Current conventional treatment for patients with acute myelogenous leukemia results in a high percentage of clinical responses in most patients. However, a high percentage of patients still remain refractory to primary therapy or relapse later. This review examines the search for new agents and new modes of therapy. In Section I, Dr. Estey discusses new agents directed at various targets, such as CD33, angiogenesis, inappropriately methylated (suppressor) genes, cell cycle checkpoints, proteosomes, multidrug resistance (MDR) gene, mitochondrial apoptotic pathway. He also reviews preliminary results of phase I trials with the nucleoside analog troxacitabine and liposomal anthracyclin and suggests new strategies for trials of new agents. In Section II, Dr. Jones revisits differentiation therapy and presents results of preclinical and clinical studies that demonstrate that a variety of clinically applicable cell cycle inhibitors (interferon, phenylbutyrate, vitamin D, retinoids, bryostatin-1) preferentially augments growth factor-mediated induction of myeloid leukemia terminal differentiation, as well as blocks growth factors' effects on leukemia proliferation. The combination of cell cycle inhibition plus myeloid growth factors may offer a potential treatment for resistant myeloid leukemias. In Section III, Drs. Levitsky and Borrello address the question of tumor vaccination in AML and shows that, although tumor rejection antigens in AML have not been formally identified to date, a growing number of attractive candidates are ripe for testing with defined antigen-specific vaccine strategies. Interestingly, the ability to drive leukemic blasts to differentiate into competent antigen presenting cells such as dendritic cells may be exploited in the creation of cellular vaccines. Ultimately, the successful development of active immunotherapy for AML will require integration with dose-intensive chemotherapy, necessitating a more complete understanding of host immune reconstitution. In Section IV, Dr. Slavin reviews the concept of delivering non-myeloablative stem cell transplantation (NST) and delayed lymphocyte infusion (DLI) to increase tolerance in particular in high risk and older patients, and take advantage of the graft-versus-leukemia (GVL) effect. All these approaches hold promise in reducing morbidity and mortality and differ from the older concepts aiming at delivering the highest possible doses of chemotherapy and/or total body irradiation to reach maximum leukemia cell kill, whatever the toxicity to the patient.