Should 510(k) Predicates be Actively Listed Devices?

[Mark Meer]

Question for discussion: Should a chosen predicate for a 510(k) be a device that has an active listing?

I don't see anything in any FDA guidances that requires this, but it seems like it might be prudent.

Consider a hypothetical device that has multiple 510(k)s over the years:
1. 2001: Original 510(k)
2. 2007: Update to include new indications
3. 2016: Update to incorporate new materials and technologies

Only the most recent iteration of the device (following the 2016 510(k)) is listed, and currently on the market.
However, what if my device is closer to the 2001 version? From my perspective it makes sense to select it as a predicate....but then I'm not really comparing it to what is "state of the art" for that device...

Discussion welcome.

 

[Ed Panek]

Are you trying to claim equivalence to a device that has been superseded in market technology? Does this proposed device solve some unique problem?

 

[monoj mon]

Interesting question. I have never encountered such situation so far, it would be interesting to see if anyone has done it successfully before. However, if I would be at your situation then I would still go for the device which is currently listed as the "predicate device" and its previous version (which is equivalent to the subject device) as the "reference device" and make a justification on how the technological differences with the predicate device wouldn't affect the safety and effectiveness of the subject device by referring to the "reference device". I hope it makes any sense

 

[Watchcat]

This question comes around now and then. Wish that meant I could snap my fingers and give you a quick, solid answer, but hey...this is the FDA we're talking about here.

Did you follow the whole 510(k) modernization saga last year? As part of it, FDA floated the idea of publishing a list older devices in an effort to encourage companies to claim more "modern" devices as predicates instead of the older devices on its list? The proposal was not well received, to put it mildly. Never been implemented as far as I know, and that was before Gottlieb moved on, so quite possibly never will be.

One issue raised was the obvious question of why, if the older devices were not safe and effective enough to serve as predicates, had FDA not pulled them from the market. Heh. So, yes, as far as I know, FDA is still good with you claiming SE to any "legally marketed device." emphasis mine:

A legally marketed device is a device that was legally marketed prior to May 28, 1976 (preamendments device), or a device which has been reclassified from Class III to Class II or I, a device which has been found SE through the 510(k) process, or a device that was granted marketing authorization via the De Novo classification process under section 513(f)(2) of the FD&C Act that is not exempt from premarket notification requirements. The legally marketed device(s) to which equivalence is drawn is commonly known as the "predicate." Although devices recently cleared under 510(k) are often selected as the predicate to which equivalence is claimed, any legally marketed device may be used as a predicate.. Legally marketed also means that the predicate cannot be one that is in violation of the FD&C Act.

To me this language suggests FDA means a device that has, at some point in time, met its criteria to be legally marketed, not that it is necessarily being legally marketed at the time of the claim of SE.

 

[Mark Meer]

...To me this language suggests FDA means a device that has, at some point in time, met its criteria to be legally marketed, not that it is necessarily being legally marketed at the time of the claim of SE.

I think I agree, however the concern is regarding "some point in time". Guidances change, efficacy standard change, technology changes... I could develop a device SE equivalent to a device from 1989, but would it be considered safe/effective by today's standards?

Presumably over-and-above strictly comparing specifications, indications, intended use, etc., there is an expectation to comply with the latest guidances and standards. With respect to technologies, however (i.e. if new technologies have been demonstrated to be far superior to older ones), I'm not so sure if there is anything dictating that newer ones are preferred over old.

Also, I just want to double check: If a device is pulled from the market, does the 510(k) record also get pulled? In other words, if I'm using the FDA's Online 510(k) Search, can I be confident that any results are viable predicates? Or is it possible that some entries represent devices that were pulled from the market or otherwise later found in violation of the FD&C Act somehow?

 

[Watchcat]

I could develop a device SE equivalent to a device from 1989, but would it be considered safe/effective by today's standards?

Per the "saga," if it isn't, then it shouldn't be available for use as a predicate. I think maybe both "it" and "available" are in question here. Does ANY manufacturer have a device listed under the same regulation and product code? If so, then I think you should be able to use any device with the same regulation and product as a predicate, even if the device is no longer sold.

I'm not so sure if there is anything dictating that newer ones are preferred over old.

While I have absolutely no idea. But I know I would decide which standards to use myself, and justify it based on risk analysis, rather than trying to figure out what FDA "thinks." (Starting to see that as an oxymoron lately.) On the EU side, I know the common wisdom is to ignore "recognized," and use the current version.

If a device is pulled from the market, does the 510(k) record also get pulled? In other words, if I'm using the FDA's Online 510(k) Search, can I be confident that any results are viable predicates? Or is it possible that some entries represent devices that were pulled from the market or otherwise later found in violation of the FD&C Act somehow?

Can't even begin to guess. Seems like it should have been, but...it's a bureaucracy. You aren't a RAPS member? If not, I'll be happy to post this question to its forum for you. Should at least get some guesses, maybe some pretty good ones. Maybe even an Answer.

 

[Mark Meer]

... You aren't a RAPS member? If not, I'll be happy to post this question to its forum for you. Should at least get some guesses, maybe some pretty good ones. Maybe even an Answer.

Thanks @Watchcat ! No, not a RAPS member....so much appreciated if you could!

In fact, never heard of RAPS before. Is it worth the membership price?

 

[Watchcat]

Is it worth the membership price?

Regulatory Affairs Professional Society (RAPS). Depends on what you are looking for. It has a forum and there are some knowledgeable people there. Has a strong tilt toward medical devices.

Anyway, will post your question and let you know...

 

[Watchcat]

I got three responses. Two people said that, if it is in the 510(k) database, it can serve as a predicate. One said it needs to be a currently marketed device.

No feedback specifically on recalled devices, but the reality is that FDA has no authority to withdraw the 510(k) altogether, only recall the device as manufactured by the manufacturer, and that's theoretically always temporary, as the manufacturer can always address whatever issues were the basis for the recall and get the recall lifted. Assuming your device doesn't have those issues (most commonly manufacturing, not design), no problem. If it is a design problem then you can fall back on "at least as safe and effective" and convince FDA your design is better (but still SE, ha!).

Also, I would be more confident it hadn't been recalled, withdrawn, whatever from a google search than from FDA's website. Even if they were "supposed to" pull the 510(k), not sure I'd bet the farm on that. Also, some listings for some kinds of devices actually show recalls. 510(k)s, I'm not sure, because they aren't consistent in the data they list for the different types of submissions, sigh.

Bottom line, I think you might get some push back due to the combination of the predicate being both older and not on the market It's a business decision on your part. I would probably go for it, unless the stakes were unusually high. And you can always do a pre-sub...

 

[Ed Panek]

How does that work? Suppose you have 10000 active FDA 510(k) approved devices operating in the USA. If the 510(k) is pulled how are AE reported?