Acute myeloid leukemia (AML) is a blood cancer that has remained incurable for the majority. The big challenge and unmet need in AML is the high incidence of relapse after reaching complete remission. There is clear evidence that the metabolism in AML is frequently altered, and also further adapts when patients egress from de novo to relapsed disease. There is strong heterogeneity in metabolism between patients, which not only entails the different tumor clones between and within patients themselves, but also extends to the tumor immune microenvironment, comprised of e.g. macrophages, T cells, NK cells, and stromal cells. It is increasingly becoming clear that the tumor immune microenvironment not only differs from the normal BM microenvironment, but can also act as driver of disease. This provides alternative treatment options for therapy, not only aimed at the tumor cell itself but also aimed at the tumor-supportive microenvironment. These levels of heterogeneity are currently not, but should be, taken into account in diagnosis and treatment strategies. Personalized tools to be able to do so, shall need to be developed further.
The MetaboTargetAML consortium will generate, based on existing and novel data, the first longitudinal metabolome roadmap at the single cell and spatial level of AML, providing insights into clonal dynamics, immune landscapes and metabolic states of specific cell populations in response to standard-of-care and novel targeted treatments. This roadmap will be used to generate AI-driven models to identify patient subtype-specific metabolic vulnerabilities, to be validated in ex vivo and in vivo models, and to be implemented and prospectively validated in spectral flow cytometry panels in routine diagnostics. Ultimately, we aim to identify novel (metabolic) candidates that can be utilized as therapeutic targets and/or biomarkers in routine diagnostics spectral flow pipelines for disease prediction and patient stratification.