Rivia secures $15M to build an agentic-driven platform for clinical trials

0:00Mike Butcher:Hello and welcome to Path Founders, the new tech media brand for entrepreneurs, investors and operators. We cover the makers of the code, the capital that gets involved and the consequences. Clinical trials for new drugs involve some of the most expensive and complex processes in modern industry. We're talking$2 billion per approved therapy. Yet remarkably, much of the infrastructure still runs on spreadsheets, manual review and fragmented data. But as a couple of years ago, of course, AI has begun to flood into health care, promising faster drug development and lower costs. However, as we're seeing across sectors, AI works best when it has accessed organised, accurate data.

0:49Mike Butcher:Today's guest is building what they describe as an agentic data engine for clinical trials, a system designed to unify fragmented data sets and embed AI directly into the operational workflows that actually determine whether a drug makes it to market. So welcome to PathFounders Eric Scalfaro. Hi Mike, thanks for having me. And I'm excited to talk to you because there are so many clinical trial startups, I don't even know where to begin. There's a lot out there and a lot using AI. And I think the first question is going to have to be, what's going to set you apart from these kinds of startups? Right.

1:34Erik Scalfaro:Indeed, there's been quite a wave since I started in the pharma sector just 10 years ago. we've decided to focus on one fundamental problem which is fragmented data we help biotechs and pharmaceutical companies solve this this problem by building an underlying foundation that helps unify and structure data across all data sources across all vendors and this helps them operate irrespective of the choices they make and we observe this through through our own experience This is, you know, our own pain that we're solving. We are drug developers in the founding team. And so for us, this was an acute need and acute opportunity when we founded the company four years ago.

2:25Mike Butcher:Right. Well, I mean, there are a lot of buzzwords being floating around AI and what have you, and also agentic systems. So exactly what do you mean when you say you're applying agentic systems to this sector? Right.

2:38Erik Scalfaro:So you really have to think through the structural components of what we provide to understand it. We made a deliberate decision two years ago to first focus on the initial foundation, on top of which reside workflows and then agents. In this case, we are integrating data across vendors. we are structuring it and then we're applying the very specific and unique logic of each individual trial each trial is unique each patient is unique and so there's a lot of specifications that need to happen once you've done that and only once you've done that you can start to point that data towards operational workflows and medical workflows that can then be supported by agents For example, we have an agent that helps to track if patients are coming in on time to the hospital, the so-called visits, a task that typically is done in Excel spreadsheets.

3:36Erik Scalfaro:And so having that unique specification to meet the conditions of each trial and having an agent on top of that allows to essentially multiply the capacity of clinical teams. And for us now, the objective is to really expand that across an exhaustive set of workflows in the very active conduct. We are in the active conduct of the trial where data is coming in, patients are being treated on a daily basis, and you need to understand what's happening.

4:02Mike Butcher:Okay, so in a way, the agent's doing some heavy lifting that would normally be done by a human, right?

4:08Erik Scalfaro:Correct. We are completely or nearly completely eliminating a lot of manual, tedious data tasks that are summarized in the formatting, the consolidation of data, and then processing that through workflows to the very last step where we ask the human to verify and to make the ultimate decision. And so that really helps clinical teams to achieve much more with fewer resources. So it's an efficiency component, but it also allows them to catch errors and opportunities that they potentially would have missed. The crux of this is volume. You have hundreds, if not thousands of rules across millions of data points.

4:56Erik Scalfaro:And so the human eye is incapable of detecting patterns and detecting root causes. And so agents not only have the opportunity to make things much more efficient, but they have an opportunity to really go much deeper for root cause assessments and pattern recognition, which tie both, again, to operational needs and to the billion dollar question of every single drug developer. Does the drug work and is the drug safe?

5:23Mike Butcher:So I know that you're claiming something like a 50 % cost reduction in the whole processes. Is that because you're cutting humans out of the loop and therefore less humans means less costs? Or are you doing something more across the board because of the system?

5:40Erik Scalfaro:It ties back to these two components. Indeed, it's an element of reducing the amount of headcount needed to operate a trial. So you can actually conduct more trials with the same amount of resources today. And secondly, you're preventing a lot of errors. You have to remember that clinical trials are a noisy instrument. There's dirty data coming in from all over the world. and so you need to catch those errors. Missing out on errors can sometimes impede you from completing the trial, it can impede you from making the right decisions which are all very costly and so those root cause assessments, those patterns really help prevent cost and I would say there is one other one which is less about cost but a more fundamental one, which is really the upside.

6:34Erik Scalfaro:Drug development is all about evaluating, you know, new investigational drugs. And so the earlier you do that and the more robust evidence you have, the more you understand whether the treatment is fit for purpose, i.e. does it really treat the intended disease in the way that we thought? And, you know, should this then justify further development or approval. And so Rivia gives that opportunity because we're collecting all this data in real time. We're providing specific trial logic and then agents that help sift through the noise and spots those differences. Okay.

7:17Mike Butcher:Well, you've raised a$15 million Series A led by Early Bird. There's participation by Defiant and others, Speed Invest, Amino, Nina Capital etc. You've raised this money but presumably your investors think that a Palantir or another big tech data cruncher isn't going to come along and snuff you out with their own system. What are you going to be doing that's different to a much larger player in this sector? Right.

7:51Erik Scalfaro:This ties back to the evolution of the problem and the solutions that have accompanied this through the journey. We have seen in the last decade an explosion of data in clinical trials. 400 % increase. You're collecting data from a myriad of sources. When I was at Novartis 10 years ago, it was mostly just clinical data. Now it's clinical data, lab data. You're asking patients to collect wearables, to take diaries at home. And so the underlying infrastructures have not kept up. They've not kept up to the evolution of these demands. And so the incumbent players have two choices. Either build nuts and bolts on top of their existing infrastructures or completely build from scratch.

8:39Erik Scalfaro:And so we are the first of its kind to have thought about solving this problem from the ground up. This ability to integrate any data at speed with workflows and agents. And the thing is, you need real clinical trial data to do that. You cannot just be a team of really talented engineers somewhere on the planet and think, oh, that's not a good problem. I'm going to solve it. No, you're talking about patient level data in a highly regulated environment. And so you need to have real customers, which in itself is both an enabler, because for us that feeds our data engine that's growing. And so we're going to keep ahead of the curve.

9:19Erik Scalfaro:Every new trial helps build this ontology of clinical specific logic. And so the more we progress, the more we advance ahead of the curve.

9:30Mike Butcher:So what kind of partners do you have on board and can you name them?

9:34Erik Scalfaro:So we have some tremendous biotech innovators who are spread across Europe and the US. To name a few, we are focusing mostly on oncology. Companies like Elentis Therapeutics and ScanCell in the UK are two oncology-based biotechs who are developing the first of its kind mechanism of action. So they are true innovators of its kind. More broadly, we have partners in other therapeutic areas that are also tackling new mechanisms such as gene therapies in rare diseases. And so in these cases, we are involved in the first stages of clinical development. It's often the first time that these drugs are being tested in a human.

10:29Erik Scalfaro:The step before that were mice and rats. and so they now have an ability to understand the potential of their therapies in real time and maybe to round it off this is more on a personal level for me and for the team at Rivia the most humbling thing the most exciting thing right is when we are looking at these workflows and these analytics in Rivia we see a line going down it's not just a line it's someone's tumor going down. That's a real patient that's seeing real benefits from a treatment where there are no alternatives. And so we're seeing that as quickly as the oncologist that is treating the patient.

11:14Erik Scalfaro:We've seen those impacts is an extremely humbling experience, and we're very excited to be part of it.

11:20Mike Butcher:Just briefly, a short answer. How much faster do you think it will be to develop a drug now that AI systems like yours are in the pipeline?

11:31Erik Scalfaro:You know, that's a very hard question to answer because you have to still follow scientific rigor. The trial designs are designed as a scientific experiment that need to treat a patient population across the duration.

11:47Mike Butcher:Sure, but I mean, sorry to interrupt, but what I'm trying to say is if the average for a new drug is, I don't know, three to five years. How much faster do you think AI is going to bring that down?

12:00Erik Scalfaro:I think there's a real opportunity. If you consider also the wave of adaptive trial design, which by design have different stages, you'll probably reduce that by at least 40, 50 percent. Because if you have an ability to evaluate the patient population with the same scientific rigor and collect all the data in high quality, you can probably reduce that based on some loose estimates. I think it's hard to put the finger on that one.

12:31Mike Butcher:Right. Well, hopefully it'll reduce the time for effective drugs and also the cost. Just quickly, Eric, just give us a little bit of an idea about your own background and your team's background. Exactly where did they come from? Were they more from the tech industry or from the pharma or medical sphere?

12:53Erik Scalfaro:Let me start with the latter. For us, it was very important to always have a balance between medical and clinical development expertise and best of breed software engineering skills. This component is actually quite hard to find in our industry. So we've got experienced executive physicians and we've got cybersecurity engineers in the founding team. I come from this space as well. I spent 10 years in the industry. I trained as a chemist and then spent most of my career in spreadsheets, suffering from the pains that we are solving today. at big pharma, at CROs, which are the service providers that help pharmaceutical companies and supporting also small biotechs.

13:44Erik Scalfaro:And so consistently seeing this problem permeate, I thought enough is enough. We have an opportunity. We have a real problem to solve. And finding the right team with this healthy balance was for me then the right elements to make the leap.

14:00Mike Butcher:Well, it's very interesting to see that investors have gone in there, especially on this startup, given that there's so much competition out there for this clinical trial sector, which I know so many startups have been concentrating on over the last few years. But for now, Eric Skull, Barrow, who's the CEO and co-founder of Riviera, thanks very much for joining Path Founders.

14:28We'll be right back.