All right. Thanks very much. It's my pleasure to be moderating this panel with the Lexeo team here. We'll keep it conversational, but maybe Nolan, I can turn it over to you to start and just give maybe a three-minute overview and update, and then we'll do Q&A. Sound good?
Sure. Thank you, Paul. It's always great to be here. Lexeo is a cardiac genetic medicines company. We're focused on two clinical-stage programs. The most advanced is treating a disease called Friedreich's ataxia. This therapy is treating both the cardiac and neurologic components of that disease. We have recent phase I/II data showing an improvement in the hallmark of cardiac FA, which is hypertrophy, where all of the patients that presented with abnormal left ventricular mass index or hypertrophy have now returned into the normal range. We've also shown an improvement on the neurologic scales of Friedreich's ataxia at about a 1.5-2 point improvement, which is similar to the commercially approved treatment.
In totality, we believe that this therapy has the potential to change the standard of care in Friedreich's ataxia, and we're excited to move into a registrational study next year where we can move towards a BLA in the near term. Our next most advanced program is treating arrhythmogenic cardiomyopathy, where we are focused on the PKP2 mutation, which is a roughly 60,000 patient rare disease in the U.S. For example, more than twice the size of Duchenne's muscular dystrophy, making this one of the most significant commercial opportunities in clinical stage gene therapy. We've completed enrollment of our phase one two study of 10 patients, including seven patients at the high dose. We're working towards a readout of primarily the high dose data in January, where we'll have at least five patients beyond six months of treatment follow-up, and we can evaluate efficacy.
In totality, both programs are moving quickly. The PKP2 program likely also on an accelerated approval path. I'm excited to talk about that today.
OK, great. Yeah, let's start with FA. I guess, can you just outline for people where you are with regulatory alignment? What's the conversation inside Lexeo like when we see stuff like what's happening with uniQure and all the stuff in the news about CBER and CDER? Again, talk about where you are today and your confidence that things are firmed up for the pivotal.
Yeah, so let's just go through it sequentially. We reached alignment with the FDA on many of the key parameters associated with the accelerated approval path. First, the study itself, so the co-primary endpoints. We know there will be left ventricular mass index and frataxin expression. The effect size that we need to achieve for LVMI is at least a 10% reduction in LVMI. The effect size on the frataxin endpoint we need to achieve is more frataxin expression post-treatment than what we saw pre-treatment. This is what we do know in terms of the study. We also know that the FDA is open to an endpoint of 12 months or sooner in terms of evaluating that effect. We also know that they're open to an inclusion criteria that has patients with abnormal LVMI as the focus population of the trial.
These are the things that are fully aligned. When we received our breakthrough designation in July, we decided to go to the FDA to ask a question, is there a more rapid path to a BLA than the one that we had discussed with them previously? That led to a conversation that we had very recently where the FDA showed an openness to pooling data from our phase I/II study with data to be generated in a future pivotal study. We know we will need to run a pivotal study. We will have an appropriate statistical plan agreed prior to starting that study. We also know that they're open to pooling data from our phase I. The effect of this would mean that the study may be smaller than what we had even previously guided.
Just to kind of frame that, right, essentially you'd be able to use, what, six patients from the phase one, five out of six that are already LVMI responders?
That's correct.
OK. What had been the thinking on how big the pivotal might have needed to be, like in that 20-30 ballpark?
Yeah. What we guided to previously, this was earlier in the year, we guided to 12-16 patients in the adult cohort. And then we also have adolescent and pediatric cohort. So the adult portion, which is the efficacy component of the study, previously was 12-16 patients.
OK. You never, I guess the 20-30 number I thought kind of included like everybody, right?
It included the adolescent and pediatric.
OK.
It would not have been 30. It would have been closer to 20.
I guess when you guys hear that the FDA is open to something like that, and again, understanding that we're in this kind of FDA landscape where we read a different article every day, does it make sense to run with that as your base case and say, OK, we have this data. Now we're only going to need another eight adult patients? Or how are you guys kind of strategically planning around the different scenarios?
I think the important theme or learning coming out of what we're seeing from other companies is that designing the statistical plan and aligning with the FDA before you run the study is an important aspect. The idea that the FDA is open to pooling, it's now a conversation that we need to have and we need to finalize with them as to how we integrate the pooled data into an overall BLA data set. That's a conversation that we will formalize and align with them on. I think it's OK to run with the pooling in the case that we align on a statistical plan that the FDA is comfortable with. The other benefit of pooling beyond just efficacy is that we have 17 patients treated in this phase I/II study.
From a safety database perspective, to the extent we can pool the data, that data will all be relevant from a safety perspective as well. There are a lot of advantages to pooling. Frankly, if there were not advantages to it, it is not a strategy we would have considered pursuing in the first place. I think the 12-16 patient study was itself a manageable study for us to enroll. We consider pooling or strategies associated with that because we think it will get us to a BLA even faster than what our base case plans would have suggested.
Yep. Yep. OK. What are the next steps now with the FDA?
Next steps, we will have a meeting where we will discuss the finalized study design and the statistical plan. This will allow us to have the conversation that we need to reach that final alignment. We will be reporting the results of that in early 2026.
OK. When you think about the LVMI outcome, we're talking about a responder endpoint above a certain threshold, a mean? Can you talk about what your base case is?
Yeah. For the LVMI endpoint, we anticipate now for 0.05. Just comparing the LVMI and the control group over what occurs in a year and propensity matching that to the treatment group, the control group being from a natural history study, and then just comparing. Now we're powering it to see a difference greater than 10%, which is what we agreed to with the agency.
Yep. OK. What do we see from natural history over the course of a year? Any change?
Either no change or an increase of 2%-5%.
No change or increase. OK. And that's specific to what stage of disease? Because I think with FA, you and I talked about this before, there's this increase over time, and then in the late stage, it actually atrophies.
We're taking a lot of extra measures to avoid the very end stage patients, as I think most people are in gene therapy. In FA specifically, the very end stage is when patients' ejection fractions start to drop, their LVs start to get big, their LV mass actually starts to go down. By including, we have an LV ejection fraction cutoff, we're avoiding those kind of stage four patients. Short of that, where they do start to drop, those patients also get sick and die. We're not including those patients in our trial, haven't thus far. I think by just having an EF cutoff, we avoid those patients.
OK. OK. Can you talk about just the alignment you have on frataxin and the way you're going to be quantifying frataxin?
For frataxin, it's been very consistent with the agency that any increase in frataxin is a sufficient endpoint for us. In this case, it would be a responder rate as opposed to an average or any ordinal number. We have not aligned exactly on what that responder rate would be, whether that is 60% or 70%. It is just yes, no. That is not compared to the natural history group, obviously, just what they had before versus after.
We achieve that response in frataxin expression across 100% of the patients in our phase I, irrespective of the dose. We also saw a dose response between cohorts one, two, and three, where cohort three had the highest % increase in frataxin. While we will have a responder rate, it's unlikely that we need to achieve 100% of the patients achieving more frataxin post-dose than what we saw pre-treatment. We achieved that in 100% of the patients in our phase I.
OK. OK. Just to round out the discussion, in the six patients that you think you'll likely pull into the pivotal analysis, where exactly are you with LVMI change now at 12 months? What does that kind of imply for what you'll need to see in the next group to kind of meet this primary outcome?
If you look at all doses, all three doses, the average improvement was 18% relative to the 10% effect size. If you look at the high doses only, the effect size, I think, is about 30% at 12 months. It is even, I believe, 28% at six months. Whether you look at all doses or you look at the high doses, we are clearing the 10% threshold pretty readily. This goes into how we power the study, how large it needs to be, and obviously gives confidence in our ability to achieve the effect size that the FDA would like to see.
Do you think it will be just using those high dose patients or all patients in this pivotal analysis?
I think all patients.
OK. OK, great. I want to ask a little bit more about PKP2, but just maybe one more question on FA. You've also talked about other functional endpoints as well and maybe having some sort of benefit on the neurological piece of the disease. Can you speak to that?
Yeah. So it's really interesting how this has developed. I think we designed this construct originally to primarily focus on the cardiovascular component of the disease, utilizing a cardiotropic capsid. However, we utilize a ubiquitous promoter, which was resulting in frataxin expression in all organs. And the reason for that, these patients are frataxin- deficient in all organs, including in skeletal muscle. We did see with this therapy in preclinical studies some transduction of dorsal ganglia. Obviously, we're transducing skeletal muscle as well. What we're seeing now clinically is that that transduction of skeletal muscle is actually likely resulting in a neurologic benefit that we're beginning to see on the mFARS scale, which is the Friedreich's Ataxia Rating Scale that's commonly used to evaluate the progression of the neurologic disease. So we're getting about a 1.5-2 point improvement in the mFARS scale at 12 months.
This is roughly similar to the commercially approved treatment. The interesting thing is we're also getting a deepening effect as time progresses. As we get out to patients that are at roughly two years of treatment follow-up, we're getting an even deeper and more significant effect on the neurologic scale. I think the benefit of this is now we have a therapy that is reversing the cardiovascular disease pathology, and patients that present with hypertrophy are now normalizing. We're also showing a benefit similar to the commercially approved treatment on the neurologic scale. In totality, this represents a step change in the standard of care for the disease. We think it will represent a very attractive commercial asset from a launch perspective.
Yep. Yep. Who do you see as the early adopters of this? And do you feel like there's going to be demand over time to treat patients even before they have cardiac symptoms?
Yeah. I mean, I'll start, and maybe I'll pass to Eric. From a commercial perspective, I think we see the abnormal LVMI population, which is roughly 40% of the adult population, as the early adopters. These are the patients that have the most progressed cardiac disease, likely the ones that are exhibiting symptoms related to cardiac disease. So that 40% is likely the earliest adopters. I think we will progressively move into patients that are earlier in the cardiac disease, so those with elevated wall thickness but not yet an elevated LVMI, and then those that have elevated troponin but maybe do not yet have manifestations of the disease in cardiac MRI. I think we'll then, with the neurologic data to support it, move into patients that do not yet have cardiac involvement via the effect size we're achieving in the neurologic disease.
Obviously, we talked about the adolescent and pediatric cohorts, which should allow us to treat patients of all ages. I think we'll have a therapeutic profile here that could allow us to address the entirety of the patient population. I don't know, Eric, if there's anything else you want to add.
I'm bullish. I've taken care of genetic cardiomyopathies for a decade. I see adolescent parents. They're going to want their kid to get a medicine as soon as they can. We see them do shame. We see them do all these things, especially with a safe profile. I think there's going to be, from the patient perspective, they're going to want it. As we work with providers and payers, it's getting them on board with that. I can't see people wanting to wait until they get sick. That's not usually the attitude, especially with a drug with a clean safety profile.
Makes sense. On the safety side, Eric, I mean, I think at one point you guys were unsure if you were going to go all the way up to the top dose, unsure if you needed to, but also just kind of unsure around, is there going to be safety risk with expressing frataxin in the liver, given that more of any capsid goes to the liver than maybe the target organ? What's led you to be comfortable with kind of moving forward with the high dose here?
Just the collective safety, inhuman safety profile that hasn't shown significant changes in LFTs at all that are clinically meaningful. I think still remaining logarithmically lower than where we see other gene therapies that are getting in trouble with liver, in terms of frataxin being 30%-50% times lower than where you see in mice problems with frataxin toxicity in the liver using gene therapy. The collective data gives us a lot of reassurance. We haven't really seen a blip. That's just using steroids alone. Now we have data two years out. If this was a frataxin tox in the liver, I think that would be manifesting in our earliest patients.
Yeah. I mean, we have no treatment-related SEEs at our highest dose. We have a single treatment-related SEE across the entirety of the study. And it's even deemed to be possibly treatment-related. I think it's a very compelling safety profile for any therapy, but it's a compelling safety profile in particular also for a gene therapy.
Yep. OK. OK. Yeah, maybe let's talk about PKP2. Can you go through, one, the initial data you've generated so far, but from treating some patients and getting kind of part of the way through this study, what have you learned about clinical endpoints in PKP2 and what the right endpoints are to kind of consistently show benefit across patients?
Do you want to speak to the data we've shown out of cohort one and then I may go.
Yeah. I mean, I think broadly speaking, we know there's roughly five criteria to diagnose ARVC. That includes arrhythmia, whether that's PVCs or sustained or nonsustained ventricular tachycardia, EKG changes like T-wave inversions, QRS lengthening, and then structural changes like right ventricular ejection fraction. Consistent with everything that we've done thus far across these gene therapy programs is looking at the totality of data as being the most meaningful in early stage diseases. You don't want to hang your hat on any one endpoint. You want to see things moving in the right direction because that minimizes the play of chance element in small numbers of patients. We're looking at all of these things, capturing them. The question that we have, frankly, is like, what is the temporal relationship between these endpoints? Could some occur sooner?
We think six months is the earliest where we expect to see anything. Could some structural changes take even longer? Time will tell. This is first in man. We do not have corollaries with restoring desmosomal function, so how long it takes for structural changes. Those five endpoints that diagnose the disease, I think, are most relevant.
Got it. Just on the cohort one data, we had data earlier this year and then gave some additional updates on cohort one. We, in effect, have two biopsies from two patients. One exhibited an effect in PKP2 expression that got them to, I think, roughly 70% of normal. In that patient, we saw a response in PVCs and some of the other endpoints that Eric mentioned. We had a patient with less PKP2 expression that had a partial response. Then we had a third patient that was not a responder. That's the low dose. We've since escalated to the higher dose. It's 3x higher than the dose that I just described.
Did the third patient not see an increase in PKP2?
Didn't biopsy. Didn't get biopsied.
Didn't get biopsied.
Yeah, had one biopsy and then didn't want to get a second one.
Got it. Got sent to the next biopsy. We're now at a three times higher dose. We'd expect to see a higher level of activity from a protein perspective. We'd hope to see a more consistent response rate across the patients in that cohort.
OK. Do we know what protein expression level we're solving for in this disease?
Now, I think it's tough with this one because it's a heterozygous disease. The patients are presenting with baseline protein anywhere from 20% to 40%. It's a lot harder to determine how much increment of protein is ultimately going to produce the clinical result, which is why we focused on a readout that could have a sufficient amount of actual clinical data in it. We'll have five patients that are at six months or more of treatment follow-up so that we could get an understanding of how the clinical picture is evolving and not have to only rely on the biopsy data to demonstrate that. I don't know if there's anything else there.
What's your latest thinking on what the right registrational endpoint or endpoints could be?
Yeah. I think it's early to say. We're looking at across all of the endpoints I mentioned. Some of them are ordinal. Some are not. They're related to all of them, especially, I should say, the arrhythmia endpoints are related to quality of life. So ventricular contractions, obviously sudden death, sustained and nonsustained VT. Patients notice them. That directly relates to their anxiety, quality of life, their risk of death. It's literally preventing sudden death. I think any of those could be good registrational endpoints. There are a number of other statistical things we could think about. Like, do we do RIN ratios? There's ARVC calculator scores, all these things that are out there. We need to see the data. What we want to start with is what are the let's show biologically that we're impacting disease.
We work backwards from that with the data we have and with regulators and patients and providers to provide something that convinces them.
Yeah. On a number of these endpoints, do we have a good sense of what a clinically significant change actually is?
I think that what a clinically significant endpoint is, some of them, I would say something like VT, like sustained or nonsustained VT, any difference is going to be meaningful. Reducing shocks will be meaningful. It's how often that event rate occurs over a year period. We are gathering natural history data to say what's normal and how variable those things are. In terms of reduction in PVCs, I think any reduction is important to patients because they feel PVCs. I think reductions are meaningful. I don't think there's a particular set point. It depends where people start. Some people have 10,000 PVCs a day, and some people have 1,000. If you're going from 10,000 to 8,000, that might still be very meaningful.
Is that like a highly variable day to day for a given patient?
It doesn't seem like it's highly variable day to day. Though we are collecting over a seven-day period to smooth that out a bit. They can, at the beginning of the disease, they tend to have this spike when they're first diagnosed. We are not enrolling patients at the very beginning to avoid that issue of getting a snapshot that's inaccurate.
Yeah. I mean, I would just say probably in the range of endpoints you could look at in cardiac studies, you put LVMI on one end, PVCs on the other. It probably is more variable than something like LVMI. To Eric's point, I think there's a spike at the beginning of the study and probably a little bit more stable later on.
OK. OK. Makes sense. On the safety side, have you treated all the next cohort at this point to clarify?
Yeah. We've treated all the patients in phase I/II study.
Should we assume no news is good news on safety?
We have been giving quarterly safety updates in our quarterly prints. We have been, throughout the course of enrolling this study and our FA study, communicating to the street about the safety profile we're observing across both of our treatments. The safety information related to this, the most recent is, as of last week, the safety update that we've given. I think it's a pretty good picture.
Yep. OK. OK. Just remind us, where are you dosing in PKP2 relative to kind of the ranges for other gene therapies and where that sort of safety risk gray zone is?
Yeah. Our high dose is 6e13 vector genomes per kilogram. Where you begin to see more significant gene therapy-related safety issues, this is liver injury and significant complement activation, you begin to see this typically at 1e14 vector genomes per kilogram, maybe at 8e13. Interestingly, AAVrh10, the serotype we're using, has a feature of has not exhibited complement activation in any significant manner across our studies or others. We have that advantage. We also, as we've been providing through our safety updates, have had no assays related to liver injury. Not only are we at a modest dose, but we're at a modest dose that has produced protein expression across both our FA and PKP2 programs and has done so without the liver injury and complement assays that you would see common to other gene therapy programs.
I think we're in a good dose range with both therapies, actually.
I also like giving a lot of credit to our CMC group. We look at our empty to full capsid ratios. We're, I think, the best in class. By having that be the best in class, the actual total amount of virus that patients are seeing is significantly less. I think when you start to see these adverse events, I always wonder if it's actually due to the increase in empty capsids, which we thankfully see very little of compared to others.
Why are you able to do that better?
We've spent a lot since the beginning of the company, we focused on in-house development of our process. As we developed this SF9 process, we focused on things like ensuring we have a low empty capsid ratio. We focused on yields. We designed this process, actually, to meet the commercial demand of PKP2, which is 60,000 patients times, call it a mid-13th per kilogram dose. We knew that we needed a substantial commercial scale to supply that market. You need a process that's high yielding, low empty capsid so that you can, with a reasonable number of batches per year, supply that type of market. We designed that process from the beginning, knowing we were attempting to treat indications like this one. We're also using the same process for our Friedreich's ataxia program.
While we've produced the vector for our pivotal study using the same 200-liter SF9 suspension process, we've produced several GMP batches for PKP2 already with the same process. It's a platform process that can be excellent.
I think SF9 versus HAC, it's like pretty straightforward. This was a big difference. People recognize that. It's helping us in cost of goods, in safety. It's a win for us thus far all across the board.
Yeah. I'd say inherently, SF9, the baculovirus, would have a lower empty capsid ratio just by virtue of how the gene of interest is produced. That is inherent. We have added other elements to our upstream and downstream process to be able to achieve that. We have two posters on our website that speak to this, both the yields and also some of the process work that we have done on CMC.
Yep. OK. Any questions from the audience? On the commercial opportunity for PKP2, do you think all 60,000 patients would actually be candidates for gene therapy? Or is there like heterogeneity here or like a gradation of severity where we can kind of segment this market?
Yeah. I think the adoption of this one is a little different than FA. Trying to understand who will be the earlier adopters will depend a lot on safety. By being best in class in safety, even if there are multiple programs, you can imagine you have a disease that's not quite as morbid as FA, that the safest one is going to be the one that patients are going to be flocking to. You can see that almost in the speed in which we're enrolling in the trial as well. I think the early adopters are going to be what we're seeing are younger patients that are having a lot of symptoms, patients that want to, frankly, one of the worst parts of this disease is the impact on exercise because you're told that you shouldn't exercise at all when you have ARVC.
Patients that are looking at a life of ping pong and bowling as their only physical activity have been really excited about being in a, that being said, we're not promising anything regarding exercise at this point, but just the potential of something that changes the actual trajectory. If they take standard of care medicines or ablations, all those things, et cetera, no one's telling them not to exercise or to exercise again. By actually changing the trajectory, which gene therapy promises, it gives at least that glimmer of hope of being able to be more physically active.
I'd add, just from an access perspective, most of the patients within 10 years of diagnosis go on to require transplants. A transplant is roughly, in aggregate, a $2 million cost undertaking for a payer. In the instance where you can introduce a gene therapy that resolves a lot of the symptoms and disease burden that Eric just described, but also prevent or delay them from requiring a transplant, you're actually looking at a cost offset that could make a lot of sense from a payer perspective as well. There are a lot of advantages, not only addressing the symptoms in the near term, but addressing some of the structural elements of the disease that prevent a patient from progressing to require a transplant.
Makes sense. Last 30 seconds, you want to talk about cash runway?
Sure. We're very well capitalized. We closed out Q3 recently with a balance of $122 million in cash and marketable securities. We raised in October for $154 million. Net of proceeds there, adding roughly $134million-$135 million to the balance. With a number in the mid-200s, that gets us runway into 2028.
That's great. All right. Thanks, guys. Good seeing you. Appreciate it.