Pivekimab sunirinine (PVEK) combined with FLAG-IDA for medically fit adults with untreated adverse-risk Acute Myeloid Leukemia (AML) or other high-grade myeloid neoplasms (NCT06034470)

Abstract

Abstract Background: The need for more effective therapies for adults with adverse-risk AML is unquestioned. We hypothesized that combining an intensified chemotherapy backbone with a CD123-targeted therapeutic could improve outcomes in such patients given the high CD123 expression among neoplastic myeloid stem and progenitor cells, including those underlying adverse-risk leukemias. Pivekimab sunirine (PVEK, IMGN632) is a first-in-class antibody drug conjugate comprising a high-affinity CD123 antibody, cleavable linker, and an indolinobenzodiazepine (IGN) payload. PVEK has shown anti-leukemia activity in early phase trials both as a single agent and in combination with azacitidine and venetoclax. Here, we conducted a phase 1 trial evaluating the safety and preliminary efficacy of PVEK combined with the high intensity chemotherapy backbone FLAG-IDA for patients with untreated adverse-risk AML. Methods Patients (Pts) ≥18 years (yrs) with newly diagnosed AML or MDS/AML exhibiting CD123 expression were eligible if they had adverse risk disease by the ELN 2022 classification, did not harbor a mutation in FLT3, were fit for intensive chemotherapy as defined by treatment-related mortality (TRM) score of ≤13.1 (ie, predicted 28 day mortality of ≤13.1% with standard chemotherapy), and had adequate organ function (LVEF ≥45%, AST/ALT < 3x upper limit of normal [ULN], bilirubin < 1.5x ULN, creatinine clearance >60 mL/min). Pts with active uncontrolled infection or concomitant illness with expected survival ≤1 yr were excluded. Prior hypomethylating therapy for antecedent low grade myeloid neoplasm (≤10% blasts) was allowed. Cohorts of 3 patients were assigned to 1 of 3 dose levels (DL) of PVEK (DL1: 0.03 mg/kg on day 1; DL2 0.045 mg/kg on day 1; DL3: 0.045 mg/kg days 1 and 22). Other chemotherapy included G-CSF (days 0-5), fludarabine (30 mg/m2 days 1-5), cytarabine 2 g/m2 (days 1-5) and idarubicin 10 mg/m2 (days 1-3). A second, identical course of FLAG-ida with PVEK was offered if complete remission (CR) without measurable residual disease (MRD-; defined as < 0.01% by multiparameter flow cytometry [MFC]) was not achieved with cycle 1. Post-remission therapy for patients achieving a CR or CR with partial or incomplete hematologic recovery (CRh or CRi) included high-dose cytarabine and PVEK at the same DL as during induction. Three doses of dexamethasone were administered prior to PVEK for infusion reaction prophylaxis. Dose-limiting toxicities (DLTs) were (1) grade ≥4 organ toxicity or (2) prolonged severe myelosuppression (ANC < 500/µL, platelets <25k/µL lasting ≥42 days in the absence of residual disease). A modified BOIN design was used for dose selection, targeting a DLT rate of ≤25%. Isotonic regression on the first 12 pts treated was used to select the maximum tolerated dose and was used for remaining pts (up to n=30 total). The data cut-off was 7/15/25. Results: Among 20 pts consented as of the data cut-off, 1 pt had CML and was excluded. The remaining 19 (median age: 60 [range: 23-76] years; median TRM score: 2.1 [range: 0.5-9.1]) met eligibility criteria and received study therapy. Three, 6, and 10 pts were treated at DL1, DL2, and DL3, respectively. At DL3, 1 DLT was observed (hypoxic respiratory failure, possibly secondary to aspiration and volume overload). One grade 3 infusion reaction occurred that responded to diuresis. Neutropenic fever was a universal treatment emergent adverse event, occurring in all pts. Three pts developed grade 4 infections (C. difficile colitis, Klebsiella bacteremia following re-induction therapy; and disseminated fungal infection that proved fatal). Other grade ≥3 AEs observed in >1 pt included hypokalemia and syncope/fall (n=2). Eighteen pts were response evaluable, with CR/CRi observed in 15/18 pts (83%). Observed responses were: CR (n=11), CRi (n=4), partial remission (n=1), morphologic leukemia-free state (n=1), and persistent disease (n=1). Among CR/CRi pts, 11 of 15 were MRD- by MFC for an overall MRD- CR rate of 11/18 = 61%. The 60-day mortality was 0%. Overall, three pts have died; the estimated 1-year overall survival is 73%. Conclusion Addition of PVEK to FLAG-IDA was tolerable at all explored dose levels and resulted in high response rates in a population that is notoriously challenging to treat. Dose escalation is complete, and additional recruitment to DL3 is ongoing to better estimate response and outcomes, with updated results to be presented at the meeting.