Human myeloid leukemias are life-threatening disorders. Although with current therapies the bulk of the tumor is readily eradicated, it is particularly the rare population of leukemic stem cells (LSCs) that is relatively quiescent and is very difficult to target. As a consequence, relapse of the disease occurs frequently resulting in poor survival rates. Nevertheless, remarkable differences do exist between subsets of patients. It is important to realize that leukemias are rather heterogeneous disorders that can be caused by a multiplicity of genetic and epigenetic changes. One of the most important challenges in the field today lies in establishing which of the differences between LSCs and normal hematopoietic stem cells (HSCs) provide attractive targets for the identification, targeting and ultimately the selective eradication of LSCs. Given the heterogeneity of the disorder, it will be important to address these aspects in detail in a leukemia subtype-specific manner.Central to this proposal is the establishment of a human xenograft mouse leukemia “clinic” in which all important human leukemia subtypes will be represented. LSCs from patients as well as from cord blood and bone marrow model systems that we have generated will be equipped with luciferase-GFP tracers in order to allow detection of leukemic development in alive immunodeficient mice, as well as with unique barcodes that will allow clonal tracking. Within this mouse clinic, in a leukemia subtype-specific manner, we will be able to: 1) identify (novel) leukemic stem cell markers and evaluate their targetability; 2) evaluate whether stem cell intrinsic versus extrinsic signaling can be used as targets in alternative approaches to eradicate LSCs, and 3) study clonal evolution from de novo to relapsed leukemia. Our studies should provide insight into the biology of leukemia and novel rational approaches to treat leukemia patients more successfully.