Importance to FDA

In response to the quickly changing drug development landscape, FDA is concentrating efforts to advance Complex Innovative Trial Designs (CIDs). CIDs include complex adaptive, Bayesian, and other novel clinical trial designs. CID has design elements and/or analysis approaches that generally require computer simulations to determine the statistical properties of the trial (e.g., power, Type I error). FDA administers a CID Pilot Meeting Program (CID Program) to support the goal of facilitating and advancing the use of CIDs. The CID Program offers sponsors, whose meeting requests are granted, the opportunity for increased interaction with FDA staff to discuss their proposed CID approach, and fulfills a performance goal agreed to under the Prescription Drug User Fee Act (PDUFA VI), which was enacted as part of FDA Reauthorization Act of 2017 (Public Law 115-52).

A goal of this program is to work with sponsors to maximize clinical trial efficiency while using scientifically sound methods to determine the design for the question and population of interest. FDA’s goal is to extend using CIDs, where appropriate, from exploratory studies to clinical trials intended to provide substantial evidence of effectiveness to support regulatory approval of new therapies. A common feature of many CIDs is the need for simulations to estimate a variety of trial parameters. CIDs do not change FDA’s expectations that clinical trials be sufficient to evaluate safety and effectiveness of drug use in the intended population, including pertinent subsets, such as gender, age, and racial subsets. Advancing the use of CIDs requires further research as described in the examples below.

Examples

FDA applies multiple strategies to address the regulatory science needed to facilitate implementing CIDs:

• Evaluating the use of master protocols, which may include umbrella, basket, or platform trials. These trials allow for the evaluation of multiple therapies in a single disease, a single therapy in multiple diseases, or multiple therapies in a single disease, with therapies entering or leaving the trial based on a decision algorithm, respectively.
• Using Sequential Multiple Assignment Randomized Trials to provide a statistical framework for evaluating potentially complex treatment algorithms.
• Evaluating Bayesian approaches for the potential to increase clinical trial efficiency. For example, Bayesian trial designs may incorporate data external to the trial in a formal mathematical framework to maximize the use of information sources.
• Exploring new approaches for statistical analyses of oncology trials, where the COVID-19 pandemic has affected participation/enrollment and the estimate specified in the original trial design.