The application of machine learning (ML) models for the detection of measurable residual disease (MRD) in acute myeloid leukemia (AML) would have the potential advantages of being objective, reproducible, and fast. Vahedi and colleagues applied a fully automated ML approach for MRD detection by multiparameter flow cytometry with the aim of it performing at least as well as manual MRD gating, while reducing the cost and time of analysis. Their work constitutes a relevant contribution to the development of a fully automated ML approach in multiparameter flow cytometry residual leukemia monitoring in AML.

Disease relapse remains the most common reason for transplant failure in patients with myelodysplastic syndromes (MDS) undergoing allogeneic hematopoietic stem cell transplantation. Maintenance therapies with hypomethylating agents are studied to reduce the risk of relapse by strengthening the graft-versus-leukemia effect. A retrospective study by Smallbone and colleagues investigating oral decitabine/cedazuridine maintenance in the post-transplant setting for patients with high-risk MDS demonstrates a promising efficacy profile with myelosuppression as the primary toxicity.