Table1_Utilization of Pharmacokinetic/Pharmacodynamic Modeling in Pharmacoepidemiological Studies: A Systematic Review on Antiarrhythmic and Glucose-Lowering Medicines.DOCX

Introduction: In human pharmacology, there are two important scientific branches: clinical pharmacology and pharmacoepidemiology. Pharmacokinetic/pharmacodynamic (PK/PD) modeling is important in preclinical studies and randomized control trials. However, it is rarely used in pharmacoepidemiological studies on the effectiveness and medication safety where the target population is heterogeneous and followed for longer periods. The objective of this literature review was to investigate how far PK/PD modeling is utilized in observational studies on glucose-lowering and antiarrhythmic drugs.
Method: A systematic literature search of MEDLINE, Embase, and Web of Science was conducted from January 2010 to 21 February 2020. To calculate the utilization of PK/PD modeling in observational studies, we followed two search strategies. In the first strategy, we screened a 1% random set from 95,672 studies on glucose-lowering and antiarrhythmic drugs on inclusion criteria. In the second strategy, we evaluated the percentage of studies in which PK/PD modeling techniques were utilized. Subsequently, we divided the total number of included studies in the second search strategy by the total number of eligible studies in the first search strategy.
Results: The comprehensive search of databases and the manual search of included references yielded a total of 29 studies included in the qualitative synthesis of our systematic review. Nearly all 29 studies had utilized a PK model, whereas only two studies developed a PD model to evaluate the effectiveness of medications. In total, 16 out of 29 studies (55.1%) used a PK/PD model in the observational setting to study effect modification. The utilization of PK/PD modeling in observational studies was calculated as 0.42%.
Conclusion: PK/PD modeling techniques were substantially underutilized in observational studies of antiarrhythmic and glucose-lowering drugs during the past decade.

引言：人类药理学包含两大重要分支学科：临床药理学与药物流行病学（Pharmacoepidemiology）。药代动力学（Pharmacokinetic）/药效动力学（Pharmacodynamic）建模（简称PK/PD建模）在临床前研究与随机对照试验（Randomized Controlled Trial）中具有关键应用价值。然而，该建模方法在针对药物有效性与用药安全性的药物流行病学研究中却鲜有应用——此类研究的目标人群异质性较强，且随访周期较长。本综述的研究目的为探究PK/PD建模在降糖药与抗心律失常药物的观察性研究（Observational Study）中的应用程度。
方法：本研究于2010年1月至2020年2月21日期间，对MEDLINE、Embase及Web of Science数据库开展系统性文献检索。为计算PK/PD建模在观察性研究中的应用率，本研究采用两套检索策略。第一套策略中，我们从95672项针对降糖药与抗心律失常药物的研究中，按纳入标准随机筛选出1%的样本进行筛查。第二套策略中，我们统计了采用PK/PD建模技术的研究占比。随后，我们将第二套检索策略纳入的研究总数除以第一套检索策略中符合条件的研究总数，以此计算应用率。
结果：通过对数据库的全面检索及纳入文献的参考文献手工检索，本系统性综述的定性综合分析最终纳入29项研究。29项研究中几乎全部采用了药代动力学模型，仅2项研究构建了药效动力学模型以评估药物有效性。总计29项研究中有16项（占比55.1%）在观察性研究场景中采用PK/PD模型开展效应修饰（Effect Modification）分析。经计算，PK/PD建模在观察性研究中的应用率为0.42%。
结论：在过去十年间，抗心律失常药物与降糖药物的观察性研究中，PK/PD建模技术的应用存在严重不足。