Pharmacokinetic modelling of the plasma protein binding of mycophenolic acid in renal transplant recipients

Renal function and the plasma albumin concentration have been shown to correlate with clearance of total mycophenolic acid (MPA). The hypothesis for the underlying mechanism is that low plasma albumin concentrations and accumulation of the glucuronide metabolite of MPA (MPAG) decrease the binding of MPA to albumin. The subsequent increase in the unbound fraction (f(u)) of MPA (MPA(u)) produces an increase in total MPA (MPA(t)) clearance. This study aimed to develop an empirical population pharmacokinetic model to describe the relationships between renal function and albumin concentration and MPAG, MPA(u) and MPA(t), in order to provide insight into the mechanism by which renal function and plasma albumin affect the disposition of MPA. 774 MPA(t), 479 MPA(u) and 772 total MPAG (MPAG(t)) plasma concentrations were available from 88 renal transplant recipients on days 11 and 140 after transplantation. Data were analysed using non-linear mixed-effects modelling. Time profiles of MPA(u) and MPAG(t) concentrations were adequately described by two 2-compartment pharmacokinetic models with a link between the central compartments, representing the glucuronidation of MPA(u) to form MPAG. MPA(t) concentrations were modelled using: [MPA(t)] = [MPA(u)] + [MPA(u)] * theta(pb), with [MPA(u)] * theta(pb) representing the bound MPA concentration, where [MPA(t)], [MPA(u)] and theta(pb) represent MPA(t) concentration, MPA(u) concentration and a factor that correlates to the total number of protein binding places, respectively. According to this equation, f(u) = [MPA(u)]/[MPA(t)] = 1/(1 + theta(pb)) * theta(pb), and therefore [MPA(t)], was significantly and independently correlated with creatinine clearance (CL(CR)), the plasma albumin concentration and the MPAG(t) concentration (all p <0.001). A reduction in CL(CR) from 60 to 25 mL/min correlated with an increase in f(u) from 2.7% to 3.5%, accumulation of MPAG(t) concentrations from 50 to 150 mg/L correlated with an increase in f(u) from 2.8% to 3.7%, and a decrease in plasma albumin concentration from 40 to 30 g/L correlated with an increase in f(u) from 2.6% to 3.5%. No significant correlations were detected between MPA(u) clearance and the plasma albumin concentration or CL(CR). The model shows that low CL(CR), low plasma albumin concentrations and high MPAG concentrations decrease MPA(t) exposure by affecting MPA binding to albumin