Peritransplant lysine methyltransferase 2a–rearranged measurable residual disease dynamics as a prognostic marker in adult acute myeloid leukemia

Abstract

Background

This study examined the impact of measurable residual disease (MRD) dynamics in adults with lysine methyltransferase 2a rearrangement (KMT2Ar) in acute myeloid leukemia (AML) during the peritransplant period (before and early after allogeneic hematopoietic stem cell transplantation [HSCT]).

Methods

This study involved 144 adult patients with AML with KMT2Ar who underwent HSCT between 2015 and 2024. Patients were enrolled if they survived without relapse for at least 3 months post-HSCT. MRD was measured via real-time quantitative polymerase chain reaction.

Results

The cohort was categorized into four groups on the basis of MRD status before and after HSCT, respectively: MRD negative/MRD negative (n = 81), MRD negative/MRD positive (n = 2), MRD positive/MRD negative (n = 43), and MRD positive/MRD positive (n = 18). Significant differences were found in 3-year overall survival (OS), leukemia-free survival (LFS), and cumulative incidence of relapse (CIR) (all p < .001), with the MRD-negative/MRD-negative group having the best outcomes (88% OS, 85% LFS, and 10% CIR), whereas the MRD-negative/MRD-positive group (0% OS, 0% LFS, and 100% CIR) and MRD-positive/MRD-positive group (29% OS, 31% LFS, and 69% CIR) had the worst outcomes. The MRD-positive/MRD-negative group had an intermediate prognosis (63% OS, 58% LFS, and 32% CIR). Peri-HSCT MRD dynamics served as an independent factor, which distinguished patient groups with differential risk probabilities of mortality and relapse. C-statistic values were greater when peri-HSCT MRD dynamic testing was used than when isolated pre- or post-HSCT MRD was used to predict CIR (0.68 vs. 0.58 vs. 0.56), LFS (0.74 vs. 0.69 vs. 0.64), and OS (0.73 vs. 0.68 vs. 0.66).

Conclusions

Compared with single–time point MRD assessments, peritransplant MRD dynamics confer superior prognostic accuracy.