Mycophenolic acid (MPA), the active compound of mycophenolate mofetil (MMF), is used to prevent graft rejection in renal transplant recipients. MPA is glucuronidated to the metabolite MPAG, which exhibits enterohepatic recirculation (EHC). MPA binds for 97% and MPAG binds for 82% to plasma proteins. Low plasma albumin concentrations, impaired renal function and coadministration of cyclosporine have been reported to be associated with increased clearance of MPA. The aim of the study was to develop a population pharmacokinetic model describing the relationship between MMF dose and total MPA (tMPA), unbound MPA (fMPA), total MPAG (tMPAG) and unbound MPAG (fMPAG). In this model the correlation between pharmacokinetic parameters and renal function, plasma albumin concentrations and cotreatment with cyclosporine was quantified. tMPA, fMPA, tMPAG and fMPAG concentration-time profiles of renal transplant recipients cotreated with cyclosporine (n = 48) and tacrolimus (n = 45) were analyzed using NONMEM. A 2- and 1-compartment model were used to describe the pharmacokinetics of fMPA and fMPAG. The central compartments of fMPA and fMPAG were connected with an albumin compartment allowing competitive binding (bMPA and bMPAG). tMPA and tMPAG were modeled as the sum of the bound and unbound concentrations. EHC was modeled by transport of fMPAG to a separate gallbladder compartment. This transport was decreased in case of cyclosporine cotreatment (P < 0.001). In the model, clearance of fMPAG decreased when creatinine clearance (CrCL) was reduced (P < 0.001), and albumin concentration was correlated with the maximum number of binding sites available for MPA and MPAG (P < 0.001). In patients with impaired renal function cotreated with cyclosporine the model adequately described that increasing fMPAG concentrations decreased tMPA AUC due to displacement of MPA from its binding sites. The accumulated MPAG could also be reconverted to MPA by the EHC, which caused increased tMPA AUC in patients cotreated with tacrolimus. Changes in CrCL had hardly any effect on fMPA exposure. A decrease in plasma albumin concentration from 0.6 to 0.4 mmol/l resulted in ca. 38% reduction of tMPA AUC, whereas no reduction in fMPA AUC was seen. In conclusion, a pharmacokinetic model has been developed which describes the relationship between dose and both total and free MPA exposure. The model adequately describes the influence of renal function, plasma albumin and cyclosporine co-medication on MPA exposure. Changes in protein binding due to altered renal function or plasma albumin concentrations influence tMPA exposure, whereas fMPA exposure is hardly affected.