Optimizing induction regimens in younger patients with AML: The UK AML19 trial

The phase III UK National Cancer Research Institute AML19 trial (ISRCTN78449203) aimed to determine the optimal induction chemotherapy regimen in younger patients with newly diagnosed acute myeloid leukemia (AML) without adverse risk cytogenetics.¹ Here, we summarize results from this trial, published by Russell et al.¹ in the Journal of Clinical Oncology.

Study design and patient population¹

• In total, 1,479 patients with newly diagnosed AML or myelodysplastic syndrome with excess blasts 2 were randomized to receive either fludarabine, cytarabine, granulocyte colony-stimulating factor, and idarubicin (FLAG-Ida; n = 738) or daunorubicin and cytarabine (DA, n = 741).
  • Of these, 1,033 were randomized to receive single dose gemtuzumab ozogamicin (GO) on Day 1 (GO1, n = 514) or a fractionated dosing schedule on Day 1 and Day 4 (GO2, n = 519).
• Median age was 51.5 years, with 14% of patients aged >60 years.
• Overall, 30%, 26%, and 12% of patients had NMP1-mutated, FLT3-mutated, or core binding factor-AML, respectively.
• The primary endpoint was overall survival (OS)

Key findings¹

• There was no difference between response rates and survival outcomes in patients randomized to GO1 or GO2; therefore, these groups were combined for comparisons between FLAG-Ida + GO and DA + GO.

Induction response

• Both response rates after two courses, and 30- and 60-day early mortality rates, were similar between treatment arms (Table 1).
  • More patients treated with DA + GO underwent allogeneic hematopoietic stem cell transplantation vs FLAG-Ida + GO (Table 1).

Table 1. Response, early mortality, and allo-HSCT rates between DA + GO vs FLAG-Ida + GO in the AML19 trial*

Survival outcomes

• When compared with DA + GO, FLAG-Ida + GO reduced relapse, improved event-free survival (EFS), and relapse-free survival (Figure 1).
• OS was similar between treatment arms, and the cumulative incidence of death in remission was higher with FLAG-Ida + GO vs DA + GO (Figure 1).

Molecular subgroup analysis

• Treatment with FLAG-Ida + GO improved OS vs DA + GO in patients with NPM1 mutations, and in those with FLT3 mutations, but not in patients with core binding factor-AML (Figure 1).
• FLAG-Ida + GO was also associated with achievement of measurable residual disease negativity after course two vs DA + GO in the peripheral blood (88% vs 77%; p = 0.02) and the bone marrow (56% vs 37%; p = 0.004).

Figure 1. 3-year survival outcome rates by treatment arm in the AML19 trial in A all patients and B within molecular subgroups* 

AML, acute myeloid leukemia; CBF-AML, core binding factor-AML; CIDCR, cumulative incidence of death in remission; CIR, cumulative incidence of relapse; DA, daunorubicin and cytarabine; EFS, event-free survival; FLAG-Ida, fludarabine, cytarabine, granulocyte colony-stimulating factor, and idarubicin; GO, gemtuzumab ozogamicin; mut, mutated; OS, overall survival; RFS, relapse-free survival.
*Data from Russell, et al.¹

Safety

• After course two, FLAG-Ida + GO was associated with a longer median time to neutrophil recovery (40 vs 27 days; p <0.001) and time to platelet recovery (46 vs 28 days; p <0.001) vs DA + GO.
• When comparing GO1 and GO2, time to neutrophil recovery (median, 29 vs 29 days; p = 0.23) and to platelet recovery (median, 27 vs 29 days; p = 0.07) were similar.