A step-by-step tutorial on creating an approved PMCF plan
Dr. Nadine Leistner and Dr. Derk Arts show medical device manufacturers how to build, justify, and execute an approved PMCF plan under the MDR.
Derk Arts, MD PhDCEO, Castor
Derk Arts, MD PhDCEO, Castor
About this webinar
This webinar provides medical device manufacturers and clinical affairs teams with a step-by-step framework for building a Post-Market Clinical Follow-up (PMCF) plan that meets notified body expectations under the European Medical Device Regulation (MDR). MEC ABC Chief Scientific Officer Dr. Nadine Leistner and Castor Founder Dr. Derk Arts walk through an approved PMCF plan for an established, single-use urinary tract lubricant. The session covers how to define the PMCF scope, conduct a claim and clinical data gap analysis using the Instructions for Use and Clinical Evaluation Report, derive measurable parameters, and justify sample sizes when no specific scientific hypothesis exists. Finally, Dr. Arts demonstrates how to set up a study and standardize electronic case report forms in Castor to execute PMCF data collection.
Key takeaways
- Because the MDR frames PMCF as a proactive activity rather than strictly a clinical study, not all clinical data must be collected within a formal study setting.
- Manufacturers should begin PMCF planning with a two-step analysis that extracts claims from the Instructions for Use and maps them against high-quality studies in the Clinical Evaluation Report to identify data gaps.
- Case report form measurements must align with existing literature and medical society guidelines while remaining feasible in routine clinical practice so that new data can be pooled with existing evidence.
- When an established device has sufficient safety and performance data and no underlying scientific question for a power calculation, manufacturers can justify sample sizes using empirical values, literature, and clinical opinion combined with an annual interim analysis.
- Selecting standardized sites that apply gold-standard care and enrolling every consecutive patient helps prevent selection bias and isolate device performance from user or procedural variables.
- Addressing electronic data capture and standardizing forms early in the PMCF planning process enables manufacturers to reuse forms across studies and replace paper-based workflows.
Full transcript
...devices, wearables and EHR systems, electronic patient reported outcome forms to administer surveys and collect clinical data into a central platform and other custom solutions. Now, I'd like to introduce our speakers for today's event. Derk Arts has over 10 years of experience in medicine, research and data management. During his MD training, Dr. Arts identified a major problem in medical research. Due to the unavailability of affordable, user-friendly data capture tools, researchers were deviating to non-compliant alternatives, and it reduced their data quality, security and reusability. To solve these issues, he founded Castor, a health tech company that enables manufacturers to optimize clinical trials through the capture of high quality clinical data that is reusable, thus maximizing their impact. Dr.
Nadine Leistner has over nine years of experience in pre-clinical and clinical research. Dr. Leistner has the idea of creating a platform to bring the medical device industry and physicians closer together in order to preserve innovation under the new MDR and strengthen diversity on the market. During the company's journey, she identified the pillars for the bridge that is needed between medical device industry and clinical research. As Chief Scientific Officer of MEC ABC, she has led a team of consultants in the field of clinical affairs for medical devices. Now, I'd like to pass it over to our presenters to begin. You may begin whenever you're ready.
Derk Arts: Thank you so much, Ryan. Yeah, thank you so much, and welcome everyone. It's great to see so many people online, almost 200 already, amazing turnout. And very happy to be bringing you this this content. I've been speaking to a lot of manufacturers at various events that that I've attended over the past years, and it's clear that the challenge around MDR, and particularly PMCF and PMCF planning, is is huge, and I'm very happy to be able to bring you what I think is an amazing presentation and overview of how you can get to to a successful PMCF plan that will be approved by your by your notified body.
So this is really meant to to help you on your journey to to manage your your device life cycle and to to be successful on the European market. And so in this presentation, the majority of the time will be spent on reflecting on an approved PMCF plan and walking through what's what in there made the notified bodies positive and and helped the approval, and what are some other approaches that could be considered, which we maybe would result in a slower approval process. So really getting a good understanding of what are these little things that the notified bodies are looking for.
And then at the end, I will briefly show you how the proposed solution in the PMCF plan can be realized in in our platform, and then we move on to to questions. So make sure to ask your questions throughout the throughout the process. This is meant to be an interactive webinar. And I'm very happy to to address any of your questions, although I think most of them will be meant for Nadine, but I'll be there to to guide the questions to her. So enjoy. I hope it's very educational and helpful. Nadine, please just take it away.
Nadine Leistner: Okay. So my screen is visible?
Derk Arts: It is.
Nadine Leistner: Oh, wonderful. Like everybody starts with in the whole PMCF discussion is with the definition of PMCF and how we understand it. I know that the wording PMCF study is established because of the MEDDEV guideline 2.12, but in the Medical Device Regulation, there are wordings or sections where it's written as a PMCF activity. And that's is this is the whole path which is going through the PMCF plan we develop with our customer. And the clinical data is defined in Medical Device Regulation very clearly, and it's clearly defined that not every clinical data has to be collected in the setting of a clinical study. And that was was one starting point of the whole strategy for getting the PMCF approval with the things we wanted.
And we all know that this is a proactive collection within the intended purpose, and I want you to keep that in mind for the whole webinar because this was one of the hardest points to discuss with the notified body, and we wrote a section about it how we understand it and why this is so important. And as I recapitulate the whole PMCF planning process, there are three key things we we had in mind or where we were focusing on. And the first thing was the scope.
It is very hard just to define which scope you have for PMCF, but it's one of the most important things you you have to do at the start of the PMCF planning because you need to know which question you want to be answered and why you need to get them answered. And the second thing is are the key factors like population, indications, and all the things you want to or you have to define to to really collect the data. And the third part is the realization part: how to get those data, which should be addressed very early as well, and should be written down in the PMCF plan.
Post-market clinical follow-up is a proactive collection of clinical data, and so it has the function to answer questions for data you still need. So the first part we were looking for is do we have special need of clinical data, special population we want to investigate, or did we have a medical question we need to be answered? And these questions you have to answer yourself as well because those questions you can answer with post-market clinical follow-up if you have some of those. But for most products and many customers of us, there are there are no specific question for PMCF activity activity, like in our example I will explain later later I will explain later.
And for this specific question for activities, do I have enough clinical data? I know this question is the most asked question in the whole PMCF and MDR world because of Article 61 and the maintenance of the products and the and the transition from MDD to MDR certificates. And if you have a safe product and you have enough data, the question is what do you measure in a PMCF activity? What should you write down in the plan? And what we did at first was the analysis of present clinical data because PMCF planning or activity is nothing isolated from the from the rest of the technical documentation. So here's one of the first polling questions.
Ryan: Yes, thank you very much. So the first question that we have for everyone today, which you should be seeing on your screen right now, the question asks, what is your scope of PMCF plan? Your answer options are A, special medical question; B, missing data Article 61; or C, fulfilling general requirements. So take some time right now to take a look through these answers, think about how it might apply to yourself or your company or those who may be in the room viewing this presentation along with you. Again, the question is, what is your scope of PMCF plan? All right, we'll close the poll in just a second and take a look and see how you all answered. All right, so 61% of you said fulfilling general requirements was the scope of your PMCF plan, 29% said missing data Article 61, and 10% said special medical question. Thank you very much for participating.
Nadine Leistner: So this is like I would have expected it, and that's that's the reason why I talk about the analysis of present clinical data. We analyze the present clinical data because of in two steps. The first step was and is always with all our customer projects now to do a claim analysis, and the main resource of the claim analysis is the instruction of use for use, because you have to have enough clinical data for every claim you make.
And the second step is to do a really important clinical data gap analysis, and this is based on the claims we identified in step one and on the clinical evaluation report, because the main document and the main resource of if you have enough clinical data is the clinical evaluation report, as you can imagine. We brought an example, and I just want to tell you something about the example. It's a single-use product, a lubricant for urinary tract procedures, and it's very simple in function, and it's sold million of times and is on the market for many years. And this is I think a typical case for those who answer with the general requirements, the 60%.
And as you see in the color codes, you see that we thought about the claims you can get from the intended use per IFU. Einstein gel is a sterile, disinfectant lubricant for lower urinary tract procedures. Sterile is a clear safety parameter, and disinfectant and lubricant are mixed safety and performance parameters, but clear performance parameters you need to address in the PMCF. Einstein gel is a fictive name. And as we identified those claims, we go went back to the to the clinical evaluation report as a present source, and we just looked at those studies which were favorable. And because this is a quality criteria, only the favorable studies with the evidence-based approach and the statistical methods are approved are used for this rearrangement.
We rearranged the clinical evaluation report per claim. As you see on the left column, the lubrication, disinfectant properties, and safety are the claims. And then in the middle, you find the studies, very fancy names I created, and and then the number of interventions and patients were assigned. With this overview, you show how many interventions or patients you have per claim, and you can with this rearrangement of the clinical evaluation report data how many data you have or how many data you will need for further PMCF activities.
Derk Arts: So Nadine, a quick question. When you say favorable study, do you mean favorable outcomes or do you mean of high methodological quality?
Nadine Leistner: You can yeah yeah, you can use every study which you used in your clinical evaluation to show safety and performance, which met all criteria from literature research.
Derk Arts: Yeah, and so, but is there is it do you see that sometimes there's sort of a selection of studies, so that some studies are being ignored, or is that something a notified body frowns upon? Could you could you speak about that, that
Nadine Leistner: Yeah, the the the good thing is that the process of process of creating a clinical evaluation report is so so so binding from the MEDDEV 2.7/1 that if you do the whole process properly, then you only get studies which are with a high quality, and you highly need to re- rely on those data because this whole process were were were performed with those studies. So it's high class studies, and now you can use them for everything you want because they are approved by this legal process which is regulated.
Derk Arts: Yeah, and so when you say when you say favorable, it's not necessarily favorable outcome for the device, but it's favorable in the sense that it's high quality methodological study. Yeah, excellent.
Nadine Leistner: Yes. Yeah. And when you do this and you have this table, we you have to describe and discuss the amount of interventions and patients per claim, because with the table nobody can can answer or see what you have meant with this tables, because every notified body or auditor needs an interpretation, your your own discussion of what you found and what you is what you think your interpretation is. And if you discuss this in the plan, there are one cases which can occur. The first one is that you find found one claim where there's no data or few data. Then going back to the scope theme, you have a perfect medical question you can put in a in a PMCF study.
You have a question, you need more data, and it's more more easy or it's just one thing you can easily do. But in most cases, like I think in the 60% in the poll, you have enough data, the product is safe for many millions are sold, and the question is what do you do now? And then the scope of the PMCF should be an overview to show that you to show continued acceptability of identified risks and you do just do the general requirements fulfillment. So to sum it up: case one, you have a super scope to address. And for case two, you need to have another strategy where I will re- refer to later.
But keep that in mind, you have a you have a you have a clear you have clearly shown that you have enough clinical data, depending on your claims and your clinical data. So here is the next polling question.
Ryan: Thank you very much, Nadine. So our next question that you see on the screen right now says, do you have enough clinical data for product maintenance? Looking for answers such as yes; no, need additional data; or not sure. Please take a moment right now to think about, as I said, how this applies to you and your company. Do you have enough clinical data for product maintenance? We'll just let this run for another few more seconds, and we'll take a look to see how everyone has answered. All right, let's take a look. So, do you have enough clinical data for product maintenance? We had 44% as not sure, 40% no, need additional data, and 17% at yes. Thank you very much.
Nadine Leistner: So thank you very much for your participation.
Derk Arts: Nadine, I had a quick question about one of your previous slides. For the safety claims in your example, what was what was the data source? I'm assuming that was available literature, but maybe I'm misinterpreting that?
Nadine Leistner: Yeah, I just go back to the slide, if that's okay.
Derk Arts: I think it's the next slide, the one where you
Nadine Leistner: Next slide. There in in good written clinical evaluation reports, later there is a list of safety and performance studies, and there is written those studies address safety. And those safety parameters are very general, but every every intervention which was examined and there is no side effects can function as safety data, because it was an intervention, or the product was applicated or used, and there did not occur any side effects.
Derk Arts: Yeah, but this is all literature review, right? I think that's what's
Nadine Leistner: Yes, it's it's all a literature review.
Derk Arts: Yeah, and the the method for that review is dictated in the in the articles you mentioned.
Nadine Leistner: Yeah.
Derk Arts: Alright, so I think there's a couple of audience questions on that. I think they're addressed now, so we can move on. Thank you.
Nadine Leistner: Okay. So now I want to guide you through the key factors. When we write clinical evaluation reports, I'm always asking of do you have an indication or a population, and did you define users? Because those questions are very important for the PMCF planning, and is something you should address at the beginning of your PMCF planning actions. And the thing I think many people or many products or people have is to define the right parameters to measure the the amount of right clinical data. And that is the next part I want to discuss with you. And we defined the parameters in two different steps. The first step was the deduction of claim of parameters through the claims.
So the claims we identified in the first part, in the scope part, are now used to deduce some parameters. And the second step is what is the how do we define the measurement of the parameters? Because from sterile or disinfectants, you don't get any parameter you you can imagine. And this I want to guide you through with the next section. And on the left, we have the the slide from the from the first part, and and now we see that the original reference in the row one, disinfectant, is the claim that the product has disinfectant properties. And the parameters can be the incidence of post-operative urinary tract infection. You really need to understand the product and how clinical research functions to just deduce it.
I just have those examples because it's very hard to explain how I did it. But for lubrication, you see in the in the second row that lubrication you can parametrize with the haptic feeling and with the problems that that no problems occur with inserting off the devices or catheters. And this is one of the hardest part you should discuss in the teams. And in the last row on the right, column on column on the right, you see short term, long term. This has an this has a really important function for the planning of PMCF activities, and which schedule, time schedule, what to communicate to the sites, and it's very important to even show this.
Derk Arts: So Nadine, would you say you always need a subject matter expert to be able to derive parameters?
Nadine Leistner: No. I think you're just in the the manufacturers has the knowledge in house, and I think they have to discuss which parameters are the best. And then it's just a direct link to the literature, because there's so many studies in the clinical evaluation report which are discussed, and look at those studies. And if you outsource it, then you talk with the one you have outsourced the the question, the clinical evaluation report.
Derk Arts: But I can imagine that the the parameters that you define can be a source of discussion with the notified body. Is that correct? Because you might be you might be choosing the parameters that you know your your product performs well on.
Nadine Leistner: Yeah, the notified body won't give you the parameters, but you can present the parameters and ask if they are right, but I don't think that that they will interfere with it. It's your choice. Your knowledge, yeah. That is the thing you have to cook yourself.
Derk Arts: All right, that's good, that's good.
Nadine Leistner: So I go on? And if you have a parameter, as you've seen on the left side, the incidence of post-operative urinary tract infection, you still need, on step two, as we discussed, the definition of measurement, because the incidence is not in a CRF form or a measurement form. And the decision for the questions you have in a case report form of the study are based on three main things. The first one is what are the measurement ranges in the literature, the evaluated one in the clinical evaluation report, which have very high quality, and the best is that those in literature are really similar to those in the medical society guidelines, because you're always right if you use parameters like in the medical society guidelines.
And the third thing you have to keep in mind is the feasibility in routine, because sometimes some fancy things you can measure, but you won't get any data or any recruitment rates that are sufficient if the feasibility is not giving. And in in our case, the report of urinary tract infection was defined as the question from the physician if it's symptomatic or asymptomatic and the laboratory values for leukocytes and nitrite. I know it's a very simple example, but I think so everybody understands it.
Derk Arts: We've got a couple of questions. Sorry, that that I think is more efficient to ask them now instead of afterwards. Because there's some questions about the short term and long term, if it refers to the time involved in the study, they don't necessarily understand. But then also, expand on the PMCF planning, the key factors short term and long term. So I think there's a few questions still around this short term, long term concept. Could you elaborate a little bit more so to make sure everyone in the audience understands what that pertains to?
Nadine Leistner: Yeah, short term is in our understanding everything which is directly within the intervention or the application of the device or minutes after, but when you see when you talk in a study schedule visit plan, it's directly when the patient is there. And long term is everything when when the patients has to come in a second time or longer. So this is no depending on the long term, short term risk class association, it's just for the scheduling of the PMCF, which is very important for the planning.
Derk Arts: Yeah, I think that's a very clear answer. Thank you.
Nadine Leistner: Okay. So what you get if you if you develop a measurement range and a parameter for every claim, you get a PMCF parameter set, as we as we called it. But the the reason to to have those sets is that you show that you can measure every claim you you make. It's not that you have to measure every claim, it's that you know what you want to measure or you can measure everything.
And that depends on the PMCF plan if you want to measure every parameter of the parameter set, but you can show the awareness to the notified body that you know, okay, I have those claims, I know how to measure, and this is my set, and for PMCF activity number one, I pick claim one, two, three or all of them, and you have you have shown that you know what you are doing. And coming back to our example is that it's a product which is sold million times and very very long on the market. And and as you've seen, the parameters you measure were very simple.
So our strategy was to measure every parameter of the parameter set so that we could show that we measured all safety and performance parameter, which you see on the right side, is on the from the first one of the first slides, to confirm safety and performance. And in cases you cannot measure everything because the the time is not there or the resources is not there there, then then justify which you use and which not. And I would recommend to make it on a risk-based based approach, because all the Medical Device Regulation is about the risk-based approach, and just justify what you want to measure and what not, and justify why you use this.
And with this plan, you see those arrows, you confirm safety and performance, show that you continue acceptability, and you detect the emerging risks.
Derk Arts: I have a question here on the disinfectant properties. The question is, because I think that's also the safety parameter, so with a risk-based approach, I would assume you would prioritize measuring that parameter. But the question, also because it's quite simple to to prove, they're asking, could we not use preclinical tests, for example, and so rely on another way to prove the disinfectant properties?
Nadine Leistner: Yeah, we discussed it a lot, because many things you should could measure in in preclinical setting, but this claim is so essential for the function. It's a claim a a clear performance claim of a product which with this with these three things and one is of those is so important that we decided that we use this. If you do it on a preclinical site, then try it, cook something and present it to the notified body, and you will see what the notified bodies will will say to that.
Derk Arts: Yeah, I think that makes a lot of sense, although given the the amount of projects they have to assess, it's probably good to at least have an estimate of of how likely they are to to agree with you, because otherwise you might end up sort of back in line for for your next proposal. But okay, that's a very clear answer, thank you.
Nadine Leistner: Yeah. And now I'm coming back to one of the most important questions of the PMCF world is the sample size. And I take you back to the first slides of how we understand that the PMCF activity, and for those products which are so long on the market, are safety and performance is shown in many studies, you have data, they are on the market, and the PMS shows there are no problems occur, there is no scientific question underlying. And if there is no scientific question underlying, it's you can't define an end point or perform a power calculation.
The power calculation is for all the other cases: if you have a medical question, if you have a hypothesis, or if you have something to be to be understand or to be to be understood. Try to justify a reasonable sample size with empirical values, literature-based, clinical opinion, etc., because just don't post a number or write down a number and say we want to do 10 or something like that. Just try to yeah, it's all about the cooking of the document and to just offer something to the notified body. These are the things the the notified body will discuss with you many times.
But with the understanding of PMCF as activity in the first site, and you have no question, and you have clearly shown with the claim analysis and the gap analysis that you have no question, you are safe, then try to justify a sample size which is not calculated. I will on the fourth bullet point will edit as well. And we combined the justification for the sample size, which is not the power calculation, with an annual interim analysis to show that we can react if something occur, if some risks occur or some side effects occur. We show with the annual interim analysis we are have understood the risk-based approach, and we will act as surely and with the PMS system that we change something.
And if you finished your first current PMCF activity, you will be able to justify or calculate another sample size with current data. The especially the first time is very hard to to numerize, to to get some numbers or calculations, but in the second second part or the second analysis, it's much more easier, and then you justify another sample size, or you have a question you need to to calculate. So here's the next polling question.
Ryan: Yes indeed, we've got another polling question for everyone here today. The question asks, how many PMCF plans do you already have approved? Your options are 0, 1 to 3, 3 to 5, or not sure. Please take a moment to think about your answer and answer as appropriate. The question again is, how many PMCF plans do you already have approved? All right, we've got quite a lot of responses. Let's go ahead and take a look at those results. 43% came in at 0, 28% at 1 to 3, 14% 3 to 5, and 15% say not sure. Thank you very much for participating.
Nadine Leistner: Yeah, thank you very much. Sorry. And to finish the key factors, going back to the indication and the population. The whole PMCF concept is also to generate real-world data or real-world evidence. And to measure the real-world evidence, you need to to have a defined population and defined indication. I know many products which have a very broad intended use, where they don't specify for whom those products are, but for those PMCF activities, it's very important to have a defined population and a defined indication. And to stay in study terminology, try try to make it possible that you do a all-comer study with on-label use within the intended use. And the selection bias is the second point I want to talk about in this context.
Many products like this lubricant, for example, have have a problem that the effects and the performance and the safety is affected by other devices or the user. I think of surgical devices or many other things. They have the problem just to measure performance when other things influence the the the own performance of the device. And what we did in this plan is that we said that we use want to use standardized sites where we are sure that the gold standard is applicated, because then we are we are more sure that that the usage of the other devices and the users have this have a good quality, and we really can identify possible side effects or performance problems.
And we used in this case also or we wrote down in the PMCF plan also that we surely will investigate every adverse event like it's described in the PMS system. And then, per site, we said include every subsequent patient, so that no selection bias will occur in the in the in the sites. Coming back to the realization part, where I just shortly will talk about because this is something I will pass over to Derk. We really early thought of how to get those data. During writing the PMCF plan, think really early about how you get those data, where do you get those data, which systems, which which if so software.
And for us, it was very clear that we don't use a paper-based CRF because that wouldn't make sense. And especially for long-term PMCF activities to to to generate general requirements fulfillments, it doesn't make sense to have some paper-based systems, and so we really got early in contact with, example, Castor EDC. Yeah, and I'm almost done with my part, but I want to summarize step-by-step what you should do and give an overview. As I said, define indication and population as I described it for the PMCF activity and the all-comer aspects of doing of collecting real-world data. Discuss your medical claims within the team, and adjust if there are medical claims which you cannot cannot prove.
And then, use the identified claims to discuss your clinical data from the clinical evaluation report, because PMCF plan is not isolated from any technical file, all is written or all links are going back to the clinical evaluation report, the same as the clinical development plan, which is coming with the Medical Device Regulation as well. And then, write down and justify and describe every decision you made for the PMCF: then the scope, the sample size duration, write it down. And therefore, you need some really clear statements. You are the manufacturer, and you have the knowledge, and you know what you do. And that's that's the part you have to show in the PMCF plan.
And then address the realization as well as the plan, because the notified body likes the plans, but he wants to know that you really collect those data and how you collect those data, eg site structure or data management. And one of the most important part is that you have to tell a story. You you have to show with this document that you know what your product is about, that you know it is safe, and you know what you want to measure on the PMCF activities, and you show that you know what the risk-based approach is and what the Medical Device Regulation is all about. And this you show with with the consistency of every section in the PMCF plan.
And now I'm coming back to the don'ts. Don't plan a sophisticated study without reference to the clinical evaluation report or the instruction for use. This is because so many people just think, okay, I'll do a PMCF study and something fancy the physicians say or which I think is good, but has no reference to clinical evaluation report, makes absolutely no sense, because the clinical data you collect in the PMCF has to be this has to be matching with the with those in the clinical evaluation report. And that is something like in the second bullet point: don't measure parameter which don't match to the existing clinical data. The parameters have has to be very similar or even matching with the existing clinical data to pool those data.
Otherwise, you have different parameters and can't pool those data or or prove that you're claim that you have enough data for any claim. And then don't measure parameter which are not feasible in clinical routine. Some fancy parameters or some physicians will say, yeah, I need those MRT data or CT data, but you know this is too cost-expensive or it's not it's not feasible, it's disturbing clinical routine, and nobody will do this. And above all, don't forget ethical standards and data protection, because some of those activities will result in a clinical study and data protection.
And so keep in mind which kind of data you have, like pseudonymized or anonymized data, and write everything down that you that you have in mind, because this is all about tell that you know what you are doing, and why you are doing it. This the why is so important under Medical Device Regulation.
Derk Arts: So if you have to give like one one reason why this this plan was approved, maybe one one or two, what do you think are the key factors that determined the success of this PMCF plan? Was it the was it the story and the why, or do you think it's something else?
Nadine Leistner: Yeah, I think the the the the key factor was that we discussed the medical claims and showed that we have enough clinical data for the product, and the scope was defined and written down. And then we just thought of how to measure it and had some some answers and the whole strategy was was written down in the PMCF plan.
Derk Arts: Thank you.
Nadine Leistner: Okay, so for my part, I'm through with it, and I will hand over to Derk.
Derk Arts: I think was there one polling question in between here or?
Ryan: Yes, that's right. We've got one more poll question that we'd like everyone to participate in. The question that you see in front of you right now says, for your PMCF plans, do you already know how you're going to execute them? Simply like to know yes, no, or not sure. Again, the question, for your PMCF plans, do you already know how you're going to execute them? Everyone's getting much faster at this. Thank you very much for participating. Let's take a look and see how you all answered. All right, 39% came in at yes, 33% not sure, and 28% following at no. Thank you very much for participating. Back to you, Derk.
Derk Arts: Is my screen being shared?
Ryan: No, you should see a pop-up right now. Let's try that again for you.
Derk Arts: Yes, here we go.
Ryan: There we are.
Derk Arts: And so I just wanted to briefly touch on how do you actually how do you organize data capture, and how do you make sure that whatever you put in your plan will happen in real life? And so of course, one way to do that is to use a system like Castor's that makes it very easy for you to set up a new project and to import or create forms that you know as defined by your PMCF plan. So what I'm showing you here is our system, and what I'm doing here is I'm setting up a new study. We already defined the the forms that were mentioned in the PMCF plan.
And because what I would recommend the viewers is to ensure that they try to standardize as much as possible. So we are really about data standardization, maximizing the impact of data, and so if you look at your entire product suite, try to try to see what overlap there is, and to what extent you can standardize forms so you can also reuse them going forward. So what I'm doing here is I've I'm logged into the system, I'm about to create a new study. So I've named it, I created an institute, a country, saying it's an observational study, and I'm going to go ahead and create a study.
And I had a colleague of mine actually build some of these forms that I will now be importing, just to show you how how fast it can be, especially if you've ensured that you build your build your forms ahead of time and standardize across across studies. And so what you already see is that I've now imported my my forms here, and I have a full study available here with with eCRFs, with with surveys that are defined, and you know this will allow me to to quickly kick off data collection. See if I can create a record here. And what I think is is key here that the sort of the simplicity for you know creating a project and reusing forms.
As you can see here, I'm jumping straight into the system, and I can start capturing data. Now, of course, you know there's a lot more to it. You need to you know find sites, start up the sites, train them, but I think it's it's key to to have a strategy around your data and select a system that makes it easy for for your sites to participate, to capture the required data to support your claims, and to be very efficient about it, and because this does not have to be a huge burden for you.
By using technology, by using standardization, and by by following the instructions that Nadine gave you, this this this challenge can be managed, and we can actually capture PMCF data at scale in an economical way. And with that, I would like to jump to the remaining questions from the audience, so I will stop sharing my screen. And of course, if you're interested in our platform, you can always reach out. We're happy to give you a demo. You can also try out the system yourself. So there should be there shouldn't be any reason not to to experiment with with our platform to support you with your PMCF activities. So on to the questions. I've asked a few already.
Let me see if if more came in what I was presenting. Expanding the chat window here. And we might not get through all the questions, and we will ask Nadine to answer some of the questions afterward, so and and share them with the with the audience. But let's let's just kick it off and take it from there. So I think a first interesting question is who approved this PMCF plan? Are we allowed to share that, Nadine?
Nadine Leistner: I don't want it, no.
Derk Arts: All right, so that's going to remain a mystery for everyone. And there was a question around the existing body of evidence. So, I'm not sure if this was understood correctly, but I want to make sure that it's clear for the audience. The question is, can we really cherry-pick studies from the CER? I'm not sure if that's that's what you said, but maybe you can reflect on that question.
Nadine Leistner: Yeah. It's not cherry-picking because it's just a rearrangement. The safety issues are widely discussed in the clinical evaluation report, and you not took those studies away which which haven't the bad results of your product. You just use them for performance those which show something about performance. And with the use of only those favorable studies which you within the five steps of MEDDEV evaluated and appraised, you use only those, but you don't let anything away, and you just have to show the performance and not the safety on this side. And if there will if there will be occur any concerns regarding safety, then this will result in the some statements of the benefit-risk conclusion in the clinical evaluation report.
So the the whole document clinical evaluation report is never cherry-picking, but for those regarding claims, you just rearrange those, but you don't let anything out of it, but it's more focus on performance than on safety for the PMCF.
Derk Arts: Thank you. Moving on, some questions around claims. It's a tough question, I think, but interesting. There's a question around an example of how this approach would apply to software as a medical device. I'm not sure if you have any experience with that, Nadine. Is there anything you can you can say around that? Does the same approach apply?
Nadine Leistner: We have some early projects, so I have no really experiences, but I could imagine that the claim understanding of claims should only should also work for the for the software as medical device devices. I think that the data collection or the realization could be much more easier because you can directly transfer those data from the devices or from the user or usage of the app, but the claim concept of PMCF and Medical Device Regulation should be applicable to those software products as well.
Derk Arts: Okay, that's I think very clear. Then a question, a broader question, for an implantable device, an intraocular lens that's been on the market for 60 years, is PMCF data still required? It's the audience question. It's quite quite a common one.
Nadine Leistner: Yeah, you you can discuss this with the notified body, but to my knowledge, you have to proactively collect clinical data for every product. It could change, or it depends on your notified body and how to justify, but in this in example, I repeat it, there were millions of sold products, and very safe, very simple product, and we have had to show a PMCF plan to really collect some data.
Derk Arts: Yeah, makes sense. Question about the when you're a plan is about a specific medical question, so when you are answering a specific medical question, wouldn't that likely be a new parameter?
Nadine Leistner: That depends on the question. I imagine, yeah, not it's not always a parameter, it's just to to prove a claim, or just to In this case, you can think of a new property you want to give the gel, and you want to prove that this property is as well a performance thing, and there's a there's a hard border regarding clinical investigation and within the intended use, but if you collect the data you have, and you don't have any intervention or something, then you can you can answer those medical questions. But it's medical question don't lead to any parameters as a must, from my point of view.
Derk Arts: Clear, thank you. Then on the sample size and on study design, a few interesting questions here. So what is the clinical opinion for justifying the sample size? Do you need a recommendation from a clinician, or can it just be justified against available literature or both? Can you comment on that?
Nadine Leistner: I would say the more the better. Use literature and clinical opinion, and and write it persuasive.
Derk Arts: Okay, so it's it's important to to to leverage your your KOLs in in that situation. And so when communicating with the notified bodies, do you have to provide a synopsis of a synopsis of a planned clinical study, or can you just initially present a high-level overview of the study without, for instance, the sample size calculations or a very well-defined endpoints?
Nadine Leistner: That depends on the state of certification. I think later on after the 26th of May, you have a really clear concept and really implementation plan and synopsis as well, yes. But now for now, under MDD and and the certificates which are renewed, there were different concepts which work, I think.
Derk Arts: Okay, and then we have another interesting question, which I think would be open to interpretation, but so if you're trying to sort of provide real-world evidence, right, which is of course the point of PMCF, why would you just select sites that are experienced and quote-unquote the gold standard, and why not use all trained users in/physicians, because this will give you the most realistic evaluation of safety and performance in in a given real-world setting? It's an interesting question.
Nadine Leistner: Yeah, there there's an imbalance between the real-world evidence and the real effect of the product, because this standardized gold standard site design or strategy is just really important for those products which are not the main products of the procedure. If you have an implant or something which is the main part of the intervention, then it's clearly you would not just focus on the gold standard on the real world. It's those concepts working sometimes against each other, but sometimes not. But for those products, it's much more important to measure the real effect of the product than just to to have a list of adverse events which don't have anything to do with the product itself.
Derk Arts: Yeah, that's that makes a lot of sense to me, so hopefully also to the to the audience. So then a question on novel devices. So a lot of examples in your presentation leveraged existing literature to to support specific decision decisions in the process. So how do you handle this with a novel product where there's only a pivotal study and there's no substantially of substantial equivalent device available?
Nadine Leistner: Yeah. Then the the the question is if the novel device has to be investigated with a clinical investigation, then the the then the concept of parameter definition will be similar, because also for the clinical investigation, you have to define which parameters or which endpoints you have. And the decisions on which parameters you choose for those devices, for the novel devices, will occur there also. And the literature thing which which you don't have an equivalent device also depends on the guidelines and the literature and the studies what is was what is measured there. But I think there are no better justification than the academia and the medical society guidelines you can use as a manufacturer.
You you have to take this as the lowest hanging fruit, because that they are quality standards for medical guidelines and for literature. So this is the best part you can you can take the decisions or justifications from, from my side point of view.
Derk Arts: Okay, great. Thank you. And then a more broader question, somewhat more generic. Someone is looking for a crisp definition difference between post-marketing surveillance and post-marketing clinical follow-up. So maybe you could offer one, because of course there's a lot of information available on Google, but sometimes it's hard to sift through it all and get a succinct definition.
Nadine Leistner: Yeah, I think the the keyword is proactive, and the other one is reactive. So the PMS reacts if something occur, and there are those deadlines when to to to show to to make a call, yeah, we have a problem, we have a side effect, and the other thing is a proactive prospective, I think one definition could be prospective data collection, and the other is retrospective, so perhaps it's more understandable with this context.
Derk Arts: Well, I hope that helps our audience member. And I think we have time for a few more questions. I see another interesting one here. So there seems to be an overlap between clinical evaluation and the activities you are describing, and when doing clinical evaluation according to the MDR, are you already doing this this mapping that you described of claims and and parameters and and data? And do I understand correctly that the examples are for products approved under MDD or AIMDD?
Nadine Leistner: Yeah. The first thing is that they are overlapping things, that that makes sense, but if we got a project for writing a clinical evaluation report, we early have the medical claim situation in the in mind, and address those question really really early during the writing process of the clinical evaluation report, because we know later it the the question will arise and then the clinical evaluation report and literature route is performed and we had missed some search strings, and therefore, we address those medical claim analysis really early in the process of writing writing clinical evaluation report. And the yeah, and the current PMCF plan was approved under MDD with an aspect of being ready for the Medical Device Regulation.
Derk Arts: Okay. And then another question: can PMCF be used to collect information about known off-label use? That's interesting.
Nadine Leistner: Yeah, that's a very tough question, because as we understand it, you are you have to measure off-label use. This could lead to to expand your intended use if you have many off-label uses. Ideally, you would add a question if if the or you will write down if there was an off-label use, because this this should be done, but but I will not recommend it now, because this is very hard decision, and and if some physicians do something you can't influence, and then you are then you have to broaden your indication, your intended use because of those physicians. Yeah, it's very hard, but but you should, you really should, but the time will will bring the the answers for those Medical Device Regulation question, I think.
Derk Arts: Great, thank you. And I think we can start with our last question, and then rest assured that we will answer the remaining questions offline. So let's say I have a large number of publications on my device, but the the literature is older than let's say 10 years, could you still use that data to justify sufficiency? The question says.
Nadine Leistner: Yeah, that depends on the quality of the literature and the methods those data were collected. If they they collected on a with a highly scientific and statistical method, that doesn't that doesn't influence any with the publication dates.
Derk Arts: Yeah, no, I exactly, so we just look for studies that are of high methodological quality, and should be should be fine. Okay, then I think we have room for one more, the last one that just came in. If can we use real-world data for PMCF purposes? When I hear real-world data, I think about EHR data, for example, so actual routinely collected data. And do we use quantitative quantitative or qualitative analysis on on those data? Do you have any experience with with an approach that involves real-world data?
Nadine Leistner: Not yet, I have to admit, no.
Derk Arts: Okay. So this remains this question remains unanswered. And so we started with an unanswered questions, and we ended with an unanswered question, which is kind of nice and symmetric, I like it. So Nadine, thank you so much for your excellent presentation and for answering the vast majority of all the questions, and there are more. I learned a lot of things, and I've attended a lot of conferences on the topic already, so I think we accomplished our goal of of having a very educational webinar. I hope the audience agrees, and with that, I would like to hand it over back to Ryan to to wrap up the webinar. Thank you everyone for attending.
Nadine Leistner: Thank you.
Ryan: Yes, and thank you both of you for the incredibly insightful presentation. However, as he said, we have reached the end of the webinar. Now, of course, if we couldn't attend to your questions, the team at Castor will follow up with you, or if you have further questions, you can direct them to the email address on your screen. Thank you everyone for participating in today's webinar. You will be receiving a follow-up email from Xtalks with access to the recorded archive for this event. A survey window will be popping up on your screen. Your participation is appreciated, and it will help us to improve our webinars.
Now I'm going to send you a link in the chat box, and you'll be able to view the recording of this event on that page, and you can share this link with your colleagues when they register for the recording here as well. So I encourage you to do that. Now please join us in thanking our speakers, Dr. Derk Arts and Dr. Nadine Leistner. We hope that you found the webinar informative. Have a great day, everyone. Thank you for coming.
Nadine Leistner: So, bye.
