{"title":"涉及多个外部控制数据源的早期随机临床试验的治疗效果评估。","authors":"Heiko Götte, Marietta Kirchner, Meinhard Kieser","doi":"10.1080/10543406.2025.2512984","DOIUrl":null,"url":null,"abstract":"<p><p>Augmented randomized clinical trials are a valuable design option for early phase clinical trials. The addition of external controls could, on the one hand, increase precision in treatment effect estimates or reduce the number of required control patients for a randomized trial but may, on the other hand, introduce bias. We build on previous work on augmented trials with one external control data source in time-to-event settings and extend it to multiple control data sources. In a comprehensive simulation study, we evaluate existing and novel analysis options mainly based on Bayesian hierarchical models as well as propensity score analysis. Different sources of bias are investigated including population (observable and unobservable confounders), data collection (assessment schedule, real-world vs. clinical trial data), and time trend as well as different types of data like individual patient data (with or without baseline covariates) or summary data. Our simulation study provides recommendations in terms of choice of estimation method as well as choice of data sources. Explicit incorporation of the above-mentioned sources of bias in a simulation study is relevant as the magnitude of deviation from the ideal setting has a significant impact on all investigated estimation methods.</p>","PeriodicalId":54870,"journal":{"name":"Journal of Biopharmaceutical Statistics","volume":" ","pages":"1-19"},"PeriodicalIF":1.2000,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Estimation of treatment effects in early phase randomized clinical trials involving multiple data sources for external control.\",\"authors\":\"Heiko Götte, Marietta Kirchner, Meinhard Kieser\",\"doi\":\"10.1080/10543406.2025.2512984\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Augmented randomized clinical trials are a valuable design option for early phase clinical trials. The addition of external controls could, on the one hand, increase precision in treatment effect estimates or reduce the number of required control patients for a randomized trial but may, on the other hand, introduce bias. We build on previous work on augmented trials with one external control data source in time-to-event settings and extend it to multiple control data sources. In a comprehensive simulation study, we evaluate existing and novel analysis options mainly based on Bayesian hierarchical models as well as propensity score analysis. Different sources of bias are investigated including population (observable and unobservable confounders), data collection (assessment schedule, real-world vs. clinical trial data), and time trend as well as different types of data like individual patient data (with or without baseline covariates) or summary data. Our simulation study provides recommendations in terms of choice of estimation method as well as choice of data sources. Explicit incorporation of the above-mentioned sources of bias in a simulation study is relevant as the magnitude of deviation from the ideal setting has a significant impact on all investigated estimation methods.</p>\",\"PeriodicalId\":54870,\"journal\":{\"name\":\"Journal of Biopharmaceutical Statistics\",\"volume\":\" \",\"pages\":\"1-19\"},\"PeriodicalIF\":1.2000,\"publicationDate\":\"2025-06-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Biopharmaceutical Statistics\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1080/10543406.2025.2512984\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"PHARMACOLOGY & PHARMACY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Biopharmaceutical Statistics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/10543406.2025.2512984","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"PHARMACOLOGY & PHARMACY","Score":null,"Total":0}
Estimation of treatment effects in early phase randomized clinical trials involving multiple data sources for external control.
Augmented randomized clinical trials are a valuable design option for early phase clinical trials. The addition of external controls could, on the one hand, increase precision in treatment effect estimates or reduce the number of required control patients for a randomized trial but may, on the other hand, introduce bias. We build on previous work on augmented trials with one external control data source in time-to-event settings and extend it to multiple control data sources. In a comprehensive simulation study, we evaluate existing and novel analysis options mainly based on Bayesian hierarchical models as well as propensity score analysis. Different sources of bias are investigated including population (observable and unobservable confounders), data collection (assessment schedule, real-world vs. clinical trial data), and time trend as well as different types of data like individual patient data (with or without baseline covariates) or summary data. Our simulation study provides recommendations in terms of choice of estimation method as well as choice of data sources. Explicit incorporation of the above-mentioned sources of bias in a simulation study is relevant as the magnitude of deviation from the ideal setting has a significant impact on all investigated estimation methods.
期刊介绍:
The Journal of Biopharmaceutical Statistics, a rapid publication journal, discusses quality applications of statistics in biopharmaceutical research and development. Now publishing six times per year, it includes expositions of statistical methodology with immediate applicability to biopharmaceutical research in the form of full-length and short manuscripts, review articles, selected/invited conference papers, short articles, and letters to the editor. Addressing timely and provocative topics important to the biostatistical profession, the journal covers:
Drug, device, and biological research and development;
Drug screening and drug design;
Assessment of pharmacological activity;
Pharmaceutical formulation and scale-up;
Preclinical safety assessment;
Bioavailability, bioequivalence, and pharmacokinetics;
Phase, I, II, and III clinical development including complex innovative designs;
Premarket approval assessment of clinical safety;
Postmarketing surveillance;
Big data and artificial intelligence and applications.