EUVIMEDEUVIMED entdecken
Open Access

Sample Size Reduction by Applying ML Based Causal Inference Methods

Ahrim Youn iD, Gang Han, Emma Gerard, Neeraj Poonia, Sergio Olmos, Bingzhi Zhang, Meehyung Cho, Andres Houseman

Pharmaceutical Statistics · 2026

DOI: 10.1002/pst.70105 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Pharmaceutical Statistics, 2026. Verfasst von Ahrim Youn und 7 weiteren Autor:innen. Beteiligte Einrichtung: Evidence Generation and Decision Science, Sanofi, Morristown, New Jersey, USA. Hauptthemen laut MEDLINE: Machine Learning, Randomized Controlled Trials as Topic, Research Design, Clinical Trials, Phase III as Topic. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We conducted a comprehensive comparative analysis of causal machine learning (ML) methods to assess their utility in improving the efficiency of clinical trial designs with or without historical data. Specifically, we compared standard ANCOVA analysis used in a Randomized Controlled Trial (RCT) with several causal ML methods, including PROCOVA, TMLE, DML, and GRF. PROCOVA is gaining popularity in RCT design and requires a historical data for prognostic scores, but other methods can be applied with or without such data. Our primary focus was on strict RCT setting without borrowing historical control data, though we also explored the impact of borrowing data. The historical data used consisted of placebo data from two Phase 3 Ophthalmology studies with a continuous primary endpoint. We employed a generative AI approach, specifically Generative Adversarial Networks (GANs), to simulate RCT data from the historical data under various scenarios, varying treatment effects with and without treatment effect heterogeneity, RCT sizes, and bias. Results showed that causal ML methods can increase power even without borrowing historical data. For example, TMLE increased effective sample size by 21% in one scenario. In scenarios with borrowing of controls, PROCOVA increased power while controlling type 1 error, showing robustness to model misspecification.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningRandomized Controlled Trials as TopicResearch DesignClinical Trials, Phase III as TopicHumansSample SizeGenerative Adversarial NetworksComputer Simulation

Stichwörter der Autor:innen

causal machine learning (ML) methodsexternal data borrowinggeneralized adversarial networks (GANs)simulation

Autor:innen und Institutionen

  • Evidence Generation and Decision Science, Sanofi, Morristown, New Jersey, USA.
  • Evidence Generation and Decision Science, Sanofi, Gentilly, France.
  • Evidence Generation and Decision Science, Sanofi, Hyderabad, India.
  • Evidence Generation and Decision Science, Sanofi, Barcelona, Spain.
  • Biostatistics and Modeling, Argenx, Boston, Massachusetts, USA.

ORCID-Profile: Ahrim Youn

Zitieren

APA

Youn, A., Han, G., Gerard, E., Poonia, N., Olmos, S., Zhang, B., Cho, M., & Houseman, A. (2026). Sample Size Reduction by Applying ML Based Causal Inference Methods. Pharmaceutical Statistics. https://doi.org/10.1002/pst.70105

Vancouver

Youn A, Han G, Gerard E, Poonia N, Olmos S, Zhang B, et al. Sample Size Reduction by Applying ML Based Causal Inference Methods. Pharmaceutical Statistics. 2026. doi:10.1002/pst.70105

BibTeX
@article{Youn2026,
  author = {Youn, Ahrim and Han, Gang and Gerard, Emma and Poonia, Neeraj and Olmos, Sergio and Zhang, Bingzhi and Cho, Meehyung and Houseman, Andres},
  title = {Sample Size Reduction by Applying ML Based Causal Inference Methods},
  journal = {Pharmaceutical Statistics},
  year = {2026},
  doi = {10.1002/pst.70105}
}
Kostenlos starten →