PMCF Masterclass Elevating Post-Market Clinical Follow-Up

We're gonna get started in a few seconds. People are joining now. Okay, welcome. Everyone to the Pm. Cf, master class elevating post market clinical. Follow up with Bianca looters from Q. Sir Rachel Meade, from Psi, Gavin Quigley, from BSI, Marcus Tor from party, Pasco, and I'm sean Smith from middeck, leading voice. I'm gonna drop some housekeeping notes into the chat. In the meantime, if you want to drop into the chat where you're calling in from

please do. So I'd also like to remind everybody to please subscribe to Med tech leading voice. It's free meant tech leading voice.com and I'm gonna go ahead and put the notes in and turn it over to Bianca to introduce yourself. Welcome.

thank you, Shawn, and thank you for inviting me. I'm Janka. Let us. I work at Qive, a consultancy company in the Netherlands. and that focuses on clinical, regulatory and quality support for medical device manufacturers. and I'm head of clinical at Qusav, which covers the CO. Doing clinical studies for manufacturers and the medical whiting team, and PMCF. Team, who focuses on especially clinical inflations at this moment and collecting clinical data in such a way that it collects sufficient clinical data under the MDR. But with as limited effort as possible, not only for the manufacturers, but also for the healthcare professionals.

Okay, my name's Rachel Meade. I'm from the BSI notified body. My background is in research and development and medical devices. I spent about 20 years in industry before joining BSI where I worked as a scheme manager and technical specialist and particularly clinical evaluation specialist. So that's been my focus for the past few years. And I'm now a clinical, regulatory lead.

Hello, everyone! I'm Gavin Quigley. I am one of BSI's internal clinicians, and I work along with Rachel and the clinical regulatory team to provide clinical oversight. My background is as a consultant neurosurgeon from the UK. With an interest in functional neurosurgery so deep brain simulation and that kind of thing. and I'm looking forward to talking to everyone today.

And Hi, everyone thanks, Shawn, for inviting me to present today. My name is Marcus Torre. I'm the Pmc. Lead at Erdie Pascoe, based in London, in the Uk. My team help provide Pmcs survey solutions. The medical advice industry from design through implementation and reporting. Having conducted well over 350 PM. Tf, surveys since 2019.

Fantastic. Thanks. Everyone for introducing yourselves. I'm gonna go ahead and turn it over to Bianca. Are you gonna get us started with the screen share.

hey? Thank you. Alright. So I'm going to start off the master class today. So first of all, if we just move forwards, one slide. what I tend to do is go back to the basics. Sometimes I think it's really important, even if we feel we're very familiar with the subject, to go back to the fundamentals, to really understand what's driving some of the decisions that are being made, and and some of the questions that some of you receive from the notified body. So first of all, I think a really good point to start is Mdr article 2. So this is the article that provides a lot of definitions. And this is where we see introduction of the the term post market clinical. Follow up

so you'll see there. Article 2 48 clinical data is defined, and there are several indents, and I want to draw your attention particularly to that final indent. So you'll see the yellow highlighting. It talks about clinically relevant information coming from post market surveillance in particular, the post market clinical follow up. So why am I going back really, to these fundamentals? It's important to understand that post market clinical follow up is a source of clinical data. I think it's it's really key that we we all recognize that this is indeed a very valuable source of clinical data. So let's move forward another slide.

I'm just going to look at some other sections of the Mdr text. So, for example, in Mdr annex 3 technical documentation on post market surveillance, where it talks about a post market surveillance plan and what that is required to cover, and one of those indents. Again, it mentions. Pmc. F. So it says, a Pmcf. Plan as referred to in Part B of Annex 14 or a justification as to why Pmc. F. Is not applicable. We're going to come back to this statement shortly. Well, let's move on a bit further. So also, under Mdr. Annex, 14 part B post market clinical follow up. Is the specific topic of that annex. And

in Clause 5, you see, Pmc F shall be understood to be a continuous process that updates the clinical evaluation. So again, this is really emphasizing the point that this is a source of clinical data. Pmc, F is really providing that necessary data to update the clinical evaluation. And then it's mentioned in other references as well. So chapters 2 6, 7. You see, there are many areas of the Mdr text where Pmc. F. Is mentioned. So let's move forward one more slide. There are also various guidance documents that are available, and some of these are older documents, and many of you will be familiar with these documents from the past.

So medev. 2, 12. Slash 2. Rev. 2. This is very much focused on post market clinical. Follow up studies, and it's a guide for manufacturers and notified bodies. So if you aren't familiar with that that is available also Medev. 2.7 slash. One Rev. 4, of course, refers to it in the context of clinical evaluation, and then we have some of the newer guidances the Mdc guidances that are more specific, of course, to the Mdr. So you have 2027, which is a Pmcf. Plan template. and then 2028, which is a Pmc. Evaluation report, and that's guidance on for both manufacturers and notified bodies on Pmc. Evaluation reports.

So you might be wondering why I've included in that list, Mdc. 2021, 6. So this is guidance called Q. And a. On Clinical Investigations. And it was recently updated, I think, in December last year. and it's interesting to note that this, indeed, has some very valuable information relating to Pmcs study. So that's another voluble source of information relating to Pmc. F. So moving on from here. According to medd. Dev. 2, 12, slash 2. Rev. 2. And again, many of you will be familiar with this document already. We take a look at when Pmc. F. Is required. So, according to that, there are several scenarios that I've presented here on this slide.

And some of them have been highlighted. So first of all, we have a case where a C mark was based on equivalence. We also have a situation where it's a novel technology, or there are long term safety performance characteristics that are unknown in respect to that device. Or perhaps the device is designed for use within a high risk population or there are risks being identified from other data sources, so it might be not directly relevant to the device. It may be something that could be relevant to the device. And then there are other questions about assessing the performance of safety, of a device in in a more representative population. So perhaps it didn't cover the entire population when the preclinical study was designed and and executed.

And then, of course, there may be questions about the occurrence of clinical clinically relevant events later on, in the time frame of of the use of that device. So perhaps delayed reactions that we want to understand as manufacturers. So it's really important to consider when it's required and when it's maybe useful to consider carrying out some kind of Pmcf. Study. If we just click forward once, then we see that there is an obligation in certain circumstances to carry out a PMCF study. So, indeed, when A. C mark was based on equivalence, when there was a novel technology, or when there are long term, safety performs unknowns related to that device, and last, but not least, when it's a high risk population.

But I'd like to make one further remark on this slide if we just click forward again. we see that it's important to note that the decision to conduct Pcs studies. It should be based on identifying residual risks or uncertainties around the long-term safety or performance that might ultimately impact the benefit risk ratio associated with that device. So moving on from here. Thanks. So some common misinterpretations. So I mentioned earlier that I come back to this statement. So this is from Mdr. And X. 3, technical documentation on post market surveillance, where it talks about the post market surveillance plan requirements.

So where it says a Pmcf. Plan as referred to in Part B of Annex 14 or a justification as to why, and Pmcf. Is not applicable. I want you to think here about inserting the word study. So if we click once hopefully, we'll get a little arrow. I want you to think about inserting the word study there after Pmc F. And why? Because actually, a Pmc plan is always required. So this is a common issue that notified bodies come across is that manufacturers don't submit a Pmc. Plan because they don't think Pmc. F. Is required or is applicable to their device. But the default position is that a Pmcf. Plan is always required, but there may be circumstances where a manufacturer may justify that a Pmcf. Study is not required. There may be more general activities that are appropriate for that device depending on the risk profile of the device. But the manufacturer is always expected to document an appropriate justification for the activities that they're planning in respect to Pmc F.

So very importantly. Pmcf, we know it's not new. These are not new requirements, and they have been clarified in the context of the Mdr. But it is something that is is always expected to be considered if we move forward again. So we're gonna move into discussing a couple of cases now to help exemplify what we've been discussing so far. So the first example here is a manufacturer of a legacy device. So by legacy I mean a device that's been certified and placed on the market under either the Mdd. Or the aim Ddd. And in this case there is insufficient data to cover a particular variant. and the manufacturer is now proposing to conduct a Pmcs study before applying for Mdr. Certification. So I'll allow you to discuss, maybe in the chat or just have a think to yourselves, you know. Where do you think this is acceptable. So maybe put yourself in the shoes of you know somebody working for the notified body. What would you think about that proposal from a manufacturer. Is it acceptable?

Yeah. So I think some of you already commenting. If you want to just click one more time, Biancher, and then we can see the answer. So yes, I think in in general, as a notified body, we would certainly not. Particularly challenge that approach. I think that would be something we would potentially welcome. Obviously, whether the manufacturer can then use that clinical data from the Pmcs study to support that particular variant, and would depend on it being of sufficient quality and quantity. But ultimately II think that proposal we would not reject. So let's now consider some different scenarios. So the first scenario, if we can bring that one up, please.

So in this case the manufacturer is proposing to source the data from a device registry. So is this a good approach. So maybe those of you who had experience of of using device registries if you want to comment in the chat again. Okay, so we're seeing a little bit of diversity here. So okay, so some say, yes. Some say no. If we can pull the answer up. Okay, maybe a typical notify body response. But maybe yes, there are some some considerations and some you know, pros and cons of this approach. So we're we're gonna talk about those shortly. But let's now first of all consider another scenario. So the manufacturer is now proposing to source the data from a Pmcf. Survey.

Is this a good approach. What do we think about that? Okay? So again. yeah, it's a bit. But yes, no, maybe yeah, lots of different. So again, if we pull up the answer hopefully, yeah, I mean, I think it's it is very much, maybe. And again, we're gonna get into the detail of that shortly. When we hand over to to Bianca and Marcus. But first I'll I'll hand you over to Gavin, who's going to talk a little bit more on this topic.

Thank you. So I'm just gonna look. I mean Rachel's kind of tease, like some of the details that we need to think about. It's worth repeating that the Mdr text. it's a series of interwoven things that are dependent upon each other, and it's really important that we link them together as best we can. You can't just look at it in isolation. You have to look at it with the guidance you have to look at what refers to what particular annex. You know it's it's hard to get our heads round. But talking about things where we have a device that's previously certified under Mdd. Or Aimd. Guidance has been issued in the form of Ndcg. 2026,

which really wanted to get clarity for manufacturers, notified bodies to get devices that were already on the market through into Mdr. Certification. So if you could just next slide, please. can you push the slide on, Rachel? Sorry. Bianca. I apologize at night. There's a deliberate mistake in this slide. Can anybody see it in the chat? And that's because I've called it rank one of M. Dcg, 21. It's clearly rank 2

but I will correct that in slides. I think I accidentally deleted one of my slides. But you know these things happen

so rank. One is really where we have, you know, very good clinical studies, randomized controlled trials, covering all the bases, all the possible use scenarios. and let's be honest, they're pretty unusual. Even the best randomized, controlled trials always have a weakness because they have an ideal group that you're going to pick. No one would ever spend the money on a randomized control. Trial and kind of have very loose criteria, so we would never really look an expectation of every device having this sort of evidence. So we start to say, Well, okay, where? Where? Where should you be aiming, or evidence for devices that are coming through from Ndd. Aimd into Mdr.

And really, this document, this guidance document, tells you what types of evidence what the weaknesses and and the the limitations of that evidence are. And and really this is. you know, a way of making sure that these standard of care devices are getting getting through you. And and that was really the aim of the guides, and we we we'll look at this guidance, and we expect manufacturers to be adhering to this type of guidance, because it seems a very logical common sense approach. So perhaps if you could just skip on, I'm not gonna say too much more. Thanks, Bianca. Yeah. So rank 3 is obviously, then, we talked about registries.

Registries, I think, from a clinical perspective, are great. but they have limitations. It depends what the aim of the registry is, and how it was set up. Some registries are set up for reimbursement purposes, some are set up to try and work out. If a particular device is having a long term problem. If you think about things like the joint registries they were set up in the wake of problems like metal or metal hip. So they were set up to look at long term problems. And they can be a really good source of data. And, in fact, for most long term implants our expectation, at least clinically, is that where possible devices should be in registries, if those registries are indeed appropriate, and I think

this is the type of thing we're seeing

both from

government pressure, but also from the Commission. There's an expectation that this is something where we can get long-term data. The data can be maintained. And we can see what trends are developing. So registries for specific devices. Absolutely well, clearly. we're not gonna talk about registries for scalpels. That's just not practical. We are talking about generally for implants or or longer term devices. So if we could skip out to the next one, please. Yeah, I mean, there's just a couple of sort of highlights there. Really, you know, can all the device variance indications come from the data?

There are obviously device variants that are used infrequently because of size or anatomical concerns, or a particular group of patients who, perhaps less common in. We have to think of ways of supporting those particular variants, and some large registries will capture those. But smaller registries may not. So can I have the next slide, please. Bianca. Right? 4. This is kind of where the guidance says that. you know, for

for class 3 devices.

This is the the sort of th. This is. This is the lowest level, if you like that we can kind of accept and we should be looking for we we need to see level. We need to see at least level 4 and above.

And I think the important point here is that we're talking about high quality surveys. So these are

properly sized, properly, statistically validated, and that they have relevant clinical endpoints. That's really what we will be focusing on. So are those surveys valid, and will they cover safety and performance? So perhaps just if you skip on the next click form. Yeah, thanks, being kind of the next one. and then one more slide for me. So just to repeat myself once again. class 3 legacy devices and implantable legacy devices that aren't well established technologies and well established technologies is a specific definition, as you know. should have this level for data as a minimum.

And

the important point here is that this reliance on complaints and vigilance. This Pms type data is just not enough. That is not, we will not. And we will not accept that as being sufficient data. You know, we know that complaints and vigilance data is just not. It's not well enough recorded by clinicians for manufacturers to have, and I can say that with my hand and my heart having practiced for 20 odd years that I very rarely filled in complaint forms, because in the end of the day it was extra work. I was busy trying to get on to the next patient. So it's that reactive stuff is not really what we're really saying. And I think what the guidance is saying is, look. we are looking for manufacturers to prospectively collect

Pmc, F data. That's what we really want. We want to see evidence that. And you know, useful clinical data is being collected as we go along. And I think that reassures notified bodies. It reassures patients and ultimately manufacturers get useful data from that, and we would encourage. And you know I would encourage that both in my hat. if you like. for as a clinician, but also from a notified poly. That's what we really want. And I think that's pretty much me you'll be pleased to hear

I'm going to go ahead and launch the first poll question if I may. you should see it on your screen. While we queue up the next speaker. I just like to remind everybody to check the housekeeping notes in the chat, to use the chat liberally, or the QA. Function? We encourage your questions and feedback throughout the event, and afterwards. So you know, don't. Don't be shy, as they say and looks like a lot of you are filling in the poll. Question. Number one. Thank you very much. There are 350 people on this call. So welcome everyone and

for those of you joining late. we have 4 amazing perspectives on post-market clinical follow-up today. And excellent speakers. Delighted to have everyone here. Thank you all for taking time out of your day to join us again. We'll let the poll run for 30 more seconds. and then we're gonna get on with the rest of the presentation. So boom.

okay.

I'm gonna end the poll.

I know that some of you are still silly and ish. Okay. share results. Here are the results of the poll. I, are you able to see them? A, okay. So 48% clinical studies or trials? 31% patient registries, 64% surveys are questionnaires, 71% literature reviews, 31% analysis of real world data and 69% post-market surveillance and 5% other. So I'm gonna stop sharing

and move on with the presentations. Thank you.

Sean, myself and Danker will now build upon what Rachel and Gavin have just run through looking at a couple of case studies. And we're gonna build upon the the initial case study that Rachel presented, and just a quick reminder on what that case was. So we're looking here a manufacturer of a legacy device the device in question. It's gonna be a pick capital which is chosen because it's pretty simple device. But it's interesting in terms of its high risk. Classification being a a class free device is currently Mgd certified a. And the manufacturer is looking to for and to apply to Mdr certification. But there is

not currently sufficient clangor data, particularly on a specific variant of the device. So. as Rachel asked before to consider whether the registry might be applicable a Pmc. Yesterday or attend an investigation. We're gonna between myself and look at the options. And then, present The process of running through which of those approaches to choose, and a bit more detail. So, Bianca, I'll hand it to you for the next slide.

Thank you. The first thing I always start thinking about, especially for legacy device is. But what are the timelines that I still have? We are talking here about the class 3 device, which means that you must comply with the Mdr. The Mdr. Certified at the end of 2,026. That's not that long any more. and that's 3 years from now. but that's not 3 years for you. knowing that the notified body takes 12 to 18 months on average, to review an MDR. File, including your responses. that's quite some time. and you don't want to be the one that's the last and less than a notified body.

and that just doesn't get a time slot in there review queue. So from that perspective. maybe also have some slept left, and so that would bring you from 12 to 80 months, maybe to 24 months before the end of 2,026 to submit to the notify body. That means that you'll have less than a year left. So if you don't have sufficient clinical data. Now, what are your options?

If you think about the survey.

you need to develop a plan, you need to collect the data and need to report them. The development of a plan is simple, and then the plan for a full clinical investigation. And whether that's a simple observational study. or more complex study that differs a little bit. But for clinical study definitely more work than for the survey doing a survey in 6 to 12 months is definitely doable, so that would be nicely in time to still include in your clinical evaluation that you submit to your notified body. But if you then think on a clinical investigation, and for a pick, you want to have at least 6 months follow up after you introduce the pick into the patient.

And if you start thinking to day preparing documents, doing submissions to the Ethics Committee definitely required from a data protection perspective includes them yet, and wool them, and we pour them. It brings you too far over 2 years from now. which means that it will be difficult to use those data for your Mba submission. Keep in mind that at the moment you have an ongoing study and you have a decent data set available. and at the time of submission, or that becomes available shortly after. There may be options. But if you need to have your full clinical data set to support, that you have sufficient clinical data

being ready after February 2,026 will give you a very short time line for the notified body to refuse you and I quite a risk to be too late.

Honestly. so we're gonna look at the the device in question with just a little bit more detail. So you mentioned it's pick after it's a class free device. Kony, Mdd. Certified with the Mdr review ongoing with that lack of sufficient clinical data. In addition to this, we've got the intended purpose, so it is indicated for shorter long term access to the central Venus system. The intranas therapy power injection or contrast media and the intended users of health care professionals that are trained in using these type of devices be that nurses positions or physician assistants the patient population is both the adult and pediatric patients. And we've also been

provided with the sales volumes. You can see the devices sold across Germany, France, UKUS. And China, with volumes of varying quantities. And that's an important consideration. We'll look at shortly. Dagger handed Chief the the right inside.

Yes. So let's assume we are going to do a Pmcf survey high quality. and then let's start with what Gavin already explained, needs to have relevant end points except as criteria and a statistically sound sample size. And of course, you need to be able to collect those data where to start. If you have a decent clinical investigation report that already includes those endpoints and acceptance criteria it also will tell you if you have any clinical gaps. So if you are looking for specific variants that need to be covered, your CEO should tell you that those need to be covered.

So. based on a proper, CEO. You can define your endpoints, your acceptance criteria. you can you. You do that based on what you have included in your clinical information state of the art part and then you can dis determine a sample size from there on. It's not that difficult to design a PMCF. Plan, and with a survey protocol. and of course creates a list of questions that covers your end points because it's nice if you have a list of questions, but if it isn't linked to the end points, you still will not be able to show that your end points are met in the end.

Then you are going to collect the data. Why the Pmc F survey report and use those findings. To update the clinical documentation, which is the CER. Risk management and and Pmc, F. So that's the overall process. But then the next question is. what do you need to consider an actual practice. I'll for class 3, device level 4 data, which means a high quality survey and points acceptance criteria sample size. But it's nice if you want to collect those data. But you need to be sure that you can collect the data. So the first question is, can they be collected because those are included in the patient charts.

most class 3 devices. The answer will be, yes, if the answer is no. The first question is, do you have to right end point for a PMCF. Survey. because if you want to collect some very specific clinical data for specific endpoints. then A. PMCF. Survey may not be the proper medium to do so. Also keep in mind. If we talk about a hip implants the patient goes to the hospital gets a new hip. Relatively quickly goes home. May come back somewhere in a month, 6 months, 2 months, 6 weeks, 2 months.

Half a year a year, depending on local clinical practice. but they will not come back every year. Usually they will only come back at the moment they have issues. So if you want to use your PMCF. Survey to support a lifetime of your hip implant. that's not going to work because those data are nuts collected, the patients don't come back to the hospital. So then, if if you know what the clinical gaps are, and then the pick example that is, the pediatric population. Which means usually the smallest franchises and the largest franchises which I use least because they are on the other side of the spectrum from a size perspective.

so that's what you want to focus on. Where does endpoints? the ability to be placed, used and removed? Which means you are looking at the full use of the device. and also, of course, look at safety. Is there any adverse events? Keep in mind. If you are looking to doing a survey, you will only collect those data on adverse events that the health care professional feels is important enough to document in the patient chart. You, of course, can look at any clinical claims. and includes of label use and misuse. and the countries to target

and keep in mind the list of countries where the device was sold at decent weights. As explained by Marcus. and includes Germany and the UK. The US. May be an option. This only will work at the moment the intended use and clinical practice and patient population in the US. And in the EU. Is comparable. If that's not the case that cannot be justified, and you will end up with survey data that are not helpful for EU submissions.

Thanks, Bianca, I'm going to let Sean just run the the second poll, so there must be a quite good time for that question, so surely going to go ahead and give up on it to you seconds to answer that question.

So while the audience is tackling the second poll question, which is, have you done? Pmc. F surveys? Or are you doing? Pmc, F surveys? And while that is running? If any of you want to jump into any of the QA that has been submitted feel free. or we can just So this is interesting. It looks like about a 60 40 split between yes and no.

I'm gonna end the poll in a few seconds.

but I want to give you a little bit more time. so again, if you have questions. Please submit them in the Qa. We may not be able to get to all of them. During the call. We probably won't but we will try to follow up after the fact. And so I'm gonna end the poll and share the results. Which is exactly 60, 40 split and there we go. I'm gonna close this and next slide.

Great thanks, John. Interesting to yeah. See that that split hopefully. This required some insights into appropriate serving design. So as behind conventions, we've just we've determined a high quality survey is appropriate for this device. We we put together our survey protocol, our survey plan. We've got some robot, robust statistical rationale for our sample size. We determine what our end points are and what accepts substance criteria is, and we've also looked at who our end users are and which countries are gonna target. So the Survey protocols Survey plan is in place. The next stage is actually to put together the survey questions and the survey design.

I've got this table here which will outline some of the critical sections that need to be considered within the survey itself. And I'm gonna start with some information on the device. So we saw earlier the clinical, personal, clinical gaps, especially around the lar, the smallest and largest French sizes and variants of the device. so very important within the survey to cover this, so the question could be what French size if the pic was used with response, options being the available sizes of that device. So 4 French, 6 French, 8 French, and 10 French and rationale for this question to ensure that we get coverage across those variances device sizes. Particularly across those smallest and larger sizes that we're looking to collect data on

acceptance criteria. This isn't one of our primary or secondary end points, so no need for any acceptance criteria. But we would recommend some minimum quotas to be set to ensure that you are going to get representation and data collected on those 2 sizes, especially in which there is a current clinical gap and insufficient data so that will cover that initial clinical gap. And if you move on the anchor will look at the next thing which is patient information. Again, we had a gap there around the pediatrician population. so there would be a question in the survey around patient age. The response. Options here are patient age ranges, and I think it's important that we specify why we're collecting ranges, not

just asking for the the patient's age, and that's really due to trying to mitigate any risk of collecting patient, identifiable information in these surveys. And so we need to be conscious of that, and age ranges and help mitigate that risk. Again, the rationale is pretty similar to the the French sizes where we're wanting to get coverage across different patient populations. Particularly where that too little gap is on the paediatric population. And similarly, not because it is not one of our primary secondary end points. Nervous action's criteria for minimum quotas to ensure enough data collected on the paediatric population is recommended. Also.

Next, we're going to move into the so core critical bits of the of the survey itself, the performance and safety sections initially starting performance, which we defined as our primary endpoint. That was whether the catheter could be placed, used and removed. And this this question, just for the sake of this and fit all on one slide, we're gonna focus on the placement part. So be a section around device placement, and a question whether the the pick can be placed in the peripheral bachelor system without complication in accordance with the eye view. responses being yes or no, and if no was selected, a follow-up question or questions around which issues were encountered, and how the procedure was completed rationale here, and this provides procedural success supports the intended purpose of the device as the primary endpoint.

and because it is our primary endpoint, we would need to define it. That sets into criteria which would be determined by looking at the state of the art benchmark devices in literature to help determine what the acceptable rate would need to be, and here it is greater than or equal to 90%. Second endpoint is around safety, and we mentioned, that is around adverse events so ensuring we're collecting any adverse events that occurs while the paper is in use. Again, a yes or no initial response question. And if Yes were selected there'd be follow up questions for respondents, or end users to either select or list which adverse events are encountered, and follow up questions on the causality. So whether it is actually related to the device or not? And the severity. So what impact that might have had, whether a clinical or patient impact.

The rationale here provides information on freedom from adverse events courting our secondary endpoint, and similarly, it should have some level of acceptance criteria, maybe not the one that helps determine our sample size. That's usually from the primary endpoint. But it should have a a level based upon again the state of the art and benchmark devices in which it should fall within. Here, for example, it would be less than or equal to 4% of the overall adverse event rate then in the survey, then sort of core parts. But there's other questions that can be asked, and we don't want to turn into a marketing survey. But we do want to try and collect relevant information. And it could be, for example, testing

clinical claims it could be seeing to see if it had collected any off label usage. And an example here is looking at how easy or difficult it was to perform the procedure, using the capita with responses being a one to 7 like her rating scale. where one is extremely difficult and 7 7 is extremely easy. If one or 2 selected. Maybe some follow up questions as to why it was difficult, and to try and have a bit more granularity there as well again. This isn't one of the primary or second gen points. So no need for an acceptance criteria on this question. So hopefully, this just gives everyone an idea of. Okay. You've got your plan in place. You define your endpoints, you you talking about why you need to put together a survey. But then, ensuring that your survey questions. Align with what you with that plan.

The worst thing is is is to plan a survey and just go ahead and not really think about how it's going to be analyzed. And the data you can collect can actually be aligned with those end points. So ensuring there's those critical key sections are covered. Making sure those questions align with those end points, but also trying to build questions around that that maybe fill any other gaps you may have, or be able to test any other clinical claims or metrics that you might want to collect in your survey. So the surveys now been designed the next stage is to consider how the survey is actually gonna be implemented. And I think this should be considered at an early stage when you put in together the survey plan on the Survey protocol.

because the it's all well and good, having a really thorough, robust survey kind of protocol. But if you can't actually get it to the end user and get collect the data. It's not doing much help, is it? So? Thinking about things starting with the programming. So how is the survey going to be hosted? Is it going to be online? If so, you need to make sure, with your survey design that there's relevant programming notes. There's routing and logic and the survey to make sure that if one question is being shown related on a previous answer that that is going to be correctly programmed. which software service software is gonna host. The survey need to have considerations around Iso 1, 4, 1, 5, 5, and and software validation, and so on. And also to make sure when it is programmed, there's extensive link testing again.

to make sure all the questions are being shown correctly. The worst thing would be to have missing data, especially on your end. Primary end point, for example, when it comes to the analysis. Recruitment's an important piece as well does the manufacturer have. Well, how are you going to get the survey in the end user's hands? So does the manufacturer have an existing relationship with the customers? If so, is that easy, or is it? There's some complex logistics in terms of actually distributing that survey, and also does that impart some bias to the survey responses? If we're collecting adverse events. Essentially, if the manufacturer is directly sending out the survey is there an issue there?

Could you utilize panels of healthcare professionals? And are you considering, compensating people their time to actually complete the survey? Once the survey is launched. Field work management is quite resource, heavy, actually, in terms of staying on top of quotas, timelines, quality control checks and the data so ensuring that that's planned for causingly. And then, as the data's coming in as well thinking about how it's going to be analyzed, how it's going to be tabulated, how the data is going to be split, and then looking at it statistically, to feed back into those endpoints and ensure the analysis is done appropriately in accordance with those.

So there's a lot to think about, not just planning the survey, but actually thinking about how it's going to be distributed and how the data is going to be collected.

Let me go ahead and run as another poll question, if I may. letting me launch this poll question. and has been launched in this meeting. Are you seeing the poll? Question or not?

Now

stop sharing watch. Sorry. Just a reminder that the slides will be available along with the recorded presentation to everyone who attended. Whoever everyone who registered. I'm sorry within the next 24 h. And additionally, there's gonna be a short 4 question survey at the end of this thing, you know, asking you how you enjoyed it, etc. So if everybody will

kindly hang into the end. That would be the oh, wonderful! I I'm gonna end this survey shortly. Marcus, is there anything you'd like to say over the next 30 s.

And no, II was just looking through the questions now. So I was. I was gonna start potentially answering, answering some of those under quite a few of them. So pick one right now might be tricky but I'll hand over to Yank. Maybe get started on this slide. Just to think about what to do. With that data once collected because we have another case to run through, and I'm conscious of the time as well.

Great. I'm gonna go ahead and end the poll. Share the results. 68%, yes, 32%. Now. are you planning on doing a Pmcs study. There we go.

That's good news, because that should mean that we have sufficient clinical data in many submissions to come in the future. At the moment you have your data collected, you, of course, analyze them. create a report. Make sure you look at them, not only as Pmc F data, but really feed them into your Pms system, your Ps UR. For plus 3 device. Make sure you've also feed them into your clinical evaluation process. risk management and update all documents that may require you and update. Depending on your analysis of the data in and

Ems process, the clinical inflammations. And this may include updates to risk management design labeling the Ssap and plans for Pms and Pmcf the clinical evaluation and the technical documentation. So be prepared. That's a long list. Try to do that as efficient as possible. Let me end up. It's a waiter.

Thank you. Okay, so we're now going to move on to a second case. So in this case, we've got a manufacturer of a Class 3 device who claimed equivalence. and when placing their device on the market originally. and this manufacturer determined that there is no requirement to conduct a Pmcs. Study. So this is a little bit, you know, intended as revision, I guess from the the start of the the webinar this afternoon. Do we think this is acceptable? So feel free to make comments in the chat. couple of people coming back. Yeah. It seems like a resounding no, so far. So so that's good. So let's let's pull up the answer and confirm no. So I think we've known already for for a long time already that the expectation is that if a manufacturer is reliant on equivalence to bring the device to market. It is expected that the manufacturer is then, in the post market phase, going to be actively pursuing the gathering of clinical data pertaining to that device.

So and and we've got to think about it, even if the device really doesn't have many differences between it and the claimed equivalent device. Regardless of these small differences. Those can lead to significant changes over time in terms of performance and safety. So this is why it's really important to make sure that a manufacturer is actually carrying out. Pmc, F. In the post market phase. So let's look now at another example. So we've got the situation where the notified body has come back with questions to the manufacturer. and they've now decided. Okay. Well, perhaps we do need to do something. So they've decided to conduct a Pmc F. Study

following an observational design. Do we think that's acceptable? Then so again feel free to comment in the chat and see a few. Okay? So again, a little bit of diversity in your response, yeah, okay, so sometimes, yeah, maybe not. Yes, no, maybe. Okay. So I think I'll hand back now to Biancher and Marcus to to go through this in in a bit more detail.

Yes, I think this is the second case we're gonna look at. I think it it pretty much pretty repetitive. What rachel said here, but a reminder. We're gonna look at again. Similar device hit capita pass free. And the end. Mdr review. But this this time looking at the equivalent room for clinical data and initially, no specific Pmc. Covered in the Pmc plan. wonder whether that's acceptable or not. We just discussed that on the previous slide. But the anchor is going to go into a little bit more detail in terms of the considerations what Pmso activities should be considered, and how to approach him for this this case.

Yes. Oh. the advantage in this case is that it is an equivalent device. Political evaluation is accepted, with the exception that you don't have a PMCF. Plan with specific Pmc f activities forward. So that means that there is no urgency to do this clinical investigation, or at least to collect clinical data, which also gives you the time to think a little bit more on what is beneficial for my company. you still need to have VoIP class 3, device at minimum level, 4 data. but you can also use it.

Look at other inputs and then think on the marketing department always wants to have all those nice additional marketing claims. Can I include one or more of them in my study or in my data collection plan. What about reimbursement? Other jurisdictions? So think on. If we are now looking at the EU, can I use those data to go to the rest of the world and costs. knowing that many manufacturers spend already a lot of money to transition to the Mdr. Is it an option to do more than a Pmcf survey? Or do you have that many products that a Pmcf survey is not in all cases feasible anymore?

We have other similar devices. Do you have accessories. Do you have any data gaps? really try to look at the big picture? Because if you have similar devices, you may be able to benefit from at least looking at them as a bigger group. The one observational study, instead of many Pmcf surveys. may be helpful. think about dividing your Pmcf. For all those similar products. If you have 3 or 4 similar products, it may be helpful to divide them over a 3 year time frame, or over a 5 year time frame, instead of saying, we need to do them all every year.

It's you as a manufacturer who decides what a requirement is. and to provide a justification to the notified body. Why this is considered acceptable so definitely give it a try. Looking at high quality Survey versus an observational study, and also called a registry. And this is not the national registries, for example, for joints. That's Gavin talked about. This is really registry. I do as a manufacturer and observational study. At the moment you look at the costs. If you only have one device, a study is definitely more expensive than a survey. But if you have more similar devices and the more similar devices you have.

the lower the cost per device will be at the moment. You put them all in an observational study, because you only need to start up once instead of starting up many cases. the quality of the data of an observational study is higher than when you look at a Pmc. F survey. Even if it is a high quality survey, then you will end up, most likely in the level 3. If you do a brilliantly designed perfect study, maybe even a little bit higher. so higher quality data. But that may also mean that it is more valuable from a marketing perspective, more valuable from a reinvestment perspective depending on which endpoints you include.

If we talk about accessories that's very difficult usually to cover in any survey. You will be able to collect some data. but no one will include everything. If you are doing an observational study, you definitely can collect decent data sets from a clinical and perspective for those accessories. For additional claims. You really, a survey only takes half an hour per individual patients at maximum to complete for a healthcare professional. So that also means that you can't ask 60 questions in an observational study. You can put more questions

to the healthcare professional than what you can in a survey advantages. But keep in mind it's a clinical investigation, so more expensive than a survey. Per what you do something that's also important to consider. Be careful in what you promise. and let's be careful in what you promised from. I'm going to do a clinical investigation. Then you need to do a clinical investigation. because that's what you said. So don't over promise, but also not

over promise from a time line perspective. Keep in mind the slides that we shared earlier, where. 2 years from a clinical investigation perspective. If you still need to start. Is not that long? So don't tell your notified body I'm going to start preparing for the study today. I want to have 6 months follow up of 100 patients, and the report is done at the end of this year. That's not feasible at all. So keep that in mind.

I'm gonna launch the poll question while you're talking. If that's okay.

Yes. And offer to Marcus. Now.

yeah, thank you both. Yeah, just to wrap up here. I think we're looking at different cases. And think about surveys. But also Pmc updations. Think about what surveys can be useful. But also what that Pmc can use for? Is it solely for European data? Slowly, for the Mdr. I think there's different options that it's a really good option for collecting your sort of standard specific, continuous Pmc and data. We've we've looked at legacy devices moving to the Mdr we're looking at closing clinical gaps in the clinical data, and it's really good solution for that. Also. For example, you might bring out a new updated model of the device. And once you use survey as

an option to provide complementary data. And also there's there's potentially 2 routes. So you may have a device that's and yeah, certified. And you want to launch it in the Us. And through FDA submission, or vice versa. And Pmcs surveys could be a good option for collecting that data to use as evidence, and for that as well. And then, finally, you can add in into surveys questions around supporting marching claims and clinical claims as well. So there's there lots of different options that especially high quality surveys can be used for the anchor mentioned earlier about where data can be collected, European data being sort of gold standard. But if there is justification for similar patient characteristics as a a similar intended use, and also not dissimilar healthcare system.

There's also an option to collect data from other markets, especially for supplementation, and we are seeing the Us. Especially being one of those markets that is accepted. So just a sort of final summary in terms of what Pmc data, especially high quality surveys, could be used for

fantastic. There'll be a short survey that will follow this, so please hang on. I We'll try to follow up with the questions. Sorry we didn't get to all of them. but we're at time any closing remarks from the panelists.

Yeah, some great questions. Thank you. Everybody for participating in the chat.

Yeah, I've tried to answer some of those as we were going, but it's hard to the more so we'll we'll all get together and try and answer questions to between the 4 of us. But again, if anyone wants to reach out directly linked in email and feel free to to follow up and be more than happy to answer any questions specifically on answer the implementation.

Yes, happy to support you as well. I will make sure, indeed, that we get questions answered and provided to you. and thank you for listening.

Thank you. We will try our very best to get some answers. Obviously we will. As a notified body, we will give you reasonably generic answers. We can't go into specific devices, as you know. I know there's a lot of discussion about, you know. Pmc, F. And the plan, I think the plan is, the plan may not be a very detailed plan depending on the risk and category of your device. What we're saying is we would expect to see it documented. It may be very similar to your Pms plan, but it need not be overly complex, and I don't want you to, you know. I don't think we're trying to get you concerned. But remember, if you're a 2 A to be 3 manufacturer, you will need to produce Ps. Yours.

and that requires a Pmcf Update on an either yearly or by IM to yearly basis. So you know, you must engage with Pmcs. If you're gonna produce Psu's. And I think that's the focus of the Psu. And I know that is a little bit different. Compared to Ndd. But we will do our very best to get the questions, and I will fix my slides, I promise.

Thanks so much. Thank you. Everyone.

Time show organizers.

Thank you.

Bye-bye.