Regulatory Open Forum

This message was posted by a user wishing to remain anonymous

Hi all, 
we would like to have one section of our pre-sub focused on the suitability of the clinical evidence required by FDA for the clearance of our device (yes, our 510k requires clinical data, such a bad luck...).
Which type of clinical documents should we submit to FDA? Would a synopsis of the protocol (with sample size, primary and secondary outcomes, inclusion criteria, etc.) be sufficient for their review and assessment? Is there anything else needed/useful (e.g., a complete protocol, informed consent, etc.)? Also, since we will propose only OUS data, I was planning to add a rationale on why such data could be representative of US population. 

What else would make sense to include? and what is the right/smart way to formulate such type of question (which basically is: is our data (foreign data, number of patients, outcomes, etc) sufficient for you to support our intended use?)

ps. I have heard people saying that your should try to have the same number of patients as your predicate, but this does not sound right to me knowing how the FDA thinks. My point of view is that as far as your study makes statistically sense, the FDA should not focus their assessment on the comparison of the number of patients.  

Any insight/example/template on this topic is greatly appreciated. 
Thanks

My general approach to your questions regarding what to submit is to provide enough information that FDA can fully answer your questions. My experience is that reviewers are reluctant to fully answer questions about study design issues unless the questions are focused on specific issues. If you ask, for example, whether foreign data is acceptable the package should provide enough background information on medical practices and populations for them to provide a helpful answer. Depending on the kind of question, a synopsis of the protocol may be sufficient. However, FDA can be reluctant to answer vague questions about a protocol if the synopsis or background information does not provide enough information. Also, I have found that the best approach for questions is to ask a Yes/No question as to whether FDA agrees to your plan. You may get a lot of "No" responses, but they will explain why they don't agree.

The idea that the same number of patients as the predicate being sufficient is reasonable advice for developers that don't have the experience and expertise to come up with a better number. So, you are absolutely right that an appropriate justification for the sample size is better than just referring back to what the predicate used.

------------------------------
Glen Park PharmD
Executive Director, Regulatory Affairs and Quality Assurance
Jersey City NJ
United States
------------------------------

Original Message:
Sent: 26-Nov-2019 09:41
From: Anonymous Member
Subject: pre-submission clinical protocol

This message was posted by a user wishing to remain anonymous

Hi all,
we would like to have one section of our pre-sub focused on the suitability of the clinical evidence required by FDA for the clearance of our device (yes, our 510k requires clinical data, such a bad luck...).
Which type of clinical documents should we submit to FDA? Would a synopsis of the protocol (with sample size, primary and secondary outcomes, inclusion criteria, etc.) be sufficient for their review and assessment? Is there anything else needed/useful (e.g., a complete protocol, informed consent, etc.)? Also, since we will propose only OUS data, I was planning to add a rationale on why such data could be representative of US population.

What else would make sense to include? and what is the right/smart way to formulate such type of question (which basically is: is our data (foreign data, number of patients, outcomes, etc) sufficient for you to support our intended use?)

ps. I have heard people saying that your should try to have the same number of patients as your predicate, but this does not sound right to me knowing how the FDA thinks. My point of view is that as far as your study makes statistically sense, the FDA should not focus their assessment on the comparison of the number of patients.

Any insight/example/template on this topic is greatly appreciated.
Thanks