WEBVTT
00:00:00.200 --> 00:00:05.450
Hello and welcome everyone to SCRS Talks provided by the Society for Clinical Research Sites.
00:00:05.610 --> 00:00:16.670
I'm Jimmy Bethel, the chief site success officer with the Society, and today I am joined by my friend Catherine Gregor, who is the chief clinical trial officer at Florence Healthcare.
00:00:17.050 --> 00:00:23.449
And we're gonna talk today about Operation Trial Blazer, not to be confused with Trailblazer.
00:00:23.489 --> 00:00:37.929
Trial Blazer, and it's really a convergence around sponsor expectations and trying to bring together the site reality from the perspective of one of our stronger supporting service providers.
00:00:37.929 --> 00:00:43.619
But before we get into that, Catherine for our listeners, a little bit about you and maybe a little bit about Florence as well.
00:00:43.820 --> 00:00:44.259
Sure.
00:00:44.270 --> 00:00:45.030
Thanks for having me.
00:00:45.409 --> 00:01:00.267
I am a chief clinical trial officer at Florence, and for those of us who don't know Florence is a software provider that provides technology to better connect sites and sponsors to allow seamless document and data exchange, hopefully speeding up trials and getting us to cures faster.
00:01:00.466 --> 00:01:09.507
Before I joined Florence, I spent the majority of my career, which is about two decades now, at the site operations side, running clinical trial sites.
00:01:09.527 --> 00:01:13.527
Anything from standalone independent community sites to for-profit health centers.
00:01:13.917 --> 00:01:21.864
And the last appointment I had prior to joining Florence was I was the director of the Cancer Trials Office at Vanderbilt-Ingram Cancer Center here in Nashville, Tennessee.
00:01:22.438 --> 00:01:22.787
awesome.
00:01:22.798 --> 00:01:23.487
Thanks, Catherine.
00:01:23.528 --> 00:01:24.438
Thanks for being here.
00:01:24.628 --> 00:01:30.918
Like I had mentioned, Florence has been along for the ride with SCRS for several years now in a pretty direct capacity.
00:01:30.918 --> 00:01:31.977
I thank you for your support.
00:01:31.977 --> 00:01:34.367
And I'm excited to talk about Operation Trailblazer.
00:01:34.567 --> 00:01:37.737
we know Florence works with tens of thousands of sites.
00:01:37.737 --> 00:01:41.337
I think the statistic that I have is around 65,000 sites.
00:01:41.927 --> 00:01:55.947
So with those organizations, What are you hearing from them about this project, about this operation, and how does it maybe match up to what sponsors are talking about on their side as well, sitting in that place where you're bringing these two constituency groups together?
00:01:56.147 --> 00:01:57.576
Yeah, it's an interesting question.
00:01:57.876 --> 00:02:03.066
On the site side, I think there's a lot of interest and a lot of excitement around what is coming.
00:02:03.466 --> 00:02:13.757
The document that was released for Trailblazer itself points very specifically to academic centers, which I think are important to include in the operations where we're trying to figure out how to better speed trials.
00:02:14.106 --> 00:02:16.356
But it doesn't address the full site network.
00:02:16.426 --> 00:02:26.727
So I think there's some revisions coming and trying to prompt the agency to think about that if we want to be successful at changing how trials get done, we also have to look at where they're actually occurring and changing that dynamic as well.
00:02:27.026 --> 00:02:55.507
So most of my site constituents are waiting with bated breath to see what's gonna happen because this initiative, combined with the RFI you heard about earlier this year around how to optimize for phase one trials using AI and things like that, are pointing to this fact that the agency is looking for ways to make a digital data transfer easier because they know that's how we're gonna speed up trials is by getting rid of the paper and the people that have to manage that data, not getting rid of them, but better optimizing them to do cooler things like actually see patients.
00:02:55.825 --> 00:02:58.110
So I think sites Are waiting to see how that's gonna play out for them.
00:02:58.310 --> 00:03:01.849
Whereas sponsors, on the other hand, have a very clear directive, right?
00:03:01.860 --> 00:03:06.538
Figure out how to get the data to the agency faster so that they can make go/no decisions faster.
00:03:06.549 --> 00:03:14.330
So instead of waiting to full data lock, can you look at data signals coming out of your trial data to make a decision of whether or not a drug is likely to succeed?
00:03:14.699 --> 00:03:18.780
So they are much more tactically thinking about what do I need to do?
00:03:18.789 --> 00:03:20.729
How do I set up my systems?
00:03:20.729 --> 00:03:25.669
How do I partner with sites that have the ability to play in the current infrastructure?
00:03:25.759 --> 00:03:32.710
Digital infrastructure, access to cloud-based services, the ability to do integrations and APIs and things like that.
00:03:32.710 --> 00:03:36.449
And that naturally is gonna be your larger centers because they have capital and infrastructure there.
00:03:36.780 --> 00:03:49.479
But they too are also worried about if we continue to perpetuate this model of just going to the traditional academic model, that you're gonna repeat the same problems we've had in the past around, speed to enrollment, access to different types of patients, those sorts of things.
00:03:49.479 --> 00:03:57.530
So they're also trying to figure out how do we pick up the excitement of where we're headed without leaving a good portion of the, the site ecosystem behind.
00:03:57.729 --> 00:03:59.979
It's such an interesting situation that we're in.
00:03:59.979 --> 00:04:02.569
It's not necessarily novel, right?
00:04:02.569 --> 00:04:12.039
We've been talking around this or about this concept, but seeing it come to life and seeing, something succinctly like I try to say, it's like we talk about it, but someone needed to put a pin in it.
00:04:12.090 --> 00:04:14.819
And- and actually harden and"Hey, this is an issue.
00:04:14.909 --> 00:04:25.579
Here is a path that we can take that might yield some outcomes and solutions." Because we know, Catherine, sites have been asked to move faster and do more for a long time.
00:04:25.579 --> 00:04:33.250
That seems to be the constant trend here when we look at things like protocol complexity and speed around startup and trial execution.
00:04:33.439 --> 00:04:44.160
So what feels genuinely different here than versus maybe some things we've done in the past, and where does that pressure, kind of land first when, in terms of what we're talking about?
00:04:44.360 --> 00:04:46.139
Yeah it's another interesting question.
00:04:46.180 --> 00:04:58.160
So in my seat and from my lens like having sat at the site seat for a long time, having sat in technology now and talking to sponsors in both of those roles, there's a lot of concern about we all know that we need to do things faster.
00:04:58.430 --> 00:05:02.649
We know that we need the sponsors to operate a different way when they design their protocols.
00:05:02.649 --> 00:05:06.550
We know that we need the sites to be able to operate in a new model that is much more tech-enabled.
00:05:06.750 --> 00:05:15.129
But there is a lot of concern about what we can do and not do from a regulatory standpoint because guidances for what they are guidances, and they're intentionally written to be broad and vague.
00:05:15.449 --> 00:05:18.100
And what that means that there's a whole wide spectrum of interpretation.
00:05:18.300 --> 00:05:29.350
And so when something comes out that it's not clearly says,"Do this one thing and this other thing will happen," sites and sponsors then are stuck trying to figure out what, where the lanes are, and we have different definitions of risk.
00:05:29.750 --> 00:05:39.750
So a lot of the process and waste, I think in trial startup in particular, and then also in conduct, is trying to mitigate this perception of risk and whether or not we're gonna get in trouble, right?
00:05:39.800 --> 00:05:41.963
We talk about we can decentralized clinical trials.
00:05:41.963 --> 00:05:43.523
We can do technology.
00:05:43.523 --> 00:05:47.843
We can do different types of technology to collect the same data and accept the variability.
00:05:48.043 --> 00:05:56.762
That's all been said, but everybody wants to know, like, how far we can deviate from the norm in both of those, and the agency to date has never given the boundaries, right?
00:05:57.153 --> 00:06:13.963
And so when I talk to the agency and say, they're like,"We're giving you all these options to do these really innovative things, and sites aren't picking it up, and they're pushing back." I'm like sites are pushing back'cause they're afraid to get in trouble from you, FDA." They're like we tell the sponsors it's okay." And I was like the sponsors aren't picking it up either because they're afraid you're not gonna approve their molecule.
00:06:14.163 --> 00:06:20.012
So you agency have to come out with very, not prescriptive, but clear statement that says,'We want you to do these things.
00:06:20.012 --> 00:06:36.031
We think it is the right path, and there's going to be some acceptance of variability in the data, and we're willing to work with you on figuring out what that's gonna look like in the move forward.'" For me, this year in particular is the first time you've seen the agency really come out with, okay, the HHS is pushing this down.
00:06:36.031 --> 00:06:45.011
They're saying that this is across OHRP, this is across FDA, this is across HHS We have to figure out how to go faster, and that means that we need to not be afraid of breaking things while we're doing it.
00:06:45.211 --> 00:07:00.641
And so I think that is a shift in the dialogue that is important and I'm hoping continues because that's actually how we're going to move past this kind of stalemate of whose job it is to do what and why we have to take all these extra steps'cause I think most of those steps are tied to risk mitigation and not actual patient safety
00:07:00.841 --> 00:07:02.290
A lot of excellent points, Catherine.
00:07:02.290 --> 00:07:08.600
A- I actually just had a conversation with someone at the FDA where we were talking a little bit about something along these lines.
00:07:08.600 --> 00:07:16.730
It was in terms of protocol complexity and involving research sites, and this individual was almost confused as to why this doesn't happen.
00:07:17.050 --> 00:07:22.281
We've- Seen it happen with silos where we bring sites in to advise on protocol development.
00:07:22.331 --> 00:07:30.331
That, to your exact point, that takes a little bit of time to do at the beginning when we're trying to move the fastest.
00:07:30.382 --> 00:07:35.291
And so sponsors aren't, we'll say they don't feel the need to do that in a lot of cases.
00:07:35.291 --> 00:07:37.482
They proceed with their standard processes.
00:07:37.531 --> 00:07:45.230
But again, this person was confused as to why that doesn't happen, and my answer to this individual was exactly the same as yours was because they don't have any pressure to do right?
00:07:45.307 --> 00:07:52.327
the FDA is not asking or even mandating or suggesting in some ways that they do this so they're not going to do it.
00:07:52.478 --> 00:07:58.887
And then, like you said, there's little incentivization on the site side to participate in that activity either.
00:07:58.927 --> 00:08:03.918
No one has really done the work to understand what is the brass tacks around that such action?
00:08:03.918 --> 00:08:05.937
So it speaks very much to your example.
00:08:06.137 --> 00:08:06.887
Yeah, absolutely.
00:08:07.048 --> 00:08:08.206
It's just clarity, right?
00:08:08.300 --> 00:08:21.608
I sat across from Commissioner Mackary last winter before his departure and he's"I don't understand, why can't we go faster?" And he's like,"We publish, The results on people who mess up." And I was like Publishing the results on the people who get it wrong isn't exactly the best tendency.
00:08:21.608 --> 00:08:25.668
You should publish the results on the people who've done innovative approaches and gotten their approval," right?
00:08:25.668 --> 00:08:26.728
You have that access.
00:08:26.769 --> 00:08:30.928
You can open up that lane by telling us"Look, this is okay." And he seemed...
00:08:30.939 --> 00:08:34.879
That seemed to resonate with him, and maybe that's part of what, my little nudge is what changed the world.
00:08:34.989 --> 00:08:38.624
But having conversations like that, it's like it's not,"Don't tell us how we're gonna get in trouble.
00:08:38.634 --> 00:08:42.114
Tell us how we're gonna succeed using these new paths that you've created for us."
00:08:42.313 --> 00:08:42.663
Yeah.
00:08:42.724 --> 00:08:43.423
And that's tough...
00:08:43.443 --> 00:08:51.224
We all know that in this industry, from the top down FDA sponsored, doesn't matter who it is, that's a tough situation for us to overcome.
00:08:51.514 --> 00:08:55.053
But I wanna talk a little bit more about data here and some outcomes, Catherine.
00:08:55.053 --> 00:09:07.124
And I know your data shows, from what I've seen, that CRAs spend anywhere from 40 to 60% of the visit time on admin related tasks- instead of the actual, clinical monitoring, right?
00:09:07.144 --> 00:09:11.224
The things that, the important part of what the CRAs need to do.
00:09:11.403 --> 00:09:12.933
So what does that look like then?
00:09:12.964 --> 00:09:15.073
How does that, affect the sites?
00:09:15.144 --> 00:09:20.384
And who ends up absorbing some of that extra work or some of the tasks related to that?
00:09:20.583 --> 00:09:21.033
Yeah.
00:09:21.043 --> 00:09:30.124
I I think it looks like for the sites when your monitor shows up and hasn't reviewed the reg binder, so they spend a day and a half going through the reg binder'cause it's enormous.
00:09:30.323 --> 00:09:31.364
And then they have...
00:09:31.423 --> 00:09:36.644
But they're only there for two days, so then they spend the last six hours of their visits looking at patient data, which is actually the valuable point, right?
00:09:36.653 --> 00:09:38.193
Look at the patient data, tell me where there's trends.
00:09:38.394 --> 00:09:43.014
There's also, trying to pull files, and show and explain where to find things.
00:09:43.014 --> 00:09:56.543
And you just spend a lot of time, I don't wanna say babysitting'cause it makes it seem infantile, but you spend a lot of time showing people where things are and talking about why things are the way they are and how they're filed, rather than what the data says and what's happening within the study conduct itself.
00:09:56.884 --> 00:10:01.864
Also, chasing physicians and PIs for meetings is a waste of time for both the monitor and the coordinator as well.
00:10:02.264 --> 00:10:14.433
So in my team at Vanderbilt, I know it was my data team when they had heavy monitor days got slowed down because they had to spend time with the monitors, which means they weren't entering data which then puts them behind on data lock, and then it becomes a whole vicious circle trying to catch that up.
00:10:14.734 --> 00:10:16.053
And same with my reg coordinators.
00:10:16.104 --> 00:10:25.153
They spend a lot of time working through the binders and explaining where things are and how and what the status of things are in process rather than doing regulatory work, which is where I want them.
00:10:25.354 --> 00:10:27.764
And so I think it's kind of a waste to both parties, right?
00:10:27.764 --> 00:10:33.219
The CRA wants to have the in-depth conversations with the physician about these are the trends that we're seeing across your study team.
00:10:33.229 --> 00:10:35.030
These are the things in your patient profile.
00:10:35.041 --> 00:10:36.600
What can we do to help with enrollment?
00:10:36.620 --> 00:10:47.441
Those are value-based conversations that I think that team would rather be having as well, instead of trying to find somebody to show them where a data point lives in Epic or what piece of paper got misfiled in a different binder.
00:10:47.640 --> 00:10:48.010
Yeah.
00:10:48.010 --> 00:10:50.010
It's an interesting perspective.
00:10:50.010 --> 00:10:51.073
and a reality.
00:10:51.354 --> 00:10:54.543
An unfortunate reality of where we sit, and that's it's exactly right.
00:10:54.563 --> 00:11:00.859
When we spend all this time why can't we take care of some of this stuff before, right?
00:11:00.859 --> 00:11:07.969
Or- I think digitization, that's I would hope has enabled us to do this, and you would know succinctly on this, Catherine.
00:11:08.149 --> 00:11:12.818
But what they should be doing on site is the things that they can't do remotely, to your point.
00:11:12.869 --> 00:11:23.889
Your exact point, and that is having the conversations with the sites reviewing where the patient-related data have seen Again, like you said, exactly seeing those trends alerting to safety.
00:11:23.889 --> 00:11:34.019
But it gets It's important that it doesn't get lost in the, in translation or buried beneath this mound of admin work that they need to do and then, oh, shoot, I have to do SDV.
00:11:34.178 --> 00:11:39.099
And they're clamoring through and burning through the patient stuff, which is actually the most important thing.
00:11:39.099 --> 00:11:51.808
So it is such an interesting place that we've found ourselves in because then that burden does that burden, how much of that burden, I should say then, gets pushed off onto the site itself and the other part of that is in an uncompensated manner
00:11:52.009 --> 00:12:01.948
And I think, and the other thing that's a barrier with monitoring is there's some barrier to acceptance of remote monitoring because the data itself is not natively digital at all sites, right?
00:12:01.948 --> 00:12:07.178
There's still sites that are working in paper, and then they have to scan that paper and upload it in order to make it digital.
00:12:07.178 --> 00:12:08.019
I use air quotes.
00:12:08.038 --> 00:12:09.609
You can't see them, but they're effective.
00:12:09.969 --> 00:12:11.208
And then the monitor...
00:12:11.229 --> 00:12:12.778
So the monitor can review it remotely.
00:12:12.798 --> 00:12:23.139
That's time waste on the site to scan in paper to enable a faster review on the monitor side, which inevitably, I do think saves some time at the visit, but it's burden doesn't go away, right?
00:12:23.139 --> 00:12:23.948
It just shifts.
00:12:24.149 --> 00:12:26.379
So we have to think about it from both ends.
00:12:26.379 --> 00:12:33.519
How do we get data natively digital from the beginning so that it can be transferred back and reviewed remotely by any interested stakeholder, not just monitoring.
00:12:33.519 --> 00:12:35.479
Also, your physicians can review data remotely.
00:12:35.479 --> 00:12:38.778
They don't have to wait for you to bring them the patient's source binders and things like that.
00:12:38.979 --> 00:12:46.229
And then you start to create more efficient use of the time that you do have to pay attention to what's in front of you when you can do it remotely.
00:12:46.428 --> 00:12:49.798
And that's what technology should be doing, right?
00:12:49.828 --> 00:12:56.349
That's- a perfectly good use case of a positive value add when it comes to decentralization and technology.
00:12:56.349 --> 00:12:58.609
And when it's done right finger, fingers crossed.
00:12:58.649 --> 00:13:03.438
Again you're close to that space that we can start to see some of those efficiencies play themselves out.
00:13:03.589 --> 00:13:19.259
another statistic here, Catherine, is 95% of sites we see still rely on email-based document exchange, and nearly two-thirds then say that sponsor technology just really just duplicates something that they already have or that they're already doing.
00:13:19.318 --> 00:13:29.009
And it becomes a, it becomes an isolated duplicative step outside of their process when it really doesn't have any time savings or value add for them is another way to say that.
00:13:29.009 --> 00:13:36.269
So if that fragmentation and things like that just as one example, you and I both know that's probably just one thing, is the real bottleneck.
00:13:36.278 --> 00:13:44.739
What has to then change on the sponsor side for something like TrialBlazer to actually reduce burden instead of adding to it?
00:13:44.938 --> 00:13:46.278
It has to happen in two places.
00:13:46.278 --> 00:13:49.558
It has to happen on the sponsor side, and it has to happen in the technology space.
00:13:49.558 --> 00:13:51.619
I wanna call vendors to the carpet on this one as well.
00:13:51.629 --> 00:13:53.688
Might, as somebody who operates in that space.
00:13:54.068 --> 00:13:54.879
It's two things, right?
00:13:54.879 --> 00:14:02.058
So the sites who've invested in technology for themselves, they have native digital, they have a workflow, they're in the tool, the data is there.
00:14:02.068 --> 00:14:13.269
We have to make it easier for that data to move from their system into the sponsor system so that the sponsors can aggregate it, which means that sponsors have to be accepting of data records existing outside of their control, right?
00:14:13.298 --> 00:14:17.259
If I have an eBinder or I have an eSource, or even my EMR, right?
00:14:17.259 --> 00:14:23.198
That data is sufficient for clinical care and for document retention from an FDA standpoint for the site.
00:14:23.198 --> 00:14:26.261
Why is it a different standard for the sponsor So that's one thing.
00:14:26.272 --> 00:14:29.581
And then I think it is on the vendors.
00:14:29.611 --> 00:14:40.172
The heaviest lift is the problem, is that we don't make it easy for data to be portable between systems, which means that if we don't do that, we're just, again, moving burden from one side to the other.
00:14:40.182 --> 00:14:53.201
So it's either the monitor has to operate in different systems and spend time moving in and out and moving documents between site systems into theirs, or the sites have to do it on their end and take between sponsor systems to move it into their system.
00:14:53.402 --> 00:14:57.392
And in the past, we know it's supposed to have been API, and can we create APIs?
00:14:57.392 --> 00:15:04.631
And APIs are great, but APIs have to be configured which is not easy for everybody to do, and they have to have mutual customers, which is a whole other thing.
00:15:04.831 --> 00:15:15.961
But I think what's really gonna open up for us is AI and the ability to read data and audit trails and move data more seamlessly without an API, so MCPs, things like that will hopefully start to reduce some of that friction.
00:15:15.961 --> 00:15:21.902
Because until that happens, we are basically asking one side to absorb work at the expense of the other
00:15:22.101 --> 00:15:30.032
Well, and it's such an interesting, a-another interesting, perspective and fact really with whose burden is it, right?
00:15:30.042 --> 00:15:36.162
And this concept of the sponsor could do it or the site could do it to, as in an example of what you just talked about.
00:15:36.172 --> 00:15:38.341
So whose ultimate responsibility is that?
00:15:38.361 --> 00:15:49.751
when is it appropriate for the site to be, we'll say, taken from arguably more important tasks, and then into things like that.
00:15:49.831 --> 00:15:55.562
And, we don't need to get into it today, Catherine, because I, it's probably another 30-plus minute conversation.
00:15:55.562 --> 00:16:04.422
But then where does AI come into that equation to ease some of that administrative or execution burden on the site and the sponsor, right?
00:16:04.422 --> 00:16:05.142
It kinda takes...
00:16:05.162 --> 00:16:08.981
it's a third party in a way that kinda takes that burden off of both sides.
00:16:09.182 --> 00:16:09.731
Exactly.
00:16:09.782 --> 00:16:11.471
I literally just had this conversation this morning.
00:16:11.772 --> 00:16:20.596
My site heart wants to say that it's the sponsor's responsibility to absorb the burden because they have the resources and the money and more- to win out of it than the site does, right?
00:16:20.796 --> 00:16:29.245
But then I also have my empathetic sponsor hat on of for the smaller pharma and biotechs, like they're just trying to get to the end of the day with their single molecule and make something happen.
00:16:29.245 --> 00:16:31.216
And so we're also asking them to pick it up.
00:16:31.546 --> 00:16:35.275
But I think it's an unfair paradox that we have to ask anybody to pick it up, to your point.
00:16:35.275 --> 00:16:36.635
So I think AI is...
00:16:36.706 --> 00:16:40.734
my true hope for AI is that it can solve the data movement problem.
00:16:41.053 --> 00:16:50.333
Because if we can query databases, and we can look at blinded clean rooms of data and query against it without ever having to ask a question of a site and without de-identifying that site.
00:16:50.364 --> 00:16:58.743
We could be able to know what's happening by surfacing some of those signals and audit trails in a way that is safe and meets data privacy and all those other things.
00:16:59.373 --> 00:17:08.271
we need to lean into that more as an industry and figure out how to make AI and the cloud work for us in a way that APIs have not in the past to, to get that through.
00:17:08.836 --> 00:17:09.777
I couldn't agree more.
00:17:09.936 --> 00:17:14.221
And again, I think it's maybe a conversation for us to set up in another time.
00:17:14.250 --> 00:17:18.750
I'm sure we'll end up talking about it and being in the same room when those conversations continue to happen.
00:17:18.750 --> 00:17:45.080
But as we move to start to close our conversation here, Catherine my final question is as you've moved from ETMF pass rates from a, a place of 65% to nearly 99% across the network of sites that you have in the technology system, With over 173 billion in research assets licensed out to China last year- the question isn't just whether the US can compete on incentives.
00:17:45.280 --> 00:17:51.961
Where and how can sites absorb the volume if conditions really showed up tomorrow?
00:17:52.020 --> 00:17:54.060
If things, started to shift for us?
00:17:54.260 --> 00:18:04.661
So the move with the volume of work that's moved out to China, it's moving there because of the speed of enrollment, which means that they've optimized their health networks to enroll patients at scale, right?
00:18:04.980 --> 00:18:05.951
We don't do that here.
00:18:06.131 --> 00:18:14.550
I think that's not gonna change for us, so I think it's going to be more sites doing more studies and still having smaller portions of patients to manage across them in order to beat speed.
00:18:14.750 --> 00:18:21.500
But it is going to be leveraging technology and AI to help the people on the ground in front of those patients move faster.
00:18:21.500 --> 00:18:28.633
So eTMF pass rates means that basically the site has put in a document, they're able to get any queries resolved in real time.
00:18:28.633 --> 00:18:31.192
We're not waiting six weeks between monitor visits.
00:18:31.202 --> 00:18:32.502
They can see it electronically.
00:18:32.833 --> 00:18:39.002
Also, we're creating prompts to tell sites when things might be wrong, so we're not waiting for a monitor to expose those.
00:18:39.012 --> 00:18:45.073
We're pointing it out to the coordinator"Hey, you scanned this page in backwards on page six of this 20-page PDF.
00:18:45.272 --> 00:18:49.282
Hey, this consent form doesn't match the language in your contract.
00:18:49.333 --> 00:18:52.603
Hey, this signature doesn't match the name on the CV." Things like that.
00:18:52.633 --> 00:19:00.833
And all of that, they're small little things, but those small little things take up time and data cleaning, and if we can remove all of that, the documents move through faster.
00:19:00.853 --> 00:19:09.962
That's a signal that we are making the people doing the end work more effective and smarter by giving them cues of where things have opportunity or errors have occurred.
00:19:10.002 --> 00:19:14.573
Because we're all human and we're moving very fast in order to keep up with the demand of delivering clinical care.
00:19:14.772 --> 00:19:21.292
And so we're trying to make it easier for those mistakes to get identified and corrected immediately rather than prolonging themselves.
00:19:21.612 --> 00:19:24.175
So I think things like that will help with our speed.
00:19:24.576 --> 00:19:32.955
I think optimizing a trial selection based on actual performance data of the site, not just reported performance or historical performance data.
00:19:33.155 --> 00:19:40.096
The one thing I always hated is people would come to my site and say it took you 260 days to do this study." The last time I did a study with you, it was 10 years ago.
00:19:40.105 --> 00:19:41.986
You wanna know how I'm, how I can do it today, right?
00:19:41.986 --> 00:19:42.276
I don't...
00:19:42.675 --> 00:19:44.536
We want sponsors to have real-time data.
00:19:44.566 --> 00:19:53.155
We want sites to know, be able to talk about what their performance is, and we can select sites that actually can meet criteria and have the ability to execute on the trials that are coming down.
00:19:53.226 --> 00:19:55.796
And I think those two things will help make us more competitive
00:19:55.996 --> 00:19:57.226
I couldn't agree more.
00:19:57.276 --> 00:19:59.365
It's about the context too Catherine?
00:19:59.486 --> 00:20:19.346
Being able to help the, whoever's on the other end of that information understand effectively what that means to your, a, a great example that you just gave there around that site selection process is really important because performance is, I don't want to say arbitrary.
00:20:19.481 --> 00:20:30.321
there's a large portion of it that is not, but there's a portion of it that does oftentimes need understanding and conversation and additional context to really understand what the entire picture of that means.
00:20:30.321 --> 00:20:34.451
And we're moving so fast, like you said, that sometimes that context gets lost.
00:20:34.451 --> 00:20:38.271
But I think it's a great place for us to conclude our conversation.
00:20:38.561 --> 00:20:43.832
A lot of really interesting points, Catherine, that you brought up here, things for us to really think about as an industry.
00:20:43.832 --> 00:21:01.301
I'm excited to hear and see where Operation Trailblazer continues to go as, your organization sits at a purposeful crux to move some of these challenges in the right direction, and like I said, really put a pin in them so that we can start to make change as a result.
00:21:01.501 --> 00:21:02.261
Yeah, absolutely.
00:21:02.261 --> 00:21:04.551
I'm excited as well, and I am hopeful.
00:21:04.551 --> 00:21:09.852
I remain the eternal optimist that we will figure this out and get out of our own way and go faster.
00:21:09.882 --> 00:21:10.761
So here's hoping.
00:21:10.862 --> 00:21:11.961
yeah, here's to hoping.
00:21:11.961 --> 00:21:13.551
That, I couldn't agree more with you, Catherine.
00:21:13.701 --> 00:21:14.271
So thank you.
00:21:14.511 --> 00:21:18.291
Thanks again for being with us here today and and for participating in the conversation.
00:21:18.491 --> 00:21:18.991
Thanks, Jimmy.
00:21:18.991 --> 00:21:19.571
I appreciate it.
00:21:19.771 --> 00:21:29.582
And for everyone that's listening, be sure to check out other site-focused resources made available to you and the entire community on our website, myscrs.org.
00:21:29.612 --> 00:21:40.942
This includes additional podcast episodes, website webinars rather, and a plethora of thought leadership opportunities Thanks again for listening, for tuning in, and until next time
00:00:00.200 --> 00:00:05.450
Hello and welcome everyone to SCRS Talks provided by the Society for Clinical Research Sites.
00:00:05.610 --> 00:00:16.670
I'm Jimmy Bethel, the chief site success officer with the Society, and today I am joined by my friend Catherine Gregor, who is the chief clinical trial officer at Florence Healthcare.
00:00:17.050 --> 00:00:23.449
And we're gonna talk today about Operation Trial Blazer, not to be confused with Trailblazer.
00:00:23.489 --> 00:00:37.929
Trial Blazer, and it's really a convergence around sponsor expectations and trying to bring together the site reality from the perspective of one of our stronger supporting service providers.
00:00:37.929 --> 00:00:43.619
But before we get into that, Catherine for our listeners, a little bit about you and maybe a little bit about Florence as well.
00:00:43.820 --> 00:00:44.259
Sure.
00:00:44.270 --> 00:00:45.030
Thanks for having me.
00:00:45.409 --> 00:01:00.267
I am a chief clinical trial officer at Florence, and for those of us who don't know Florence is a software provider that provides technology to better connect sites and sponsors to allow seamless document and data exchange, hopefully speeding up trials and getting us to cures faster.
00:01:00.466 --> 00:01:09.507
Before I joined Florence, I spent the majority of my career, which is about two decades now, at the site operations side, running clinical trial sites.
00:01:09.527 --> 00:01:13.527
Anything from standalone independent community sites to for-profit health centers.
00:01:13.917 --> 00:01:21.864
And the last appointment I had prior to joining Florence was I was the director of the Cancer Trials Office at Vanderbilt-Ingram Cancer Center here in Nashville, Tennessee.
00:01:22.438 --> 00:01:22.787
awesome.
00:01:22.798 --> 00:01:23.487
Thanks, Catherine.
00:01:23.528 --> 00:01:24.438
Thanks for being here.
00:01:24.628 --> 00:01:30.918
Like I had mentioned, Florence has been along for the ride with SCRS for several years now in a pretty direct capacity.
00:01:30.918 --> 00:01:31.977
I thank you for your support.
00:01:31.977 --> 00:01:34.367
And I'm excited to talk about Operation Trailblazer.
00:01:34.567 --> 00:01:37.737
we know Florence works with tens of thousands of sites.
00:01:37.737 --> 00:01:41.337
I think the statistic that I have is around 65,000 sites.
00:01:41.927 --> 00:01:55.947
So with those organizations, What are you hearing from them about this project, about this operation, and how does it maybe match up to what sponsors are talking about on their side as well, sitting in that place where you're bringing these two constituency groups together?
00:01:56.147 --> 00:01:57.576
Yeah, it's an interesting question.
00:01:57.876 --> 00:02:03.066
On the site side, I think there's a lot of interest and a lot of excitement around what is coming.
00:02:03.466 --> 00:02:13.757
The document that was released for Trailblazer itself points very specifically to academic centers, which I think are important to include in the operations where we're trying to figure out how to better speed trials.
00:02:14.106 --> 00:02:16.356
But it doesn't address the full site network.
00:02:16.426 --> 00:02:26.727
So I think there's some revisions coming and trying to prompt the agency to think about that if we want to be successful at changing how trials get done, we also have to look at where they're actually occurring and changing that dynamic as well.
00:02:27.026 --> 00:02:55.507
So most of my site constituents are waiting with bated breath to see what's gonna happen because this initiative, combined with the RFI you heard about earlier this year around how to optimize for phase one trials using AI and things like that, are pointing to this fact that the agency is looking for ways to make a digital data transfer easier because they know that's how we're gonna speed up trials is by getting rid of the paper and the people that have to manage that data, not getting rid of them, but better optimizing them to do cooler things like actually see patients.
00:02:55.825 --> 00:02:58.110
So I think sites Are waiting to see how that's gonna play out for them.
00:02:58.310 --> 00:03:01.849
Whereas sponsors, on the other hand, have a very clear directive, right?
00:03:01.860 --> 00:03:06.538
Figure out how to get the data to the agency faster so that they can make go/no decisions faster.
00:03:06.549 --> 00:03:14.330
So instead of waiting to full data lock, can you look at data signals coming out of your trial data to make a decision of whether or not a drug is likely to succeed?
00:03:14.699 --> 00:03:18.780
So they are much more tactically thinking about what do I need to do?
00:03:18.789 --> 00:03:20.729
How do I set up my systems?
00:03:20.729 --> 00:03:25.669
How do I partner with sites that have the ability to play in the current infrastructure?
00:03:25.759 --> 00:03:32.710
Digital infrastructure, access to cloud-based services, the ability to do integrations and APIs and things like that.
00:03:32.710 --> 00:03:36.449
And that naturally is gonna be your larger centers because they have capital and infrastructure there.
00:03:36.780 --> 00:03:49.479
But they too are also worried about if we continue to perpetuate this model of just going to the traditional academic model, that you're gonna repeat the same problems we've had in the past around, speed to enrollment, access to different types of patients, those sorts of things.
00:03:49.479 --> 00:03:57.530
So they're also trying to figure out how do we pick up the excitement of where we're headed without leaving a good portion of the, the site ecosystem behind.
00:03:57.729 --> 00:03:59.979
It's such an interesting situation that we're in.
00:03:59.979 --> 00:04:02.569
It's not necessarily novel, right?
00:04:02.569 --> 00:04:12.039
We've been talking around this or about this concept, but seeing it come to life and seeing, something succinctly like I try to say, it's like we talk about it, but someone needed to put a pin in it.
00:04:12.090 --> 00:04:14.819
And- and actually harden and"Hey, this is an issue.
00:04:14.909 --> 00:04:25.579
Here is a path that we can take that might yield some outcomes and solutions." Because we know, Catherine, sites have been asked to move faster and do more for a long time.
00:04:25.579 --> 00:04:33.250
That seems to be the constant trend here when we look at things like protocol complexity and speed around startup and trial execution.
00:04:33.439 --> 00:04:44.160
So what feels genuinely different here than versus maybe some things we've done in the past, and where does that pressure, kind of land first when, in terms of what we're talking about?
00:04:44.360 --> 00:04:46.139
Yeah it's another interesting question.
00:04:46.180 --> 00:04:58.160
So in my seat and from my lens like having sat at the site seat for a long time, having sat in technology now and talking to sponsors in both of those roles, there's a lot of concern about we all know that we need to do things faster.
00:04:58.430 --> 00:05:02.649
We know that we need the sponsors to operate a different way when they design their protocols.
00:05:02.649 --> 00:05:06.550
We know that we need the sites to be able to operate in a new model that is much more tech-enabled.
00:05:06.750 --> 00:05:15.129
But there is a lot of concern about what we can do and not do from a regulatory standpoint because guidances for what they are guidances, and they're intentionally written to be broad and vague.
00:05:15.449 --> 00:05:18.100
And what that means that there's a whole wide spectrum of interpretation.
00:05:18.300 --> 00:05:29.350
And so when something comes out that it's not clearly says,"Do this one thing and this other thing will happen," sites and sponsors then are stuck trying to figure out what, where the lanes are, and we have different definitions of risk.
00:05:29.750 --> 00:05:39.750
So a lot of the process and waste, I think in trial startup in particular, and then also in conduct, is trying to mitigate this perception of risk and whether or not we're gonna get in trouble, right?
00:05:39.800 --> 00:05:41.963
We talk about we can decentralized clinical trials.
00:05:41.963 --> 00:05:43.523
We can do technology.
00:05:43.523 --> 00:05:47.843
We can do different types of technology to collect the same data and accept the variability.
00:05:48.043 --> 00:05:56.762
That's all been said, but everybody wants to know, like, how far we can deviate from the norm in both of those, and the agency to date has never given the boundaries, right?
00:05:57.153 --> 00:06:13.963
And so when I talk to the agency and say, they're like,"We're giving you all these options to do these really innovative things, and sites aren't picking it up, and they're pushing back." I'm like sites are pushing back'cause they're afraid to get in trouble from you, FDA." They're like we tell the sponsors it's okay." And I was like the sponsors aren't picking it up either because they're afraid you're not gonna approve their molecule.
00:06:14.163 --> 00:06:20.012
So you agency have to come out with very, not prescriptive, but clear statement that says,'We want you to do these things.
00:06:20.012 --> 00:06:36.031
We think it is the right path, and there's going to be some acceptance of variability in the data, and we're willing to work with you on figuring out what that's gonna look like in the move forward.'" For me, this year in particular is the first time you've seen the agency really come out with, okay, the HHS is pushing this down.
00:06:36.031 --> 00:06:45.011
They're saying that this is across OHRP, this is across FDA, this is across HHS We have to figure out how to go faster, and that means that we need to not be afraid of breaking things while we're doing it.
00:06:45.211 --> 00:07:00.641
And so I think that is a shift in the dialogue that is important and I'm hoping continues because that's actually how we're going to move past this kind of stalemate of whose job it is to do what and why we have to take all these extra steps'cause I think most of those steps are tied to risk mitigation and not actual patient safety
00:07:00.841 --> 00:07:02.290
A lot of excellent points, Catherine.
00:07:02.290 --> 00:07:08.600
A- I actually just had a conversation with someone at the FDA where we were talking a little bit about something along these lines.
00:07:08.600 --> 00:07:16.730
It was in terms of protocol complexity and involving research sites, and this individual was almost confused as to why this doesn't happen.
00:07:17.050 --> 00:07:22.281
We've- Seen it happen with silos where we bring sites in to advise on protocol development.
00:07:22.331 --> 00:07:30.331
That, to your exact point, that takes a little bit of time to do at the beginning when we're trying to move the fastest.
00:07:30.382 --> 00:07:35.291
And so sponsors aren't, we'll say they don't feel the need to do that in a lot of cases.
00:07:35.291 --> 00:07:37.482
They proceed with their standard processes.
00:07:37.531 --> 00:07:45.230
But again, this person was confused as to why that doesn't happen, and my answer to this individual was exactly the same as yours was because they don't have any pressure to do right?
00:07:45.307 --> 00:07:52.327
the FDA is not asking or even mandating or suggesting in some ways that they do this so they're not going to do it.
00:07:52.478 --> 00:07:58.887
And then, like you said, there's little incentivization on the site side to participate in that activity either.
00:07:58.927 --> 00:08:03.918
No one has really done the work to understand what is the brass tacks around that such action?
00:08:03.918 --> 00:08:05.937
So it speaks very much to your example.
00:08:06.137 --> 00:08:06.887
Yeah, absolutely.
00:08:07.048 --> 00:08:08.206
It's just clarity, right?
00:08:08.300 --> 00:08:21.608
I sat across from Commissioner Mackary last winter before his departure and he's"I don't understand, why can't we go faster?" And he's like,"We publish, The results on people who mess up." And I was like Publishing the results on the people who get it wrong isn't exactly the best tendency.
00:08:21.608 --> 00:08:25.668
You should publish the results on the people who've done innovative approaches and gotten their approval," right?
00:08:25.668 --> 00:08:26.728
You have that access.
00:08:26.769 --> 00:08:30.928
You can open up that lane by telling us"Look, this is okay." And he seemed...
00:08:30.939 --> 00:08:34.879
That seemed to resonate with him, and maybe that's part of what, my little nudge is what changed the world.
00:08:34.989 --> 00:08:38.624
But having conversations like that, it's like it's not,"Don't tell us how we're gonna get in trouble.
00:08:38.634 --> 00:08:42.114
Tell us how we're gonna succeed using these new paths that you've created for us."
00:08:42.313 --> 00:08:42.663
Yeah.
00:08:42.724 --> 00:08:43.423
And that's tough...
00:08:43.443 --> 00:08:51.224
We all know that in this industry, from the top down FDA sponsored, doesn't matter who it is, that's a tough situation for us to overcome.
00:08:51.514 --> 00:08:55.053
But I wanna talk a little bit more about data here and some outcomes, Catherine.
00:08:55.053 --> 00:09:07.124
And I know your data shows, from what I've seen, that CRAs spend anywhere from 40 to 60% of the visit time on admin related tasks- instead of the actual, clinical monitoring, right?
00:09:07.144 --> 00:09:11.224
The things that, the important part of what the CRAs need to do.
00:09:11.403 --> 00:09:12.933
So what does that look like then?
00:09:12.964 --> 00:09:15.073
How does that, affect the sites?
00:09:15.144 --> 00:09:20.384
And who ends up absorbing some of that extra work or some of the tasks related to that?
00:09:20.583 --> 00:09:21.033
Yeah.
00:09:21.043 --> 00:09:30.124
I I think it looks like for the sites when your monitor shows up and hasn't reviewed the reg binder, so they spend a day and a half going through the reg binder'cause it's enormous.
00:09:30.323 --> 00:09:31.364
And then they have...
00:09:31.423 --> 00:09:36.644
But they're only there for two days, so then they spend the last six hours of their visits looking at patient data, which is actually the valuable point, right?
00:09:36.653 --> 00:09:38.193
Look at the patient data, tell me where there's trends.
00:09:38.394 --> 00:09:43.014
There's also, trying to pull files, and show and explain where to find things.
00:09:43.014 --> 00:09:56.543
And you just spend a lot of time, I don't wanna say babysitting'cause it makes it seem infantile, but you spend a lot of time showing people where things are and talking about why things are the way they are and how they're filed, rather than what the data says and what's happening within the study conduct itself.
00:09:56.884 --> 00:10:01.864
Also, chasing physicians and PIs for meetings is a waste of time for both the monitor and the coordinator as well.
00:10:02.264 --> 00:10:14.433
So in my team at Vanderbilt, I know it was my data team when they had heavy monitor days got slowed down because they had to spend time with the monitors, which means they weren't entering data which then puts them behind on data lock, and then it becomes a whole vicious circle trying to catch that up.
00:10:14.734 --> 00:10:16.053
And same with my reg coordinators.
00:10:16.104 --> 00:10:25.153
They spend a lot of time working through the binders and explaining where things are and how and what the status of things are in process rather than doing regulatory work, which is where I want them.
00:10:25.354 --> 00:10:27.764
And so I think it's kind of a waste to both parties, right?
00:10:27.764 --> 00:10:33.219
The CRA wants to have the in-depth conversations with the physician about these are the trends that we're seeing across your study team.
00:10:33.229 --> 00:10:35.030
These are the things in your patient profile.
00:10:35.041 --> 00:10:36.600
What can we do to help with enrollment?
00:10:36.620 --> 00:10:47.441
Those are value-based conversations that I think that team would rather be having as well, instead of trying to find somebody to show them where a data point lives in Epic or what piece of paper got misfiled in a different binder.
00:10:47.640 --> 00:10:48.010
Yeah.
00:10:48.010 --> 00:10:50.010
It's an interesting perspective.
00:10:50.010 --> 00:10:51.073
and a reality.
00:10:51.354 --> 00:10:54.543
An unfortunate reality of where we sit, and that's it's exactly right.
00:10:54.563 --> 00:11:00.859
When we spend all this time why can't we take care of some of this stuff before, right?
00:11:00.859 --> 00:11:07.969
Or- I think digitization, that's I would hope has enabled us to do this, and you would know succinctly on this, Catherine.
00:11:08.149 --> 00:11:12.818
But what they should be doing on site is the things that they can't do remotely, to your point.
00:11:12.869 --> 00:11:23.889
Your exact point, and that is having the conversations with the sites reviewing where the patient-related data have seen Again, like you said, exactly seeing those trends alerting to safety.
00:11:23.889 --> 00:11:34.019
But it gets It's important that it doesn't get lost in the, in translation or buried beneath this mound of admin work that they need to do and then, oh, shoot, I have to do SDV.
00:11:34.178 --> 00:11:39.099
And they're clamoring through and burning through the patient stuff, which is actually the most important thing.
00:11:39.099 --> 00:11:51.808
So it is such an interesting place that we've found ourselves in because then that burden does that burden, how much of that burden, I should say then, gets pushed off onto the site itself and the other part of that is in an uncompensated manner
00:11:52.009 --> 00:12:01.948
And I think, and the other thing that's a barrier with monitoring is there's some barrier to acceptance of remote monitoring because the data itself is not natively digital at all sites, right?
00:12:01.948 --> 00:12:07.178
There's still sites that are working in paper, and then they have to scan that paper and upload it in order to make it digital.
00:12:07.178 --> 00:12:08.019
I use air quotes.
00:12:08.038 --> 00:12:09.609
You can't see them, but they're effective.
00:12:09.969 --> 00:12:11.208
And then the monitor...
00:12:11.229 --> 00:12:12.778
So the monitor can review it remotely.
00:12:12.798 --> 00:12:23.139
That's time waste on the site to scan in paper to enable a faster review on the monitor side, which inevitably, I do think saves some time at the visit, but it's burden doesn't go away, right?
00:12:23.139 --> 00:12:23.948
It just shifts.
00:12:24.149 --> 00:12:26.379
So we have to think about it from both ends.
00:12:26.379 --> 00:12:33.519
How do we get data natively digital from the beginning so that it can be transferred back and reviewed remotely by any interested stakeholder, not just monitoring.
00:12:33.519 --> 00:12:35.479
Also, your physicians can review data remotely.
00:12:35.479 --> 00:12:38.778
They don't have to wait for you to bring them the patient's source binders and things like that.
00:12:38.979 --> 00:12:46.229
And then you start to create more efficient use of the time that you do have to pay attention to what's in front of you when you can do it remotely.
00:12:46.428 --> 00:12:49.798
And that's what technology should be doing, right?
00:12:49.828 --> 00:12:56.349
That's- a perfectly good use case of a positive value add when it comes to decentralization and technology.
00:12:56.349 --> 00:12:58.609
And when it's done right finger, fingers crossed.
00:12:58.649 --> 00:13:03.438
Again you're close to that space that we can start to see some of those efficiencies play themselves out.
00:13:03.589 --> 00:13:19.259
another statistic here, Catherine, is 95% of sites we see still rely on email-based document exchange, and nearly two-thirds then say that sponsor technology just really just duplicates something that they already have or that they're already doing.
00:13:19.318 --> 00:13:29.009
And it becomes a, it becomes an isolated duplicative step outside of their process when it really doesn't have any time savings or value add for them is another way to say that.
00:13:29.009 --> 00:13:36.269
So if that fragmentation and things like that just as one example, you and I both know that's probably just one thing, is the real bottleneck.
00:13:36.278 --> 00:13:44.739
What has to then change on the sponsor side for something like TrialBlazer to actually reduce burden instead of adding to it?
00:13:44.938 --> 00:13:46.278
It has to happen in two places.
00:13:46.278 --> 00:13:49.558
It has to happen on the sponsor side, and it has to happen in the technology space.
00:13:49.558 --> 00:13:51.619
I wanna call vendors to the carpet on this one as well.
00:13:51.629 --> 00:13:53.688
Might, as somebody who operates in that space.
00:13:54.068 --> 00:13:54.879
It's two things, right?
00:13:54.879 --> 00:14:02.058
So the sites who've invested in technology for themselves, they have native digital, they have a workflow, they're in the tool, the data is there.
00:14:02.068 --> 00:14:13.269
We have to make it easier for that data to move from their system into the sponsor system so that the sponsors can aggregate it, which means that sponsors have to be accepting of data records existing outside of their control, right?
00:14:13.298 --> 00:14:17.259
If I have an eBinder or I have an eSource, or even my EMR, right?
00:14:17.259 --> 00:14:23.198
That data is sufficient for clinical care and for document retention from an FDA standpoint for the site.
00:14:23.198 --> 00:14:26.261
Why is it a different standard for the sponsor So that's one thing.
00:14:26.272 --> 00:14:29.581
And then I think it is on the vendors.
00:14:29.611 --> 00:14:40.172
The heaviest lift is the problem, is that we don't make it easy for data to be portable between systems, which means that if we don't do that, we're just, again, moving burden from one side to the other.
00:14:40.182 --> 00:14:53.201
So it's either the monitor has to operate in different systems and spend time moving in and out and moving documents between site systems into theirs, or the sites have to do it on their end and take between sponsor systems to move it into their system.
00:14:53.402 --> 00:14:57.392
And in the past, we know it's supposed to have been API, and can we create APIs?
00:14:57.392 --> 00:15:04.631
And APIs are great, but APIs have to be configured which is not easy for everybody to do, and they have to have mutual customers, which is a whole other thing.
00:15:04.831 --> 00:15:15.961
But I think what's really gonna open up for us is AI and the ability to read data and audit trails and move data more seamlessly without an API, so MCPs, things like that will hopefully start to reduce some of that friction.
00:15:15.961 --> 00:15:21.902
Because until that happens, we are basically asking one side to absorb work at the expense of the other
00:15:22.101 --> 00:15:30.032
Well, and it's such an interesting, a-another interesting, perspective and fact really with whose burden is it, right?
00:15:30.042 --> 00:15:36.162
And this concept of the sponsor could do it or the site could do it to, as in an example of what you just talked about.
00:15:36.172 --> 00:15:38.341
So whose ultimate responsibility is that?
00:15:38.361 --> 00:15:49.751
when is it appropriate for the site to be, we'll say, taken from arguably more important tasks, and then into things like that.
00:15:49.831 --> 00:15:55.562
And, we don't need to get into it today, Catherine, because I, it's probably another 30-plus minute conversation.
00:15:55.562 --> 00:16:04.422
But then where does AI come into that equation to ease some of that administrative or execution burden on the site and the sponsor, right?
00:16:04.422 --> 00:16:05.142
It kinda takes...
00:16:05.162 --> 00:16:08.981
it's a third party in a way that kinda takes that burden off of both sides.
00:16:09.182 --> 00:16:09.731
Exactly.
00:16:09.782 --> 00:16:11.471
I literally just had this conversation this morning.
00:16:11.772 --> 00:16:20.596
My site heart wants to say that it's the sponsor's responsibility to absorb the burden because they have the resources and the money and more- to win out of it than the site does, right?
00:16:20.796 --> 00:16:29.245
But then I also have my empathetic sponsor hat on of for the smaller pharma and biotechs, like they're just trying to get to the end of the day with their single molecule and make something happen.
00:16:29.245 --> 00:16:31.216
And so we're also asking them to pick it up.
00:16:31.546 --> 00:16:35.275
But I think it's an unfair paradox that we have to ask anybody to pick it up, to your point.
00:16:35.275 --> 00:16:36.635
So I think AI is...
00:16:36.706 --> 00:16:40.734
my true hope for AI is that it can solve the data movement problem.
00:16:41.053 --> 00:16:50.333
Because if we can query databases, and we can look at blinded clean rooms of data and query against it without ever having to ask a question of a site and without de-identifying that site.
00:16:50.364 --> 00:16:58.743
We could be able to know what's happening by surfacing some of those signals and audit trails in a way that is safe and meets data privacy and all those other things.
00:16:59.373 --> 00:17:08.271
we need to lean into that more as an industry and figure out how to make AI and the cloud work for us in a way that APIs have not in the past to, to get that through.
00:17:08.836 --> 00:17:09.777
I couldn't agree more.
00:17:09.936 --> 00:17:14.221
And again, I think it's maybe a conversation for us to set up in another time.
00:17:14.250 --> 00:17:18.750
I'm sure we'll end up talking about it and being in the same room when those conversations continue to happen.
00:17:18.750 --> 00:17:45.080
But as we move to start to close our conversation here, Catherine my final question is as you've moved from ETMF pass rates from a, a place of 65% to nearly 99% across the network of sites that you have in the technology system, With over 173 billion in research assets licensed out to China last year- the question isn't just whether the US can compete on incentives.
00:17:45.280 --> 00:17:51.961
Where and how can sites absorb the volume if conditions really showed up tomorrow?
00:17:52.020 --> 00:17:54.060
If things, started to shift for us?
00:17:54.260 --> 00:18:04.661
So the move with the volume of work that's moved out to China, it's moving there because of the speed of enrollment, which means that they've optimized their health networks to enroll patients at scale, right?
00:18:04.980 --> 00:18:05.951
We don't do that here.
00:18:06.131 --> 00:18:14.550
I think that's not gonna change for us, so I think it's going to be more sites doing more studies and still having smaller portions of patients to manage across them in order to beat speed.
00:18:14.750 --> 00:18:21.500
But it is going to be leveraging technology and AI to help the people on the ground in front of those patients move faster.
00:18:21.500 --> 00:18:28.633
So eTMF pass rates means that basically the site has put in a document, they're able to get any queries resolved in real time.
00:18:28.633 --> 00:18:31.192
We're not waiting six weeks between monitor visits.
00:18:31.202 --> 00:18:32.502
They can see it electronically.
00:18:32.833 --> 00:18:39.002
Also, we're creating prompts to tell sites when things might be wrong, so we're not waiting for a monitor to expose those.
00:18:39.012 --> 00:18:45.073
We're pointing it out to the coordinator"Hey, you scanned this page in backwards on page six of this 20-page PDF.
00:18:45.272 --> 00:18:49.282
Hey, this consent form doesn't match the language in your contract.
00:18:49.333 --> 00:18:52.603
Hey, this signature doesn't match the name on the CV." Things like that.
00:18:52.633 --> 00:19:00.833
And all of that, they're small little things, but those small little things take up time and data cleaning, and if we can remove all of that, the documents move through faster.
00:19:00.853 --> 00:19:09.962
That's a signal that we are making the people doing the end work more effective and smarter by giving them cues of where things have opportunity or errors have occurred.
00:19:10.002 --> 00:19:14.573
Because we're all human and we're moving very fast in order to keep up with the demand of delivering clinical care.
00:19:14.772 --> 00:19:21.292
And so we're trying to make it easier for those mistakes to get identified and corrected immediately rather than prolonging themselves.
00:19:21.612 --> 00:19:24.175
So I think things like that will help with our speed.
00:19:24.576 --> 00:19:32.955
I think optimizing a trial selection based on actual performance data of the site, not just reported performance or historical performance data.
00:19:33.155 --> 00:19:40.096
The one thing I always hated is people would come to my site and say it took you 260 days to do this study." The last time I did a study with you, it was 10 years ago.
00:19:40.105 --> 00:19:41.986
You wanna know how I'm, how I can do it today, right?
00:19:41.986 --> 00:19:42.276
I don't...
00:19:42.675 --> 00:19:44.536
We want sponsors to have real-time data.
00:19:44.566 --> 00:19:53.155
We want sites to know, be able to talk about what their performance is, and we can select sites that actually can meet criteria and have the ability to execute on the trials that are coming down.
00:19:53.226 --> 00:19:55.796
And I think those two things will help make us more competitive
00:19:55.996 --> 00:19:57.226
I couldn't agree more.
00:19:57.276 --> 00:19:59.365
It's about the context too Catherine?
00:19:59.486 --> 00:20:19.346
Being able to help the, whoever's on the other end of that information understand effectively what that means to your, a, a great example that you just gave there around that site selection process is really important because performance is, I don't want to say arbitrary.
00:20:19.481 --> 00:20:30.321
there's a large portion of it that is not, but there's a portion of it that does oftentimes need understanding and conversation and additional context to really understand what the entire picture of that means.
00:20:30.321 --> 00:20:34.451
And we're moving so fast, like you said, that sometimes that context gets lost.
00:20:34.451 --> 00:20:38.271
But I think it's a great place for us to conclude our conversation.
00:20:38.561 --> 00:20:43.832
A lot of really interesting points, Catherine, that you brought up here, things for us to really think about as an industry.
00:20:43.832 --> 00:21:01.301
I'm excited to hear and see where Operation Trailblazer continues to go as, your organization sits at a purposeful crux to move some of these challenges in the right direction, and like I said, really put a pin in them so that we can start to make change as a result.
00:21:01.501 --> 00:21:02.261
Yeah, absolutely.
00:21:02.261 --> 00:21:04.551
I'm excited as well, and I am hopeful.
00:21:04.551 --> 00:21:09.852
I remain the eternal optimist that we will figure this out and get out of our own way and go faster.
00:21:09.882 --> 00:21:10.761
So here's hoping.
00:21:10.862 --> 00:21:11.961
yeah, here's to hoping.
00:21:11.961 --> 00:21:13.551
That, I couldn't agree more with you, Catherine.
00:21:13.701 --> 00:21:14.271
So thank you.
00:21:14.511 --> 00:21:18.291
Thanks again for being with us here today and and for participating in the conversation.
00:21:18.491 --> 00:21:18.991
Thanks, Jimmy.
00:21:18.991 --> 00:21:19.571
I appreciate it.
00:21:19.771 --> 00:21:29.582
And for everyone that's listening, be sure to check out other site-focused resources made available to you and the entire community on our website, myscrs.org.
00:21:29.612 --> 00:21:40.942
This includes additional podcast episodes, website webinars rather, and a plethora of thought leadership opportunities Thanks again for listening, for tuning in, and until next time
