Search

We propose an adaptive hybrid control causal (AHCC) design to leverage historical control data to reduce the sample size demanded by standard randomised controlled trials (RCT). Under the causal inference framework, we define the causal estimand of the average treatment effect and derive the corresponding estimator based on the trial data and historical control data. The AHCC design takes a multistage or group sequential approach. The number of patients randomised to the concurrent control is adaptively adjusted based on the amount of information borrowed from the historical control data. At each stage, based on the interim data, the contribution of the historical control data, quantified by the effective sample size, is updated and used to determine the randomisation ratio between the treatment and control arms for the next stage, with the goal to resemble a standard RCT upon the completion of the trial. Simulation studies show that the AHCC design has desirable operating characteristics. For example, it saves on sample size when substantial information can be borrowed from the historical control, and it maintains power when little information can be borrowed from the historical control. [ABSTRACT FROM AUTHOR]

For multiple rare diseases as defined by a common biomarker signature, or a disease with multiple disease subtypes of low frequency, it is often possible to provide confirmatory evidence for these disease or subtypes (baskets) as a combined group. A novel drug, as a second generation, may have marginal improvement in efficacy overall but superior efficacy in some baskets. In this situation, it is appealing to test hypotheses of both non-inferiority overall and superiority on certain baskets. The challenge is designing a confirmatory study efficient to address multiple questions in one trial. A two-stage adaptive design is proposed to test the non-inferiority hypothesis at the interim stage, followed by pruning and pooling before testing a superiority hypothesis at the final stage. Such a design enables an efficient and novel registration pathway, including an early claim of non-inferiority followed by a potential label extension with superiority on certain baskets and an improved benefit-risk profile demonstrated by longer term efficacy and safety data. Operating characteristics of this design are examined by simulation studies, and its appealing features make it ready for use in a confirmatory setting, especially in emerging markets, where both the need and the possibility for efficient use of resources may be the greatest. [ABSTRACT FROM AUTHOR]

Existing phase II clinical trial designs focus on a single scalar endpoint, such as a binary, continuous, or survival endpoint. In some clinical trials, such as pain management studies, the efficacy endpoint of interest is measured longitudinally. We propose a Bayesian phase II design for such clinical trials. We model the longitudinal measurement process using Bayesian hierarchical model, where subject‐specific trajectory shrinks toward the population trajectory to borrow information across subjects. The Bayesian penalized spline is used to model subject‐specific and population trajectories without making strong parametric assumption on their shapes. We use the area under the curve of the trajectory as the summary of the treatment effect over time. The design takes a group sequential approach and takes into account both statistical significance and clinical relevance. Bayesian criteria is proposed to make interim and final decisions based on the evidence of statistical significance and clinical relevance. The proposed design is highly flexible and can accommodate trials with one or multiple longitudinal endpoints, as well as a longitudinal primary endpoint with a secondary endpoint. Simulation study shows that the proposed design is robust with desirable operating characteristics. [ABSTRACT FROM AUTHOR]

Personalized medicine, or tailored therapy, has been an active and important topic in recent medical research. Many methods have been proposed in the literature for predictive biomarker detection and subgroup identification. In this article, we propose a novel decision tree-based approach applicable in randomized clinical trials. We model the prognostic effects of the biomarkers using additive regression trees and the biomarker-by-treatment effect using a single regression tree. Bayesian approach is utilized to periodically revise the split variables and the split rules of the decision trees, which provides a better overall fitting. Gibbs sampler is implemented in the MCMC procedure, which updates the prognostic trees and the interaction tree separately. We use the posterior distribution of the interaction tree to construct the predictive scores of the biomarkers and to identify the subgroup where the treatment is superior to the control. Numerical simulations show that our proposed method performs well under various settings comparing to existing methods. We also demonstrate an application of our method in a real clinical trial. [ABSTRACT FROM AUTHOR]

By generalizing the proportional hazards model, we introduce a new function β( t ), which we call the proportionality function, and which we show plays a role in studying aspects of the randomly censored model. We develop an asymptotically efficient nonparametric estimator of β( t ), establish its uniform consistency, and obtain a weak convergence result. Furthermore, a confidence band for β( t ), based on the bootstrap, is developed. The results are applied to an actual dataset. [ABSTRACT FROM AUTHOR]