Acute myeloid leukemia (AML) blasts require hematopoietic growth factors for their survival. Growth factors mediate signal transduction through signal transducer and activator of transcription (STAT) proteins. We have demonstrated that STATs are constitutively activated in approximately 50% of AML cases at diagnosis. Blasts with constitutive STAT3 activity have a unique gene profile and are resistant to apoptosis. We have shown that disease-free survival is significantly shorter in patient with, compared to without, constitutive STAT3 activity. Most of the patients in this study were treated with our in-house clinical trial using high-dose cytarabine and idarubicin. Arsenic trioxide (ATO) has growth suppressing activity in acute promyelocytic leukemia. In other types of AML, ATO induces apoptosis, leading to designation of ATO as an orphan drug for AML. However, the precise mechanisms of action of ATO are unknown. We have discovered that ATO down-regulates constitutive STAT3 activity in AML cell lines. We hypothesize that ATO similarly down-regulates STAT3 in blasts from AML patients and thus enhances their sensitivity to undergo apoptosis. We propose to measure the baseline and changes in STAT3 activity in AML blasts during in vivo therapy with ATO. We will accomplish this goal by performing a phase I study of ATO administered over one hour followed by high-dose cytarabine and idarubicin in patients with newly diagnosed AML < 60 years old. We will determine the maximum tolerated dose of ATO in this study and study the effects of in vivo administration of ATO on STAT3 activity, induction of apoptosis and changes in gene expression profiles in AML cells. In addition, we will attempt to identify the mode by which ATO controls the activity of STAT3 and how this effect alters the gene profile patterns and induces apoptosis. ? ?