Using Bayesian Dynamic Borrowing to Maximize the Use of Existing Data: A Case-Study

Edwards, Dawn; Best, N.; Crawford, J.; Zi, L.; Shelton, C.; Fowler, A.

doi:10.1007/s43441-023-00585-3

Cited by 3 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is crucial to establish an agreement on analysis and design priors, with the latter being utilized to assess the operating characteristics of the trial design under all conceivable scenarios. In this regard, a graphical approach can be used to help select design parameters, including the degree of discounting for the pilot study data [ 170 ].…”

Section: External Data Borrowingmentioning

confidence: 99%

Using Bayesian statistics in confirmatory clinical trials in the regulatory setting: a tutorial review

Lee

2024

BMC Med Res Methodol

View full text Add to dashboard Cite

Bayesian statistics plays a pivotal role in advancing medical science by enabling healthcare companies, regulators, and stakeholders to assess the safety and efficacy of new treatments, interventions, and medical procedures. The Bayesian framework offers a unique advantage over the classical framework, especially when incorporating prior information into a new trial with quality external data, such as historical data or another source of co-data. In recent years, there has been a significant increase in regulatory submissions using Bayesian statistics due to its flexibility and ability to provide valuable insights for decision-making, addressing the modern complexity of clinical trials where frequentist trials are inadequate. For regulatory submissions, companies often need to consider the frequentist operating characteristics of the Bayesian analysis strategy, regardless of the design complexity. In particular, the focus is on the frequentist type I error rate and power for all realistic alternatives. This tutorial review aims to provide a comprehensive overview of the use of Bayesian statistics in sample size determination, control of type I error rate, multiplicity adjustments, external data borrowing, etc., in the regulatory environment of clinical trials. Fundamental concepts of Bayesian sample size determination and illustrative examples are provided to serve as a valuable resource for researchers, clinicians, and statisticians seeking to develop more complex and innovative designs.

show abstract

Section: External Data Borrowingmentioning

confidence: 99%

Using Bayesian statistics in confirmatory clinical trials in the regulatory setting: a tutorial review

Lee

2024

BMC Med Res Methodol

View full text Add to dashboard Cite

show abstract

Pharmacometrics-Enhanced Bayesian Borrowing for Pediatric Extrapolation – A Case Study of the DINAMO Trial

Sailer,

Neubacher,

Johnston

et al. 2024

Ther Innov Regul Sci

View full text Add to dashboard Cite

Bayesian borrowing analyses have an important role in the design and analysis of pediatric trials. This paper describes use of a prespecified Pharmacometrics Enhanced Bayesian Borrowing (PEBB) analysis that was conducted to overcome an expectation for reduced statistical power in the pediatric DINAMO trial due to a greater than expected variability in the primary endpoint. The DINAMO trial assessed the efficacy and safety of an empagliflozin dosing regimen versus placebo and linagliptin versus placebo on glycemic control (change in HbA1c over 26 weeks) in young people with type 2 diabetes (T2D). Previously fitted pharmacokinetic and exposure-response models for empagliflozin and linagliptin based on available historical data in adult and pediatric patients with T2D were used to simulate participant data and derive the informative component of a Bayesian robust mixture prior distribution. External experts and representatives from the U.S. Food and Drug Administration provided recommendations to determine the effective sample size of the prior and the weight of the informative prior component. Separate exposure response-based Bayesian borrowing analyses for empagliflozin and linagliptin showed posterior mean and 95% credible intervals that were consistent with the trial results. Sensitivity analyses with a full range of alternative weights were also performed. The use of PEBB in this analysis combined advantages of mechanistic modeling of pharmacometric differences between adults and young people with T2D, with advantages of partial extrapolation through Bayesian dynamic borrowing. Our findings suggest that the described PEBB approach is a promising option to optimize the power for future pediatric trials.

show abstract

Bayesian Dynamic Borrowing in Group-Sequential Design for Medical Device Studies

Chiaruttini,

Lorenzoni,

Gregori

2024

Preprint

View full text Add to dashboard Cite

Background Medical device technology evolves rapidly, with shorter lifecycles compared to pharmaceuticals. This acceleration necessitates swift safety and efficacy assessments to keep pace with technological advancements. In this context, leveraging clinical data from previous device versions is crucial to reducing enrollment periods and accelerating development, particularly because medical devices often share similar mechanisms. Bayesian Dynamic Borrowing has emerged as an approach that adjusts the weight of historical information based on the congruence between past and new data, enabling unbiased data augmentation. Methods This study explores the efficiency of a new study design algorithm that combines Bayesian Dynamic Borrowing with Group-Sequential Design theory. A phase 4 clinical trial on a new medical device for the patent foramen ovale closure has been used as motivating example, and 4 past studies on the control device have been used for the prior elicitation. Simulations were conducted under both the assumption of exchangeability (congruent scenarios) and non-exchangeability (incongruent scenarios) between historical and current control data to evaluate the design's operating characteristics. Results The proposed algorithm, when tested under the congruent scenarios, demonstrated its ability to reduce the expected new enrolled patients (2,790 vs. 4,848 under H0, and 3,846 vs. 4,848 under H1) while maintaining both Type I error and Power at their nominal values. Additionally, the asymmetric early stopping boundaries allow a high percentage of trials to be stopped under the null hypothesis (71% at the first interim). The simulations under incongruence scenarios demonstrate how the proposed algorithm discounts the prior information and reduces the expected borrowed sample size dropping from 395 to 13 and from 575 to 68 in scenarios with the higher degrees of incongruence, under Null and Alternative hypotheses, respectively. Conclusions In summary, this paper underscores the potential advantages of incorporating Bayesian Dynamic Borrowing with Group-Sequential Design within clinical trial design for medical device studies. The proposed method effectively discounts historical data, maintains control over Type I error and Power, and ensures ethical considerations through early stopping boundaries. In addition, this approach offers considerable flexibility through parameter customization, facilitating more effective collaboration between statistical and clinical specialists.

show abstract

Using Bayesian Dynamic Borrowing to Maximize the Use of Existing Data: A Case-Study

Cited by 3 publications

References 13 publications

Using Bayesian statistics in confirmatory clinical trials in the regulatory setting: a tutorial review

Using Bayesian statistics in confirmatory clinical trials in the regulatory setting: a tutorial review

Pharmacometrics-Enhanced Bayesian Borrowing for Pediatric Extrapolation – A Case Study of the DINAMO Trial

Bayesian Dynamic Borrowing in Group-Sequential Design for Medical Device Studies

Contact Info

Product

Resources

About