Welcome once again to TD Cowen's Genetic Medicines and RNA Summit. I'm Phil Nadeau, one of Cowen's biotech analysts, and it's my pleasure to moderate a fireside chat with Voyager Therapeutics. We have with us today Al Sandrock, the Chief Executive Officer, and Todd Carter, the Chief Scientific Officer. Al and Todd, I'll maybe hand it over to you to begin. Could you give us a brief state of the company overview? What are the biggest strengths, biggest challenges, and what does Voyager need to achieve to drive outperformance over the next 12- 24 months? Well, thanks, Phil. Thanks for inviting us to speak today. So Voyager is all about trying to make transformative medicines for bad neurological diseases, and, you know, look, we believe that this is the era where we're starting to make headway for some terrible diseases, for example, ALS, Alzheimer's disease. We have clear-cut examples of drugs that are now disease- modifying, that have been approved or soon to be approved, and including some neurodevelopmental disorders as well, not just neurodegenerative disease. So we have drugs for spinal muscular atrophy and Friedreich's ataxia, so very exciting times for neuro. We have... I would say the reason to invest in Voyager could be summarized in terms of the four Ps. We have a robust pipeline. We have already entered the clinic with VY-TAU01. We're doing our phase I studies now, and we are marching towards getting three INDs for three gene therapy programs in 2025, next year. One of those three is a wholly owned program for ALS. The second P would be platform, and Todd will hopefully speak about that in a minute, but we have a platform that generates these novel blood-brain barrier-penetrant capsids that is the basis for some exciting programs that we have in our own pipeline, but also in our partner's pipeline. And as I said earlier, we have three development candidates that are marching toward INDs. The third P would be partnerships, so we have 13 partnered programs, all leveraging these novel capsids derived from this TRACER platform for gene therapy. And the last P would be potential, and there we talk about the fact that we have, that we have found the receptor that mediates the transfer of these capsids across the blood-brain barrier. What we're now doing is investigating whether we can leverage these receptors to get other kinds of macromolecules across the blood-brain barrier, like proteins and oligonucleotides. So we would have then multiple modalities that can do neurogenetic medicines. And I would say that finally, you know, look, we're in a very strong cash position. We ended last quarter with about $400 million in cash, and we have runway into 2027, which should enable us to get these three programs into the clinic, these three gene therapy programs into the clinic, and hopefully get some data from those, because you know, you have to start right off in patients at doses that could work in gene therapy trials. Moreover, we'll have, from our lead program, the anti-tau, some data on whether or not our antibody can block the spread of tau progression in Alzheimer's disease. Maybe starting with the second P before focusing on the first P, on the technology platform, Todd, for those less familiar, could you describe Voyager's foundational technology for CNS gene therapy delivery, and maybe in particular, how does Voyager drive CNS gene therapy expression following an IV administration? Absolutely, and thanks from me for the invitation to speak as well. So, TRACER is our platform to identify, to discover novel capsids that have greatly improved blood-brain barrier penetrance. What we're trying to do is harness the vasculature, which it broadly gets through the brain, and harness that breadth of distribution to subsequently deliver our gene therapies throughout the CNS. The way we've done that is we've made mutations in particular places on the surface of the capsid that we know won't completely disrupt the capsid, and we've screened millions of these different variants to identify those that not only cross the blood-brain barrier, not only get to the other side, but also get into cells and express. That's a really important component of what we've been trying to do to identify new gene- therapy- capsids that can deliver and express these payloads. We've done this importantly in non-human primates, and not just a single species of non-human primate. One of the issues in the field has been that it has been possible to identify a capsid that might work well in one particular species, or even a subset of strains in mouse, for example, but we've seen very little. It's been much more difficult to identify something that's cross-species. Our platform is really entrenched in the idea that we not only identify these improved capsids in a single species, but across multiple species of non-human primates, and we've got families of capsids that can deliver not only across those, but into rodents as well. We also look not only at where we want to deliver, but where we want to not deliver, and what I mean by that is, we wanna see increased delivery into neurons or astrocytes of the CNS, but we want to de-target those parts of the body that we don't want to express in. For example, the liver is a common tissue that AAV can typically go to, but you can see problems with toxicities. So, we've identified capsids that have greatly improved delivery into the brain, but reduced delivery into the liver and other parts of the periphery that we want to avoid, while still maintaining that high delivery into neurons and glial cells.... And then I guess finally, I'll say that this is an iterative process. I mentioned we screen these millions of variants. What we can do is, once we've identified capsids, we can continue to improve upon them. And so at ASGCT this year, the Cell and Gene Therapy Conference, we presented data off of our second generation capsids, where we've taken our first generation capsid families and continued to improve upon them. What we've seen is a really, Shown at the ASGCT conference, is that we can get a really robust transduction of 50%-75% of cells across a wide variety of different brain regions. This is all with a single intravenous dose, at a very clinically relevant dose of about 3 × 10^13 vector genomes per kilogram. We can see even higher transduction of particular important cell types. We've seen upwards of 95% transduction, in cells such as Purkinje neurons, 98% of dopaminergic neurons in the substantia nigra, which are known to be very important for diseases like Parkinson's disease, and over 80% of spinal motor neurons. What that has allowed us to do, and enabled us to do, is both internally and with some of our partners, we've managed to nominate three development candidates over the past several months to now move forward toward INDs in 2025. That's a perfect transition to the first P, the pipeline. Maybe starting at a high level before diving into the specific candidates, can you describe how Voyager approaches R&D portfolio construction? How does Voyager balance the magnitude of unmet need, depth of scientific understanding, probability of technical success, market size, all the relevant factors? How does Voyager balance them in determining what programs to pursue? Yeah, thanks, Phil. So I think the key thing, there's a lot of promise in neuro, in the diseases I mentioned, because we understand the targets, and I think we start with the fact that we want to go after highly validated targets, either targets that others have already shown by drugging them, that they will work, but also highly validated by human genetics. We also, as you said, focus on high unmet need, because we wanna be sure that we can get it approved, fairly expediently, or very expediently, if possible, and get a return on investment. But the key thing in neuro is to deal with the risk. Everybody understands that neuro holds great promise. Nobody argues with the unmet need, but people see it as risky, especially if you're talking about gene therapy. The key is, we only pursue programs where we can de-risk very efficiently. So if I have to do large studies, or that take too long, and they are too expensive to de-risk, we're not gonna do it. We're gonna let somebody else do that. So we've chosen programs where we can get into the clinic quickly and de-risk efficiently. And so the efficient path to proof of concept or proof of biology, at least, is critically important. Starting with the tau programs, maybe for those less familiar, could you describe VY-TAU01, and maybe go into detail on why an antibody targeting the C- terminus of the tau protein is promising for Alzheimer's? Yeah, thanks. So ironically, our first program is not a gene therapy. It's actually a regular monoclonal antibody that... We now know that monoclonal antibodies, when given IV, do get into the brain and can cause very meaningful biological, if not clinical, changes. So it's a simple monoclonal antibody, but it's differentiated from the other anti-tau antibodies that went before us, that targeted the N-terminal, and that failed in the clinic. Our approach has been, we make antibodies against pathological forms of tau, so we avoided antibodies that also bound to normal tau. But we still had a number, half a dozen or so, that bound to pathological forms of tau very, very specifically, scattered across the tau molecule. And so we say, "Well, how do you pick among those?" And so we then employed an animal model, where we inject human pathological tau from Alzheimer's brains, and we look at the spread of tau in that, in the brain. And we chose the C- terminal antibody because it very robustly and consistently blocks the spread of tau. The other thing I like about this model is that if you test the N- terminal antibodies that had already failed in the clinic, they also fail in the animal model. So the animal model seems to have very good negative predictive value, certainly did for those failed N- terminal antibodies. What we don't know is whether it has positive predictive value, and of course, we'll find out when we get into the clinic, but we're hoping it does. You've recently announced that the first participant's been dosed in the phase I single ascending dose study. Can you remind us what the design of that trial is, and what does Voyager hope to learn from the SAD portion before moving on to the MAD? Yeah, so it's, as the SAD implies, it's you give one dose, single dose, and what we're trying to do is to go to higher and higher doses and measure essentially two things: plasma PK, pharmacokinetics, and safety. And we're doing it in normal, healthy volunteers, so we're not gonna get an efficacy read out of this. The thing what we wanna do is we know what exposures we need in the blood in order to get adequate exposures in the brain based on preclinical studies. So what we're trying to determine is whether we get the PK that we hope to get, and so that prepares the way for the multiple ascending dose, which will come next. Yeah, and maybe discuss the multiple ascending dose in a bit more detail. Sure. What efficacy measures can you collect, and what would you consider a success in the MAD portion that would warrant further development? Yeah, so for the multiple ascending dose study, we are gonna go do it in patients with Alzheimer's disease, and the reason why we've chosen Alzheimer's disease, because there are other tauopathies that we could have chosen, but we went after Alzheimer's because we can go, as I said earlier, it's important that we can rapidly de-risk. And in Alzheimer's disease, people have already employed tau PET imaging to look at the spread of tau in the brain of patients with Alzheimer's disease. So we're gonna take... We're likely to take patients, you know, we haven't disclosed all the details, but we're gonna probably take patients in the relatively early stages of Alzheimer's, probably Braak stage II or III, you know, patients where the tau is still mostly in the medial temporal lobe. We know there's a very stereotypical spreading pattern from the medial temporal lobe to the lateral parts of the temporal lobe and to other parts of the cerebral cortex. So what we can do is do tau PET imaging, and we don't think it's gonna take more than 20-25 patients per group. So if we hone in on the right doses, we can do a very efficient 1-year or so look at 20-25 patients per group and see if we can replicate what we saw in the animals where we blocked the spread of tau. Now, what's important here is that we do believe that the spread of tau is what's responsible for the, for much of the dementia, and that's based on a lot of human studies that... And in fact, the anti-amyloid antibodies that are now approved, they also block the spread of tau. So it could be that the anti-amyloid antibodies work because they affect tau secondarily. What we also know from those studies, and if this came out at the donanemab FDA Advisory Committee, that people who have a higher tau burden when they went into the trial have less of an effect with the anti-amyloid antibody. So clearly, especially when you have moderate to high, higher levels of tau, you've gotta have some other approach, otherwise you're not gonna fix Alzheimer's disease. Yeah, I think- That's, that's the concept. Interesting article this week, I'm sure you saw in the New England Journal, looking at the mutations in APOE- Yeah ... where there was, I believe, no effect on beta- amyloid levels, but fewer tangles- Yeah ... tau tangles in those patients, which is- Yeah, it's a very cool article because it looked at APOE Christchurch heterozygotes. There had been previously a paper where it looked at an APOE3 Christchurch homozygote, who had a very, you know... She had a brain full of amyloid but did not get demented like all of her relatives. It's remarkable, if you have a PSEN1 mutation, you get, start to get symptoms between age 47 and 49. The 95% confidence interval is 47-49. Very precise if you carry that mutation, but if you, if you have the, the APOE Christchurch, you're, you don't... You, you have a delay in the onset of dementia. And so far, each of these patients that have had a delay, the brain is full of amyloid, but the tau is not progressing the way you would have expected it, as it did in all the other relatives. So clearly points to tau, I think, as the cause of the dementia. Voyager also has a tau silencing gene therapy program. Can you provide more details on the mechanism of tau silencing and the status of that program? Yeah, Todd? Yeah, absolutely. So, our plan is to use our novel BBB-penetrant capsids to create an IV-delivered one-time dose, vectorizing an siRNA to knock down the tau mRNA, and subsequently reduce the protein. In some ways, this is similar to our SOD1 program, where we're knocking down SOD1 with a vectorized siRNA. In addition, it's similar to a BIIB080, which is an ASO that Biogen is pursuing. That's an intrathecally administered ASO that requires repeat administration. But the early data has suggested that they're seeing reductions in the spread of the tau pathological reduction in pathology overall. So we're harnessing that same type of mechanism, the knocking down or reduction of the mRNA, only we're doing so with the intent of a single IV-administered gene therapy. We've seen and shown data, where we can do this very well in the mouse with a BBB-penetrant capsid in an IV dose, where we're seeing quite substantial knockdown throughout the CNS, both at the mRNA and protein level for tau. We're anticipating an IND for this program in 2026. In general, we think that tau is as Al just described, kind of the next really key target for Alzheimer's disease. In Alzheimer's disease and neurodegeneration in general, we've seen remarkable progress over the past several years, with the first drugs that actually can impact the disease in a disease-modifying way. We really think tau is going to be very important for the reasons that Al mentioned. We're pursuing it in two different ways, the tau antibody that Al described, and this knockdown. The tau antibody that will target particular pathological epitope, and the intent is to target that epitope in an extracellular space that the antibody can get to and inhibit the progression, the cell-to-cell spread. The tau knockdown approach, however, should knock down tau at the mRNA and protein levels in each cell, so that there's an intracellular effect, which will have both an intracellular and an extracellular reduction. We're very much looking forward to progressing that program and showing more data as we have it. May I add, Phil, that you know, as 'cause I know you know this program, but BIIB080 is mechanistically similar to that vectorized siRNA approach that Todd mentioned. And you know, there's data that's coming out of that program, the BIIB0 80 program, that if you knock down tau, actually, you decrease levels of tau in the brain, but you actually seem to show a trend toward clinical slowing of clinical progression or clinical worsening. So it's some early days, where it looks like if you can lower tau in the brain, you can have a positive effect on Alzheimer's disease. That'll need to be confirmed, of course, but we're pretty excited by that preliminary data. I think the whole field is. Before moving on to the rest of the pipeline onto tau commercial question, what are your plans here to develop and commercialize the VY-TAU01 itself, and then maybe throwing in the gene silencing program too? Would you consider exploring a partnership, and if so, at what point in development? Yeah, we would have to get a partner, Phil. I mean, I can't imagine Voyager trying to commercialize a product for Alzheimer's disease globally. I think that would be pretty hard to imagine for a small company like Voyager at this time. And if, and then if you say, "Well, if we're not gonna- if we need a partner to commercialize, then we also probably need a partner for phase III." So the idea would be that if we can at least achieve proof of concept, and then, you know, then we would have to get a partner to help us with the later stages of clinical development and eventual commercialization. Right, moving on to the SOD1 gene silencing program, can you describe that program and update us on the status of VY-9323? Sure, yeah. So that sort of, you know... We're trying to follow the path laid out by tofersen, which is an antisense against SOD1 that was just recently approved by FDA about a year ago. And, you know, what's remarkable about tofersen, actually, is that for the first time in ALS, about 30%-40% of patients seem to be improving muscle strength. And I can tell you, as a physician that used to treat ALS patients, I never saw patients improve muscle strength. And so the fact that, you know, 30%-40% of people can improve muscle strength by knocking down the expression of SOD1, is like a dream come true for a clinician like me. And so, what we're hoping to do is to vectorize an siRNA, very similar in concept to the way Todd described vectorizing an siRNA for tau. This time, we're vectorizing an siRNA for SOD1. Conceptually, it's very similar in some ways to what tofersen is doing in the spinal cord and brain. And what we're hoping to do is to leverage neurofilament and SOD1 levels. So we have a path laid out by Biogen in terms of the biomarkers we need to make sure we de-risk early. Again, the same theme over and over, we wanna de-risk early in as few patients as possible. And here, the biomarker may even actually lead to accelerated approval, as it did for tofersen. When could an IND be filed, and can you discuss the likely design of the initial clinical trials for VY-9323? Well, so, the IND, we're hoping to file next year. In fact, we already nominated a development candidate for it. We have to do the GLP tox studies, and we have to interact with the FDA. But we plan to file an IND next year. In terms of the clinical design, you know, it's not a very common disease, right? I mean, we estimate worldwide, maybe 1,000 patients, in the US, Europe, and Japan. Might be a total of 1,000, roughly. And so we're not gonna be able to do very large studies, and we don't want to anyway. We think that the concept will be, we're gonna pick some doses that... and we're gonna start with doses that have a chance of working, and we're gonna see if we can lower SOD1 levels in the spinal fluid. We know quantitatively how much we have to do that because the first one provides that clear-cut level of evidence that we need. And if it lowers SOD1 levels in the spinal fluid, we're gonna also look at plasma biomarkers and particularly plasma neurofilament, and that would be, we think, proof of concept. And then we'll approach the agency and talk about whether or not accelerated approval would be possible. But we're not, we're... and you know, and as you know, Peter Marks at CBER has been talking about how for very serious life-threatening diseases that are rare, that he's open to accelerated approval pathways with not, not, not a huge number of patients. So I don't wanna get too specific, but that's sort of the concept. Maybe moving on to the Neurocrine collaboration. For those less familiar, can you remind us of the financial terms of that collaboration? Who's responsible for which activities in development and commercialization? So basically, we had two deals with Neurocrine, one that preceded my coming to Voyager, and one that happened after I came. The first one was where Friedreich's ataxia was the target disease.... plus two additional undisclosed targets. There was a $165 million upfront payment, a total of $1.3 billion in milestones, and then additionally, royalties. And Voyager has a right to opt- in for 60/40 economics in the United States. The second one, and by the way, it really pleases me that they doubled down after working with Voyager. I think that speaks a lot to the partnership and to their belief in our work. But the second deal was GBA1, for GBA1 mediated diseases, Gaucher's and Parkinson's, plus three undisclosed targets. That brought in $175 million upfront cash, a total of $4.2 billion in milestones, and here we have a 50/50 opt-in right in the U.S. for GBA1. So these, you know, there's a joint steering committee, we work closely together. So in addition to bringing in upfront non-dilutive revenue, they take it off our PNL. So Neurocrine pays for everything up until the IND, and then they take... And then we do, but we actually make the gene therapy, both the capsids and the payload. And then they take over, and then after the IND, they take over and basically do the clinical development on their own, but we have those opt-in provisions after phase I for both FA and GBA. Great. And maybe last question on Neurocrine. What's the most recent guidance as to when those programs could enter the clinic? So, both FA and GBA, we have nominated development candidates, which, by the way, started to bring in some of those milestones we're talking about. And, that's always nice to have. And both development candidates occurred, nominations occurred in the first quarter of this year, and again, both are, are on track to, get the INDs next year. Great. Maybe moving on to some, corporate questions. In terms of business development generally, how do the deals with Novartis and Neurocrine exemplify your strategy moving forward, given how modular the gene therapy platform is, how does Voyager plan to balance collaboration deals versus developing wholly owned pipeline candidates? Yeah, so first of all, I would say that I talked about the risk inherent in neuro drug development, as well as the risks inherent in gene therapy. You know, one of the things we wanted to do was to. And I talked about how we deal with risk in our own clinical development in R&D, but I also felt it was important to de-risk, as a corporation. So having 13 partner programs, where our partners are basically, you know, in exchange for some upfront cash, also paying for essentially, everything. We get milestones, and we get royalties, some very nice returns, potentially down the line, if these drugs work. So 13 partner programs. It was also a way of de-risking, if you will. You know, there's so many targets for the brain and spinal cord, many more than we could ever pursue on our own. So in addition to de-risking and getting non-dilutive revenue and taking it off our P&L, there's no way we could pursue this, all these diseases. So we sought these partnerships. Actually, many, many, many times they came to us when they saw our capsids. I would say that, however, I wanna balance that with the fact that I wanna have some wholly owned programs ourselves. I don't wanna give away our entire future. So I think we'll continue to be mindful of that balance, the proper balance. And I'm always open to talking to any potential partners, particularly those that we've already partnered with as well, because, you know, for me, it's validation, when when Novartis did two deals with us, too. The second one was just in January of this year. And look, these are... Neurocrine and Novartis are both experts in neuro drug development, so I think we're blessed to have great partners like that. Last question, what do you think investors are missing about the story? What's most misunderstood about Voyager in the investment community? I think that if you just think, "Oh, gene therapy, neuro, high risk," I think that's what people probably immediately compute. I think I tried to explain how we've dealt with the risk, because the unmet need is tremendous, and the promise is tremendous, especially now, when we see evidence, tangible evidence, that we can actually make huge inroads into helping patients. So I think this, this concept of de-risking, both in terms of the portfolio partnerships, but also how we do drug development, that could be something that people might be missing. Great. With that, I think we're out of time. I'd like to thank Al and Todd for a very interesting discussion. Thank you, Phil. Thank you.