Medidata NEXT London 2026 LIVE
At its heart, Medidata NEXT is a conversation where leading experts in clinical research come together in open dialogue. Representing the biggest names in healthcare and life sciences from around the world, they explore industry challenges and share their excitement for what the future holds.
That made NEXT Innovation Day London the perfect venue for our latest live edition of from Dreamers to Disruptors. Recorded straight from the show floor, Medidata leaders and a broad range of guests break down the power and incredible potential of AI across the clinical trial process, combining diverse perspectives into a united vision for the next era of healthcare technology.
Session 1 Speakers:
- Edward Ford | Vice President, Client Engagement, Medidata
- Anthony Costello | CEO, Medidata
- Nitish Mittal | Partner, Everest Group
- Krzysztof Wróblewski | Head of Global Clinical Solutions Technologies, Senior Director, AstraZeneca
Session 2 Speakers:
- Aryana Hosseinkhani | Vice President, Head of Product Marketing, Medidata
- Lisa Moneymaker | Chief Strategy Officer, Medidata
- Chris Standley | Vice President, Vendor Governance, Worldwide Clinical Trials
Session 3 Speakers:
- Wayne Walker | GM, Head of Data Experience, Medidata
- Robyn Orie | Sr. Director, Data Innovation, Caidya
- Cassie Kendrew | Chief Operating Officer, EMS Healthcare
- Amardeep Heer | Medidata Site Insights Board Member | Principal Investigator, Lakeside Healthcare
How are bold ideas born, and which ones survive to eventually shake up the status quo?
We'll hear straight from our industry's greatest visionaries, who are making waves,
and learn how they turned their dreams into disruptive reality. This is from Dreamers to Disruptors, a podcast powered by Medidata.
Welcome to London Medidata NEXT Innovation Day, we are just mere steps from Westminster Abbey here at the Queen Elizabeth Center,
and this is actually a culmination of our next Innovation Day series. We kick this off in New York.
We've been to Paris, we've been to Milan, we've been to Madrid, San Francisco, and we're excited to be here in London with this illustrious panel.
So excited to have you all here. The energy here in the room is amazing.
Folks are passing through, milling around, having great conversations, networking with this great community,
and the a lot of that buzz is centered around AI and how AI is going to really impact and drive the future of this industry, the future of clinical research.
And I'm really excited to have this discussion with you. Here we have Nitish Mittal from Everest Group. We have, we have Krzysztof,
...
go for it,
Krzysztof Wróblewski.
There you go
from AstraZeneca. That was good. And, of course, we have the Medidata CEO, Anthony Costello, and we're just.. we actually just stepped off of the keynote stage,
and so I really want to start and get just kind of initial reactions from each of you as to what you heard on the opening keynote this morning.
And, Nitish, I'll start with you. What was your reaction to what you heard?
Yeah, I think the keynote Ed was really interesting because it spoke about the power of the partnership for me.
What really stood out beyond the platform, beyond all the great announcements, was the power of the partnership that was the digital thread of the connective tissue
in a lot of the case studies, a lot of the references that we saw today, and to me that is what will take to make AI real move from pilot to production.
Krzys?
Great energy, amazing keynote, Anthony. It was great to see the productware advancements,
especially taking into consideration the whole transformation from the protocol throughout the system and how the data can cause very intriguing presentations, interesting product concepts.
Anthony, how'd you feel up there, man?
I felt great today. Felt really good. The energy in the room is palpable. Yeah, and I think it's because of what you guys are mentioning.
You know, we are focused on partnership. It takes a, it takes a village to build these kind of AI technologies, and then to get them launched and used at scale,
and that's really, that's really what we're trying to do with our customer base, and you can feel the energy here, as, as all the other NEXT conferences that we're doing this year,
there's a lot of excitement around the adoption of AI, and our customers are looking for ways to change the way that they work with these kinds of tools, and you can definitely feel it here today.
Yeah, I mean, I think the desire for change is rooted in this idea that we touched on on the stage.
A lot of these challenges that the industry has faced, they're not new, it's not like it's stuff that's popped up in the last 18 months. This is stuff that we've been talking about for years.
I mean, the high cost of running a trial, the complexity, and trials, as we know, are only getting more complex, and the need, and the for to getting patients, finding the right patients, and then keeping them in the trial.
These are all just a snapshot of some of the longstanding challenges that the industry has faced, and as we've pushed towards new innovative solutions to solving those challenges,
I think AI and bringing the capabilities that are in resonant in this new innovative technology will help us tackle that.
Part and parcel with that is the reality of bringing regulators along in that journey as well, and so maybe we start the conversation there.
How, what do we, what do we feel like is the reaction that we're getting from regulators as we're trying to interact some of these really innovative approaches in our trials?
Are they willing to kind of accept some of these new things, and what some of the hurdles that we might face? Nitish, I'll turn it over to you, maybe to start. Yeah,
...
I think that's a very critical question at this point in time, because I think what we need more than anything else is the trust infrastructure to make AI work, and a lot of that will come from regulators, but industry participation as well.
I feel with regulation, what's happening is I almost think there is a great convergence happening in the industry, where you know data, AI, regulatory actions, and regulatory stakeholders are all in the same room, having the same dialog.
There doesn't seem to be a big lag in terms of actions and ideas to make it work. So, I'm seeing a lot of rapid dialog, rapid prototyping.
If you take the UK, for instance, you know what the MHRA is doing with the AI Airlock initiative is really interesting, so it's really moving fast.
But show me your homework. So there's a lot of active collaboration, which I think will be crucial to building this trust infrastructure.
Yeah, yeah. Maybe Anthony, I know one of the things that links up with that is this idea of AI as a black box, and a lot of times there's a perception of, oh yeah, we prompt the AI, the AI spits out some things,
and we were talking about consumer-grade AI, the worry about hallucinations, and can I trust this data, maybe react to what Nitish said,
and position how Medidata's AI sits a little bit differently from that, is we can actually show the work as we're giving some of these recommendations. Yeah, I mean,
...
I think that's the most important part of how we've designed Dot and our AI program. It is open, it's transparent about how it's doing the work, what's the data source, what am I presenting,
and it really, it does operate as a co-pilot, so it is making suggestions and recommendations, and, and you know, we like to think of it as steering users, but there's still a human in the loop.
There's a way to validate what it is that Dot is telling you, and there's a way to accept it, or to modify it, and ask maybe a more particular question.
The other thing that's really special about the Medidata AI program is we've trained our agents on probably the cleanest data set in the universe, right, because we're talking about historical clinical trials data.
These operational data and clinical data are perfectly queried and cleaned and perfected and regulatory ready, so when you train AI on a data set that's that clean, it doesn't mean that we don't still have to go through heavy validation to make sure Dot's working,
and we know many of our customers are heavily involved in the validation of these AI tools as they adopt them, but the training ground for our AI is very different than a lot of AI out there being trained on real world data
or data that aren't quite as pristine as clinical research data, so we we we really think that's a differentiator and something that will help customers adopt these technologies faster.
Adoption is the other maybe term of the day as a sub-bullet around around AI and Krzys, this is where I want to kind of get your perspective.
You're at a large organization, there's probably a large spread of on that spectrum of adoption. I know kind of where you are on that spectrum of let's test these new technologies and figure out a way to activate them.
I'm sure there are folks, maybe not as excited within your org to maybe do some of that. Talk a little bit about what you're seeing around that adoption.
How are you making the case for value on AI within your organization, and bringing folks on that journey?
...
I would like to start, actually, with the previous perspective when it comes to the validation and collaboration with regulators, I fully view the continuity that it takes a village,
and I think that this perspective of collaborating between sponsors, technology providers and regulators is so super crucial. This to your point, we see this with openness in regulators, but it's not yet exactly committed to your point.
When it comes to the adoption, I think that what I observe across the industry when I talk with my colleagues, as well, there you go, is the perspective that everybody right now is excited about AI.
A lot of energy and sponsor work went into enabling our employees and stakeholders with AI skills, with AI capabilities. Right now, it's all about transforming this energy, transforming this level of knowledge into real-life cases.
What was presented today during the session about one of the challenges is a real return of investment. How to make those use cases work, how to select those that are very impactful for us,
not only the ones that are compliant and can help with our clinical process, but also the ones that really make sense, that we are not burning those tokens only for the sake of AI magical transformation.
Okay, is there any, like, is there a particular use case that jumps out in your mind? You know, we talked about this idea of kind of these four areas.
You can deploy AI, whether it's simulating your study before you even dose a patient, starting up that study, collecting and analyzing your data.
These various use, is there any one particular area that jumps out to you as ripe for disruption by using and introducing AI, and one that actually will produce value?
...
So for sure starting an automation with a digital protocol, and consequences of the digital protocol – what we saw today in the presentation, and your products, this is pretty interesting.
I think that this is something that could have a positive impact in the overall value stream. Second, digital twinning of the study simulations are a great area,
but again, for me, it's balancing out those technological advancements versus the reason versus costs of it. Today at the presentation I here had a very intriguing part of continuously running a simulation that's amazing,
it's like a life and insights on our study, but the question is, what's the what's the cost, and and and for me it's going to be always this balance of technological advancements.
How we can make our standards better, faster, safer for patients, and on the other hand, what is the price of?
Yeah, I mean, stay tuned, right? Because part of what we're going to have up on the stage is diving deeper and really bringing in examples where we're seeing folks have that exact conversation within their organization,
and when you think about, you know, generally speaking, right, the cost of an amendment, like how much does it cost every time we amend that study, that cost is significant, especially when you talk about a phase three trial,
so that cost benefit analysis and that calculation, the numbers can start to look very appealing with a very kind of low bar for entry, right? If you're impacting, if you impact just one amendment, we can almost pay for this technology, and so why not try it?
Nitish, I want to bring you in on this conversation around specific use cases. Are there things that from an industry perspective that might be the places that the industry maybe is moving to more quickly across this continuum?
Is it more on the kind of data management side? Are you seeing it in the operations side? What's kind of your perspective.
Absolutely, I think you know that's the question of the day to press this point around ROI and value, and what we're seeing is earlier with maybe generative AI, people who are taking a more of a spray and pray approach.
Now with agentic, they really want to be coned in on which are those use cases where there's disproportionate economic patient safety value versus investment.
We're seeing two areas emerge. One, you know, it sounds like a no-brainer, but data cleaning, you know, to Anthony's earlier point around having the most pristine clean data sets,
but data cleaning is also now getting into things like ontology, semantics, and it's a living and breathing exercise to make sure both your own data and synthetic data is in one place, and you're able to use that to power your context.
In my opinion, a lot of the large language models, you know, and other things might become more utilities in a few years. So, what separates your AI initiative versus your competitor? It's your context and your data.
So, there's a bunch of initiatives around that to make this a more living and breathing exercise. And second to your point around protocol amendments, I think there are a bunch of initiatives around this which can help create a self-funding loop for AI,
and I think to your point around, you know, if you save one protocol amendment, you can create a business case for funding the AI transformation, so it's finding those low hanging fruit, starting at protocol amendments, there are other things around synthetic controls.
So, I think there's a bunch of unused value, which is now becoming possible to address, because the technology is putting a spotlight on it.
I think the universe agrees with what both of you are saying, because in the midst of this conversation, the sun has come out here in London.
No small feat. We're getting the full experience of London here today. It’s the tube strike. It was raining, but now the sun has come out. When we started talking about AI, it must be a sign.
This is the crystal ball segment of the of the podcast, so I want you to think about a year in advance. We're back here next year for London NEXT, what is the big change, innovation, progress that you see that we'll be talking about next year?
If we, if we time hopped?
Since nobody's starting, I'll go first, because I'm afraid everybody else will say my point, and I won't have anything left.
But more seriously, my hope a year down, and who knows, by the way, right, even as an analyst and an advisor, I keep saying with the pace of velocity of, or the velocity of change, you know what you know on Monday is outdated by Thursday.
So having said that, I will still look at my crystal ball and say my hope is that agentic AI, and you know, maybe the next buzzword around the corner will stop being a buzzword and become more of a plumbing board, more of a default feature in our workflows.
That to me is the, is the hope, and that means, you know, faster protocol amendments, authoring, how we find learn less busy work and more productive work for people in ClinOps, so that to me is the hope where agentic AI becomes embedded,
and that's why I love some of the framing around Dot, and that steering. I love the framing around the steering, which helps us do more productive work, not just busy work.
I love it. Plumbing. Okay. Thoughts. Crystal ball, where are we gonna be talking about in 12 months?
You look terrified when asking the question. Okay, for me, in a year's time, I think that we're gonna evaluate some of those business cases, and we're gonna have a reality check.
What plays out, what makes sense? I hope that we're gonna utilize this digital protocol impact on the whole technological landscape.
How fast we can automate, how fast we can get prepared for the study, and utilize those operational efficiencies to your point.
My hope is we're gonna be able to translate some of those technological advancements to patient benefit to our last question,
were you asking about the things that I saw that were interesting, is to see that we are trying to address a patient experience, and here AI with this personalization of their, of their lives,
of how they are going through the sit pass, ultimately, and connecting this fitness perspective, going through from curing into a help management, ultimately, this is something where I see an opportunity. Yep, obviously very hot.
That's great. That's great. We have a tradition on the podcast, and Anthony, I'm gonna start with you, because you're familiar with our tradition on the podcast, which is we close each episode with a dream as to what we want to disrupt next.
Okay, this is from Dreamers to Disruptors, is the name of the podcast. So, Anthony, what do you dream to disrupt next?
Yeah, well, I'm gonna, I'm gonna merge your last two questions together.
Sounds good to me.
A little bit. So, you know, if we go back a year, I think we were all just starting to get adjusted to the idea that AI is really here,
and at Medidata we spent the last couple of years launching all of these things that you saw on stage today into real production, so that they're actually accessible to our customers.
My hope for the next year, and I'll make it a dream, is that we move through this phase of, I guess, I'll call it pilot.
Pilot purgatory, I think I've called it.
Pilot purgatory testing, evaluating, looking for the ROI, which, of course, is important, and at next London, a year from now, we have more customers describing actual use case examples of where they're seeing the savings.
It's hard to build these tools. It's easy for us to get on stage and talk about all the great things that you could do, but the real proof, I think, for the next year is,
can we get customers and partners that are showing the evidence that these things change their businesses and ultimately accelerate the clinical trials, so that's my hope, and we'll see how it goes in the next year.
All right, man, I think you all have given us a lot to think about in terms of the ways that we can deploy AI in a meaningful way, the context that we're seeing in the market,
and what the promise really for AI is to really embed it into the workflows, not just look at it as a cool add-on, but to, as you put it, Krzys, build it into the plumbing of how organizations build their clinical research kind of team and approach,
and I think that will help us serve the mission that certainly Medidata has in all the organizations here around improving human health and making sure that people have access to care to keep them healthy,
and when they are sick, to get them back to a state of health that can help them enjoy their lives and the special moments that they have with patients. Anthony, Krzys, Nitish, thank you so much for being here with us today.
This is just the first in a series of live podcasts we're doing here live from NEXT London. We'll have more for that. We have a great vibe in the audience, we have a lot of great booths here,
so fully you all get a chance to enjoy the rest of the time here and interact with some more folks again. Thank you for being here with us, and we're signing off here from our first live podcast session from Dreamers with Disruptors. NEXT London.
...
Hi everyone, welcome to our live podcast from NEXT. We actually did the first one of these from our flagship event, NEXT New York, back in March.
We're so excited to have it on the road, as you can see from our beautiful backdrop, that's not fake, that's Westminster Abbey. We are here in London, so we're at NEXT London right now.
I am joined by our own Chief Strategy Officer, Lisa Moneymaker. Great to be here, and we're also thrilled to have one of our greatest and latest partners join us,
Chris Standley, the VP of Data Governance from Worldwide Clinical Trials. Thank you for joining us.
Yeah, delighted to be here.
So, Chris, why Medidata? Tell us why you guys went all in with Medidata as your vendor of choice.
Well, we've known we've worked with Medidata off and on for a long time, so we're well aware of Medidata to partner, and we've had a great partnership with Medidata, but I think the organization Worldwide's very much reached an inflection point,
and my job Worldwide is to try and make sure that our teams, our operational teams, have all the best tools to deliver studies, delight the customers, delight their shareholders, but also particularly the patients and spouses,
and to do that, I think we're an inflection point with both the growth of Worldwide, but also with tools like here, and I think our take on this is that to get best value out of this we need two things, we need a robust partner,
we need a partner who that we can rely on, a mature party we can rely on with robust processes, but we also need an end-to-end, a more end-to-end holistic integrated solution,
because the value in implementing AI stems from having all of the data in one place, having it all mastered, and having it all available consistently to the AI tools, which then can then operate across the whole platform in a coordinated fashion.
So when we sat down and looked at that as a key objective for the company, or, and you know, our IT teams wanted to try and empower our staff with the best tools, they came to come to people like me to try and look across the vendor landscape,
look across the partners we've worked with, and we felt Medidata was by far and away the best fit for that.
Great. You just gave me some great sound bites. Lisa, on the heels of that, how do you see this partnership really impacting our strategy?
It's been a critical partnership, so I think the industry is at a really big inflection point, as it often is, but we hit these cruxes, and Chris touched on those, the demands of AI.
What it means for companies to think about insourcing versus outsourcing, and what it means to be able to make value out of tools like AI in an evolving data ecosystem means there's more opportunity than ever to think about transformative deals like this,
where we can help a company who is looking to become the place to be for trials to be conducted, the CRO of choice, and so what does that mean for them to bring technology to that, to think about what has traditionally been a manpower-based ecosystem,
and how can we use AI to bring technology to bear with that? So it's a perfect opportunity for us to get to showcase a lot of the work that we've been doing with a partner who is as excited for the kind of change that that brings
as we are to really be willing to rethink business processes. To not just say I'm going to bring tech but I'm going to do things the same way that I had before and so it lets us kind of unleash our inner innovator in a really exceptional way,
because we've got a partner who's willing to use it in the way that we have been thinking about it, as opposed to in this kind of piecemeal.
So, I guess Chris, back to you. Partnership, relationship, it's ever a one-sided string, as Lisa was saying, bringing that innovation to you guys. What do you need from us?
So, we really see this as a partnership where we're trying to dissolve the boundaries between the two organizations. We want the two organizations to work together deeply in step, and from our perspective, we want to bring our expertise.
Worldwide is a very sophisticated therapeutically focused CRO with a lot of expertise internally on running clinical trials and serving a wide variety of customers, particularly small, medium-sized biotechs,
and so we want to sit down with Medidata and look at how we can bring that knowledge to bear using harnessing the technology and the expertise that Medidata has, harnessing the vision that Medidata brings, and you know, particularly around things like AI, the capabilities, but also the rigor.
You know, I was sitting at, I spent some time, was not going to spend some time yesterday in the partner advisory board, and one of the things that really stood out for me there, for example, and there's an example of why you know we value this partnership so much,
is the rigor you're bringing to data privacy, security, validation, and so on, and so we are confident that we can adopt this technology, we can use it to drive the business forward, we can use it to bring medicines to market faster, but we can, and we can do that
safe in the knowledge that the technology we use it to do that is robust and has, you know, good safeguards around it, and we don't need to worry about the way it's implemented, but it comes to, you know, working issues.
Absolutely, yeah. And Aryana, I think it's a, it's a critical thing that we think about. We didn't just arrive in the clinical development game last year, and say we've got some great AI.
We've been focused on data privacy and patient privacy and data security, and what it means to uphold this data in the degree of trust that it needs, and to have systems that are validated right out of the gate.
And so we take that and we bring it to our philosophy around AI, so that it's not simply, you know, what could be done, but what can be done in a controlled way with the right trust, and that goes to how we develop the models, how we test models, the data access that we give them to.
We spent time at the advisory board yesterday, really talking about this in a more revealing way, I think, than we have in the past, and the response we got was we actually Medidata. It was a learning point for us.
We need to be more upfront with that with customers anyway, because we have spent so much time with it, and that development of trust is what makes a trusting partnership.
And so it allows us to do the sorts of development that we want to do, and have a trusting partner who says, I can use this.
You both mentioned AI. I mean, it's the hot topic. As a marketer, any session that has the word AI in it is the most attended session.
So, staying on the heels of that being a hot topic, where do you see the future being within clinical trials within Worldwide and the adoption of AI?
Well, when I said at the outset that it's an inflection point that that's that's kind of a jaded and overused term, but I really do think that within the industry we're at the point now where it's gained starting to gain traction and starting to have a meaningful impact, and we'd see it across the board.
It's funny when you sit down internally with our operation teams, and you have conversations with them about pain points and challenges, the use cases for this stuff just come flooding out.
There is no shortage of opportunity for integrating AI. I think that the challenges historically has been having the technology to back it up, having the data sciences, the rigorous data sciences, and the robust validation behind it, to be able to rely,
and so that's why I think we're really looking to gain from this partnership, and the use cases are all over there. It's everything from trying to empower our CRAs to feasibility, and it's not about it's not about having this conversation all the time with our staff internally.
It's not about replacing them. It's not about not needing staff. All of those staff have all of our team members have a huge amount of expertise, a huge amount of experience, and we want to use the AI to free them up to take advantage of it
and to take away a lot of the routine work and drudgery preparation reports, and all of this sort of stuff, so that they can use those years of experience to best effect, you know, and that's the way.
Yeah, yeah, what we think about it – Because you make a great point, there's no shortage of use cases that we could work on, we try and make sure that we're focused on how is this improving quality in how the trial that's run or in the data collection?
Can it make you go fundamentally faster, so can you execute the trial overall faster, or can we help you be more efficient so that your people can handle more trials in parallel or more tasks and empower them into those insights?
So I think that's allowing us to think about this idea of impact. How do we make sure that as we're spending our precious time and energy and roadmap on these features, are they the kinds of features of the true label deliver impact into the trial,
and when we think about kind of the two factors that you could bring to that, it's can it help do something where the work is being done, so can it assist someone in doing work, or can it provide unique knowledge or value to them in insight that they could otherwise not have found, or have difficulty.
So, there's sort of the amplifying the intelligence factor, and then helping people do work, and those two factors allow us to really balance and drive into where are we going to bring AI where it's necessary, because there's lots of innovations we can do that have nothing to do with AI as well. You continue to focus on those roadmap items together.
And I think that's for us, that's a big part of the attraction of the partnership, because the opportunity to collaborate on identifying those cases, because I think we have, as an organization, we have the same, we share exactly the same philosophy.
We don't want to adopt AI for AI’s sake, we don't want to just stick a sticker on everything, saying “AI here,” you know. So it's the contemporary equivalent of a flaming logo. We don't want to do that.
What we want to do is implement AI in a meaningful way that actually delivers, you know, return on investment surely pleases this CFO, but, but also return on investment in the sense that for our clients, our sponsors, the trials are getting delivered faster, the treatments are getting to market faster, our patients are having a better, a fair patient journey, a better experience.
I want to highlight, you said implementing AI in a meaningful way, Medidata Plus Lisa, yeah, tell me just a little bit, very briefly, like what is this glow up we gave Dot? What is Medidata Plus? Our audience probably hasn't heard of it yet. Just, yeah, let's say really quickly, what is it?
So Medidata Plus is new this year. Dot we released into our platform last year. Dot is the indicator of AI at work within our platform, so anywhere where an insight is brought to you by Dot, there's an opportunity to chat with Dot, Dot is the orchestrator taking you from place to place.
What we did spend a lot of time thinking about and researching is how do we make AI as impactful for our customers, and some of the things we thought about were you need to have broad access to the data, so we've got to help with the underlying data platform type tools that bring your data together, so that you can make sense of it.
That's where AI really shows its value. And then, how do we make sure that all the different user types can get access to this AI, that we're not thinking about throttling it?
And did you run out of tokens by noon on Tuesday, and now you don't have any for the rest of the week, and so we introduced Medidata Plus, which is the offering that allows us to bring to customers and say, you get it all, you get all of our AI in any product that you're using,
all the AI that will continue to come on a platform data ecosystem, that means you're getting the right insight at the right time, we are going to help you move between those workflows, so it's our means of making sure everyone within the company that needs access has access to the right data at the right time, driven by AI, where applicable.
And we're all in. Yeah, that's that's why we signed this partnership, is we're all in. We recognize the value in that platform-centric approach to you need access to all of the data, you need the interconnections between the different components of the platform to really make the best use of this and maximize value add.
Yes, and that's that's why we've signed up for Medidata Plus, because you know, the fastest, as fast as you can develop it, we want to start the
It's a match made in heaven.
Yes.
I believe it's going to have a meaningful impact.
Yeah, it's, it's, it's definitely going to be a journey. I mean, a lot of this stuff is pretty cutting edge in it, and a lot of the stuff we've seen today is really exciting, but it's, it's, you know, it's rolled out, it's just come out, it's being rolled out,
and obviously we then have to adapt our internal processes, potentially to cocoon some of it, and we have to, you know, all of this stuff, but and some of it's about completely rethinking the way we go about doing business.
Some of this to best get the best value out of the AI. It's not necessarily just a bolt on, it's about thinking about how we completely re-engineer the entire process of conducting this study, so it's going to take time, but I would love to come back, and yeah, the initial signs where we've sat down and looked at some of this initially are really encouraging. You've seen some really transformative metrics coming out of some of the work we've done on CDS.
...
So, with I think got a couple minutes to go, I do want to end with one of my favorite questions, but please do your time responding to it. I'll start with you, Chris. What do you dream to disrupt next?
What do I dream to disrupt next? My gosh, I haven't the faintest idea. I don't know. Retirement planning, okay?
Getting into viticulture?
Yeah. Yes, I was just before the podcast I was saying that a chap I used to work for retired and bought a vineyard in New Zealand.
Amazing.
So we can all aspire.
There we go.
We all aspire for that. Yeah. No, I mean, I think I think this, it's been really invigorating the whole AI piece for the industry recently has been really invigorating, because you know there's a wave of trends that float, you see through flow through the industry over time, many of which the impact of which is relatively limited, but I think AI is fundamentally different, a fundamental foundation shift, paradigm shift, and so I think it's, it's really interesting thinking about the way the industry works in terms of the impact and AI has my gutta.
Viticulture for you as well?
A little dream to disrupt. So you know we've spent a lot of time with technology defining what could be possible, and we have to work hand in hand with CROs and sponsors for what they are trying to achieve, to be possible,
but there's a third party in all of this that, if we're honest, has made our industry more of slow followers as opposed to even fast followers or early adopters, and that's the regulations, and you know, I think that we are on the cusp of actually the first major change I've seen in that space as we speak,
so for those of you, I mean, I know we're across the pond, but the FDA has released its real-time clinical trials pilot that's going on right now, and there's companies globally that are looking to participate,
and I think this is an effort to fundamentally say why have we still conducted trials in the same way for the last 25, 30, 40 years? Can we have different signal identification? Can we have a better dialogue with regulators that takes a lot of the tools that we've been working on, centralization of data out of EHRs, centralization of data across the clinical paradigm, looking for endpoints, looking for safety signals, and then making faster decisions upon them.
I think we actually have the opportunity to change with that. The regulators needed to lean into it, and they're starting to show signals of that, and so I think that lets us do some highly exciting things with the tools that we've been talking about today, the AI-based signal detection, but the centralization of data, we've got all the right things to actually transform what it could look like to run a trial, and I think that's the big piece that's ready for disruption.
Well, thank you both so much for your time. This was really exciting. I'm hoping we can continue the conversation once you guys have fully joined us and started utilizing our tools. We're excited to hopefully take this continuously on the road with our next one, hopefully popping up in the next location we take next. Thank you both so much.
Welcome back to Dreamers to Disruptors here at NEXT London. Thanks, panel. We'll introduce you here in a second, but the premise of this podcast here today is around we're collecting more and more data.
It's actually creating a lot of noise in our industry, in our organizations, and in our operational teams, and we're not necessarily running any sort of better clinical trials or getting better outcomes,
so we're really going to be talking about today shifting from sort of simple data collection to truly orchestrating an experience. So let me quickly introduce our panel here. Robyn Orie, we just did a panel on stage here at London NEXT.
She is the VP of Clinical Data Management at Caidya. We've got Cassie Kendrew, who is the COO of EMS Healthcare. Did I get your title wrong?
You just gave me a really nice promotion.
Oh, great, we'll work on that. Don't worry about it. And then, lastly, but not least, I got Dr. Amardeep Heer, who is the Lakeside Healthcare and Site Insights Board member.
So, let's kick this off. Let's just have some fun with this. So, firstly, let's go into the first question here. So, Robyn, from a CRO perspective, what is the actual operational and financial cost when data review, central monitoring, and risk management are all live in separate vacuum tubes of data, different silos everywhere.
It's pretty big. You have, you have everybody, everybody working on the same data, but in their own little silo. So data management is looking at it one way with one set of output, medicals using a different set of output, central monitoring is using their output, and nobody's talking to each other.
So, either you're all actioning the same signal in four to five different ways, or you all think everybody else is actioning the signal and nobody's looking at it, and then all of the rework that that results in, we obviously can't charge for that. So, as a CRO, that's a big impact to our budget.
Yeah, Study absolute.
So Cassie, sites, clinical care facilities – actually to both of you, if you can – they're not IT help desks, we hear that all the time, right?
So, how is this explosion of separate, disconnected dashboards, all the technology that is being used by sites and point solutions solutions actually impacted your teams and their relationships with patients, which is ultimately what we're all here for?
Yeah, we were just talking about this off camera, actually.
I think, and it was touched upon in the last session, but actually the problem with all of the tech that we have, and one of the things that you were saying is, you know, we are really big advocates for tech, we love it, the teams get so excited about new tech,
and there'll be a new eCOA or something coming on, and the teams are really genuinely delighted about it, and then you can see the wind kind of go out of their sails, when the reality of, you know, they're getting more questions from participants is not actually straightforward, it's really dysfragmented or fragmented, yeah, so that's kind of thing.
The reality of tech is that actually, you know, a really, I suppose, another example, and very often we've, we've almost kind of normalized behaviors that wouldn't be acceptable in other kinds of industries.
So, for example, participant IDs, we've got five different systems on a study at the moment, and the ID has to be manually entered into all of those different systems, and that's the participant's first kind of view of you, and you're running from kind of one tech platform to another, you've got multiple workflows you're trying to work through, and actually the first time you're in front of a participant, you really are, it's all about building trust, and it kind of can erode that trust if you don't have a slick operation.
So, I think we don't really have sandbox environments to work through. And actually, that would massively help, you know.
Kay Duran, in the, in the US, talks about the patient negative one, that would be huge for us if we were able to work through all of the, you know, the participant experience kind of from start to finish on each of those different visits. I don't know what your thoughts are.
Clearly, I mean, our patients will probably expect us to be able to do that anyway, and when we don't do that, and it's visible to them, frankly, it can be quite embarrassing, really.
So, you know, we're recruiting them to a clinical trial, and actually we haven't had the opportunity to actually go through exactly what we're expecting the tools to do, but from a data point of view, I mean, we're all professional, we want to collect high-quality data for, you know, the sponsors and CROs that we're working with, and if the tech's not working, we're not getting the right data.
There's data queries being raised that you know we then have to go back and forth about how to answer those, and that takes time, and that's not funded either as well for sites, so all that extra time that's been taken, where actually if things did run smoothly, both for the patient and for the site, you know, the sponsor would be happy, their data management teams would be happy.
I think the whole environment, the whole ecosystem would be a lot happier if you know we didn't have these sort of issues and barriers.
Yeah, so Doctor Heer, we'll, we'll stay with you here a second. Was there… we're all doing more complex trials, they're not getting simpler.
Was there an aha moment for you where all of these disconnected workflows, all of these disconnected data acquisition tools, when you think about all of your trial ecosystem, I'm on a clinical trial now, they're running about 22 trials across six different sponsors with about six different technologies on each and every study, it's a real burden to them.
Was there an aha moment that it's just not sustainable to have that sort of landscape of technology in the modern clinical trial complexity?
Yeah, absolutely. I mean, we, we run up to 20 trials a year at the same time. Different stages. So we've got recruitment, we've got follow on, we've got close down, and all of those will have different systems,
so you know when we onboard a new member of staff, you know, it takes six months actually for them to get their accesses, get their trainee, and that's before they've actually really started getting into the day to day activity, so you know onboarding people, and then the worst thing is if somebody leaves, you know, replacing somebody, you know, you know,
so again, just having one system, or just very few systems, even if it was one system per study, doesn't have to be the same study system, we'd love it to be on Medidata, I'm sure, you know, but you know, if it was one system that we could train our staff on, and they could, they'd have accesses to everything, and they could transfer that training, so next time a study comes on,
you know, they don't have to be retrained, they might have to be retrained or updated on a new version, but you know, all of that makes things much more streamlined, and actually that is the one for sites,
and because that makes it easier for us to recruit patients at the end of the day, that's what we're trying to do, is to recruit patients to time, to target, on budget for our clients.
Absolutely, that's what we want you focused on, Cassie. To you, the aha moment, where it's just not working with this sort of very disparate landscape.
I think it's absolutely the same, we are, we need to be operationally agile, so we're a mobile network, site network, and we use mobile units, we take them into the community, every other part of our kind of operation is really, really agile.
Bringing people on just isn't, because that training burden, you know, you're talking at least two weeks for every study, and if you've got multiple studies running through a site, and you're looking at, say, months before you can get people up and running.
And I think for us, we realized that the more complicated protocols are, and the more data sources in kind of systems there were, there's only so much good people and workarounds can actually cover up,
and I think, as sites as well, we're not in control of our kind of the systems, and you kind of, it's forced upon you, so that's fine, but where we are, and have been in control, is we've implemented CRIO, so we've done that in the last six months,
we've just rolled it out on the first three studies, and that least we have got control over an element, and that simplified that. So our whole moment for us was this just isn’t workable, and it's certainly not scalable. So at least if we control the bit that we can control, yeah, I think that's kind of the help up.
Yeah, so Robyn, you come in a little different angle than Cassie and Dr. Heer, what's the.. what was the aha moment for you?
So, I don't know if there was one specific moment, but we have... we have early phase trials that have over 20 different external data sources, so when you put that together with all of the EDC data that you're collecting and all of the reconciliation that needs to happen, and that doesn't even count as AEs,
so all of those data points that have to come together, and you have, you know, data management doing just the high level recon between the EDC and the lab files, the missing samples, there's the CRAs that are looking at the lab, if the lab is big enough.
The lab provided, you know, dashboards trying to find where things are at. Was it collected? Was it sent? Where what's happening? And just the disconnect there of who's managing what, and multiple issue logs and vendor communications like sites are getting the same queries from data managers and the CRAs
and you know the external vendors are getting everybody's going to frustrated, so we've we've we have to figure out how to streamline all of that data into a central location and make sure everybody's looking wheat the same thing and those those you know, for the central labs that have those dashboards and everything, that's not validated, so you have to take what you get out of there with a grain of salt, you're taking the reconciliation and the sample tracking that you're doing there at a risk.
Yeah, yeah. So let's pivot a little bit here, maybe, and we are when we talk about a unified data ecosystem, we're really not talking about sort of integration through APIs, because Cassie and Dr. Heer, to your point, your staff are still in those systems, we just may reduce duplication, which is great duplication of entry, but it still means there's another system there.
So we're really talking about a cross-functional paradigm shift. How does moving away from sort of reactive data cleaning and review and reconciliation weeks after the fact to proactive sort of real-time identification of signals really mitigate the risk of conducting clinical research for you? We'll start with Robyn, so Cassie and Dr. Heer can think about it from their perspective.
Sure, so you know from a site perspective, you guys have the patient come in, you have the patient visit, you enter the data, and it could be six to nine months to longer before anybody even looks at it and starts to issue theories,
and then you guys have to go back through medical records and try to figure out what might have been happening, so you know it's hard to go back and retrospectively get that information, so with. So with the centralized platform being in that proactive signal detection, it means we're looking at the data sooner,
and you know it might not be immediate, but hopefully within, you know, a few days or a week, or at least a reasonable amount of time, that's when we're coming back to you guys with queries and getting that data cleaned and getting it locked down, so we can move on to the next visit, and the next subject, and you know, not going back a year later, and trying to figure out what might have been going on.
Yeah, I think from our point of view, it's, you know, we don't want to get it to query stage, we want to be identifying it before it gets to you, so it's, and I think CRIO is definitely helping us with that as well, you know.
Our PIs have got oversight and ought to be units, we've got physicians on each unit, and then the PI is, of course, necessarily on one of those sites, but how's that overview of the data immediately.
So, from a data integrity wave, you're able to question stuff, able to identify hopefully where there's kind of any anomalies, and then also it kind of supports with us looking at and training and looking at trends and look at things in the moment, so being a bit more proactive about kind of query resolution and what we can put in place in as mitigation strategies.
I think for us also, yeah, I mean it's managing our teams as well, so you know, being every… having that oversight of, you know, what is the timeline of data entry?
Are we meeting the KPI that we've been given by the sponsor or the CRO to get the data on within the first five days, or you know, a data SA data within 24 hours. How quickly are we responding to queries? You know, and actually performing, managing our own team.
So, I should say, look, you know, this is what you need to be focusing on. This is the, you know, prioritizing the work, you know, you know, we're doing, for example, doing a fast recruiting high volume vaccine study, you know that data needs to be on there as quickly as possible, queries need to be resolved as quickly as possible.
And having access to those tools to be able to manage that, and that's one of the things I have for Site Board, of the things I’ve related, you need to be able to give the site to empower the site to be able to see all of this in, you know, one view, so they can see all of their Medidata studies, and they can see, actually, you know, we're not hitting the five day rule for this particular study.
Who's responsible for that? Do we need to put any extra resource in? Do we need to put training in, etc. whereas we're at the moment, we're waiting for the CRAs to come. When they come into the visit, they give us that feedback, but that, there's a time delay for that, you know.
So we want to be more active, we want to be proactive. You want to deal with it at the time, rather than two or three months later.
And to build on that, like there's always been those KPIs for the sites, the data has to be entered within X amount of days. The queries have to be turned around, but that until recently hasn't been on the CRO side.
Could take us a year to review the data, and nobody was saying to us, “What's going on?” But now with all this new technology and the trends with the signal identifications, we're now feeling that heat to get to get those things identified faster.
And actually, we want to know that we want to know how our site is working, and if there's problems, we want to know them sooner, so we can do something about it.
Ultimately, you know, your staff didn't get involved in sort of healthcare or clinical research to start doing the same thing over and over and over again. They want to be with the patients, they want to advance the research, so you know it's very important that we tie that together.
So we're going to close out. We've got a couple of minutes left with a real quick fire round that you, I want each of you to finish the sentence, and Dr. Heer, we'll start with you. “The data task we do manually today that will be completely automated and invisible tomorrow is…?” Go.
Is a completely integrated sort of system between our clinical systems and our research systems?
We want patients who we look after routinely, automatically being given the opportunity to take part in a clinical trial, and those systems will be fully integrated, so we're not using loads of different systems, using one system.
And mine would be manually moving data between multiple systems, so no one should ever have to enter the same data twice.
Yeah.
So speaking of the manual, manually entering data, just the different types of manual review listings, all of that data is going to be in one place. All of that data review is going to be in one place,
and it's really going to allow us to focus on that kind of critical to quality data, and get rid of all of that noise that's not really necessary for the analysis, which means we're not sending unnecessary queries to the sites for them to answer,
like nobody, it's not really critical if you know this date is off by, you know, one day, like that's not going to blow up the whole protocol.
Yeah, well, I know that 20 minutes has gone really, really quick. I want to thank each of you for this very interactive panel. Thanks for attending the Innovation Day here in London. Thanks for what you do for our industry, each and every one of you, because ultimately we all want to get these treatments to patients quicker. Everybody is waiting for the output from the work that we're ultimately doing, and we really thank you for your participation, the inclusion in that. So that's a wrap for Dreamers to Disruptors here in London, and we'll see you next time. Thank you.
Thank you for tuning in to today's conversation. If you've been enjoying this podcast, please subscribe to our YouTube channel and follow us on Spotify, Apple, or wherever you get your podcasts.
If you have questions for me or thoughts about the episode, drop them in the comments. I do read them. Thanks again for listening to from Dreamers to Disruptors, and we'll see you again next time.