Trials with Maya Z

  Hello and welcome to the next episode of Trials with Maya Z, brought to you by TrialHub, a data intelligence platform that helps clinical research organizations and sponsors plan clinical trials. This podcast is about how we can make clinical trials more successful and patient-friendly. Friendly, I am your host Maya C and in every episode, I will be interviewing a leading expert from various industries in order to discuss some of the major challenges and brainstorm how we can solve them.

Let's get started.  

Hello everyone. And welcome again to another episode with Trials with Maya Z. I'm your host Maya. And today I have a very special guest, Daniel Schwartz from the Institute of Biostatistics and Analysis, who is here to tell us how biostatistics can actually change clinical trials and not only.

Daniel, giving the words to you to introduce yourself and tell us a little bit more about your background and how you ended up working in clinical trials.

Yeah. Hello, Maya. Thanks for inviting me. I'm really looking forward to this, interview. My background is coming from biomedical engineering. So, originally, I was educated at the University of technology, but later I transferred to the Masaryk university, where together with the university and my business partner, Peter, we set up an Institute of Biostatistics and Analysis, which is in fact an SME or a spinoff, spinoff of the university.

And today after 10 years, I can say that we are a full-service CRO active in clinical research. And, from a background in biomedical engineering, a lot of technology, and at Masaryk University now I am also an associate professor at the Department of Simulation Medicine. And I enjoy having some doctoral students in simulation medicine.

But also in neuroscience. So this is part of the career, but the main role here is really the Institute of Biostatistics and Analysis.

Super exciting, Daniel. I wonder, tech background, what led you to start a clinical research organization? Isn't that more, let's say, service-heavy and less technology? Or maybe that's not how you see things?

No, I think you are completely right. Anyway, we were always more people in the Institute and my role was really the technology. So, still, I am the CIO here and director for RnD. And, really for the service, I mean, the regulatory project management and data management, we have other people, so it's teamwork.

And my role was always the technology, innovations, and enhancements, and visions. And yeah, and the other parts it's my colleagues.

So maybe that will sound like quite a vague question, but because again, you have this interesting background, I wonder, from your perspective, what is the role of technology in a CRO? And, actually, your CRO is, I would say, relatively new compared to Other CROs. So I wonder what brought you to start doing this like CRO in general.

And from the very beginning, how did you integrate technology into CRO?

Yeah. Thanks for this question, because this is probably what sets us apart from other CROs. We have a strong R&D component, which consists of computer scientists and data scientists. And from the beginning, we decided with the business partner, also with the university that we would try to build a completely new EDC system from scratch.

Previously at the university, we were using open-source systems, which at least I perceived as not good enough for such a complex area as clinical research. So we were thinking of starting a serious business also in software development. At first, we developed the system only for our projects, for ourselves, for our data managers, and It's been now three years since we transferred really to the market with the product.

So, in parallel with providing the CRO services, we started also the business with the system itself. And today, I can say that we have several users, several small CROs who are already using our system as their EDC.

And I wonder. There are so many different providers of EDC, and then again, we decided we need to build our own. What is the difficulty with EDC in general? What is, let's say, the secret sauce that actually so many providers haven't achieved so far, like today?

Yeah, you are, right? There is a big competition. Anyway, I think that it's a little bit different than 10 years ago when we were deciding. So, unfortunately, there were many companies. Anyway, I think that making the software really as a CRO and not as a software house, it helps us to develop a product that has clinical data management I would say in the DNA of the product itself.

And this is something that we are getting, as feedback from our customers, because they are really saying that we see that this is not a product from software developers, but from a CRO, because the processes are there, we don't have to bend our processes to software, but we take your software and it's like we are.

But this was also a part of the motivation to develop really software that would be tailored to our processes because we believe that we were already good 10 years ago, which I think today is a little bit funny, but, anyway, yeah, that was the decision in that time.

I love what you said about clinical data management as a DNA, and I'll tell you I'll take probably like a few seconds a minute to actually explain why I love that, because for all my Career and especially since I started doing the podcast I've been interviewing a lot of people and a lot of them were indeed representing the sponsor point of view.

And a lot of us at the end of the day, what is a clinical trial? It's a hypothesis we try to prove throughout the whole process. And the data is what actually validates if our hypothesis was right. Or not. So without the data, you actually don't have a clinical trial or actually, yeah, you don't have a clinical trial.

So that's why, when you said clinical data management as a DNA, it's 1 of the things that it's super important for us. I thought, what a beautiful thought because I guess we all need to have clinical data management in our DNA. If you want to be successful. When we speak about clinical trials, speaking about data management, Daniel, what are these main mistakes companies make when they're thinking about their data management?

Have you collected any observations, any stories that you can share with us? Which are these main mistakes companies make when they're thinking about their data management? Why many companies end up being not very happy with the final result?

Yeah. From my experience, I can see that companies are not so much unsuccessful in data management because usually when you run a business, you are very careful about all aspects. But what I can see is that data management is somehow underestimated in a bigger organization like academia.

And as a spinoff of Maastricht University, We have still many contacts with the scientists and academic community. And very often, we see that very good scientists, level one scientists, they are still collecting data in Excel. And, after two years of collecting data in Excel, they are trying to do something with that.

And finally they, Always arrive at some kind of CRO or some kind of tech company. And they are trying to do, things which are really tough because the data management wasn't there at the beginning. So I would say. If we can generalize this, the biggest mistake is when you don't employ data management, the proper data management into the design of your project from scratch, from the beginning. This is the biggest mistake.

If you start with data management from the beginning, you can always make all the changes because change management is part of data management, but if you don't have data management set up at the beginning, this is probably the biggest mistake. And yes, it can be also in academia and maybe in companies, but I can see it mainly unfortunately in academia.

  So, Daniel, as a data management provider, how much do you need to understand the regulatory expectations like yourself, or do you solely have to rely on your clients?

Yeah. So speaking from the Institute of Biostatistics and Analysis, I can say that we have or estimate that we can have around 180 documents. I mean, control documents, SOPs, forums, et cetera, and, at least a quarter of it- more than 40 is focused on processes in data management. So I think these documents capture all the regular regulatory staff, which is really needed. So if you consider us as a CRO, very focused on data management. Yeah, we need to understand regulatory affairs and besides the data management component, we also have a regulatory affairs manager here. So, Yeah, we need to do it.

We need to grasp all the aspects of data management and regulatory is, yeah, definitely one.

  I hope you're enjoying this episode. If you find this topic relevant, you'll find it helpful to book a demo with our team on trialhub. com. Since 2019, we've supported more than 3, 000 clinical trials with country, site, and patient feasibility. We'd love to show you how and help you get your trial right from the start.

And now, back to my guests.

 I guess then it's fair to say that whoever is looking for an EDC provider, one of the things that you need to do to check in there like checkbox is how much they understand the regulatory requirements before you actually go for the tech. So that's why I was asking. I wonder what is the biggest challenge actually for creating an EDC platform. Is it the tech or is it something completely different? Because my observation, Daniel, maybe you like prove me I'm totally wrong here, is that in clinical trials, most of the tech solutions are not actually tech challenges. It's more like the complexity around stakeholders, the whole industry in general.

So yeah. Tell me more about EDC. What's the main challenge when building EDC software? Yeah.

I think that your understanding is very similar to mine It's not a big thing Tech challenge, the biggest challenge in creating a new EDC and providing it to the market is to have a good team of computer developers, who have some history in the company because as the system is very complex, it's not an e-commerce system.

It's a very complex system. So you have to have guys who are here quite long because there is a memory and some persistence needed if you want to create such a complex system like EDC. And our EDC system is not just only the EDC. There are a lot of features, which we can say they are on the border between EDC and CTMS.

So there are kind of also clinical trial management stuff in the CLADE-IS. So it's, yeah, it's very complex. And for such a complex system, you need to have, developers who have a longer history and At least in our region, this is the biggest challenge, to have developers who want to stay with you longer than I don't know, two or three years.

Hmm.

Interesting. That's very interesting. And definitely an unexpected answer. But I can imagine that

this is actually what contributes to the overall quality at the end of the day. And I have this very interesting question. Imagine you put yourself in the shoes of the sponsor, for example, or whoever is actually conducting the clinical trial.

Maybe it's not like the sponsor. Maybe it's a CRO, another CRO company that's using your EDC solution. What would be the reasoning behind choosing a bigger EDC provider versus a smaller organization, smaller company, EDC provider again?

Why would you choose a small one versus a big EDC provider?

Okay. So, I would say that the benefit on the side of the smaller one is flexibility. Also, I can say we are more eager to work than the bigger players. I don't know if you have the same experience, but the smaller teams are always more flexible and somehow more eager to work.

So, yeah. This is the first thing. And the second, which came to my mind is the price. Usually, the smaller players try to be competitive with the pricing, for example, in our case, I think our prices are competitive enough. So, we try to add some value and besides providing our clients with the system, we provide them also with the capacity of our senior data management staff.

Interesting. And why would you choose, for example, if you're now like the sponsor, and you would consider a big EDC, why would you choose a bigger EDC provider? That's a valid

It's a tricky one, because I would never choose the bigger anyway. Yeah, I think that you want to hear something about that. And I believe that it might be the case that the bigger players, they must be compliant with more systems. I mean,

Okay.

So I can imagine that there are some exotic countries, I mean, in our region can be exotic. For example, Japan, I don't know the regulation body in Japan, but a bigger company, a global one, doesn't have any problem with being compliant with all legislation in Japan. So this is something that might be the reason for choosing a bigger player if I want to become global.

So, if are looking for an EDC provider another thing on the checklist is which are these countries that you comply with. Because you can be a small player in Japan only and comply with Japanese .. Actually that raises another question.

How much are different like smaller EDC providers collaborating together so that if you have a global clinical trial, but then local EDC providers, is it even possible to have this smoothness, like seamless experience as a client of like multiple EDC providers?

Is it even possible?

I don't know if I respond directly what you are asking for, but, In our region, I'm now in Brno in the Czech Republic. There are several small CROs and I can say that we have a very good collaboration because we as a CRO focus on data management and on the EDC, we are providing them with the technology.

There are other CROs, which are more focused on decentralized studies. They have CRAs. in the terrain, what we don't have, for example. So, we can make a good let's say mixture of smaller CROs and ask for bigger businesses. And yes, we are trying sometimes and sometimes we are also successful.

So, there is cooperation and I'm very glad about that, that we are not only competitors, but also cooperators. If that answers your question.

My question was more about other EDC providers. Is it even possible? It's like thinking aloud. Basically, is it even possible for multiple EDC providers to work on one in the same clinical trial and still provide the same level or less similar quality level of the data? At the day, imagine like the clinical trial is involved in the Czech Republic and Czechia, and Japan. And so you use, let's say, your EDC, like you're the EDC provider in Czechia, but then the EDC provider is a Japanese company, for example. Is that even feasible and possible? Mm

yes, it is possible and feasible, but from the business perspective if I were the sponsor, I wouldn't go for that. Anyway, there are, already ontologies or data models, which if you agree on them, like for example, OMOP, if you agree on them, and both or all the EDC systems, which are involved are compliant, with this data model, then it's possible.

Got it.

From my experience, there will be always a kind of risk. So there in such a case, I would see the biggest risk in data integrity.

Interesting. All right. And because everyone speaks now about AI, there is no way I don't ask you, what is the role of AI in EDC and data management in general? Are you rather positive or negative? And like I said, in our conversations before, there is no right or wrong answer. There's just the perspective.

I also wonder very often, and I want to hear your opinion, AI and EDC. Is that a good marriage or maybe not so much?

Oh, thanks for asking. This is a good question. AI is a buzzword. And yes, if I consider myself a neuroscientist, I have always tried to employ machine learning into my algorithms, into my research. And so when it, when the generative AI came here. I was very happy about that.

And still I am, anyway, I would make a big border between AI and machine learning. Machine learning is only a subtype of AI and I see machine learning as a very good opportunity for making EDC systems and making the quality of data better. And this is also

something which we tried to do. So we employed some machine learning algorithms into the EDC system, which is now capable to ensuring quality control in terms of detecting anomalies in the data. In a multidimensional space. So we are looking for some patterns, which are kind of suspected anomalies.

And then, yeah, in this way, with the use of machine learning algorithms, we are making the quality of the data better, so this is machine learning. I can't see now a better role of AI in EDC than this, but maybe maybe I'm just missing something.

Yeah. But anomaly detection. Yeah. We will see. At the end

of the day we all think is get out of the buzzword and think more of the practical applications of technologies because technology is no good without actually the so-called jobs to be done. Like, what do we need to achieve? So thanks for this answer.

I have one last question. It's my favorite question that I ask every single guest, which is if there is one thing that can make clinical trials more patient-friendly and successful, what would that be from your perspective? From

my perspective as a technology CRO, the first thing that came to my mind is, of course, enhancing the usability of the platforms, that we use, for example, for collecting prompts. I mean, the patient-reported outcome measures, which are essential to understand the impact of the interventions, on the quality of the care, or from the patient's viewpoint.

Yeah, there's the usability and the other thing if I can make some example from our projects, which are more, let's say, observational, there's also what I would call proactive engagement. We have a very powerful example in our history when we, in a critical situation, were able, for example, to call patients into one of our studies.

It was a special subgroup and we were able to call them into action because they were in danger. It was in COVID pandemic times. And with the doctors, we identified a subcohort, that was in danger and we were able to proactively help them to avoid, exacerbating the disease. So it was a moment when I was touched by our own business, but yeah, I remember that as a powerful

Daniel, thank you very much for sharing that story. I agree with you every single time when we can see what's the impact of our business,

with like actual people, no matter if patients are not like actual people that always like. brings another perspective of whatever we're, whatever activity we're doing, business, nonbusiness like NGO, whatever we're doing, but the impact is important.

Thanks for mentioning your perspective as well. And, like providing this answer and in general for the conversation that we had, I hope the audience found insightful our conversation around EDC CROs and technology and CROs and good luck with your Institute.

Good luck with your institute. Yeah. Thank you very much, Maya. It was really a pleasure to talk to you. Thanks for inviting and yeah, I hope that the listeners will enjoy

Absolutely. Thanks, Daniel.

  Hope you enjoyed listening to Trials with Maya Z. If you're interested to hear more about how clinical trials can serve patients globally, subscribe to the podcast on Spotify, Apple Podcasts, and Google Podcasts. Have a great day.