Hello, everyone, welcome to Oppenheimer's 36th Annual Life Science Conference. I'm Jay Olson, one of the biotech analysts here at Oppenheimer, and it's a pleasure to welcome you to our discussion with Voyager Therapeutics. It's an honor to introduce the legendary Al Sandrock, CEO of Voyager. With that, I wanna thank you so much for making time for us here today, Al. It's always a pleasure to catch up with you. I'll turn it over to you in case you'd like to make some, I guess, opening remarks for anyone in the audience who's not completely familiar with Voyager and maybe bringing us up to speed on that amazing letter that you wrote to shareholders to start the year.
Great. Thanks, thanks, Jay. Thanks for inviting me to your conference. It's a pleasure to talk to you, as always. Let me start with that letter. You know, we sent out a letter at the beginning of the year just to summarize what we think is a pretty important year for us coming up. There's basically 3 pillars of value. The first is centered around a particular tau target, which is tau. We have 2 shots on the tau target that I'd like to talk about. 2nd is gene therapy, that this is the year we're gonna put 2 different gene therapy assets into the clinic utilizing our novel blood-brain barrier-penetrant capsids. One is a Neurocrine partner program. One is a wholly owned Voyager program.
Finally, the 3rd pillar would be our emerging NeuroShuttle platform, where we're taking receptors that we know can deliver capsids into the brain and making shuttles that will bring all sorts of other drugs into the brain. That's what we said in the letter, and it was pretty well-received, Jay, 'cause when we went to JPM, almost everybody I talked to had already read it, so they kinda knew our story. By the way, I would also add that, you know, we're well capitalized, and that we have cash into 2028. With that, I'll let you ask the questions.
Okay, absolutely. Thanks for that setup, and I totally agree with you. Your letter was very well-received, and I think it made its way around the investor community. Even those who may not be shareholders of Voyager were very interested in and fascinated by that letter, and yeah, really some really visionary comments, not just about Voyager, but about the industry as a whole. Thank you for setting that up. Yeah, maybe just to start with the first pillar, if we could please dive into your anti-tau antibody program, VY7523. There was a lot of discussion following UCB's bepranemab phase II results at CTAD last year, and then more recently, J&J discontinuing posdinemab. I guess, how should we think of tau as a target?
We get a lot of questions from investors about external validation of tau as a target for Alzheimer's, and also just, I guess, the ideal modality for targeting tau.
Yeah, no, I've been thinking about tau for a long time. You know, it turns out that I think this year we will learn from third-party data a lot more about tau. You just mentioned one of them, J&J. They will be presenting data at an upcoming meeting in March, the ADPD meeting. We'll also hear from Biogen with their antisense oligonucleotide and perhaps some other compounds, because, you know, many companies are targeting tau. I think the reason for that is that, first of all, the accumulation of tau, or what I should say, is the spread of tau in the brain, happens in a very stereotypic manner in Alzheimer's disease. In fact, the staging of Alzheimer's disease is based on how far tau has spread in the brain.
This characteristic spread, we think, is responsible for the neurodegeneration in the brain. In fact, there are some examples of patients who have a lot of amyloid in their brain, but who are not demented and don't get demented until the tau starts to spread. It's very clear that tau spread in the brain, the misfolded, aggregated tau that's spreading in the brain, the pathological tau that's spreading in the brain, is actually better correlated with clinical decline than actually even amyloid. You know, Jay, the funny thing is, back in the day when we were developing anti-amyloid antibodies, and I was in my former company, I got a lot of criticism for developing anti-amyloid even when there were a lot of failures.
You'll remember the first few anti-amyloid antibodies failed, and of course, it turns out anti-amyloid antibodies work. You know, already back then, I was already thinking that we should be also looking at tau. Lo and behold, the anti-amyloid antibodies do work, but there's still a big unmet need because they certainly aren't a cure, and we'd love to get better treatments. Anyway, we think it's a pretty important target, and we'll learn more, like I said, this year. You touched on bepranemab. You know, that was a UCB antibody that for the first time actually showed. It was kind of a mixed story in the sense that it's the first time that it showed an effect on the spread of tau.
As I said, that's what we're trying to impede, is the spread of pathological tau in the brain. The bepranemab drug was the first one that showed an effect on that. 2 earlier antibodies, N-terminal directed antibodies, had failed to block even the spread, but that one did, and it actually also had an effect on ADAS-Cog, which is a tried and true clinical measure that people have been using for decades, actually. That actually hit on the as a secondary endpoint, but the trouble was that the primary endpoint, CDR-SB, was negative. That's why I say it's a mixed picture. Now we're seeing at least a drug that works in terms of the biology of blocking the spread of pathological tau.
Now, what we don't know is, how much biology do you need to get a clinically meaningful effect? We see, as I said earlier, a hint of an effect, but maybe we need to block it better to see a more consistent effect, shall I say, across clinical measures. I'm looking forward to seeing the data from J&J, which is just a couple of weeks away now at the ADPD meeting. What I wanna know is, does it block the spread as well as bepranemab, or better, or worse? If it doesn't block the spread at all, well, then it's not gonna work, right?
If it blocks the spread of pathological tau and does it better than bepranemab, and it still doesn't have a clear-cut clinical effect, now I'd say we'd be worried that the antibody approach may not be the right approach. That's why the 2nd shot on goal that I mentioned earlier, the VY-1706 gene therapy also is of interest, and Biogen's BIIB080 is the best test of that approach. In that situation, instead of using an antibody, which will only bind extracellular forms of tau, the knockdown approach will decrease the expression of all forms of tau throughout the brain. Biogen's phase I data showed an inkling of efficacy, actually, but the trouble was there was no placebo group or control group.
Boy, it sure looked like it had a interesting clinical effect, potentially a very large effect when you compare it to natural history data. Also, it had a dramatic effect on the pathological tau by tau PET imaging. It actually decreased. It didn't just block the, slow the, impede the spread of pathological tau, it actually decreased the pathological tau in all areas of the brain that they looked at. Quite a dramatically different effect, and that's why I think that's also a very important test for tau as a target.
Okay, understood. Thank you so much for that, super comprehensive overview and covering the different approaches to targeting tau. One of the things that comes up sometimes with investors is how to going back to the antibody approach, thinking about how to sequence treatment with tau antibodies versus anti-Aβ antibodies. We've always kind of been fascinated by the donanemab studies, where they stratified patients by tau, and I know they got kind of a broad label for treatment of Alzheimer's, regardless of tau levels, as long as they were amyloid positive. The data didn't look all that great in high tau patients. How should we think about the sequencing of anti-amyloid versus anti-tau antibodies in Alzheimer's patients?
Yeah, it's a very interesting question. You know, they had a pretty rich discussion at the FDA advisory committee meeting, you'll recall on this. The effect of the donanemab was much better in the low tau burden group than in the high tau burden group. I think the committee was starting to think, "Well, should we maybe only approve it for the low tau?" Or contemplating that possibility. The problem is, then you'd require tau PET imaging on everybody, and there was no approved tau PET ligand. It would've been a morass. It's hard enough to do the diagnostics now to get people on anti-amyloid. I think that's why they avoided that.
Nevertheless, that does suggest that you wanna catch the tau before the burden in the brain is too high, right? Before the burden of pathological tau is too high. So the way I've been thinking about it is that we have some early Alzheimer's trials, even before people have symptoms, with the anti-amyloids. You know, both the Biogen anti-Aβ drug, lecanemab, and the Lilly drug are in this, if you will, pre-symptomatic stage. Certainly people before they even have mild cognitive impairment. I'm looking forward to those results. My bet is that the anti-amyloid antibodies will work better in that group. I think that the tau antibodies probably also will work better before the tau has spread very far.
BIIB080, though, shows that you can actually decrease the tau even after it's already been spread in the brain. It could be that you wanna start with anti-amyloid really early, perhaps even before symptoms, the way we treat cholesterol as a biomarker before we have the first heart attack or any symptoms. Then maybe you want to treat with tau if the anti-amyloids don't prevent Alzheimer's disease from appearing. Once Alzheimer's disease starts to appear, you probably need to then use the tau approaches. That's how I've been thinking about it. We'll see if that's true.
Okay, excellent. Appreciate those insights. Yeah, we're definitely looking forward to those 2 studies reading out. Also looking forward to, as you mentioned, the detailed results from J&J's Phase 2b AUTONOMY study of donanemab coming up at ADPD. I guess, where would you like to focus on that data set, and what kind of read-across would you like investors to take away from that data set to VY7523 when we see the full data for donanemab?
We already know from their press or their announcement that it didn't have an effect on clinical measures, so it didn't work on the clinical outcomes. That's all we know so far. What I'm very interested in knowing is whether or not they actually had a biological effect on the spread of pathological tau. As I said earlier, if it doesn't affect the spread of pathological tau, I, you know, it's not surprising, it's not gonna work in clinically. If, on the other hand, it does work on the spread of tau, then I think it's kind of a referendum on whether the antibody approaches work. If I had to bet, at this point, I would bet that it didn't block the spread of tau very well, because we actually looked at that.
We had an antibody against a very similar epitope at Voyager, which we could have chosen to put into the clinic, and we didn't choose it because it doesn't block the spread of tau very well, that antibodies against that epitope. I think it, as in the case of anti-amyloid, epitope really matters. In the case of Aβ42, N-terminal drugs are the only ones that worked. The mid domain and the C-terminal antibodies didn't work. Here, I'm hoping it's the opposite, that the C-terminal antibodies, like ours against the, our anti-tau, will work, and we already know the N-terminal ones don't. Whether the mid domain or the MTBR antibodies work, we'll see, but I think epitope's gonna end up mattering quite a lot if antibodies work at all.
Okay, excellent. Appreciate those predictions. That's good to know. Great setup.
Well, we'll see if I'm wrong, but I our animal, ours 'cause we have a spreading model in the animal using human Alzheimer's material, and that epitope didn't work as well, so that's what I'm betting. Of course, sometimes animal models don't predict the humans very well, so we'll see.
Okay. All right. I guess with regards to your own clinical progress with 7523, congrats on completing the enrollment in the MAD portion of your study. You presented data at CTAD on the SAD portion. Can you just remind us about how the emerging clinical profile of 7523 looks based on what you've observed so far?
So far, what we've done, what we've looked at is preliminary safety and the pharmacokinetics. In terms of safety, it was very safe, and actually all the anti-tau antibodies, the safety profile looks quite good. That's good. The pharmacokinetics are exactly as we would have predicted for a typical antibody, in the sense that the brain-to-plasma ratio, if you will, or CSF to serum ratio in this case, is 0.3%, which is right in the middle of the 0.1%-0.5% range of typical monoclonal antibodies. Also, the half-life would support the once a month, every 4-week dosing that we're employing in the multiple ascending dose study.
Okay. All right. Sounds good. Since we're anticipating getting a look at the MAD portion of the phase 1 study for VY7523 in the H2 of this year, one question that we get is, will you include tau PET imaging data, and how would you like to set investors' expectations ahead of that data readout later this year?
Yeah, no, in fact, MAD was designed specifically to look at tau PET imaging. It's powered to see an effect on slowing the spread of pathological tau in the brain based on tau PET imaging. In fact, that's really the only biomarker that I would rely on for, Because fluid-based biomarkers are available, but sometimes the fluid-based biomarkers are positive, and the tau PET imaging is negative. To me, it's the spread of pathological tau that we wanna impede. so that's the key endpoint. I hope we're better than the bepranumab, because, as I said, bepranumab gave mixed results. For better than bepranumab, I think that will be pretty exciting.
You know, I've always said we would need a partner going into Phase 3, and Alzheimer's is too large an endeavor for a small company like Voyager. Hopefully we'll find a partner. We also have this shuttle technology, and we've seen how shuttled anti-amyloid antibodies can actually be even better than the unshuttled antibodies. We have the option to make a shuttled anti-tau antibody as well. That's something that we could potentially think about as well.
Okay, those are good options to have. Anything that you can share with us on the baseline characteristics for the patients in the MAD portion of your study? I know you said that tau PET imaging is the biomarker to look at, but are there any other biomarkers that you plan to show that we should pay any attention to?
The population will be early Alzheimer's disease, so people with mild cognitive impairment and early mild dementia. Very typical of the patients enrolled in the other tau trials, actually also the amyloid trials. That's not unusual. It turns out that's a pretty good population to use tau PET imaging to look for impeding the spread. That's why we chose that population. I don't think it'll be unusual when people see the population we chose. Sorry, your second question was what? Oh, the other biomarkers. Yeah. We will look at some fluid-based biomarkers. You know, people, there's all different isoforms of pathological tau. There's a variety to look at.
I would say that those are potentially, I would view, as sort of target engagement type biomarkers because, you know, for example, Biogen published that in their trial with the one of their antibodies that failed to affect any clinical measure or tau PET imaging, it actually had a 40% effect on the most popular, if you will, tau fluid-based biomarker, the MTBR tau. There was a 40% decrease in that fluid biomarker, but there was no effect on tau PET imaging in that study. They actually presented that at last year's ADPD meeting. That told me that, boy, you can't rely on that fluid-based biomarker because there was a clear-cut effect on that, but there was no effect on tau PET imaging.
That's why I say I'm banking more or I'm placing more emphasis on the tau PET imaging.
We will look forward to that. Once you see that data, how do you plan to make a go, no-go decision on VY7523? Once you have that decision, if you do plan to advance into phase 2, can you maybe talk about any initial thoughts on a new study design?
I think we'd have to get a partner. Even phase 2 is, you know... I don't wanna, it's too big for a small company like Voyager to venture too far beyond the multiple ascending dose study on our own. The design, I think, would be early, you know, you know, as you know, the amyloid field is moving to earlier and earlier stages. I think for tau, we could stay in that early MCI and mild dementia patients, 'cause you can go later with tau treatments, I think. I think we'd stay with that. That's what I would do. Of course, we'd have to wait for our partner to tell us what they would wanna do and have a discussion.
That's what I would have a propensity toward studying that population.
Okay. All right. Yeah, we definitely appreciate your partnership strategy. We've always thought that the Biogen-Eisai partnership is kind of a model for sort of not only risk sharing, but resource sharing in development of large-scale Alzheimer's studies and clinical development programs. Definitely appreciate that.
Well, yeah, Jay, I mean, look, Alzheimer's is such a huge opportunity, right? Sharing the upside doesn't bother me, but I would like to share the risk and the cost, right? Sharing the upside makes sense for such a large opportunity, I believe, in particular.
Agreed. I totally agree. Unless there are other comments you'd like to share on the antibody approach, definitely wanna talk about VY1706, your tau silencing program with your innovative TRACER capsid. We're super excited about that. Can you maybe remind us about the preclinical data that you've reported so far? How does the degree of tau mRNA or protein reduction compare to other tau silencing programs? You mentioned BIIB080. How do those 2 compare?
Yeah. BIIB080 was about a 60% reduction, and we're targeting 50%-70%. In fact, that's what we find in our non-human primate studies, that we get that at a pretty low dose, too. The gene therapy that we wanna go to lower doses because, obviously, it has a higher likelihood of being safer, and also it's better on cost of goods, right? Remarkably, at 1.3 × 10¹³ vg/kg, which is about an order of magnitude lower than people typically use for intravenously delivered AAV, we see 50%-70% knockdown, our target knockdown range, which is right whereby BIIB080, which is what BIIB080 achieves. We also, on top of that low dose, we have a capsid that de-targets the liver by 30-fold.
'cause as you know, liver toxicity can be very dangerous for patients. In fact, we had some other companies that had some patients who died from liver failure. We're very conscious about the safety, and we're very conscious about trying to get to that level of reduction that BIIB080 achieved, and hopefully, we'll see some nice treatment effects of that level of reduction later this year when BIIB080, their the data are shared.
Okay. Okay, that makes sense. Yeah, we're definitely looking forward to that as well. Anything you could share with us with regards to your FDA interactions on the NDA filing thus far? I guess, you know, similar question, any initial thoughts on your Phase I study design?
Yeah, well, all I can we did have a productive interaction with FDA about a year ago, actually. You know, they helped us think about the GLP toxicology studies, and we even contemplated some initial trial designs. You know, we expect to file an IND in the Q2 of this year, we expect to start in the clinic in the H2 of this year, first patient in the H2.
We haven't really disclosed the design of the trial yet, but based on what I just said earlier about how much I believe in tau PET imaging and how Biogen's BIIB080 has such a remarkable effect on that you can bet that we're gonna be looking at tau PET imaging very closely with our gene therapy asset.
Okay. All right. Well, since you mentioned BIIB080, definitely want to follow up on that, and the Phase 2 CELIA study, results expected mid-year. I know people ask you this, it's a tricky question. Any predictions on the study outcome? I guess, do you expect anything that you learn from BIIB080 to influence your view on VY1706?
Well, BIIB080, you know, Biogen has already published, actually, the paper actually just came out recently. They have been presenting it for quite a while, but you, they saw a pretty interesting effect on multiple clinical outcome measures. Even though there was no control group in the study itself, when you compare it to natural history studies, or to the placebo group of a similar study in terms of the patients enrolled, it's a pretty dramatic effect, much bigger than the anti-amyloid antibodies. I'm hoping that they reproduce that in this well-controlled trial, randomized controlled trial. you know, the prediction would be that they would have a bigger effect on CDR sum of boxes than any of the anti-amyloid treatments so far, including the shuttled anti-amyloid antibodies.
I think that could be quite exciting for the field.
Okay. All right, that makes perfect sense. I know a lot of people are excited about that. We kind of talked a little bit about this, but maybe if you could highlight some key differences between VY-1706 and BIIB080 in terms of, well, first of all, the route of administration, but also potentially the biodistribution in the CNS.
BIIB0 80 is an intrathecally administered antisense oligonucleotide. It's either gonna be every 3 months or every 6 months, they're testing both, but that's a repeated intrathecal injection. All antisense oligonucleotides injected into the lumbar cistern will produce a gradient, so that the exposure is going to get lower and lower as you go up the neuraxis, up the spinal cord, and still lower as you get up to the brain, and lowest in the deep gray structures. Biogen's already showed that you get enough into the cerebral cortex to get a dramatic effect on tau PET imaging. There is gonna be this gradient. Ours is a 1-time IV drug, and it just produces in the brain the siRNA that will continually be produced.
There is no gradient, 'cause we leverage the vasculature to get broad distribution into the central nervous system. Those are the key differences. Mechanistically, though, they're pretty similar in the sense that they both decrease the expression of all forms of tau.
Okay. All right, also, we did touch upon this a little bit earlier, but can you just talk about where the anti-tau mechanism fits into the future treatment landscape, and who is the ideal patient population for anti-tau therapy?
I think that it's either going to be given to patients who don't respond very well to anti-amyloid antibodies, or it's gonna be maybe combined. In fact, one company is already testing the concept, this combination idea. If the anti-amyloids is working in a particular patient unless it's a cure, you're probably gonna wanna add something that will stop Alzheimer's in its tracks, and I think that's where the tau treatments would come in handy. The other thing with tau, by the way, is that we only talked about Alzheimer's disease. There's a whole host of other tauopathies that we can also think about going after, such as PSP. In fact, there's a lot of data that suggests that traumatic encephalopathy is a tauopathy.
There's a whole host of other tauopathies that we could go after. Very exciting, actually.
Okay, and that's super helpful, and, yeah, a lot of opportunities there. Maybe, just to follow up on one point you made earlier about the anti-amyloid studies in pre-clinical Alzheimer's populations. Assuming those are positive, would you wanna move, a tau approach into the pre-clinical setting?
We might want to, but I'm not sure you need to because tau spreading happens later. Even when you get somebody in the mild cognitive impairment or early Alzheimer's stage, it's still early in terms of the natural history of pathological tau in the brain. You may not need to go that early, but of course, you'd have to go. You know, somebody may wanna do that experiment, 'cause my prediction is if you could go earlier, it would be even more effective, but you may not have to in the case of tau.
Okay. Okay, makes sense. I know we're running out of time, but if I could squeeze in a couple of questions on your NeuroShuttle platform. First, congrats on all the progress that your team has made in characterizing ALPL as a novel target for crossing the blood-brain barrier. You had some early characterization of ALPL versus transferrin receptor shuttles. Can you just highlight the key differentiating features for your ALPL shuttle?
The key thing is that we have a very different pharmacokinetics. Transferrin receptor shuttle drugs have a very short half-life. Unless it's bound to something in the brain, it clears very rapidly. That's because TFR shuttle drugs go to many other tissues and not just the brain. It goes to heart, it goes to muscle, and some people have actually leveraged the fact that it goes to muscle, as you know. The other thing is that there are hematologic adverse events, and you see decreased reticulocyte counts and stuff. ALPL is differentiated in the sense it has a much longer half-life because the distribution in the body is different, and we've seen no effect on reticulocyte count. We're still learning.
It's still early, but I think so far, we see different clear-cut differentiation from transferrin receptor. Yeah, it could be quite an important new shuttle that we could use for a whole host of other diseases. For example, I mentioned tau. You don't want a short half-life if you need to maintain coverage for 24 hours, 7 days a week. Many antagonists, you wanna maintain antagonism, 24/7, and a short half-life is not great for that. You don't wanna have to be injecting it every week either, right? So that's where I think for certain targets, ALPL may be the preferred shuttle.
Okay, that makes sense. Definitely appreciate the potential benefits for different targets. What about therapeutic modalities? Are antibodies the modality that you think works best with the ALPL shuttle, or are there other modalities that you're thinking of?
Yeah. We're looking at antibodies, we're looking at peptides. We're also looking at oligonucleotides. Certainly, you know, our first foray will probably be more protein therapeutics, but we're definitely also looking at oligonucleotides, because, as I said, you know, this intrathecal injection, you know, some of these ASOs have to be injected every month, which is a quite a burden to the healthcare system, not to mention the patients. Also, you get this gradient. If we could shuttle it, we would not have to deal with the gradient. We could have less frequent injections, IV, much more, much more convenient. That would be the goal.
Okay. I apologize, we've gone a little bit over. Maybe one last really quick question. Any comments you could share with us on your partnered programs with Neurocrine on FA or GBA1 gene therapies?
Neurocrine has said that the FA program, using one of our novel TRACER-derived capsids, will also go into the clinic this year. I think they're planning to file an IND this year, and they've indicated they're gonna move into the clinic this year. That's our 2nd gene therapy, if you will, that I already mentioned, shot on goal. They also say the GBA program is progressing. They haven't said when they're gonna enter the clinic yet. They've also stated that a couple of the other programs that we have partnered with them are also progressing.
Okay. All right. Well, thank you for the extra time here. Apologies for going over a little bit. It's always a pleasure catching up with you and, learning from your pearls of wisdom, and congrats on all the progress. Thank you so much, Al.
Thank you, Jay. Thank you. Goodbye.
My pleasure. Thanks, everybody.