Impact of detectable measurable residual disease on umbilical cord blood transplantation.

Abstract

The impact of measurable residual disease (MRD) on cord blood transplantation (CBT) outcomes has remained debated. To address this issue, we assessed the impact of measurable MRD at CBT on outcomes in large cohort of patients with acute leukemia. Inclusion criteria included adult patients with acute myeloid (AML) or acute lymphoblastic leukemia (ALL), CBT as first allo-HCT in first or second complete remission (CR) at transplantation, and known MRD status at the time of CBT. Data from 506 patients were included in the analysis. Among them, 317 patients had AML and 189 had ALL. Positive MRD was reported in 169 (33%) patients while the remaining 337 patients were MRD negative at CBT. At 2 years, relapse incidence was 18% in patients with MRD negativity vs 33% in those with MRD positivity at transplantation (P < .001). Two-year leukemia-free survival (LFS) and overall survival (OS) were 57% and 60%, respectively, in MRD negative patients, vs 38% (P < .001) and 48% (P = .004), respectively, in those with MRD positivity. There was no interaction between the impact of MRD on OS and LFS and diagnosis (ie, ALL vs AML), single or double CBT, and reduced-intensity or myeloablative conditioning. On multivariate analysis, MRD positivity was associated with a higher risk of relapse (HR = 1.8, P = .003), comparable non-relapse mortality (P = .44), worse LFS (HR = 1.4, P = .008) and a trend towards worse OS (HR = 1.3, P = .065). In conclusion, these data suggest that novel strategies that are aiming to achieve MRD negativity at CBT are needed for leukemic patients with positive MRD pre-CBT.