Aouam Karim, Chadli Zohra, Hammouda Mouna, Fredj Nadia Ben, Aloui Sabra, May Mezri El, Boughattas Naceur, Skhiri Habib, Chaabane Amel
*Department of Pharmacology, Faculty of Medicine of Monastir, University of Monastir; and †Department of Nephrology, University Hospital Fattouma Bourguiba of Monastir, Tunisia.
Ther Drug Monit. 2015 Aug;37(4):524-30. doi: 10.1097/FTD.0000000000000173.
Limited sampling strategies (LSS), using few sampling times after dosing, have been used to reliably predict tacrolimus area under the 12-hour concentration-time curve (AUC). Because the pharmacokinetics of tacrolimus is subject to significant changes over the exposure time to this drug, it can be hypothesized that the reliability of the LSS would also change. This study aimed to develop a reliable and practical LSS allowing the estimation of tacrolimus AUC in Tunisian kidney transplant recipients taking into account the posttransplantation time.
Thirty Tunisian patients were enrolled into 3 groups (10 in each group) according to the posttransplantation period: period 1: between 1 day and 3 months, period 2: between 3 and 12 months and period 3 over 12 months, as defined by the European consensus conference on the therapeutic drug monitoring of tacrolimus. Samples were collected just before and 0.5, 1, 2, 4, 6, 8, and 12 hours after tacrolimus administration. The full pharmacokinetic profiles obtained from these timed concentration data were used to choose the best sampling times. Error indices (mean absolute prediction error and the root mean squared prediction error) were used to evaluate the predictive performance.
Among the 1-point estimations, the C4-predicted AUC showed the highest correlation with the measured one during period 1 and period 2 (r = 0.94 and 0.91, respectively) but not period 3 (r = 0.76). The C0-predicted and the measured AUC become less and less correlated from period 1 to period 3 (r = 0.81, 0.75, and 0.66), respectively. Only the model including the C0/C2 provided a high correlation between predicted and measured tacrolimus AUC regardless of the posttransplant period (r = 0.95, 0.96, 0.98 and root mean squared prediction error = 4.1, 5.8, 4.2 during periods 1, 2, and 3, respectively).
Our data clearly indicate that the predictive performance of LSS is prone to change according to the posttransplantation time. A 2-time point LSS was found to be sufficient to predict tacrolimus AUC. The LSS using C0 and C2 is reliable, accurate, and practical to estimate the AUC of tacrolimus regardless of the posttransplantation time.
有限采样策略（LSS），即在给药后采用较少的采样时间点，已被用于可靠地预测他克莫司12小时浓度 - 时间曲线下面积（AUC）。由于他克莫司的药代动力学在该药物的暴露时间内会发生显著变化，因此可以推测LSS的可靠性也会改变。本研究旨在开发一种可靠且实用的LSS，以便在考虑移植后时间的情况下，估计突尼斯肾移植受者的他克莫司AUC。
根据移植后时期，将30名突尼斯患者分为3组（每组10名）：时期1为1天至3个月，时期2为3至12个月，时期3为超过12个月，这是由欧洲他克莫司治疗药物监测共识会议定义的。在他克莫司给药前以及给药后0.5、1、2、4、6、8和12小时采集样本。从这些定时浓度数据获得的完整药代动力学曲线用于选择最佳采样时间点。使用误差指数（平均绝对预测误差和均方根预测误差）来评估预测性能。
在单点估计中，C4预测的AUC在时期1和时期2与测量的AUC相关性最高（分别为r = 0.94和0.91），但在时期3相关性不高（r = 0.76）。从时期1到时期3，C0预测的和测量的AUC之间的相关性越来越低（分别为r = 0.81、0.75和0.66）。只有包含C0/C2的模型在无论移植后时期如何的情况下，预测的和测量的他克莫司AUC之间都具有高度相关性（在时期1、2和3中，r分别为0.95、0.96、0.98，均方根预测误差分别为4.1、5.8、4.2）。
我们的数据清楚地表明，LSS的预测性能容易根据移植后时间而变化。发现两点LSS足以预测他克莫司AUC。无论移植后时间如何，使用C0和C2的LSS在估计他克莫司AUC方面都是可靠、准确且实用的。
