypercalcaemia with acute kidney injury (AKI) accompanying a new diagnosis of AML, subsequently demonstrated to be secondary to leukaemic blast production of active vitamin D (calcitriol) with gross over-expression of vitamin D related genes. This represents a novel pathogenic mechanism causing hypercalcaemia in a myeloid malignancy. Case Report: AA, a 68 year old male presenting with fatigue and found to have circulating blasts was subsequently diagnosed with acutemyelomonocyticleukaemia(78% blasts on bone marrow biopsy). Additionally, he had marked hypercalcaemia (calcium 3.3mmol/L, normal range 2.1-2.6mmol/L) and AKI (creatinine 263umol/L, normal range 60-110umol/L). Given the rarity of AML-associated hypercalcaemia, extensive investigations were undertaken to elucidate the cause. In search of a second concurrent malignancy, AA underwent computed tomography and positron emission tomography scanning, with subsequent biopsy of FDG avid vocal cord nodules; but only benign pathology could be demonstrated. Parathyroid hormone (PTH) levels were appropriately suppressed (0.9pmol/L, normal range 1.6-6.9pmol/L) and levels of PTH-related peptide and serum ACE were normal (500pmol/L, upper limit of normal: 190pmol/L). Both the hypercalcaemia and kidney injury proved refractory to multiple therapeutic strategies including aggressive hydration with an average of over 2.5L of crystalloid per day, as well as intravenouspamidronate. However, as depicted in Figure 1, there was a precipitous response following the initiation of chemotherapy (idarubicinandcytarabine, 7+3 regimen). Within several days, AA's serum calcium levels returned to normal levels, and his kidney function followed a similar pattern of improvement shortly thereafter. The rapid resolution of serum calcium levels also mirrored peripheral blast clearance, and repeat testing of calcitriol levels showed progressive improvement towards a normal concentration. AA achieved complete remission following induction chemotherapy and remains leukaemia free after consolidation chemotherapy and current maintenanceazacitidine. His hypercalcaemia has not recurred and his renal function remains normal. Having excluded other causes of hypercalcaemia and given the dramatic response to chemotherapy, we hypothesised that AA's AML blasts were secreting calcitriol. Accordingly, quantitative PCR was performed on AA's stored leukaemic cells for genes essential to vitamin D metabolism: the vitamin-D receptor (VDR), CYP24A1, and CYP27B1 (1-α-hydroxylase). RNA was extracted from AML cells using the QIAGEN RNeasykit. cDNAwas synthesised from 400ng of RNA using the Roche First Strand cDNASynthesis Kit. Gene expression was assessed by quantitative real-time PCR, relative to the housekeeping gene GAPDH. AA's leukaemia cells demonstrated markedly elevated expression of these vitamin-D related genes compared to healthy control CD34+ cells and four other independent primary AML cells (Figure 2, labelled AML 1-4), which were selected for absence of patient hypercalcaemia from our institution's tissue bank (Figure 2). Conclusion: Hypercalcaemia secondary to secretion of calcitriol can be a manifestation of lymphoid malignancies, however our case is the first documented occurrence of this phenomenon in a myeloid cancer. The PCR studies demonstrated striking overexpression of vitamin D related genes in leukaemia cells, resulting in the patient's hypercalcaemia and AKI. This finding represents a novel mechanism for a rare complication in AML. Figure 1 Figure 1. Figure 2 Figure 2. Disclosures No relevant conflicts of interest to declare.

Description: Despite the relative rarity of haematopoietic stem cells (HSCs) within the blood system, functional heterogeneity is paramount to their ability to sustain lifelong blood production. The quiescent HSC sits at the functional apex possessed with self-renewal properties and the greatest repopulation output. We previously identified the gene, Ap2a2 as an enhancer of HSC function and its protein as a potential cell fate determinant in HSC asymmetric cell divisions (Ting SB et al., Blood 2012). Mechanistically, we hypothesise Ap2a2 induces a state of HSC quiescence. Using the Tet-On histone H2B-GFP mouse model (Foudi et al., Nat Biotech 2009), we have shown Ap2a2 to be highly and differentially expressed in the predominantly, G0 dormant CD150+48-LSK GFPhigh as opposed to the more cycling GFPlow HSC subpopulation. Competitive transplantation of Ap2a2- versus empty vector-transduced H2B-GFP HSCs results in a three-fold increase of the CD150+48-LSK GFPhigh HSC subpopulation. To further confirm the importance of Ap2a2 in haematopoiesis, we have constructed Ap2a2-LacZ reporter and constitutive Ap2a2 knockout (KO) mouse lines. The Ap2a2 LacZ reporter with b-galactosidase flow cytometry staining of bone marrow subpopulations confirmed high endogenous Ap2a2 expression in the CD150+48-LSK long-term (LT-) versus CD150-48-LSK short-term (ST-) repopulating HSCs. Interim analyses of the constitutive Ap2a2 KO mice have revealed two obvious phenotypes: 14% of Ap2a2-null mice termed "non-survivors" are smaller, paler with failure of fetal liver (FL) development and die between E18.5 and weaning, whilst the remaining 11% are adult viable "survivors". However, at E14.5, Ap2a2-null compared to Ap2a2-wild type fetal livers showed less absolute total FL cells but increased CD150+48-LSM FL HSCs. This was quantitatively correlated via limiting dilution assay assessed at 16 weeks post-transplant with a two-fold increase in Ap2a2-null HSC numbers (1 in 78,917 versus 1 in 150,891, p=0.027). This suggests Ap2a2 has a role in FL HSC differentiation and/or fate with potential impairment of symmetrical versus asymmetrical HSC divisions currently being studied. When E14.5 FL cells were competitively transplanted, the Ap2a2-null HSC had impaired donor reconstitution function measured at 16 weeks post-transplant (19.8% versus 48.6%, p=0.015). Ap2a2-null versus wild-type E14.5 FL cells showed equivalent numbers of primary in vitro methylcellulose colony assays but loss of secondary colonies upon re-plating indicative of loss of in-vitro HSC self-renewal. Importantly, although the Ap2a2 adult "survivors" exhibited normal quantities of bone marrow HSC subpopulations, when functionally assessed, Ap2a2-null adult "survivor" HSCs showed loss of in-vivo HSC self-renewal in secondary transplantation assays. To investigate potential cellular mechanisms, we studied the cell cycle state of Ap2a2-null and wild-type E14.5 FL cells and identified that Ap2a2-null "non-survivors" had a relative loss of quiescent G0, specifically in the LT-HSC (and not seen in the ST-HSC) subpopulation throughout all of (E14.5 to E18.5) FL development. In contrast, the LT-HSC subpopulation in FLs of Ap2a2-null "survivors" had an initial loss of G0 at E14.5 but a compensatory increase in LT-HSC G0 by E18.5. Our preliminary data suggests Ap2a2 is a crucial factor for the quiescent LT-HSC subpopulation, and we propose that both during the highly proliferative fetal liver stage of haematopoiesis and adult HSCs under stress that Ap2a2 maintains a critical balance of dormant ("deep-sleeper") HSCs to ensure global HSC function. Disclosures No relevant conflicts of interest to declare.

Description: Aim: Cardiac amyloidosis is a protein deposition disease that can be difficult to diagnose and has a poor prognosis if diagnosis or treatment are delayed. The two major subtypes are AL and transthyretin (ATTR). Both have vastly different treatments so confirming the correct amyloid subtype is crucial. A tissue biopsy is usually required for the diagnosis of amyloidosis and to distinguish between the subtypes. No blood test, echocardiography or cardiac MRI can reliably distinguish between AL and ATTR. With a cardiac biopsy, distinguishing AL from ATTR can be challenging with immunohistochemistry, and time consuming with mass spectrometry. Recently, Gillmore et al demonstrated bone scintigraphy with 99mTc-DPD tracer can reliably diagnose ATTR, avoiding endomyocardial biopsies to confirm the subtype in most cases. [1] 99mTc-HDP is a tracer similar to 99mTc-DPD and is more readily available in Australian and the USA. We sought to examine the use of 99mTcHDP bone scintigraphy in Australia and determine the accuracy of this tracer to diagnose cardiac amyloidosis and distinguish between the AL and ATTR subtypes. Methods: All patients with confirmed ATTR or AL who had 99mTcHDP bone scintigraphy were analysed. Results were correlated with histology, NT ProBNP, Troponin-T, free light chains, cardiac MRI and echocardiography. Grading was conducted with Perugini scoring.[1] Results: 25 patients with amyloidosis diagnosed by cardiac MRI and/or biopsy, had 99mTcHDP bone scintigraphy. 18 were confirmed ATTR, 7 AL. Two ATTR patients had hereditary disease (Thy60Ala and Gly109Lys), the remainder were wildtype. 17 (94%) patients with ATTR and 2 (29%) AL had positive scans. The negative ATTR patient had localized bladder disease only with a normal echocardiogram and cardiac biomarkers. All ATTR patients with positive scans had Perugini scores of 2 or 3, including the 2 patients with hereditary mutations, while the two positive AL only had scores of 1. All 11 patients with amyloid features on cardiac MRI had positive scans. Mean NTproBNP values for ATTR and AL were 530pmol/L and 1396pmol/L respectively. The two AL amyloidosis patients with positive bone scans had higher NTproBNP values. No ATTR patient had a detectable plasma cell dyscrasia. Conclusion: Bone scintigraphy with 99mTcHDP tracer is an easily accessible, rapid and non-invasive method of diagnosing cardiac amyloidosis. In our small series, all patients with Perugini scores 2 or 3 had ATTR, while those with negative or Perugini score 1 scan either had AL or no cardiac amyloidosis. There was a suggestion that AL patients with higher NTproBNP scores were more likely to have positive scans. This suggests that the 99mTcHDP tracer can be used like 99mTc-DPD to aid in the diagnosis of cardiac amyloidosis, and, as suggested by Gillmore et al, can confirm the ATTR subtype in those with no detectable plasma cell dyscrasia and Perugini score 2 or 3 scans, thus hastening accurate diagnosis and avoiding cardiac biopsies. Reference: Gillmore JD, Maurer MS, Falk RH, Merlini M, Damy T, Dispenzieri A, et al. Non-biopsy diagnosis of cardiac transthyretin amyloidosis. Circulation. 2016 Jun 14;133(24):2404-12 Disclosures No relevant conflicts of interest to declare.

Description: The ability of hematopoietic stem cells (HSC) for self-renewal is in part dependent on asymmetric cell divisions (ACD). However, the molecular insights remain elusive. Our current knowledge on mechanisms of ACD is predominantly derived from studies in the invertebrate systems of D. Melanogaster and C.Elegans. These include a cluster of polarity related genes, which function as cell fate determinants (CFD). The identification of Numb as a CFD gene in the mammalian stem cells of the brain and retina, and muscle satellite cells suggests conservation of aspects of ACD. Based on published data, a candidate list of 72 potential CFD genes was chosen for expression profiling. The specific populations of mouse bone marrow cells used for quantitative-PCR (q-PCR) studies were: (1) Long Term Repopulating-HSC (LTR-HSC) sorted for CD150+48−41−Lin− compared to progenitors (CD150+/−48+41+Lin−); (2) LTR-HSC derived from Sca+Kit+Lin−Rholow CD49b− compared to progenitors (Sca+/−Kit+/−Lin-Rhohigh CD49b+/−) and (3) the primary Leukemia Stem Cells (LSC), FLA2 and FLB1 with Leukemia Repopulating Cell (LRC) frequencies of respectively, 1 in 1.4 and 1 in 350 (P. Austin and G.S., unpublished). To validate the purity of LTR-HSC, limiting dilution assays were performed on the sorted populations. RNA obtained from all independent populations were initially pre-amplified before q-PCR. Subsequently, a subset of the highly expressed genes was re-analyzed with non-amplified RNA derived from the FLA2 and FLB1 cells. The q-PCR results from both the LTR-HSC and LSC populations show that a significant number of potential CFD genes, in particular those related to polarity and cytoskeleton function were highly and differentially expressed. It is assumed that the LTR-HSC has predominantly ACD. By using HoxB4 retroviral (MSCV) mediated over-expression to enforce symmetric cell divisions (SCD) and immunofluorescence (IF) analysis, we assessed the endogenous cellular expression of 2 of the highly expressed genes from the aforementioned q-PCR studies namely, Numb and Par6B. In a small cohort of LTR-HSC infected with the MSCV vector alone (representing ACD), the majority of cells expressed both Numb and Par6B in a cortical and asymmetric distribution. In contrast, infection by MSCV-HoxB4 virus of LTR-HSCs (representing SCD) resulted in the re-localization of both Numb and Par6B to be predominantly diffuse and symmetrical. Furthermore, using the FLA2 and FLB1 leukemia, with their respective LRC frequencies (above) as another cohort of cells that represent ACD (FLB1) and SCD (FLA2), similar re-localization results were obtained. Specifically, in FLB1 cells the predominant endogenous expression of Numb and Par6B was cortical and asymmetrical, in comparison to FLA2 cells where it was diffuse and symmetrical. Quantitative analysis of IF data suggested a higher expression of Numb and more so, of Par6B in FLB1 compared to FLA2 cells. Western blot analysis of Par6B in FLB1 and FLA2 cells supported the IF data. These preliminary results show potential CFD genes are highly and differentially expressed in LTR-HSC and LSC. Also, that during ACD there may be an up-regulation of CFDs together with a redistribution of these proteins to the cortical membrane, in either an asymmetrical or symmetrical localization. The inference of re-localization and up-regulation of in particular, Par6B may in itself be deterministic of HSC fate or may reflect a read-out of the cell fate choice, the latter of which could then be utilized to further identify other HSC CFD genes.