Everybody, Luca Rossi, Senior Biotech Analyst here at RBC Capital Markets, and today is our great privilege to have Lexeo Therapeutics as part of our 2024 Global Healthcare Conference. Representing the company, we have Nolan Townsend, Chief Executive Officer, as well as Sandi See Tai, Chief Development Officer. Nolan and Sandi, thanks so much for joining us. How are you guys doing?
Very good. It's great to be here. Thanks for having us.
Great, great. I have a long list of questions here, but maybe Nolan, before we go into the details, it would be great if you can give us a little bit of a big picture overview of what progress has the organization made over the last few months, and most importantly, what's ahead here for Lexeo.
Yeah, so I think 2024 is a transformational year for the company. We have three clinical data readouts across our three most advanced programs. The first readout will be in the FA Cardiomyopathy Program, which I think we'll spend some time talking about today. The next readout will be in our Arrhythmogenic Cardiomyopathy Program, treating PKP2-mediated cardiomyopathy or arrhythmogenic cardiomyopathy. And then the third readout will be in our APOE4 Homozygous Alzheimer's Disease Program. Over the last few months, I think we've had a couple of product updates. One was related to our Friedreich's Ataxia Program, where we licensed incremental clinical data associated with a separate clinical study that was ongoing via our scientific founder. This will increase the end of our readout. It also will help support future regulatory conversations.
More recently as well, within this half of the year, we've initiated our clinical trial for the Arrhythmogenic Cardiomyopathy Program as well. So that's another key milestone in the last couple of months. So I think a lot of progress, and we're excited about the readouts that are forthcoming this year.
That's great. That's great. That's a great overview. Maybe can you expand a little bit more on the recent Cornell deal, wide-definition strategy for the organization? You already mentioned you're going to have a more robust data package as well as regulatory conversation, but it would be great if you can expand on it. And also, as the capsid being manufactured at the same place between the Cornell study and your trial, or how should we think about the manufacturing part?
Yes, it's the identical drug product being studied across both. It's the same manufacturing source, the same product. In fact, the Module 3s or the CMC sections are identical across the two as well. So there's consistency in the product candidate. The endpoints in the studies are pretty much identical. The inclusion criteria in the studies are pretty much identical. The only major difference between the two studies is that the Lexeo study has cardiac biopsies included in it, and the Cornell study does not. However, we will be accumulating data from both studies to support the mid-year readout. We'll also be accumulating data from both studies to support future regulatory conversations.
Okay, okay, okay, that's helpful. Can you maybe do a little bit of a preview of the data as we see it in mid-2024? Can you maybe remind us how many patients are we going to see in that data set? And what do you think would be like an underwhelming, on par, versus exceptional data set for that update?
Sandi you want to handle the second part.
Yeah, sure. So in total, across the two trials, we have the 11 patients dosed. So we would be looking forward to reading out additional data in terms of the expression of frataxin, in addition to understanding some of the clinical data on the biomarkers, whether it be the imaging biomarkers, and we are conducting echoes and MRIs, and in addition to blood-based biomarkers such as troponin.
Okay, okay, that's helpful. Can either of you contextualize the elevation and expression that we've seen so far? Again, the data set is relatively small, but we're generally seeing an increase, I think, from 1 nanogram per milligram to maybe 2 ng/mg in the context of the general public maybe having levels between 20 and 60 ng/mg . So how should we think about that? In other words, where do you think is the minimum expression that is required to actually make an impact on patients?
Yeah, so there are a number of vantage points that we can consider when thinking about what target level might be enough to lead to a clinical benefit. Probably one of the main ones is actually from a mouse model, a murine model, the YG8-800 FA model. And in that mouse model, there's about 5% of normal frataxin expression. And within that model, we're seeing fairly normal cardiac output in that model. So that informs that maybe that's a lower-end threshold for what might lead to clinical benefit. We also are hearing from our KOLs as well that they believe that a small, modest amount increase in frataxin should lead to clinical benefit. And we're certainly taking learnings from pathobiology, from other diseases such as hemophilia, for example, where severe is less than 1% factor activity.
And raising that level of activity to where it's above 5% significantly reduces the risk of bleeds. And there are other disease paradigms, too. For example, DMD and Becker's, where having some modest degree of expression changes the clinical phenotype. So we believe that all these various data points point to, at least shows us that, at least with a small increase in potentially in frataxin, could lead to meaningful clinical benefit.
It's also consistent with our preclinical studies. I think we've seen now the biodistribution picture clinically tracking roughly in line with the preclinical studies. From cohort one to cohort two, we increased the dose 3x. We saw a 5x change in protein. That's what we saw preclinically. We've now seen that clinically with the tissue samples that we've since reported.
Got it, helpful. Obviously, expression is super important. We've already talked about it, but we're going to see a lot of other data in mid-2024. Maybe can we talk about the left ventricular mass index for a minute? One, why is that an important biomarker? And two, where is the bogey here? I think if you look at mavacamten, they've shown a 15%-20% reduction in left ventricular mass index. And I think Rocket for Danon appreciates different diseases versus Friedreich's Ataxia. They have shown a 25%-30% reduction in left ventricular mass index. Are those the right bogey for us to think about it as we see the data in mid-2024?
Sure. So I'll start off a bit. Why LVMI? Why is it important? So the cardiomyopathy that's associated with FA is generally a hypertrophic phenotype. And so across many diseases, the association between LVMI and clinical outcomes such as survival, that association has been established. I think that established the LVMI as a clinically meaningful surrogate in terms of what that means potentially for patient outcomes. And what's that threshold that's clinically meaningful? I think what we've seen from certain other conditions that are hypertrophic cardiomyopathies, that roughly a 5 g/m² reduction is clinically meaningful. And I think from what we've seen in our first couple of patients that we reported data that we've disclosed data on, we are already seeing about a 9 g/m² squared decrease in LVMI, which is roughly about 10%.
So we have to take into account the degree of hypertrophy that we see in these patients, which may be different than what's seen in Danon patients.
I'd also add, the two patients, they got close to or within the normal range. So you wouldn't expect to see a more dramatic improvement than the patients reaching normal. A 20% improvement probably wouldn't even have been appropriate in the case of the patients that she was just describing. I also think even in the alignment reached on this Danon program, that there was an effect size that was lower than the amounts you mentioned that the FDA viewed to be meaningful. I think we're excited about the possibility of showing improvement across hypertrophy. I think LVMI will be an important part of the picture here.
Got it. Similar question, maybe for troponin I. Again, can you remind us why it's important? And two, again, what's the bogey? Again, I'm just looking at Onpatro and what Onpatro showed for HELIOS-B. Sorry, for Apollo-B, I think it was a 15% reduction in cardiac biomarker. Is that, again, the right bogey to keep in mind as we see your data in mid-2024?
Troponin is well recognized as an established marker of cardiomyocyte injury. And so I think it's a clinically meaningful marker. What we've seen from the first couple of patients that have been disclosed, and granted, it's just a couple of patients, but early on, we had already seen, on average, about a 40% reduction in the troponin. We know that the measure is quite variable. So what we're looking forward to in our mid-year readout is looking for directional and consistent findings across the board.
Okay, okay, helpful. How are you thinking about dosing? What's your appetite to go to higher doses? I think when I look at your preclinical package, I think going to higher doses is tempting because you've seen this nonlinear relationship where you actually increase the dose by twofold, and then you see the expression going up way higher than that. At the same time, we've seen a lot of companies getting in trouble when the dose is too high; you would [get a] clinical hold and whatnot. So how are you thinking about go-no-go decision to stick to the mid-dose or to actually go to the high dose?
Yeah, so I mean, it always comes down to threading the needle, right? Feeling that you can achieve adequate clinical benefit for patients and balancing that with the safety profile. So I would say we intend to be thorough and thoughtful in our approach to it. I don't think it's necessarily reflexively going to the higher dose, but looking at what we're achieving from protein expression all the way through to the imaging and blood-based biomarkers, in addition to potentially having some early data from CPET that could give us the full picture and make an informed decision.
What I'd add is the clinical translation appears to be running ahead of the preclinical models, meaning at the 2e11 per kilogram dose, preclinically, we did not see dramatic improvements in survival. There were no major differences in some of the cardiovascular outcomes, measures, and cardiac output relative to the untreated group. And yet, at that same dose, we saw this 10% improvement in LVMI. We saw these improvements in CPET and so on clinically, right? So then we've increased the dose. We're now giving five times more protein. The question there is, what effect size are we seeing in cohort two and to the point that Sandi raised? But then the other question is, do you really need cohort three, or are you achieving an effect size across these first two doses that could be sufficient to take forward?
I think the good thing is we have the room to go up if we need to. We also want to look carefully at the data that we're producing now to ensure that we're not meeting our target product profile with this dose and therefore do not need to escalate. That's a question that I think will be answered in part by the data that we'll be reading out here at mid-year.
Got it. If you do decide to go higher, could you potentially change your immunosuppressive regimen? As part of that strategy, I find it kind of interesting that for PKP2 with arrhythmogenic cardiomyopathy, in addition to steroids, you're obviously using sirolimus. We're seeing, generally speaking, more companies in gene therapy using more aggressive immunosuppressive regimens. Could that be a possibility? Or how are you thinking about the immunosuppressive regimen that you're using and potentially tweaking that if needed?
Yeah, I mean, to date, the tolerability profile has been very good. We haven't had any treatment-related serious adverse events. Certainly, we continue to monitor the safety in an ongoing way in our programs. I think that the doses that we're using, even if we go to the higher dose, is still, in relative terms, even to what we're doing in our other program, is still a lower dose that we don't believe we would need to consider, at least at this point, based on what we've seen. We don't think we would need to go beyond the corticosteroids, the prednisone, within the current trial.
Yeah, the PKP2 study is more than one log higher than the highest dose we were sort of considering in this program. It's substantially higher doses that we're starting with there.
Got it, got it. Maybe one or two more on FA before we pivot to PKP2. Obviously, early days, but how are you thinking about the next interactions with the FDA? Obviously, Rocket, a different disease sets a regulatory precedent, if you will, with LVMI and evidence of expression as the co-primary endpoint for their trial. And obviously, relatively small n with just 12 patients. Is the combination of those two endpoints what you're going to ultimately pitch the FDA? Or you haven't made up your mind yet? Or any thoughts?
Yeah, I think we've definitely had some thoughts. I think we're at a good time where the agency seems very open to having discussions and considering, I think, some flexibility in terms of what could be considered for an accelerated approval. And to your point, I think what we've seen in terms of alignment with Rocket gives us some confidence that probably some combination in terms of frataxin expression and expression, along with a measure of hypertrophy in combination, seems fair to think that it could be somewhere along those lines.
I think the DMD program, Sarepta's, is another good regulatory precedent. The accelerated approval based on dystrophin levels is obviously an interesting precedent. We have others for the hypertrophic cardiomyopathies looking at hypertrophic endpoints. I think there are a number of precedents out there. Probably the Danon one is the closest to this particular circumstance. I think there's a lot to look at out there as precedents.
So probable. Then maybe just related to endpoints, but I want to make sure we're clear on this. There's no way the FDA asked you for mFARS because Reata is the comp here, given, again, a totally different mechanism, totally different approach. Is there any scenario where doubt could potentially be the case?
I think that's unlikely. I mean, they squarely evaluated the neurologic component of the disease. We're squarely evaluating the cardiovascular disease. So I would expect these would be more cardiac-focused endpoints rather than something like mFARS.
Okay, okay, helpful. Maybe pivoting to PKP2, maybe just big picture, why are you fired up about this indication and this approach? And then maybe you guys already alluded to it. You're using way higher doses than what you're using for Friedreich's Ataxia. Can you remind us why?
Yeah. So just some background. This is a genetic cardiovascular disease. There's about 60,000 patients in the US. This makes it, for example, more than twice the size of Duchenne's muscular dystrophy, making it perhaps one of the largest targets in cardioskeletal gene therapy. This is a disease of arrhythmias. And there's an endpoint, premature ventricular contractions, which are a precursor to major arrhythmic events, which is a pretty objective endpoint to evaluate in a clinical study. Preclinically, we've shown that we can improve the phenotype across a CRISPR-Cas9 mediated disease model for the disease. We showed dramatic improvements in survival. We showed also improvement in premature ventricular contractions at the doses which we intended to take forward into the clinical study. We've since launched a clinical trial. We're working towards a readout by the end of this year showing protein expression in our first dose cohort.
I think, in general, this is a very intriguing target and indication because, one, the size. But secondarily, it's a highly morbid disease with mortality risk. I think about 50% of the patients go on to be eligible for transplants after they're diagnosed. So you can see that even from a cost-effectiveness picture, the ability to deliver gene therapy could impact the cost of treating the disease. But I think the unmet need here is very substantial. So it takes on a lot of the attributes of what we've seen from other cardiac gene therapies that are out there today. So we're excited about the opportunity. We think as our data and others start to read out, either later this year or next year, there'll be a lot of focus on this program and this target.
Got it, got it. Super helpful. Can you maybe talk about your initial dose? I mean, when you look at your preclinical package, you clearly have shown that your capsid has enhanced tropism for the heart, even semi-head-to-head versus Rocket. However, I think when you look at your starting dose versus them, it feels to me that you're starting roughly in the same zip code as their starting dose. One, is all that factual? And two, if that is factual, what drives that decision?
Maybe less focused on the comparison. But I would say we're starting at a 2.13 × 10¹³ dose. I understand one of the other programs is starting at 8 × 10¹³. That is a 4x difference in dose. So it's a pretty substantial difference. And I believe this would be the lowest starting dose of any of the three clinical programs that are out there, which I think is reflective of the cardiac tropism profile of AAVrh.10, which we've demonstrated both preclinically and both two different large animal models, but in particular, in the actual disease model associated with this indication. So we've shown a greater improvement in the phenotype with AAVrh.10 versus other commonly used capsids. So I think we'll have to look for the clinical data and what expression level we're seeing. And that will, I think, guide to whether we're seeing the same effect clinically.
But I think across all of the studies we've run, we've been able to support this cardiac tropism profile of AAVrh.10.
Okay, okay, that's helpful. I think you already alluded to it. But I think it's important to contextualize the data and your approach versus the competitive landscape. Can you just maybe remind us what are some of the key differentiating features of your program versus either Rocket or Tenaya?
I mean, it really comes down to the capsids and the dosing. That's the main differentiation here.
Yeah. I mean, our perspective is for a gene therapy where you're seeking to treat, call it, 60,000 patients. And there may be undiagnosed patients here. So we may be looking at something more than that. We need to ensure we have a safe product. And that safety comes both from the dose, so the sort of total amount of virus that you're administering into an adult patient, but also the immune suppression. So the immune suppression itself can yield safety events. So in our view, the program that's able to achieve correction of the disease at the lowest dose with the least immune suppression is likely to be the safest program. And therefore, that's likely to be the program preferred, ultimately, by patients.
And so that's what we focused on from a product profile perspective, is ensuring that we're able to deliver against a therapeutic profile to correct the disease, but we're able to do so in a safe manner. And for gene therapy, that means, frankly, just lower doses. And that's where the cardiac tropism profile of the capsid comes into the picture. So that's what we're working towards here. And what we hope to be able to show in the data towards the end of the year is the expression profile at these doses.
In addition to expression, any other key endpoints that we should be focused on? Do you think the year-end data will have already some initial readout on maybe reduction in arrhythmias versus baseline? Or is there any other endpoint that we should be focused on as we see that data year-end?
I think for now, biodistribution is what we're able to guide to. There may be some others there. But we'll have to evaluate that as the study progresses.
Okay, okay, that's helpful. Let's maybe talk too about the CNS approach here. Maybe actually, big picture, can you just give us a sense of your contribution of your spend between the cardiac programs versus the CNS program? And maybe big picture, is this something that you want to pursue by yourself? Or is that something that you're potentially looking for a partner? Walk us through all those dynamics.
Yeah, I'd say the vast majority of our capital allocation is going into cardiac. There is still some spend on CNS. But I think it's relatively modest. We've completed enrollment of the phase 1/2 study. And so that's the majority of the spend. And obviously, the study with completing enrollment takes on a different spend profile than it is when you're actually enrolling patients. We have a modest amount of spending on some of the earlier preclinical APOE programs. But again, we're talking about very modest amounts. I think, in general, we see Alzheimer's and the APOE4 program and population as a very important unmet need, even with the amyloid antibodies in existence. I think there still remains a pretty substantial unmet need for APOE4. So there is a need for precision medicine for APOE4 homozygotes.
I think, though, given the nature of this indication and how capital-intensive it is, it's probably better suited for us to take it forward alongside a partner. So I think at the right point, that is something we would be looking for. But I would note we have data reading out later this year. We have yet to have a regulatory engagement related to this program to talk about the next study. So I think these are likely to be the next steps and developments that would play out here ahead of such a partnership.
Got it, got it. Super helpful. Maybe as we think about the data toward year-end, I mean, we already have seen some initial data around biomarkers. Obviously, are we going to see anything on function toward year-end? Are you capturing ADAS-Cog or MMSE or some of the other important scales or CDR-SB? Or how should we think about those endpoints? Go ahead.
So we are capturing all three of those that you mentioned, the ADAS-Cog, MMSE, and CDR-SB. So it's going to be helpful to see longer-term data, meaning through the first year on those measures. But of course, the study is not powered for that. So we would just be looking for trends, directionality in those readouts. But what we would expect include what you've seen before, which is in terms of looking at APOE2 expression relative to APOE4 and in addition to the CSF biomarkers like Aβ-42, pTau, tTau. And we are also doing PET imaging as well for amyloid and tau.
Got it. Running out of time, maybe one last question, Nolan, if I may. Big picture, what's the most common pushback that you get from investors? And what's your pushback to that pushback?
I think probably just the question on how much frataxin do you actually need to correct the disease. I think that there has been historically a focus on, do you need wild-type levels? Or what's the baseline amounts of frataxin? I think historically, that number has come out of peripheral tissues versus the actual tissue of the organ that mediates the disease. So I think that sort of area of questions are ones that we receive. I think our response to that is similar to what Sandi mentioned earlier. All of the preclinical evidence that we've seen suggests that 5% of frataxin or more in the heart does result in an improvement of the phenotype. So this is across our studies. It's across other published papers.
And then there's this question also of, if that is the case, and it is this low amount of frataxin that can result in a benefit, then what effect can we begin to see in our clinical trial? We have seen already at the lowest dose cohort where we did not expect to see a meaningful clinical improvement. We have begun to see that already. So I think there's a lot of lines of evidence here, both preclinically and clinically, that support the idea that you don't need a lot of frataxin to correct the disease. So I think that's what we're excited to be able to explore here with the media readout. And we'll be excited to share more at that point in time.
Super. I have a lot more questions but no more time. So thanks again for joining us. And good luck for the rest of the conference. And thanks, everyone, for listening to the conversation.