Pharmacokinetic study of mycophenolate mofetil in patients with systemic lupus erythematosus and design of Bayesian estimator using limited sampling strategies

Background: Monitoring of the area under the plasma concentration-time curve (AUC) of mycophenolic acid (MPA) has been developed for individual dose adjustment of mycophenolate mofetil (MMF) in renal allograft recipients. MMF is currently used as an off-label drug in the treatment of systemic lupus erythematosus (SLE), but factors of its exposition may be different in these patients and need to be determined for therapeutic drug monitoring (TDM) purposes. Objective: The aim of the study was to develop a maximum a posteriori probability (MAP) Bayesian estimator of MPA exposition in patients with SLE, with the objective of TDM based on a limited sample strategy. Methods: Twenty adult patients with SLE given a stable 1 g/day, 2 g/day or 3 g/day dose of MMF orally for at least 10 weeks were included in the study. MPA was measured by high-performance liquid chromatography (HPLC) coupled to a photodiode array detector (11 plasma measurements over 12 hours post-dose per patient). Free MPA concentrations were measured by HPLC with fluorescence detection. Two different one-compartment models with first-order elimination were tested to fit the data: one convoluted with a double 7 distribution to describe secondary concentrations peaks, and one convoluted with a triple 7 distribution to model a third, later peak. Results: A large interindividual variability in MPA concentration-time profiles was observed. The mean maximum plasma concentration, trough plasma concentration, time to reach the maximum plasma concentration and AUC from 0 to 12 hours (AUC(12)) were 13.6 +/- 8.4 mu g/mL, 1.4 +/- 1.2 mu g/mL, 1.1 +/- 1.2 hours and 32.2 +/- 17.1 mu g . h/mL, respectively. The mean free fraction of MPA was 1.7%. The one-compartment model with first-order elimination convoluted with a triple 7 distribution best fitted the data. Accurate Bayesian estimates of the AUC(12) were obtained using three blood samples collected at 40 minutes, 2 hours and 3 hours, with a coefficient of correlation (R) = 0.95 between the observed and predicted AUC 12 and with a difference of <20% in 16 of the 20 patients. Conclusion: A specific pharmacokinetic model was built to accurately fit MPA blood concentration-time profiles after MMF oral dosing in SLE patients, which allowed development of an accurate Bayesian estimator of MPA exposure that should allow MMF monitoring based on the AUC(12) in these patients. The predictive value of targeting one specific or different AUC values on patients' outcome using this estimator in SLE will need to be evaluated.