The relationship between the dose of tacrolimus, trough tacrolimus blood concentration, and selected clinical end-points (acute rejection, nephrotoxicity, and other toxicities) was examined in a prospective, multicenter clinical trial to validate the use of an enzyme-linked immunosorbent assay (ELISA) for monitoring whole-blood concentrations of tacrolimus in liver transplant patients. A total of 111 subjects from six transplant centers were evaluated over 12 weeks posttransplantation. In addition to trough tacrolimus blood concentrations, hematocrit, ALT, AST, GGTP, alkaline phosphatase, total bilirubin, serum creatinine, BUN, serum potassium, serum magnesium, blood glucose, and serum albumin were also measured. The relationship between trough tacrolimus blood concentrations and clinical endpoints was analyzed using both a logistic regression model and a Cox proportional hazard model. By logistic regression analysis, a statistically significant (P = .0465) relationship between increasing trough tacrolimus blood concentrations and decreasing risk of acute rejection was demonstrated over a 7-day time window. Nephrotoxicity and other toxicities also demonstrated statistically significant relationships with trough tacrolimus blood concentrations. The results of the Cox analysis were consistent with the logistic regression analysis. Using receiver operator characteristic curves, trough tacrolimus concentrations as measured by the ELISA method were able to differentiate the occurrence of nephrotoxicity and toxicity from nonevents. To minimize nephrotoxicity of tacrolimus, it is necessary to maintain trough blood concentrations below 15 ng/mL. This study demonstrates that the ELISA method used to measure tacrolimus blood concentrations in this study provides information of predictive value for managing the risk of nephrotoxicity, other toxicity, and rejection in liver transplant patients.