Interim phase II results from the ongoing AG221-AML-005 study on enasidenib plus azacitidine in patients with newly diagnosed acute myeloid leukemia with IDH2 mutations

Isocitrate dehydrogenase 2 (IDH2) mutations are one of the most common mutations in AML, occurring in 8–19% of patients.¹ Enasidenib (ENA) is an oral, small-molecule inhibitor of mutant (m) IDH2 protein approved in the United States (US) for use in adult patients with relapsed/refractory mIDH2 AML. In a large phase I/II study, a cohort of newly diagnosed (ND) AML patients receiving ENA monotherapy showed a complete remission (CR) rate of 18%, an overall response rate (ORR) of 30.8% and a median overall survival (mOS) of 11.3 months.²

Similarly, azacitidine (AZA), a hypomethylating agent, is approved for the treatment of newly diagnosed patients with AML not eligible for intensive chemotherapy, inducing ORRs of 19–30% with a mOS of about 10 months.³

Given the enhanced effect on cell differentiation obtained in vitro with the combination of ENA + AZA, a phase I/II study (AG221-AML-005) was initiated to evaluate the combination of ENA + AZA in patients with mIDH2 ND-AML who are not candidates for intensive chemotherapy (NCT02677922).

The results from the phase Ib portion of the study were presented at the 23^rd Congress of the European Hematology Association, at the 61^st American Society of Hematology Meeting & Exposition,  Courtney DiNardo from the University of Texas, MD Anderson Cancer Center, Houston, US, reported the interim results from the randomized, phase II portion of the study.⁴

Study design⁴

• Adult patients (age ≥ 18 years) with mIDH2 ND-AML who were ineligible for intensive chemotherapy and with no prior hypomethylating agent treatment were randomized (N= 101) in a 2:1 ratio to receive ENA + AZA (n= 68) or AZA only (n= 33) in repeated 28-day cycles
• All patients received subcutaneous AZA 75 mg/m²/day from Day one to Day seven of each treatment cycle; patients randomized to ENA + AZA also received continuous ENA 100 mg/day
• The primary endpoint was ORR
• The secondary endpoints include CR, overall survival (OS), event-free survival (EFS) and safety

Patient characteristics

• Median ages were 74 years (range, 62–85) in the ENA + AZA arm and 75 years (range, 57–85) in the AZA-only arm
• About three quarters of the patients presented the IDH2-R140 mutation (75% in the ENA + AZA arm and 73% in the AZA-only arm). The most common co-occurring mutations were DNMT3A and ASXL1
• The majority of patients had intermediate-risk cytogenetics (80% in the ENA + AZA arm and 90% in the AZA-only arm)

Results⁴

Efficacy

• At data cut-off (August 2019), 22 patients were still receiving their randomized treatment
• Response rates were significantly higher in the ENA + AZA arm vsAZA-only arm:
  • ORR 71% (95% confidence interval [CI], 58─81) vs 42% (95% CI, 26─61) (p= 0.0064)
  • CR 53% (95% CI, 41─65) vs 12% (95% CI, 3─28) (p=0.0001)
• The median number of treatment cycles was 10 (range, 1–26) in the ENA + AZA arm and six (range, 1–28) in the AZA-only arm
• The time to first response was approximately two months. In responding patients, the median duration of response was 24.1 months (95% CI, 11.1─not reached) vs 12 months (95% CI, 2.8─14.6) in the ENA + AZA arm vs  AZA-only arm
• The most common reason for study discontinuation was progression of disease (31% in the ENA + AZA arm and 52% in the AZA-only arm). Discontinuation due to adverse events (AEs) was rare in both arms (6%)
• At a median follow-up of 14 months:
  • Median OS was 22 months in the ENA + AZA arm vs 22.3 months in the AZA-only arm (hazard ratio [HR]= 0.99 [95% CI, 0.52─1.87], p= 0.9686)
    • Median OS was not reached in the ENA + AZA arm who achieved CR
    • Median EFS was 17.2 months in the ENA + AZA arm vs 10.8 months in the AZA-only arm (HR= 0.59 [95% CI, 0.3─1.17], p= 0.1278)
  • Maximal mIDH2 variant allelic frequency (VAF) reduction from baseline was stronger in the ENA + AZA arm (-83.4%) compared to the AZA-only arm (-17.7%, p= 0.0008), with more than 20% of patients in CR in the ENA + AZA arm having complete mIDH2 VAF clearance
  • Median maximal 2-hydroxyglutarate reduction, as a marker for inhibition of IDH2 enzyme activity, was significantly higher in the ENA + AZA arm vs the AZA-only arm (-97.8% vs -54.3%, p= 0.0004)

Safety

• The most common treatment-emergent AEs (TEAEs) in the ENA + AZA arm vs AZA-only arm were:
  • Thrombocytopenia: 62% vs 44%
  • Nausea: 69% vs 38%
  • Anemia: 53% vs 44%
  • Vomiting: 49% vs 47%
• Treatment-related Grade 3–4 TEAEs in the ENA + AZA arm vs AZA-only arm were similar, except for the occurrence of IDH differentiation syndrome (IDH-DS) in the combination arm:
  • Thrombocytopenia: 37% vs 19%
  • Neutropenia: 35% vs 22%
  • Anemia: 19% vs 22%
  • Febrile neutropenia: 15% vs 16%
  • IDH-DS: 10% vs 0%
  • Infections: 18% vs 31%
• In the first 60 days, five (7%) patients died in the ENA + AZA arm and one died (3%) in the AZA-only arm

Conclusions⁴

• The combination of ENA and AZA was effective, with a significant improvement in ORR and CR rates
• Although the median OS was similar between treatment arms, for ENA + AZA patients who achieved CR median OS was not reached
• The combination treatment was well tolerated:
  • Patients should be monitored for infections and anti-infective prophylactic treatments should be considered; in case of suspected IDH-DS, corticosteroids should be initiated