Measurable residual disease assessment as an early end point of treatment efficacy in acute myeloid leukemia: a multinational PETHEMA study of 2623 patients treated with intensive therapy.

Measurable residual disease (MRD) is an early endpoint of treatment efficacy in some hematological malignancies, but several questions remain unanswered to endorse its use in acute myeloid leukemia (AML). We leveraged PETHEMA real-world outcomes of 2623 newly-diagnosed AML patients fit for intensive therapy and in complete remission (CR) after one cycle, to investigate if MRD assessment using flow cytometry is more sensitive than morphological CR, what is the optimal MRD cutoff, the individual-patient-level association between MRD negative CR and overall survival (OS), and if MRD negative rates after different induction regimens associate with different OS. MRD was detected in 68% of patients and presence of MRD levels ≥0.01-<0.1% associated with inferior OS. MRD negative CR using the 0.01% cutoff showed independent prognostic value and was significantly correlated with OS at the individual patient-level (Global odds ratio 1.23; 95% CI, 1.02-1.40). However, significance was lost in specific subgroups defined by genetic risk and post-consolidation treatment. The relationship between treatment effect in MRD negative CR rates and OS after different induction regimens was generally observed. Altogether, these results support formal studies to investigate the role of MRD as an early endpoint of treatment efficacy in specific subgroups of AML patients.