As at the previous PEAC meeting, the public attendees were seated at roundtables and one part of the meeting was dedicated to roundtable discussions based on a scenario provided by CDRH:
You are employed by a medical device company that manufactures Device C. Device C, designed to cure lung cancer, has completely changed how physicians treat this disease and has exponentially increased survival rates for lung cancer patients.
In your position, you are responsible for conducting postmarket surveillance on Device C (that is, evaluating potential safety concerns for the marketed device). You received an email from one of your field technicians that Device C may be causing strokes in patients by causing an irregular heart rhythm like atrial fibrillation.
Your boss does not want to pull Device C from the market (for example, issue a voluntary recall) unless there is evidence of a true safety concern; however, your boss is concerned about possible lawsuits and patient deaths if Device C is associated with strokes. You would like to explore the issue a bit more before making your recommendation.
At this point, I am nearly lost to the rest of the conversation. I realize it's only a discussion scenario, but seriously? This is a device that has exponentially increased lung cancer survival rates. How could a stroke possibly be a true enough safety concern to warrant a recall?
To make matters worse, several people who are clearly asking themselves the same question immediately raise the question of how soon after treatment these strokes occur. Off the cuff, the FDA discussion leader says "well into the future," probably without consideration of the clinical implications of this statement, but because a lot of the questions are about use of RWE as an alternative to a long-term postmarket study.
So now I am completely gone. (You all know how I can get.) I express my astonishment that anyone would even consider a recall of this device, regardless of when it might occur, but if it is going to occur at some time well in the future, the idea of a recall is just, just…mind boggling.
I ask, what patient would say, "Oh no, I don't want the treatment that exponentially increases my chances of surviving this cancer, because if I do survive, then someday I might get a stroke? Noooo, I want the treatment that exponentially reduces my chance of survival now, so I won't risk getting a stroke in the future." What doctor would ever recommend this?
Across the table, a bright-eyed millennial RA professional from one of the large device companies is bobbing her head in enthusiastic agreement with my every word. I take a breath and she jumps in and starts talking about reviewing manufacturing records. I have no idea why, but since she was so enthusiastic about my comments, I bob my head reciprocally at hers.
Later I realize that she and I were probably making different assumptions about the potential source of the problem. She was thinking more along the lines of a manufacturing problem leading to the recall of one or more lots. That actually wouldn't have been so bad, since patients would still have had access to the treatment. I, on the other hand, was thinking of a design problem, so I talked about looking at principles of operation and mechanisms of action.
I finally came in for a landing, and we proceeded to discuss the scenario and the specific questions that FDA wanted us to address, which regrettably did not include "Would you be out of your mind to recall this device," lol. One was the potential use of registries in this scenario. I'm skeptical you need a registry to find lung cancer patients, because most of them are followed closely for years. I would just contact oncologists. But I decided it was high time I got with the program, so I let this go.
After the allotted time for discussion, the FDA discussion leaders summarized the comment at their tables. I was reassured to learn that I was not alone in my reaction to the idea of a recall. No one addressed this directly, since it wasn't one of the questions, but the general theme found its way into the summaries, nonetheless:
There was less discussion about doing a randomized clinical trial than the ethics of denying treatment to a group that could potentially benefit from the therapy…
We actually took a little bit of a different route…looking at the benefit-risk assessment overall, given that this was a breakthrough treatment for lung cancer, versus the risk of stroke
…may not even be that important to patients with lung cancer if they've had such an exponential improvement…
My favorite summary comment was on the issue I had let go
:
"How could you use the internet to find patients with the condition to participate in your clinical trial?"
Why use the internet, right? Why not directly contact the oncologists…
I especially liked this answer because there is a get-rich-quick crowd that is drooling over the potential of patient engagement via the internet, and I want them to just go away.
------------------------------
Julie Omohundro, ex-RAC (US, GS), still an MBA
Principal Consultant
Class Three, LLC
Mebane, North Carolina, USA
919-544-3366 (T)
434-964-1614 (C)
julie@class3devices.com------------------------------