Posttransplant MRD and T-cell chimerism status predict outcomes in patients who received allografts for AML/MDS

Abstract

Allogeneic stem-cell transplant allows for the delivery of curative graft-versus-leukemia (GVL) in patients with acute myeloid leukemia/myelodysplasia (AML/MDS). Surveillance of T-cell chimerism, measurable residual disease (MRD) and blast HLA-DR expression may inform whether GVL effectiveness is reduced. We report here the prognostic impact of these biomarkers in patients allografted for AML/MDS. One hundred eighty-seven patients from FIGARO, a randomized trial of reduced-intensity conditioning regimens in AML/MDS, were alive and relapse-free at the first MRD time-point and provided monitoring samples for flow cytometric MRD and T-cell chimerism, requested to month+12. Twenty-nine (15.5%) patients had at least 1 MRD-positive result posttransplant. MRD-positivity was associated with reduced overall survival (OS) (hazard ratio [HR], 2.18; P = .0028) as a time-varying Cox variable and remained significant irrespective of pretransplant MRD status in multivariate analyses (P < .001). Ninety-four patients had sequential MRD with T-cell chimerism results at months+3/+6. Patients with full donor T-cell chimerism (FDTC) had an improved OS as compared with patients with mixed donor T-cell chimerism (MDTC) (adjusted HR=0.4; P = .0019). In patients with MDTC (month+3 or +6), MRD-positivity was associated with a decreased 2-year OS (34.3%) vs MRD-negativity (71.4%) (P = .001). In contrast, in the group with FDTC, MRD was infrequent and did not affect the outcome. Among patients with posttransplant MRD-positivity, decreased HLA-DR expression on blasts significantly reduced OS, supporting this as a mechanism for GVL escape. In conclusion, posttransplant MRD is an important predictor of the outcome in patients allografted for AML/MDS and is most informative when combined with T-cell chimerism results, underlining the importance of a GVL effect in AML/MDS.