Mutant Isocitrate Dehydrogenase 1 Inhibitor Ivosidenib in Combination With Azacitidine for Newly Diagnosed Acute Myeloid Leukemia

Abstract

Purpose: Ivosidenib is an oral inhibitor of the mutant isocitrate dehydrogenase 1 (IDH1) enzyme, approved for treatment of IDH1-mutant (mIDH1) acute myeloid leukemia (AML). Preclinical work suggested that addition of azacitidine to ivosidenib enhances mIDH1 inhibition-related differentiation and apoptosis.

Patients and methods: This was an open-label, multicenter, phase Ib trial comprising dose-finding and expansion stages to evaluate safety and efficacy of combining oral ivosidenib 500 mg once daily continuously with subcutaneous azacitidine 75 mg/m² on days 1-7 in 28-day cycles in patients with newly diagnosed mIDH1 AML ineligible for intensive induction chemotherapy (ClinicalTrials.gov identifier: NCT02677922).

Results: Twenty-three patients received ivosidenib plus azacitidine (median age, 76 years; range, 61-88 years). Treatment-related grade ≥ 3 adverse events occurring in > 10% of patients were neutropenia (22%), anemia (13%), thrombocytopenia (13%), and electrocardiogram QT prolongation (13%). Adverse events of special interest included all-grade IDH differentiation syndrome (17%), all-grade electrocardiogram QT prolongation (26%), and grade ≥ 3 leukocytosis (9%). Median treatment duration was 15.1 months (range, 0.3-32.2 months); 10 patients remained on treatment as of February 19, 2019. The overall response rate was 78.3% (18/23 patients; 95% CI, 56.3% to 92.5%), and the complete remission rate was 60.9% (14/23 patients; 95% CI, 38.5% to 80.3%). With median follow-up of 16 months, median duration of response in responders had not been reached. The 12-month survival estimate was 82.0% (95% CI, 58.8% to 92.8%). mIDH1 clearance in bone marrow mononuclear cells by BEAMing (beads, emulsion, amplification, magnetics) digital polymerase chain reaction was seen in 10/14 patients (71.4%) achieving complete remission.

Conclusion: Ivosidenib plus azacitidine was well tolerated, with an expected safety profile consistent with monotherapy with each agent. Responses were deep and durable, with most complete responders achieving mIDH1 mutation clearance.

FIG 1.

Treatment duration, response over time, and IDH1 mutation status. Responses were assessed by the investigator. Complete remission (CR) with partial hematologic recovery (CRh) was derived by the sponsor and defined as meeting all CR criteria except absolute neutrophil count > 0.5 × 10⁹/L (500/μL) and platelet count > 50 × 10⁹/L (50,000/μL). The last dose date (for either treatment, whichever was later) was applied for the treatment duration calculation for any patient who was still on treatment as of the data cutoff date. (*) Patient continued on commercially available ivosidenib. (†) Patient had mutant IDH1 (mIDH1) clearance in peripheral blood mononuclear cells (PBMCs) only (bone marrow mononuclear cells [BMMCs] not available); all other patients had mIDH1 clearance in both BMMCs and PBMCs. (‡) Patient received ivosidenib + azacitidine for 23 days before withdrawing from study treatment and died 14 months after end of treatment (unknown cause); no additional therapy after ivosidenib + azacitidine was reported. CRi, CR with incomplete neutrophil recovery; CRp, CR with incomplete platelet recovery; HSCT, hematopoietic stem cell transplantation; MLFS, morphologic leukemia-free state; NA, not assessed; PR, partial remission; SD, stable disease.

FIG 2.

Baseline co-occurring mutation analysis and association with clinical response. (A) Baseline IDH1 variant allele frequency (VAF) levels as detected by next-generation sequencing (NGS) in either bone marrow mononuclear cells (BMMCs; n = 17) or peripheral blood mononuclear cells (PBMCs; n = 20). VAF in patients achieving complete remission (CR)/CR with partial hematologic recovery (CRh) was compared with that of patients with non-CR/CRh responses (including stable disease [SD]) using Student’s t test (two-sided). VAF levels in neither BMMCs (P = .89) nor PBMCs (P = .17) were associated with clinical response. (B) Mutations co-occurring in ≥ 5% of patients at baseline in order of frequency in BMMC and/or PBMC samples, depending on sample availability. (C) Heat map showing baseline co-occurring mutations identified by variant type and patient characteristics (mutant IDH1 clearance by BEAMing (beads, emulsion, amplification, magnetics) digital polymerase chain reaction, cytogenetic risk, de novo or secondary acute myeloid leukemia) and grouped by best overall response and altered pathway. NGS data were derived from BMMCs (n = 17) or from PBMCs (n = 6) if no screening bone marrow sample was available. IDH1-MC, IDH1 mutation clearance; MLFS, morphologic leukemia-free state; NE, not evaluable; RTK, receptor tyrosine kinase.