Transcript
Speaker: Welcome to Clinical Data Talks, a podcast brought to you by CRS-Cube. I'm your host, Sylvain Bertelot. Join me and industry experts as we discuss the latest trends impacting the world of clinical data.
Speaker: Now I've had the pleasure of recording podcasts in the past with today's guest, and I know we're in for a treat. She's the Executive Director of Clinical Data Sciences at Premier Research, where she's instrumental in linking technology with quality and efficiency improvement.
Speaker: She's also a key member of ACRO's RBQM Working Group, and we're going to talk about their recent work in a few moments. I know she's also passionate about mentoring the future generation of clinical data scientists and challenging the status quo.
Speaker: Let's welcome Jennifer Stewart. Hi Jennifer, how are you doing? Good, thanks. How are you? Thanks so much for having me. Lovely to connect with you again.
Speaker: And you. It's always a pleasure. It wasn't under the the same podcast, but we we talked about ACRO at least once when when we recorded podcasts. So I always love hearing from you and what you're doing with ACRO.
Speaker: Which leads me to my first question. For those who don't know about ACRO, could you tell us what it does in our industry, please?
Speaker: Yes, absolutely. So I've had a pleasure to be a part of ACRO for four four or five years now. And is the Association of Clinical Research Organizations. so it is a industry organization that collaborates with regulators, make sure that there is a positive, supportive regulatory environment.
Speaker: um And we advocate for for safe, ethical, high quality research to support participants and help them benefit from treatments and therapies. um ACRO has many committees within its organization spanning from regulatory to data privacy,
Speaker: artificial intelligence and machine learning and others. And members of various ACRO member companies sit in on these committees alongside ACRO employees. So I am a part of the risk-based quality management working group.
Speaker: um Risk-based quality management is a pretty common term in the industry these days, but it's an adaptive approach focusing on the greatest area of need and where there's the greatest potential to impact patient safety and data quality. So our working group is trying to promote greater adoption of our BQM in the industry.
Speaker: um As a part of ACROM, I've had an opportunity not just to be a part of podcasts like this, but to attend various conferences, participate round tables, even FDA listening sessions.
Speaker: um We have publications out there that feature data that we've collected about risk-based quality management and clinical trials. So it's a really great forum for um moving forward innovation in the industry, advocating, of course, for patient safety, um but but being efficient and um having the ear of regulators so that we can kind of help move um adoption with their support.
Speaker: And so specifically for me, I helped kind of move the needle both on the education and training and the adoption Our working group leverages the data that we collect about trials that are conducted by our various companies. And then we also try to find innovative ways um to keep the industry moving forward.
Speaker: Nice. That sounds very interesting. I'm intrigued by FDA listening sessions. What do they look like? So there are forums once or twice a year where We are invited, ah members of ACROR are invited sit in and talk to them about the research that we've done and what we're seeing in practice in the industry. So especially with ICHD6R3 coming out, um it's been several years in the making. You know, a draft of R3 came out a couple years back at least. We've had the opportunity to give feedback on drafts of those documents.
Speaker: they're They're out there for public consumption and for feedback. So we just take advantage of that opportunity. And then we talk to them about the data that we collect from our annual survey and where we've seen movement and where we still see some sort of resistance, where we think they can be clear about what their recommendations are.
Speaker: um being more visible in the industry. For example, um some of the conferences I've been at recently, we've had members of the FDA or EMA or other regional regulatory groups sit in on panels and speak to topics like this, encouraging the industry to be comfortable to make a change. And so things like that are topics in those listening sessions where we really have the opportunity to collaborate.
Speaker: and hopefully influence some of the regulations and guidances that are published by the agency and and similarly with other regulatory Nice. that I like that. Seeing like, I always, I'm happy when I hear about collaboration across industry. i imagine that at Accro, you get to talk to ah some of your competitors, but actually i imagine that you're trying to move the needle in the same direction together. So I really like that.
Speaker: Yeah, yeah, exactly. it's It is a unique forum in that We are all working together to elicit change and we we run into some of the some similar challenges.
Speaker: We bounce ideas off of each other, right? Of course, we're all running a business and we're competitors in certain arenas, but in this particular forum, we are all working together to try to better are the environment,
Speaker: bring drugs to market faster. And and it's it's quite a unique and so far pretty successful organization. They've they've done a great job, not just in the RBQM space, but in a lot of the different areas of focus to move things forward.
Speaker: Mm-hmm. And you've mentioned surveys. so I know you run surveys with the AmbQM Working Group.
Speaker: um And when we talked about it, you had something interesting to share about the the adoption of centralized monitoring. Could you share a bit more about that, please?
Speaker: Sure. So since have run a landscape survey, looking at risk-based quality management, um, components of risk-based quality management that are used on clinical trials and across the, uh, seven member companies, CROs that are member companies of Acro.
Speaker: Um, each year we pull the data associated with the trials that are ongoing in that calendar year. um So it includes studies that are new in that year, as well as ongoing studies. So we've done this for six years and we are preparing the seventh right now. So which will reflect the 2025 calendar year.
Speaker: So we use this survey to evaluate how risk-based management has been adopted over time and use that as a proxy for how, you know, how RBQM is being adopted across the clinical track industry.
Speaker: From year to year, there have been somewhere between 4,000 6,500 active studies. And in a given year somewhere between and starving. So with the collection of data over time we've We've adjusted with the industry and and what's happening, but we've tried to remain as consistent as possible so that we can show changes from year to year.
Speaker: um Not surprisingly, so you asked about centralized monitoring. Not surprisingly, over the course of 2019 to 2024, we've seen what we call traditional trials really fall off. So in 2019, when we started, only about half of the trials that were ongoing at that point used some component of risk-based quality management.
Speaker: And now from the 2024 data, which was the last readout we had, there was negligible amount, something like 4%. So we've seen the impact of COVID um on adoption of remote monitoring, for example.
Speaker: We're also anticipating that as we pull the data for 2025, we'll see some early effects of ICHG But what's been interesting is kind of post pandemic normalized industry of such a thing exists, 2022, 23, 24 data didn't show as much of a change in the adoption of certain components of RVQM as we would have liked. So from 2019 to 2022, there was a significant swing in certain things. um
Speaker: Risk assessments from the time that they started to be required we've seen those all go all the way up to above 90%. And really we think every trial now is having a risk assessment. It is a requirement. So um that is a part of what has moved the needle.
Speaker: But what's interesting about other components like centralized monitoring is that we've we've seen an increase, but it's kind of leveled off around 50%. just under of trials
Speaker: leveraged centralized monitoring, which means the other half didn't. And that was surprising to us. We saw slight increases in that in those couple of years, but at the same time, there wasn't much of a change to reduced SDV and reduced SDR actually declined. So that was very surprising as well. So, so in 2024, there were only a third of data, a third of studies that had a reduced SDV strategy. So again, that means two thirds.
Speaker: we're relying on 100% STD still, and less than a quarter with reduced SDR. So three quarters are still doing 100% source data review.
Speaker: um So even though research has shown there's a lack of benefit of transcription checking, that the key points of data no longer sit in EDC necessarily, or that there are other ways to look better at data, there still was a really high number of trials where um, where there was a high rate, uh, if not, you know, hundred percent, uh, SCR and SDV.
Speaker: And so then even more surprisingly is that in 2024, some of the numbers dropped from 2023, including reduced SCR. Um, so, so we think, we hope that we're making an impact on the industry with all of the the various um, forums and strategies and publications. Um, but we were a little bit disappointed to see that while some of the components of, of, um, adoption of risk-based quality management, um, have improved over time, there's a little bit of kind of an idling or leveling off now.
Speaker: And that's really where we want, need, hope to see a change, especially in light of the latest revision to ICHE
Speaker: Yeah, you'd expect that no no trial would require 100% 100% HDR, you know. and i And it's interesting because, i mean, i I sit on the technology provider side, And that there's the there seems to be this constant to and throw between needing technology to advance to support reduced SDV, for example.
Speaker: But then the actual adoption that takes time, that may not be fast expected or even may be... fast as expected or even um may not be there there may not be a will to actually implement those technology advances. So i find that interesting. Have you been able to investigate the reasons why the centralized monitoring is underutilized? It's an interesting question. We have some theories. Keep in mind, you know, what I'm going to share are my opinions. ah Everyone has an angle on that. it's hard to actually
Speaker: prove them. ah But one of the things is kind of to your point earlier that there's a perception that the cost to invest in technology to implement some of these strategies are too great to overcome, especially for smaller companies.
Speaker: So in reality, what we've seen, especially now that we're a few years into this is that yes, there might be an initial investment either in technology costs itself, or in procedure changes, right? There's change management associated with shifting the way something has always been done to something new.
Speaker: Um, what we've seen over time is that there is a return on investment and reducing time and an operating costs. Um, the more that those tools and processes are used, people are more efficient or more familiar. There's less, um, you know, that upfront cost is kind of a one-time cost.
Speaker: So, um, Anyways, it it does seem like there is a perception that cost is still an inhibitor. What's more prohibitive, in my opinion, is the tendency for um for companies to be risk averse. So this is where we're trying to advocate with regulators, because despite what guidance is, day what requirements are out there, it still seems like there's not a willingness
Speaker: to look at less than everything, right? this This concept of not looking at every data point or doing transcription checking of everything. um And where we saw reduction in SDR, so source document review, and perhaps that's kind of a catch-all or a a backup plan for centralized monitoring.
Speaker: So even where there is centralized monitoring, One of the benefits of leveraging centralized monitoring is that you can reduce s SCR and SCV and target intentional places where those reviews need to take place versus all the time of looking at everything.
Speaker: But still, yeah it it seems like there's some fear, some lack of trust, some general risk aversion to leaning on the technology to only looking at at the data that's or the processes that matter most.
Speaker: In reality, you know, some of the data in a trial matters more to the outcomes to signal safety concerns. And so that's really, you know, where we we want what we want the impetus for adoption of rvcum to be.
Speaker: um And again, with working with the regulators, think feedback in one of the roundtable sessions I i was in was that during the review of draft R3 feedback, there was a lot of misinterpretation or over-interpretation. People generally taking a conservative to approach conservative approach to what was written in the guidance.
Speaker: And so one of the things the ICH D6 R3 final version and is is supposed to do is kind of remove some of that ambiguity or at least be very clear about specific topics where they've seen that that people aren't adopting or or or taking what they've written in the direction that they expect.
Speaker: so Um, so, so those are my thoughts. There maybe other reasons, uh, from the perspective of someone who's in the data side of a CRO where we are promoting, you know, uh, the use of, um, subject visit sampling and, uh, reduced of source data review and verification and centralized monitoring.
Speaker: pretty much on every study. um it's it, it's difficult to understand, how, um, some of, some of these perceptions, right. I get it. It's just how our industry is. We're very slow to change, but it's a little bit challenging for all of the data.
Speaker: And all of the the outcomes that we've seen that point to this being a positive, beneficial, cost-effective solution, but still it hasn't been adopted.
Speaker: yeah Yeah. Yeah. And it's interesting what you said about R3 because R3 pushes for a risk-based approach and, and I mean, I don't know. i haven't been in those discussions. So obviously I can only imagine, but like I've been in project management, for example, where you have a risk-based approach, you're not going to fully validate everything. So here it's a similar um similar concept that you don't need to check every single thing.
Speaker: and you need to focus on what is high risk. And that's clearly stated through like through throughout R3. So I'm surprised to hear that people have interpreted that taking um a careful approach in a way in implementing risk-based.
Speaker: ye Yeah. Agreed. Even R2, which has been around for decade, no risk based approaches were called out and, you know, not requiring 100% You know, that it's been more than 10 years that those things were out there and introduced. And so, so we hope that the changes and of the attention R3 has gotten will help us turn the corner. But unfortunately, it's it's not really a new concept, which is why it's a little bit frustrating. Yeah.
Speaker: Yeah. So you you mentioned ah roi And that's something I'd like to talk a bit more about, because my expectation is that if you implement a risk based approach properly, that should reduce your monitoring costs overall, because you're focusing on areas of risk. You don't necessarily need to to do monitoring as much monitoring as as what
Speaker: you did in the past. So ah you talked about upfront cost, a but then you've got this ah ROI. Is it also your as expectation that it should reduce your costs of of monitoring?
Speaker: And is it something that you've been able to measure or estimate? That's a good question. And and the answer is yes Yes, that's my expectation.
Speaker: and yes, we have seen some of cost reduction, some cases where cost reduction was demonstrated. um so The purpose of centralized monitoring is earlier detection, right improving data quality, increasing patient safety, having greater visibility, um and reducing unnecessary resource use. so Yes, particularly with monitoring it, it doesn't mean that we're not doing monitoring where we're approaching monitoring in a different holistic way that doesn't depend on boots on the ground to be executed completely.
Speaker: role of of the CRA or the site monitor has evolved, right? It's a very important role, especially for relationships with sites and for things that are difficult to identify, um when Um, looking at, uh, you know, aggregated data or remotely, but from a data monitoring perspective, yes, the costs associated with travel and time on site, uh, plus the delays, right. The time it takes to.
Speaker: someone on site, complete the review, right. All of those processes have added days or weeks, um, to the cycle times by depending on, um, uh, some.
Speaker: physical presence somewhere, for example. So, you know, a key component of ICHE 6 or 3 is the concept of proportionality or risk proportionality. And, um you know, where the risk is low, taking the opportunity to costs, whether that's with time or travel or, you know, um the type of resources that are involved, involved um you know, there's there's definitely opportunity to save.
Speaker: And we have seen some trials that end-to-end beat the the typical cycle time. So, of course, enrollment is always a dependency for a trial timeline.
Speaker: um But from a, you know, kind of apples-to-apples comparison, being able to go from um hitting a mark to delivering an interim analysis or a an output for a DSMV or blast patient last visit out to lock. There's a lot of opportunity to condense that.
Speaker: um so So there are various cost savings here, not just straight up dollars, but a lot of components that contribute to the overall investment or costs being reduced. and and i heard in one particular case that there was the potential to reduce costs of up to 30%.
Speaker: Now, our landscape survey does not capture anything associated with costs. It's all um activities associated and other metadata for for those trials.
Speaker: But I have, you know, i've I've been in forums where I've seen presentations or case studies that have shown the opportunity that centralized monitoring has given a company to reduce the cost.
Speaker: um Something else that's interesting is that, you know, we don't publicly share where we've failed. So a lot of a lot of what we have in the industry is anecdotal.
Speaker: Actually, this is something that um I've heard in forums, even with regulators, you know, how, how can we What can we show, what can we share that either, you know, when something didn't work, it wasn't linked to the reduction or in SDV or the adoption of central as monitoring and vice versa, how can we prove the success is other than just saying,
Speaker: you know, products have been approved and their results were positive and these particular trials used a reduced, a targeted, a proportional risk-based approach. um So again, kind of based on anecdotal discussions, we have not heard that use of a risk-based approach um translated to a lack of approval. Yeah.
Speaker: Just that we've heard some anecdotal case study successes um with either shortened cycle times or reduced kind of bottom line costs.
Speaker: Yeah. And I imagine that it's difficult to measure anyway, because you don't know, two trials are the same, so you can't compare an apple to apple. ah So going back to what you're doing with Acro, do you have any um expectations of how to improve the adoption of centralized monitoring?
Speaker: um Specifically with ACRA, we're continuing to gather data. We're continuing to evaluate in particular now what we see with R3 being effective. So we have a delay in the readout of the data, right? We wait till the year is complete and then start gathering the data and and and eventually start sharing it. so in really the middle or latter part of this year, we'll have all of that data gathered and and in a way that potentially could feed a publication or presentation. So we are eager to see if the work that we've done in collaboration with FDA and EMA and other you know colleagues, technology companies too in the industry, if if that moves the needle um and using as much
Speaker: evidence, right? Leveraging data to show what's working, i think is key, not just because I'm a data person, but that's you know, that's evidence that's, we're in research, we we understand that concept.
Speaker: um We do, you know, have have ah view access to regulators. So we intend to continue to have um, discussions, share with them, give feedback onto any sort of other, um, publications that they're planning, any guidances that come out, um, and just continue to encourage greater visibility, and of the regulators themselves and, and to have conversations with them if they're in a forum where they can speak with them, because, uh, I think, you know, every little bit of encouragement helps.
Speaker: Um, and ultimately, ah someone from the agency had said at some point that they, they expect that the clinical trial industry is delivering clinical trials more efficiently, more effectively, um that there is a huge cost and there is a long runway for products to get to market.
Speaker: And they believe that these concepts that are, um, are summarized in R3 like proportionality, like taking a risk-based approach, um not doing things the way we've always done them, not depending on 100% STV, that those are ways to bring products to market quicker without compromising quality or safety. And so, you know, how, where, wherever we can get that concept across,
Speaker: and including showing the cost to run clinical trials, not just the cost savings of of central monitoring, but the overall cost and time, it it is mind blowing to me still how long it takes and how much it costs and where there's an opportunity to produce both of those and still have um the same positive, safe intended outcome is should be reasoned in itself for us to be involved.
Speaker: Yeah, that's a clear incentive to to implement it because, of course, everyone wants to bring drugs to market faster because the the the cost of delay is even greater than the cost of running the trial itself. So, yeah.
Speaker: Well, very, very interesting. Thank you. um So I've got one last question for you. ah What's the best piece of advice you've received that you consistently apply in your work?
Speaker: That's a great question. And linked to this concept really is focusing on what matters. and It's true in the industry, but it's also true in life. I'm a data person.
Speaker: So I can easily flip into everything needing to be perfect in line, correct. um But in reality, some things just are more important than others. And and in life, that may be values or you know where my time is spent.
Speaker: um Similarly for this industry, right we need to prioritize ah the things that add value um and be okay with this concept of of progress over perfection, right? We need to keep things moving forward.
Speaker: And it's okay if there's a data point that isn't perfect, if it has no impact on patient safety or on the outcome of the trial.
Speaker: And really, if we are in this to bring treatment to patients and and better um people's lives and and the treatment that's available then we all need to focus on that
Speaker: Wow, i'm I would have fallen on the floor if I wasn't already sat down. We've heard it that from a data scientist, it's okay if not all data points are perfect. I love that. You can call me. Thank you so much, Jennifer. It's been so nice talking to you. i really love the work you're doing with ACRO and I love hearing the the progress you're making.
Speaker: ah so yeah, thank you so much for sharing and joining me on the podcast. Sure thing. I appreciate you having me and giving me another forum to share what we found and advocate for taking a risk-based approach in clinical trials. And more importantly, just appreciate being able to connect with you and and look forward to whatever our next endeavor is.
Speaker: Thanks so much, Sylvan. Me too. And thank you all for watching or listening to us. You can find more clinical data talks episodes on our website.
