I'm Sam Semenkow, one of the Biotech Analysts here at Citi, and it's my pleasure to be hosting Voyager Therapeutics. I'm joined by Al Sandrock, President and CEO, Nate Jorgensen, CFO, Todd Carter, CSO, and Trista Morrison, Chief of Communications. Thank you all for being here today. I'm actually going to turn it over to you, Al. I think you have a couple of slides you want to run through. Can you hear me now? Great. We have some forward-looking statements on this slide. What's the investment rationale for Voyager? We would summarize it on this slide here. We have an exciting pipeline of neural assets. We expect four programs in the clinic in 2026, and one of those programs, VY7523, we expect data in the second half of next year, which is our anti-tau antibody in a fairly decent-sized, multiple ascending dose study where we'll look at the effects of our antibody on the spread of pathological tau in Alzheimer's patients. Among those programs, we have four wholly owned programs in Alzheimer's disease, including the most recent entry, which is APOE. We have programs targeting amyloid as well as t au and APOE. Neuro has been considered risky by the industry, and our strategy to mitigate the risk in neuro is based on several factors. One is we focus on validated targets, targets validated predominantly by human genetics. We take programs forward where we think we can de-risk early in the clinic, where we think we can get human proof of concept early in relatively small, early- stage trials. We're also seeking to find drugs that have the potential to have transformative effects for patients, for high unmet needs. Right now, we are mainly employing AAV gene therapy. Our anti-tau program is an antibody, but the remainder of the programs are AAV gene therapy that are delivered IV across the blood-brain barrier. Today, we're going to show for the first time data from our emerging platform, a shuttle platform designed to optimize delivery of multiple modalities into the CNS. Finally, we believe that to do Neurow el, we have to have great people. We've been able to hire some of the very best people with a lot of experience, a good track record of getting drugs approved. We also work with partners that are very high-quality, and we are very proud of the scientific engagement we have with these partners. All this is based on a nice amount of cash that gets us, the cash runway is into 2028, which doesn't include any of the potential milestones that could total up to $7.4 billion. As I said, I want to focus the rest of this talk on our emerging platform, which is related to non-viral forms of treatment, the so-called shuttle program, and we're calling this the Neuro Shuttle. The reason why we are focusing on this as our additional platform is that we have seen evidence that TfR conjugation can really make some amazing drugs. We have, for example, the anti-amyloid example of gantenerumab. gantenerumab, the unshuttled antibody, had modest efficacy. Only 28% of patients could become amyloid PET negative, and that's even with about a quarter of the patients getting the main side effect, which is ARIA. If you look now at trontinemab, which is a shuttled gantenerumab, we see that now 91% of patients get a very good amyloid removal, and less than 5% of patients get ARIA. Shuttling this antibody transforms it into a much better product, and yet there is room for improvement, we believe. Everybody right now, to my knowledge, or most people are focusing on TfR, which, as the example of trontinemab shows, can be very beneficial. Each shuttle is predicted to have different pharmacokinetics. For example, TfR provides rapid brain uptake, but with fast clearance, and biodistribution is going to be different based on the shuttle. TfR has high expression peripherally, and there's a peripheral sink that affects its distribution. Finally, each shuttle is expected to have its own safety issues related to the endogenous function of the receptor to which the shuttles bind. For example, TfR is a critical regulator of iron homeostasis, and we see evidence of hematologic adverse events. What are the chances that the very first shuttle that we discovered for BBB penetration is the best one and is going to apply for every indication for every target? Seems unlikely. For that reason, we've developed the second platform at Voyager, which leverages our first platform. As you know, Voyager has this TRACER AAV discovery platform where we search for AAVs that, sorry, where we look for novel capsids that penetrate the BBB. This is called TRACER. These capsids bind to certain receptors that the capsids leverage to get them across the BBB. What we do is we search for receptors that these capsids leverage to get into the brain, and then once we've identified the receptor, we make ligands against the receptor to see whether they can perform as shuttles. The first platform, the AAV TRACER platform, actually leads to the Neuro Shuttle platform. We start with receptors that we already know can get AAV across the blood-brain barrier. There is a very high likelihood that they will perform as shuttles as well. These shuttles can now deliver all sorts of modalities into the brain, as seen, for example, with TfRs, including protein therapeutics as well as oligonucleotide and peptide therapeutics. The next slide shows some data that's publicly available on TfR shuttles. On the left is data from Denali, in mice. On the right is data from Aliada, which was acquired by AbbVie, with a TfR shuttle in non-human primates. On the top, we see the PK in the brain. In the gray, you see the control immunoglobulin. If we look at, for example, anti-based one that's been linked to their shuttle, the so-called ATV shuttles, you see that the brain penetration is much better in the orange and the blue relative to the gray, but within seven days, 90% of that brain PK is gone. If you look at the plasma as well, in seven days, you see a big drop in the plasma concentrations as well. That's likely due to the peripheral sink for TfR shuttles. On the right, we see the same thing in non-human primates with the Aliada antibody in non-human primates, similar PK in the brain and serum. The shuttle that we are working on initially, the first discovered receptor, is called ALPL. We are making ligands against ALPL to see if we can shuttle various modalities into the brain. Here we're looking at anti-ALPL versus anti-TfR in mice. In the green is what we're seeing with TfR antibodies. We see a very high initial brain penetrance, and then by seven days, we see most of the TfR antibody has gone from the brain. In the anti-ALPL, we don't see quite the same Cmax, but we see a very nice AUC, and that's so for at least the 21 days. We haven't actually carried the experiment on any beyond that yet, but we see a very nice PK where the antibody concentrations are maintained. On the right, we have the plasma PK, where again, the TfR antibody is rapidly cleared. The anti-ALPL antibody has a much slower clearance from plasma, which reflects the fact that it does not have the same peripheral sink issue that the TfR antibody has. The next slide shows some of the adverse events associated with TfR targeting shuttles. As I said earlier, TfR is very important in iron metabolism. If we look, for example, at the trontinemab data in Cohorts 3 and 4, and you see the doses there, you see that 19.7% or 10.5% in Cohorts 3 and 4 respectively show evidence of anemia from their studies. On the other hand, the next slide shows data in mice that we see no effect on reticulocyte count, and we would not predict that anti-ALPL shuttles would have anemia issues when we get into humans. However, is there a slide missing here? Oh, yeah. I forgot to mention that whereas we don't expect to see anemia in humans with the ALPL shuttles, there may be other safety issues that we have to be careful of. ALPL and its normal role is to, it plays a key role in skeletal and dental mineralization as well as liver function and neurotransmitter synthesis. What we know in humans is that decreased levels of ALPL to below 30% residual activity can result in bone hypomineralization or cardiovascular complications. If we look at the human genetics databases, it does seem that partial loss of function is tolerable, but if we get below 30% loss of function, there may be adverse events related to mineralization. We are obviously looking very carefully at this to identify any potential safety profiles associated with ALPL-based shuttles. This slide shows that shuttles can be quite value-creating. For example, I already talked in the box there about Aliada. AbbVie acquired them for $1.4 billion, and that was for a single asset as well as the platform for a phase I Alzheimer's disease asset. We expect that the shuttles create value by increasing efficacy, by increasing on-target delivery in the brain, improve safety by reducing peripheral exposure, lower COGS by reducing the dose needed, and that ALPL is the first of multiple receptors that we have identified based on the capsids we've discovered that cross the blood-brain barrier. It's become a key priority for Voyager to develop this platform because we're very interested in neurotherapeutics and optimizing delivery for a variety of modalities. My last slide, just to remind you, our pipeline, as I said, we focus on validated targets, and we believe there's high- potential value. Our most advanced program is the anti-tau antibody, as I said, in a multiple ascending dose study scheduled to read out, second half of next year. Next year, we hope to also introduce into the clinic a tau silencing gene therapy, VY1706. We have multiple programs in the research phase, two additional programs targeting Alzheimer's disease, and one targeting amyotrophic lateral sclerosis. I should also point out that we have a number of gene therapy programs that are partnered. Neurocrine expects to file INDs this year for frataxin for Friedreich's ataxia, as well as GBA-1 gene therapy for Gaucher's and Parkinson's disease. We have multiple other programs, both with Neurocrine and Novartis, and one with Alexion. We hope to be announcing soon that we have Voyager Neuro Shuttle programs as well to add to this pipeline. With that, I'll turn it back over to Samantha. Al, thank you very much for walking us through those slides. It seems like you've made a lot of progress on the Neuro Shuttle program. Correct me if I'm wrong, but this is the first preclinical data that you've shared from the program. Maybe you could help me just frame a little bit about the potential here, maybe just starting with ALPL specifically, but also more broadly as you add in more receptors for the program. You know, what can you do with this technology? Yeah, so, you know, in the NeuroS phere, what we've, of late, we have identified a lot of targets that are of high interest to us that can address some of the most important diseases with the highest unmet needs. We know the targets we want to go after. The only way to really get good drugs into the brain against these targets currently is with small molecules. However, there are many targets that are undruggable with small molecules. We have these newer modalities, oligonucleotides, antibodies, peptides, but their delivery into the brain is suboptimal. What we're hoping to do with the shuttles is to get these newer modalities into the brain with a systemic delivery. We think, because many of these disorders affect multiple regions within the brain, for example, intrathecal administration of oligonucleotides, the issue there is that you get a steep gradient. You get very high exposures in the spinal cord where you typically inject in the lumbar cistern, but you get less and less drug exposure as you get up in the brain. You can still get enough, potentially, to get some effects in the cerebral cortex, for example. We're all looking forward to seeing the data from BIIB080, the anti-tau, the tau- lowering, antisense oligonucleotide. We believe that if we can optimize delivery with an IV-delivered shuttle, oligonucleotide or antibody, that we can do much better for patients. When you think about ALPL specifically, are the ideal applications for that receptor a broad delivery to the CNS, or is it targeted to certain cells within the CNS, or how do we think about that? The way I think about it is that, as you saw, you optimize the AUC. You have a nice steady brain PK with very little clearance over 21 days. For example, if you wanted to shuttle in an anti-tau antibody, there we would want to maintain the blocking the spread over time. You may want to use the ALPL shuttle because you can maintain that coverage, if you will, over many weeks. You can certainly then dose less frequently. An anti-tau shuttled anti-tau, if we get to the point where we would want that, could be, for example, where an ALPL shuttle would be optimal. I think it's going to depend on the disease and the target. Is there anything else that makes ALPL an ideal receptor for the shuttle program? I'm going to turn it over to Todd, but you know, for me, as with any of these novel shuttles that we're going to pursue, we start with a receptor that we know can get AAV, which is a very large molecule, into the brain. I think we start with a very promising receptor, and each one's going to have its own safety, distribution, and pharmacokinetics. Hopefully one day we'll be able to tailor the treatment to the indication of the target. Todd? Yeah, thanks, Al. I think what's really interesting is that, using that TRACER technology, we've identified these multiple receptors. As Al mentioned, we've already confirmed they can get something large, AAV, across the blood-brain barrier. What we're seeing is now that we've taken that, and in the context of ALPL, we are able to get other modalities across the blood-brain barrier. We've made that leap to blood-brain barrier penetrant capsids, taken that, and converted that into other modalities. The patterns that we're seeing are quite different between transferrin receptor as a modality or as a target for that BBB penetrance and ALPL. It may be that, again, I think Al pointed this out, that different diseases might require different receptors or different patterns. For the case of ALPL, where transferrin has this initial high delivery and then it rapidly clears, ALPL seems to go in and then maintain that level of delivery for quite some time. That's probably related to the distribution of the receptor across the periphery, where if the receptor is present in a wide variety of cells in the system, that results in the quick uptake and the rapid clearance, whereas ALPL doesn't have that. It's more limited in its expression. That results, we think, and the data seem to show so, in this elevated levels and longer, much longer half-life. It's very interesting to consider that different receptors are going to have these different profiles. With a variety of these receptors, we may be able to really tune into the right ones for the right diseases. Got it. That's very interesting. For the PK data specifically, I noticed that there wasn't much of a Cmax for ALPL. Is that ideal? Could you imagine, you know, an indication where you don't want a large Cmax and maybe conversely ones where a Cmax would be a benefit? Yeah, I mean, there are some diseases where you would want a strong Cmax. When I think of a drug that has a high Cmax and a rapid clearance, you're kind of basically pounding that target repeatedly over time. In some situations, that may not be optimal. In certain situations, it may be actually adequate and desirable. For example, trontinemab works pretty well, right? The anti-amyloids, once they get into the brain and they bind to amyloid plaques, they probably persist. In fact, we know they persist in the brain, because they bind to aggregated forms of A-beta. Also, they trigger microglial activation, and it's the phagocytosis of amyloid that makes it effective. There, having a high Cmax may actually be adequate or optimal for all we know. We don't know yet. On the other hand, having a steady, for many drugs, for example, if you imagine an antagonist against a receptor, you want to maintain coverage over the entire 24-hour period, right? For most of the drugs that I can think of, you may be wanting to optimize the AUC rather than the Cmax. Again, it's going to depend on the target and the situation. Right. That makes sense. You mentioned a couple of safety concerns, potential safety concerns for ALPL. Is there any concern for delivering the cargo to a non-target cell as well? It may not be the delivering the cargo that's as much of a concern as interfering with the function of ALPL itself. That's something we're going to have to learn as we go forward here. Todd, any other thoughts? Yeah, that's something that we're closely evaluating. We're doing some studies to look at that. So far, there are no huge red flags, and we're happy with our data to date. In terms of particular cells, Al mentioned the particular tissues and regions that we'll need to look at carefully. We're in the process of doing that, and we hope to be able to share some data in the future. Got it. The idea is to not have a sync going to some of those other locations rather than across the blood-brain barrier. Both for safety and for that clearance issue that we've been talking about. If you're looking forward, how far from the clinic is a potential shuttle program, or even nearer term, how soon could you maybe start to disclose discovery or development candidates for the shuttle program? Right now, this is fairly early stage research. We're excited about what we're seeing, so we want to share it, but we're not talking about timelines quite yet. Hopefully, in the near future, we'll be able to do that. Is there a potential to consider strategic partnerships at this stage, or is it also still a little bit early for that? Our history is that we love strategic partnerships, and we have a strong track record of doing a lot of those. We love working with companies that have strong scientific expertise in an area. I'd love to be able to combine our expertise in shuttles with companies that have other expertise, for example, in certain modalities. Got it. I look forward to that. Al, before I switch topics, is there anything else on the Neuro Shuttle program that you wanted to say before we move on? Only that the ALPL is the first of multiple receptors we've identified through the TRACER platform. One day we hope to have a variety of shuttles fit for purpose. What are the chances that the very first shuttle ever discovered, TfR, is going to be the ideal shuttle for every single neuro program? Unlikely, right? I think having a variety of different shuttles will one day benefit patients greatly. Is identifying and characterizing additional receptors, so you've done it once so far that we know of publicly, is the process very plug and play, or is it more iterative? Do you want to talk about the pandemic? We have identified multiple receptors through our programs, through our capsids. I like to think of them as they're capsid-validated receptors, in that we've shown they can get something, namely a large capsid, across the blood-brain barrier. Our work always includes work in not only rodents, but non-human primates, so we can show that we can do that. The other advantage of having that receptor identified is that we can show that these things work against the human form of the receptor too. That really greatly increases the chance of success in the clinic. I mentioned we have a number of these that we've identified, and ALPL is just the first of these that we've shown that we can make that sort of reverse translation from capsid to other modalities. We have every expectation that we'll be able to do that with some of our other ones. Your question implies that it's very important for us to identify the receptors, even if we weren't doing a shuttle program, just to be sure that our AAVs can work in humans. To know what the receptor is, knowing that there's a human homolog is very, very helpful. We have every incentive. We've always had this strong incentive to identify receptors. It's not quite plug and play. It takes a lot of hard work, but we've been able to identify several other receptors. Got it. That makes sense. I'm sure there's a lot that goes into it, a lot of hard work. Let's switch gears a little bit and talk about your current pipeline. I'm wondering, Al, if you could just speak to your overall strategy for Alzheimer's disease. You have assets across a range of targets and modalities. I'm wondering how you envision each of these approaches to be utilized in a real-world setting, potentially. I think we're at the dawn of disease-modifying treatments for Alzheimer's disease. The three anti-amyloid antibodies that have been approved certainly indicate that. We know that that's not going to be the solution for every patient. On average, there's about a 30% effect on slowing of decline. We need better therapies, or we need therapies that can be added to the anti-amyloid treatments. That's why we have two programs targeting tau. We think that the spread of tau is a very important piece of pathology that we have to address. In fact, one would argue that if it weren't for the fact that the anti-amyloids don't also impede the spread of tau, they might not be quite as effective. Once the tau starts to spread, I believe it's going to be very hard to affect the spread of tau unless you target tau itself. Amyloid probably triggers the spread of tau, but once the tau starts to spread on its own, we may need treatments that target tau. APOE is a very interesting target from our point of view because we know that there are many patients who are APOE4 carriers and that if you're homozygous for APOE4, your likelihood of getting Alzheimer's disease is enormous. It's almost like a single gene mutation. We met one patient who visited us. She showed a picture of her family, and more than half of her family members were already affected by Alzheimer's or are already showing signs of cognitive decline. These patients typically present earlier, and the progression can be more rapid. Why not then prevent the onset of Alzheimer's? If the likelihood is enormous to get Alzheimer's, why wait till you have first symptoms? We're very excited by the possibility that if you're an APOE4, certainly if you're homozygous, why not decrease the expression of the harmful allele, E4, and increase the expression of the, what's often called protective allele, E2? I think the bottom line is we think multiple approaches are going to be needed to properly address this disease. Got it. Okay. Your first asset in the clinic, you've talked about it a little bit in our conversation already, but for VY7523, the tau antibody, next year, the second half, 2026, is when you've guided for the key tau PET data. Maybe you could just talk about what the expectations of what we should see there are. Yeah, I think that what I'm hoping to see is what we see in the animal model that we employed, which is that it blocks the spread of pathological tau. This is an interesting animal model where you take a mouse that's transgenic for human tau, and you inject pathological forms of human tau into the mouse, and we look at the spread. The two N-terminal antibodies that failed in the clinic in humans failed in that model. UCB's antibody actually worked in that model, and we were saying, if that antibody fails in the clinic, that's going to affect our program. Lo and behold, it actually blocked the spread of tau. I see the UCB data as a positive event for the tau targeting treatments in that it blocked and impeded the spread of tau. It actually also had a positive effect on ADAS-Cog, one of the key cognitive measures in Alzheimer's. It did fail on the primary endpoint of CDR, some of the boxes. What I'm hoping is that we'll see not only that we blocked the spread of tau, our antibody is a little bit more robust. That's why we chose the epitope that we chose, than the UCB antibody. It also is specific for pathological forms of tau. The UCB antibody is not specific for pathological forms of tau. That turned out to be pretty important in the amyloid story, as you'll recall, antibodies that don't recognize the pathological forms of amyloid don't work as well as the ones that do. I'm hoping that we'll not only see a blocking of the spread of tau, but hopefully we'll see a clear-cut effect on the clinical outcomes. Our trial is not powered for seeing an effect on the clinical outcomes. The main thing we're hoping to see is an effect on tau spread. There are other companies that have much larger trials that I think will maybe shed more light on the connection between blocking the spread of pathological tau biologically and the clinical effect. Got it. Okay. Hypothetically, is it possible to see a Neuro Shuttle version of VY7523 in the future, or is that not something that's currently in the plan? It's not in our plan right now, but if we see proof of concept with VY7523, then that's certainly an option. As I indicated, the PK characteristics of the ALPL shuttle would potentially make it ideal for an anti-tau program. Interesting. That hasn't started yet. Right. Perhaps we'll see that disclosure one day in the future. For your APOE program, you mentioned a little bit about, you know, using it commercially. I wonder if you could just talk a little bit about how you designed that asset. I believe you've guided to some initial preclinical data this year. I'm curious if you could share a little bit about what to expect in that disclosure as well. Yeah, I'll start it. I'm going to ask Todd to jump in, but the concept was we wanted to maintain levels, overall levels of APOE, decrease the harmful allele, E4, and increase the protective allele, E2. I believe we've actually achieved that, Todd. This approach is a gene therapy approach with a one-time dose or IV TRACER-based capsid we can deliver broadly throughout the brain. Al mentioned earlier the huge genetic component of APOE and particularly APOE4 alleles in contributing to Alzheimer's disease. What we're trying to do is convert folks in a sense from APOE4 carriers to the more protective APOE2. To do that, we need to reduce, knock down APOE4 in those carriers and then replace it with an expression of APOE2. What we've been doing is designing a vector and payload that can do just that. The trick is to knock down the APOE4 to a substantial level while delivering a physiologically relevant amount of APOE2. We don't want to overexpress, nor do we want to underexpress. We need to design those vectors quite carefully. The data that we're looking forward to showing later in the year at a scientific conference is data that we expect to show just that, that we can deliver the appropriate amount of APOE2 while knocking down APOE4 to a very substantial level. The other component is that our TRACER capsids, and we've shown these data before, is astrocytes are the primary site of expression of APOE in the brain. We know we can deliver to 90%+ of astrocytes across multiple different brain regions in non-human primates and rodents as well. This is a program we're quite excited about. Yeah, looking forward to seeing that data. Maybe a big picture question. What can we just expect from Voyager as a whole? Let's talk over the near term, call it next 12 months. I think we've talked a little bit about this, but maybe just reiterate. What I'm more interested in is what's the longer- term picture of what Voyager will look like with all of your current pipeline programs, the potential of the Neuro Shuttle, your ongoing partnerships. What can we look forward to that future looking like? We want to be a multimodality neurotherapeutics company, leveraging highly validated targets, often by human genetics, leveraging the use of biomarkers that, for example, PET imaging I've been talking about, but also blood-based biomarkers like neurofilament to get rapid de-risking in the clinic, all done by great people that we've been able to hire. We are a multimodality neurotherapeutics company, where, as you noticed, we believe delivery is something that needs to be optimized to do neurotherapeutics properly. We have optimized delivery, we believe, for AAV capsids so that we can do gene therapy better. I'm hoping that we can then leverage the receptors discovered through the TRACER p latform to bring in all the other modalities that currently are not optimally delivered into the brain, but they do address some of the undruggable targets, undruggable based on small molecules, but key targets that we would like to drug with these other modalities. Got it. Nate, you want to add anything to that? I think you're right. The only other comment I'll make on that is that I'd like to talk about how I have a Ph.D. in neuroscience. One of the reasons why I joined Voyager is because they're doing just what Al described. They're taking neuro and they're de-risking it to biomarkers and clinical development, making it a risk profile more like a rare disease company where they take those approaches. I think it's going to be a very exciting future in the next few years. Got it. Nate, just one more for you. I've left you alone for the whole session. Cash runway, if you could remind us. Yeah, as of the second quarter, we have $262 million, which gets us into 2028. Last time I checked, we're in 2025, so that's a couple of years of cash runway. That's good, which will get us through some really nice clinical readouts that we talked about earlier, namely VY7523 or tau PET second half of next year. Of course, three gene therapy programs are going to be in the clinic next year. It's going to be very exciting with that and in terms of the third-party readouts as well. All right. I think we're at time. Thank you very much for being here. I think this was a great discussion. Thank you, Samantha.