Department of Pharmacology and Toxicology, Radboud Institute for Health Sciences, Radboud University Medical Center, Geert Grooteplein Zuid 10, 6525 GA, Nijmegen, The Netherlands.
Royal Dutch Pharmacist Association, The Hague, The Netherlands.
Clin Pharmacokinet. 2022 Dec;61(12):1705-1717. doi: 10.1007/s40262-022-01181-8. Epub 2022 Nov 11.
More than half of all drugs are still prescribed off-label to children. Pharmacokinetic (PK) data are needed to support off-label dosing, however for many drugs such data are either sparse or not representative. Physiologically-based pharmacokinetic (PBPK) models are increasingly used to study PK and guide dosing decisions. Building compound models to study PK requires expertise and is time-consuming. Therefore, in this paper, we studied the feasibility of predicting pediatric exposure by pragmatically combining existing compound models, developed e.g. for studies in adults, with a pediatric and preterm physiology model.
Seven drugs, with various PK characteristics, were selected (meropenem, ceftazidime, azithromycin, propofol, midazolam, lorazepam, and caffeine) as a proof of concept. Simcyp v20 was used to predict exposure in adults, children, and (pre)term neonates, by combining an existing compound model with relevant virtual physiology models. Predictive performance was evaluated by calculating the ratios of predicted-to-observed PK parameter values (0.5- to 2-fold acceptance range) and by visual predictive checks with prediction error values.
Overall, model predicted PK in infants, children and adolescents capture clinical data. Confidence in PBPK model performance was therefore considered high. Predictive performance tends to decrease when predicting PK in the (pre)term neonatal population.
Pragmatic PBPK modeling in pediatrics, based on compound models verified with adult data, is feasible. A thorough understanding of the model assumptions and limitations is required, before model-informed doses can be recommended for clinical use.
超过一半的药物仍被超适应证用于儿童。药代动力学（PK）数据对于支持超适应证剂量非常重要，但对于许多药物，此类数据要么稀疏，要么不具有代表性。基于生理学的药代动力学（PBPK）模型越来越多地用于研究 PK 并指导剂量决策。构建用于研究 PK 的化合物模型需要专业知识并且耗时。因此，在本文中，我们研究了通过务实的方法将现有的化合物模型（例如，为成人研究开发的模型）与儿科和早产儿生理学模型相结合来预测儿科暴露的可行性。
选择了七种具有不同 PK 特征的药物（美罗培南、头孢他啶、阿奇霉素、丙泊酚、咪达唑仑、劳拉西泮和咖啡因）作为概念验证。使用 Simcyp v20 通过将现有的化合物模型与相关的虚拟生理学模型相结合，预测成人、儿童和（早产）新生儿的暴露情况。通过计算预测与观察 PK 参数值的比值（0.5 至 2 倍的接受范围）以及通过预测误差值的预测误差值进行视觉预测检查来评估预测性能。
总体而言，模型预测婴儿、儿童和青少年的 PK 能够捕捉临床数据。因此，认为 PBPK 模型性能具有高度可信度。当预测（早产）新生儿人群的 PK 时，预测性能趋于下降。
基于经成人数据验证的化合物模型的儿科实用型 PBPK 建模是可行的。在推荐模型指导剂量用于临床应用之前，需要对模型的假设和局限性有透彻的了解。
Zotero 插件
