the time course of intravenously administered cyclosporine (1 mg kg-1) and its metabolite AM1, AM9, and AM1c were examined in the blood, liver, and spleen of naive Lewis rats. Cyclosporine concentration versus time data for all three tissues were qualitatively similar, following a biexponential model C = Ae-gamma 1t + Be-gamma 2t with maximum cyclosporine concentrations reached at 1 h. Whole-blood cyclosporine clearance, terminal half-life, mean residence time, steady state volume of distribution, and hepatic extraction ratio (calculated from blood data) were similar to previously reported results. Cyclosporine in the liver showed the largest area under the concentration-time curve, mean residence time, and disposition and terminal half-lives. Spleen cyclosporine mean residence time and and terminal half-life were not significantly different from blood parameters. Metabolites AM1, AM9, and AM1c showed almost parallel time courses in all three tissues. The hydroxylated derivative AM9 was the major metabolite found in all tissues, with twofold greater levels in the liver compared to the blood and spleen. Slightly less AM1 was found in the liver relative to blood and spleen, where it was present in equal amounts. AM1c levels in the liver were not different from those in the spleen and were greater than observed for blood. The results obtained above were reflected in preliminary studies using liver transplanted rats treated with multiple doses of cyclosporine. Both blood and liver biopsy levels of CyA, AM1, and AM9 post-transplant showed twofold to fourfold decreases from day 3 ( samples taken 4 h post-CyA-dose) and concentrations were not significantly different from similarly sampled naive controls. More importantly, the metabolite/CyA ratios did not vary significantly between liver and blood in the two groups. For naive rats, and liver transplanted animals not undergoing rejection, changes in blood cyclosporine levels seem to predict variations in tissue concentrations.