Results of the Phase 1B study of pegargiminase (ADI-PEG 20) in combination with AZA-VEN in newly diagnosed high-risk AML patients not candidates for intensive chemotherapy

Abstract

Abstract Background Tumor metabolism is a promising area for drug development in AML. Pegargiminase (ADI-PEG 20), a metabolism-targeting therapeutic consisting of the arginine-degrading enzyme arginine deiminase combined with polyethylene glycol, demonstrated high synergy with VEN-decitabine in pre-clinical models and anti-leukemic activity as monotherapy and with cytarabine in clinical trials. Efficacy was most notable in tumors deficient in argininosuccinate synthetase (ASS1), the rate limiting step in the urea cycle to form arginine. AML is expected to be sensitive, with approximately 90% of AML patient samples deficient in ASS1 expression. We conducted a phase 1A/B trial evaluating triplet therapy with ADI-PEG 20 + azacitidine and venetoclax (AZA-VEN) in relapsed/refractory (R/R) and newly diagnosed (ND) AML (NCT05001828). Methods The phase 1A dose escalation cohort in R/R AML (n=13) was previously reported. Here we present for the first time dose expansion results in ND AML patients (pts) who were older or unfit for intensive chemotherapy and had high-risk features by WHO 2022 criteria. ADI-PEG 20 was given at the RP2D of 36 mg/m2 intramuscularly (IM) on day -3 (priming dose) and weekly starting C1D1 with AZA-VEN in 28-day cycles, continuing during AZA-VEN treatment holds. AZA was given per SOC and VEN for up to 28 days in C1, 21 days in C2 and beyond, with subsequent reductions in VEN duration allowed. Azoles were permitted with VEN dose adjustment. Secondary and therapy-related AML were allowed; up to 2 cycles of prior HMA were allowed. Key study objectives were safety, objective response, and OS. Patients were followed for survival after end of study treatment (EOT). Results 25 ND AML pts were enrolled. Median age was 71.9 years (range 58.1-86.4), 64% of pts were male; 64% were white, 32% Hispanic, 4% unknown. All pts had adverse ELN risk. 10 pts (40%) had TP53 mutation or complex monosomal karyotype (CMK) (TP53+CMK, n=8; TP53 without CMK, n=1; CMK/TP53 wild-type, n=1). 6 pts had RAS pathway mutations (1 overlapping with TP53). No pts had FLT3-ITD. Overall, 16 unique pts (64%) had TP53, CMK, RAS pathway (K/NRAS, PTPN11, CBL, NF1), or KMT2Ar, and 9 other pts (36%) had secondary-type mutations +/- MRCA (excluding CMK). 5/25 pts (20%) had blast phase MPN. 23 pts were evaluable for efficacy as defined by the protocol (minimum 21 doses of VEN, 7 doses of AZA, and 3 doses of ADI-PEG 20). Of these, 16 had best response of CR (69.6%), 1 CRh (4.3%), 4 CRi (17.4%), and 2 MLFS (8.7%), resulting in an objective response rate (CR+CRh+CRi; composite CR [CRc]) of 91.3% (21/23). All evaluable pts had a response of MLFS or better (23/23). Of CRc pts, 11 (52.4%) achieved MRD-negativity by MFC. Median duration of CRc (DOR) was 10.4 months (95% CI: 2.5-NR). Pts were censored for DOR at date of last on study response assessment (updated sensitivity analysis will be presented). Median OS was 19.1 months (95% CI: 7.8-NR), and 14/23 pts (60.9%) were censored for OS analysis (still alive). Median follow-up was 16.9 months. In the entire cohort (n=25), safety was consistent with the expected profile of AZA-VEN. The addition of ADI-PEG 20 was well-tolerated, with a notable absence of toxicities associated with PEG-asparaginase. No hematologic or non-hematologic DLTs were observed. 30-day mortality was 8% (2/25), with 1 early death from pneumonia and 1 subject who elected for hospice before completing C1 (both were TP53 mutated). Reasons for EOT were, completed 24 months of study treatment (n=1), proceeded to allo-HCT (n=5), physician decision (n=5), subject withdrawal (n=3), adverse event (n=3), progression (n=1), relapse (n=4), and death (n=1). 2 pts continue on study treatment in CR/CRh, and 9 were removed from study in CR/CRh for either allo-HCT, active surveillance, or maintenance therapy. Baseline plasma arginine level (n=8) was 81.3 µM (mean, SD 48.6), decreasing rapidly after the first ADI-PEG 20 dose (day -3) to 0.1 µM (SD 0.4) on C1D1. Arginine levels remained depleted throughout C1, with a mean level of 0.7 µM (SD 0.8) on C1D22. As compensatory biology, plasma citrulline levels increased proportionally with ADI-PEG 20 treatment. ConclusionADI-PEG 20 was easy to administer and tolerable in frontline combination with AZA-VEN, resulting in the rapid depletion of circulating arginine. Rates of CRc and MRD-negativity were high, with durable responses observed. OS was better than expected, in this very-high risk AML population.