Wang Jian, Edginton Andrea N, Avant Debbie, Burckart Gilbert J
Pediatric Clinical Pharmacology Staff, Office of Clinical Pharmacology, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, MD, USA.
University of Waterloo, Waterloo, Ontario, Canada.
J Clin Pharmacol. 2015 Oct;55(10):1175-83. doi: 10.1002/jcph.524. Epub 2015 Jun 9.
Selection of the first dose for neonates in clinical trials is very challenging. The objective of this analysis was to assess if a population pharmacokinetic (PK) model developed with data from infants to adults is predictive of neonatal clearance and to evaluate what age range of prior PK data is needed for informative modeling to predict neonate exposure. Two sources of pharmacokinetic data from 8 drugs were used to develop population models: (1) data from all patients > 2 years of age, and (2) data from all nonneonatal patients aged > 28 days. The prediction error based on the models using data from subjects > 2 years of age showed bias toward overprediction, with median average fold error (AFE) for CL predicted/CLobserved greater than 1.5. The bias for predicting neonatal PK was improved when using all prior PK data including infants as opposed to an assessment without infant PK data, with the median AFE 0.91. As an increased number of pediatric trials are conducted in neonates under the Food and Drug Administration Safety and Innovation Act, dose selection should be based on the best estimates of neonatal pharmacokinetics and pharmacodynamics prior to conducting efficacy and safety studies in neonates.
在临床试验中为新生儿选择首剂非常具有挑战性。本分析的目的是评估利用从婴儿到成人的数据建立的群体药代动力学（PK）模型是否能预测新生儿清除率，并评估为进行有信息量的建模以预测新生儿暴露量需要何种年龄范围的既往PK数据。来自8种药物的两个药代动力学数据来源被用于建立群体模型：（1）来自所有2岁以上患者的数据，以及（2）来自所有年龄大于28天的非新生儿患者的数据。基于使用2岁以上受试者数据的模型的预测误差显示出过度预测的偏差，预测的CL/观察到的CL的中位数平均倍数误差（AFE）大于1.5。与不使用婴儿PK数据的评估相比，使用包括婴儿在内的所有既往PK数据时，预测新生儿PK的偏差有所改善，中位数AFE为0.91。随着根据《食品药品监督管理局安全与创新法案》在新生儿中开展的儿科试验数量增加，在对新生儿进行疗效和安全性研究之前，剂量选择应基于对新生儿药代动力学和药效学的最佳估计。
