Paper Details

ABSTRACT

Background: Mutations in isocitrate dehydrogenase 1 (IDH1) are detected in 7–10% of patients (pts) with newly diagnosed (ND) acute myeloid leukemia (AML). Ivosidenib (IVO) is an oral, selective inhibitor of mutant IDH1 (IDH1mut) with established efficacy in relapsed/refractory AML and in pts with ND AML unfit for intensive chemotherapy (IC). A single-arm phase 1/2 study showed that IVO plus IC is feasible (Stein et al. Blood 2021) but a prospective assessment of the comparative therapeutic value of the combination is lacking. The phase 3, multicenter, double-blind HOVON150/AMLSG29-18 trial evaluates IVO versus (vs) placebo (PBO) in combination with IC, followed by single-agent maintenance, in pts with ND IDH1mut AML or myelodysplastic syndrome with excess blasts 2 (MDS-EB2). Here, we report the results of the prespecified interim superiority analysis of the primary endpoint, event-free survival (EFS).

Methods: Eligible pts (≥18 years, ECOG performance status ≤2, FLT3 wildtype) were randomized 1:1 to receive IVO or PBO, stratified by age (≤60 vs >60 years) and disease status (de novo vs secondary). Induction therapy consisted of two cycles: cycle 1 with 7+3 (cytarabine plus daunorubicin) and cycle 2 with intermediate-dose cytarabine (IDAC) plus daunorubicin, with IVO (500 mg once daily from day 1 and given continuously thereafter) or PBO. Pts achieving complete remission (CR), CR with incomplete hematologic recovery (CRi), or morphologic leukemia-free state (MLFS) received consolidation with either chemotherapy (up to three cycles of IDAC or one cycle of mitoxantrone/etoposide) with IVO/PBO, or autologous hematopoietic cell transplantation (HCT), or allogeneic HCT, followed by maintenance therapy with IVO or PBO according to the original randomization for up to 2 years. The primary endpoint EFS was defined as time from randomization to failure to achieve CR or CRi after induction, relapse after achieving CR or CRi, or death, whichever comes first. Key secondary endpoint was overall survival (OS); other secondary endpoints included remission rates, relapse-free survival, measurable residual disease, quality of life, and safety. A prespecified superiority interim analysis of the primary endpoint was conducted in December 2025 after approximately 67% of the total required EFS events had occurred. EFS was analyzed in the intention-to-treat population using a stratified Cox proportional hazards model and the stratified log-rank test.   

Results: Between September 26, 2019, and September 19, 2024, 487 pts were randomized to IVO (n=244) or PBO (n=243) across 191 centers in Europe and Australia. Baseline characteristics were balanced between arms. Median (range) age was 61 (21–77) years, with 29% of pts aged >65 years. Overall, 95% had AML and 5% MDS-EB2; 48% were female, 37% had ELN 2017 adverse-risk disease, and 35% underwent alloHSCT per protocol. At a median follow-up of 24.4 months (mo), median EFS was not reached (NR; 95% CI, NR–NR) in the IVO arm vs 10.8 mo (95% CI, 9.3–13.3) in the PBO arm (HR 0.32; 95% CI, 0.22–0.48; P<0.001). Estimated 24-mo EFS was 74% (95% CI, 67–79) with IVO vs 35% (95% CI, 28–42) with PBO. The prespecified superiority boundary was crossed, demonstrating that the trial met its primary endpoint. EFS benefit of IVO was consistent across age groups (>65 and ≤65 years). CR/CRi rates after induction were 88% vs 76%, and CR rates 80% vs 68% in the IVO and PBO arms, respectively. Among pts achieving CR/CRi, relapse occurred in 7% of the IVO arm vs 33% of the PBO arm, while 10% vs 14% of pts in CR/CRi died. Preliminary safety data were comparable between the treatment arms. Adverse events (AEs) grade ≥1 considered related to IVO or PBO occurred in 54% and 43% of pts, respectively. Serious AEs were reported in 62% vs 51%. Investigator-reported differentiation syndrome grade ≥2 occurred in 7% of pts in both arms, and QTcF prolongation grade ≥3 in 12% of the IVO arm and 7% of the PBO arm. Median time to neutrophil recovery (≥1.0x10⁹/L) was 26 vs 28 days after induction cycle 1 and 29 vs 27 days after cycle 2 in the IVO and PBO arms, respectively. Median time to platelet recovery (≥100x10⁹/L) was 25 vs 26 days after cycle 1 and 36 vs 34 days after cycle 2. The interim OS analysis is planned for Q3 2026 and will evaluate updated outcome and safety data.

Conclusion: The addition of IVO to standard IC in patients with ND IDH1mut AML results in a highly significant and clinically meaningful improvement in EFS.