Scott Miller, PhD, is a biostatistician at Clinipace.
Thanks again to everyone who attended the webcast, Navigating Regulatory Biostatistical Requirements through the Trial Lifecycle, I recently hosted with my colleague, Ron Marks, PhD, Clinipace CSO, director of biostatistics, and cofounder. We shared information on common statistical issues and potential solutions for the planning, conduct, analysis, and submission phases of a clinical trial.
Following the webinar, we’ve addressed many of the excellent questions from attendees in a series of blog posts. In the first group of Q&A posts, we answered questions regarding the trial planning phase, including protocol deviations, adaptive designs, and alpha spending functions in interim analyses. In addition, we responded to questions about trial conduct.
In our final two posts, we’re addressing questions related to trial analysis and submission. If you have any additional thoughts or related questions to ask our experts, please share your comments below.
Q: When planning for the end of Phase 2 analyses, is there a way to get questions about data analyses answered by someone at the FDA?
A: If you have been in contact with the FDA Review Division, you can email them with a request to have them weigh in on a particular question. Clearly state what type(s) of reviewers you would like to weigh in on the given set of questions to avoid confusion. For example, if you do not specifically request input/review by a statistical reviewer, it may not be reviewed by one.
For device trials, it’s possible the FDA would have to review too much supplemental information to weigh in via an email. In such cases, it is easier to submit the information and questions as a Pre-Submission.
In general, it is almost always preferable to provide specifics about the rationale or question rather than a vague “Can we do this?” The answer to a general question like that is usually, “Yes, you can do whatever you want.” When you provide details, the response will oftentimes be more specific and more helpful in understanding any potential concerns the review division may have with your given approach. This can help you frame a submission that specifically addresses those points.
Q: A blinded study is running. Based on monitoring of blinded data, additional new analysis not in SAP will be interesting. Please comment if one should revise the SAP and provide rationale to the FDA for change prior to study end or do as post hoc analysis.
A: The answer to this question depends on two important factors. One factor is the stage of study; updating the SAP in this situation would be more acceptable in a Phase II study rather than Phase III study. In a Phase II study, it is understandable you might learn something new about your product or study requirements during study conduct that could lead to important additional information for future studies. In this situation, adding a new endpoint or changing an existing secondary endpoint could be justified to the FDA through the SAP and CSR. You certainly could justify adding new exploratory (i.e. observational) endpoints and analyses. You would just need to limit the new information to future study planning, and not to drawing new definitive conclusions used for registration purposes.
For a Phase III, the situation is different. This is your registration study, so it would be less likely the FDA would approve changes to the SAP and analysis. If there is any area where you feel you do not have complete information to finalize a critical endpoint or analysis, you might propose an interim analysis (or adaptive design) to learn the needed information early in your study, and then finalize the protocol when you have recruited enough subjects to obtain the needed info. But changing your SAP and protocol for the primary or critical secondary endpoints (those geared toward FDA approval or for labeling) would likely be frowned upon by the regulatory group.
The second critical factor is planned use of the results of the post hoc findings. The FDA or any regulatory authority is likely going to limit any post hoc findings to planning for new studies and will almost surely not accept any statistically significant findings as definitive for registration or labeling purposes. The regulators would likely insist on you running a new study with the a priori objective of now proving what you observed in this study. Only then would the findings be accepted as true.
I believe studies often times can yield new and potentially interesting information in addition to what was identified during protocol development. This new information should be assessed in an ongoing study, if possible, as long as the limitations of such findings are understood.
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 our final post regarding trial analysis and submission as well as the related eBook!