Episode #122| The Mini Brain That Learned Pac-Man: NAMs with Dr. Lowry Curley

A brain organoid that can learn to play Pac-Man sounds like a stunt until you realize what it represents: measurable learning, rewiring, and human-relevant neural function that animal models often fail to predict. We talk with Dr. Lowry Curley, founder of Luna LifeSci and former CEO and co-founder of Axosim (now 28 Bio), about why neuroscience drug development breaks so often and how NAMs are finally giving teams better tools to make safer calls earlier.

If you’re building in biotech, investing in drug discovery, or just trying to understand what replaces animal testing next, this conversation gives you a clear map of the technology, the incentives, and the milestones to watch. Subscribe, share this with a colleague, and leave a review with the biggest NAM question you want answered next.

Show notes: 

YouTube: https://youtu.be/suVxiYZgPVE

Transcript

Welcome And Guest Background

SPEAKER_00 0:00

Please listen to the disclaimer at the end of this podcast. Hello, hello. Welcome to the Lattice Podcast. Our guest today is Dr. Laurie Curly. And he is the founder of Luna LifeSci. He is also former CEO and co-founder of Axosin, now known as 28 Bio. And Dr. Curly has a PhD in biomedical engineering from Tulane University. And he is an absolute expert in neuroorganoids. And we certainly will talk a lot about it. He has more than six patents. And he also is a member of 3R Collaborative and FNIH validation and qualification network. Welcome to the pod.

SPEAKER_01 0:44

Thank you so much for having me. Excited to speak with you.

SPEAKER_00 0:48

So the interesting story behind how we met is actually I encountered 28 Bio first. And that is one startup that kind of stuck into my mind because it can play Pac-Man, the video game using the Organoid, which is unheard of. I don't think anyone even knew this was happening. So, but before we go there, I like to just talk about the new approach methodologies a little bit. We hear

What NAMs Actually Mean

SPEAKER_00 1:14

this buzzword everywhere now on social media and podcasts. Would you like to just expand on the term a little bit so people can understand a little bit more what it is?

SPEAKER_01 1:24

Yeah, absolutely. So you are correct. New approach methodologies everywhere. NAMS. There's actually been multiple publications on what NAMS is because some people use different terminology. But essentially, what it is is any tool that can be used to generate human-relevant data in place of an animal. So this can be anything from in silico. We have incredibly powerful AI models that can mine an incredible amount of data. They can start to give us insights into safer, more effective drugs. And then after you have the data, you do need to prove that it works biologically. And this is where a subset of NAMS, which is called microphysiological systems, comes into play. So a lot of people think of organona chip, organoids as NAMS, they're a part of NAMS, but it's a lot more than that. Where I spent my time was the organona chip and organoid space. But as I'm a part of the validation and qualification network, some of these bigger organizations, I'm able to branch out and see how NAMS really encompasses so much more than I originally thought.

SPEAKER_00 2:33

You know, for people who are not in the scientific or pharmaceutical industry, the first impression they had about NAMS is like this is a way we try to avoid abusing animals or killing cute furry animals for testing. But what are some of the major reasons that we we need NAMS and kind of shift away from animal models?

SPEAKER_01 2:55

You know,

Why Animal Models Fail Humans

SPEAKER_01 2:56

I think that is important, right? Even as a scientist, no one likes to test on animals. But unfortunately, there just have not been better tools. And what the FDA has set up to do is make sure drugs don't get into humans that are unsafe and cause problems. And again, historically, we've not had tools other than animals to do that. With the advances in tissue engineering and bioengineering, we now have those tools to solve the biggest problem. 89% of drugs that are developed before clinical trials that are tested, you know, in the lab, tested in animals, finally get FDA approval to go into humans, 89% of the drugs fail. So I don't really know a lot of other places where you have a 10% success rate and keep going, right? But that's just been the way it is. And neuro, where I've spent most of my time, it's actually a 94% failure rate. So that's an even more difficult space that just shows you a mouse brain is not a human brain. And we now have tools that are more human relevant to be able to solve that. And as we do that, we're not going to replace animals tomorrow or next year. So I think it is important to remember that. They are still a part of it. What we're able to do is start to replace and run in aside animals. So you have to do less animal testing. And really, as that grows, as we prove that these technologies really are superior, I think that's when you'll see this replacement uh beginning to happen.

SPEAKER_00 4:27

Like you said, human brain is obviously different from rat brain, but across the board, 90% failure, almost 90% failure, is quite amazing that people still stuck with it for a long period of time. Where do you see, I mean, just looking at the failure statistics, where do you see the main reason animal models fail, other than the obvious that we're not exactly the same? Typical.

SPEAKER_01 4:53

I mean, most often it's they fail to catch a drug that becomes toxic in humans. So you'll put this into animals, you know, increasingly from mice to rats, you know, then beyond that, and everything says, oh, this drug is not causing any problems. It's okay, it's not hurting the liver, it's not causing any kind of you know, heart cardiovascular problems. But then when you put into humans, there have been a lot of examples where all of a sudden it's causing a big, big problem. And there have been instances where people have died because they missed something. You know, often people have really bad side effects, and it's just because they didn't catch something. There there was such a disconnect in the biology, no, there was no way of predicting what was gonna happen.

SPEAKER_00 5:43

I see. So it's not just an efficacy issue, but also safety issue. And interestingly, this morning I was doing some research about the history of animal testing. We actually have almost about a hundred years of at least in the US of preclinical testing. And it started off because there was some drug that was toxic to kids and they were given to pediatric patients and killed a lot of patients. And that's when the 1937 disaster was in fact the starting point of animal testing. So it started off because of a safety concern. So now we're evolving now to improve upon that, it seems like, for both safety and FXT issues. Now

A Seizure That Changed Everything

SPEAKER_00 6:24

you spent almost a decade or more than a decade in Axosim or now 28 Bio. It's interesting to talk to you because you have basically created a company from science to commercialization stage to the point where you were talking to 20 pharma, large pharmaceutical companies to adopt this technology. I think our audience would love to hear that story of how you bring the technology from the lab all the way to start talking about serious players in the field.

SPEAKER_01 6:57

Yeah, absolutely. It was quite a journey, you know, still is, I would say. But to take a step back of why I got in the field, and I'll say to build a company and to do all that takes a lot of work, takes a lot of effort. You really put yourself out there. And I dedicated myself doing this because I actually had a neurological issue myself. So in college, I had a seizure, uh, it was out of nowhere, no family history, no warning. Uh, and ultimately I was diagnosed with what's called an AVM. It's an arteriovenal malformation. So essentially, in one part of my brain, the arteries and the veins were so tangled up that blood flow got trapped in there. I'd had it my whole life, but for whatever reason, this one day, the pressure got so high that it caused a seizure. And you can imagine that that was quite a life-changing moment. And it just really had me start digging in and realizing we just don't know much at all about the brain. You know, the surgery I ended up having was just radiation. So they shrunk this thing so small the blood wouldn't go into it anymore. So they didn't cure it, they didn't know what to do. They fixed it, right? I'm okay. But that drove me to really question how do we learn more about the brain? How do we make sure we get treatments out there to solve issues much worse than that? And that's what led me to the field of tissue engineering, being able to build models. This was before Franklin NAMS and MPS. This was in 2007. I realized I could build models of the nervous system and start to really look at regenerative medicine. How does it react to injury? And I worked with a PhD advisor, and we were able to develop our initial technology, which is called a nerve on a chip. And that was the start of the scientific basis for it. As I mentioned, originally it was for regenerative medicine. So we were actually looking at if you have an injury to a nerve in your arm, for instance, it doesn't grow back. So how can we perhaps figure out some biomaterials to put in to allow that nerve to reconnect? Well, once we developed that technology to a certain point, pharmaceutical companies started coming to us and saying, well, that's really interesting for an injury model, but what if we tested drugs on that? You know, what if we use that to look at, for instance, a lot of cancer oncology drugs cause damage to the peripheral nervous system. So they can cause chronic pain in your arms, loss of use of your feet. We basically heard them, right? And I was crazy enough to say, hey, what if we do this as a company? And so what we went out and did, my professor, along with me, was talk to, I mean, we talked to dozens and dozens of uh pharmaceutical scientists and said, hey, you know, what are the problems you have? Essentially customer discovery. It's something that's really popular now. And we realized we had a product and there was a market for it. So we had this fit. And uh in 2004, we basically said, okay, let's take the plunge. You know, we're convinced there's something here. I'm convinced it can make a difference, and we started the company.

SPEAKER_00 10:11

But it seems like the decade-long of the company building, there are quite a few milestones. First, you developed in-house technologies, but also you also use the strategy of licensing, merger and acquisition strategies to grow the portfolio. Why don't you tell us a little bit about that?

SPEAKER_01 10:30

Yeah, no, absolutely.

Funding A Platform With Proof

SPEAKER_01 10:32

So obviously that was the genesis uh in 2014. And we originally built the company off of federal funding. So small business innovation research grants, STTRs, technology transfer grants. And so we were able to basically get the money, you know, specifically to commercialize these technologies to start. And so what we did there was we started testing drugs in this model to be able to start to say, we are more accurate than animals, right? We've tested all these drugs. We accurately said what was going to happen in humans. So that was the beginning of the company. And being in neuroscience, I told you that 94% failure rate. We realized we could grow beyond just this safety toxicity question to start modeling different diseases. And as first-time entrepreneurs, you know, we kind of thought we have this platform, it can do everything, right? Let's build it to do multiple sclerosis, let's build it to do ALS. And we realized you have to be realistic about your technology, know the weaknesses, know the strengths,

Licensing And Acquiring To Scale

SPEAKER_01 11:36

right? And so we said, okay, we want to be the organ ownership company in neuroscience. How do we do that? And this is what led us to look and say, okay, what other technologies are out there at universities? Uh, what other companies are out there, you know, that might have opportunities. And that led us first to license a technology from Johns Hopkins University, a lab of Thomas Hartung is the leader of that lab. He is really a pioneer in the NAMS space. And that was a brain organoid model. So that was the first time we really truly got into the brain. And then we continued to grow with that model. You know, we would show this to pharmaceutical companies. And whereas with the safety side, they said, okay, prove it to me. You've tested all these drugs. Okay, I believe it, you know, and they needed models so bad they were willing to go out on a limb, be early customers, help us develop it alongside them. But then when we started talking to the efficacy disease modeling people, they said, we are so desperate for a solution. You've got data that convinces us. Let us help you develop it. So they were really a much lower barrier to entry. Uh and there's unfortunately so many diseases of the nervous system that gave us a lot of opportunities. So, you know, we grew off of those two platforms for many, many years. And then we were always looking around what else is out there, talking to customers, what else are you looking for? And we had the opportunity in 2023 to purchase a company that had been a competitor of ours called Stamonics. They rebranded as Viant Bio. And so we did a lot of work to say, is this technology complimentary? You know, is it going to cannibalize the other products that we have, or does it address its own market? You know, we we brought the team on board. We did a lot of work to say, is there a culture fit? So we we really did our due diligence with the acquisition and and realized very quickly, like, yes, this is absolutely a good fit. And I raised money from investors to acquire the company, and we onboarded this team and this new platform, and we're able to even further expand our reach within the neurospace.

SPEAKER_00 13:54

So at

Explaining Organoids To Skeptics

SPEAKER_00 13:55

this point, I want to talk a little bit about the technology of solved organoid. Organoid is obviously not an organ, is some people say a clump of sales. How do you explain organoid to people who are open to adapt? Let's say people who are very used to the conventional 2D culture system.

SPEAKER_01 14:15

So it is interesting. There's actually two customers we found ourselves convincing, right? There was the people who are used to 2D cultures, and then there's the people who are used to using animals. And they asked very different questions. So, you know, you you mentioned that, and it was interesting for us to learn, you know, different people were worried about different aspects. You know, and for a lot of the 2D people, they knew where their limitations were. And so we didn't necessarily have to convince them too much, especially in neuro, because there's just there at that point there were so many studies that if you do it in 3D, you know, the neurons are able to interact with each other in a way they just can't do in a petri dish. And so what we had to convince them there, when you do this in an organoid, it's more expensive and it's lower throughput. So you can't test as many drugs in the same amount of time. And so we had to convince them that it is worth slowing down a little bit, spending a little bit more money to get that data. And we were able to convince them of that. And then, you know, with the in vivo people, a lot of them just were so entrenched in that that we again had to sell them with the data, but often it was a different piece of the data to say, we know now where the limitations are, and we had to ask a lot of questions to understand not everyone was going to buy it, right? But where is the biggest need? Where is the biggest gap? And then what data do we need to do to generate to convince that customer? I don't understand this technology yet. Okay, I know it's just a piece of an organ, right? It's not an entire brain, but I think it can solve a problem that nothing else can right now.

SPEAKER_00 15:56

Yeah, you brought up a good point, which is probably a lot of people understand that organoids is more superior model than a 2D culture. And even in our cohort of startups, there are a lot of investors question the comparable metrics between animal, the in vivo testing, and organoids or any of the NAM system. Just from your perspective, can you just dig a little bit deeper on what are some of the classic examples people ask? And some some of the features that you figure out is actually more superior in organoid models.

SPEAKER_01 16:31

So it's

Electrophysiology Readouts That Matter

SPEAKER_01 16:32

a really good point and a good question. Just because you put all these cells into an organoid, that doesn't necessarily make it better. You really we had to spend a lot of time making sure not only could we get the right cells into it, but that we could get them to organize in a way that was more human relevant. So it really takes a lot of engineering to get the nerve the neurons, the brain cells to connect the right way. And it wasn't just having one cell type. We had to add astrocytes, the supportive part of the brain, and a lot of work went into getting the ratios of that correct. So, first, it's not trivial to build the right biology. We spent a lot of time doing that. Then the question is, okay, what data am I gonna extract from this? What are the endpoints I'm gonna get? There's limitations in 2D, what you can look at electrophysiologically, which in the brain, that's what matters. In 3D, we were able to look at how do each of these neurons talk to each other and then extrapolate that into an endpoint that looked more like a human endpoint. So we were able to correlate apples to apples. This drug is doing this, maybe causing a seizure in our mini brain. You saw seizures in humans, you know, there's a correlation there where 2D, you could not model that. And in animals, they'd already shown the animal just didn't catch that. And then there are other ways we're able to track that where we can study an organoid across time. So you can add a drug for weeks and weeks and test that continuously, where in an animal, you can't do that. You have to, you know, unfortunately, in the experiment, sacrifice the animal for every single time point. So we easily were able to prove we can do things that you just can't do with them.

SPEAKER_00 18:19

And certainly a mice cannot play the Pac-Man.

SPEAKER_01 18:23

And my that is very true. Yes. Once we start getting those kind of readouts, then you're not talking about how are these neurons talking to each other, but what does it mean the way they're talking to each other? And so those are really the ultimate endpoints that start to look more human relevant.

SPEAKER_00 18:39

And my high-level understanding of 28 bioles technology right now is a combination of organoid and microphilic chips. And also you have some visualization tools and, like you said, electrophysiology detectors to get these data. Can you just like make maybe uh give a more professional explanation of your current platform?

SPEAKER_01 19:01

Yeah, no, absolutely. So it starts with the biology that I've been describing. We have these organoids, we can make thousands of them at a time. They look the same every single time. So that was where it started. That took us years and years and years to develop. Along the way, we also developed customized embedded electrode arrays, is what we call them. So the on a chip part for us actually isn't microphluidics so much because the brain doesn't really feel the same forces as, say, the heart, you know, or the lung. But what that allowed us to do was measure the electrical activity at multiple regions of this brain. And that's not a trivial thing to do. So we built the 3D biology, but how do we make sure we're measuring every single part of that? So we developed that actually for our nerve on a chip technology originally to model the peripheral nervous system. But we realized if we put these brain organoids on it, we could talk to and record from the brain in multiple different areas. And so once you're

How A Brain Organoid Learns

SPEAKER_01 20:04

able to do that, you can essentially stimulate a part of the brain and then measure how the rest of the brain reacts. For something like Pac-Man, you know, you have all of these sensors essentially on this miniature brain. And if you stimulate one particular part, you hook it up to a computer, and you know, essentially what you can do is teach it if I do this particular movement, I get shocked, right? And if I do a different movement, I don't. And that actually causes the brain to rewire where that then the fundamental building block of learning.

SPEAKER_00 20:42

That's fascinating. No, but would you say that neuro is more advanced in terms of the microphysiological system than the other organ systems?

SPEAKER_01 20:53

I wouldn't necessarily say that because I would argue it's the most difficult of all biology to develop. Something like a liver on a chip or a liver organoid model, people have been working on those for a very, very long time. And so there is a lot of data out there proving the predictive nature of it, the accuracy. And with the brain and the nervous system, it took us a long time to get the biology right, to make it look as close to a brain as we can get it. But there's still such a far way to go to get that to really model the brain. So we have it now as one of the basic building blocks of the brain. And I think we've done a great job of that. But if we want to start modeling more complex diseases, if we really want to teach it to play uh, let's say Minecraft instead of just Pac Man, you know, there's a lot more work to go. So it's come a really long way. What we've built, because the gap between animals and humans is so big, what we've built has been adopted in a big way, but there's still a lot of growth, I think.

SPEAKER_00 22:01

Yeah, just

Organoid Intelligence And Ethical Lines

SPEAKER_00 22:02

something fun on the side is uh when I was doing my research for this interview, I came across a paper talking about organoid intelligence. There's apparently an organoi intelligence community out there. It's kind of crazy. Uh you want to explain what that is, just really briefly for the audience?

SPEAKER_01 22:19

That absolutely, yeah, that is a a big audience and a growing audience. So if you think about what I said of making these things play Pac-Man, what we're doing is we're stimulating one part of this brain, and then ultimately we're we're shocking it in certain ways to tell it you did something good, so a reward, you know, or you ran into a wall, you did something bad. And that causes this miniature brain, the neural networks, to rewire. So it literally changes the way they connect with each other. That's the building block of learning, like I said. And so you can pretty easily extrapolate if you make these things complex enough and they're able to rewire and learn, you know, where is the threshold for intelligence? Uh, and I think the community is doing a great job of debating that, discussing that, making sure it stays ethical. But it's something that you do need to talk about. And I'll say one of the other aspects of organoid intelligence, our brain is the most powerful computer, the most powerful processor out there. So if you can harness these mini brains to utilize their processing power, you know, you can start to replace some of your traditional power grid, electrical generation, fossil fuels, all of that. And with AI, you know, we just need more and more storage, more and more processing power. So can you do that in this small scale that is more powerful than anything we've built before?

SPEAKER_00 23:52

Yeah, I always said our brain is a lot more powerful than the quantum computing that people are working on right now.

SPEAKER_01 23:59

Yes, absolutely.

SPEAKER_00 24:00

Probably underpaid relatively to that.

SPEAKER_01 24:04

That's safe to say, yeah, where a computer can do a lot, you know, basically in parallel, it's not able to iterate the way that we are. It's not able to have that complex learning. And, you know, that's just something that we can do much more energy efficiently.

SPEAKER_00 24:20

Yeah, fascinating. And, you know, when

FDA Pathways And ISTAND Momentum

SPEAKER_00 24:23

you were talking about the story of founding 28 Bio, you mentioned some timeline, which I think it was interesting. You founded the company in 2014, right? That's what you mentioned. Yes. But actually, I think the government, the US government, started to be very interested in shifting away from animal testing, I guess, as l as early as 2000. And then 2020, which coincidence coincidentally was the pandemic timeline, they started this I stand, right? At innovative science technology alternative. Or I can't really I forgot what it stands for, but that's basically the is it it so and then you also have these customers who you say are desperate. Are you are they desperate because the failure rate is so high, or do you think the policy is also driving part of that need? I think.

SPEAKER_01 25:12

I think right now it has been because they have a problem that needs to be solved. And they're able to go a certain amount of the way to uh using these in conjunction or instead of animals for internal decision making. So they don't go to the FDA or run FDA studies until they've chosen, let's say, the one version of their drug that they're convinced is going to work. So they adopted technologies like ours early on to work from thousands to hundreds to tens of versions of a drug and select the one. So that's why they were able to adopt that originally. But it becomes a much bigger market when you get into that level of study that everyone is required to do for the FDA. So there's some pharma companies that were comfortable using this before the regulations change, and there's some that are waiting on that. And it's taken a long time to get there. You know, the ISTAN program, like you said, it has existed for a while. It's just now hitting its stride. So this is a way of the FDA to work with companies like 28 Bio, you know, like LiverChip, Lung Chip, and basically say, let's design a study together to say, if you generate this data and it shows that you are accurate, then we will say this is cleared to be in the FDA filings. So then the pharmaceutical companies have the green light, right? They know that the FDA is going to look favorably, assuming their data is good, on it. And so that's going to really open up the floodgates, I think. It's still, you know, early. They had their first company go through it. It was a company called Integral Molecular. They designed this experiment, they uh ran all the data, they worked with the FDA to analyze the data, and they basically said, okay, you can utilize this. Uh basically, they want to look at, you know, is a drug going to hit the wrong tissue? So tissue cross-reactivity. So it's a relatively simplified compared to an organ on a chip. But uh the first company called Emulate got their liver on a chip, basically their LOI letter of intent accepted in 2024. They worked with the FDA to design the experiment. They launched that, formally generating the data. And from what I hear, they're packaging up the data right now. So I think we're we'll have our first MPS model, uh, you know, having gone through this program, and and there are many other companies now working on this. CN Bio, there's a consortium of liver chip companies, and so, you know, I think the floodgates are are getting closer. They're starting to open, let's say. And and yeah, it takes time, but biology takes time, and the FDA kind of understandably does need to be conservative because there's real implications if you know they they use the wrong tool.

SPEAKER_00 28:13

Absolutely. It feels like there is a acceleration in the field in the last couple of years. I mean, 2020, okay, 2020, we had I stand, 2021, we have the FDA modernization 2.0. I'm not sure what 1.0 was, but uh and then just almost right now in July 2026, we just had a modernization act 3.0. And then in April 2025, there is this roadmap put out by FDA that suggests we need to just off-shift to the SNAMS in three to five years, which is extremely ambitious in my opinion. But I'd like to hear what your thought on this is. I mean, clearly the policymakers are trying to push towards this direction. Do you think one, scientifically we can get there? Two, is the commercial space actually ready for it? I mean, are people writing checks to buy these models?

SPEAKER_01 29:06

So I'll start with uh, you know, just addressing in general what you said. The last two years, it's been incredible how much work has been done, how much policy has changed. And that I think is the first step towards opening these regulatory floodgates. Yeah, you mentioned, you know, in April of last year, they had this roadmap that really spelled out we want to, for certain applications in three to five years, you know, have these human relevant, not wholesale. I think that's is something just to make sure people talk about because it can be a little bit of a trap if we say, oh, we're gonna be done with animals in five years, because it gets expectations really, really high when the reality is let's pick the places that we know animals don't work and let's plug in tools that do work there, generate momentum. Uh, and that, yeah, that's that's exactly what's happening. So, really, I would say that laid the groundwork. But then there was still a lot of the foundation that needed to be built, a lot of the infrastructure that had to grow to be able to accommodate that. Because, of course, as as you said, it's not so easy as just, okay, we're gonna flip a switch. But what I'm really excited about is in March earlier this year, the FDA built on that. So there's a roadmap that's really important, that gives us the 10,000-foot view. They released a draft guidance, which may not sound as exciting as a roadmap, but it's actually incredibly powerful. And there they spelled out to pharma companies if you want to utilize these organs on a chip, you can. We've already said that. Here's how we want you to choose. And a big part of it is context of use. So you have these models, as we talked about. It's not an entire brain, it's not an entire liver, right? It's a it's an organoid or it's that on a chip. But there are some areas where that is maximally effective. And so that's context of use. So find the context of use where this tool works better than anything else that's out there. Once you've identified that, write up what that means, then you know, design your experiments to prove to us that it works. And once you've spelled that out, you can go to the FDA and make your case. So that is a way to do it as a one-off. I stand that we talked about as a way to do it all at once, but pharma can can already go to the FDA and say, we're convinced of this model. We know exactly where we're applying it, and the FDA can say, okay. And so I think that was a really big moment. April of this year, the FDA actually came up with their one-year progress report from the April announcement last year, and they said, you know, we're getting the regulatory agencies in place that we're all talking to each other, we're all talking the same language. And so they're on track, which is which is amazing. I mean, they've opened the standardization of Organoids Center, and they've a group called Arriva or an institute called Arriva that's coordinating all of this. And and so the infrastructure is there. And then, you know, your next question was is the commercial need there? I would say, yes, there is enough commercial desire that companies like 28 Bio and others have been able to build to a certain size. And it's a really good size. You know, these companies are generating significant revenue. But until the FDA gives the green light, there's a little bit of a threshold, right? But I think we're on the cusp right now of the FDA for some of these tools saying, okay, we're allowing it. And that's when every single pharma company starts to use it, right? Not just the ones who are progressive. So we're we're on the cusp of, I think, that growing exponentially.

Startup Playbook For Faster Adoption

SPEAKER_00 32:56

So if you're a startup founder now in the space, what would you do to kind of help acceleration along?

SPEAKER_01 33:04

That is a very big, uh, very big and broad question, you know. But what I'll say, what's difficult to do when you develop tools like this that can do so many different things? It's really difficult to pick the one application or the one disease that you can do better than anyone else, right? With with our nerve on a chip and our brain organoid, you have the ability to do Alzheimer's, ALS, multiple sclerosis, Parkinson's. You can't do them all at once. So what I see the successful companies doing now is starting to hone in on what they do best, what they do better than anyone else, and develop that. So that's where I would start. If you're uh you know a startup or you know, developing new technology in this space, find what you do better than anyone else as quickly as you can. And then now we've moved past, okay, prove to me you have the biology, so it looks enough like a brain that you can convince me to prove to me that you can run the same study 10 times and get the same drug. So now people are worried about reproducibility, scalability. Are you building in features that lets you test 10, 100, 1,000 drugs as you grow? So it's really focusing on different aspects of what we need for full integration and full-time adoption of these technologies.

SPEAKER_00 34:24

Well, in addition,

Data Sharing And The Devil’s Case

SPEAKER_00 34:25

in addition to technological startups, uh there are also big organizations, some big CROs, pharmaceutical companies. Who do you think is gonna be the ultimate winner of this entire space? I mean, just the wild gas, I guess.

SPEAKER_01 34:42

You know, I don't think there's gonna be one winner, if I'm if I'm really being honest. Uh I think that within pharma, you have some institutions, some companies like Roche, Genintech, uh, Black So Smith Klein, you know, Takeda, some of these have been ahead of the curve in adopting these technologies. They've adopted them, you know, they've validated them, they have experts now who can run these studies. And so win to them, a win to them, looks like getting better drugs to patients, faster, better chance of success, right? So now that they've teed this up, they're starting to put drugs into the clinic that they chose based at least in part of some of this data. So there are some of them that are going to start to win in a way that benefits all of us. You know, when you talk about all of the different organ-own a chip companies, because there are many of them, right? Everything from heart owner chip, liver owner chip, kidney owner chip. We always wondered what would be the ultimate exit, right? How are we gonna make money, make our investors money? One thought was there would be a roll-up where there would be kind of an amalgamation or you know, all of these would come together under one umbrella. That hasn't happened yet. I think it still could. Then the other, like you said, is there could be a CRO who comes along, scoops all of this up so that they can own the space. We've yet to see which one of those is gonna win. But I think once the FDA really makes their decisions, things could move, you know, move very, very quickly.

SPEAKER_00 36:19

Yeah. I'm glad that you said the w the word roll up before I did. Um you clearly have been thinking about it when you were even the founder of 28 Bio because you were already starting acquisition as a startup, which is rare for startups. Uh, but you realize that you need to grow in terms of the portfolio and I guess diversify or roll up. So yeah, I think I think this is uh actually a question a lot of the investors also raised as well is if the industry needs a roll-up. Now, I think in 2026, stem cell research and therapy named three obstacles towards NAMS. One is uneven validation across toxicology endpoints, two is incomplete global data sharing, and three is culture inertia. What do you think is the biggest blocker of the three?

SPEAKER_01 37:13

You know, I think the cultural inertia, I think we're getting over that. I don't see that as the biggest, uh, the biggest issue now. We've talked about all of the things the FDA is doing, you know, NAMS, bringing this into formal programs to identify these. The UK also uh made their own announcement of a strategy in November of last year, putting money behind this. So I think the writing is on the wall enough that every pharma company is at least looking for a strategy here, even if they haven't jumped in wholesale. You know, the data sharing, I think, is a really good point because if you're going to get this accepted by the regulators, not just in the US with the FDA, but in Europe too, right? In the UK and Japan, those regulatory bodies are talking to each other, but they're kind of one-offs, right? There's not a formalized body yet that is sharing this data, saying, hey, we've, you know, we have qualified this liver, let's say. If if I'm Europe, you know, I can hear the US and say, oh, that's great, but I have to see the data myself, right? I have to make an independent decision, informed, of course, by the FDA. And right now there's not a direct pipeline. And that is going to be really critical. I will say that is being developed. Those institutes and those countries are coming together and talking. I was just at the MPS World Summit and I heard all of them saying, here's how many NAMs we've seen, you know, here's how many we've accepted. Interestingly, Australia has jumped out to the forefront in sort of opening up the regulatory pathway to make it easier. So, you know, when we all start working together even in a bigger way, you know, I think that data is able to be shared, the wins are able to be shared, the decision making is gonna happen faster. And so I would say really the data sharing and the decision making is the hurdle now from a regulatory perspective.

SPEAKER_00 39:20

And when you say data sharing is about the regulatory, it's between the regulatory bodies or is uh among the companies that you're talking about?

SPEAKER_01 39:28

Uh this is regulatory bodies that I'm talking about in this case. You know, when we're working with pharma, largely they're not talking to other pharma companies because things are proprietary. On the talk side, it's pre-competitive, so that's a little bit different, but we've published with Takeda, you know, we've published with some big pharma, and so they will share some of the data, but not all of it. So that's a slow path because they just have to be careful. And so in this case, yeah, I'm talking about the regulators. Um, you know, the more pharma we'll publish, the better, without a doubt. But I think as soon as the FDA starts allowing these and encouraging these with a formal path, then you know, we're we'll start breaking down walls quickly.

SPEAKER_00 40:09

Now, if you were to argue against NAMS, to say we still need animal models, what would those arguments be just to play the devil's advocate?

SPEAKER_01 40:19

Yeah, no, and it's an important point. I said before, right? I think we can't let the hype get too big. If we just wholesale replaced everything, you know, with organona chip with NAMS, that can be quite uh, you know, computational as well, we would still not be at 100% prediction rate. We might fall off a little bit. And I say that because a mini-brain, a brain organoid, a liver organoid, a lung organoid, that's not an entire system. So you do still need situations where you take a drug, you metabolize the drug, the drug flows into organ A, then organ B ultimately goes to the brain. But to get to the brain, it has to go through the blood-brain barrier. And we just can't recreate all of this. So you have to know where the tools' strengths are, plug them in. But then also we need to be very, very open-eyed, clear-eyed about where the weaknesses are and build the tools before we plug them in there. And that is where we will still need animals. So I would say let's make sure we're using these where they need to be used and know when animals are still needed for now.

SPEAKER_00 41:33

Now, if you were to say what one thing the NAM community could do to make things better or improve faster, what would that be?

SPEAKER_01 41:42

So, you know, as far as what the NAMS community themselves can do, I think that's difficult to say because we have to remember these are startups, right? We're having to raise money from investors. They want their money back. The field's been around for a while. Maybe it's taking a little longer than we thought for wholesale adoption. And so you can't expect, I think, the MPS or the NAMS companies to do it themselves. So, what I'm encouraged by is MPS companies are being sort of brought along by regulatory bodies, we said, and those are starting to understand we need to give grants, we need to give these companies money to be able to run these. But then there's also non-governmental agencies like the Foundation for the National Institute of Health, which is not officially affiliated, their validation and qualification network, which I'm a part of. So that's actually bringing in MPS companies, NAMS companies, and saying, all right, let's design experiments together, not just with regulators, but with pharma, with CROs. And so now you're seeing these big adopters looking at the data in real time. So what I'm seeing, and I would encourage more even for emerging MPS and AMS companies, find these programs, engage with these programs early because you're giving people a sneak peek at your data and you're just gonna open your market quickly. Easy to say, but hard when you're a startup and you only have so many resources.

SPEAKER_00 43:12

Absolutely. Now, this is a good segue into your current role as uh of advisor to startups. Why don't

Luna LifeSci And Founder Focus

SPEAKER_00 43:21

you tell us a little bit about Luna Psy Luna?

SPEAKER_01 43:24

Yeah, no, absolutely. So Luna Life Psy, I started after I left 28 Bio, and really I brought my experience, everything I'd learned 10 years of building a company, realized pretty quickly I could see around corners that some of the companies I was working with couldn't. And what I was able to do was found this consulting firm, and I work with other entrepreneurs, founders, you know, technologies, not just in the NAM space. I realized quickly diagnostics, uh, even medical devices, they face a lot of the same. Hurdles. So I'm able to work with these companies and say, hey, let's develop your commercial strategy now. Let's develop your go-to market. Let's develop your market positioning now when maybe they wouldn't have the time or the experience to do that. They would figure it out, but it might take a lot of time. I say, hey, work with me. I'm going to take some of the lift off you. I'm going to bring my experience and I'm going to get you where you need to go faster. So I'm really enjoying that close, you know, staying very close to the technology entrepreneurship, but getting to see a lot of different technologies, work with a lot of different founders, which I really enjoy. And then that also lets me work with some of the regulatory bodies, like the validation qualification network I mentioned, uh the three Rs, where I'm able to see holistically what are we doing in NAMS.

SPEAKER_00 44:50

Now, not to give all your seekers away, but uh if you were to say, you know, the one mistake that you have made yourself and you like to warn entrepreneurs about, what would that be?

SPEAKER_01 45:01

Focus. 100% easy answer. Focus. When I say I work on go-to-market commercial strategy, what that means is find the customer who has the biggest need, find your value proposition. So how you can sell to them and then develop the data, the marketing, the pathways to sell to them as fast as possible. You have a tool that can do 20 different things. It's so easy to try to do 20 things at once because you have 20 different markets. You have 20 different types of customers. Sure, but which customer within each of those markets are you really selling to? How do you convince them? How do you get to them? I help focus, drill into the one that you can get to the easiest, and figure out how do I have reproducible revenue, get that scaling, and then move on to the next market? You can't do it all at once and you shouldn't. So focus is the easy answer there. Easy to say, not easy to do.

SPEAKER_00 46:00

That's an excellent advice, by the way. Uh now we're reaching the end of this interview. I

Milestones To Watch Next

SPEAKER_00 46:06

like to hear what do you think in the next couple years, in terms of the NAM space, what are some of the milestones to watch out for that you think could be irreversible for the industry?

SPEAKER_01 46:16

I mean, I think the biggest set of milestones really is going to be coming out of I stand. Uh, this is where you're gonna get a laundry list of technologies, multiple companies, multiple types of technologies that the FDA says, okay, we'll accept data from this. We believe that data, and pharma's gonna start using that because it is cheaper, you know, it is faster, it is more predictive, but they're scared to go out on a limb. So I'm I'm excited to see who the first kind of bolus of companies and technologies are gonna be there. And then I know that there are a lot following suit, a lot of them that haven't announced it yet. So I'm excited to see what that steady uh train of technologies looks like. And then also I think the validation and qualification network I talked about, they're doing the same thing, but directly with pharma. And so it's faster, right? It the the FDA is conservative, they need to be, but this is your big pharma, you know, your big CROs working with companies. And so as these companies, they've announced seven different uh platforms now coming through that program. So it's a good number. And they're already generating data. Pharma companies are already saying, okay, we're gonna start working this, we're gonna internalize this. And so you have an increased adoption coming both preclinically and then from the regulatory side. So I'm excited to see which technologies take the forefront there.

SPEAKER_00 47:41

And and how do a startup be able to participate in some of these organizations?

SPEAKER_01 47:48

So, I mean, you just have to watch for their calls for applications. You know, a lot of times there are grants attached to it, so it's money. You know, the the FNIH validation qualification network, VQN, they put out posts when it's time to apply for the next group of technologies. I stand, you just need to reach out to them. Um, you know, the NIH, NCATs, they all have grants that you can you can participate in. So really you just need to be proactive about it and and watch when these opportunities are coming because they're coming a lot more quickly.

SPEAKER_00 48:21

I I know it sounds it feels like surfing right now, the big wave is coming, so you better catch that. Thank you so much, Lori, for this very informative conversation today. I'm sure a lot of people will really appreciate the education. And how do people get in touch with you?

How To Reach Dr. Curley

SPEAKER_01 48:36

Yeah, so first, thank you very much. I really enjoyed this. Uh, if people would like to get in touch with me, I'd love to hear from other people building companies, technologies, whether I can give some advice or formally help. Uh, but my email is laurie.curly at lunalifesci.com. You can see my name for how to spell it there. It's a little bit different. Uh, or you can go to lunalifesci.com and see how we can help. But always enjoy, yeah, speaking to the next generation of entrepreneurs and scientists.

SPEAKER_00 49:06

We were sure include those in our show notes. Thank you so much, Laurie. Thank you, Jenny. Talk soon. Bye-bye.

Disclaimer

SPEAKER_00 49:14

This podcast is for educational and informational purposes only. The views expressed do not constitute medical or financial advice. The technologies and procedures discussed may not be commercially available or suitable for every case. Always consult with a licensed professional.