By Michael Drues, Ph.D., President, Vascular Sciences
On Oct. 1, 2015, the U.S. Food and Drug Administration’s (FDA) revised version of the guidance document Refuse to Accept Policy for 510(k)s (RTA) goes into effect, replacing the prior version issued Dec. 31, 2012.
The timing of the revision seemed odd, given that, according to an FDA report released this summer, the RTA rate declined from 58 percent during the second quarter of 2013 to 37 percent during the second quarter of 2015. Overall acceptance rate, when RTA first and second cycles are combined, was 84 percent in 2013 and 90 percent in 2014.
However, between January and June of this year, 69 percent of 510(k) applications were rejected the first time out of the box. Many of those were rejected based on shortcomings in the RTA checklist. That is something that should never happen.
While the recent RTA checklist guidance is applicable specifically to 510(k) submissions, the general principles are applicable to all medical device regulatory submissions.
FDA’s Latest Guidance
Training and increased FDA and industry experience regarding the RTA process have contributed to the decreased RTA rate while improving the quality of 510(k) submissions. This latest guidance is applicable to all three types of 510ks —traditional, special, and abbreviated.
Each form of 510(k) submission has its own checklist within the new RTA guidance. For a traditional 510(k) submission — a category which includes the majority of 510(k) submissions, about 70 percent — the checklist is 24 pages long.
For a special 510(k), most commonly used when a company is making a modification to an existing device design or a manufacturing method used to make a device, the checklist is 10 pages long. Finally, the abbreviated 510(k) checklist, used in only about 4 or 5 percent of 510(k)s, is 26 pages long.
The most important takeaway from this RTA guidance is simple: Follow it. Use the checklist.
To understand the RTA, we need to first understand the overall review process and how the RTA fits into that. When a medical device manufacturer makes a submission to FDA, essentially, it is a 2-step process. The first step is the administrative review, and the second — and much more important — step is the scientific review.
The administrative review is where FDA makes sure that all sections of the submissions are present, all of the signatures are present, the pages are numbered, and so on. The scientific review is where the testing methods and statistical analysis are reviewed, etc.
The RTA focuses almost exclusively on the administrative review. Although FDA’s most recent RTA is specific to 510(k) submissions, the general process is applicable to all types of submissions —510(k), PMA, de novo, the humanitarian device exemption (HDE), the custom device exemption (CDE), and all seven of the major pathways to market that we have to choose from in the medical device industry in the U.S.
Returning to the recent guidance specific to 510(k) applications, it’s not enough simply to understand what the regulations says: We also have to try to understand what FDA is trying to accomplish. Stated bluntly, the vast majority of companies’ submissions to FDA are very poorly written.
I don’t just mean from a scientific, engineering, or a medical perspective, but much more fundamental than that: Sections are missing, tables of contents are missing, page numbers are missing. In some cases, peoples’ names and addresses are missing. Which brings us to my first piece of advice…
Don’t leave out any section of the submission, even if that section does not apply to your product.
Put yourself in the shoes of the reviewer. If that reviewer is going through the RTA checklist and he or she sees that you have a particular section missing, they don’t know whether it is missing because it is not applicable or because you simply forgot about it.
Case in point: A few months ago, I had a three-page submission kicked back on administrative review, under the RTA, because it did not have a table of contents — a three-page submission! In the end, I told the company to add a one-line table of contents that just read, “submission: pages 1-3,” and we met that regulatory requirement.
If you have a section that’s not applicable, don’t just leave it blank: Write “not applicable.” Make it obvious to anybody reading the submission that you didn’t leave it out because you forgot about it. In some cases, I advocate adding a sentence or two about why that section is not applicable (e.g., “This section is not applicable because our latex gloves do not contain software.”).
Again, this extra information is not a regulatory requirement, but it is something that I often do because I want to minimize the amount of questions that I get back after the review.
Use Current Forms
One of the companies with whom I'm working recently gave me a submission to review prior to sending it to FDA. I noticed that the cover sheet template they used was actually about five years old and, after doing some homework, I discovered a new version of the cover sheet. Interestingly, when we compared the current version and the previous version of the cover sheet, the only difference between the two was the date of the version at the bottom of the page.
Some people might say, “Why would it be important in such a case to make sure we are using the current cover sheet?” Because, on administrative review using the RTA, one of the questions could be, “Is this sponsor using the current form?” If the answer is “no,” that submission is going to get kicked back.
By the time that submission is returned to the manufacturer, updated by the manufacturer, and resubmitted to FDA, you may be looking at a three- or four-week delay, maybe longer, for a very trivial thing. The vast majority of problems that companies run into, especially under administrative review with the RTA, are not only preventable — they are predictable.
Practice Makes Perfect
I find it fascinating how many companies treat the FDA as their beta test. In other words, the first people to see that company’s submission, outside of their own organization, are at the FDA.
Although this is certainly the industry’s standard, it’s totally wrong. Something I do with some of the companies I work with — because I also consult for the FDA — is to read through their submission, sit through their presentation and, to be blunt, bash the heck out of it.
If they're going to make a mistake, it’s better for them to make it in front of me, rather than when it really matters. This simple practice, using the RTA checklist as a sanity check, allows companies to greatly mitigate the probability of getting their submissions kicked back by FDA.
What To Do If Your Submission Is Rejected
First, ask yourself why the submission was rejected. Was it because of the RTA? If so, that’s an easy fix and, usually, although you lose a few weeks or even a month or two, it’s not the end of the world.
It becomes much more interesting when you get a submission kicked back or rejected because of scientific review. First and foremost, you want to work directly with your reviewers to address their concerns. Maybe you are not sufficiently explaining what it is that you're doing and why.
For example, you can often bolster your case by including letters from subject matter experts (SMEs), perhaps to better explain why the statistical methodology you used makes sense. I use SME letters frequently because it doesn’t cost much time or money. I make it very clear to the experts that I do not want them to laud my device; I simply want them to verify that my method is generally accepted in the statistical community for this particular application.
If, despite all reasonable efforts to address the issue, you remain at an impasse with the reviewers, you have the option to take your case to the Ombudsman’s office. The Ombudsman is not part of CDRH; they report directly to the office of the commissioner, acting as an independent arbitrator.
You (the manufacturer) present your case to the Ombudsman, the reviewers present their case, and the Ombudsman will make a determination as to who is right. The reason I only advise this course of action as a last resort is because it can lead to you winning the battle (i.e., the Ombudsman’s office makes a determination in your favor), but losing the war — you may encounter those reviewers again in the future, and they may be none too thrilled to see you again.
I would like to say that doesn’t happen in the real world, but I also didn’t fall off the turnip truck yesterday; people are only human. Bottom line, the Ombudsman pathway is a legitimate pathway. Manufacturers have the right to use it. Just use it as a last resort.
If a submission is kicked back because of scientific review, then that is the FDA is doing its job exactly as it’s supposed to. Perhaps there was an honest disagreement in terms of how you analyzed your data, statistical methods you used, or the design of your test methodology, and that’s fair game.
But, if a submission is kicked back on administrative review — in other words, because of the RTA guidance — then, quite frankly, that’s the company’s fault. Be thorough in completing your RTA checklist for any regulatory submission, use the proper forms, and conduct a “beta test” before sending everything off to FDA.