Molecular Measurable Residual Disease before Transplantation Independently Predicts Survival and Relapse Risk in Adult KMT2A-Rearranged Acute Myeloid Leukemia

Introduction

Patients with KMT2A rearranged (KMT2A-r) acute myeloid leukemia (AML), occurring in 5-10% of adult cohorts, are assigned to intermediate and adverse risk categories at diagnosis. Allogeneic hematopoietic stem-cell transplantation (allo-HSCT) appears to be the only potentially curative therapeutic option. However, the value of molecular KMT2A-r gene measurable residual disease (MRD) status before allo-HSCT in adult cohorts has rarely been evaluated.

Methods

Two hundred and five adult KMT2A-r AML patients (≥18 years) who achieved complete remission and subsequently underwent allo-HSCT between January 2015 and January 2023 were included. TaqMan-based reverse transcriptase‒polymerase chain reaction was used to detect the KMT2A fusion genes, EVI1 and other genes in bone marrow samples. Fusion and internal control (ABL1) transcript levels were measured synchronously, and quantitative results are expressed as the percentage of fusion genes and ABL1 transcript levels. For KMT2A-PTD, we defined KMT2A-PTD/ABL1 ≥ 0.08% as MRD test positive and KMT2A-PTD/ABL1 < 0.08% as MRD test negative. For other KMT2A-r genes, we defined the detection of any level of KMT2A-r transcripts as MRD-positive. In addition, we used an EVI1 ≥ 8% cutoff value to define abnormal overexpression (EVI1+; as opposed to EVI1 negative [EVI1-]). Eight-color multiparameter flow cytometry (MFC) was used to identify MFC-MRD with a sensitivity of 0.01%, and any measurable level of MRD was considered positive. The endpoints were overall survival (OS), leukemia-free survival (LFS), cumulative incidence of relapse (CIR) and non-relapse mortality (NRM).

Results

Among the 205 included patients, 108 (53%) patients were female, and the median (range) age at HSCT was 37 (18-67) years. KMT2A-r MRD positivity (KMT2A-r MRD) was detected in 91 (44%) patients, and 114 (56%) were KMT2A-r MRD-negative (no KMT2A-r MRD). Patients with KMT2A-r MRD had inferior 3-year OS (57.7% v 82.1%; P<0.001), LFS (50.4% v 78.6%; P<0.001), CIR (39% v 16%; P<0.001) and NRM (11% v 5.3%; P=0.13) than those of patients with no KMT2A-r MRD. According to the multivariate models, molecular MRD independently predicted OS (hazard ratio [HR]=3.77; 95% CI 2.08 to 6.85; P<0.001), LFS (HR=4.06; 95% CI 2.41 to 6.84; P<0.001) and CIR (HR=2.78; 95% CI 1.59 to 5.00; P<0.001) after allo-HSCT. The post-HSCT survival of KMT2A-r AML patients is heterogeneously depending on the type of KMT2A fusion partners, being more favorable in patients with t(9;11) and t(10;11) than those patients with KMT2A-PTD, t(11;19) and t(6;11), and this ability to predict CIR was retained in the KMT2A-r MRD subset rather than in the no KMT2A-r MRD subset. Deregulated EVI1 expression at diagnosis defined a dismal prognostic subset in the KTM2A-r MRD group, with 3-year OS rates of 41.9% and 70.9% (P=0.032), LFS of 32.4% and 61.0% (P=0.027) and CIR of 64% and 29% (P=0.009) for EVI1+ and EVI1- patients, respectively. The combination of molecular MRD and MFC-MRD stratified four risk groups according to differential OS (HR=1.27 for KMT2A^negative/MFC^positive [P=0.70]; HR=2.95 for KMT2A^positive/MFC^negative [P<0.001]; HR=3.05 for KMT2A^positive/MFC^positive [P=0.003], with KMT2A^negative/MFC^negative as reference), LFS (HR=1.38 for KMT2A^negative/MFC^positive [P=0.57]; HR=3.05 for KMT2A^positive/MFC^negative [P<0.001]; HR=3.14 for KMT2A^positive/MFC^positive [P=0.001], with KMT2A^negative/MFC^negative as reference) and CIR (HR=2.09 for KMT2A^negative/MFC^positive [P=0.17]; HR=3.02 for KMT2A^positive/MFC^negative [P<0.001]; HR=4.30 for KMT2A^positive/MFC^positive [P<0.001], with KMT2A^negative/MFC^negative as reference), suggesting that flow cytometry-based MRD detection provided more refined prognostic information on molecular MRD for post-HSCT outcomes.

Conclusions

Residual KMT2A-r prior to HSCT independently predicted the risk of survival and relapse, donor lymphocyte infusion or posttransplant maintenance therapies should be considered.

No relevant conflicts of interest to declare.