Background
Phase 3, double-blind, placebo-controlled randomized controlled trial (AGILE). N=146 patients with newly diagnosed IDH1-mutated AML ineligible for intensive induction (age ≥75, or age 18–74 with ECOG PS 2–3, cardiac/hepatic comorbidities, or creatinine clearance <45 mL/min). IDH1 mutations drive oncogenesis via 2-hydroxyglutarate overproduction and epigenetic dysregulation; ivosidenib selectively inhibits mutant IDH1, reducing 2-HG and enabling differentiation. Randomized 1:1 to ivosidenib + azacitidine vs placebo + azacitidine. Published NEJM 2022.
Interventions and follow up
Arm A: Ivosidenib 500 mg PO once daily continuously + azacitidine 75 mg/m² SC/IV days 1–7 of each 28-day cycle
Arm B: Placebo PO once daily + azacitidine 75 mg/m² SC/IV days 1–7 of each 28-day cycle
Primary endpoint: Overall survival (OS)
Median follow-up: 12.4 months
Arm B: Placebo PO once daily + azacitidine 75 mg/m² SC/IV days 1–7 of each 28-day cycle
Primary endpoint: Overall survival (OS)
Median follow-up: 12.4 months
Results
mOS: 24.0 vs 7.9 months (ivosidenib+aza vs placebo+aza), HR 0.44 (95% CI 0.28–0.68), P<.001
CR rate: 47.2% vs 14.9%, P<.001
ORR (CR+CRh+CRi+PR): 62.5% vs 18.9%
EFS: 12.5 vs 1.6 months, HR 0.33
12-month OS rate: 60.3% vs 36.1%
CR rate: 47.2% vs 14.9%, P<.001
ORR (CR+CRh+CRi+PR): 62.5% vs 18.9%
EFS: 12.5 vs 1.6 months, HR 0.33
12-month OS rate: 60.3% vs 36.1%
Adverse events
Class effects: Differentiation syndrome any grade 25.3% vs 7.4%, grade ≥3 3.8%; QT prolongation any grade 24.1%, grade ≥3 7.6%
Hematologic/other: Grade ≥3 AEs 82.7% vs 82.3% (balanced); neutropenia grade ≥3 33.3% vs 37.0%; febrile neutropenia grade ≥3 27.8% vs 24.7%; nausea/diarrhea any grade ~30–40% with low grade ≥3; discontinuation due to AEs 14.1% vs 17.5%
Hematologic/other: Grade ≥3 AEs 82.7% vs 82.3% (balanced); neutropenia grade ≥3 33.3% vs 37.0%; febrile neutropenia grade ≥3 27.8% vs 24.7%; nausea/diarrhea any grade ~30–40% with low grade ≥3; discontinuation due to AEs 14.1% vs 17.5%
Conclusions
Ivosidenib + azacitidine significantly improved OS (24.0 vs 7.9 months, HR 0.44) and CR rates (47.2% vs 14.9%) in newly diagnosed IDH1-mutated AML ineligible for intensive chemotherapy, establishing this doublet as a standard of care for this biomarker-selected population. The three-fold OS improvement represents one of the largest survival gains in elderly/unfit AML.
Key Limitations
Small N=146 limits power for subgroup analyses. Differentiation syndrome (25%) requires vigilance and prompt corticosteroids; QT prolongation necessitates ECG monitoring and drug-interaction assessment. IDH1 mutation testing must be validated and rapid before starting, and not all mutations respond equally. AGILE enrolled only unfit patients, so ivosidenib + intensive induction is untested. Concurrent mutations (e.g., FLT3, TP53) affect prognosis and were not the focus of the design.
Clinical Context
AGILE led to FDA approval of ivosidenib + azacitidine for newly diagnosed IDH1-mutated AML in adults ineligible for intensive induction (May 2022). It is one of several precision-medicine frontline AML approvals alongside midostaurin + 7+3 (RATIFY, FLT3-mutated fit patients). In IDH1-mutated AML eligible for intensive therapy, ivosidenib +/- 7+3 or allo-SCT considerations apply; AGILE pertains to unfit patients only. IDH1 clearance at end of induction is a favorable biomarker for CR durability. ESMO-MCBS: not assigned.