Investing In Tomorrow's CNS Breakthroughs With Breyer Capital's Bret Bostwick, M.D.

0:03Bret Bostwick, M.D.: Welcome back to the Business of Biotech. I'm your host, Ben Comer, Chief Editor at Life Science Leader, and today I'm speaking with Dr. Bret Bostwick, M.D., Venture Partner at Breyer Capital, a global venture capital and private equity investment firm. Bret is a physician scientist and board-certified medical geneticist who has worked in academic medicine at Baylor College of Medicine and Texas Children's, and he's worked in the life sciences industry. At Alnylam, he co-invented CNS-targeted RNA therapeutics and has leadership experience in drug discovery, translation, and clinical development.

0:41Bret Bostwick, M.D.: In addition to his role at Briar Capital, Brett is a principal at Translational Medicine Solutions and an AI advisor to Meta. I'm excited to have Brett on the show today to talk about his work in CNS and what's happening in the category, his role as an investor at Briar Capital, how to avoid mistakes during the drug discovery and translation process, and where AI can help. Brett, thank you so much for coming on the show. I appreciate you being here. Ben, it's a real pleasure. I have been a long, long time fan of your podcast, and so it's wonderful to be a guest. Oh, well, thank you so much. I'm thrilled to have you.

1:19Bret Bostwick, M.D.: I wanted to start with your background, Brett. I'm curious how you would describe your experiences as a physician and professor maybe first, and then what led you to move over to the life sciences industry as a director of clinical research at Alnylon? Yeah, I mean, let me actually take you back a hair further because I think where like my passion first started was the Human Genome Project, right for many of us medical geneticists so take you back to year 2000 um the human genome rough draft is published and it's 3.2 gigabases of information and i remember at the time i was a teenager and i was thinking about my small town library right and that was about the amount of information in my small town library 3.2 gigabases yeah and just thinking about how like that amount of information sits in every cell in your body, of which there's trillions of cells.

2:16So I kind of first became like obsessed with this scale, just like the idea that we all had a recipe book. And so then the question was, where is the right application? Of course, in like year 2000, we thought, you know, all diseases would be cured by the human genome, something that's maybe like reminiscent of the technology of today.

2:35Bret Bostwick, M.D.: But really, you know, then it was diving into figuring out what all these genes do. And so I joined a fruit fly lab when I was a teenager and started thinking about what are the basic applications of genetics and medicine. And then obviously, did a bachelor's of science in genetics and then headed to medical school where I thought the best application layer here would be in human medicine. I had the great opportunity to go to Baylor College of Medicine for residency. And that is, I still think, the premier place to train as a medical geneticist. You've got HHMI faculty advising you. It's a real kind of incubator for cutting edge what's going on in genetics.

3:19They also have Baylor Genetics, which is a diagnostic sequencing lab that, you know, runs like a company. So while you're training there, you also see kind of the company building principles around how technology can turn into revenue models and you can build a business. So I stayed on as faculty at Baylor College of Medicine and was part of the undiagnosed disease network, looking at these rare phenotypes and trying to figure out, mostly in children, what was the underlying drive of their disease. We quickly, at that point in time as a field, got to this awkward limitation where we were able to pinpoint the exact reason that a child had a disease, for example.

4:03We'd be able to find the typo or the misspelling in the gene, but we were kind of unable to do anything about it. And you could see we were just having the tools that could cure that same disease in flies and in mice, but we just didn't have the ability to translate these tools to humans. So that's what got me kind of, you know, really obsessed about what is that translational gap that we have? Why are so many mice cured and so many people aren't? And so I realized that it's really hard to do drug development from within the walls of academia. And so I had another great opportunity to transition to alnylam pharmaceuticals.

4:43I went there right at the point of their first FDA approved drug and was there, I think, in a timeline of about five years, there were five alnylam invented drugs that became FDA approved. So really had a chance to see like the A to Z drug development process while I was there, while also working on an area in the central nervous system, which was more like a startup within that ecosystem. We were we were designing the first siRNA that was going to target the human brain and thinking about, you know, really big translational challenges in the field.

5:19Bret Bostwick, M.D.: Yeah, I'm sure that must have been a really exciting time with, you know, so many new discoveries coming in. And I have to ask, though, before we move on, what lab did you get into as a teenager, a fruit fly lab? Yeah, I was in Barry Gnetsky's lab and he was he was a fantastic mentor. And he classified some of those early channelopathies or those genes that are important in human seizure etiology or cardiac arrhythmias. And believe it or not, those those same genes exist in fruit flies and they have seizures just like the people. And it's a much easier model system to work in. Yeah. So you just knocked on the door as a teenager and asked if you could help out.

6:01Bret Bostwick, M.D.: Like, how did that work? Yeah, pretty much. That's exactly what I did. Yeah. Yeah, that's it. Yeah. And, you know, the initial work was, you know, was a lot of flipping flies. So moving them from one agar-based vial to another so they could have their growth media and trying to do both, you know, prospective and retrospective mutagenesis, right, where we induce mutations, then we classify phenotypes and try to isolate the gene. Back then, it was very hard to find what gene was driving a phenotype, unlike today, where we can kind of sequence all of it. Yeah, right. Well, I want to get back to CNS in just a second.

6:39Bret Bostwick, M.D.: I think it's a really fascinating therapeutic area, especially right now, the last couple of years, what's been happening in that space. But just to get us kind of up to your current role, after Alnylam, you worked as both a consultant and an advisor, and now as a VC investor. Were there things that you learned at Alnylam, Brett, that you wanted to share with other developers? Can you maybe describe an example of some of the consulting projects that you worked on at Translational Medicine Solutions are that you're still working on? Yeah, I mean, the list of things I learned at Alnylam, I mean, it was an incredible place to be.

7:23And as a medical geneticist, it was like a kid in a candy shop, because Alnylam focused on genetically validated drug targets. So I've learned rigorously what that actually means. It's a term thrown around a lot today, but it really is a way of changing what is a traditionally 90 % failure rate in our field of pharmacology into a much lower rate of failure. And that is a rigorous understanding of what human genetics teaches us about when you drug a particular target, how effective that might be for a particular disease and how safe it might be. So that was like a huge learning. Also important, I think equally important now is the culture.

8:09You need the right group of people. So fantastic science is not enough. There is outstanding science all the time that falls off the tracks and the company dies because they don't have the right people around the table or they don't have the right culture in place. So I learned at O 'Neill and that kind of always keeping your eye on the patient at the end of this process, developing a shared vision within a team, and then being extremely thoughtful around how you compose that team and the team members in it is as crucial of an ingredient in success as the chemistry or the IC50 or all these other metrics we talk about in drug development.

8:50So those were two big takeaways. And then also the impact of personal mentorship. I mean, John Maragonori was the CEO when I was there. And Akje Vaishna was president and head of R &D. They remain good friends today. And they provided absolutely outstanding mentorship to not only me, but dozens, hundreds of people who are now spreading their wings in biotech.

9:14Bret Bostwick, M.D.: On the first piece about target verification, I want to ask one follow-up on that. You know, from, I guess, a scientific perspective, when is target verification simply mean that you have this hypothesis, we're going to target, you know, this cell or this tissue or this receptor, and then you're actually able to target it with a molecule? And is that what, you know, verifies a target or validates a target? I forget the exact phrase you used. Yeah, no, I'm talking about genetic validation of a target. And so we could name different targets, right? But take like, for example, ApoE, which is, you know, a genetically validated target for Alzheimer's disease.

9:56So why do we know this? So the best genetically validated targets kind of give us the yin and the yang of the scenario. So we know that individuals that are E4, E4 homozygous have, you know, a 10 to 20 fold higher rate of Alzheimer's disease. We know that individuals have the E2 allele are protected, but even more important is in some rare individuals where they have a autosomal dominant form of Alzheimer's disease, which means they're almost certainly going to get Alzheimer's disease by the time they're 40 or 50, there's rare ApoE mutations that can prevent them from ever getting Alzheimer's disease.

10:39And so like it's, it is not only so this is the kind of rare genetic evidence that we can use to really bookend, what's the level of confidence that drugging this target might be able to impact clinically, probably the most prototypical case of a genetically validated target is PCSK9, discovered out of the Dallas cohort study. But as we can go on to discover more and more of these genetically validated targets, these targets have less risks as we go into clinical trials because we've already learned some of what modifying that target will do based on the natural experiments that have happened in human genetics.

11:18Bret Bostwick, M.D.: Great. Yeah. Thank you for that explanation. I wanted to ask, I guess, next thing about the circumstances that led to you joining Breyer Capital as a venture partner. I mean, were you, it sounds like you've been interested in kind of the entrepreneurial side of medicine, you know, for a while. But, you know, thinking about like the funding cycles, leadership that you've just spoken about, operational aspects of company management, is that something that you kind of grew into or had been sort of paying close attention to, you know, throughout, you know, the various stops in your career? Yeah, I mean, I had always been interested in the business aspects in particular, because that's the way in which a lot of great ideas fail.

12:03And so I was always interested in like, like why that was. So when I started translational medicine solutions, we were helping with translational strategy to a number of biotech and, and, and pharma clients. And what I was surprised by was a lot of venture capital groups wanted to use TMS's platform in order to inform their investment strategy and investment risks. So what we tried to do was instead of... Today, let's say you have a new drug and you want to say, what are the chances this will work in clinic? You basically pull together five KOLs and you say, hey, what do you think the chances are this will work?

12:42And one says 20%, one says 50%, and you kind of take their average. So what we tried to do is using large language models and AI, we tried to really ground probability of success in an empiric data set where we use Bayesian statistics. So what's the pretest probability this will work? Apply some facts we understand about the process, the biology, the risks. And then what's the post-test probability? And so it turns out venture capital groups, because biotech investing is such a binary risk, it often either fails or it works, right? They really like bookending probability of success. So it ends up, I think that, you know, that got me in conversations with a lot of VCs.

13:26And then I met Morgan Cheatham, who's head of healthcare and life sciences at Breyer Capital. At the time, he was at Bessemer. And he's a medical geneticist as well. He has a decade of experience in venture capital. And we obviously just hit it off. And there's probably that joke, like, two medical geneticists walk into a bar and we hope to write like the ending of that joke, you know. But yeah, so we teamed up. And when I heard he was going over to Breyer, I was thrilled to join him.

13:55Bret Bostwick, M.D.: Excellent. I want to ask a little bit, a couple more questions about the CNS therapeutic area. I mentioned I think it's a really fascinating category for drug development in recent years, you know, after a kind of long pipeline drought. I think it's fair to say on RNAI therapeutics specifically, what what makes that a compelling modality for for CNS diseases, in your view? Yeah, I mean, we could talk about maybe as a separate what all the challenges are within the CNS specifically. But RNAI is a category of therapeutic that we call programmable therapeutics. and you know for centuries in medicine we we didn't program drones we discovered them right we looked at soil microbes and exotic plants and sea urchins and venom and through a process of like pretty clumsy trial and error um and with like a lot of serendipity required to be successful like there were small molecules discovered what we're doing today in medicine is fundamentally different.

14:59We're basically taking a molecule, we are writing code, and we're executing that code in the hardware of human biology inside the system. And by doing that, we're able to prescribe the effect that we want on the biology and the pathway with precision that we never had before. And within then that category of programmable therapeutics, my biggest passion is around any modality that uses Watson-Crick base bearing. So this is the famous, you know, complementarity between two bases. When we design drugs that are either interfacing with DNA or RNA, we take advantage of the most beautiful chemistry ever invented by evolution called Watson-Crick base bearing.

15:47This is like a digital handshake that happens at a level of specificity that is like not comparable to other modalities or classes. So we don't talk about whether it's 50 to 1 discrimination for its target or 100 to 1. We talk about things like, is it a million to 1 or 10 million to 1 discrimination for its target? And so by being able to exploit Watson Creek base pairing as the primary chemistry in your drug design, you take advantage of all the years of evolution, having made that a highly specific interaction, you can not only program a drug, you don't have to discover it, not only can you target upstream of disease biology pathways, but you have a way to do so with a specificity that until the last few years was unheard of.

16:39So yes, I'm a huge bull behind, you know, oligonucleotide therapies, siRNA, I think being one of the best, but also, you know, editing at the DNA level as well. I think that these are also exciting modalities and platforms.

16:54Bret Bostwick, M.D.: And you, you know, you mentioned some of the broader challenges in CNS, and I think continuing with this example, you know, you have to know, you have to have a verified, you know, gene or genetics that is giving rise to a certain condition. You mentioned APOE and Alzheimer's, but, you know, of course, there are a number of mental health conditions that really suffer from a lack of clear biomarkers. You know, a lot of the side effects are kind of subjective, which makes it really difficult. And I'm curious, you know, I mean, is that what you see is the biggest challenge in the CNS category for drug developers?

17:30Bret Bostwick, M.D.: Or are there other kind of key outstanding challenges in CNS that need to be overcome? Yeah, I mean, I probably answer this question a little different than most drug developers, But I do not think that understanding what the target is the primary obstacle in the CNS. Okay. I think that a lot of outsiders look at CNS drug development and they think of the brain as like just another organ. But it is like an entirely different biological universe in that thing. I mean, from evolution of non-human primate to the human brain, something tremendous had to happen in a very compressed evolutionary timescale.

18:12And what that was, was a series of system changes I can describe that really make the brain a substantially different organ and neurons, a substantially different cell type than most of the cells in the rest of the body. So one is we all know about delivery and the moat around the brain. You know, we talk about moats in business all the time.

18:34Bret Bostwick, M.D.: The blood brain derriere you're referring to? Yeah, exactly. Like we got a physical moat around this thing. It's designed to be an immunological fortress. Its intent is to keep all foreign material out, including most of the drugs we design. And so the first main obstacle you have is just how do you get your drug there? So we did it at Alnylam through intrathecal delivery. So giving the drug into the fluid that surrounds the spinal cord so it can float up to the brain. Now we have BBB shuttles and other methods and ways to get into the brain. Number two is scale. I think the brain is still like poorly understood relative.

19:10We were talking earlier about the scale of the genome. So like take a motor neuron that sits in your spinal cord. The cell body sitting in your spinal cord in the middle of your back, but like let's say it's your sciatic nerve. The end of that axon is like all the way down in your ankle. So it's a three foot long cell. So to put that three feet into proportion, if the cell body is like a tennis ball and it has a little thin cable coming out of it, an axon, that would stretch for four miles.

19:43Bret Bostwick, M.D.: Wow. Okay. So imagine you deliver a drug to the cell body, but there's this long tube going down for four miles and there's a very complex set of biology that shuttles things back and forth. And so the scale here makes treating in the CNS complicated. And then lastly, I alluded to this idea that, evolution had to pick up its pace relative to the human brain. And so the way it did that is it couldn't make the genome much larger, right? We're already at 20 ,000 genes. So what it did in the CNS is it made the transcriptome larger. So like of our 20 ,000 genes, a typical cell in the body maybe expresses 3 ,000 or 4 ,000 of them.

20:25The mature neurons express like 10 ,000 of these genes. But then instead of like one gene leading to one protein, these genes undergo alternative splicing and they make dozens or hundreds of distinct mRNAs, which then each go on to make dozens or hundreds of unique proteins. And so you have like this exponential multiplier on the transcriptome that makes the diversity just so much greater than other cell. There's much more RNA content in a neuron than any other cell type. On top of that, it also had to like fine tune all this orchestration around synapses and everything else. So it did so in part by having employing a lot of long non-coding RNAs and micro RNAs.

21:10And these are like RNA species that we probably least understand in biology. And then there's one last confounder, which is that the brain has this like postal system where like it can take contents, mRNA content from one cell, package it, the same we call a neuronal exosome, but like, you know, an envelope or a letter, and it sends it somewhere else in the brain. So imagine you design a drug that's supposed to go in and silence a particular gene. You can silence it in that one tennis ball cell, but it might receive letters from neighboring cells that give it a replacement copy of that gene. And so the combination of all this makes me think of the brain not as a...

21:53The transcriptome isn't like a static library, but you know those old videos of the New York Stock Exchange trading floor where everyone's going nuts, holding up their pieces of paper? I think of the brain transcriptome as a high-frequency trading floor. And it's also the dynamics of that situation, the complexity with the dynamics overlaying it that makes CNS drug development so hard. So I think it's actually the organ itself, not so much what do we target or what modality or the pharmacology, but I think it's just this thing that we have in our body that is just orders of magnitude more complex than the rest.

22:35Yeah.

22:36Bret Bostwick, M.D.: Yeah, that's really interesting. But to get back to modalities, maybe aside from RNA or genetic therapies generally, Brett, are there other therapeutic approaches, other modalities in CNS that excite you or that you're kind of following or maybe potentially investing in? Well, all of these precision tools I described, so whether it's an ASO, an SIRNA, a base editor, a prime editor, an epigenetic editor, gene therapy, all these things are great tools. And then you have to look at the application of the problem you're trying to solve and where you're trying to go in this brain to find the right tool to do it.

23:22But I don't think, you know, we're very modality agnostic at Briar Capital. There, just because a tool is great, doesn't mean it's the right tool for that particular job, right? And so it's really about articulating exactly where the path of biology exists, where you're trying to get there. And once you're there, what are you trying to do? And then you select the tool, you know, from the tool bag to do the job.

23:46Bret Bostwick, M.D.: I wanted to ask you, just given your work on CNS, there have been some deals, some pretty big deals this year in the psychedelic drug development space. AbbVie acquired, let's see, a product from Gilgamesh. And then also Eli Lilly did a big deal with a Ty Beckley not too long ago. And I'm curious, you know, if that's something that Briar Capital is looking at or if there's a kind of investment thesis on psychedelic drugs for for CNS disorders? I mean, do you do you, I guess, anticipate psychedelics becoming potentially like a substantial part of the treatment landscape, at least for mental health disorders?

24:33It's a good question. We do not have a particular thesis on it. I do have a good friend who is a psychedelic medicine physician. So I'm always learning a lot from her. You know, these are mostly drugs that, you know, are in the small molecule class, right? That have been discovered through that other process I told you, you know, by an exotic plant or a sea urchin or what have you, right?

24:55Bret Bostwick, M.D.: A mushroom, of course. I think that we're starting to study them as, you know, real contenders to address symptoms in the CNS. And I think, you know, that these are appropriate trials to run and we'll see where it goes. We don't have any, you know, these are kind of the opposite of programmable therapeutics. These are more these molecules that were found by serendipity. And of course, yeah, nature gave them to us and we should figure out what they do and supportive of those investigations. But I think we tend to focus more on, you know, where we can write that code and execute that code in a very intentional way.

25:31Yeah.

25:32Bret Bostwick, M.D.: And I think that answers at least part of this next question I wanted to ask you, Brett, which is and I know there will be listeners of the show that want to hear about the kinds of biotech companies that that Breyer is targeting for investment. Can you talk a little bit about that overall investment model at Breyer just in terms of maybe company types or clinical stage or typical investment amounts? Yeah, I mean, again, we have the privilege of being able to kind of approach everything. So we're looking for what's exciting out there. So our process in general is we start with the science, not kind of at a fluffy level, but really like what's the biological ground truth of the pathway they're trying to target.

26:18We look for founders to come in and really articulate unmet need or the translational gap that we want a clean problem statement in the cleanest, clearest way. And then we want them to explain the question you just asked, which why is this modality or this tool the absolute right one to go and fix this problem? Right. So in that, we start with the science. And then, of course, we want to pursue precision. So these are these scalable, programmable medicines. We like the use of these. We don't want to do incremental. If a company thinks they can make Tylenol 2 % better, that's great. They should go do that.

26:59We're really trying to find breakthroughs. We're trying to back high-risk breakthrough technology that can paradigm shift the way doctors approach a particular disease area. um and so we really then kind of we often say we allow the need to shape the form but we allow the the need of what you're trying to intervene on to shape the form and the modality by which you go at it so we're modality agnostic and then increasingly if we can treat medicine like an information science um and you know biology used to be about observation and now biology is becoming engineering We like the idea of seeing companies that are starting to think about medicine as an information science.

27:45And so that's where we look. So now late stage, early stage, of course, we love to be early. We love people to come with us with just an idea. They don't have to have a full business plan yet. Come with an idea. We could talk through it. If you can ever do something that really helps people, there's a business on the other side of it.

28:05Bret Bostwick, M.D.: Yeah. Okay. So you'll come in at the very earliest stages then, like if you find a founder with the right idea, that that would be somewhere where you could potentially come in. What about exit? Is there, you know, do you get into late stages of clinical trials or are you kind of done by that point or exited by that point? Well, no, I mean, we stick with our companies. We're a long term partner. So one of, you know, one of the advantages, right, of being at a family office is we don't have like any particular fund cycle. You know, we're an evergreen fund and we continue to stick with our companies and we can be patient.

28:39Sometimes it takes a decade. You know, great science is not always fast. And so we can be patient partners. And so a number of companies in our portfolio, of course, grow and they IPO or they exit. And, you know, that's great. That's a great outcome. And that usually means that patients are getting, you know, that medicines are getting to patients. and hopefully those medicines are being studied so we can really learn. It's great if they work, but if they don't work, we can learn exactly why they fail so we can all learn as a group how to do better next time.

29:13Bret Bostwick, M.D.: Got it. I want to dig in a little bit on the preclinical space and translation. And I know it's probably difficult to speak generally about this topic, Brett, but based on your experiences, I wonder if there are any kind of repeat mistakes that you see happen across different companies during the translation process? I mean, like, let's say you're, you know, you're advising a new client that's ready to translate a discovery into a development candidate. You know, what are the kind of first questions that you ask or first advice you would give? Yeah. I mean, companies, especially preclinical biotechs, they spend a lot of time in animal data, mouse data and rat data.

29:58It's, you know, mice lie. They lie. They're not a very good proxy for what will work in clinic. Human genetics doesn't lie, right? So, and going

30:09Bret Bostwick, M.D.: to clinic will give you the most valuable learning your company's ever had. So, you know, advice to free clinical companies is from the get-go, chart the quickest path, the fastest, most agile path you can to get to clinical trial, that first in human trial. Because not until you give your new idea, your modality, your platform to the first human that you really start to learn what it does in this organism, right? This animal species. And then when you get to clinic, that is not a finish line. That's actually the starting line, right? Because the whole point is to study what these things do in people.

30:45But I would, you know, in as you're thinking about how to prioritize as an early company, I would not lean so heavily on what a mouse did when you gave a drug. I would look more carefully at what human genetics has to teach us about the target, what you have to pharmacology to go after the target. Do you have a clear biomarker that will tell you whether you're hitting it when you go to clinic? And then is there a defined clinical regulatory strategy that can guide you through those first couple of phases of development?

31:17Bret Bostwick, M.D.: On your comments about, you know, animal models, this is a topic that's come up a number of times on the show, which is that you're right. Often the results don't translate to humans. And I think to their credit, the FDA has been, you know, really pushing to move toward non-animal models, NAMs and various parts of the, you know, the preclinical development process. What's your take on that? Are you encouraging companies to seek out some of these new kinds of processes outside of animal models? Is it not there yet? What's your opinion on that? Yeah, I mean, I think as a field, right, it's very important for us to figure out how can we go faster.

32:04And the reality is that when there's shortages of non-human primates and their studies, they're really expensive. early biotech, you have a finite cash runway. You got to figure out your fastest path. I mean, I'm sure you've talked to guests about a lot of biotech that's moving over to China to do first in human studies. And there's reasons for that. It's not just the finances of it, but it's also the speed by which you can move forward and the kind of data package that's required before you can go into humans. It's a very different data package there than it is here. So I mean, I think like the question as to whether or not these other ways of de-risking a molecule will pan out to de-risk in human trial.

32:48We don't know the answer to that yet, but I think these are the right steps in the right direction. This is about being agile, being quick, quick to fail, quick to learn, and then hopefully quick to succeed. Right. And that's the cycle that I think these technologies are enabled. world.

33:03Bret Bostwick, M.D.: I'm glad you brought up China. We have heard from a lot of guests on this show, particularly about preclinical early stage capital moving to China for the exact reasons that you just mentioned, Brett, a faster and cheaper ecosystem for new drug candidates. Well, first, how are you thinking about Chinese biotech as an investment target at Breyer? And is there a kind of longer game in China? Maybe I'll speak for the general ecosystem, right? My take, and I've been part of running trials over there, both advisory, consulting, capacity, etc. The Chinese biotech system has matured tremendously in just the last few years.

33:48And it's now a hub, a global hub for drug development, cutting edge drug development. I think that's a good thing. I think it's great that multiple geographies on this globe are going to produce drugs that can help patients. I think that that's great. And if you look at how they did that, it was not an accident, right? They kind of deliberately built this through regulatory reforms, through clinical trial infrastructure, building these tertiary quaternary care centers where you get this talent density, much like American universities, where then kind of this talent density has emergent properties where they're able to kind of come up with great new ideas.

34:28And I think the West is now kind of feeling the opportunity cost here a little bit of like slower execution. So what do we do here? Execute, execute, execute, right? I just told you like, you know, race to getting that first patient dosed, go, go, go. That's what you should be focused on if you're a US biotech. I think that there are, you know, the world is huge. And so, you know, there's great science happening all over the place. And fantastic science has a way of bubbling up. So it doesn't stay isolated geographically for long. The lessons become learned, the drugs disperse, and anything that's working really well gets its attention.

35:12So, yes, of course, we keep an eye on the global landscape. And, you know, where appropriate, you know, our portfolio companies will might choose to run a first in human trial over in China because it's a faster path to getting to information. They might also choose to come and run a trial here in the U.S. using some of these new pathways where you can have, you know, more agility and quicker time to clinic here on U.S. soil.

35:39Bret Bostwick, M.D.: Yeah, the FDA is experimenting with a number of pilot programs to try to bring up that competitive level. And I take your point. I think anywhere in the world that innovative new medicines that are going to help patients are coming out should be celebrated. And, you know, at the same time, I hear U.S. biotechs, you know, U.S.-based biotechs in the preclinical and early clinical development phases, you know, wondering, seeing, you know, the statistics about the amount of capital moving over, shifting over to China. Are there any, I guess, tips maybe, Brett, that you would have in terms of what leaders at U.S.-based companies could do that are in that kind of preclinical valley of death early stages that could make them more attractive to investors?

36:27Bret Bostwick, M.D.: And maybe it's what you already said, you know, get into the clinic and get some data as fast as you can, right? Yeah, I mean, my sense here, right, is like I mentioned part of our thesis is like we don't invest in incremental. We want paradigm shifting breakthroughs. So if you're a US biotech company today, and you are going to try to run the same playbook as a Chinese biotech, but do it cheaper, if that's your business strategy, that's going to be tough. But if you're a US biotech today, and you are truly innovating, you have a platform where you have a nice flywheel, where you're learning and learning deeper and deeper every day, and you're having breakthroughs and you're innovating, that becomes very hard to rip off, right?

37:10And so I just think that what you have to do is you have to be on the cutting edge of doing something big, breakthrough, that is not easy to reproduce. And then all of that becomes noise. So as long as that's what they should focus on doing, tackle a bigger problem, take a higher risk, aim for a breakthrough, Aim for something that's really going to be impactful and not incremental and and no copycat strategy because that's just not not likely to succeed.

37:42Bret Bostwick, M.D.: I want to ask you about AI. And I know Breyer is heavily invested in various forms of AI, different companies using AI. You talked about translational medicine solutions, creating this kind of empirical AI platform for looking at, you know, newly discovered drugs or really early stage potential drug candidates. And I wonder, you know, what you see as kind of the most promising uses of AI in early stage biotech and maybe kind of what's most overblown, because, you know, like all of us, I'm constantly hearing about AI and, you know, some of it I can follow more closely than, you know, some other things.

38:25Bret Bostwick, M.D.: But, you know, what what would you say about, I guess, how AI is changing early stage biotech? What's most promising? What's most overblown? Yeah, I mean, we really have seen in like the last year or two, AI has really hit its like adoption phase, right? So a couple of years ago, everyone was piloting some AI somewhere. And now it's just all in use across the full biotech stack. um so in biotech like we see utility right uh discovery better molecule design you know like efficiently building compelling preclinical packages um and so i think we've seen some gains there but like what's not useful is like an ai model that can discover like a trillion new compounds on a weekend because like what do you do with that right it's like congrats I think, so let me go to overblown and then maybe I can circle back.

39:18But like what's definitely overblown is the statement that the primary bottleneck in delivering medicines today is discovering new medicines. That is not the primary bottleneck. So drug discovery is easy. It's easy to discover drugs. We have known pathways. What is really hard is drug development. So that's once you have the molecule to put it in clinical trials, that's where 90 % of the costs sit. That's where all the risks sit to humans. And that's where 90 % of things fail. And so we really haven't seen AI in a meaningful way come in and compress biomedical timelines. So we are focused, right?

40:01So there's a problem statement, right? So we love companies that come to us and pitch how they'll fix that part. Because discovery of trillion molecules with no budget and no ability to quickly and cheaply test them in humans, it's still going to cost probably$10 to$20 million per molecule to figure out whether you can even get through phase one, right? And so what we really have to start doing is figuring out how we can do clinical trials, compress biomedical timelines, and adopt some of the AI efficiencies into that part of our drug development process.

40:35Bret Bostwick, M.D.: I have one more question about AI, and it's honestly not really related to biotech. But I want to ask you, because you're somewhat familiar with AI, I've been hearing this concept of cognitive surrender lately. And it's this idea that, you know, the more we outsource our thinking capacity to AI, the more that capacity weakens and deteriorates. What do you make of that? Am I kind of off base with this? Do you think that's real? Or if you're not concerned about that, are there any other AI concerns that you're watching? Well, I mean, I think with the adoption of any new technology, right, and then a high frequency of human interaction with it, learnings emerge.

41:21there are probably many things that as they became automated like our ability to do it like maybe call it long division or you know you pick your thing right sure maybe that things waned i think the larger question would be like the impact of your workflow has the productivity and impact at the end of your workflow actually been detrimented by ai so how you got there is less important as what you're yeah i mean i could tell you like when i drive out of my driveway if i don't have my gps on i can't even find the grocery store that i most frequently go to well you know it's a great

41:53Bret Bostwick, M.D.: point i was because i was thinking you know art we had a rear a rear view camera that like um went on the fritz and wasn't working it's like you know i never had a rear view camera when i was driving like growing up but now you know you take that away and all of a sudden you have to use your mirrors and like little stunned for a second so yeah i mean so like cognitive yeah i think i think that's right so for like gps i think we all know that you know or at least i'll speak for myself. I've lost the ability to navigate like I used to be able to. But that's a trade-off, right? That I am happy to make because that frees up other compute in my brain to tackle other things.

42:29I think more so than the cognitive pieces, I am concerned about automation complacency. So when AI is kind of like, because I see it used in clinical decision-making, but you know, when AI is like tell you the next best step, like, okay, do this, do this, do this and they're right and they're right and they're right over and over, you eventually kind of almost click enter. It's like you get a little bit complacent. So there needs to be thoughtful systems in place where there's a necessary human in the loop for verification. I think it's really important that we don't allow those humans to become complacent.

43:04And I think about this in the doctor-patient relationship where right now there needs to be a human in the loop. But for other areas i mean where there doesn't need to be a human in the loop i don't know if it takes over the cognitive burden of of that then that frees up your mind to think about other things right no that's no i'm not that concerned about it i guess would be the answer yeah uh well um

43:28Bret Bostwick, M.D.: last question for you uh brett what are your top priorities uh for the rest of this year and maybe the first half of next year 2027 yeah i mean i i'll get like real granular and then i'll bring it up a level but like like the last few weeks i've just been nerding out on the blood brain barrier um i think there's been real progress made here you're probably aware of of some of the shuttle technology that's out there um the shuttles are using tfr to get across the transferrin receptor largely right now and i think that that will only be a solution for a very small subset of where we want to deliver in the brain because certain brain cells also have tfr receptors on them and other certain brain cells we might want to target don't have it.

44:12So it's not going to be the holy grail for every drug to be delivered. So I'm very interested in other receptors, you know, CD98, LRP1, but then also other ways across like interactions with the glycocalyx at the BBB to get across through scavenger receptors or the paracellular ways to get across through like BBB relaxers. There's a whole new group of ideas spawning out of academia right now where they're trying to find ways to get across the blood brain barrier. And so I just have like this pet project rabbit hole to go down to like understand that landscape and then try to pick a few that I think might be winners in this space.

44:51So that's like my, maybe like between now and Thanksgiving, like my personal goal, maybe on a larger, like end of the year, beginning of next year, we're still big on this I think we coined it our in vivo renaissance but this idea of moving the manufacturing of drugs from these big bioreactors outside the body to back inside the body using mRNA technology we think that this is going to be transformative I mean I think last week maybe way longer back when this podcast airs but last week I think Novartis acquired Abigin so that's an mRNA encoding a T cell engager, a CD19 TCE. And, you know, that's a great example of exactly what we're talking about, where like right now we take these biologics and we have bioreactors, multi-billion dollar factories, we make them, they aren't that shelf stable.

45:48You come into an infusion suite and, you know, you have to get them every so often. And there's all these liabilities where you might be able to make that same thing from within the body and programmed in an intentional way. So, and of course, obviously in cell therapy, this is kind of the leading edge of this, where we've moved ex vivo car T now to in vivo. And so we'll continue to make bets in that space. And we're really optimistic that, you know, some of these, you know, the type of thinking around the in vivo renaissance will yield some like really truly transformative drugs here in the coming years.

46:26Bret Bostwick, M.D.: It's a fascinating area. And Brett, I really appreciate you taking the time to come on the show today. Thank you so much. Yeah, thanks, Ben, so much. It's been a pleasure to be a guest and I'll continue to be a listener. Thank you. We've been speaking with Dr. Brett Bostwick, MD, venture partner at Breyer Capital. Breyer Capital, my apologies. I'm Ben Comer, and you've just listened to the business of biotech. Find us and subscribe anywhere you listen to podcasts and be sure to check out our weekly video cast of these conversations every Monday under the business of biotech tab at lifescienceleader.com.

47:00Bret Bostwick, M.D.: We'll see you next week. And thanks, as always, for listening.