Department of Pediatric Pharmacology and Pharmacogenetics, Hôpital Robert Debré, Assistance Publique-Hôpitaux de Paris, Université Paris VII, Paris, France.
Ther Drug Monit. 2011 Dec;33(6):681-7. doi: 10.1097/FTD.0b013e318235d067.
Limited sampling strategies (LSS) for estimating the area under the curve (AUC(0-12h)) of tacrolimus and optimizing dosage adjustment are not currently used or fully validated in pediatric patients, although the method is of real benefit to children. The objective of the present study was to develop and validate reliable and clinically applicable LSS using Bayesian estimation and the multiple regression analysis for estimating tacrolimus AUC in pediatric kidney transplant patients.
The original tacrolimus pharmacokinetic dataset consists of 50 full profiles from 50 pediatric kidney transplant patients. Two LSS based on Bayesian estimator or multiple regression analysis to calculate tacrolimus AUC were developed and then compared. External validation was prospectively performed in an independent validation group, which consisted of 42 full pharmacokinetic profiles from 20 pediatric kidney transplant patients.
Bayesian estimators using C(0h), C(1h) or C(2h), and C(3h) gave the best predictive performance, the external validation having a mean prediction bias of 1% and mean imprecision of 5.5%. The multiple regression analysis using C(0h), C(1h), and C(3h) gave the best correlation (r² = 0.953) between estimated and referenced AUCs with a mean prediction bias of 4.2% and mean precision of 8.3% in external validation dataset.
The prediction of AUC using developed LSS was unbiased and precise. The age and time after transplantation did not influence the predictive performance. Such LSS approach will help guiding tacrolimus therapeutic drug monitoring based on AUC in pediatric kidney transplant patients.
目前在儿科患者中尚未使用或充分验证有限采样策略（LSS）来估计他克莫司的曲线下面积（AUC(0-12h)）并优化剂量调整，尽管这种方法对儿童确实有益。本研究的目的是开发和验证使用贝叶斯估计和多元回归分析来估算儿科肾移植患者他克莫司 AUC 的可靠且临床适用的 LSS。
原始他克莫司药代动力学数据集由 50 名儿科肾移植患者的 50 个完整图谱组成。开发了两种基于贝叶斯估计器或多元回归分析的 LSS 来估算他克莫司 AUC，然后进行比较。外部验证在一个独立的验证组中进行，该组由 20 名儿科肾移植患者的 42 个完整药代动力学图谱组成。
使用 C(0h)、C(1h)或 C(2h)和 C(3h)的贝叶斯估计器具有最佳的预测性能，外部验证的平均预测偏差为 1%，平均不精密度为 5.5%。使用 C(0h)、C(1h)和 C(3h)的多元回归分析在外部验证数据集之间具有最佳的相关性（r²=0.953），估算的 AUC 与参考 AUC 之间的平均预测偏差为 4.2%，平均精度为 8.3%。
使用开发的 LSS 预测 AUC 无偏且精确。年龄和移植后时间不会影响预测性能。这种 LSS 方法将有助于指导儿科肾移植患者基于 AUC 的他克莫司治疗药物监测。
