The use of arsenic-containing compounds in cancer therapy is currently being re-considered, after the recent approval of arsenic trioxide (Trisenox) for the treatment of relapsed promyelocytic leukemia (PML). In an attempt to prepare a carrier system to minimize the toxicity of this drug, the aim of this study is to prepare and characterize liposomes encapsulating arsenic trioxide (ATO). For this, we prepared different types of liposomes entrapping ATO: large multilamellar (MLV), sonicated (SUV) and dried reconstituted vesicles (DRV). The techniques used were: thin film hydration, sonication and the DRV method, respectively. Two lipid compositions were studied for each liposome type, EggPC/Chol (1:1) and DSPC/Chol (1:1). After liposome preparation, drug encapsulation was evaluated by measuring arsenic in liposomes. For this, energy-dispersive X-ray fluorescence spectroscopy or atomic absorption was used. In addition, the retention of the drug in the liposomes was evaluated after incubating the liposomes in buffer at 37 degrees C. The experimental results reveal that encapsulation of ATO in liposomes ranges between 0.003 and 0.506 mol/ mol of lipid, and is highest in the DRV vesicles and lowest in the small unilamellar vesicles, as anticipated. Considering the in vitro stability of ATO-encapsulating liposomes: 1) For the PC/Chol liposomes (DRV and MLV), after 24 hours of incubation, more than 70% (or 90% in some cases) of the initially encapsulated amount of ATO was released. 2) The liposomes composed of DSPC/Chol could retain substantially higher amounts of ATO, especially the DRV liposomes (54% retained after 24 h). 3) In the case of PC/Chol, temperature of incubation has no effect on the ATO release after 24 hours, but affects the rate of ATO release in the MLV liposomes, while for the DSPC/Chol liposomes there is a slight increase (statistically insignificant) of ATO release at higher temperature.