×

Download Premium Content

First Name
Last Name
Company
Country (Citizenship)
State
Opt-In to Future Emails
Thank you!
Error - something went wrong!

Panel: Wearables and Sensors

July 22, 2019

Transcript

MODERATOR

We’re going to switch gears here a little bit into a panel discussion. We’re going to try and keep this relatively streamlined and shave some time off of the schedule so we can get done on time. But we’re going to actually stay with the theme of connected devices and wearables.

You’ve met Jen and Bill. I want to quickly introduce Mark Wade from Transperfect. Mark is a published expert of eCOA, including two e-books, and is a regular talking head for major trade publications for both best practice in clinical translation and cultural adaptation for COA instruments. So thanks, Mark, for joining the panel.

I’m going to start off just talking, maybe keeping with the theme of what Jen introduced. Where do we think are the biggest opportunities in wearables right now? Maybe Mark, we’ll start with you. We haven’t heard from you yet.

MARK WADE

Thank you. I think, if we’re talking about sensors again, I think the biggest opportunity is in the low hanging fruit, the very simple measurements. If you look at eCOA, and look at the penetration of eCOA, we’re talking about the last ten years at least. And not every single sponsor is using eCOA. So the fact that it’s that slow advancement, I think it’s going to be definitely very simple [unclear 01:24]-type things where we’re going to see measures being used in wearables.

MODERATOR

I think we’ll move forward just to keep the conversation going. Bill, since you have the microphone, what are we hearing from regulators on this topic?

BILL BYROM

I’m looking at Nikunj because I was going to cite the meeting that we had back in October 2017, and I’m just wondering, you would have been FDA then,  you were there, were you? Right, okay so you can see if I’m being honest about how I interpret what the FDA are thinking. Thank you, excellent. As part of the ePRO Consortium, we did a piece of work about the same time as the CTTI work actually on a similar topic in terms of what evidence would you need to support the selection of a particular sensor for a clinical trial and then also to support the use of an endpoint you might derive from the data. And we wrote that up and then published it in Value in Health, so that publication is available. We presented it to FDA as part of a critical path innovation meeting. And there were a couple of other topics, but it really was about wearables and sensors. And my take-home from the meeting was, they liked the recommendations that were in the paper, and they appeared to be favorable in terms of their opinions towards the use of these types of approaches. I think the concern was, very much to what Jen had said at the beginning is, we’re not going to use a wearable just because it’s a cool thing to do, we want to make sure that you start with the concepts of interest of your trial, and if it turns out that actually using a wearable or a mobile sensor is the best way to measure that, then by all means go ahead and use that technology. And yes, maybe the sort of thing that CTTI and ePRO Consortium are saying with regards to the evidence to support that would make some sense. So my take-home is, the regulators are pretty positive.

And actually just moving on from that a little bit, the patient-focused drug development meetings which were in October which I attended, again, wearables came up a lot in that and again there was a lot of discussion with the regulator and others about the use of wearables. And it again appears to be there is an optimism about it. I don’t think there’s any reluctance to use it so long as we implement these things in the right way, just as we would for any endpoint in any measurement methodology that we use in a clinical trial.

 

JEN GOLDSACK

I’ll add to that briefly and say that, for folks who don’t know, the Clinical Trials Transformation Initiative, CTTI, is a public-private partnership. It was cofounded by FDA and Duke University and it’s got industry members, it’s a large multi-stakeholder organization. And the reason that they have a whole program on mobile technologies is because FDA came to CTTI and said, you’re a multi-stakeholder organization, your focus is quality and efficiency of trials, can you take on this work on this project. And so that work was initiated by FDA, it was Leonard Sacks and his team with some good support from others who came in and asked CTTI to take on and champion that work to really understand it, not just for industry, so that FDA could better understand it. And I’m only repeating what I’ve heard, but the language that I’ve often heard from FDA colleagues is, we want to learn together, this is a new frontier, it’s going to take a lot of different experts. So I think that that’s really important.

 

[05:13]

And when we did launch some of the CTTI work last June at a full-day meeting at FDA, we also had Francesca Coretta from EMA was part of the day and in prepping for that meeting and in reviewing all of the materials, it became apparent during our conversations and she said it at the meeting, that there’s really good alignment at the meeting between certainly—I can’t say FDA, I’m not FDA, it’s nothing to do with me—but between what was in the CTTI recommendations and how EMA are thinking. So again it’s really nice to see alignment too between different regulatory agencies.

MODERATOR

Mark, anything to add?

MARK WADE

No, I was actually—I have a question. And it was to do with your presentation. Sensors, I love by the way that you broke into mechanical and clinical, I like that idea. My question is, there are certain, as we move forward, do we need to get 510(k) licensing for these devices? Does that make sense? The fact that we’re now going to use these to capture primary data, what do we need to do with these devices?

JEN GOLDSACK

Again, as part of the working group that came up with the meat behind the framework that I presented, that was a really detailed discussion we had many times. And our conclusion was no, it doesn’t have to be a medical device. And if you think about what the definition of a medical device is, it’s to diagnose and treat. Now there are going to be, I would have thought that if you want to start putting sensor rays in someone’s brain, the safety profile ought to be checked, it ought to be a 510(k) cleared device. But I think it’s much more about being fit for purpose. And I actually think that this idea of a medical device being better, especially something as simple as an accelerometer can be a bit of a misnomer. That clearance, a cleared device—and I stumble over the language sometimes, if I don’t have it exactly right forgive me—it tells you a lot about the safety profile but it’s a cleared device for a very specific application. If you then take it and start using it as a measurement tool it doesn’t tell you squat about how good it is as a measurement tool for this new indication. So I think in some cases it’s a bit of a red herring in fact and that it’s certainly not the case that if you’re using it as a measurement tool for a trial outcome and not to inform a treatment decision, I don’t think it has to be. It should be fit for purpose, and there should be a lot of evidence that it’s fit for purpose.

MARK WADE

Yeah, I mean I think there’s a whole thing around consumer and professional grade. And I’d be concerned about things like, if it’s a longer study, things like if the software is updated, is this updatable. Is the battery good enough. Simple things like that need to be tested. Like consumer devices are great, but I would not really like to see consumer type devices in a trial environment, really.

BILL BYROM

Yes, and I think those examples you brought up are really good ones, and that’s probably one of the reasons why we might not be using consumer devices. But it would fail on the assessment criteria that we lay out, how we choose a device that is appropriate. Well, I want to make sure I’ve got full control of software updates, I want to make sure that I’ve got full control over the source data, whatever cloud system it’s living in, and these sorts of things, which we might feel if we look at a Fitbit or something else, we don’t have that control and therefore we eliminate them. And I agree, I don’t think these things need to be market cleared. However, we might find it a little easier to distribute them in global clinical trials if they are market approvable in different territories that the study is running in. And Juanita is sitting at the back there, and she’s involved in our diabetes devices business. And okay, they are medical devices and you can get them into countries where you don’t have a marketing approval on kind of what we say is a clinical trial waiver. But the idea is that, because you’re using them in a very controlled environment of a clinical trial, and you’re not selling direct to consumers, then you can actually get them into these countries. But that’s not clear cut and it’s not always going to be the case. So it would be easier if they all had market approvals in the different markets that we were wanting to conduct our studies in. But really that isn’t practical.

[10:00]

JEN GOLDSACK

Can I ask a devil’s advocate question, which is the nature of the technologies, which may not have device clearances, would you face challenges going into markets? Would those markets care if it was just an accelerometer?

BILL BYROM

I don’t know is the answer. I don’t think we know yet. But I think in principle I think we’re all in agreement, we don’t need to select something that has a market approval or a 510(k) or a CE mark. But it has to meet the criteria that we define in terms of what makes it an adequate and suitable device, and that we’ve looked at the liability and the measurements that it produces and all that sort of stuff, and if we can tick all of those boxes then we should be comfortable we can support and defend its use to create endpoints for clinical trials. And that’s really the bottom line, I think.

MODERATOR

That’s great. This is all building up to, we have regulatory receptiveness, we have a framework for decision making, what are the next steps? What’s the next frontier in order to move on from exploratory endpoints?

JEN GOLDSACK

I would say that it’s two pronged. I think that not—again I go back to this idea of not wanting to overreach and fail. And I think Mark had a really thoughtful response to your question about where is the best use. I’d add a non-therapeutic or measurement bent to that in terms of application and I go back to this idea of perhaps internal decision making, adding it to the mix as you make go/no-go decisions at the end of Phase II because you’ve included this in Phase II. It’s low risk, because the data may be used for different purposes. But you start to build a body of evidence and that’s what’s critical. You can’t do any of the steps, right, you guys know this certainly better than I do because you’re measurement experts. You need a body of data, you need to see how these measures behave in your patient population of interest before you’ll be brave enough and frankly before you’ll have the evidence to use them in a different way. So I think, having the courage to use them in those sorts of environments and start to build the body of evidence is important.

The other drum that I would bang is, I see a lot of what feels like endless sort of pilots. You do a 30-person pilot, there’s really great data, there was good compliance, oh my goodness, who could believe it. And then you move on to the next pilot. And I don’t see any sort of collaboration between—I’ll continue to use the Duchenne’s example—the six or seven pharmaceutical companies that have R&D pipelines in these measures coming together and saying, you know, our IP isn’t in a measure, our IP is in the molecule that’s going to save these kids. So let’s come together, let’s throw our hats in the ring, let’s build this measure together. We’ll understand it much more quickly, we’ll get much more quickly to regulatory acceptance because we’re using one measure, and then we’ll take it back and we’ll do what we do best, which is develop drugs. So I also think, and if anyone has the magic wand to make that happen, wave it, but I think that’s really important too, at least among pharma companies, realizing that the IP for them at least isn’t in the measures and sort of holding them close. I’m not sure there’s too much value in that.

MODERATOR

Maybe Mark?

MARK WADE

I think, I agree with that totally. The body of evidence, and I am going to compliment you and Willie Muehlhausen, the body of evidence will start building up. And then it will take someone like these two guys, who knows, who will actually collate and analyze all that data and produce wonderful research which will say, look, we’ve collated all this data and we actually now know it works. We can prove the point. I think that will make a difference. But that’s going to take time. It was 2004 when the ISPOR guidelines came out, and it was 2009 when it was finalized. It was 2013 when it was ratified. So let’s not hold our breath.

MODERATOR

No comment here?

BILL BYROM

It’s all been said, it really has. I agree with it.

MODERATOR

Cool, okay.

[END AT 14:27]

 

Previous Video
Keynote:  T.J. Sharpe
Keynote: T.J. Sharpe

T.J. Sharpe, patient advocate, keynote address at eCOA Forum.

Next Video
Panel:  Role of Electronic in Patient-Focused Drug Development
Panel: Role of Electronic in Patient-Focused Drug Development

T.J. Sharpe, Nikunj Patel, and Ashley Slagle discuss patient-focused drug development.

×

First Name
Last Name
Company
Country (Citizenship)
State
Opt-In to Future Emails
Our team will be in contact shortly
Error - something went wrong!