Regulatory Open Forum

 View Only
Expand all | Collapse all

When the perfect predicate was never commercialized

  • 1.  When the perfect predicate was never commercialized

    Posted 18-Dec-2017 00:29
    So, here's a new one for me...

    I have a new device that requires a 510(k) - there's only 1 other device (cleared in 2009) that has the exact same intended use within the ProCode.  However, that device was evidently never commercialized and the company no longer exists, so we can't purchase it to conduct head-to-head testing for SE.  Additionally, the sponsor submitted a 510(k) Statement in lieu of a 510(k) Summary,  I was able to track down the person on the clearance letter and send a request for information, but have received neither acknowledgement nor response.  I've also submitted a FOIA request for the device's original submission, but nothing received to-date - and no guarantee that the info I need won't be redacted.  

    As I suspect this is not an isolated incident, I'm hopeful that someone on RegEx has been through a similar situation and can share their experience. 

    Can I leverage standards related to this type of device (and assume that this predicate would meet the current standard)?   Am I up a creek and need to identify a Plan B?   Realistically, I could probably use a different predicate, but that means so much extra work to demonstrate that the differences don't raise new S&E questions = time and money I don't have.  The predicate I'm targeting is exactly what I need, so I'm hopeful for a creative workaround.  

    Looking forward to hearing about actual experiences,

    Tina


    ------------------------------
    Tina O'Brien RAC, MS
    Director of Regulatory Affairs
    Aroa Biosurgery
    Auckland
    New Zealand
    ------------------------------


  • 2.  RE: When the perfect predicate was never commercialized

    Posted 18-Dec-2017 15:14
    Edited by Vidya Jayaraman 18-Dec-2017 15:15
    Hi Tina,

    May be you do not need to test with competitor device for this submission? can't you just demonstrate compliance to ISO standards for your procode through regular V&V testing and make a case for equivalency based on the information you already have? Have you had a pre-submission talk with FDA?

    regards,

    ------------------------------
    Vidyalakshmi Jayaraman
    Regulatory Affairs Specialist
    MA, United States
    ------------------------------



  • 3.  RE: When the perfect predicate was never commercialized

    Posted 18-Dec-2017 15:22
    ​I'd love to be able to leverage the standard, but have no evidence that the predicate complied.  Could I reasonably assume that the current standard would be equally or more stringent than whatever performance data was submitted for the predicate, without knowing for sure the nature of that data? 

    Yes, I am planning a QSub to discuss with FDA. 


    ------------------------------
    Tina O'Brien RAC, MS
    Director of Regulatory Affairs
    Aroa Biosurgery
    Auckland
    New Zealand
    ------------------------------



  • 4.  RE: When the perfect predicate was never commercialized

    Posted 19-Dec-2017 07:21
    ​​Hi Tina

    I feel for you as I was in exactly this situation. Just to make things more interesting, our predicate was

    a) the only predicate and
    b) cleared on the basis of split predicates (on the advice of FDA, as we found from our FOIA request).

    What that meant was that the de novo path was closed to us since there was a predicate, but side-by-side testing was obviously not possible.

    Safety testing was clear-cut according to various ISO standards.

    As far as performance testing went, I put the following sentence into the submission:


    "Side-by-side bench testing to compare the performance of the proposed device with the predicate device was not feasible owing to the fact that the predicate device was never placed into commercial distribution despite a successful 510(k) clearance. The performance comparison is therefore based on promotional material made publically available by the predicate 510(k)-holder and the redacted 510(k) submission for Kxxxxxx."

    I was waiting to see if FDA was going to insist on side-by-side testing; obviously they couldn't and didn't.
    It worked for us. Not being able to test side-by-side or evaluate performance standards ended up a non-issue.

    Good luck!



    ------------------------------
    Jean Bigoney PHD, RAC
    Managing Member
    Nu Device Consulting LLC
    Murphy NC
    United States
    ------------------------------



  • 5.  RE: When the perfect predicate was never commercialized

    Posted 19-Dec-2017 07:43
    Tina, we've used unmarketed predicates many times.  It's still a lawful predicate.  Don't share that with FDA, it may erode confidence in an otherwise lawful predicate.  You do not need to purchase it and do head-to-head testing.  Just hire knowledgeable folks to figure it out and make your pitch to FDA  at the Pre-Sub.

    Mark DuVal, JD, FRAPS



    Sent from my Verizon, Samsung Galaxy smartphone





  • 6.  RE: When the perfect predicate was never commercialized

    Posted 19-Dec-2017 08:32
    Hi Tina,

    I am running into exactly the same issue with a new client I am having early discussions with.  Thankfully, there is a Special Controls Guidance document (a little out of date, but appropriate), and FDA has stated an Abbreviated 510(k) might be used to demonstrate conformance with the Special Controls. This is consistent with FDA Least Burdensome guidance recently released and their new proposed approach to using conformance to consensus standards, instead of predicate.

    The client still has to generate data to meet the Special Controls, so in the interim I am thinking I would consider a pre-sub to confirm the consensus standard/lack of clear predicate approach (due to non-marketing, etc) and see if FDA is expecting anything else besides the Special Controls which I would not anticipate.

    This is an interesting quandry indeed  - but not everything needs to be De Novo.

    I would suggest you contact the reviewing Division to see if you can use Least Burdensome and conformance to consensus standards and/or Special Controls instead of that perfect non-marketed predicate.

    Best regards,

    Ginger Cantor, MBA, RAC
    Founder/Principal Consultant
    Centaur Consulting LLC centaurconsultingllc@gmail.com
    +1 (715) 307-1850





  • 7.  RE: When the perfect predicate was never commercialized

    Posted 20-Dec-2017 13:45
    Thanks everyone - this information is quite helpful!  I'm confident we'll be able to present a solid strategy for our QSub.  

    Happy holidays!!!

    Tina

    ------------------------------
    Tina O'Brien RAC, MS
    Director of Regulatory Affairs
    Aroa Biosurgery
    Auckland
    New Zealand
    ------------------------------



  • 8.  RE: When the perfect predicate was never commercialized

    Posted 21-Dec-2017 21:56
    ​​I find this discussion very interesting, because it involves an area in which I have limited experience and knowledge.  However, that's never stopped me before. :)

    I've never been under the impression that head-to-head testing was required to support a claim of substantial equivalence. Certainly this is the most unambiguous approach, but the requirement is for the two devices to have the same "technological characteristics," which I think can be (which is not to say whether FDA does or does not) interpreted pretty broadly.  I think other approaches are often used, including the use of standards and historical data.  I have wondered if the perception that head-to-head testing is required might stem from the fact that many 510(k)s claim SE to another device manufactured by the same company, in which case, of course you do head-to-head testing, and eventually the perception develops that head-to-head testing is how it is done?

    I must say, I don't know about a pre-sub.  I'm also under the impression that pre-subs are not routinely requested for 510(k)s., but maybe my impression has become outdated since the release of the guidance...and while there is still no fee for a pre-sub. 

    I thought pre-subs were usually requested for 510(k)s only if there was some special concern or complexity about the 510(k) and the demonstration of SE.  Of course, there is a special concern in this case, which is that you can't do head-to-head testing because your predicate is not available, because it was never marketed.  I concur with Mark's wisdom that it would probably be best not to point out to FDA that your predicate was never marketed.  If the only reason for the pre-sub is to try to confirm that head-to-head testing isn't necessary, then it seems to me it could be a short hop to the reason you want to leverage standards instead of doing head-to-head testing.

    If you think you can put together a good 510(k) leveraging standards, I might be more inclined to just do that.  What's the worst that can happen?  That's not a trick question, because the answer to this question is in large part a business question, and can only be answered with a knowledge of your company and how this device fits into its business goals.  Assuming it's not a one-device start-up, where the entire company lives or dies on timely clearance of this one device, the worst might not be very bad at all.  But that's something only you can determine, in consultation with your management.

    ------------------------------
    Julie Omohundro, ex-RAC (US, GS), still an MBA
    Principal Consultant
    Class Three, LLC
    Durham, North Carolina, USA
    919-544-3366 (T)
    434-964-1614 (C)
    julie@class3devices.com
    ------------------------------



  • 9.  RE: When the perfect predicate was never commercialized

    Posted 22-Dec-2017 07:39
    Hi Julie,

    Often the devices are similar enough to  perform equivalently in the "clinic" but you are going to get the "show me tge data" argument from FDA, and rightly so. A paper based high level comparison is usually not going to work for SE.

    However, absent  clinical studies, the easiest way to show equivalence for similarities is to run head-to-head performance testing. Or if that cannot be done, then test your own device to key technical standards if any are applicable (e.g., ISO Standards for dental materials or response curves for ECG) and if you can't test the competitor then reference their marketing claims on their  performance to the same tests.

    FDA has really turned 510(k)s almost into mini-PMAs. It really depends on the review group, but certainly physical devices usually include key performance testing, and comparable head-to-head is what bolsters the equivalency claim. It also is used to support your marketing materials, if done well (scientifically valid,  validated well performed testing).

    Best regards,

    Ginger Cantor, MBA, RAC  
    Centaur Consulting LLC centaurconsultingllc@gmail.com
    (715) 307-1850





  • 10.  RE: When the perfect predicate was never commercialized

    Posted 22-Dec-2017 11:31
    ​Hi, Ginger.  Happy to have your thoughts on this.

    Sigh.  I wish I could tell clients categorically that a paper-based high-level comparison absolutely, positively will NOT work for SE, but I can't.  I do my best to discourage them, but often there is nothing left to do but send it to FDA and let FDA tell them, hopefully in three weeks, rather than three months.  (And then later kindly turn a deaf ear to their complaints about how long it is taking to get clearance, lol.)

    Otherwise, yes, I should think data would be needed to support SE, but not necessarily data from head-to-head testing.

    To me, the difference between a PMA/De novo and a 510(k) is the argument you are using the data to support, not the amount or type of data presented.  I'm doubtful that a full understanding of this difference has seeped all the way through FDA yet.  Most reviewing divisions and reviewers have not had a lot of experience assessing data from the perspective of supporting safety and effectiveness, rather than a claim of substantial equivalence.   This will become more common as with De Novo Rising, and presumably more internal training will soon be provided to support that type of review, now that FDA is getting user fees for De novos.



    ------------------------------
    Julie Omohundro, ex-RAC (US, GS), still an MBA
    Principal Consultant
    Class Three, LLC
    Durham, North Carolina, USA
    919-544-3366 (T)
    434-964-1614 (C)
    julie@class3devices.com
    ------------------------------



  • 11.  RE: When the perfect predicate was never commercialized

    Posted 22-Dec-2017 09:01
    I agree with Julie that often companies should actually skip the Pre-Sub.  It often empowers FDA to sit on your development team and pontificate what they think the requirements should be and suggest their own development plan (and inputs).  They can become architects of regulatory perfection asking for more than you are willing to give.  If you do a decent job identifying requirements you can then hand it to them to react to what you've done.  Its harder for them to reject your file, but by no means  of course.  If however it involves clinical trials and if you seriously miss the mark in their estimation, you may be back to the drawing board.  Its a calculated risk in making the decision to Pre-Sub or not to Pre-Sub, that is the question.  

    As for Pre-Subs, we typically deal in novel devices that are 510(k)s, de novos and PMAs and as a firm we do about 3 to 4 per month.  I'd say 80% are 510(s) and de novos, because there are far fewer PMAs proportionately.  Many clients have enough uncertainty that they want to do a Pre-Sub.  We have  Client Alert that uses a Gilligan's Island theme, something like "When a three hour tour can turn into a shipwreck."  It's still good but a bit dated.   I need to update it.  

    Mark DuVal, JD, FRAPS
    DuVal & Associates, P.A.

    Sent from my Verizon, Samsung Galaxy smartphone





  • 12.  RE: When the perfect predicate was never commercialized

    Posted 22-Dec-2017 11:40

    I use the term "novel" to refer to a device that has no predicate, semantically excluding the possibility of a novel 510(k) device. 

    I would happily consider using some other term to refer to devices that have no predicate, but so far "novel" is the only one I've seen being used in this manner, if only within certain circles.  Perhaps there are other terms being used in other circles?



    ------------------------------
    Julie Omohundro, ex-RAC (US, GS), still an MBA
    Principal Consultant
    Class Three, LLC
    Durham, North Carolina, USA
    919-544-3366 (T)
    434-964-1614 (C)
    julie@class3devices.com
    ------------------------------