5 Mistakes To Avoid When Filing Your FDA Presubmission
By Sharon Kvistad and Allison Scott, Navigant Consulting
The U.S. Food and Drug Administration (FDA) presubmission program is an incredibly useful tool, but like most “incredibly useful tools,” the key to its usefulness lies in knowing how to use it. Hammers and screwdrivers are great tools, but they are worthless and frustrating for someone who doesn’t understand how to use them. So it goes with the FDA’s presubmission program.
The FDA has expanded its pre-IDE program since 1995 by allowing, and sometimes requesting, sponsors to submit certain information for FDA to review and comment upon for applications other than IDEs. This was moderately confusing, since everything was referred to as a pre-IDE document, even when there was no IDE involved. Fast forward to 2012, when the FDA issued draft guidance for a presubmission program for applications reviewed by either the Center for Devices and Radiological Health (CDRH) or the Center for Biologics Evaluation and Research (CBER), and officially changed the designation of submissions to Q-submissions (or Q-subs, if you prefer the trade jargon).
This is not a re-hash of the presubmission guidance, but rather a retrospective of presubmission documents we have reviewed in the course of consulting engagements, and some of the most common errors we have seen during those engagements. Our intent is to help you avoid some of these errors before filing your presubmission. And remember — you are allowed to submit more than one presubmission for a specific device.
Error #1 – Request For Classification
The FDA has made it pretty clear that it will not provide a determination of classification via a presubmission. There is another vehicle, called a 513(g), that is used for determination of classification (and requires a fee).
However, if you want the FDA’s free input on your device classification, provide your rationale or justification as to why the device is a specific classification, such as Class II. Then, consolidate that information into an “agreement” question around your determination of device classification — something like this: “Does the FDA agree with our determination of Class II based on this, this, and this?”
Error #2 – Expectation Of Data Review
Again, the presub guidance is pretty clear that the FDA will not review data. Note that we didn’t say FDA won’t read it; they just will not pass judgement on your data. Data review is reserved for the formal submission process, whether it is a clinical or a commercial application. So, when a sponsor provides us with a presubmission that is loaded with data, either from bench testing or clinical study, our advice is to put the data in an appendix. Or, better yet, eliminate the data altogether, unless seeing the data is paramount to understanding the discussion.
If you desperately need feedback on “data,” lay out exactly how you intend to conduct (or have conducted) testing, and the type of result that the testing will yield, and then formulate your question around that information. Ask about endpoints, statistical analysis, test method, etc., but do not ask “Is this data sufficient?” because you will get a party-line response about presubmissions not being the vehicle for data review.
Error #3 – Lack Of Focus
This is probably the most common error we see in presubmission construction. We can’t quantify the number of presubs we have reviewed that left us feeling dizzy and with absolutely no clue as to what the sponsor was trying to achieve. If there is one thing we have learned in working with the FDA, it is that confusion on the part of the FDA results in feedback that is a little like opening the Ark of the Covenant in Raiders of the Lost Ark — the sponsor is hit with all kinds of stuff that they never intended or envisioned and, unfortunately, you can’t undo what has been done.
In this instance, your best option is to send in a second presubmission that attempts to clarify what you messed up the first time by focusing on a specific aspect of feedback that you need/want. Again, you can submit more than one presub. So, if you want to get clarification on the classification, indications for use, appropriate predicate device, testing requirements, clinical data structure, and FDA’s opinion on whether the sky is blue, chances are your submission isn’t focused or will be too much for FDA to digest. Pick one or two topics per presub. Baby bites.
Error #4 – Too Much Detail
We’ve read many presubs that practically included instructions on how to build the device, yet at the end, we still didn’t know how the thing worked. Sponsors are excited about their technology — we get it. But, you may not be addressing an audience that has the same depth of knowledge in your specific area of expertise, in which case you will lose the reader. Or, the FDA could call in someone who thoroughly understands your technology, resulting in an academic or technical discussion that, while interesting and thought-provoking, may not support the ultimate objective of the presub. Provide sufficient detail in a language and format that can be understood by a lay-person, as well as an expert. If the FDA wants or needs more specifics, it will ask for them, allowing you to tailor the detail provided.
Error #5 – Pointless Or Confusing Questions
When constructing a presub, the sponsor is asked to put forth questions that they want answered by FDA. We think a sponsor should create these questions first, before they ever start drafting the main document. Too often, the questions come after the document is drafted (or even finalized), and no one has spent any time making sure that 1) the questions make sense, and 2) the information needed to address the questions is in the presub.
Some of you may be incredulous and think “Really? Who does that?” The sad answer is, “lots of people.” We’re not suggesting that questions can’t change once you have drafted them — some may need to be adjusted. They may not be specific enough, or they are too specific, so that you may not get the breadth of feedback you seek.
Maybe what you thought was a good question doesn’t matter anymore, for a variety of reasons — we have seen that scenario a lot: The sponsor asks a question about a specific element of device operation, the FDA responds, then the sponsor says “Oh yeah, we changed the device. It doesn’t work that way anymore.” HUH? There are many examples we could share, but you get the point — check and cross-check your questions against content, and vice-versa.
We firmly believe that the presubmission process is one of the best tools that the FDA has provided to sponsors in recent years — and there is no user fee! Many sponsors assume that they will obtain the same type of feedback in the formal submission review process, but that is flawed logic. FDA has a clock ticking during formal review and, more than once, a sponsor has sought detailed guidance only to have the FDA say “Oh, so sorry — we’re out of time.” Don’t be that sponsor. Use the process, take advantage of the free advice, and don’t ignore what you are being told. Presubs can relieve the uncertainty of the formal submission process immeasurably.
About The Authors
Allison Scott is a managing consultant at Navigant Consulting, Inc. with over a decade of experience providing regulatory support to U.S and global health IT, medical device, diagnostics, and combination product companies. In addition to her regulatory experience, Allison has assisted various pharmaceutical and medical device companies in reviewing of their compliance programs. She can be reached at 317-228-8719 or at allison.scott@navigant.com.
Sharon Kvistad is an associate director in Navigant's Healthcare and Life Sciences Disputes, Regulatory, Compliance, and Investigations practice, focusing on FDA regulatory matters. Kvistad has over 30 years of experience in U.S. and global regulatory affairs as they relate to medical devices, with domestic submission preparation experience including IDE, PMA, PMA/S, HUD/HDE, and 510(k). She can be reached at 317-228-8715 or at sharon.kvistad@navigant.com.