Centre for Applied Pharmacokinetic Research, School of Health Sciences, University of Manchester, Manchester, UK.
Certara, Princeton, New Jersey, USA.
Clin Pharmacol Ther. 2024 Sep;116(3):814-823. doi: 10.1002/cpt.3291. Epub 2024 May 13.
Precision dosing strategies require accounting for between-patient variability in pharmacokinetics (PK), affecting drug exposure, and in pharmacodynamics (PD), affecting response achieved at the same drug concentration at the site of action. Although liquid biopsy for assessing different levels of molecular drug targets has yet to be established, individual characterization of drug elimination pathways using liquid biopsy has recently been demonstrated. The feasibility of applying this approach in conjunction with modeling tools to guide individual dosing remains unexplored. In this study, we aimed to individualize physiologically-based pharmacokinetic (PBPK) models based on liquid biopsy measurements in plasma from 25 donors with different grades of renal function who were previously administered oral midazolam as part of a microdose cocktail. Virtual twin models were constructed based on demographics, renal function, and hepatic expression of relevant pharmacokinetic pathways projected from liquid biopsy output. Simulated exposure (AUC) to midazolam was in agreement with observed data (AFE = 1.38, AAFE = 1.78). Simulated AUC variability with three dosing approaches indicated higher variability with uniform dosing (14-fold) and stratified dosing (13-fold) compared with individualized dosing informed by liquid biopsy (fivefold). Further, exosome screening revealed mRNA expression of 532 targets relevant to drug metabolism and disposition (169 enzymes and 361 transporters). Data related to these targets can be used to further individualize PBPK models for pathways relevant to PK of other drugs. This study provides additional verification of liquid biopsy-informed PBPK modeling approaches, necessary to advance strategies that seek to achieve precise dosing from the start of treatment.
精准给药策略需要考虑药代动力学（PK）方面的个体间变异性，这会影响药物暴露，还需要考虑药效学（PD）方面的个体间变异性，这会影响在作用部位达到相同药物浓度时的反应。虽然用于评估不同水平的分子药物靶点的液体活检尚未建立，但最近已经证明可以使用液体活检来对药物消除途径进行个体特征分析。应用这种方法结合建模工具来指导个体化给药的可行性仍有待探索。在这项研究中，我们旨在根据 25 名肾功能不同的接受者的血浆液体活检测量值来个体化基于生理学的药代动力学（PBPK）模型，这些接受者之前曾作为微剂量鸡尾酒的一部分口服给予咪达唑仑。基于液体活检输出，从人口统计学、肾功能和相关药代动力学途径的肝表达构建虚拟双胞胎模型。模拟的咪达唑仑暴露（AUC）与观察到的数据一致（AFE=1.38，AAFE=1.78）。三种给药方法的模拟 AUC 变异性表明，与液体活检指导的个体化给药（五倍）相比，均匀给药（14 倍）和分层给药（13 倍）的变异性更高。此外，外泌体筛选显示出与药物代谢和处置相关的 532 个靶标（169 种酶和 361 种转运体）的 mRNA 表达。与这些靶标相关的数据可用于进一步个体化与其他药物 PK 相关的途径的 PBPK 模型。这项研究为液体活检指导的 PBPK 建模方法提供了额外的验证，这对于从治疗开始就寻求实现精准给药的策略是必要的。
