Voyager Therapeutics, Inc. (VYGR) Earnings Call Transcript & Summary

Hello, everyone. I'm Pete Stavropoulos biotech analyst at Cantor and welcome to our first of 2 webinars with Voyager to discuss its blood-brain barrier crossing technologies. It's an exciting time for the field. The blood-brain barrier has been a sort of a bottleneck in CNS drug development preventing many promising therapeutics from reaching their targets in the brain. Today, however, there's a wave of delivery platforms designed to actively settle therapeutics across the barrier and is gaining momentum. Over the past 5 years, we've seen more than 25 partnerships and strategic transactions centered around these technologies, highlighting the growing recognition of the potential One of the biggest examples of this trend was AbbVie's $1.4 billion acquisition of Aliado Therapeutics in October 2024. And there are numerous other examples. These platforms are already demonstrating potential to dramatically expand the reach of CNS drug development, and they will unlock a lot of value from neuro indications. We're excited to have with us Voyager Therapeutics who has a differentiated blood-brain barrier crossing platform and is developing a pipeline of brain Penetrant therapeutics. So today, we have with us Todd Carter, Chief Scientific Officer; and Mihalis Kariolis, Vice President of Non-Viral Therapeutics. So welcome. Thank you for taking the time, and we'll start off with an introduction of yourselves and a snapshot of Voyager for those not familiar.

Thanks, Pete. Really appreciate the opportunity to join and talk about this. It's work we're very excited to share. I'm Todd Carter. I'm CSO. I've been here at Voyager, not quite 10 years, been in biotech for a couple of decades working on whole set of different kinds of modalities. I'm looking forward to the discussion today. Mihalis?

Yes. And thanks for the intro Pete. I'm excited for the discussion today and share the enthusiasm about the brain delivery field. My name is Mihalis Kariolis, as you mentioned, Vice President of Non-Viral Therapeutics here at Voyager. I've been with Voyager for about 1.5 years now. And prior to that, spent the better part of 9 years at Denali helping with their transport vehicle platform and the associated programs.

Awesome. A little bit of background about Voyager and where you are currently?

Yes. So Voyager, we're a CNS-focused neurogenetics therapy company. And we're mostly known for our novel blood-brain barrier-penetrant gene therapy capsids. These are entirely novel capsids that came out of what we call our TRACER platform that we may have the opportunity to talk about a bit. But we're able to dose systemically using intravenous injections and get broad delivery across the CNS, including the brain. And we have lower delivery to off-target tissues such as the liver. Now that work led us into the shuttle platform space based on the receptors that we identified that were responsible for that blood-brain barrier penetration. This year, 2026 is particularly important to us. We like to talk about it as the year of tau because the whole field, if you look across it, we have many readouts in the field that are going to be important around tau. So we have two programs in that area that are tracking in the clinic. One is not a blood-brain barrier-penetrant technology, but it's a tau monoclonal antibody, where we'll have tau PET data reading out in the second half of the year. And the second is a single-dose gene therapy using one of our novel BBB penetrant capsids to deliver a vectorized siRNA that knocks down tau mRNA protein. And so that will be going into the clinic later this year. We also have a partnership with Neurocrine for a gene therapy for Friedreich's ataxia That's, again, with one of our novel intravenous blood-brain barrier-penetrant capsids that will be moving into the clinic this year. So all of this really is we have that tau antibody, but everything else is based on our BBB crossing technologies, one that is the AAV capsids and the second is the platform that Mihalis is helping build around developing shuttles to get a variety of things across the blood brain barrier.

Next slide. How should we think about the blood-brain barrier's role in CNS drug development? How has this barrier historically limited development of therapeutics for neurological indications.

The blood-brain barrier, I think as most of us know, excludes 98% or greater percent of small molecules and large therapeutics, like proteins, enzymes, antibodies, et cetera. People have taken a variety of approaches to do this to get whatever it is you want to get into the brain. You can dose very high to get into the brain. That's what people tend to do with the antibodies. The benefit of antibodies is that they can be highly specific, so relatively safe. And that's where the efficacy is seen with things like aducanumab, lecanemab, the anti-amyloids. One benefit of the tau antibody is those appear to be remarkably safe. So we're able to push the doses pretty high on antibodies. People also use alternative routes like intrathecal or direct brain injection. People have tried that or have been using that for gene therapies for all of the nucleotide-based therapies, like ASOs or siRNAs and while there are the likely opportunity there, particular diseases that can see benefit, it's pretty challenging to get quite broad delivery and safe distribution that you usually want for most neurogenetic diseases. And so we're pretty excited by these next-gen CNS therapies like our novel brain penetrant and and capsids like the Shuttles to enhance brain penetration and get that broad delivery with either lower doses or improve the therapeutic window.

Can you just briefly walk us through the mechanisms that the brain sort of uses to transport molecules across the blood-brain barrier that you're going to eventually leverage?

Yes. And I'll say a couple of things, and then maybe Mihalis can jump in and give us his thoughts as well. But basically, what we know is getting things into the brain, the body needs to do that. There are active processes that we hope to piggyback on. The brain needs to transport things across the blood-brain barrier. And these different opportunities, different receptors gives us the opportunity to identify these novel ways of doing that. Mihalis?

Yes. Just to go a little deeper on that point, Todd, I think you set up the challenge of the blood-brain barrier for drug delivery, but it's important to recognize that it doesn't only keep out the drugs that we're trying to get in. It's actually a barrier to all of the macromolecules the brain needs to properly function. And so in order to maintain the integrity of that barrier for safety, but allow the brain to get access to everything that it needs to properly function. There's this network of active transporters that the body has evolved over time to bring things these important nutrients from the blood and actively bring them into the brain. And that's such a critical concept because we can actually begin to exploit these endogenous pathways for drug delivery. We can engineer platforms that bind to those receptors and hitch your ride in. And to highlight something that Todd briefly mentioned, the blood-brain barrier is so extensive. There are over 400 miles of vasculature within the BBB. And so you can achieve this really uniform drug distribution and delivery across the CNS and into deep brain regions by leveraging this native biology that exists already. And something to highlight, Pete, is that when we often think about receptor-mediated transcytosis or that process of pitching a ride in and leveraging these native receptors, we often go to those pathways that are well studied in the field like transferrin receptor or Glut1 or CD98 but the brain needs so many different things to function. There are many RMT pathways. And I would suggest that there are a number that the field has yet to figure out. And so what's really exciting about the time we're in now, is the concept of leveraging RMT for drug delivery to the brain is starting to be proven out. The different ways that you can go about doing it. I think we're still discovering newer and newer pathways in. So we're at this really neat stage where this process has been proven out. And now we get to go see how many different ways we can go and leverage it.

Awesome. How do you see these delivery technologies or changing the success rate of neuroscience drug development?

Yes. So I think the delivery has historically been a problem. Todd mentioned you having to dose really high in order to get drugs where they need to go. I think one of the most recent examples you can point to for how transformative these approaches can be is actually gantenerumab. The gantenerumab is an amyloid antibody that was tested on its own and didn't really achieve the clinical success that the field was hoping for. Now Roche have ended their TfR binding brain shuttle to that exact antibody, and it has now set the standard for what plaque reduction in amyloid clearance looks like, both in terms of safety with respect to ARIA, the kinetics of plaque clearance as well as the total reduction. And so this was an antibody that absent active shuttling was not efficacious in the clinic and now is starting to set the bar for what amyloid therapeutics look like. So I think it highlights how when you can lower the doses, you can get more drug into brain. We end up widening the therapeutic windows, and we're able to actually start to ask the question will these drugs be efficacious. There's not a lack of good therapeutic targets in the CNS. Historically, it's been a lack of an ability to get the drugs that hit those targets there. And so with these brain delivery platforms, we're going to start to see much safer, more efficacious drugs because we're being able to deliver them to where they need to go. Todd...

Also add in a little bit that I think classically, small molecules have traditionally only been a drug class that we could get effective delivery. And as we mentioned earlier, the vast majority of those can be tough to deliver because various targets can be challenged. The things you need to do to make something bring penetrant to make the other characteristics you want. It might be somewhat mutually exclusive. And we have these other routes that we can use intrathecal or direct injection, but those can be difficult daunting for patients and health care systems or just not suitable for delivery. So I think one of the things that we're hoping to see and that there's evidence already, is that we're able to access targets that are beyond the targets that can be typically achieved with small molecules, for example, or maybe even antibodies. And so that opens up a whole new class of targets that we can begin to interrogate.

I mean you just mentioned gantenerumab and there, how they put it on to the shuttle suddenly changed success. But overall, do you think historical drug failure is were due to drug target and biology versus the delivery limitations and entry into the CNS.

So that's a good question. I think that my hedging answer to that is we don't know. And the reason is because we couldn't test -- the hypothesis weren't tested. We weren't getting good enough delivery to know whether we were accessing the right targets. So I think that over the past decade or a little more, we've seen these wealth of targets that Mihalis mentioned be identified. That's really been the key part of the genomics and genetics revolution I think we've seen in the periphery and in cancer, a lot of progress now that we can go after those. What we need to do now is test that in the CNS, and we need these delivery mechanisms to be able to test that.

All right. which disease areas of neuroscience are most likely to benefit from this shuttle technologies? And what's sort of the low-hanging fruit?

Yes. So I don't know, in neuro, I don't know if there is a low-hanging fruit. I think -- but there are opportunities on some things like gantenerumab is a good example. You had a drug class these anti-amyloids that clearly can be efficacious and there's a clear test of if you improve delivery, can you improve efficacy and, in that case, even reduce the safety concerns. So one area are things where we have hints of success can we rapidly test those to see if we can get even better success or greater therapeutic indices. I think others are in Neurodegeneration. We have a whole set of these genetically identified targets. And those are some of the first places to check. And that's certainly something, as I mentioned, we see ourselves as a neurogenetics company, and that's why because the -- some of the biggest derisking on targets, I think, are the genetically validated targets. Nature has already done the experiment for you in some ways to say that these genes and these targets are important for the disease. And so things that are genetically validated, I think, are going to be the equivalent of a low-hanging fruit. I think there are others with chronic diseases where if -- for something like a gene therapy, where you're one-and-done or a shuttle-based approach where you can go intravenously and get the broad delivery, that's another place where the mode of delivery ends up being pretty important because some of these diseases, if you want to go in and get direct injection into the spinal column or into the brain in repeated ways, those just aren't really going to be feasible. So I think that there are opportunities there. Mihalis, do you have any thoughts on this?

Yes. I think you highlighted a couple of real drivers there, right? One of them is up until a few years ago, these delivery technologies still had yet to bear out clinically. And so in order to not stack the risk too much, it's about taking therapies that had hints of success, whether it's the amyloid or what's seen a lot of interest is actually the lysosomal storage disorders where peripheral enzyme replacement therapy has been quite effective to treat the peripheral manifestations of diseases, and that has really been the proving ground for these delivery technologies. Now what's exciting is with the approach relatively derisked now for brain delivery, we can start to go into some of these additional areas that Todd mentioned. The genetically validated targets that may have been intractable and asking the platform question and novel biology on the therapeutic target may have seemed like a bridge too far. It's no longer the case because these delivery platforms have proven out. So I'm excited to see where we can go particularly in these new spaces with the genetically validated targets.

So one thing that I might just comment on is with tau knockdown in particular, last year or 1.5 years ago, Biogen reported some very early but very, very promising data with their tau knockdown ASO. It's intrathecally delivered. It hits the Tau mRNA with an ASO. If the data that we're expecting later this year, that data looks as positive as the preliminary readout suggested that it could be. That's a great validation for a knockdown approach, and then we're able to move very quickly in with our gene therapy. And so one and done with an intravenous delivery with our gene therapy there. So that's another example.

So there are a number of active transport technologies that are emerging to shuttle biologics across the blood-brain barrier. What are some of the key design principles that you believe will determine whether these platforms succeed?

Right. That sounds like a Mihalis question.

Yes. There are a couple of things to think about, right? One of them is what receptors are you targeting? And I think that's one of the things I'll highlight about Voyager's approach is our unbiased screen to identify functionally validated receptors gives us some unique angles to ask the delivery question in terms of what can you target in order to achieve it. But at the end of the day, for me, it's three big things. It's really about delivery to the brain broadly biodistribution and of course, safety. And what do I mean by those? So for delivery, it comes down to how much drug can you actually get into the brain where it needs to go? And what does that delivery profile look like? So using some real examples, we're very familiar with what TfR can do. It can rapidly deliver drug to brain and then you see a fairly rapid clearance peripherally and in brain, and that's due to the expression of TfR peripherally and that binding and basically clearing out a lot of your TfR binding drug. And so what that does is it can limit the extent of brain delivery over time. On the other hand, what we've shown by targeting ALPL is that we have some differentiated kinetics because that high clearance associated with peripheral binding is not there. So binding to ALPL can achieve sustained brain uptake for weeks after dose. Now it's important to note that both profiles are really valuable but it's about having the optionality and building up different types of platforms that get you different delivery profiles. And that's pretty linked to biodistribution as well, that second pillar that I think about, which -- when we think about technologies for their CNS delivery, sometimes it's easy to forget that a lot of these receptors are expressed peripherally and throughout the body as well. And so how much expression outside of the BBB is there can actually have a huge impact on the duration of exposures and where your drug goes, just like we were talking about for TfR. And you can even go deeper on that biodistribution question and say, okay, if this is a good target to deliver you to brain, where is this target expressed within the brain and how does that influence the kinetics and the distribution of your CNS penetrant drug once you've actually got it across the blood brain barrier. And so that delivery and biodistribution angle, I think, dictate a lot of the ultimate efficacy in your drug and last and certainly not least, which is kind of co-mingled with those other two is safety. So safety is really tightly related to where your blood-brain barrier target is expressed and certainly relative to biodistribution. Where are you pushing your drug by adding this binding to it. And a lot of it comes down to on target. So for a given RMT target, what is its native function? And how are you engaging it? are you doing something that disrupts the important physiology that's there naturally. The brain has evolved to express these receptors for a critical reason. It's on the brain vasculature because it's important. So we don't want to be messing with the native function of those receptors. And then the other thing with respect to safety that we need to start thinking more about or continue to think more on is the co-expression of therapeutic targets and blood-brain barrier targets. Again, the blood-brain barrier targets are often expressed in a lot of peripheral tissues. And we can generate novel biology for binding the RMT target with the shuttle and a therapeutic target a Fab bound antibody, for example, and sometimes bringing those receptors close together can be a good thing and sometimes that can lead to some unwanted effects. And so thinking very carefully from a safety perspective about how we're engaging these receptors, not just at the blood-brain barrier, but in concert with the overall therapeutic as one molecule is something we need to continue to really be focused on.

Awesome. I'll kind of ask a question, the odds are you're not going to answer. Have you identified any receptors that sort of are a little bit more selective towards crossing the blood-brain barrier and let's say, have limited or no expression in the periphery?

So we disclosed ALPL as a receptor that we're building the first version of the neuro shuttle on, and we look forward to discussing some of the other receptors that are coming in soon.

But I would add on ALPL in particular, one of the reasons that we're excited by itt is we have this very differentiated PK profile where transferrin we know goes up as pretty well, but it is rapidly cleared. ALPL, we get more into the brain, and then we're able to maintain that much longer because we don't have this peripheral clearance. . But we believe that's because ALPL is not expressed as broadly as transferrin is. So there's an example where we think the different expression patterns are really driving important differences in how the receptor works.

That's great. All right. So I guess part of my question was, will these platforms sort of be a plug-and-play? Or will target an overall therapeutic approach and modality impact the overall design of these candidates. In other words, even the need to explore multiple formats and it's less of a plug-and-play though you sort of alluded to the sensitivity of the design...

Yes. Maybe I'll take this, Todd, and then if you got some stuff to add. So ultimately, what we want to do by building platform technologies is to invest in the first therapeutic and hopefully make the subsequent ones a little bit more plug-and-play building on the derisked platform. I think the reality is that every therapeutic is going to require a few tweaks to it. So having a toolbox within your platform of different flavors to match the platform to the therapeutic will likely have to be the case. What I will say is that the first program through will derisk the overall approach of binding a particular receptor, so we can rapidly leverage that platform de-risking to have more confidence in subsequent programs in the pipelines we build on top of that. And then within therapeutic modalities, I think Pete, you were alluding to this, it's probably going to be a little bit more plug-and-play. So for an oligo delivery platform, for example, you work things out like the conjugation sites and chemistries and the affinities required. You'll tweak a little bit when you move from oligo to oligo just to make sure that the safety and the exposures and the stabilities hold when you change in the actual oligo moiety. But that will be more or less plug-and-play. You're not rebuilding the platform from the ground up each time.

Okay. So how should we be thinking about the relationship between the blood-brain barrier receptor sort of binding affinity and valency at the blood brain barrier and the sort of the efficacy of transcytosis into the brain.

Yes. So I mean, that's a great question because at its foundation, it highlights the importance of engaging blood-brain barrier receptors in an optimal way. It's not just enough to bind to a receptor. You have to do it in a way that's tailored to match the receptor's native biology. Remember, we're not looking to create any new biology with these platforms. we're looking to take advantage of a natural process that already exists without ideally altering it at all. And so the short answer is that every receptor is going to have its own native biology, it's expression, it's trafficking, it's kinetics, what causes it to internalize. And so approaches are going to have to be empirically determined for every receptor. And I'll give you one example, just like hypothetical, if a receptor takes 10 minutes to move across the cells from the blood side to the brain side. You don't want to bind to it so tightly that it takes you an hour to fall off, for example, you want to match that affinity and more specifically, the kinetics of binding so that you're falling off in about 10 minutes when that receptor makes it to the other side. And so these considerations are some of the most impactful design criteria when you're thinking about building a platform that determine how effectively these blood-brain barrier receptors can actually be exploited. But every receptor is going to be a little bit different, and it's about leveraging the native biology and understanding the native biology enough first to figure out how to leverage it.

Okay. I guess -- I mean you sort of again alluded to it, but like do different receptors, ALPL, transferrin and CD98 have sort of different optimal affinity windows based on the intracellular trafficking kinetics?

Absolutely. And I think you see that play out in some of the data that we've shown with the ALPL platform and that the field has shown on CD98 and TfR, where just the kinetics of brain uptake alone give you a lot of information about the receptor turnover and kinetics of trafficking that you're exploiting and they're going to be very receptor specific. So again, it just comes back to engaging a receptor in ways that leave it's native trafficking is untouched as possible.

All right. How well does the affinity-exposure relationship observed in rodents translate to nonhuman primates and then ultimately to humans.

Yes. So this is one of those questions that it's fun to get now as opposed to maybe 3 or 4 years ago because we're at a really exciting time where we are generating more and more clinical data with the TfR based approaches. So we can actually start to answer the totality of your question, not just the preclinical side of it. And I think at a high level, it looks great for TfR. A couple of things that maybe I'll highlight a lot of times, we start in the mouse because that's how we can derisk some of the biology of new receptors and iterate quickly. But the mouse has a notoriously leaky blood-brain barrier relative to higher order species like nonhuman primates and certainly humans. And so when we look at mouse data, there are much higher levels of passive drug in brain so in a lot of ways, we underestimate the need for and the value of these delivery platforms when all we're looking at is mouse data. But it certainly becomes more clear when you pull it up into nonhuman primates the important of and the value of active delivery, you get meaningful drug concentrations of brain. So as I mentioned, that relationship has scaled really well for TfR from mouse all the way to humans. And we've seen evidence from mouse to nonhuman primates as a field, not just for TfR, but for CD98 as well, and we're looking forward to generating some of that data on ALPL. And I'll point out that we can get additional insights for the mouse to nonhuman primates actually from the capsid world and from some of our own TRACER capsids where that correlation is now held. We've seen from mouse to nonhuman primates for ALPL as far as our TRACER capsids do. So early days overall for drawing those relationships and those correlative but so far, what we see in mouse for the receptors that have been studied in nonhuman primates is fairly predictive and certainly, the nonhuman primate data to the clinical data relative to TfR and what's been published has held well and is pretty exciting because, again, to me, it proves this idea of receptor-mediated transcytosis not that it's a TfR specific thing, but that is this idea of leveraging native receptors more broadly. And so it really opens the aperture for what you can go after and do.

All right. So how should we think about selection of isotype backbone and Fc function? Is this something that will be sort of 100% modular and plug-and-play?

Depends on how you design your platform from the start, a lot of these antibody fragment-based shuttles are done with that in mind where it is plug-and-play. The beauty of these fragments, whether they're Fab or single domains or scFvs, is that you can append them to just about anything. And when we're talking about therapeutic antibodies specifically, that shuttle moiety is fairly independent from the isotype that you're looking for or certainly Fc modifications if you're -- if you want to modulate and optimize Fc-gamma receptor binding and effector function, for example, or FcRn binding to confer differential clearance rates. So these shuttle platforms broadly can be thought of as plug-and-play to different antibody backbones.

All right. Is there a possibility of marrying this tech to extended half-life or not?

Yes. So I would clarify this tech and what you mean by that

Blood-brain barrier-penetrant technology.

Yes, broadly speaking -- and I'll explain why I wanted that clarification. So it's an intriguing idea to try to get better exposures in blood for your transport platform because ultimately, what dictates brain delivery is how much you have in blood. If it's not in blood and you don't saturate the uptake mechanism, there is no brain uptake. So the more you can keep around in blood the longer, the more brain uptake you have. The challenge to reducing that to practice can come from the magnitude of clearance that some of these brain transport receptors are capable of. So transferrin receptors clearance rates are so high because of how fast or how broadly it's expressed peripherally and how fast it recycles that applications of these half-life extending technologies that by two, maybe threefold improvement in terminal half-life on a monoclonal antibody, can compete with the clearance of TfR. But if you move to a more neutral or a receptor like ALPL where we don't see that type of accelerated or higher clearance peripherally, then it becomes a really intriguing idea because you're going from -- we've shown data out to 3 weeks now. So with TfR, if you're doubling 3 or 4 days of exposure, that's meaningful. But if you're doubling weeks upon weeks of exposures with ALPL now, now it puts you in a whole different regime in terms of what you might be able to achieve for dosing. So the short answer to your question is, yes, you should be able to marry these. I think the more nuanced answer has to do with you've got to do it on a platform that doesn't have such high clearance in order to really see the value of doing it. And that's one of the things we're excited to look at for ALPL.

Okay. So what are the key safety risks regulators are sort of watching out for with these shuttle technologies? Is it risk due to target receptor expression patterns. Is it design of the Fc region and effective function? Or is it specific to the target, say, amyloid beta or a combination?

It's all of the above. I mean, we're still early days in these active transporters. And again, sometimes we think about these things in pieces, we have a blood-brain barrier transporter, but that's expressed everywhere else. We have a therapeutic and sometimes that's expressed outside the CNS. When you put them together, and it's no longer a therapeutic in a shuttle. It's just multi functional molecule that can do a lot of different things. So I think all of the things you touched on are going to be critical aspects of safety as we continue to translate these obviously, top of mind for the field and the agencies right now are the anemia risks seen with TfR. A lot of it is going to continue to be the acute IRRs that could potentially come from binding the receptor particularly if you have a factor function on these molecules and triggering an immune response. We have to be careful about that. And that's the acute over time under chronic settings, it's more about how are you impacting that native receptor that you're trying to exploit for brain delivery. What's the normal function? Are you altering it? Does that lead to any detrimental effects and that's part of the anemia for TfR. And I'll highlight what I kind of said earlier as well I think one of the unique things about these shuttled molecules is that they're multi specific. And so you can end up getting new biology and new pharmacology that comes from bridging whatever your therapeutic is supposed to do with your shuttle technology and so if we're binding a blood-brain barrier receptor and a therapeutic target, not at the blood-brain barrier, what's happening? Are you creating some unwanted effect and so all of these things are top of mind, certainly as we engineer and characterize new platforms and certainly when we start to translate them clinically.

Maybe I'll just add a couple of things as well. Just as we look at transferrin and ALPL, so Mohalis' point about you've got in some ways almost two different on-target potential safety concerns that you have to consider. There's the on target for the shuttle aspect and then there's the on target for the therapeutic that Mihalis mentioned. And if you think about transparent, the anemia, et cetera, for something like ALPL we know that we don't see evidence of any issues with ALPL. We don't expect to because we don't express on the cell type that drives that transferrin. And we have ALPL expression and other things in bone, et cetera. So we know where to look, and we're doing all the right experiments to look at all that. But then on the target side of things, the benefit, hopefully, is that you're delivering more of your therapeutic to the places you want to deliver as well as maybe to some other places that you don't necessarily need to deliver. In the case of something like gantenerumab, where suddenly you're delivering parts that actually reduced the ARIA risk associated with anti-amyloid, which was a bit of a surprise, but a welcome one in that context. So we have these different things to look for and both the shuttle-based on target, the payload-based on target and then of course, the off targets associated with all those.

All right. Next Slide. I had a series of questions, but I want to make sure that we get to Voyager's candidate. Which of these sort of receptors do you see the most and least promising and why?

When you say which of these receptors, so transferrin versus ALPL or?

All of them, like out of those, which ones do you think -- I mean, obviously, we know that transferrin -- you guys have been successfu. What's left over we hear that other groups are targeting, do you think has the least probability of success or perhaps some toxicity known versus a higher probability of actually allowing it to be leveraged.

So I'll make -- I'll say a couple of things maybe on the little bit on capsid side, too, and then ask Mahalis to jump in with his thoughts. On one of the things that we're actually really excited about, we've talked -- we're talking a lot about ALPL on our platform side. But our capsid work which is what led to this. It was particularly we find interesting because we screened millions of different variants and identified capsid across the blood-brain barrier without knowing what's driving it. It was an empirical approach. What we've been able to do is to take that and then identify the receptors responsible for delivery across the blood-brain barrier. ALPL is the first one, but we have multiple other receptors now that we know can deliver these large -- very, very large macro molecules that are AAV capsids across the blood-brain barrier. So now we have a diverse set of receptors that we can then explore for delivering antibodies, oligos, et cetera through. Now ALPL is the first one through and that was, in many ways, a test case. Could we apply an AAV receptor, BBB receptor for other things. And the data that we're showing we're generating say that we can. So that makes us very excited for the others. The other thing that's really interesting from a Voyager platform perspective is that these receptors we're identifying are not anything we would have expected to find. So these empirically identified receptors are not things that we would have chosen or we think anybody else would necessarily have chosen, given the typical things that people look for. And so we're really excited about those opportunities. So in a way, what I'm saying is for these novel things that we now have in our toolbox, I don't know the answer to your question because we need to do more work to be able to look at it. What I do know is that if we take ALPL is the first one through with Voyager specific set, we're seeing very differentiated kinetics, very differentiated PK profile that gives us what we think a good opportunity. We don't think any single receptor is going to be perfect for all diseases, and so we're going to need this toolbox. Mihalis, do you have any thoughts on kind of the existing set that's out there?

I'll just put a finer point on what you just said because I think that's worth repeating. There is -- I don't think that there's one or two or maybe even three receptors that are ideal. When we think about the opportunity that's in front of us, it's about expanding the scope of what we define as neuro therapeutics by thinking about all the different types of drugs that we want to deliver all the different indications. And when you start to do that, the number of different types of delivery profiles and capabilities becomes pretty big. And it's pretty inconceivable to think that TfR or TfR and another receptor is going to solve all those problems in the most efficient and optimal way. So Pete, it's really about figuring out how many of these receptors can actually mediate meaningful drug delivery to brain, what are those properties in terms of the kinetics of distributions, all the things we talked about before. And how do those uniquely enable the types of drugs we want to get across. So I don't necessarily view them as good or bad receptors. I view them as what are they uniquely good at doing? And how many of those can we stack up to enable all the different types of neuro therapeutics that we want to go after for patients.

Got it. All right. You briefly touched on this Todd. But anything else you want to add in terms of the discovery process and what were sort of the key steps and insights that led you to focus on ALPL as a receptor of interest? .

So it was what I mentioned before, that empirical approach, we identified and have identified multiple capsid families that use different receptors. And so it will be multiple capsids quite often that target it and these capsids are related. I mentioned that we know by definition, if it gets an AAV across that it can get a large macromolecule across. Of course, with AAV, essentially, what we're doing is we're adding a function, and we've shown this in our studies with ALPL and presented this that most of the activities of the AAV capsid already are left intact. So we know that we can deliver an AAV across the blood-brain barrier and then other aspects, the other intrinsic properties is that AAV capsid essentially still apply. So it can get into cells and do all the kind of complicated biology that the capsid needs to be able to do to deliver its vector genome into a cell to get it into the nucleus, it uncoats, it forms an episome and begins to express. Different receptors may be expressed -- well, they're expressed in the vasculature they would have to be to deliver across the blood-brain barrier, but they can be expressed in different ways in different cell types on the other side. And so what you can see is the opportunity to have different kinds of profiles where a particular capsid -- sorry, a particular receptor set might give you delivery into neurons once it crosses, might get to delivery into both astrocytes and neurons. And they give you delivery into other tissues of interest peripherally. So you might want something that gets into the brain and muscle for example. Transferrin might be an example, assuming it has the other kinetics and PK characteristics, good for whatever disease you want to go after, but there might be other diseases where you want different sort of subsets of tissues that you want to target as well. And the exciting thing about the receptor discovery that's really become an engine for us is that we get these novel receptors that have these different profiles based on their capsids that we wouldn't have ever predicted in the first place.

What's the natural function of ALPL receptor and what feature of that biology make it sort of well suited for targeting?

So that's an interesting question. I don't know that a specific biology does. Maybe Mihalis has some thoughts on that, too. It's a phosphatase. So it's what we call it a GPI-anchored protein, meaning there's a particular attachment to the membrane that can be cleaned and it could be released into the blood stream. It's interesting because as you look at the capsid side of things, there are multiple GPI-anchored proteins that have been involved in getting capsids across the blood-brain barrier but we now know that transferrin in sort of the reverse direction can be used to get capsids across. It started as a shuttle approach, and now we know we can get capsids across as well, and that is not a GPI anchored protein. And so there's a diversity there. The specific phosphatase biology of it is probably not what's important, I don't believe. I think its involvement in that transcytosis that trafficking across the blood-brain barrier is what's important about it. And then why this particular protein is involved in that, I think we don't know. And that's, again, what we're learning is that there are these multiple receptors that we wouldn't have hypothesized to be important to turn out to be.

All right. Is ALPL, do you view it as mechanistically similar to TfR or CD98 in terms transcytosis or are there just meaningful differences that you should highlight?

Yes. Go ahead, Mihalis.

Yes. There are probably differences. But to Todd's point, we still don't really understand for ALPL and I would argue for TfR and even CD98, Glut1, some of the other receptors, the actual mechanism of transcytosis once it gets within the cell, I think what's worth highlighting is that in that set alone TfR is a single-pass transmembrane iron trafficker, CD98 heavy chain is actually a chaperon for the light chain amino acid transporters, ALPL as a GPI linked phosphatase. Glut1 is a multi-pass glucose transporter. So all of these very phenotypically and functionally diverse receptors have shown the ability to be brain transporter. So just looking at the diversity of inherent native biology, I would say there's a mechanistic difference there. And it speaks to the diversity of things the brain needs and just the opportunity to go using approaches like TRACER, figure out what else we can leverage and exploit to get these brain uptakes. But there are certainly some differences we just need to do a little bit more work to mechanistically understand what they are and how these things work.

So how does engaging a GPI-anchored receptor changed the intracellular trafficking pathway compared with like classical transmembrane receptors like TfR.

I'll go back to, we probably don't know a lot of the answer to that yet. Todd mentioned that a lot of the early work in the AAV field for brain delivery. Certainly, the work in mouse early for things like Live6a that never actually translate into humans because we don't have that. The early AAV work ALPL included, highlighted a few different GPIs. So ALPL is not uniquely the GPI that's been shown to confer brain delivery. So there's certainly a class property for these types of anchored proteins that what makes them special in terms of being able to transport into the brain. I mean, unless Todd, you have other ideas, I think that's still something that we need to figure out and mechanistically is poorly understood.

Yes. Much of transcytosis, I think, is sadly a black box to the field right now. And it's interesting, you mentioned -- I mean your question, Pete, was around how do these change transcytosis. I don't -- we don't know that they do and Mhalis pointed out that the goal would be not to, right? We want to piggyback but not alter in a perfect scenario. Of course, we knew that doing anything can perturb things to some degree. We want to minimize that as much as possible. But right now, I think we still have a lot to learn.

Right. I'm not sure if this is a black box as well, but like when you do have GPI-anchored receptors like ALPL, are there any advantages in terms of avoiding lysosomal degradation?

I think right now, we don't know that, we don't know the answer to that. I think we're seeing -- we can most likely and Mihalis, correct me if you see it differently. But I think we still end up in the lysosome. And in fact, lysosomal based degradation can be -- you can use that to your advantage in some situations. If you want to release your payload or things of that nature. So I think there are both opportunities and challenges associated with getting into and through the lysosome.

Yes. I would just highlight that I would caution against a binary view and not that, that's what we're discussing right now, but these trafficking pathways as much of a black box as they are. They're very complex. And it's not that each receptor has a singular pathway that it takes. There's some amount that gets recycled back to the cell surface. There's some amount that goes across the cell in transcytosis, and there is some amount that will go naturally to the lysosome just to regulate the kinetics of native receptor expression. And so it's about how much of your receptor goes to each one of those different pathways that dictates the amount of drug delivery to brain. And we just need to do more work to understand how the GPIs are doing it relative to other things like the single-pass membrane receptors like TfR.

I'll comment just there as part of an earlier question that you've got this particular slide up. One of the things about ALPL and our ability to bind and what we saw coming out of the capsid screens is as part of our capsid screens, because of what Mihalis mentioned, where the first brain penetrant capsids were identified in mice, but they only worked in mice, they didn't translate into nonhuman primates. And in fact, they didn't even translate broadly throughout all strains of mice and so we've built in the need to identify and screen for things that were cross species and cross species across nonhuman primates, different nonhuman primate species and when possible into rodents. And so what we identified, we identified capsids that were able to do so. And it turns out that it's pretty important. It's important because once we started discovering the receptors, we understood why and it made sense. We were binding parts of the receptor that were conserved between species -- the ALPL capsids were across species, again, in mice, multiple nonhuman primate species. And once we knew the receptor, we could test the human ortholog of the receptor and show that we were able to also bind the human form that in in vitro studies, mediated transcytosis in vitro with the human as well as the nonhuman primate mouse. But also can give us a leg up on knowing where and what to target on the shuttle side. So it gives us a lot of information and ability to move the shuttle-based platform forward pretty quickly.

All right. So from a safety consideration, what are the theoretical safety concerns for ALPL engagement? Any potential safety signals from human genetics or loss of function mutants?

So we mentioned this a bit earlier, but ALPL is important, has an important role in mineralization skeletal and dental mineralization. And folks that carry mutations in their ALPL gene. If it gets around 30% or lower activity, overall, that can result in this hypomineralization and some cardiovascular complications. And so that gives us something to look for. We know that the -- when you look at the genetics, there are data out there where you can evaluate kind of the risk. So if you look at databases with Nomad and others, they can give you an evaluation of the risk of loss of function and based on the carriers in the population. And when you do that, transferrin comes up is very highly concerning. And we know that that's true that transferrin -- that humans have a pretty low tolerance for loss of function in the transferrin gene. ALPL can tolerate much more loss of function, but it's still something that we need to take a look at. And so we're doing those experiments and evaluating that to look in the bone, in particular for any issues there.

Alright. Next slide. One of the attractive features of the NeuroShuttle platforms is apparent applicability to different payloads, different payload types. How flexible is the system when paired with different therapeutic cargoes such as enzymes, antibodies or oligos?

So we're getting started with this. And Mihalis, maybe you can answer that.

Yes. So what we've shared to date is the delivery of antibodies using these shuttles, and it's looking quite differentiated. I think we've mentioned it a couple of times now with its exposure profiles relative to what can be achieved with existing platforms, say, like we're targeting TfR, and that's something we're really excited about. We look forward to sharing some more data in the future as we continue to see how well ALPL pairs with some of these other therapeutic modalities that are top of mind for all of us.

All right. So next slide. In the preclinical studies, you just mentioned now and before different PK sort of PK. You demonstrate sustained brain exposure for ALPL shuttle lasting about three weeks. Do you believe that sustained brain exposure is primarily driven by improved transcytosis or by little clearance once the antibody reaches the brain.

Yes. This is a good question because I think the field has learned that there are differences when all you're doing is looking at concentration. Certainly, I think the lessons out of the CD98 data, where there is a lot of retention in brain as opposed to continual active uptake -- so this has been a question on top of mind for us. And we believe that this extended exposure profile that we're seeing both in blood and brain is not only differentiated from TfR but is a direct result of having this less clearance from ALPL periphery. And what this allows us to do is keep those blood concentrations higher for longer. And that's critically important because in order to get into the brain, you need concentrations in blood that allow you to saturate in our case, ALPL binding and that uptake mechanism at the blood-brain barrier. So this is what happens with the NeuroShuttle and the data that you have on the slide here is that the exposures are high enough peripherally that we're taking advantage of active uptake and what we're showing here for at least three weeks post dose. And we've run the studies out for three weeks here, but are looking forward to seeing just how far can we continue to take advantage of this active uptake? Again, it comes back to that preservation of peripheral exposure. So this looks to us to be continual uptake over time as opposed to retention in brain in the acute week or so. Right?

Do you expect a linear relationship between systemic dose and brain exposure? Or does sort of transport saturate at higher concentrations?

Yes. I mean, just theoretically, every receptor is going to have a saturating amount that you can use to get in. It's expressed at a certain level, and it turns over at a certain frequency. And those two things together will dictate how much drug you can get into brain by binding it. So I would expect a linear relationship up into that point of saturation. I think what we're seeing here for the data on ALPL is certainly saturating. And then as you start to dose down and get below that saturation threshold, I would expect a linear dose.

Next slide. In preclinical studies, are you seeing differences in regional brain distribution following ALPL-mediated delivery?

So one of the things that I'll comment back on that we touched on briefly before is these data, along with what we see with TfR and other targets, I think, highlight the value of leveraging receptors and RMT at the blood-brain barrier because that 400 miles of brain vasculature that makes up the blood-brain barrier gives you access as long as your receptor expression profile is pretty uniform, which ALPL seems to be to all of the different brain regions across the CNS and access to the deeper brain regions in particular that are really hard to get to if you're not leveraging an active process. So we don't really see a lot of regional distribution, and I think it's tied back to this ability to leverage the totality of the blood-brain barrier for this type of transport.

Are there any cells that NeuroShuttle targets within the CNS? And is there any way to sort of enrich in certain cell types versus others? say you want an ASO?

Yes, it's a good question, and this is a theoretical one about what do you want your shuttle to do and how active is it once you get into the brain? We talked about that a little earlier. So this data is showing actually binding to and colocalization with neurons and cell uptake in neurons. So we were a little bit surprised at from the start because there's not reported to be a lot of ALPL expression on neurons. This data and the histology has repeated out enough that we're very confident in the conclusion that we're associating with neurons in this case and seeing uptake. But I think it just comes down to what you want your platform to do or you're seeing broad delivery and the ability in this case to associate with the cells in the parenchyma.

All right. you briefly touched on this a little bit earlier in terms of the way that you've validated across species. But how confident are you that this cross-species conservation will translate to consistent brain barrier transport efficacy in humans.

Yes. So the biology of the receptor appears to be fairly conserved both in terms of function as well as expression from these mouse all the way to human. So I think the confidence is pretty high that what we're seeing in terms of delivery here will be recapitulated as we move it to cynos, and we look forward to doing those experiments in nonhuman primates, which will be a big I think, confidence boost and derisk on the way to human.

And I'll just add to that and second that because it's hard to imagine better data packages to suggest that kind of translatability. We know what the receptor is. We know where we bind, we can show across these species where it works. So I think we're about as -- it's about as translatable as it can be without just testing it in humans.

All right. Next slide. The data suggests that ALPL shuttled antibodies achieved plaque engagement comparable to transparent settled antibodies after a single IV dose looking at the images, both the shuttle seems -- or appears to achieve broad brain distribution. The ALPL shuttles show continued uptake between day 3 and 14. What does that sort of tell us about the pharmacokinetics of this receptor transport and the durability of receptor engagement.

Yes. So I think this speaks to something that we were talking about a couple of questions ago, Pete, which was that are we just doing receptor uptake in retention in brain? Or is this actually active uptake over the entire PK profiles that we show? I think this data, in particular, is a really good one to point to active uptake over those entire three weeks in this case that we were showing, where we do see some active transport on the acute 3-day time point. But really, the value is the area under the curve of drug exposure or the AUC here. And these data were generated two weeks after dose, what you have on the slide. I think what's really exciting is to think about, and these are experiments that we're looking forward to doing. But if we're already at TfR's level two weeks after dosing. And we know that we have weeks to go in terms of sustained brain exposures and uptake post single dose. What does this start to look like after a month or even two months after dosing. That's where you can start to really take advantage of those extended exposure profiles to drive target engagement and ultimately, therapeutic efficacy from a single dose that we just can't with a TfR-based approach.

So did you see any differences in regional penetration or cellular localization ALPL versus the TfR approaches?

Certainly not regional distribution. I think TfR is hard because it's expressed and there's a lot of catabolism and cell uptake once it gets into the brain. But certainly, in this model of immunodecoration, we don't really see any differences in where you can access the plaques that you decorate from a spatial perspective from the two platforms.

Got it. So I guess a little bit of speculation or a question of like, is it correct to say there's gradual accumulation in the brain? And if so, could this lead to an improved sort of safety profile if you were to target amyloid beta, say, lower rates of ARIA?

So definitively, this data suggests that there is a slower immunodecoration of plaques over time. Now by two weeks, we're at what TfR can do acutely. So we are achieving similar levels and hope to be able to show more. But the relationship to ARIA, I think is one that's still a little bit of a open-ended question in the field as to what is actually causing it? And how are the results Todd mentioned trontinemab's lack of ARIA, a reduction of ARIA was a little bit of a surprise to the field. So without having a much clearer understanding of the mechanism of ARIA and what's driving its reduction with TfR targeting -- it's just pure speculation at this point as to whether other approaches, how that's going to translate out.

Next slide. So in certain diseases, such as Alzheimer's, ALPL expression appears to be elevated. To what extent could higher ALPL levels in Alzheimer's allow for like lower effective doses or improved tolerability versus targeting another receptor, again, probably speculation, but we're likely to speculate.

Well, Todd, do you want to take this one?

I mean I'll say just a few words. So we've been looking at this quite a bit as we think about our capsid programs that are moving forward as well as the shuttle stuff. And then the changes that have been reported, they're not huge. They're not massive changes. We know that different nonhuman primates have slightly different levels of ALPL expression. We can show kind of a consistent level of delivery through them. When we look at different ages of monkeys, we can show a pretty consistent level of delivery. And so we don't see that as necessarily a problem because we know that we can get into younger animals quite well. And if we look, say, a mice with some amount of increase maybe 1.5 to twofold levels of difference. But overall, we don't think it's going to be a challenge. There might be a little opportunity there that you mentioned, Pete. But overall, I think it's just more of a good thing that we're able to predict. We think we're going to be able to predict what doses we're going to need to go into humans with and be able to trust that.

Okay. Two more questions, quick ones, I promise. What would be the next key milestone or what are the next set of key milestones for the NeuroShuttle platform over the next two years? And when can we expect to hear additional receptor candidates?

So one of the things that we hope to be sharing more and more data as we continue. So things like nonhuman primate data. We mentioned that we had a program, but we haven't disclosed what that program is. And so a milestone for us would be to share the program and the status of that program and really building the key data sets that we are in the process of doing and sharing those at forums like this and at conferences.

Have you shared when we should expect to hear about the first candidate?

No.

Okay. I'm impressed. All right. So how should we view the NeuroShuttle program in terms of potential BD activity? Are you thinking about companies bringing assets to you and sort of coupling it to technology? Or are you thinking about licensing now specific candidates? How should we be thinking about this platform and BD activity?

So Pete, that's a great question. And I think we can point to what we've done on the capsid side as our model for this because we've had -- we have examples, I think, of most, if not all, of what you just described in the casid side. So we've got partnerships with Novartis, with Neurocrine, with Alexion on the capsids. We have programs where we're collaborating directly with a partner like we're working on the programs together. And in a sense, there's a payload and a delivery mechanism or a delivery capsid is very similar. And that's enabled us to bring in over $400 million of non-diluted revenue over the past four or five years. So we've been pretty successful with that on the capsid side. And I think we look to be doing the same sorts of things. So with Novartis, they licensed our capsids for their SMA program. They have Zolgensma already in the clinic. So they're looking at it for perhaps follow-ons for that. And in that case, that's just a license where they're taking the capsid and doing their own payload work. We have Huntington's program at Novartis, we're working closely together. We've got the FA program moving into the clinic with Neurocrine, where we've collaborated on that. So I really think our CEO likes to say we're open for business in all reasonable discussions. But we're hoping to do all of those things much as we've done with the capsid platform as the NeuroShuttle matures.

All right. Any closing remarks I'd like to make or highlight something?

I think that just the excitement of us this around this, the novelty of the receptors that we're finding and the first one through showing this very differentiated PK profile is really exciting on the NeuroShuttle side. On the capsid side, we'll be seeing what those do in the clinic this year, both on FA and tau knockdown. We have these two programs on tau, One on gene therapy and one on an antibody that we'll be making substantial progress in hitting inflection points this year. And so really, we've got a lot going on. And I think in about a month or so, maybe we've got a deep dive with you, Pete specifically on the capsI'd that will be exciting to talk about.

We'll be excited. I'm looking forward to seeing some proof-of-concept human data for the capsid. Thank you very much, Todd. Thank you very much, Mihalis. I appreciate your time, and thank you to the audience for allowing me to keep you wait.

Thanks, Pete.

Thanks, Pete.