Search

Background Chimeric antigen receptor (CAR)-T cell therapy has revolutionised the treatment of relapsed or refractory B-ALL in children and young adults with unprecedented response rates. However, primary resistance and relapse are unresolved challenges that limit long term benefit in a significant proportion of patients. Death receptor mediated extrinsic apoptosis is a key component of CAR-T cytotoxicity and impairment of this system, of which CAR-T derived TRAIL (tumour necrosis factor related apoptosis inducing ligand) is a key initiator, is a principal driver of primary resistance. Arginine deprivation with the therapeutic enzyme ADI-PEG20 (pegylated arginine deiminase) sensitises cancers deficient in the enzyme argininosuccinate synthase (ASS1) to the apoptosis initiating activity of TRAIL through tumour cell surface upregulation of death receptors DR4 and DR5. Whether this effect could potentiate the TRAIL-DR activity in CAR-T therapy has not been explored. Aim We tested the hypothesis that ADI-PEG20 treatment can sensitise susceptible B-ALL to anti-CD19 CAR-T through priming of death receptor mediated apoptosis signalling. Methods The effect of ADI-PEG20 on cell survival and death receptor expression in B-ALL cell lines and primary samples was analysed by flow cytometry. Second generation anti-CD19 CAR-T cells with a CD28 costimulatory domain were generated by retroviral transduction of activated peripheral blood mononuclear cells (PBMC) from healthy donors. For CAR-T co-culture experiments, B-ALL cell lines were pre-treated with ADI-PEG20 before washing and re-suspending in arginine replete media prior to CAR-T cell addition. Results To establish potential susceptibility of B-ALL to ADI-PEG20 we measured expression of ASS1, which inversely correlates with sensitivity to the drug, using combined in situ immunohistochemistry (n=6) and RT-qPCR (n=7). ASS1 deficiency was consistently seen in this series of primary samples suggesting the potential utility of ADI-PEG20 in B-ALL, with comparable expression levels to those seen in a cohort of primary AML samples proven to be sensitive to the drug (figure 1a). Next, to examine variation in ASS1 expression between genetically defined subtypes of B-ALL we re-analysed transcriptome data from a cohort of 215 patients treated on the ECOG E2993 trial, filtered into a network of 58 genes generated according to known or predicted interaction with ASS1. We found an enrichment of Philadelphia chromosome positive (Ph+) and Philadelphia-like (Ph-L) samples in the cluster characterised by lowest ASS1 expression along with high HIF1A expression, matching a recurrent pattern reported in other ADI-PEG20 sensitive tumours. This therefore predicts that among B-ALL subtypes, Ph+ and Ph-L are likely to be most sensitive to therapeutic arginine deprivation. We then functionally confirmed, using in vitro cell line (n=3) and in vivo patient derived xenograft models of B-ALL (n=2), that ASS1 deficiency is required for ADI-PEG20 sensitivity. Using the ASS1-low, Ph-L cell line MUTZ-5, we established that ADI-PEG20 induced apoptosis accompanies cell surface upregulation of both DR4 and DR5 expression. Upregulation of DR4 was observed to follow an upwards trend after treated cells were washed and re-suspended in arginine replete media, suggesting that transient arginine starvation can commit ASS1-low B-ALL to a state of apoptotic priming (figure 1b). With confirmed engagement of arginine starvation and death receptor upregulation we tested the synergy potential of ADI-PEG20 pre-treatment of MUTZ-5 followed by CAR-T, utilising calculated combination drug indices (CDIs). Across independent PBMC donors (n=3) we observed greater potency killing of CD19 + leukaemia cells in the combination treated co-cultures when compared to the single agent treated conditions, with CDIs consistently less than 1 confirming a synergistic effect (figure 1c). Conclusion Our study proposes a synergistic interaction between the arginine depleting enzyme ADI-PEG20 and anti-CD19 CAR-T for the treatment of ASS1 deficient B-ALL, whereby priming of death receptor signalling may underlie enhanced CAR-T cytotoxicity against CD19 + tumour cells. These data support an emerging framework for CAR-T optimisation based on targeting of the death receptor mediated extrinsic apoptosis pathway and can inform future refinements in the development of cellular immunotherapy. Figure 1 Figure 1. Disclosures Bomalaski: Polaris Pharmaceuticals Inc.: Current Employment. Maher: Leucid Bio: Other: Chief Scientific Officer. Gribben: Abbvie: Honoraria; AZ: Honoraria, Research Funding; BMS: Honoraria; Gilead/Kite: Honoraria; Janssen: Honoraria, Research Funding; Morphosys: Honoraria; Novartis: Honoraria; Takeda: Honoraria; TG Therapeutis: Honoraria.

The strategy of enzymatic degradation of amino acids to deprive malignant cells of important nutrients is an established component of induction therapy of acute lymphoblastic leukemia. Here we show that acute myeloid leukemia (AML) cells from most patients with AML are deficient in a critical enzyme required for arginine synthesis, argininosuccinate synthetase-1 (ASS1). Thus, these ASS1-deficient AML cells are dependent on importing extracellular arginine. We therefore investigated the effect of plasma arginine deprivation using pegylated arginine deiminase (ADI-PEG 20) against primary AMLs in a xenograft model and in vitro. ADI-PEG 20 alone induced responses in 19 of 38 AMLs in vitro and 3 of 6 AMLs in vivo, leading to caspase activation in sensitive AMLs. ADI-PEG 20-resistant AMLs showed higher relative expression of ASS1 than sensitive AMLs. This suggests that the resistant AMLs survive by producing arginine through this metabolic pathway and ASS1 expression could be used as a biomarker for response. Sensitive AMLs showed more avid uptake of arginine from the extracellular environment consistent with their auxotrophy for arginine. The combination of ADI-PEG 20 and cytarabine chemotherapy was more effective than either treatment alone resulting in responses in 6 of 6 AMLs tested in vivo. Our data show that arginine deprivation is a reasonable strategy in AML that paves the way for clinical trials.

Introduction Malignant cells require amino acids for a wide range of core functions. Amino acid deprivation using enzymatic degradation has been used to induce remission in acute lymphoblastic leukemia for decades. Amino acid deprivation may also benefit patients with acute myeloid leukemia (AML). We have previously shown that AML cells lack of argininosuccinate synthetase 1(ASS1), a key enzyme in the pathway that produces arginine (1). Here we tested the effect of an arginine depleting agent, pegylated arginine deiminase (ADI-PEG 20) on primary AML cells in a xenograft model of AML. Methods NOD/SCID/interleukin 2 gamma chain null (NSG) mice were transplanted with 6 primary AML samples. 12 weeks after transplantation of AML mice received either ADI-PEG 20, cytarabine, ADI-PEG 20 plus cytarabine or vehicle. ADI was administered weekly for 4 doses and cytarabine was given for 10 consecutive days (0.2mg per day, roughly equivalent to 40mg in humans). Five weeks after treatment started, mice were killed and the percentage of AML in the bone marrow was determined by flow cytometry. Blood was collected to quantify plasma arginine using liquid chromatography-mass spectrometry/mass spectrometry. Results Plasma arginine levels were depressed following ADI-PEG 20 administration confirming that arginine depletion was achieved in vivo. In all six experiments the combination of ADI-PEG 20 and cytarabine induced a significant reduction in levels of AML compared to control (Figure 1). Critically the combination of ADI-PEG 20 and cytarabine was significantly better than cytarabine alone in three of six experiments. Conclusion Our experiments show that arginine deprivation by ADI-PEG 20 can decrease the leukemic burden in mice transplanted with primary AML cells. The combination of ADI with cytarabine had a greater effect than cytarabine alone in half the experiments. These results provide the rationale to test ADI-PEG 20 with cytarabine in clinical trials. 1. Peter W. Szlosarek, Fiona Luong, Andrew Clear, David Taussig, Simon Joel, Maria Calaminici, Silvana Debernardi, Jude Fitzgibbon, John S. Bomalaski, Arthur E. Frankel, and Dominique Bonnet. Pegylated arginine deiminase (ADI-PEG 20) as a potential novel therapy for argininosuccinate synthetase-deficient acute myeloid leukemia. Cancer Research: April 15, 2011. AACR 102nd Annual Meeting 2011, (AACR Abstract # 467) F. M-M and L. A-M contributed equally D.B., P.W.S. and D.C.T contributed equally Disclosures: Bomalaski: Polaris Group: Employment, Equity Ownership. Szlosarek:Polaris Group: Research Funding.

Abstract 4640 Background. Arginine is critical for the growth of certain human cancers and is an intermediary in the synthesis of various bioactive molecules. Argininosuccinate synthetase-1 (ASS1) catalyses the rate-limiting step in arginine biosynthesis, the conversion of citruline to arginine. Argininosuccinate lyase (ASL) is an enzyme that catalyses the reversible breakdown of arginosuccinate producing arginine and fumerate. Down-regulation of ASS1 has been demonstrated in several solid cancers and has been shown to confer sensitivity to arginine deprivation (arginine auxotrophic tumors). Methylation-dependent silencing of the ASS1 promoter may account for ASS1 down-regulation and arginine auxotrophy. Arginine deprivation using arginine deiminase (ADI-PEG20) is currently considered as a novel therapeutic intervention for cancer, and early clinical trials have shown promising results in solid tumors. In this perspective, we investigated for the first time the methylation status of the ASS1 and ASL CpG islands in multiple myeloma (MM) and analyzed for clinical relevance. Methods. Genomic DNA was extracted from bone marrow aspirate samples from 46 MM patients (28 male, 18 female, median age 64 years) obtained at diagnosis. Methylation-specific PCR (MSP) was employed to study the methylation in the ASS1 and ASL CpG islands. DNA was isolated and bisulphite modified using commercially available kits (QIAmp DNA mini kit, Qiagen and EZ DNA methylation kit, Zymo Research respectively). Control methylated and unmethylated genomic DNAs were included in each experiment. Ten bone marrow samples, with no neoplastic cells, from patients with borderline thrombocytopenia served as negative controls. Logistic regression analysis was used to measure the association between gene methylation and sex, age, ISS stage, presence of extramedullary disease, renal failure (eGFR Results. Methylation in the CpG island of ASS1 was detected in 71.7% patients (95% CI 57–82%) and of ASL in 37% patients (95% CI 24–51%), while simultaneous methylation in both genes was present in 10 patients. None of the two genes were detectably methylated in the control group. Patients with ASS1 methylation were less likely to have bone disease (p=0.04, OR=0.22) and extramedullary disease (p=0.05, OR=0.22). There was no statistically significant difference in overall survival by methylation status of the studied genes. Conclusions. Arginine biosynthesis genes ASS1 and ASL are methylated in MM. Methylation of ASS1 was found to be more frequent and negatively associated with bone or extramedullary disease. Further evaluation of ASS1 and ASL methylation is warranted in MM in the perspective of expanding arginine deprivation trials in these patients Disclosures: No relevant conflicts of interest to declare.