Implementing Measurable Residual Disease Assessment in Acute Myeloid Leukemia: Methods, Clinical Decisions, and a Tiered Framework for Resource-Limited Settings

Measurable residual disease (MRD) is a well-established prognostic biomarker in acute myeloid leukemia (AML), although randomized evidence that MRD-guided monitoring improves overall survival is so far limited to prespecified molecular subgroups. This review compares current MRD methods (multiparameter flow cytometry, polymerase chain reaction, and next-generation sequencing), outlines their benefits and drawbacks, and discusses the prognostic significance of MRD at different time points during treatment. We summarize the 2025 ELN-DAVID recommendations, which address the integration of MRD status into decisions on treatment escalation or de-escalation and on the timing of allogeneic hematopoietic cell transplantation. We discuss the time points at which MRD should be assessed, including after intensive chemotherapy, after hypomethylating agents combined with venetoclax, and around transplantation. We also discuss open questions in FLT3-mutated and NPM1-mutated disease, including low-level MRD and treatment cessation. Factors affecting the accuracy of MRD assessment are also covered, such as sample hemodilution, peripheral blood testing, and the uneven standardization of flow cytometric and molecular assays. Finally, we consider how to implement MRD assessment when infrastructure, reagents, or downstream interventions are limited, and propose a tiered framework that matches modality to local capability.