Generation and Validation of a Limited Sampling Strategy to Monitor Mycophenolic Acid Exposure in Children With Nephrotic Syndrome.

Abstract

Mycophenolate mofetil (MMF) plays an increasingly important role in the treatment of children with nephrotic syndrome, especially in steroid sparing protocols. Recent publications show the relationship of exposure to its active moiety mycophenolic acid (MPA) and clinical efficacy. Performance of full-time pharmacokinetic (PK) profiles, however, is inconvenient and laborious. Established limited sampling strategies (LSS) to estimate the area under the concentration (AUC) versus time curve of MPA (MPA-AUC) in pediatric renal transplant recipients cannot be easily transferred to children suffering from nephrotic syndrome, mainly because of the lack of concomitant immunosuppressive therapy. We therefore aimed for the generation and validation of a LSS to estimate MPA exposure to facilitate therapeutic drug monitoring in children with nephrotic syndrome. We performed 27 complete PK profiles in 23 children in remission [mean age (±SD):12.3 ± 4.26 years] to generate and validate an LSS. Sampling time points were before administration (C0) and 0.5, 1, 1.5, 2, 4, 6, 8, and 12 hours after the administration of MMF. MPA was measured by enzyme multiplied immunoassay technique. There was no concomitant treatment with calcineurin inhibitors. Mean daily dose of MMF was 927 ± 209 mg/m of body surface area resulting in a mean MPA-AUC0-12 value of 59.2 ± 29.3 mg × h/L and a predose level of 3.03 ± 2.24 mg/L. Between-patient variability of dose-normalized MPA-AUC0-12 was high (coefficient of variation: 45.5%). Correlation of predose levels with the corresponding MPA-AUC0-12 was moderate (r = 0.59) in a subgroup of 18 patients (20 PK profiles, generation group). An algorithm based on 3 PK sampling time points during the first 2 hours after MMF dosing (estimated AUC0-12 = 8.7 + 4.63 × C0 + 1.90 × C1 + 1.52 × C2) was able to predict MPA-AUC with a low percentage prediction error (3.88%) and a good correlation of determination (r = 0.90). Validation of this algorithm in a randomized separate group of 6 patients (7 PK profiles, validation group) resulted in comparably good correlation (r = 0.95) and low percentage prediction error (5.57%). An abbreviated profile within the first 2 hours after MMF dosing gives a good estimate of MPA exposure in children with nephrotic syndrome and hence has the potential to optimize MMF therapy.