Limited sampling strategy to predict free mycophenolic acid area under the concentration–time curve in paediatric patients with nephrotic syndrome

Abstract

AbstractIn paediatric patients, there is no data on the recommended area under the concentration–time curve from 0 to 12 h (AUC0−12) for free mycophenolic acid (fMPA), which is the active form of the drug, responsible for the pharmacological effect. We decided to establish the limited sampling strategy (LSS) for fMPA for its use in MPA therapeutic monitoring in children with nephrotic syndrome treated with mycophenolate mofetil (MMF). This study included 23 children (aged 11 ± 4 years) from whom eight blood samples were collected within 12 h after MMF administration. The fMPA was determined using the high‐performance liquid chromatography with fluorescence detection method. LSSs were estimated with the use of R software and bootstrap procedure. The best model was chosen based on a number of profiles with AUC predicted within ± 20% of AUC0−12 (good guess), r2, mean prediction error (%MPE) of ±10% and mean absolute error (%MAE) of less than 25%. The fMPA AUC0−12 was 0.1669 ± 0.0697 μg h/mL and the free fraction was within 0.16%–0.81%. In total, there were 92 equations developed of which five fulfilled the acceptance criteria for %MPE, %MAE, good guess >80% and r2 > 0.900. These equations consisted of three time points: model 1 (C1, C2, C6), model 2 (C1, C3, C6), model 3 (C1, C4, C6), model 5 (C0, C1, C2), and model 6 (C1, C2, C9). Although blood sampling up to 9 h after MMF dosing is impractical, it is crucial to include C6 or C9 in LSS to assess fMPA AUCpred correctly. The most practical fMPA LSS, which fulfilled the acceptance criteria in the estimation group, was fMPA AUCpred = 0.040 + 2.220 × C0 + 1.130 × C1 + 1.742 × C2. Further studies should define the recommended fMPA AUC0−12 value in children with nephrotic syndrome.