Investigation of innovative clinical trial designs through plasmode simulation – a research proposal
-
Year 2024
- Principal investigator Jeff Zhan
- Host institution University of British Columbia
- Research location University of British Columbia
-
Partner(s)
Canadian Consortium of Clinical Trial Training Platform (CANTRAIN)
This project, supported by the 2024 CANTRAIN competition, focuses on developing and evaluating innovative clinical trial designs to enhance efficiency and reliability. Led by BC Principal Investigator Denghuang (Jeff) Zhan, a third-year doctoral student at UBC, this BC-based research integrates modern statistical approaches to address challenges in clinical trials.
The study builds on insights from the THREE-D trial, a groundbreaking study that evaluated treatments for patients with treatment-resistant depression (TRD). Traditional clinical trials, like the THREE-D trial, often require large sample sizes and fixed designs, which can be inefficient and inflexible. This research introduces a Bayesian adaptive design framework, which allows trials to adjust in real time based on accumulating data, improving efficiency and reducing costs.
By employing advanced simulation methods, including plasmode simulation (a technique using real-world data with controlled modifications), this project compares the performance of Bayesian adaptive designs with traditional fixed designs. Key benefits of Bayesian adaptive designs include the ability to stop trials early for efficacy or futility, more precise treatment effect estimation, and reduced sample size requirements without compromising statistical power.
The anticipated outcomes of this work include:
- Improved trial designs that are more responsive to real-world data.
- Practical guidelines for implementing Bayesian adaptive designs in clinical research.
- Enhanced methodologies that could accelerate treatment evaluation in areas like psychiatry and beyond.
This research will not only advance statistical methodologies but also contribute to the broader goals of Learning Health Systems (LHS), enabling healthcare systems to learn and adapt quickly to new evidence. By making trials faster, more flexible, and cost-effective, this work has the potential to improve healthcare delivery and patient outcomes.