5,6-Dihydro-5-azacytidine (DHAC; NSC 264880) is an analogue of 5-azacytidine that does not possess the hydrolytically unstable 5,6-imino bond of the parent compound. Thus, unlike 5-azacytidine, DHAC is stable in aqueous solution and may be administered by prolonged i.v. infusion, potentially avoiding acute toxicities associated with bolus administration of 5-azacytidine. In this study, patients with advanced cancer were treated with DHAC administered as a 24-h constant i.v. infusion every 28 days. Treatment began at a dose of 1 g/sq m and was escalated to the maximum-tolerated dose of 7 g/sq m, where the limiting toxicity was pleuritic chest pain. Other toxicities included nausea and vomiting, which were not limiting. There was no evidence for myelosuppression, nephrotoxicity, or hepatotoxicity. DHAC was measured in plasma, urine, and ascites by a sensitive and specific reverse-phase high-performance liquid chromatography assay capable of detecting 50 ng of drug per ml. Steady-state plasma levels were achieved with 8 h and ranged from 10.0 to 20.5 micrograms of DHAC per ml at the maximum-tolerated dose. Total-body clearance of 311 +/- 76 ml/min/sq m and postinfusion half-lives between 1 and 2 h were observed. Between 8 and 20% of the administered dose was excreted unchanged in urine. While ascites DHAC levels in a patient with ovarian cancer were comparable to plasma levels, postinfusion elimination was slower from this compartment than from plasma. No correlation was observed between DHAC plasma levels and duration or intensity of dose-limiting pleuritic chest pain. One patient with progressive Hodgkin's lymphoma demonstrated stabilization of disease for seven treatment cycles, and two patients with aggressive lymphoma demonstrated dramatic, although transient, disease responses. A dose of 7 g/sq m is recommended for Phase II trials of DHAC using this schedule.