Minimal Residual Disease (MRD) Detection By Flow Cytometry during Aplasia and Post-Induction Is an Independent Prognostic Parameter for Relapse-Free Survival in AML

The identification of patients at high risk of relapse is crucial for treatment stratification in acute myeloid leukemia (AML). Persistence and reoccurrence of AML cells with aberrant phenotypes can be assessed by flow cytometry at high sensitivity in more than 95% of patients. Therefore, assessment by flow cytometry enables determination of MRD status in patients without suitable molecular markers (e.g. NPM1). Previously, we demonstrated that early detection of MRD during aplasia is associated with shorter relapse-free survival (RFS) after intensive induction chemotherapy (Köhnke T et al., Leukemia 2015). Here, we aimed to validate these findings in an independent patient cohort. In addition, we assessed the prognostic value of a combined MRD determination during aplasia and post induction for improved prediction of relapse-free survival (RFS).

Flow cytometry MRD was analyzed in patients with AML (excluding APL) diagnosed between 2012 and 2016 receiving intensive induction chemotherapy (sHAM or 7+3). MRD flow analysis was performed using a NAVIOS flow cytometer. Presence ≥0.1% Leukemia-associated immunophenotype (LAIP)-positive cells (identified by comparison with bone marrow from healthy donors) during aplasia or post-induction defined MRD positivity. Kaplan-Meier estimators and log-rank tests were used to retrospectively analyze survival data. Cox's proportional hazards regression model was used to determine the influence of individual factors in multivariate analyses. Pearson's chi square test was used to compare categorical variables between MRD groups.

A total of 105 patients were included. In 3 cases (2.9% of all cases), no LAIP could be identified, and these patients were excluded from further analyses. MRD assessment during aplasia was available in 87 cases. Regarding flow MRD assessment post induction, patients refractory to induction therapy were excluded. Therefore, 72 patients had flow MRD assessments available post induction therapy. 59 patients achieving CR or CRi after induction therapy had flow MRD assessments available at both time points.

MRD positivity during aplasia was associated with refractory disease (36.2% vs. 10.0% for MRD pos. vs. MRD neg., p=0.017) and shorter median event-free survival (EFS, 5.8 vs. 15.7 months, p=0.011). Importantly, early blast clearance by morphology had no prognostic value if flow MRD assessment was available (HR for EFS 2.0, p=0.03 for flow MRD vs. HR 1.6 and p=0.17 for Blast clearance by morphology). However, differences in median RFS for MRD^pos vs. MRD^neg patients (10.3 vs. 20.5 months) for assessment during aplasia have not reached significance yet (p=0.135). MRD^pos patients post induction had significantly shorter median EFS (10.8 vs. 15.7 months, p=0.032) and RFS (7.7 months vs. not reached, p=0.007) than MRD^neg patients. Using a combination strategy for aplasia and post induction assessments, we grouped patients with MRD positivity at both timepoints (confirmed MRD positivity) vs. all others (MRD^neg or MRD^uncertain). Confirmed MRD positivity was associated with significantly shorter median EFS (6.7 vs. 15.7 months, p=0.003) as well as RFS (4.4 months vs. not reached, p=0.002). Confirmed MRD positivity was an independent predictor of EFS and RFS (HR 2.7 and 2.8, p=0.017 and p=0.018, respectively).

Taken together, patients with confirmed two-time flow MRD positivity showed lower EFS and RFS compared to MRD^neg or MRD^uncertain patients, respectively. Identification of patients with poor treatment response was improved in comparison with cytomorphologic assessment or independent flow MRD assessments at both timepoints alone. Using this approach, early identification of patients with adverse prognosis is possible in >95% of AML patients. MRD flow will be prospectively evaluated as a therapy stratification tool in patients with AML receiving intensive induction chemotherapy.