Predictive impact of residual disease detected using multiparametric flow cytometry on risk stratification of paediatric acute myeloid leukaemia with normal karyotype

Introduction Residual disease (RD) detected using multiparametric flow cytometry (MFC) is an independent predictive variable of relapse in acute myeloid leukaemia (AML). However, RD thresholds and optimal assessment time points remain to be validated. Material and methods We investigated the significance of RD after induction therapy in paediatric AML with normal karyotype between June 2008 and June 2018. Bone marrow samples from 73 patients were collected at the end of the first (BMA-1) and second (BMA-2) induction courses to monitor RD using MFC. Results Presence of RD after BMA-1 and/or BMA-2 correlated with poor relapse-free (RFS) and overall survival at 0.1% RD cutoff level. Receiver operating characteristic curve showed that RD cutoff levels of 1.3% and 0.5% after BMA-1 and BMA-2, respectively, predicted events with the highest sensitivity and specificity. In multivariable analysis, RD after BMA-2 was the strongest independent risk predictor for poor RFS (hazard ratio 2.934; 95% confidence interval: 1.106-7.782; P = .031). Conclusions Our study therefore suggests that an RD level >= 0.5% after BMA-2 has a significant predictive impact on the prognosis of AML patients having normal karyotype and thus guide the stratification of treatment strategies.