Dickinson Laura, Back David, Pozniak Anton, Khoo Saye, Boffito Marta
Department of Pharmacology, University of Liverpool, Liverpool, UK.
Ther Drug Monit. 2007 Jun;29(3):361-7. doi: 10.1097/FTD.0b013e3180683b25.
Area under the concentration time curve (AUC) over a dosing interval is considered to be the best estimate of drug exposure in a patient. However, determination of this parameter is costly and often impractical, requiring multiple samples and a great deal of time and resources. A limited-sampling strategy (LSS) may overcome some of these issues, making pharmacokinetic studies easier to perform, particularly in a limited-resource setting. The aim of this work was to develop and validate a pragmatic LSS for the accurate and precise prediction of boosted saquinavir AUC0-12 (AUC over the 12-hour dosing interval) at a dosage of 1000/100 mg twice daily. Pharmacokinetic data were obtained from 34 human immunodeficiency virus (HIV)-infected individuals stable on saquinavir/ritonavir-containing therapy, randomly split into two sets (n = 17 per set). One set was used to construct prediction models using univariate and multivariate analysis (development set), and the second was used to determine the predictive performance of the models (validation set). For single samples, 6- and 10-hour concentrations correlated best with saquinavir AUC0-12 (r2: 0.913 and 0.911, respectively), yet all single samples failed to produce precise and unbiased predictions. However, combinations at 2, 6; 0, 2, 6; 0, 4, 10; 0, 4, 12; and 2, 4, 6 hours achieved good predictive performances, and both precise [root mean squared relative prediction error (%RMSE): 6.4% to 11.9%] and unbiased [mean relative prediction error (%MPE), 95% CI: -2.7%, (-0.8)-2.7 to 1.6%, (-1.8)-4.7] estimations of saquinavir AUC0-12. Of these models, concentrations obtained at 0, 2, 6 and 2, 4, 6 hours are more practical in a clinical setting and are therefore the LSS with most potential. Provided that the technique is validated in specific patient populations, an LSS approach is a potentially useful tool to evaluate the AUC0-12 of saquinavir in resource-limited settings, reducing both costs and volumes of blood taken. It may also aid the choice of sampling times for population analysis.
给药间隔时间内的血药浓度-时间曲线下面积（AUC）被认为是对患者药物暴露量的最佳评估指标。然而，测定该参数成本高昂且通常不切实际，需要采集多个样本并耗费大量时间和资源。有限采样策略（LSS）或许可以克服其中一些问题，使药代动力学研究更易于开展，尤其是在资源有限的情况下。本研究的目的是开发并验证一种实用的有限采样策略，用于准确、精确地预测每日两次服用1000/100mg剂量的增效沙奎那韦的AUC0-12（12小时给药间隔内的AUC）。从34名接受含沙奎那韦/利托那韦治疗且病情稳定的人类免疫缺陷病毒（HIV）感染者中获取药代动力学数据，并随机分为两组（每组n = 17）。一组用于通过单变量和多变量分析构建预测模型（开发组），另一组用于确定模型的预测性能（验证组）。对于单样本而言，6小时和10小时的血药浓度与沙奎那韦AUC0-12的相关性最佳（r2分别为0.913和0.911），但所有单样本均未能得出精确且无偏差的预测结果。然而，在2、6小时；0、2、6小时；0、4、10小时；0、4、12小时；以及2、4、6小时采集的组合样本具有良好的预测性能，对沙奎那韦AUC0-12的估算既精确[均方根相对预测误差（%RMSE）：6.4%至11.9%]又无偏差[平均相对预测误差（%MPE），95%置信区间：-2.7%，（-0.8）-2.7至1.6%，（-1.8）-4.7]。在这些模型中，在0、2、6小时以及2、4、6小时采集的血药浓度在临床环境中更具实用性，因此是最具潜力的有限采样策略。倘若该技术在特定患者群体中得到验证，有限采样策略方法是在资源有限的情况下评估沙奎那韦AUC0-12的潜在有用工具，可降低成本并减少采血量。它还可能有助于选择用于群体分析的采样时间。
