Acute myeloid leukemia (AML), is a heterogeneous clonal disorder characterized by an accumulation of malignant myeloid progenitors in the bone marrow (BM), hindering normal hematopoiesis. AML exhibits dramatic heterogeneity in terms of cytogenetics, morphology, and chemotherapeutic sensitivity. Therefore, the investigation of novel, efficacious AML therapeutics will require advanced preclinical in vivo model systems, capable of recapitulating patient specific disease heterogeneity, and induction chemotherapy outcomes. A major focus and eventual outcome of this work was the establishment and development of a more clinically relevant mouse xenograft model of patient AML, that efficiently harbors patient derived xenografts (PDXs), and unlike more v prevalent SCID models can tolerate more clinically relevant doses of DNA damaging induction chemotherapy. We examined the functional utility of our newly established, advanced AML PDX model to confirm our in vitro findings that perturbation of methionine (Met) / S-adenosylmethionine (SAM) metabolism is uniquely cytotoxic to MLLrearranged (MLL-R) leukemic cells, in vivo. We demonstrate here that perturbation of Met/SAM metabolism decreases intracellular methylation potential, alters global histone methylation dynamics, impairs the expression and function of the H3K79 methyltransferase DOT1L, and induces apoptosis in MLL-R leukemic cells. We show a significant extension in the survival of mice harboring aggressive patient MLL-R leukemias, when treated with a pharmacologic inhibitor of Met/SAM metabolism and induction therapy, as compared to induction alone. The work featured in this dissertation establishes a novel chemotherapy tolerant AML xenograft model, demonstrates its translational utility, and supports the continued investigation of targeted inhibition of Met/SAM