Life Sciences

[Vlog] Perficient + Lokavant: Proactive Clinical Trial Operations

Param Singh And Devin Solanki

Param Singh: Hello, everyone. My name is Param Singh, and I’m the director of the clinical operations solution practice here at Perficient life sciences. And I want to welcome everyone to today’s video session where we’re going to give an intro to the value of the Lokavant platform and get a chance to briefly see the system and its power.

I’m very excited to have with me today Devin Solanki, who is the head of growth at Lokavant. Devin, thank you so much for joining us today. We’re really looking forward to today’s discussion and demo. Devin, I’d love for you to introduce yourself to everyone today.

Devin Solanki: Now, thanks for having me, Param. And I’m excited as well. So, I’ll give you a brief. As Param mentioned, I’m the head of growth at Lokavant, which is a clinical trial intelligence company. But prior to that, actually, I was working on the sponsor side at Roivant Sciences supporting a variety of our biotechs on everything from discovery, development, to commercialization and how we could bring technology to ultimately make our drug development process more efficient.

Then I stumbled into this space, actually, as a consultant at McKinsey, where I saw these challenges emerging on the site side, working with large hospital systems. So really ran the full gamut there and have come back on the technology side to hopefully stitch together all the problems that we saw.

The Challenge Clinical Operations Teams are Facing

PS: That’s great. Thank you, Devin. Again, I’m really glad to have you here with us today and really excited that we’re going to be able to share a brief demo of the solution today as well. Before we get into the demo, though, I do want to set the stage a little bit. We have seen in the past several years that in the life sciences industry, with the advent of new enterprise systems, cloud applications, that are really designed to make our operations much more efficient These systems are really generating an increasing amount of data.

And in most cases, there’s a significant volume of data coming in too quickly to be manually curated. Data has to be identified and has to be tagged, structured combined in increasing volume before we can effectively manage to drive intelligence for decision making from the insights from that data.

So, what’s really resulted is in clinical operations teams, instead of working to resolve the key issues and risks in our trials, they’re really forced to serve as data analysts. Right. To sift through, understand and make sense of that data before they can really implement any resolutions to the issues and the risks, and at which point it might already be too late in some cases. So, Devin, I was hoping you you could briefly discuss this challenge and these challenges and how the Lokavant team has seen this evolving in our industry.

DS: Oh, certainly, Param and that very much resonates with me. And I’ll just share sort of one piece of framing information we often use when we have these conversations, which is this trend of a tremendous amount of data and tremendous amount of sources that just doesn’t exist ten years ago.

The way that we’re running trials now, in terms of the data sources and the data flow, is and will continue to be drastically different. That’s only been accelerated by COVID. And, actually, that’s a good thing, as it is in most industries. More data means more evidence, which hopefully means potentially more therapies in the future, more effective trials and more patient centricity with a lot of the new devices out there, as an example.

But what the challenge comes in, like you mentioned, is when we use the same processes and infrastructure and systems to try to manage these type of trials. We have all these bandaid solutions where there’s a lot of reconciliation and oftentimes a lot of insights like you mentioned, will get left behind because your data sources are fragmented.

Your teams themselves are siloed by their data source or by their function on the study. And so that just leaves what we call quite a large opportunity out there, of course, in terms of accelerating trials, reducing the cost of developing these drugs, and most importantly, I think an often overlooked, but actually better data quality, not just because you’re observing this new flow of data, but without the right systems, intelligence processes in place this inflow of data and different sources creates more risk, in fact, than actually reducing it.

And that’s something that I saw on the sponsor side many times over, where we’d want to implement novel technologies. We’d want to have continuous streaming data, maybe collect on 30 or 40 different secondary endpoints that we think could be helpful for future drug development.

But the rate limiter was our clinical operations teams were getting left behind because they were stuck reconciling this data in the same way.

At Lokavant, even before we started as an official company, were actually building technology internally for our small biotech sponsors. Before we even built that technology, we looked for a solution that could help us solve, ultimately, what we thought as a two rate limiting factors.

The first is that getting insights just took way too long. We were taking weeks, sometimes even months, relying with the different vendors and partners to understand what was happening in our study when we knew it should be real time.

And the second is, as a result of this, our teams are being reactive, looking back at issues that did happen and trying to put out fires, rather than using all this information from our past experience to try to be more proactive and take an approach on what will happen. In my trial. And point solutions weren’t helping us there, for the reasons I described earlier.

What we needed to start with, and what we discovered was there’s a middle layer that you need that sort of sits on top of a lot of these source systems, whether it’s your more traditional EDC or your CRO’s CMS or something more novel like wearables or other direct data capture sources.

Having a middleware that sits on top of those, harmonizes that data into a single source of truth or a common language that you can understand not only between teams on your study, but between all of your studies. And taking one step further, to use that data to provide predictive analytics was really a part of the core platform that we built that actually drove change for our studies.

And then the second part of the diagram you see down there, those insights are only as good as they are user friendly. So we didn’t want to build this complicated infrastructure that then would require another member of our clinical team, a statistician or even a data scientist to interpret. We want to get these insights, like you said Param, in the hands of the people that can take action to them. So that’s how Lokavant has thought about this approach.

PS: Right, that’s that’s incredibly insightful and very exciting indeed. And Devin, I mean, we’ve seen it time and time again as well. So often it seems like teams are working in reverse, right, reacting to operational data that in most cases isn’t even real time, but still to trying to determine how to get a clinical trial back on track.

And so, at this point, I’d love for you to jump into the system itself and show the power of the solution to address these challenges and work towards a way to go from, again, reactive to proactive trial management.

Lokavant Demo: Moving to Proactive Trial Management

DS: Yeah, that sounds great, Param. I’ll use our risk-based monitoring solution just as an example here. You can see here, Param, is exactly kind of what you were talking about, which is the sort of power of the solution lies in the hands of how does it make your team proactive in managing their studies. The way we think about doing this in the context of our oversight platform and risk or trial oversight tool is through risk scores, simple indications of your study health and the details behind your study health, that go from zero to five.

So you can see here right away across the portfolio, which study you need to focus on because it’s high risk. But importantly, we don’t stop there because we recognize that not all risks are equal, right? You have certain tradeoffs that you can make on a study depending on what is at risk. And so we break it down to what we believe are sort of the most relevant indicators or outcomes for any given study, which is:

  • the timeline
  • the cost or budget
  • and then submitability or data quality in the study.
Life Sciences - How Artificial Intelligence Can Enhance the Clinical Data Review and Cleaning Process
How Artificial Intelligence Can Enhance the Clinical Data Review and Cleaning Process

This guide analyzes how artificial intelligence – including machine learning – can be used by pharmaceutical and medical device companies to improve the clinical data review and cleansing process.

Get the Guide

We’ve even seen now that we’ve deployed across a variety of different sponsors and CROs, folks using this to make tradeoffs on if it’s a competitive product, for example, and the timeline really matters on the study, you might be willing to spend more or increase your budget to make sure that it stays on track. Or in other cases, only data quality might be the thing that matters, for example, in an ultra-rare indication. So that’s what is at risk in my study.

And then on the right-hand side here, with these risk categories trying to answer the question on how do you resolve this? So right away, looking at this example, you can see the areas of protocol adherence. Are sites sticking to their protocol? Are we getting a lot of deviations in enrollment? From screen for our discontinuations to actually getting enough folks in the funnel are two areas that are high risk for the study and where you’d want to drill into right away.

And then importantly, to add to that, we think at the high level, you would want to see in this cockpit which countries are driving this risk and relatively which sites are driving this risk, because that’s ultimately the level you have to get down to, to sort of get to some answers. So, yeah. That’s kind of how we think about it at the highest level within the first sort of view of your your study. Which studies do I need to focus on, and where are the risks in that study?

PS: Yeah, and this is what gets me excited about it, you know that.

So, the system essentially surfaces what’s most important in your study immediately based on the criteria and those key metrics.

So again, there’s no more sifting through or churning through that data. One of the key things to note here is that this data is in your transactional system, right? These are coming from transactional systems data.

But, a team member will have to use their expertise to really sift through large volumes of data to really figure out where the risk is and what’s important, where it originated, how to fix it. And at best, in the transactional system, all they can do is really answer what has happened and how they react to it, let alone having some predictive insight into the future of the trial. So this single source of truth, and this consolidation of these risk categories, real time predictions and insights that it enables are really game changers.

I feel, I get really excited about that. And I especially get excited about how this solution then takes that knowledge a step further and points to, OK, so so now what? What do we do now? How can I maybe you can you can share this, Devin. How can sponsors and CROs determine then where to focus their attention and their conversations for that that greatest impact and resolution and getting that study back on track?

Taking Action With Real-Time Data From a Single Source of Truth

DS: That’s it’s a great, great summary and a great natural next question, Param. Once you can surface what’s happening in those transactional systems and de facto at your sites, how do you then take action against that? And that’s something we spend a lot of time talking about. So I’ll I’ll show you an example here.

The first step, like you mentioned, is being able to sort of see what is happening in real time. And I think normal approaches for this, even using technology, you typically have a periodic review of what did happen in your study. Maybe I put a couple of points on this chart, but the approach that we’ve taken at Lokavant is do the work upfront to connect directly to those transactional source systems so we can pull this data in in near real time, sometimes up to six times a day for certain type of studies.

And that’s where you get to see these views on, OK, not only is my risk high, but it’s been trending up. And if we look at the categories that you mentioned before, it might have been trending up just this past week or these past couple of days. So that’s when I know as the sponsor of the study, the CRO, the operations team, these are the areas where I need to focus on right now at the study level to make sure that there are not issues in the future in these categories.

And so the next step on that, naturally, is seeing what’s wrong, let’s say taking quality as an example. What’s going on in terms of our protocol adherence at our sites? And sure enough, you see that breakdown across the key risk indicators that, Param, you mentioned kind of roll up into this this larger number. So now we’re getting to one level of more detail. And right away across the board, you see there is risk across all different types of deviations.

But the majority, the highest risk area, there’s just too many major deviations. And again, being able have that real time trending view, you can see that these deviations have picked up over the past few months alone. The next natural question there is, OK, is this a study level problem? Is it two sites that are causing this problem? Is it 100 of them? where do I need to go to next to resolve this? So for each of these areas, rather than generating another report, we aggregate these metrics at the site level automatically as well.

So you can see some sites that are doing better and then some sites like here in New York in this example where things aren’t going so well, so if we click down here now, you’re taken to the sort of site level. And now we can talk about sort of where it’s happening with those metrics at my highest risk site. Sure enough, in this example, you’re seeing the same trend across the board and that you’re seeing it’s picked up over time.

Now, here you already have enough information to have a conversation, right, with your clinical operations team. You know who the monitor is. You know who the principal investigator is, you know when their next visit is, when their last visit was. So you have the sort of contextual information to make a call.

But in our experience, you almost need to go one step further. Because while there are a lot of issues at the site level, being able to drill down to the patient or the identified patient, rather, let’s you really get down to the details for the conversation. So that’s what we’ve done here. You know that the major deviations are an issue. And in fact, you can see over the course of the past few months, a lot of those deviations from the same patient or set of patients, and a lot of them are related to concomitant medicines.

So not only you can you say, hey, look, there’s too many deviations to your site, can you tell me what’s going on? Can I schedule a visit? You can get on the phone that day and say, hey, look, there’s this patient ID. We’re seeing a lot of concomitant meds. You’ve seen it with other patients as well.

Maybe there’s a retraining that we need to do at the site on the protocol or maybe we need to take and examine, look and see what’s going on with the sort of concomitant medicine to see if we need to adjust the protocol more broadly. And that’s sort of the power of getting down into this level of detail in just a few clicks versus a few months of work.

PS: Yeah. No, I yeah, I especially love how the solution will make our our conversations real, right, and make the actual work that our teams are doing much more fulfilling where we’re not again, we’re not forced to be data analysts.

But we’re, we’re able to have contextual business and operational conversations and actually resolve issues.

Like, you just showed that in, in a few clicks. Again, instead of just trying to troubleshoot operational data, we’re actually having these real conversations and resolving the issues at hand. So that’s that’s something that I really love about this solution. And, you know, you’ve covered. I’m sorry. Go ahead.

DS: Yeah, Param, I was going to say before you move on, I really like how you put it.

Giving the contextual relevance is what ultimately leads to the the better results.

And maybe I can just give you one example there before we move on and close up demo where this actual issue in terms of protocol adherence did pop up with one of our partners. And in fact, it was in the first couple of months of a rare disease study that was operating during the height of the COVID pandemic. So you had a lot of new technologies being used, a lot of remote monitoring for the first time for these folks. And there’s a lot of deviations. So the team was having would have struggled to cut through the noise.

But because we had the sort of predictive analytics from other similar studies using similar sources to identify when was an issue real, we actually were able to notify them using notification feature just like this one, that there were a high amount of deviations in the first month of their study and those deviations were all tied to the same sort of issue, which was patients dropping out, being lost to follow up. Now, that could happen for a variety of reasons.

Especially during COVID, but we saw a particular trend in the data that felt anomalous based on what we’ve seen from from other risk scores. So Lokavant notified that the CROs in charge of the study. They took a closer look, realized that there was a certain set of sites that weren’t being trained appropriately on the protocol, and issued a retraining really quickly.

When we talked to that team about what would have happened without Lokavant, it would have taken about three to four months to have enough data to detect is this an outlier or not? And that would have meant dozens of patients dropping out of a rare disease study, which may have either delayed it significantly or canceled it altogether. So my favorite part about this job is seeing all the technology make the lives of the clinical operations folks easier, but also then ultimately impact the patients, something like that.

Beyond the Surface of Intelligent Clinical Trials: an Intelligent Middle Layer

PS: Yeah, no, I really love that. The fact that you’re able to give us real examples of where this has actually been that impactful, that that that’s huge. I know we’ve covered quite a bit of ground, but from my understanding, we’re only scratching the surface of what’s possible really with the solution, aren’t we, Devin? I mean, you want to speak to that a little bit.

DS: Yeah, certainly, Param. I think you and I talked about that at length from from your angle, seeing across all sorts of different transactional systems and technologies. That’s just one use case, within one metric, within one sort of product. The real value of this connected clinical intelligence platform, it comes from the fact that all of this is flowing into the same sort of core integrated set of metrics and set of applications across all the clinical and operational user groups.

Now, I wouldn’t say we’ve built all the solutions yet, but we’ve got quite the suite being built. And what we’re seeing is insights when connected are much more powerful. So things like I just showed you with protocol adherence and enrollment on a clinical operation side, and then separately on our clinical data side with sort of our medical monitoring applications, our study startup applications, rolling those insights together and connecting those teams has become incredibly valuable. And every new study that’s run, every more data that runs on the platform, drives value for all of the users.

So ultimately what we’re hoping for is that in the future all trials are intelligent.

You have this selection of transactional systems you can use, which I know you guys work with many of them. Your partners do too. But there’s always this intelligent middle layer that brings you back to predictive proactive analytics while giving your patients and investigators the choice to use the systems that they want.

PS: That’s great. Well well, I know, again, we’ve only scratched the surface of what Lokavant can do. And I look forward to more opportunities to share these details around the capability of the solution with you, Devin, and with our partners and how it can be used to, within seconds, really discover what’s most important in your clinical trial with respect to risk to proactively understand the health of your study and to quickly determine where to focus that attention for that that greatest impact.

And I personally, you know, our team, we’re extremely excited about this partnership because with the power of the Lokavant platform and Perficient’s deep industry knowledge and expertise in data integration and implementation, our joint team is really uniquely positioned to partner with with sponsors and CROs confidently and quickly break free from, again, reactive processes and processes and, empower them for more proactive, efficient, insight driven clinical trial operations.

So, ultimately that’s driving towards innovation that supports better health for patients, so I’m really excited about that. So, Devin I really want to thank you for being with us today and help us give an overview and introduction to the solution in the partnership. Our team is extremely excited to work alongside yours in serving our clients.

DS: Yeah. Thank you so much. For having me, Param. And likewise, we’re really excited about where this partnership can go.

PS: Great. And thank you for everyone for joining us today. And, if you’d like to learn more about Lokavant and have a much more detailed and contextual conversation about the solution, please reach out to us and we’d be happy to schedule a more detailed demo and discussion as as it pertains to your organization. Thank you all so much for joining us.

SEE MORE: [Vlog] Intelligent, Real-Time Clinical Trial Insights: Lokavant Partnership Introduction

Learn More About Our Life Sciences Solutions

Leading life sciences organizations rely on us for strategic and technical expertise to achieve their missions. Our thought leaders are here to support you in achieving business goals and solving your most complex technology challenges. If you are interested in learning more about Perficient’s Life Sciences solutions or would like to contact us, click here.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Cassidy Rimmey

Cassidy Rimmey is a Marketing Coordinator in the healthcare industry at Perficient.

More from this Author

Subscribe to the Weekly Blog Digest:

Sign Up
Follow Us
TwitterLinkedinFacebookYoutubeInstagram