EUVIMEDEUVIMED entdecken
Crossref

Digital twins and Bayesian dynamic borrowing: Two recent approaches for incorporating historical control data

Carl-Fredrik Burman iD, Erik Hermansson, David Bock, Stefan Franzén, David Svensson

Pharmaceutical Statistics · 2024

DOI: 10.1002/pst.2376 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Pharmaceutical Statistics, 2024. Verfasst von Carl-Fredrik Burman und 4 weiteren Autor:innen. Beteiligte Einrichtung: Early Biometrics & Statistical Innovation, Data Science & Artificial Intelligence, R&D, AstraZeneca, Gothenburg, Sweden. Hauptthemen laut MEDLINE: Bayes Theorem, Randomized Controlled Trials as Topic. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 52 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent years have seen an increasing interest in incorporating external control data for designing and evaluating randomized clinical trials (RCT). This may decrease costs and shorten inclusion times by reducing sample sizes. For small populations, with limited recruitment, this can be especially important. Bayesian dynamic borrowing (BDB) has been a popular choice as it claims to protect against potential prior data conflict. Digital twins (DT) has recently been proposed as another method to utilize historical data. DT, also known as PROCOVA™, is based on constructing a prognostic score from historical control data, typically using machine learning. This score is included in a pre‐specified ANCOVA as the primary analysis of the RCT. The promise of this idea is power increase while guaranteeing strong type 1 error control. In this paper, we apply analytic derivations and simulations to analyze and discuss examples of these two approaches. We conclude that BDB and DT, although similar in scope, have fundamental differences which need be considered in the specific application. The inflation of the type 1 error is a serious issue for BDB, while more evidence is needed of a tangible value of DT for real RCTs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Bayes TheoremRandomized Controlled Trials as TopicHumansResearch DesignComputer SimulationMachine LearningSample SizeData Interpretation, StatisticalModels, Statistical

Stichwörter der Autor:innen

PROCOVA™clinical trialsmachine learningprognostic scorerobust mixture prior

Autor:innen und Institutionen

  • Early Biometrics & Statistical Innovation, Data Science & Artificial Intelligence, R&D, AstraZeneca, Gothenburg, Sweden.
  • Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.
  • BMP Evidence Statistics, BioPharmaceuticals Medical, AstraZeneca, Gothenburg, Sweden.

ORCID-Profile: Carl-Fredrik Burman

Zitieren

APA

Burman, C. F., Hermansson, E., Bock, D., Franzén, S., & Svensson, D. (2024). Digital twins and Bayesian dynamic borrowing: Two recent approaches for incorporating historical control data. Pharmaceutical Statistics. https://doi.org/10.1002/pst.2376

Vancouver

Burman CF, Hermansson E, Bock D, Franzén S, Svensson D. Digital twins and Bayesian dynamic borrowing: Two recent approaches for incorporating historical control data. Pharmaceutical Statistics. 2024. doi:10.1002/pst.2376

BibTeX
@article{Burman2024,
  author = {Burman, Carl-Fredrik and Hermansson, Erik and Bock, David and Franzén, Stefan and Svensson, David},
  title = {Digital twins and Bayesian dynamic borrowing: Two recent approaches for incorporating historical control data},
  journal = {Pharmaceutical Statistics},
  year = {2024},
  doi = {10.1002/pst.2376}
}
Kostenlos starten →