A Randomised Comparison of CPX-351 Versus Standard Daunorubicin and Cytarabine Plus Fractionated Gemtuzumab Ozogamicin in Older Adults with AML without Adverse Risk Cytogenetics: Results of the NCRI AML18 Trial

Background: In version 1 (v1) of the NCRI AML18 trial we demonstrated that a fractionated, schedule of Gemtuzumab ozogamicin (GO) combined with daunorubicin/cytarabine (DAGO2) reduced Measurable Residual Disease (MRD) and improved overall survival (OS) in older acute myeloid leukaemia (AML) patients with non-adverse risk cytogenetics compared to a single dose (Freeman, Blood 2024). The survival benefit was apparent in those with myelodysplasia (MDS)-related gene mutations as well as those with de novo AML and was dependent upon delivery of transplant in CR1.

Methods: In v2 of the trial (June 2019-Dec 2022) we compared DAGO2 with the liposomal formulation of daunorubicin/cytarabine, CPX-351 (CPX) (1:2 randomisation) for up to 3 courses in AML patients >60 years without known adverse risk cytogenetics. Following the EMA approval of CPX for secondary AML, these patients were excluded from trial entry. The primary endpoint was 5yr event-free survival (EFS) with a target of 506 patients - however, recruitment was affected by the COVID pandemic and the trial closed early. CPX induction consisted of 100 units/m² (100 mg/m² cytarabine and 44 mg/m² daunorubicin) administered on days 1, 3, and 5. DAGO2 induction was as previously reported. Patients with abnormal liver function could enter the randomization and receive DA alone if so randomized.

Flow cytometric MRD assessment was performed on recovery from course 1. In both arms, patients who were not in MRD-negative remission (CR+CRi) could enter a second course randomization for treatment intensification or not. Results of intensification in the DAGO2 arm with DA-Cladribine or FLAG-Ida have been reported (Russell, ASH 2023). In the CPX arm, the intensification randomisation compared 3 (days 1, 3, 5) vs 2 (days 1 and 3) doses of 100 units/m² CPX. If MRD-negative, patients randomised to CPX received a second course of 100 units/m² on days 1 and 3; those randomised to DAGO2 received DA(3+8) without GO.

Results: Treatment arms (CPX-351, n=295, DAGO2, n=144, including 7 with DA alone) were balanced for baseline characteristics. Median age was 68 yrs with 33% >70 yrs; 81% had de novo AML, 7% clinical secondary AML, 12% high-grade MDS. 8% had adverse-risk cytogenetics. FLT3 mutations were present in 20%, NPM1 mutations in 24%, TP53 mutations in 4% and MDS-related gene mutations in 61%. Median follow-up was 35 months.

Early all-cause mortality was comparable at day 30 between DAGO2 and CPX (at 4% and 7%, HR 1.67, 95% CI 0.66-4.26, p=0.27) but was higher with CPX at day 60 (4% vs 11%, HR 2.80, 95% CI 1.14-6.86, p=0.019).

Post-course 1 response rates were greater after DAGO2 (CR+CRi, 60% vs 47.5%, OR 0.61 95% CI 0.41-0.91, p=0.016; CR, 50.5% vs 39% p=0.02). MRD results were available for 85% of patients in CR/CRi post course 1. Of these, more attained MRD negativity in the DAGO2 arm (85% vs 66% for CPX, OR 0.35 95% CI 0.15-0.76, p=0.004). Following course 2, overall response (CR+CRi) was 85% for DAGO2 vs 78% for CPX (OR 0.64, 95% CI 0.39-1.09, p=0.095) and CR rates were 78.5% vs 68% (OR 0.59 95% CI 0.37-0.94, p=0.024).

We observed better 3yr EFS (34% vs 27%, HR 0.73 95% CI 0.57-0.93, p=0.012) and 3yr OS (52% vs 35%, HR 0.62, 95% CI 0.46-0.83, p=0.001) with DAGO2 compared with CPX. Relapse-free survival did not differ significantly by randomisation (HR 0.75 95% CI 0.57-1.01, p=0.058).

In a stratified analysis, CPX did not appear to provide a survival benefit over DAGO2 in patients with MDS-related mutations (HR 1.36, 95% CI 0.94-1.98, p value for heterogeneity 0.348) and was associated with poorer survival in patients with NPM1 (HR 2.83, 95% CI 1.17-6.82) and FLT3 mutations (HR 2.36, 95% CI 1.10-5.07).

37% of patients were transplanted in CR1 and this did not differ by randomization (37.5% and 36.6% for DAGO2 and CPX respectively) - neither did OS following CR1 transplant (57% DAGO2 vs 64% CPX at 3 years, HR 0.98, 95% CI 0.52-1.85)

Conclusions: In this population of fit older patients, primarily with clinical de novo AML and intermediate risk cytogenetics, CPX did not improve survival outcomes in comparison with DAGO2, overall or in any subgroup, including that of MDS-related mutations. DAGO2 induction was associated with higher response rates and higher rates of MRD-negativity. Notably, transplant rates in both arms were increased, compared to the AML18v1 experience as were survival rates with DAGO2.