Chronic myeloid leukemia (CML) comprises a simple and effective paradigm for generating new insights into the cellular origin, pathogenesis, and treatment of many types of human cancer. In particular, mouse models of CML have greatly facilitated the understanding of the underlying molecular mechanisms and pathogenesis of this disease and have led to the identification of new drug targets that in some cases offer the possibility of functional cure. There are currently three established CML mouse models: the BCR-ABL transgenic model, the BCR-ABL retroviral transduction/transplantation model, and the xenotransplant immunodeficient model. Each has its own unique advantages and disadvantages. Depending on the question of interest, some models may be more appropriate than others. In this chapter, we describe a newly developed xenotransplant mouse model to determine the efficacy of novel therapeutic agents, either alone or in combination. The model facilitates the evaluation of the frequency of leukemic stem cells with long-term leukemia-initiating activity, a critical subcellular population that causes disease relapse and progression, through the utilization of primary CD34(+) CML stem/progenitor cells obtained from CML patients at diagnosis and prior to drug treatment. We have also investigated the effectiveness of new combination treatment strategies designed to prevent the development of leukemia in vivo using BCR-ABL (+) blast crisis cells as a model system. These types of in vivo studies are important for the prediction of individual patient responses to drug therapy, and have the potential to facilitate the design of personalized combination therapy strategies.