Welcome, everyone. I'm Josh Schumer from the Kantar Biotech Equity Research Team. Welcome to the Kantar Global Healthcare Conference. Very pleased to introduce the management team from Lexeo Therapeutics. We have Nolan Townsend, Chief Executive Officer, Eric Adler, Chief Medical Officer and Head of Research, and Louis Tamayo , Chief Financial Officer. Gentlemen, welcome. Nolan, maybe you can kick things off, frame Lexeo and the lead programs.
Sure. Thanks for having us. It's always great to be here. Lexeo Therapeutics is a clinical-stage gene therapy company focused primarily on cardiovascular disease. Our most advanced program is treating the cardiac pathology of Friedreich's ataxia. Many would know FA as a neurologic disease. However, for 70% of the patients, the cause of death is the cardiac disease. That's our primary focus. However, this therapy is also showing benefit in certain neurologic scales. Our next most advanced cardiac treatment is treating PKP2-mediated arrhythmogenic cardiomyopathy. This is a roughly 60,000-patient disease in the U.S. For example, it's more than 2x the size of duchenne muscular dystrophy, making it one of the largest targets in, let's say, cardiovascular gene therapy. For this program, we're in an ongoing phase 1/2 study with eight patients dosed. We intend to have a data readout associated with this program towards the end of the year.
Moving quickly on both programs, I'd note that our Friedreich's ataxia program is moving into a pivotal study next year. We're excited about the progress that we're making there, both with the FDA, but also the clinical progress as well.
All right. I want to restart with kind of the big issue I think investors are struggling with in the AAV field. That is some of the safety setbacks that we've seen in other programs. Maybe as you kind of frame your approach to these indications, whether it's the vector selection, whether it's empty full capsid manufacturing, dosing, et cetera, how do you feel your position relative to this field that's started to generate a little bit of safety noise that are making some investors uncomfortable?
Yeah. It's a good question and obviously a very important one today. As we know and have always known, safety in gene therapy is linked to dose. As you get closer to doses like 1e14 vg per kg with gene therapy, you begin to see the pretty significant safety issues that we've seen across the field. From our perspective, the approach for gene therapy is choosing the right vector or capsid for the organ you're targeting and understanding the biology of the disease such that you can choose the right dose utilizing that vector. We've selected AAVRH10 after a body of work suggesting it has one of the most compelling cardiac tropism profiles of any of the known vectors.
With that vector, we've been able to, in both studies, advance a number of patients in terms of the patients we've treated and do so with nothing more than a grade 2 SAE that we've seen to date. In the FA program, we've been able to produce protein expression across 100% of the patients. We've been able to reduce certain important surrogate endpoints across the disease, but do so at a dose of 1e12 vg per kg. This is a full two logs lower than the doses that we've seen, for example, in therapies like Duchenne's muscular dystrophy. I think the ability to treat this disease at a low dose, achieve protein expression across all of the patients that is clinically relevant, speaks to the basic clinical de-risking of the capsid that we're utilizing to treat the cardiac manifestations of these diseases.
I think from a risk-benefit perspective, we're in a very good place with the FA program. We've dosed 17 patients to date. We've seen, again, nothing more than a grade 2 SAE. I think as we move into PKP2, we're utilizing a slightly higher dose for that program. This is somewhat related to the biology of that disease, which is different than FA. We're also using a cardiac-specific promoter in the construct for PKP2, whereas we're using an ubiquitous promoter in the FA construct. This requires a higher dose to get expression in the target organ. Even in this program, our highest dose is 6e13 vg per kg, which is a dose where, historically, you would have expected to see elevations in liver enzymes. However, in our program, we've seen no drug-related SAEs.
What I'm describing is a pretty compelling safety profile utilizing AAVrh.10 across multiple programs with over 20 patients dosed, many patients out to almost two years of follow-up. This is the clinical support we have behind our safety profile. If I broaden the question a bit more, I think safety is always a risk-benefit question. If you look at diseases like DMD, where there's high-dose gene therapy, obviously, it's a life-threatening disease. I think patients will opt for therapy despite the existence of safety events. I'd put FA in that same category, that it's a life-threatening disease. As I mentioned, we think the safety profile of our therapy is one that's very compelling for patients. I don't know, Eric, if there's anything else you want to add to that.
No, I think it's a good first question because to patients, to providers, to regulators, and to investors, safety is paramount. There's alignment as this being the most important thing. I think it requires intention from the beginning. Thinking about safety from the capsid you pick, which is what we did, we picked kind of, to me, a Goldilocks capsid where it targets the heart. It targets the heart probably. We've shown more than AAV9, for example. Yet, it has a complement profile that's more similar to rh74 because it's actually similar to rh74 composition level. Because of that, we're able to avoid both the LFT abnormalities largely and the complement abnormalities that we see in other programs.
I'm not sure the extent to which it's been broadly disclosed, but one adjustment that I would imagine is important to me is the empty to full capsid ratio, right? If you've got a very low ratio and a high dose, you're actually going to be dosing a much higher burden of virions.
Absolutely.
If you've got a lower dose and the opposite, a high ratio, have you been able to think about framing that from your program and some of those others that got into trouble with the 1e14?
It's an important question. We're utilizing the Sf9 Baculovirus Production System, which inherently has a lower empty capsid ratio just by virtue of the system. If you compare that to HEK-based systems, empty capsid ratio is substantially lower. We've gotten more than 80% full capsids in our programs, and that number for PKP2 is actually even higher than 80%. I think we're observing very low empty capsid ratios, which, again, is very different than HEK-based manufacturing systems just inherently.
Another potential advantage as we think about the safety profile. Can we flesh out a little bit the safety data that you've accumulated at where you think the target that I think you've indicated for FA, that target dose of 1e12? When you say you've dosed 17 patients, have they all been at that dose? As we aggregate the two programs and kind of think about the total safety database, I'm not sure if there are any other clinical trials using AAVrh.10 to help inform kind of that denominator of experience to feel comfortable.
Yeah, OK. In the FA program at our highest dose, 1e12 vg per kg, we have had no treatment-related SAEs. This includes patients that are treated out to almost one year of follow-up. Typically, in gene therapy, you would expect to see the safety events early in the treatment lifecycle, whether you look at complement or some of the liver abnormalities that have been observed. As I mentioned, we have patients at, I think, significant enough amount of follow-up to draw some conclusions about safety at that high dose. That's the dose we're taking forward into our registration study. For the PKP2 program, we have a total of eight patients dosed. We have three patients dosed at our lower dose, 2e13 vg per kg . We have the balance of patients dosed at our high dose of 6e13 vg per kg . Again, in this program, we have had no treatment-related SAEs.
I can just speak to those doses as, you know, I think, a good indicator of the safety profile we'd expect. As we steer clear of the doses where we've seen safety issues, for example, at 1e14, we find ourselves in a place where we have the safety profile that I described with, I think, a meaningful amount of follow-up to support that.
Maybe we can turn to the Friedreich Ataxia program. You've said 70% cause of death is cardiomyopathy. Do all patients develop some form of cardiomyopathy? How does it typically present? Is it a presenting feature or instead?
I would say at least 70% of patients will develop some type of cardiomyopathy over their lifetime. Of those patients, they're likely to die from that cardiomyopathy. The symptoms are often difficult to ascertain because it's confounded by the neurologic symptoms as well. Frankly, they're not exercising. Understanding exercise tolerance is challenging in these patients. They'll have symptoms like fatigue. They'll notice problems with ambulation, which is probably a combination of their neurologic symptoms, their skeletal muscle symptoms, and their decreased cardiac function. What's clear, and this is true of these kind of neurocardiovascular disorders, is that neurologists and cardiologists have to work together. As soon as an FA patient is diagnosed, they should be seeing a cardiologist. They should be getting screening tests, cardiac MRI preferably, to identify whether they have cardiomyopathy.
I'd also just add, you know, I think some of the emerging data for this program is also showing an improvement in the mFARS scale, which is a neurologic scale used to support the approval of Skyclarys, which is the existing treatment for Friedreich's ataxia. We're showing about a 3-point improvement in mFARS. I believe Skyclarys was approved on about a 2.5-point improvement in mFARS. I think we're also seeing a clinically meaningful improvement in neurologic scales associated with this disease. It's not only a cardiac therapy. It appears to be one also benefiting other aspects of Friedreich's ataxia.
I assume that is a result of penetration across the blood-brain barrier. Is that something you expected to see, or is that a surprise?
We think that we may, preclinical models suggest there could be some DRG, Dorsal Root Ganglion, transduction. If you look at the mFARS, at the components of the mFARS, a lot of them are like ability to stand, ambulation. Clearly, those things are influenced by cardiac function as well. It's probably a little of both.
Got it.
Also, adding just skeletal muscle transduction, we're using ubiquitous promoters. We're likely achieving frataxin expression in skeletal muscle where the patients are also deficient. I think all aspects, DRG transduction, cardiac transduction, skeletal muscle, this is likely what's showing the neurologic benefit.
Maybe we can talk a little bit as an autosomal recessive disease, the amount of frataxin expression you think you need to be achieving to start to normalize the outcome for patients. What is the evidence that you're able to hit that threshold, assuming it's kind of an expression threshold? You've presented data as like a % increase from baseline. That doesn't necessarily address that.
Sure.
Where have you wound up?
I think to put this in context, you have to remember the function of this protein is enzymatic. It's different than a structural protein. We can see in enzymatic diseases in general, small increases can have profound effect when patients are deficient, regardless of what the enzyme does. In this case, this enzyme is key for mitochondrial function. Mitochondria represent about a 1/3 of the heart total weight. It's shocking how much mitochondria are in the heart. That's why the disease manifests as a cardiovascular disease and as the cause of death is cardiovascular. We think from looking at preclinical models that as little as 5% of normal frataxin in the heart is sufficient for normal cardiac function. There's a YG-800 mouse, which is a humanized mutation in which 5% of normal is associated with normal cardiac output. The bar is actually relatively low.
Frankly, any increase is probably improving mitochondrial function. What's striking to us is how low these patients are with cardiac disease when they start. They're clearly less than 2%, sometimes 1%. They're almost like knockouts. Any increase we're seeing is associated with improved and striking improvements in our key cardiac biomarkers, including LV mass and troponin, whether that's a 5% threshold, lower, higher, it depends a little bit on where the patients are in the disease. Nonetheless, it's pretty consistent that as we increase doses, we're seeing comparable decreases in LV mass and decreases in troponin that are significant, an average 25% reduction in LV mass. This is not something that you, in clinical practice, see with placebo, for example, or see with the therapies that we have. These are beyond what we would think of as just a standard deviation.
We're using cardiac MRI, which is a very precise tool. Troponins dropping as high as 90% drops in troponin, 60% drops in troponin. Again, things that you wouldn't see as a play of chance.
OK. You have two studies ongoing, I think, to summarize FA and then an investigator-sponsored trial at Weill Cornell. Maybe kind of share a snapshot of the data that you've provided. Which program, or is it coming from both?
Yes, I'll clarify that. The data that we've presented to date includes data from both studies, both the Cornell and the Lexeo sponsored study. It's evaluating the same drug product, the same endpoints, evaluating the same profile of patients. The only difference between the two is that the Cornell study does not have cardiac biopsies in it, and the Lexeo study does. If you look at the data on that basis, that's the background. In terms of actually stepping through the data itself, maybe I'd ask Eric to say a few words about it.
Yeah, I mean, I think because we're using the same drug product, other than we have more biopsy data, obviously, all the biopsy data comes from the Lexeo-sponsored study. When we look at the data in aggregate, because we're using cardiac MRI across both, you see across 17 patients, as you increase dose, you see these proportional increases, or I should say decreases, in LV mass. That is dose- responsive. Specifically, what you see is as we get into our second and third cohorts, you see these earlier drops in LV mass and more significant drops in LV mass. Clear dose response relationships across all three cohorts are sustained out to the latest time point. We're not seeing any regression, even in our earliest cohorts, or maybe very minimal regression out to a year and even beyond. Closer now, we're getting two years of data, et cetera.
Maybe I can put that into context. We're seeing about a 25% average reduction in left ventricular mass index. The threshold of reduction we need to achieve in LVMI is aligned with the FDA, and our pivotal study is 10%. We're seeing frataxin expression in 100% of the patients that we've treated across all of the doses. The threshold for the FDA is that we need to show more frataxin post-dose than pre-dose. Important secondary endpoints, lateral wall thickness, we're showing a 14% improvement in wall thickness. We're showing about an average 60% reduction in troponin. I think we're showing clinically meaningful improvements on all of the surrogate endpoints. We're showing improvements of a greater magnitude than what the FDA would require for an accelerated approval.
We simply have to replicate that same result in our registration study with up to 16 patients, which we've dosed already 17 patients in our Phase 1.
I mean, I'll add, I think one of the most important things, the takeaways for us is 10% reduction in LV mass is associated with an increase in LV mass associated with a 20% increased risk of death. We're at 25% reduction, but even a 10% reduction is significant. Cardiologists who are going to be engaged with these patients recognize these biomarkers of LV mass and troponin now more and more in hypertrophic cardiomyopathy. I think we're hitting the important clinically relevant biomarkers. We're beyond this hurdle significantly.
I think what's interesting, or it may even be causing a little bit of confusion amongst investors, is that the clinical benefit seems almost to be punching above its weight of expression, where you're seeing these very robust signals. You touched upon this a little bit, Eric, but maybe you can drill down a little further as to, you know, first, how accurate are these biopsy measures of frataxin expression, recognizing that this distribution across the heart may not be homogeneous? As we think about that threshold consideration, if there is one to see an improvement, why it may be as low as it is in this context?
Sure. We're using the most precise tools we can at mass spec to measure these. Obviously, no matter what we do, we're doing repeated biopsies. They're not in the exact same spot. There's going to be a bit of noise. We recognize that. I think that's why the clinical biomarkers are probably more relevant and important. Frankly, when we talked to the FDA and showed them the data, what they saw is that they said any increase in frataxin is all we care about because we don't want to exclude patients because of these kind of creating some kind of a nominal readout on a biologic, relevant or not. It's not clear if it's relevant versus the relevant biomarkers, which are troponin and LV mass. In terms of why we're seeing this association, it's frankly because this enzyme is working in mitochondria that are seeing essentially no frataxin.
Even a little bit of frataxin can probably have profound effects on the ability of mitochondria to function for iron sulfur clusters to form. I would say, you know, it seems surprising, but when you look at the biology of something like hemophilia, we saw the exact same thing. When you have very small increases in enzymes associated with blood clotting, you can see those profound improvements. Conversely, when you just try the old gene therapy approach of giving as much protein as possible, you end up in toxicity. Frataxin, when you try to get above wild type levels, is toxic. By just giving enough for enzymatic proteins is important, different from structural proteins. Every disease needs a different approach. This is 0.01% of the RNA of the heart, very small amounts. All you need is very small amounts.
If you're dealing with something in the sarcomere, you need much more. That's why this is such a good target.
Is there a reason to think there's a bystander effect, where if you transfect one cell, the surrounding cells may not have been?
That is also something that there's some evidence of from some other programs that you could be. I think that's a hypothesis at this point that could be frataxin in exosomes or sharing. There's some evidence that mitochondria can traffic from cells or from non-cardiac cells into cardiac cells when you transduce. There's certainly some biological preclinical evidence of that. I would say beyond that, it's speculation.
Is there an abnormal morphology of these cells on histology that improves?
are iron sulfur clusters, which are hard to quantify. I would say it's not high enough at a point where we could have made an objective biomarker. That's one of the things. There is also looking at mitochondrial structure and function by electron microscopy. We've collected all of that. The problem with these things is they become subjective and not objective. It's hard to take these biomarkers and turn them into something that's reliably precise and objective. They become kind of part of the story. They're certainly there, but not something that we can use as a biomarker as a readout, I think.
Sorry, Nolan.
Yeah, I mean, the only thing I was going to add is just based on what we know about what normal frataxin is in the population around the world. Our research suggests that normal amounts of frataxin could be anywhere from 100 ng down to 20 ng. There are normal individuals walking around with no heart disease, no Friedreich's ataxia that have as low as 20 ng of protein. The question when we're talking about what is normal, what are we at in % of normal, the real question is what is the appropriate baseline to use for that. You could use averages. You could use the lowest point. I think it speaks to these wide amounts of frataxin, and even a 20-ng individual does not have disease.
If we're looking at single-digit nanogram amounts of protein, we're able to show the benefit that we're observing in the therapy that we described.
This is a mitochondrial form of cardiac hypertrophy. A lot of the changes in LV mass are likely due to reductions in mitochondrial hyperplasia. By delivering enough frataxin, you can reduce the need for mitochondrial divide, which is why you're seeing these profound reductions. You're not looking just at reduction in sarcomeric structure and content.
Got it. Now, if I'm not mistaken, you recently received breakthrough designation for this program after you kind of designed the pivotal, which I thought may be interesting validation that with the changing FDA regime, there's still alignment with the program. Maybe kind of expand upon that if that's a reasonable interpretation and then frame the pivotal trial.
Yeah, I think it is. I think for the most part, the breakthrough designation was based on the data that was presented to the market back in the second quarter. It included both the cardiac surrogate endpoints we've been describing. It also included the mFARS data. We received that designation this summer after the change in regime at the FDA. I think it just speaks to their interest in this therapy and seeing it accelerated, getting to patients in an accelerated manner. We have a regulatory update we will give on the final size of the registration study and some of the statistical analysis plan details. By and large, we're in full alignment with the FDA in terms of the path forward. I think the breakthrough designation really speaks to that alignment and their interest in the therapy and the future of the program.
Yeah, maybe frame the pivotal trial design, what you've shared thus far.
Yeah, the final statistical considerations are still to be agreed upon with the FDA. What we've agreed upon, and that is already documented in minutes, is the endpoints. The primary endpoint is a reduction in LV mass, as we mentioned, of 10%, and that would be an average reduction, as well as the majority of patients having the presence of an increased amount of frataxin compared to their baseline. One is an average, and one is just an effect of the frataxin.
An average of what?
An average of all of the LV mass reductions.
Across?
Across the patient population.
Oh, OK, that's interesting. It's not a responder analysis.
No, right.
I'd just add on to that, the inclusion criteria, we're evaluating only patients with what we call abnormal LV MI . This is two standard deviations greater than normal. These are the patients that showed the greatest benefit in our Phase 1 study. Those will be the focus of this trial. Importantly, we're also evaluating functional endpoints like mFARS and the Kansas City Cardiomyopathy Questionnaire. We think mFARS will be the appropriate endpoint to use for the full approval. The concept we're working towards is to have the same study used for accelerated and full approval, with the biomarker surrogate endpoints for accelerated and then functional for the full approval.
OK, and functional will be specifically mFARS?
Yeah, that's what we're looking for.
I guess because it captures the cardiovascular benefit too as well. OK, interesting. Why go with an average LVMI reduction as opposed to more of a responder rate?
We're doing responder rate for the frataxin expression. We think an average is what we've agreed upon, and we're far above that with 25%. We were allowed to capture those patients that are, I think it just works to our benefit to include that because we're so far above. We can be, frankly, it's a better endpoint for us to do it that way statistically. It gives us more power.
Sizes of trial and enrollment timelines?
As we said, we're finalizing that. We look when we use this average reduction of 10% and a responder rate for our frataxin expression of trial size of 12- 16.
OK, and followed for how long?
A year.
12 months. Got it. When do you expect to get underway?
We're expecting to start the study in early 2026.
OK.
We'd have data in mid 2027, we'd expect.
Got it. Just remind us, how many Friedreich's ataxia patients in the U.S. and worldwide?
I'd say 5,000. The nice thing about this disease, and I don't want to say there's nothing nice about this disease, but one of the things that will make the implementation, I think, a little bit easier than others is these patients are all identified because of the neurologic. There's not hiding FA patients out there.
Maybe we can, in the final minutes, touch on the PKP2 cardiomyopathy program. You've started to show early, but very encouraging data there. Maybe share with us the disease measures and outcomes you're looking, you're focusing on and the benefits.
Sure. Quick overview, 50,000- 70,000 U.S. patients, making this probably the largest indication in gene therapy currently that's being evaluated or one of the largest for sure. A very morbid disease characterized by extra heartbeats or premature ventricular contractions, which are a sign of impending doom, you could say, risk of dying suddenly. That's a very quantifiable endpoint for us, reduction in PVCs. Though there's still work to be done, you could see that as a primary endpoint. It's used as a primary endpoint in other small molecule trials of ARVC. There is already some consensus with the FDA of the meaningfulness of that endpoint. As you said, we have early study, early data that we've shared, two patients. I want to put that in context that it is two patients in our earliest cohort. We did see improvements in PVC reduction.
We also did biopsies and showed the increases in protein and presence of mRNA.
Got it. The next steps for that program, the number of patients per dose and dose escalation?
Yes. We've disclosed we've dosed a total of eight patients, three at the lower dose, the balance at the higher dose. We have a total of 10 patients we expect to dose in this Phase 1 study. The readout that we're working towards that will be close to the end of the year will focus on biopsies for the patients we've not yet reported on. We'll be looking at clinical biomarkers for every patient that's beyond six months of treatment follow-up. Why six months? At the three-month time point, the patients are on prophylactic immune suppression that is just being reduced. We wouldn't want to confound the results there by any immune suppression kind of interaction. We'll focus on the six-month time point for the clinical biomarkers. We're looking at things like PVCs , RV function, and others to evaluate the improvement in the therapeutic profile.
Why would immune suppressants interfere with that?
There is an inflammatory component, less of plakophilin compared to other forms of ARVC, but there is an inflammatory component. That being said, no one's ever shown that anti-inflammatory therapies impact PVCs. I think just removing the confounding in general makes it a more scientifically rigorous study.
Maybe we can close with kind of your cash runway and the milestones you're going to be able to achieve with it.
Yes, with our May raise, we're very comfortable with our runway into 2028, including the FA pivotal. We did see some elevated G&A in Q2. That matter has been fully resolved amicably in August with no admission of liability on either side. We see some additional litigation costs in Q3, but don't expect that to continue going forward.
OK. Maybe last question for coming back to the Friedreich's ataxia program. What are the next updates we can be looking for? I'm guessing it's more about the pivotal trial. Will we get other updates?
We will have all of the patients in the trial to date at 12 months of follow-up and beyond by the end of this year. The expectation would be the next update for the FA program in 2026 with all of the patients beyond 12 months of follow-up.
All right. OK, thanks so much for sharing the story of Lexeo. Looking forward to these next updates.
Thanks, Joe.
Thanks everyone.
Thank you for having us.