Hello, everyone. I'm Jay Olson, one of the biotech analysts at Oppenheimer, and it's a pleasure to welcome you to Oppenheimer's 35th Annual Healthcare Conference and our discussion with Voyager Therapeutics, a company that's doing exciting work in neurological diseases with high unmet medical needs. It's a pleasure to introduce CEO Al Sandrock and his team, Nate Jorgensen, CFO, Todd Carter, CSO, and Toby Ferguson, CMO. We've got the whole team here on an important day with breaking news. Al, thank you so much for joining us, and I'll turn it over to you.
Thanks, Jay. You undoubtedly saw our announcement this morning that we're assessing alternate payloads for our SOD1 ALS program, and thus we no longer expect to file an IND in the middle part of this year, as we had hoped. We believe the problem was the payload, not the capsid. I'd like to share some slides with you that will explain why we think that way. I want to also reiterate that we remain quite excited about our capsids and that all the other programs remain on track. Next slide, please. Of course, I'll be making some forward-looking statements here. This is sort of the summary, if you will, that we had an asset called VY9323, which was a gene therapy for SOD1 ALS.
That was basically a miRNA payload designed to decrease the expression of the SOD1 gene leveraging one of our IV-delivered second-gen TRACER-derived capsids. We tested four doses in a GLP tox study, a six-month study, and we're showing some data today from the three-month interim analysis. The long and the short of it is that we didn't really see a therapeutic window. At none of the doses did we see any signs of liver or systemic toxicity, and we had no reason to worry about immunogenicity. At the higher doses, we had a very nice knockdown of SOD1 mRNA, approximately 60% across the central nervous system. However, in those high doses, we saw late-onset neurotoxicity manifested by tremors, as well as a delayed increase in neurofilament light, NfL, and we saw evidence of neuronal degeneration by histopathology.
At the lower doses, we did not see these adverse events, but we also did not lower SOD1 mRNA to the levels that we wanted them to be lowered to. Basically, we had no therapeutic window. What we are doing is we are going to delay the program. We are not going to file the IND this year as planned, and we are looking for new payloads, better payloads. We believe that this is due to an off-target payload activity, and we are not going to advance VY9323 while we assess for alternate payloads. What this results in is an extension of our cash runway into mid-2027. Next slide, please. I am just going to share some data slides. On the left here, what I am showing are the four dose groups, and the doses range from 5E12 vg/kg to the highest dose of 1E14 vg/kg.
As I've always said, you know, if we have to go up to as high as E14 vg/kg, we think that's not great. I mean, because those are the doses where people started to have seen systemic toxicity with other capsids. We wanted to stay below E14 vg/kg, but we assessed up to that highest dose. Here we're showing liver function tests, and we saw minimal elevations in the early time points, as you might expect. In fact, on the right, we have, by comparison, an AAV9 SMN1 gene therapy at 1.1 E14 vg/kg, and you see much higher LFT elevations, if anything. We were happy with the fact that our capsids, it looks like they de-targeted the liver as we had hoped. Next slide, please.
What we saw at the two highest doses, though, was evidence of neurodegeneration, and this is some of the histopathology that I mentioned earlier. If we look on the right here, we see pictures of the white matter up in the brain and the cerebral hemispheres. We also saw evidence of neurodegeneration, nerve fiber degeneration in the spinal cord. Here we're looking in the white matter of the spinal cord. We also did see some increased cellularity in the dorsal root ganglion. This was at the two highest doses, and this is not what we expected to see. Next slide.
Not only did the animals manifest with tremors, and not only did we see histopathology changes, as I just noted, but we saw, when we looked longitudinally at plasma neurofilament, we saw first a little bump at about one week after the injection of the IV gene therapy, a minor bump and much lower than has been seen with other programs. We were delighted to see that. With a delay, we see now a peak at day 56 and it still remains high at day 92. We saw not only the emergence of adverse events, the tremors, as I mentioned, but we saw a significant increase in neurofilament. That delayed increase is sort of what you would expect from a payload problem because that's when the payload is being expressed. The capsid is cleared from the blood by day four.
Capsid problems are typically manifest in the early time points. That is one reason why I believe it was the payload and not the capsid. It is the delayed time course. The other main reason is that we tested the exact same capsid with four other payloads. Next slide. Here we are looking at the same exact capsid with two different promoters and two different miRNA payloads. Although we do see that little bump right after the capsid is injected within the first week or so in the neurofilament light, what we do not see is that delayed increase. We have, for example, doses ranging from 1.3 E13 vg/kg up to 7.1 E13 vg/kg. We do not see an increase in NFL at those later time points as we saw with the SOD1 program.
Same exact capsid, four different constructs, four different payloads, and we don't see that late rise in NFL. That is the second reason why we think that it's the payload and not the capsid, is that when you switch out the payloads, we don't see it with the same capsid. Actually, these data come from our tau knockdown gene therapy program. In fact, VY1706, next slide, we're very excited about because even at doses as low as 1.3 E13 vg/kg, we see substantial knockdown of tau expression. For example, you see those pictures on the lower right, those photomicrographs, we see quite a nice reduction in the expression of tau in neurons, which is the key cell of interest for tau.
If we quantify images like that on the left side, just directly to the left of that image, you see the in situ hybridization data, and we see anywhere from 75%-80% knockdown across brain regions by in situ hybridization in the tau mRNA. Here we have a very nice therapeutic window. By the way, additional data on this VY1706 will be presented at the AD/PD meeting in a couple of months. Next slide. This is our updated pipeline slide, if you will. The major change here is that we moved that SOD1 silencing gene therapy bar to the left. It is back in research as we search for alternate payloads. Everything else basically stays the same. All those other programs remain as they were, and they're on track.
Just to remind you that Neurocrine has two programs, one for FA, Friedreich's ataxia, and one is a GBA1 gene therapy for Parkinson's disease and Gaucher's disease. Both of these INDs are planned for this year. Those will be using the TRACER-derived capsids. Next slide. I think this is my last slide. As I said earlier, this takes our cash runway now out to mid-2027, which allows for multiple potential catalysts to occur. First, there's going to be a lot of data coming from anti-tau antibodies. As you know, Jay, our lead program is an anti-tau antibody, not a gene therapy against the C terminal. We have data readouts from Merck, Johnson & Johnson. In terms of knockdown approaches, Novartis and Biogen have phase 1b and phase 2 data planned in this time frame.
Just to remind you that our 1706, the data I just showed you, that IND is planned for 2026 as well. Finally, as I said earlier, the two Neurocrine programs using TRACER-derived capsids, the INDs are planned for 2025. I think this is my last slide. Yeah. Okay. Just to close, this is something I think you've seen before. The investment rationale remains the same. It's our pipeline, our platforms, and our partnerships. The only change that we made to this slide was that we now have cash into mid-2027. Everything else on this slide remains as we showed it in J.P. Morgan a month ago.
Okay. Great. Thank you. I really appreciate your providing that timely update. That's super helpful. Maybe just a couple of follow-up questions. First of all, it's certainly reassuring to see that the capsid is not the driver of the late-onset neurotoxicity, and you have not observed that in any other settings beyond 9323. Can you maybe talk about the factors that could have contributed to the off-target activity that led to the histopathology changes?
Sure. There are two leading theories. One is what we call over-harnessing, which has been observed and reported on previously. This is where you express so much siRNA or miRNA that you overwhelm the cell's sort of capacity to deal with siRNAs in general. That is called over-harnessing, and that is one possibility. The other possibility is that the SOD1 siRNA was not quite as specific for SOD1 as we would have liked, and that we had knockdown of other genes that may have caused the neurotoxicity. We are actually doing some studies now to try to distinguish between those two possibilities to learn as much as we can from this program. Todd, do you have anything to add?
I think you've captured it very well.
Thanks, Todd.
Okay. That's helpful. Maybe can you just talk about the process of identifying an alternate SOD1 targeting payload for your TRACER capsid delivery? Also, are you pursuing any other targets beyond SOD1 for ALS?
Yeah. Look, we have a high bar for the payload that we want to find. First of all, it has to come pretty quickly because any delay is not acceptable. Optimally, it would be a program that's either already in the clinic or near the clinic. Optimally, it would be a program that has already passed GLP tox studies. We need to see that it can knock down SOD1 quite substantially, at least more than 50%. Toby, do you have anything to add to that? Jay, in summary, we're looking outside, right? Toby?
Al, I think you've nicely highlighted, I'll just sort of reemphasize the idea that we like the biologic aspects of SOD1 ALS in terms of the biomarkers and there remains a need in the population. Al's right to highlight, we need a high bar of an asset that really is through tox and could get into the clinic quickly.
Okay. Maybe I could just add one other aspect to that is that because of the profile we saw from the capsid, we remain quite excited about the capsid's ability to deliver to where we need to go. It really is about finding that optimal payload. The capsid itself is really exciting.
Okay. That's super helpful. Al, I know you said you wanted to move quickly. Can you comment on the timing for finding another siRNA payload?
We'd like to get to an IND by next year. That's why we kind of want next year at the latest. This year would be better, but it's unlikely that it could happen this year. We'd have to do a GLP tox, right, combined with our capsid. Yeah, and it's got to be past GLP tox. If the IND is expected past next year, I don't think we're going to be too interested.
Okay. Got it. Makes sense. All right. Unless our audience has other follow-up questions, which you are welcome to send to me, I thought maybe we could talk about your anti-tau antibody, VY7523. That's great news that you've completed enrollment in the SAD part of the phase one study. There was a lot of discussion last year following UCB's bepranemab phase two sort of mixed results at CTAD. Can you just share your latest thoughts on bepranemab and whether those phase two results support the anti-tau antibody approach to treating Alzheimer's?
I will just say that we see the cup half full rather than half empty. Maybe Toby could. Toby's been really thinking hard about this program as he works on our program. Toby?
Thanks, Al. Jay, it's a great question. I think fundamentally, I certainly agree with Al on the glass half full. That data, I think at the high level, I make two points. One, in the entire population, we saw for the first time a tau antibody that decreased the accumulation of pathologic tau as measured by tau PET, particularly in the temporal lobe, but in other regions of the brain as well. That's a really important first step for the field. In that broad population as well, they had potential clinical signal of ADAS-Cog, also really important. They highlighted the potential importance of APOE and low tau in terms of if you looked at a number of clinical signals in the no APOE, low tau population.
I think we're working through that to try to understand sort of the sweet spot for population where you potentially can differentiate based on population. I think the final comment I would make is, fundamentally, we think there's clear evidence of biologic and potentially clinical activity in this study. We'll take some lessons from that study. We're really quite excited about our antibody in particular because it preferentially binds pathologic Tau, which is a differentiation point from the UCB molecule as well as its C-terminal epitope. Take those things together. We're quite excited. The program's on track. We expect to have a data readout from the SAD the first half of this year and to initiate the MAD.
Okay. Great. Thank you. That's super helpful. I'm glad you mentioned one of the key differences between 7523 and bepranemab.
Are there other differences between those antibodies that you think could translate into clinical profile differences?
I think the core ones really are the pathologic and epitope we think will matter. I would just highlight that, and Todd can comment as well here, but fundamentally, we chose our antibody based on its ability to impede the spread of tau in a preclinical model in which you inject tau and watch that spread throughout the region of a mouse brain. I think fundamentally, we believe that's an important biologic process. We like that point of differentiation as well. Todd, anything to add there?
I'll add that we did look at multiple antibodies across the tau protein, and not all of them did inhibit the spread, as Toby and Al were describing. We liked what we saw with the antibody that we chose. It was the most robust in our studies, and that led to our selection of it.
Okay. Great. I know you acknowledged this in your slides. There are a number of other anti-tau antibodies with readouts this year. Anything you'll be looking for from Merck's phase one readout or J&J's phase two readout for posdinemab in 2026? Any other updates from bepranemab that we could read across to 7523?
Let me start, and I know Toby will likely want to add some thoughts. The Merck antibody is a C-terminal antibody. In some ways, at least it's in the same region. It's not the exact same epitope, but it's pretty darn close. What we're looking for, I think, in these data is not only confirmation that we inhibit the spread of Tau by Tau PET imaging, but can we see more clarity on the clinical outcome measures? This could come from bepranemab as well. What is the relationship between exposure, so PK, and the effect on Tau PET imaging, the biology? So PK, PD. I think we'll learn more about that from bepranemab. Also, we may learn about that from these other antibodies.
The other thing is then what is the relationship between the degree of Tau inhibition or inhibition of spread and the clinical outcome? Is there a threshold effect? Is there some inhibition of Tau spreading that you need to achieve in order to get even a minimally clinically meaningful effect? What is that relationship between biology and disease? That, by the way, is agnostic to drugs. That relationship should hold for all antibodies, I would think. I would also say, as Toby mentioned, what's the right population to study? Is it earlier in the disease? Is it better if you start with low Tau burden? Is there really an interaction with APOE carrier status as the folks at UCB have suggested? These are all learnings that I think will be very important for us. Toby?
Thanks, Al. I think you've covered most of the key points I was going to raise. I think I was struck by in the UCB conversation, there was a former FDA reviewer who made this explicit point that he reminded us that as we've moved through and looked at a number of different beta amyloid therapies, really once you put together the exposure response data for the field as a whole, it really became clear that beta amyloid reduction had a relationship to a clinical effect. The importance of putting together this data across multiple programs is really quite clear. He made the interesting comment that this data reminded him of the early stages of the beta amyloid programs. I'll just double down on Al's point that really understanding that exposure response is critical.
Okay. Super helpful.
Based on the safety that we've seen from bepranemab, does that lead you to suspect you could potentially dose higher? I guess, what should investors be looking for in the results from the SAD portion of the study in the first half of the year?
Maybe I can start there, Al. I think fundamentally, I think the UCB team did a very nice job of looking at safety in the context of a large 400-plus person study. In particular, they looked at imaging-based abnormalities, potentially ARIA and others. They did not see any. I think that is an unappreciated sort of upside of that data set. In that context, I would say we feel comfortable that we can potentially dose to higher levels in our program, which is another potential point of differentiation, a greater dosing, obviously, hopefully getting greater clearance or impeding of spreading of pathologic Tau. We are excited about that potential opportunity. I think in the context of our SAD study, again, this is a single setting dose study in healthy volunteers.
Here we're looking for safety, for immunogenicity, and initial PK information enabled sort of to predict our exposures and move forward with a multiple ascending dose study in AD patients.
Okay. Super helpful. Okay. Maybe we'll shift gears for a moment to VY1706, your tau silencing program, advancing into IND enabling studies. I guess, first of all, can you just talk about how you would compare the two modalities, antibody versus siRNA? And have we learned anything from BIIB080 that might inform your thinking about the tau silencing approach?
I would say the BIIB080, we've learned a lot from BIIB080. It does look like you could have an even bigger treatment effect. The BIIB080, the comparison had to be to either natural history or to the placebo group of an equivalent trial, and actually an anti-tau trial that Biogen did. We do not have a direct comparison to placebo in the BIIB080 data that we've seen so far. Nevertheless, it looks like the treatment effect size is quite large. The other thing is, it does not look like it's unsafe to decrease the expression of tau. Of course, now we're going to see here in the second half of 2026 that Biogen is conducting a fairly large phase two well-controlled trial. I think we'll learn a lot.
Okay.
Jay, maybe I can add. I think the other key thing for the BIIB080 data that was quite remarkable was through the use of Tau PET, that program clearly showed that you could remove pre-existing pathologic tau. That really emphasized the importance of Tau PET as a biomarker. That is a key lesson for us into clinical development.
Okay. Excellent. That is super helpful. Since you have already presented some promising preclinical data on 1706, are there any additional preclinical data that investors should watch out for this year? Can you just talk about the timeline for initiating a clinical trial?
Todd, do you want to talk about any other data that we may be showing at some of the scientific meetings coming up? Later, Toby could talk about the plans for the clinical trials.
Yeah. Happy to, Al. We can expect to see some new data on the tau knockdown program at our AD/PD, as well as in other programs at ASGCT. In the next couple of months, you can expect to be seeing some additional data coming out.
The program is on track for 90 in 2026, Jay.
Okay.
You should also talk about non-viral delivery. I think that would be interesting for people to hear.
Yeah. As you know, Jay, last year, we talked about identifying a receptor for one of our leading class of capsids. What we're doing now is making ligands against that receptor. It's not TfR. It's actually ALPL, a very different receptor. What we're going to evaluate is, are there advantages of ALPL over a transferrin receptor? Can we leverage that receptor to shuttle protein therapeutics as well as oligonucleotide therapeutics into the brain? That's an emerging second platform. That's why we have platform with the plural platforms, because this would be an additional platform to add to our capsid platform. I think that could be a very exciting area of genetic medicine that Voyager could get into.
Okay. That provides a lot of optionality. Maybe just as you look ahead to the future of treating Alzheimer's disease, can you talk about how the anti-tau approach will fit in with anti-A beta approach? Which patients and where in the treatment paradigm and combinations sequencing? How are you thinking about synergies between these two approaches?
Yeah. It is still early days. I think what we are going to learn is the heterogeneity of treatment response with the anti-amyloid antibodies. What we are already hearing is some people do very well with the anti-amyloid antibodies, and some people do less well. If you get no benefit at all from anti-amyloid antibodies, you would just sequence and just move to a different treatment, such as an anti-Tau. If you are getting partial benefit, that is where you would want to add anti-Tau antibodies. If you are getting a full benefit, if the patients are completely stable in terms of disease progression or maybe even improving, there would be no need to add anything.
I think as we understand how patients respond and learn, maybe even be able to predict who's going to respond well, that will lead to, I think, either combination or sequencing or just switching to an anti-tau. It is very clear that on average, the anti-amyloid antibodies do not—I mean, there's still a high unmet need. We have on average about a 25%-30% effect size, right? There is still a lot of room to improve.
Okay. Great. That's a very helpful framework. I do want to come back to your cash runway. You provided an update today that it goes to mid-2027, and you spoke a little bit about your priorities. Can you just elaborate some more on how you tend to prioritize your capital allocation over the next two and a half years?
Nate?
We announced in Q3 that we have $345 million. That, as you mentioned, gets us into the mid-2027. That is well beyond some interesting clinical catalysts, namely what we talked about today, the tau antibody. There are these competitor readouts from the Merck tau antibody and the J&J antibody. Neurocrine, filing two INDs this year, is what they guided to. That is their guidance about when the data is going to come out, but that should not be too far beyond that. Of course, we do not spend any money on those programs. There is very limited money. That is Neurocrine. The final thing I will mention in terms of catalysts that could come up for us is BD. BD has funded the company over the last few years, and it has been the major catalyst for the investors. We expect to do more BD deals going forward.
That is how we are prioritizing our money overall. We continue to be very disciplined and try and bring in as much non-dilutive resources or funding capital as possible.
Okay. Great. That's super helpful. We're out of time, so we'll wrap things up there. I want to thank you, Al, Nate, Toby, Todd. Thank you so much for sharing your time with us on what I'm sure is a busy day for you today. That's super reassuring to hear that it's not the capsid. That's great news. Thank you for updating us today.
Thank you, Jay. Thanks for having us.
Thank you, Jay. Our pleasure.