Background
Phase 1/2, open-label, single-arm dose-escalation and expansion study. N=176 patients with R/R IDH2-mutated AML enrolled in the expansion cohort at the recommended phase 2 dose (RP2D: 100 mg QD). IDH2 mutations (predominantly R140Q and R172K) are present in ~15% of AML; they increase 2-hydroxyglutarate (2-HG) production via gain-of-function neomorphic enzyme activity. Enasidenib (AG-221) selectively inhibits mutant IDH2, reducing 2-HG to near-normal levels and restoring normal hematopoietic differentiation. This was the pivotal single-arm study supporting FDA accelerated approval. Published Blood 2017.
Interventions and follow up
Regimen: Enasidenib 100 mg PO once daily (continuous 28-day cycles) in R/R IDH2-mutated AML
Primary endpoint: ORR (CR + CRh); safety
mFollow up: 6.6 months (efficacy-evaluable cohort)
Primary endpoint: ORR (CR + CRh); safety
mFollow up: 6.6 months (efficacy-evaluable cohort)
Results
ORR (CR+CRh): 30.7% (CR 19.3%, CRh 11.4%)
Full ORR (CR+CRh+CRi+MLFS+PR): 40.3%
mDOR (CR patients): 8.2 months
mOS (all patients): 9.3 months
mOS (CR patients): 19.7 months
2-HG reduction: 90th percentile reduction achieved in most patients within 2–4 weeks
IDH2 R140Q vs R172K: R172K associated with higher CR rates (18% in R140Q vs 53% in R172K)
Full ORR (CR+CRh+CRi+MLFS+PR): 40.3%
mDOR (CR patients): 8.2 months
mOS (all patients): 9.3 months
mOS (CR patients): 19.7 months
2-HG reduction: 90th percentile reduction achieved in most patients within 2–4 weeks
IDH2 R140Q vs R172K: R172K associated with higher CR rates (18% in R140Q vs 53% in R172K)
Adverse events
Overall: Grade ≥3 AEs 77.7%.
Hepatic: Hyperbilirubinemia any grade 81.0%, grade ≥3 12.2% — class effect: enasidenib inhibits bilirubin conjugation → unconjugated (indirect) hyperbilirubinemia; NOT hepatotoxicity; ALT/AST typically normal.
IDH-specific: Differentiation syndrome any grade 14%, grade ≥3 7% — fever, pulmonary infiltrates, hypoxia, edema, rising WBC; requires immediate corticosteroids (dexamethasone 10 mg BID) ± hydroxyurea.
Gastrointestinal: Nausea/vomiting 37%/18% (grade ≥3 2%/1%); diarrhea 27% (grade ≥3 8%).
Infectious: Febrile neutropenia grade ≥3 22%.
Hepatic: Hyperbilirubinemia any grade 81.0%, grade ≥3 12.2% — class effect: enasidenib inhibits bilirubin conjugation → unconjugated (indirect) hyperbilirubinemia; NOT hepatotoxicity; ALT/AST typically normal.
IDH-specific: Differentiation syndrome any grade 14%, grade ≥3 7% — fever, pulmonary infiltrates, hypoxia, edema, rising WBC; requires immediate corticosteroids (dexamethasone 10 mg BID) ± hydroxyurea.
Gastrointestinal: Nausea/vomiting 37%/18% (grade ≥3 2%/1%); diarrhea 27% (grade ≥3 8%).
Infectious: Febrile neutropenia grade ≥3 22%.
Conclusions
Enasidenib achieved a 30.7% CR+CRh rate and 40.3% ORR in heavily pretreated R/R IDH2-mutated AML, supporting FDA accelerated approval. The characteristic class effects of IDH2 inhibition — hyperbilirubinemia (indirect, benign) and differentiation syndrome (potentially severe, requiring prompt steroid treatment) — must be recognized and distinguished from disease progression or hepatic injury.
Key Limitations
Single-arm design: No randomized comparator; the subsequent phase 3 IDHENTIFY trial failed to show an OS advantage over conventional care (HR 1.06).
Short follow-up: Median 6.6 months at the efficacy analysis limits durability assessment.
Mutation subtype: R140Q (most common) had lower CR (18%) than R172K (53%), but R172K is uncommon — modest power for subset estimates.
Class effects: Indirect hyperbilirubinemia (benign) and differentiation syndrome (potentially severe) must be distinguished from progression and hepatic injury.
Short follow-up: Median 6.6 months at the efficacy analysis limits durability assessment.
Mutation subtype: R140Q (most common) had lower CR (18%) than R172K (53%), but R172K is uncommon — modest power for subset estimates.
Class effects: Indirect hyperbilirubinemia (benign) and differentiation syndrome (potentially severe) must be distinguished from progression and hepatic injury.
Clinical Context
FDA granted accelerated approval to enasidenib (Idhifa) for R/R IDH2-mutated AML in August 2017 based on this phase 1/2 ORR. The confirmatory phase 3 IDHENTIFY trial subsequently failed its primary OS endpoint (mOS 8.8 vs 9.6 months, HR 1.06), yet approval was not withdrawn. Ivosidenib (IDH1) and gilteritinib (FLT3) address distinct mutational targets. Allo-SCT in CR1 is recommended for eligible IDH2-mutated AML patients. ELN 2022 includes IDH2 inhibitors in the R/R targeted-therapy algorithm. ESMO-MCBS: not assigned.