A reliable limited sampling strategy for the estimation of mycophenolic acid area under the concentration time curve in adult renal transplant patients in the stable posttransplant period

Abstract

In renal transplant patients, there is an established relationship between mycophenolate area under the curve and clinical outcome. The authors have developed and validated a limited sampling strategy to estimate mycophenolic acid area under the curve to 12 hours (MPA AUC0-12) in a stable renal transplant Indian population prescribed a formulation of mycophenolate mofetil (Mofilet) along with prednisolone and tacrolimus. Intensive pharmacokinetic sampling was performed in 29 patients to measure mycophenolate concentration from trough to 12 hours postdose. Subsets of different timed concentrations against total measured 12-hour area under the curve were analyzed by linear regression. Three models were identified and linear regression analysis done. After all subset regression analysis, three, four, and five time point limited sampling strategies (LSS) were developed having correlation coefficients above 0.92. Validation of the models was performed using the jackknife method and their predictive performances were tested. After validation, the correlation coefficients for all three models were above 0.901. The five-point LSS had the best predictive performance with a bias (95% confidence interval) of 0.67% (-3.45 to 4.79) and mean precision 7.73%. In all patients except one, the five-point LSS estimation for total area under the curve was within +/- 20% of the total measured AUC0-12. Trough concentration had a significant correlation with AUC0-12 (r = 0.69). However, if dosing in routine clinical practice was adjusted based only on trough concentration, 41% of our patients would require a different dose compared with monitoring using AUC0-12. The five-point LSS uses half-hourly samples from trough to 1.5 hour postdose with an additional sample at 3 hours. Ninety-three percent of our patients had a Cmax within 1.5 hour and inclusion of all the time points up to1.5 hour gave a better estimate of AUC0-12. This model simplifies area under the curve measurement with high precision in stable adult renal transplant patients.