Episode Transcript
Hello and welcome to TransPerfect LifeSci Talks. I'm Mark Wade. Today I'm delighted to talk to Paul O'Donohoe, who is Senior Director of Product for eCOA and COA and the Scientific Support for Medidata. Paul, welcome to TransPerfect LifeSci Talks.
Thank you kindly, Mark. Delighted to be here.
Two Irish guys talking, so the subtitles will be put on later [laugh] but anyway. Thanks for joining us. Can we jump right in? Because the challenges of eCOA and DCT - decentralized clinical trials, that's the watchword of the day. What are the challenges that you and I see today in adopting those studies?
Yeah, so it certainly is the watch-word, or dare I say, "buzz-phrase" of the day. Any conference you've been at in the last year or so has many panels on DCT and I do think there's kind of quite a bit of confusion that's maybe driving reluctance to really embrace so-called decentralized trials. I think people view it initially as an all-or-nothing thing that suddenly you're just going to have no site visits. Patients are going to be left on their own at home. We're only going to interact with them through SMS messages, for example. And that's very much not what DCTs are. In reality, DCTs are just a way of adjusting the study design so that more of the study activities can be done in a place that's convenient for the patient, whether that's their own home, a local clinic. But it's a way of shifting and reducing some of that burden of patients having to come in for site visits. And depending on the study you're running, that can be a very small aspect of the trial or you can decentralize large amounts of the trial. And, so, I really think we're moving towards a world of recognizing that DCTs are just a flavor of clinical trial. And we'll probably just go back to talking about clinical trials that have these decentralized aspects to them in the future.
You actually touched on two points. I want to unpack both of those, because you talk about – I'm going to talk about modalities because you're talking about how DCT has one aspect, and I love that idea by the way, that it's one arm of a study, and it should be seen that way. So, can we talk about capturing data on multiple modalities? I don't like using overusing that word, but multiple routes. Can we talk about that for a moment? Because I think you're doing some work on that, aren't you?
Yeah, absolutely. So the audience is aware, typically, or at least traditionally I should say, we've been capturing data from patients using a provision device. We give them a smartphone, they take it home, they carry it around with them beside their everyday phone. And that's what we've been doing for years. We're starting to see more and more uptake of so-called BYOD – "bring your own device" patients installing an app on their own device, submitting data that way. We're also seeing the rise of web-based data capture, so patients logging into a browser and submitting data there. But there's been a lot of reluctance within the industry, I think, to kind of embrace these different modalities. And there are some kind of hangovers from historical regulatory aspects around concerns of patients responding differently on these different modes. I think thankfully we now have plenty of evidence showing patients don't answer inconsistently. They answer the same across multiple different modes of data capture. We actually have an updated ISPOR task force paper coming in the next few months that hopefully puts this question to rest. But really I'm hoping that then allows us to introduce more flexibility in how we're thinking about this study design, allowing patients maybe to decide where they want to submit the data, where's most convenient for them to submit the data.
And that dovetails nicely into my next point that you actually brought up was the patient burden. We're putting, we're putting burden on patients. Look, that's table stakes, whether it's paper or digital, there's going to be, there's going to be burden. As to the level of burden that we're putting on patients, I think it's something we definitely need to talk about. If I digress for a second, if we look at protocols today, we are putting in many, many instruments to try and get key data points. One of my previous podcasts, I talked to an SME around instruments that are particularly fine tuned for particular oncology studies. So rather than the scatter gun, we have rifles. So can we talk about patient burden and the burden that DCT puts on patients?
Yeah, yeah. I think you're absolutely spot on. I think we have to start any of these conversations with a recognition that no matter how we're structuring the clinical trial, just engaging with the study in the first place is putting burden on the patient. So we have to start from there. We can't be naive and think we're suddenly gonna have a zero burden study. I do think one of the risks of technology is that people will use technological solutions just for the sake of using them and assume that that removes all burden and it absolutely does not. Obviously, technology can add significant burden to anyone who's struggled to get their laptop working or remember what their passwords were for the various websites that are out there. So we can't assume that technology is just gonna simplify the patient experience, but well-designed technology can simplify how a patient is engaging with their study. And that's very much how I envision the use of technology within trials. We're never gonna remove burden completely, but what we can do is improve their experience of engaging with that study and completing the various tasks that they are being asked to complete within the context of the trial, and just make it a more seamless way of providing data typically.
And, that brings us to another, another area of burden. We talk about patient burden. I mean, I'm always talking about patient burden, but there's actually a huge burden on sites as well. You know, sites have so many different technologies that they need to employ. I mean, by the time a kit arrives, they need to be put into an RTMS system, and then there's an eCOA element of that, and then there's EDC. And so that there's there all these different disparate type of technologies that the sites are overwhelmed. Sometimes I get told that from sites a lot, they're overwhelmed. So, genuinely what's the solution?
Great point. I think one of the other buzz-phrases we as an industry have been talking about for the last, what, 5-10 years, is patient-centricity. Absolutely right. We need to be patient-centric. We should continue to do so, I would just hope we're a bit more kind of meaningful, patient-centric. But I think we also need to start thinking about site centricity. I think one of the risks of this rise of technology in trials is that we're outsourcing the help desk basically to the site themselves. And these sites are typically completely overwhelmed. They have too much day-to-day work, nevermind the additional work that's been piled on them by participating in trials. And if we're not providing them the training they need to use these systems as well as making sure the systems just work well and are intuitive, I think it's a huge burden that we're placing on sites that really risks actually the success of the trial because we consistently see if the sites are against the technology, that's going to translate it onto the patients because they're not gonna want to train the patients up on using a technology.
It's simple human nature, isn't it? It really just simple human nature. Hundred percent. I got a lot from from sites where they're saying they use so many different... In a previous life, I got involved with sites deeply and, the challenge they had is that there were so many different disparate technologies that they have to remember the idiosyncrasies of each of those platforms. And it becomes a huge burden, nature of the beast. What happens? Things get... Not ignored, I wouldn't say ignored, but things get deprioritized
Completely. Yeah.
When they're trying to do their day-to-day job of just helping patients and making them, well, of course our clinical trials are gonna fall down their list of what they're focusing on. So I think there's a few different responses to that. I do think we need to spend time thinking about that site experience and again, wanna reiterate the patient experience a hundred percent. They're our ultimate end users. We spend a lot of time thinking about that and getting that right, but the site experience, they're also one of our end users and making sure that is a user-friendly experience for them. And I think one of the things we saw off the back of the pandemic where, I think we're rolling out a number of technologies in the middle of clinical trials, often in response to patients not being able to come in for site visits.
And, we could have and might have a whole separate conversation around the impact of that. But I think one of the things we saw with kind of all these disparate technologies being dropped on sites was it was a response to a specific situation, but it also created its own challenges in regards to sites trying to manage all these very unconnected systems. And so, obviously Medidata takes a very different approach of coming at it from kind of platform solution approach, where it's a single system that brings all the technologies you need within the trial, but there's pluses and minuses to both.
Sorry, for the audience. For the audience, because some people might not be aware of that. What do you mean by this platform? What does that mean?
Yeah, thank you for giving me the opportunity there. That, so Medidata is a provider of a platform solution for the life sciences. So that basically means we provide a single software solution that you can run your entire clinical trial on from beginning to end. And that includes all of the site-based technology. I'm on the patient cloud team, which is our patient-facing technology. And then there's obviously an aspect for the data analytic and site and, study management piece. So basically we have a single platform and you can do everything you need within that platform to run a clinical trial.
The flip side of that is these point solutions. So we have companies out there who do separate aspects of the trial sponsors, might work with a range of different companies for a single trial, and then try and bring that all together to have a single trial experience. And again, pluses and minuses to both of those solutions. But I think what we saw within the pandemic and kind of continue to see that playing out as we're kind of taking stock as we come out of the pandemic for certain definitions is maybe a recognition that just having all these multiple different point solutions creates a significant amount of challenge. Whereas having a kind of single platform solution does remove one of those, those burdens that we consistently see.
I agree. I mean, in that they both have different challenges. Certainly I think the holy grail for the single point solutions will be how they integrate in, let's say, a single sign on with other parts of the study technology. I mean, I get it. I'm not suggesting one is better than the other, but there's definitely plus and minuses. If the holy grail is a single sign on that one, technology can speak to the next and the next and the next, and the data can be collected in one or multiple large data pools. That's wonderful. So, I hear what you're saying.
I did wanna circle back on one point you made, Mark, if you don't mind me picking up on it. It was this really important point around seeing the huge increase in questionnaires within a study that you've seen, and I think I can certainly echo that as well, oncology particularly, but other therapeutic areas, this sense of maybe just throwing questionnaires at patients in the hope that something sticks. And I think that's really a source of burden for patients, but also sites who may be the ones who have to sit there and help administer the questionnaires. A source of burden that I don't think helps anyone to be honest. And it is actually an area where I think there's some really exciting stuff we can do with technology around being much more targeted in the kinds of things we're administering to patients kinds of questions.
We're asking them maybe even pairing that with some kind of wearable or sensor device. So removing this, tens if not hundreds of pages of questionnaires that a patient is doing, and instead getting a much more fine-grained, higher resolution insight into the patient experience. Maybe they have to complete these questions more regularly, maybe even daily, but if they're only doing five or ten of them as opposed to having to do a whole host of questionnaires every few weeks, I think there's some really interesting stuff that we'll start to see emerging within the industry over the next few years around shifting away from those kind of traditional big chunk of questionnaires, to much more kind of fine grained insightful dataset.
I totally agree. I totally agree with that. And again, we can have a whole separate conversation around this idea that a much more rifle view of questionnaires , would reduce the number of questionnaires we filled out and the frequency. And when you're talking about certain TAs, especially oncology, when do we execute those questionnaires is just before the patient has had chemo, or is it's just after? If the patient at home, they're so exhausted that the last thing they want to do is fill out a questionnaire. And if there's multiple questionnaires in that pack, then what does that mean? What does that mean to the data? So you're absolutely right. I totally agree with you. But that's a real- and I'm hoping that I get someone on the podcast, maybe a few people, and we talk about study design.
From the sponsor side, and we talk about study design and what are the drivers and what can be done. We had a very – pushing my podcast – but we had an interesting podcast recently where we talked about question banks and how you can pull questions from question banks and build your own validated instrument. And it meant that you had lesser, more rifled questions rather than these large legacy general wellbeing [ones]. And that was a fascinating conversation and, I would advise anyone to have a look at that. But it's definitely the whole idea of study design. I'm with you. Totally with you. Totally with you. I want to ask you one because again, I'm circling back to you now, so I want to ask – You touched it very briefly. We were talking about mixed modalities. You mentioned, and I'm not gonna lie to you, I've always been a huge advocate of WAP [Wireless Access Point]. And let's unpack that slightly because some people might not know the idea of WAP and what is the WAP and what is an app wrapper – can you talk to that?
Yeah. Also, I think referred to as HTML5. I actually heard you describe it as kind of bringing together the best of web with the best of app, which I think is a good description. It's an app shell that runs browser code within it. And so you get all of the flexibility that you see with any webpage that you interact with on your computer, which, is all about responsive design, making sure things fit neatly on the screen, that basically it's good user experience. So bringing all of that power, but then kind liberating it to a certain extent from the browser and having it in a more app based environment. So, as I referenced earlier, we have been using native apps within this space for a very long time now.
That's kind of the default. We're starting to see the rise of, of web-based, browser-based data capture. And we definitely haven't really seen much HTML5 driven technology. And, I think there's a couple of reasons for that. I think there's just the conservative nature of the industry. We're slow to embrace new technologies. I think there's always concerns about things like offline data capture, consistently get that question from sponsors. What if a patient is offline, how can they submit data? And obviously in the native app they can still submit data. I think there's still questions around the performance from a computer performance perspective. But yeah, it's an area, full transparency, that I haven't dug into in a huge amount just because we've been so focused on app and web.
Love to hear what you foresee, I've been a fan of this for many, many years. In fact, I spoke to you years ago about this. I'm talking about years ago. Maybe 2012/2014 –years ago. And the technology wasn't there.
Yeah.
Today we have incredibly robust wifi across western Europe and North America. Incredibly robust, incredibly fast. We have IoT now, we have 5G cell data. The technology wasn't there. But in the translation world, and I'm sure you have the exact same in eCOAs, I get a daily pushback from sponsors going "I want it done yesterday. Can you do it yesterday?" And there's a constant pressure on doing it faster. Forget about cheaper, just do it faster. The FPI dates are becoming tighter and tighter. But I do think that a WAP-based app wrapper, I call it an app wrapper. Cause it looks like an app. It smells like an app.
It behaves and walks like an app, but it's not. And I do think that that's gonna be the future because we can build them faster based on the study design. We can build them much faster. The translation piece is always gonna take time. It's very heavy lifting. It's incredibly complex. And there's only, I would argue that, because it's so specialized, one has to tread carefully, one can't expedite it to the point where you're compromising the quality. But, in your word, in the build world, when you're screen testing and all the rest, that is a bottleneck. And if we can get past that bottleneck with something like HTML5, if we can get past that bottleneck, I do think that there is, there's an argument to be made that we should be deploying that. But again, it goes back to the very first thing you said, mixed modalities.
Yeah.
Is it possible–I don't know. But is it possible to have an app wrapper, a native app, a web-based BYOD, and provision device? Is it possible to have all those modalities to capture the data from the patient and not introduce any kind of bias into the data? And I think that that's a conversation that is worthwhile having.
Absolutely, you bring up a host of really interesting points. At the risk of tangenting slightly, we consistently hear and receive pressure around timelines and cost and completely, understandably at the same time, at the risk of oversimplifying things, if this is your pivotal trial, the success of your new compound rests on this study. I think this really single-minded push towards shortening timelines is just setting oneself up for the risk of failure down the road. And again, we should be doing everything we can to shorten timelines, but, I just have my own frustrations around that. Sometimes single-minded focus on the timelines and budget piece when within the full life cycle of a compound.
Absolutely. No, that's a very fair pushback, a very fair pushback because there are different personalities involved and there are different personalities in the study teams that want FPI these days. So, the strategy that we employ, is that we stagger the countries. So we say, where's your first FPI? And they'll give us a list of countries. So we prioritize those. And, that kind of dovetails into what I'm saying, if those countries are a good candidate for a WAP-based wrapper, then we can deploy them faster and then we can work on the native app on the countries where we do need the native app. Does that make any sense at all?
No, completely.
I just think it's, like you said, it's more holistic view. It's, it's an overall view of the study. But I think everything has a place, I think everything has a place, and this all or nothing, this all or nothing DCT thing is not a great idea. In many ways.
It's not, I think, healthy for the innovation that we need within the industry. And I think, to get back to your original question around kind of mixing these modes together, and I think it neatly circles back to one of the first terms we brought it up around flexibility. I think that's absolutely the future of where we're going within clinical research. And that sits uneasily within what is ultimately an experiment. We're trying to run an experiment. And to successfully run an experiment, you need to try and minimize the variability and the variables that are impacting the outcomes. So completely recognize that flexibility can be a scary term within that context. But I think considering the sheer complexity of these studies, global trials, different patient populations, different realities on the ground in regards to technology, access, different needs for patients, across your study, I think allowing patients more choice as well as giving the study teams more choice around the technology being used is only gonna have positive outcomes.
And as I referenced, we have a huge amount of evidence now showing that patients do answer consistently across different modes, whether it's paper, web, app, I suspect these HTML5 solutions as well, and so I think that's going away as a concern. And that's really, I hope, freeing us up to decide what is the best solution for a given study. And then even within that study, what's the best solution, the best solution for a given country, or even the best solution for an individual patient. Let's give them flexibility to decide.
You preempted my last question because I was gonna ask you about like the barriers and that you're absolutely right. I mean, the barriers were this lack of urgency to pick up these new technology strategies.-I get it. I totally get it. One last thing I was hoping that you might maybe hat tip toward, I don't know, but the regulators. The regulators, I think you would agree with me, have been fantastic in the pandemic. Because they allowed us to expedite certain- so I was just curious your point of view on that too.
A hundred percent. Regulators move slowly and that's by nature. I think their job is to be the break and we push against that. I think they should be commended for how quickly they responded, but also how pragmatic their response was. You know, my maybe unfair kind of summation of what they said was, do what you need to do to ensure patient safety, obviously, first of all, but then ensure the integrity of the trial. And that really opened the door for us to roll out many of these technologies that we've had for a long time. But now we were rolling them out in a single trial, for example, and really provided us the evidence that these do work as, as solutions. So absolutely they should be commended for that. I think at the moment, as an industry, we're in a bit of a kind of taking stock from those last few years learning some of the lessons that we learned there.
But I don't anticipate us going backwards. I think the regulators recognize the importance of technology in the clinical trial space as a way of enabling patients to have better experience as a way of getting better insight into that patient experience. And, they're asking people to come and talk to them about these things. They recognize they're not the experts in this, and so they want the experts to come and have those conversations. So I do think, as much as we might moan about the regulators holding us back, I do think actually they should be commended in this area
Here. I agree with everything you're saying. I wanna ask you my last question, which I ask all my guests: If you had a magic wand, what would you do to change, or not just to change, but what would you do to expedite the adoption of DCT today?
I think I would loop back to kind of where we started with DCT maybe being a misunderstood phrase. And so I think I might use my magic wand to strike the term DCT from the lexicon and just go back about clinical trials basically. And so we would talk about clinical trials and what appropriate solution and use of technology within that clinical trial. And so we would no longer worry about clinical trials and then this mythical thing called DCTs. We'd just be talking about clinical trials and what's the correct way to use technology within for that specific study. And so I think we will end up there, but I would use this magic wand to accelerate that.
Well, we'll see what happens. Thank you, Paul. I appreciate it. Today I was talking to Paula O'Donohoe, who is Senior Director of Product for eCOA and the Scientific Support of Medidata. I'm Mark Wade of TransPerfect. Please join us for our next episode of TransPerfect LifeSci Talks.
Many thanks, Mark. Thank you.
