510(k) 10993 Biocompability Testing

[planB]

Mixi,

in addition to ISO 10993-1:2018, have a look into the FDA guidance on biocompatibility. Specifically on p. 8, this guidance says - quote

In certain situations, a sponsor may propose to use a material that has known toxicities but where the material could be acceptable for the end use. In this case, the risk assessment should include consideration of the intended use population that will use (e.g., protective mask for clinician) or be treated with the device and a discussion of potential benefits of using the chosen material as well as potential mitigations that have been considered (e.g., hermetically sealing).

A chemical analysis of the materials used in a device in its final finished form can be informative. Chemical analysis can be particularly helpful to demonstrate that chemical toxicity testing from a previously cleared or approved medical device is relevant to a device under review by the Agency. For example, in some circumstances, a chemical analysis can demonstrate that the extractables and leachables in a biocompatibility extract have not changed, eliminating the need for additional biocompatibility testing using that type of solvent.

A note on "putting test results into perspective"; this is typically and evaluation activity that either you would do or a consulting division formally independent from the test lab: an accredited test lab is exceeding its scope if they interpreted the test results, so they typically don't.

HTH,

Gerhard (from Austria)

 

[Ronen E]

@Mixi_Austria - In general, ISO 10993-1 (or the FDA) doesn't require "biocompatibility testing". What they do require is a "biological evaluation" (or biocompatibility evaluation). Testing is prescribed only once all relevant available information has been analysed and a real need is substantiated. Preferably, an experienced toxicologist from an ISO 10993-1 accredited lab would do that for you. It's not a matter of "putting things in perspective", it's actually what the standard prescribes.

Since your device is a formulated one, I would begin with a chemical characterization. If you also have access to the predicate's formulation, or can characterize it through lab work, your evaluation may conclude with a written analysis showing that any detected differences are negligible (and accounting for any residual uncertainties).

I recommend consulting with your accredited test lab regarding the utility of ISO 10993-18 (chemical characterization).

 

[Mixi_Austria]

First of all, let me thank you for all the replies!

How do I know if biocompability testing is required?
Truth is, I don't. I have looked at the summaries of possible predicates and they all did biocompability testing so I figured I'd need it as well. After what I've learned so far and what some of you have suggested, the first step would be a biocompability evaluation and I'm meeting with a lab later today to get that process started.

Why don't I talk to the FDA?
I assume you refer to the Q Subs / Pre Submission options. I need to read up on this subject but at this point I don't know how open they are to rather basic questions. So I've just been hesitating and felt more comfortable doing research online.

Gerhard: Nice to meet a fellow Austrian here! You're right obviously. We're now working with OFI; I'm sure you know them. They have a consulting division that is independent from the lab.

A little background: As you might have guessed, I'm rather inexperienced in this topic. We have been in the market for vet Supplies, water disinfection, etc. for quite some time but this is the first time we're trying to register a medical device. So please excuse me if some questions sound rather trivial.

 

[Ronen E]

I have looked at the summaries of possible predicates and they all did biocompability testing so I figured I'd need it as well.

This is a legitimate approach that sometimes works. In fact, the FDA is quite likely to ask for the very same tests unless you convince them that it's unnecessary. One thing to keep in mind is that standards (ISO) and expectations (FDA) evolve over time, and the predicate might have been cleared in a different era. ISO 10093-1 was reissued in 2003, 2009 & 2018, and the FDA has also revised it's related guidance during that period (however, not after the release of 10993-1:2018). So while "same as the predicate" might be the default response, it's not necessarily the last word. The FDA promotes the "least burdensome approach", so if you can provide a similar level of assurance with less testing (or no testing), you might make it. It will, however, be your onus to show that your approach is aligned with the standard and the FDA's interpretation of it, and indeed substantiates an acceptable level of biocompatibility. This should not scare you though - if the lab you'll work with is accredited to ISO 10993 and the toxicologist conducting the evaluation has FDA experience, they should be able to build a solid case for you.

 

[Watchcat]

My own strategy for biocompatibility has always been the same, which is to secure the services of the most qualified (i.e., intimidating) biocompatibility consultant I can find, have them work through ISO 10993-1 to determine what tests are needed, put it in writing, and sign it. This on the theory that CDRH is unlikely to challenge their official opinion. It never has, but perhaps that was simply because the expert knew what he was doing, so did CDRH's biocompatibility reviewers, and the testing was solid, so nothing there to challenge.

As a general principle I don't rely on the advice of anyone who makes money by selling me something to tell me what I should buy. This includes not relying on test labs to tell me what tests I should pay them to do. In addition, test labs are experts in doing tests, not in assessing biocompatibility. Some of the test labs will tell you differently, and I know from personal experience that some of them have employees who are very knowledgeable, but no matter what you might read on LinkedIn, companies are companies, they are not their employees. Test labs, like all companies, have their own business models to follow, designed to meet their own business needs, not yours. In addition to their (we hope) expertise in conducting tests, test labs are very up-to-date on what tests their clients are asking them to do. How representative their clients are of all device companies is unknown, and I don't take it as a given that what their clients are asking for is what CDRH is accepting, but still very helpful knowledge, IMO.

As for what I've learned recently, it seems that CDRH may have gone off the rails when it comes to biocompatibility lately. Some examples below:

______________________________________________________

This is from an email exchange with a colleague who has decades of experience in device RA, and who works with one of the larger device companies (which I think get CDRH's best attention):

Colleague: The stated goal is to modernize biocompatibility within a risk management process. While that is the goal of the standard, the FDA has taken a giant leap BACKWARDS on this topic….

Me: Ummm… Correct me if I’m wrong, but I thought ISO 10993 “modernized biocompatibility within a risk management process” many years ago?

Colleague: FDA and the ISO committee for 10993-1 (and all subcommittees) are completely out of sync, and have been since FDA issued its “guidance” on 10993-1 in 2016. It completely reverses the direction in 10993-1, although it gives it lip service. They are idiots and did this to simplify for reviewers (e.g. now they don’t have to think). It was a sad day for biocompatibility fans everywhere. The lower species of animals are probably not too thrilled either; PETA left EXTENSIVE comments on the draft guidance to no avail, SO...the cost of all this unnecessary testing went UP because the test labs had to add MORE security to protect themselves from PETA and that ilk.

______________________________________________________

Post in a device RA forum:

I'm wondering if any other device RA professionals have had recent experience with FDA regarding the calculation of cumulative contact duration for evaluation of biocompatibility for use of identical single-use devices for around 30 minutes on multiple days in a single month, with an actual cumulative contact time of less than 24 hours.

In response to our effort to provide a justification for Limited Contact, the reviewer stated: "We appreciate your calculation regarding the minutes used per test per day for determination of patient contact duration for the subject devices. However, FDA believes that patient contact duration of a medical device should be based on the total number of the body contacting days, regardless of whether the device will be used for the entire day or for only a few minutes or hours of the day." And subsequently confirmed that "this reflects current CDRH biocompatibility review position for general medical devices" (rather than just applying to the specific narrow range of devices involved in this submission).

The reviewer made reference to the Agency's 2016 guidance on the use of ISO 10993, but I can't find anything in this guidance that would result in the above policy being adopted and it is not in alignment with the FDA-recognized ISO 10993-1.

As I see it, the practical result of such a policy is that FDA would no longer recognise any medical device as having a cumulative contact duration of less than 24 hours if it was used on more than one day, even for only a few minutes, i.e. it would be impossible for such a device to be classified as 'limited contact', because any contact during a single day would count as 24 hrs contact duration.

Which garnered this response from a device RA consultant:

Sadly I have seen this in a situation recently in the last year. I wish to give you some nuggets of wisdom or insight into your question whether this is a new policy or something new in the ISO 10993 standard and guidance - but I do not have any. There are interpretations made that I do not necessarily agree with and the whole cumulative contact time has baffled me.

In this one recent application we had a similar instance where the device was used for a few minutes each day. This was interpreted into actually putting the device into Long Term use - clearly not what we were expecting. While there is cumulative use we had some discussions also with the reviewer about the length of time in between each use. The body could "normalise" and more importantly we had a biological evaluation report supporting whether the device was used for 2 minutes every day or stuck on the body continuously for 30 days, there would be no difference. Sadly they did not agree with us.
______________________________________________________

I was told by someone whose company had discussed the matter with CDRH management, "They seem to know they have a problem, but they don't seem to be doing anything about it." Whether they choose not to, or don't know how, I can't say.

Elsmar Cove seems to have no character limits, but I'm going to pick up the matter of asking in a separate post anyway...

 

[Watchcat]

As for asking FDA versus us Covers, I think it is always useful to get multiple perspectives on an issue. That's the value Elsmar Cove represents to me. It is one of a number of sources I rely on help me avoid being trapped in the little tidepool of my own experience and that of others in my immediate circle. There are others who frequent forums for the same reason. There are those who see such forums as an opportunity for free consulting, and I'm sure they get their money's worth. There are also those who simply don't want to be bothered to do their own homework. There are those who see such forums as opportunities to share their knowledge, sometimes just to share, and sometimes "just to get business." (Ronen E, that one's for you. )

When it comes to device RA, the biggest problem I see, among both those who ask and those who share, is a lack of awareness that things change, and that one size does not fit all. One of the best RA lessons I was taught very early in my career was that "FDA is not a monolith." Anyone who has worked with multiple reviewers, to say nothing of multiple review divisions, can speak to the truth of this. (And the different centers may as well be on different planets, which is why I usually make a point of referring to CDRH, rather than to FDA.) Things also change over time. When it comes to CDRH, I tend to dismiss anything that happened prior to 2016 as a basis for current and future expectations. That year is something of an arbitrary marker, but as far as I can tell, it took CDRH a solid three years to even begin to get a handle on what FDASIA had wrought, much less start applying it. It has been doing the latter for a few years now, but it is far from done, and probably never will be.

As for asking CDRH, the main advantage is obvious; it's opinion is the only one that counts. I think most people who actively avoid asking CDRH would say that the main downside is that it might tell you something you don't want to hear.

I’m inclined to think that, if you give CDRH complete information, follow your pre-sub with a submission in a reasonably timely manner, and the submission doesn't contain new or different information from that provided for the pre-sub, the non-binding answer you get in the pre-sub will virtually always be the same as the binding answer you get from the review of your submission. So often it's a matter of getting the bad news now, or getting the bad news later.

Many startups would prefer to avoid pre-subs, because they don't want to risk getting bad news early, because it might discourage investors early, while savvy investors will insist on a pre-sub, and will want to see the results. In established device companies, where different product development teams compete for internal funding, some teams may want to avoid a pre-sub for the same reason, while executive management should be inclined to want them.

Regardless, it is still a business decision, one that RA can inform, but executive management must make, after weighing the resources needed to prepare for a pre-sub, along with the impact on project timelines. An established company, generating revenues from multiple products and in no hurry to get a particular product to market, may find it makes more business sense to skip the pre-sub and go straight to submission. RA might also inform them that what is needed is clear enough that a pre-sub would serve to confirm rather than to clarify. If it skips the pre-sub, the company may have to do additional testing later, to address deficiencies, but, if it is in no hurry, no harm done. The testing will cost the same as it would have cost if done prior to submission, with the difference that you now have near certainty that it is the right testing.

If rumors of issues with CDRH biocompatibility reviews have substance, I'd be inclined to see this as a tilt toward having a pre-sub, but not before I (or my biocompatibility expert) had decided which tests were needed, and prepared solid justifications for the design of the tests planned and for not doing tests that were not planned. But that's still just a tilt. Other factors might reasonably lead management to decide otherwise. (Or unreasonably, which happens on both sides of the regulated coin.)