Use of clinical data in a 510(k) submission

[DCONN1]

Is anyone familiar with or have experience in submitting clinical data as part of a 510(k) submission which was collected outside of a traditional clinical trial? Was the FDA receptive to accepting the data as evidence for substantial equivalence?

I am aware that the FDA recently issued a draft guidance document addressing the use of such data titled "Use of Real-World Evidence to Support Regulatory Decision-Making for Medical Devices". However, I am a bit confused as to the extent to which such data would be considered "clinical data" in terms of a 510(k) submission, as opposed to a sort of literature review. Take, for example, a device which has been previously cleared by a 510(k) submission and which is then used outside of its indications for use in several independent studies. According to the draft guidance referenced above, such studies could be appropriate evidence for a 510(k) submission seeking to expand the indications for use for this device. However, in terms of the requirements for a 510(k), does this mean that Financial Disclosure/Certification forms would need to be submitted for each study?

 

[Ronen E]

Have you identified a clear requirement, applicable to your device, to submit clinical data in your 510(k)?

Substantial equivalence is usually determined based on comparing the intended use and the technological characteristics of the subject device and the predicate. I'm not aware of an overarching expectation to demonstrate equivalence through clinical performance.

 

[DCONN1]

Thanks for the response, Ronen.

Yes, I would agree that there is no expectation/requirement to present clinical data to demonstrate substantial equivalence for such a submission. However, I believe it could be beneficial to include these clinical studies as a part of the evidence demonstrating substantial equivalence to a predicate already cleared with expanded indications, especially if the results of these studies can be compared to similar studies involving a predicate device. Obviously the clinical data would not be the sole evidence presented for substantial equivalence, but it could certainly be helpful to include it.

I could be wrong; perhaps including such data just raises more questions in the minds of the reviewers and they begin focusing in on studies and faulting them for not being traditional clinical trials, and then find reasons to kick back submissions or ask for more information based on this. For this reason, I'd be interested in hearing about any experiences with 510(k)s which included non-traditional clinical data.

 

[Ronen E]

demonstrating substantial equivalence to a predicate already cleared with expanded indications

Not sure I understand what you're trying to do. Are you submitting the 510(k) to expand your device's intended use, and using a predicate with the expanded intended use? If so, I think you should focus on the similarities in the claims and in the technological characteristics. If your device is technologically different, showing that clinically it works the same won't do you much good - that's my gut feeling.

I could be wrong; perhaps including such data just raises more questions in the minds of the reviewers and they begin focusing in on studies and faulting them for not being traditional clinical trials, and then find reasons to kick back submissions or ask for more information based on this. For this reason, I'd be interested in hearing about any experiences with 510(k)s which included non-traditional clinical data.

That's exactly my concern. I don't have the specific experience you're after, but my general experience shows that it's not a good idea to provide "extras" that aren't required - it usually triggers clarification questions and requests for additional information. If you're lucky, they might just tell you that it's not required and ask that you avoid it in the future (after you've already wasted some times in corresponding).

I would try that path only if the submission is marginal and you need whatever additional support you can scrape.

Cheers,
Ronen.

 

[mihzago]

Actually, clinical data may be required for some devices as part of special controls. This is typically established through special guidance document or as part of the De Novo classification.
For example, take a look at DEN140040, one of the special controls listed on page 3 is clinical performance data. Similar requirement in DEN130051, or DEN140025.


You could potentially used data from published articles of studies that your company did not conduct or sponsor, but this will work only if the study was designed and conducted in accordance with GCP and contains all required elements, e.g. inclusion/exclusion criteria, justification for sample size, data analysis methods, description of controls, etc. FDA may also ask for raw data.
For my de novo I have used a combination of data from clinical studies sponsored by my company, and clinical data from published articles, and both were accepted. I did have to provide a lot of detailed information about how the studies from articles were conducted.
I would think that what worked for De Novo, should also work for a 510(k).


Include clinical data only if you believe clinical data is required. Otherwise it may raise questions or simply unnecessarily extend the review time.

 

[Ronen E]

Actually, clinical data may be required for some devices as part of special controls. This is typically established through special guidance document or as part of the De Novo classification.

This is why I initially asked

Have you identified a clear requirement, applicable to your device, to submit clinical data in your 510(k)?

I would think that what worked for De Novo, should also work for a 510(k).

Unless it's an overkill.

Include clinical data only if you believe clinical data is required. Otherwise it may raise questions or simply unnecessarily extend the review time.

That's exactly what I meant.

 

[Julie O]

About 10% of all 510(k)s are supported by clinical trials. I'm not a 510(k) expert, so I don't know if they were included because the applicant felt they were needed to demonstrate SE for their device, or because FDA required clinical data as a special control for this type of device. I would also not be surprised if other 510(k)s have supported by clinical data that are not from traditional clinical trials.

I think you have to decide whether the data are important in making your case for substantial equivalence. If so, I would include it. If not, don't. You are just giving reviewers more to review, and, even if that doesn't lead to more deficiencies, it can't help speed your 510(k) along.

I can't answer your question about financial disclosure without more information on the source of the data.

I would be cautious about relying on De novos as a guide for 510(k) submissions. Any clinical data included in a De novo petition would be included to support classification based primarily on risk, not to support substantial equivalence to another device.

You can search the 510(k) database for 510(k)s that include clinical trials. Given that your data are not from a clinical trial, I don't know if that would be helpful to you or not. Depending on how much information is in the 510(k) summary, it might give you a better idea of how clinical data generally are used to support a claim of SE.