Limited Sampling Model for Advanced Mycophenolic Acid Therapeutic Drug Monitoring After Liver Transplantation

BACKGROUND: The immunosuppressive drug mycophenolate mofetil (MMF), with mycophenolic acid (MPA) as active metabolite, is a nonnephrotoxic alternative to calcineurin inhibitors. Therapeutic drug monitoring (TDM) of MPA may improve clinical benefit from MMF therapy, especially in MMF monotherapy or with reduced dose of a calcineurin inhibitor. Limited data are available on TDM strategies for MPA in orthotopic liver transplantation (OLT). The authors here describe the pharmacokinetic (PK) behavior of MPA after OLT and developed a Bayesian limited sampling model for monitoring MMF after OLT. METHODS: PK data were obtained from 57 stable patients, and trapezoidal area under the curve (AUC(0-12h)) was calculated. The effect of the covariates kidney function and serum albumin concentration was studied. A TDM strategy was developed based on individualized population PKs using Bayesian estimations and limited sampling models to predict the MPA AUC. RESULTS: A relationship between MMF dose and MPA AUC was found and a 8-fold apparent clearance range of MPA was observed at the same dose level. Significant relationships of albumin concentration and creatinine clearance with MPA plasma clearance were identified (respectively, r² = 0.12 and 0.24; P < 0.05). A model with limited sampling at 0, 0.5, 1, 2, and 3 hours after drug administration showed very good correlation with trapezoidal AUC(0-12h) with acceptable bias and precision (r² = 0.92, mean prediction error = 1, mean absolute prediction error = 13; P < 0.05). CONCLUSIONS: Remarkable variability of MPA clearance in stable OLT patients exists, which can be partially explained by the patients' albumin serum levels and creatinine clearance. Systemic exposure in these patients can be accurately assessed by the Bayesian limited sampling TDM strategy.