Scott Miller, PhD, is a biostatistician at Clinipace.
Thanks to everyone who tuned in to the webcast “Navigating Regulatory Biostatistical Requirements Throughout the Clinical Trial Lifecycle” I recently hosted with my colleague Ron Marks, PhD, Clinipace CSO, director of biostatistics and cofounder. During the presentation, we reviewed common statistical issues (during trial planning, analysis, and submission phases) and discussed potential approaches to address these issues.
As a result of a great turnout and many excellent follow-up questions, we decided to address a number of questions in a series of blog posts. Earlier this week we addressed protocol deviations. Today, we’re going to address considerations for adaptive trial designs.
If you have any additional thoughts or related questions to ask our experts, please share your comments below.
Q: Can you briefly highlight statistical considerations for adaptive trial designs?
Adaptive designs are a fairly broad category that covers a wide range of very diverse topics. Essentially, an adaptive design is a pre-specified process which allows for some aspect(s) of a clinical trial (e.g. design, conduct, analysis) to be changed in light of accumulating data.
One example is covariate adaptive design, where you have multiple clinically relevant covariates of interest you need to ensure balance on, but there are too many to account for via stratified randomization. Another example is response-adaptive designs. In a standard trial, you would randomize subjects to the treatment arms based on some fixed randomization ratio (like 1:1 or 2:1) and use that ratio all the way through the trial. In some cases, you might want to adjust that ratio over time either to be more statistically precise, therefore minimizing the sample size, or to increase the likelihood of subjects receiving the best performing treatment. Yet another adaptive approach is a sample size re-estimation. In this instance, you are recognizing that the initial sample size was based on fairly imprecise treatment effect estimates, so partway through the trial, you look at the data and re-estimate the sample size. If your updated sample size is about what you initially expected, then you continue the trial unchanged. If your updated sample size is larger, then you would need to decide if you need to allocate additional resources to increase the sample size or stop the study early for futility (futility analyses will be discussed in an upcoming blog post).
Because there are many different ways of doing adaptive trials, my advice on adaptive designs would be to proceed with caution. Adaptive designs are great in some situations, but in other cases, they don’t really help you much. And, again, it comes down to a case-by-case basis of what you’re looking to do and how you’re hoping it will help you.
For example, CDER came out with a guidance document on adaptive trial designs, and they basically categorized them as, “Adaptive trial designs that we like and adaptive trial designs that we’re not so sure on.” The implication is, “If you do one of these ones that we like, we’re okay with that and won’t really have any questions for you. If you use one of these other ones, we’re probably going to have a bunch of questions for you because we’re not sure how that’s going to behave.” Additionally, the Center for Devices and Radiological Health recently came out with a draft guidance on adaptive trials from their perspective, which is actually a little bit different than the CDER document.
Sometimes adaptive designs are very helpful and can help you with shortening trial duration or balancing multiple covariates at the same time. There are other times when adaptive design isn’t really going to help because the time you save is offset by an extra two rounds of review cycles at the FDA to get them to buy in to your particular application.
If you’re interested in learning more about Navigating Regulatory Biostatistical Requirements through the Trial Lifecycle, be sure to listen to our webcast in its entirety. Check back soon for more questions regarding trial planning, conduct, and trial analysis/submission as well as the related eBook!