The US Food and Drug Administration (FDA) on Friday unveiled two draft guidances focused on adaptive clinical trial designs and master protocols for cancer treatment trials.
The 32-page adaptive design draft guidance describes the principles for designing, conducting and reporting the results from an adaptive clinical trial. The draft also advises sponsors on the types of information FDA needs to evaluate the results from trials with adaptive designs, including Bayesian adaptive and complex trials that rely on computer simulations for their design.
For the purposes of this guidance, which replaces a previous 2010 draft guidance, FDA defines an adaptive design “as a clinical trial design that allows for prospectively planned modifications to one or more aspects of the design based on accumulating data from subjects in the trial.” FDA also explains how adaptive trial designs can allow a trial to adjust to information that was not available when the trial began.
Examples of adaptive designs featured in the guidance focus on an HPV vaccine, an Ebola treatment and several other treatments and trials that proved successful in their use of an adaptive design.
The draft also features discussions on limitations of adaptive trials, how to control the chance for erroneous conclusions, estimating treatment effects, adaptive designs based on comparative data vs. non-comparative data, and adaptations to the patient population, among other sections.
Under the special considerations section, the draft also discusses simulations in adaptive design planning, Bayesian adaptive designs, adaptations in time-to-event settings, adaptations based on a potential surrogate or intermediate endpoint, secondary endpoints, adaptive design in early-phase exploratory trials, unplanned design changes based on comparative interim results and design changes based on information from a source external to the trial.
Maintaining trial integrity and regulatory considerations are also included to help sponsors navigate the process of using an adaptive trial design…