All In on Biosim AI with Simulations Plus

Announcer:
Welcome to TD Cowen Insights, a space that brings leading thinkers together to share insights and ideas shaping the world around us. Join us as we converse with the top minds who are influencing our global sectors.
Brendan Smith:
Welcome back to another episode of Machine Medicine, AI and Healthcare, TD Cowen's podcast series where we bring you the latest and most important takeaways from the state of AI and the healthcare sector today. I'm your host and TD Cowen Healthcare Analyst, Brendan Smith, and today I'm joined by Simulation Plus's Chief Executive Officer, Shawn O'Connor. Shawn, it's great to have you, and welcome.
Shawn O'Connor:
Hey, thanks for having me, Brendan.
Brendan Smith:
For anyone new to our podcast series, Machine Medicine aims to break down the use of artificial intelligence across healthcare into bite-sized digestible points one episode at a time. We're looking to highlight the biggest misconceptions and then recontextualize each piece back into the bigger picture. Today, Shawn and I are exploring how advancements in artificial intelligence are changing the game for biosimulation players, how Simulations Plus itself is future-proofing their own platform, and really what the field of AI-powered biosimulation ultimately means for the industry.
Shawn, let's just dive right in. First, let's maybe start with a quick point of distinction here. Broad strokes, what do we mean when we say biosimulation versus artificial intelligence, and really, why do next-gen biosimulation platforms, like the one you all developed at SLP, really lend themselves so well to the integration of AI?
Shawn O'Connor:
Yeah, thanks, Brendan. The misnomer there is biosimulation versus AI. AI is a tool that we've used in biosimulation since we began in the early '90s. Obviously, AI wasn't the buzzword it is today at that point in time, but early techniques in terms of machine learning are what formed the basis of our products as early as the mid-'90s. Obviously the tools have improved and advanced over the years, and as they've advanced our ability to take biosimulation approaches, in silico modeling of biology and chemical entities to be predictive across the continuum of drug development, those AI developments have improved our ability to be predictive and insightful to decision-making along the way.
Brendan Smith:
Mybe let's double-click a bit more on SLP's platform specifically, namely, your first launch within the sweeping AI rollout integration, a little bit more broadly, GastroPlus .2. Historically, what has been the real selling point for GastroPlus's value-add to your customers, and how do you envision the integration of AI really taking this to the next level?
Shawn O'Connor:
We have several platforms that support discovery, preclinical, clinical development, and even have applications after market approval for a drug. The focus and benefit of these tools is really focused on that billion-dollar cost and 12-year timeframe to get a drug to market. How can we shorten those timelines? How can we reduce the cost of a drug getting to market? Each of our applications has different use cases. Biosimulation really has built itself over the years by the introduction of new use cases on an annual basis, new decision points, new simulations, new models that dive deeper into the biology, the science, and leverage AI tools. The real leverage point today is beyond simply the ChatGPT, the information gathering capabilities of LLMs, and now is evolving into the application of agentic AI into our workflows for biosimulation.
What does that mean? What it means is that we can increase the efficiency of the scientists that's applying these tools to develop models, biological models and drug models and accelerate their process. Agentic AI can step in and automate processes that typically would require an extended period of time on the part of the scientist before he got his teeth sunk deeply into the real issues. The mechanistic processes of pulling together models, the mechanical processes, I should say, can be automated and get the scientists much farther down the line, and that frees up his time to expand the application of biosimulation. It allows an organization that is invested in biosimulation to deploy these techniques across their drug programs, across the continuum of time more rapidly.
Brendan Smith:
Yeah. You're getting at a lot of really important questions here, I think, in a lot of the general public's attempt to understand where this is today, where it's going. As you're looking forward over the next, let's say 12 to 18 months, as you all have talked about continuing to roll out and integrate more and more of these capabilities into the platform, When you look broadly across the drug development spectrum in the healthcare industry, where do you think are really the fastest or maybe nearest term and really tangible opportunities for some of these AI efficiencies that we're seeing? The real answer is everywhere, right? But over the near term and how we think about prioritizing, whether you're a researcher in an academic lab or even at some of the pharma customers that I know you all serve, where do you see as the lowest of these low-hanging fruit today?
Shawn O'Connor:
Yeah, I agree. The opportunities span the full continuum. One example of opportunity that is rich in an environment in which the drug candidates are becoming more complex, clinical trials are becoming more complex, therapies that are being developed when the batting average for successful Phase 3 clinical trials is as poor as it is, that complexity translates into cost, and so the cost of Phase 3 clinical trials is increasing. Challenges in terms of patient recruitment and other components of that call out for, how do we improve the batting average? How do we improve the success rate of those Phase 3 clinical trials if they are going to elevate in cost here? So biosimulation has been applied in protocol development and clinical trial design over time, and yet it is still untapped in terms of the efficiencies that can be impacted there in terms of selecting the right drugs to take into those clinical trials and then defining protocols that optimize the efficacy and toxicity trade-off and potential success for those trials.
Brendan Smith:
Yeah. I feel like a lot of these decisions that need to be made both on your side and frankly on the customer side, too, are in many ways tied to where the technology realistically is today. Every day we're hearing more about the blurring of lines between big tech, like Nvidia, Google, OpenAI, and big pharma, frankly, and especially as some of the pharma guys really start to ramp up their own investments into this technology itself. On that front, one thing we're asked a lot about is, does pharma's increasing investment in it come as a tailwind to where you all sit, or is it actually an either/or on their side? Where do you fall in that conversation, and do you expect the integration of AI, or maybe I should say how do you expect the integration of AI into all of this biosim capabilities ultimately impacting their decision-making on investing internally versus outsourcing or licensing?
Shawn O'Connor:
Yeah, fair question. Investment is picking up in terms of whether it's the third parties you reference or our clients themselves in AI. Where is that investment going? That investment spread out across a spectrum of efforts, but very focused today in terms of data management, the ability for a pharma company to get their arms around the availability of a wealth of information they have internally across many years of drug development, many drug programs, many drug candidates that can prove very valuable in terms of biosimulation. Getting that data into an accessible and curatable and usable format is a real tailwind in terms of biosimulation applications.
Their investment is not an investment in replicating the tools that we offer and provide to them. Their investment is in terms of building an ecosystem that can fit into... Pharma companies are very standard-operating-procedures-oriented, process-oriented, and one of the big movements in our industry is the movement of our clients to building ecosystems internally that integrate both the biosimulation tools and other AI tools and making them integrated so that they can follow a drug candidate from discovery through to approval and beyond, and those tools are easily accessible to the wealth of their participants in the program through that 10-year cycle timeframe. So these investments are very positive, both in terms of building our products into their standard operating procedures, making more and better data available as input into our tools, which leads to more accurate predictive simulations to support their decision-making process.
Brendan Smith:
Yeah, and I think your point about replicating what's already been done by external sources, by, frankly, they're already licensed partners, right? It's such an important point that sometimes I do feel like people miss when they're trying to dig into this because it's just like AI, that those letters now carry so much weight in some context and then so little in the way that it's often applied. So I feel like the way that pharma is spending their dollars, where that's actually going internally versus their license, I think it's important for people to remember it's not necessarily like they're not going to do something they don't have to do, especially if it's already readily available, commercially available, already a license that they can tap into.
In the same vein of the conversation, I think I would be remiss if I didn't bring up the latest FDA guidance that came out last week, and we've heard a lot about this over the course of 2025 altogether, really in line with the agency's efforts to start phasing out animal testing in lieu of some of these, what they're calling, new approach methodologies or NAMS. How has that recent guidance update, and in particular the one on the non-human primates last week, messaging out of FDA all year, really impacted either your business or your conversations with customers to date? What's been the industry's response over the last six to eight months?
Shawn O'Connor:
Yeah, no, I'll step back a little higher up and just say that it is a great example of what has been building momentum in biosimulation, a great example of regulatory support of further use of in-silico methods. It's what's driven the business over its 30 years of existence and will continue into the future. The NAM announcement with regard to animal testing is also a part of a number of FDA announcements of support of the use of AI tools broadly and specifically here in terms of animal testing, it's a raising of the bar. What I mean by that is that biosimulation has been used to date in the animal testing area focused in terms of predicting first in human results, characteristics of a drug, and that information used in the design of animal testing steps to reduce species populations, ensuring that the protocol for the animal test provides the appropriate output to take to the next level. The guidelines here raise the bar rather than just simply minimizing or improving animal testing, can we eliminate the animal test, and is certainly a raising of the bar, an ambitious goal that I think will have some results.
At the same time, we need to recognize these things don't change overnight. I was positively impressed with the announcement in April, getting the first iteration of the guidelines out by here in December That's pretty good timeframe in terms of the way things work in the regulatory world, so it's a real ambitious milestone that we're working towards. Clients are watchful and thoughtful, depending on where their programs are in the development phases. This doesn't mean things are going to change overnight, and in the next six months all animal testing is going to be taken away. Ultimately it's probably going to be a combination of certain therapeutic areas where the weight of evidence is sufficient to significantly impact animal testing, and in some therapies, that may not be the case. So this one will be one that builds over time. It's a new use case for biosimulation, a raising of the bar there that will create opportunity for us in the future. It's not necessarily a hockey stick that's going to occur in a short window of time. That's the nature of this industry.
Brendan Smith:
Yeah, and I think that's such an important point, too. These are the conversations we've been having so much over the course of this year, right? You don't flip a switch and all of a sudden mouse testing and NHP testing is just gone, right? It's a conversation about, if I'm used to spending X number of dollars and taking X number of years to get the amount and quality of data I need to bring a drug into Phase 1, and you can maybe spend 5, 10% less of that towards animals and use some of that last 5 to 10% to leverage some of these computational approaches and save some time, save some money upfront. It's really a conversation of maybe that's 90-55, maybe that's 80-20, and it just over time shifts, right? It's not an overnight sensation by any means.
But on that point, what do you think for some of the folks, irrespective of stage of development, but whether they're individual researchers or entire companies who are maybe dragging their feet a little bit on this, maybe they're entrenched in their animal studies or what have you, what do you think the industry really needs to see to drive a more concerted or tangible shift or monetizable shift really towards full-scale broad-based adoption of these NAMs? Is it just a matter of time? Do they need to start playing with it and see the money savings? What is that lever, you think?
Shawn O'Connor:
Yeah, it's a little bit of time, evolution, generational change. Science requires a lot of debate and evidence before the momentum changes. There's also the opportunity for, hey, when you develop better use cases, or I say examples of success, that have been driven by biosimulation, then that spurs on adoption. These situations have occurred in multiple scenarios. Bioequivalence waivers for formulation changes is an example that in our history where it took a while for people to gain acceptance there, but once it did, it was a flood of, "Hey, we no longer have to do clinical trials to make a formulation change in a drug post approval," and that created momentum in biosimulation.
Ultimately, the drug sponsor who can point and say, "Along the way from discovery, I wouldn't have identified the molecular structure if it had not been for AI approaches through to preclinical progress and elimination of an animal test, through into later stage clinical development, and here's what biosimulation did for us," and then finally the approval of a drug, and that weight of evidence that biosimulation is really what drove it through the process, it didn't take 10 to 12 years, it didn't take 1 to $2 billion, the more examples that we have of that nature, the more momentum and speed of adoption will follow.
Brendan Smith:
Yeah. Again, I think it's such an important point for people to start to wrap their head around. Maybe in that frame of mind, from where you stand in your conversations today, what do you think maybe the investment community in particular either underappreciates or misunderstands altogether, really, about integration of AI with biosimulation and its utility that you think is especially important for any investor to know heading into next year?
Shawn O'Connor:
We've touched on a couple of those points. A, biosimulation versus AI, it's not an either/or. Both are the same effort going forward. Secondly, the change dynamic in an industry that drug development is a scientific-based, regulated process, and the change dynamic is not overnight there. So in terms of an investor's view, this is not a hockey-stick environment that's just, can we time it right and tomorrow it's going to take off? This is one that's going to build slow over time. That slow, it's good growth, and the important factor there is that the runway is quite long here for long-term growth of biosimulation in the drug development world that supports a company like ours.
Brendan Smith:
I know we've covered some great ground today and it's a conversation I'm sure you and I will come back to for months and years to come, but before I let you go, one thing I like to ask all of our guests here is, if everything we've discussed today goes over someone's head, but they've made it this far and they're still listening at this point, what is maybe one point you would really want everyone listening and irrespective of what they do every day, irrespective of their background, to make sure that they really remember and take away from our conversation?
Shawn O'Connor:
I'll point to, it's wonderful to wake up every morning and be working hard to bring therapies to the market more quickly that help patients. So in terms of inspiration for myself, my company, and the inspiration for the use of biosimulation here, it's got a fantastic goal out there that we work towards.
Brendan Smith:
Well, that's great, and I think with that, I want to thank you for hopping on, Shawn, and talking us through what really is the cutting edge of this marriage between software and healthcare technology innovation. I'm sure we'll have plenty more to discuss over the weeks and months ahead as it pertains to all of this, so really thank you so much for joining.
Shawn O'Connor:
Hey, thanks for having me. Appreciate it.
Announcer:
Thanks for joining us. Stay tuned for the next episode of TD Cowen Insights.