All right, I think we're going to get started. Let me just go ahead and welcome you once again to the Cowen's 44th Annual Health Care Conference, joined by a large swath of our colleagues here on our CNS panel, CNS corporate panel, I should say. So first, all the way to our left is Will Chou from Passage Bio, then we have Arnon Rosenthal from Alector, and then Rick Winningham from Theravance Biopharma. We got Ryan Watts from Denali Therapeutics, and then Rachel McMinn from Neurogene. And I'm joined today on the panel by my colleagues Ritu Baral and Ruby Sierra. We got a lot to cover. Obviously, the whole CNS space and everything that entails continues to evolve in a lot of different fronts.
So I think maybe we'll start with a couple of thematic overview questions, and then we'll kind of dive into some of the company-specific ones as we kind of go through the discussion. So maybe kind of just first, I'll pose it so well, and we'll kind of move down the line and get everybody's thoughts that way. So maybe at kind of a high level, what do you think some of the most important kind of clinical regulatory considerations are when developing drugs specifically for CNS? And I guess putting in the context of the regulatory environment today, I mean, how do you anticipate some of that might change over the next few years, just based on some of your experience today?
Sure. So I'll just start with a couple. So first it is, can you get to where you're trying to get? So can you get to the CNS with the minimal number of trips and the highest level and least invasiveness? So that's kind of the sweet spot there. And I think the other big thing is how well can you measure the pharmacodynamic effects of your product, and how much is that linked to clinical outcomes? It's different for every therapeutic area, the strength of that link. And probably also, how good are the diagnostics around that? So we have a—we're attacking TDP-43 pathology. The diagnostics for TDP-43, they're not quite there yet, but maybe they will be in a few years.
How would you answer that question?
Yeah, so for novel diseases for which there is no disease-modifying drug, I think it's really important to agree with the regulatory agencies on the clinical readouts. We need to make sure that we know what's clinically meaningful. So agreeing ahead of time with the regulatory agencies, what's the primary and secondary readouts? What's a biomarker that really represents disease progression and disease severity? And what's the clinical, if it's a dementia, for example, what's the clinical readouts that would be acceptable on the regulatory agencies? Otherwise, you're going for things that may not be relevant to the patient and may not be relevant to the regulatory agencies.
Yeah, I would agree with that. I think the confidence that you have in the endpoint that you're studying and reaching agreement with the regulators and making sure that they share that confidence as a measure of having an impact in a disease. Then obviously, when you're looking at the next big step going from phase two to phase three, then it's confidence in your program, in your drug, that it's actually going to have the effect that you want it to have. I think the distribution on that confidence is a little wider in many CNS diseases because of the challenge of the disease process that we're trying to affect. But really, to the point that he's made on making sure that you've got agreement with regulators, I think that's absolutely essential and, in fact, is a cornerstone of a successful program.
All right, maybe I would also just carry the conversation to add a little twist in here if there's any particular differences when you're talking about this with more orphan versus broad neurological disorders.
Yeah. So I was going to start with that. So our portfolio has two—I mean, basically two divisions we work with closely. So obviously, the neuro and rare, and they almost have polar opposite views of biomarkers. And I remember, I don't know, maybe it was the early days of Denali 8 years ago, 7 years ago, at a very small meeting where Billy Dunn was present and some of our colleagues from Biogen were present. And Billy was adamant against biomarkers as a potential path for approval. I actually had a one-on-one conversation with him about amyloid and could it be used as a potential accelerated marker for accelerated approval. And it's interesting to see how that evolved over a 2-3-year period. And I mean, we all know the associated controversy with what happened.
But actually, Billy has essentially been indicated in terms of using that as an accelerated marker for approval. And why? Because lecanemab showed clinical benefit, statistically significant. Donanemab showed clinical benefit. And what you're seeing with Tofersen and NFL is the neuro division actually sees value in these biomarkers as potential for accelerated approval, as reasonably likely to predict clinical outcomes. And the irony is that you can actually run, especially Alzheimer's studies, they're relatively easy to enroll. Now, if you go to the rare division and you're working in a rare disease like Hunter syndrome or Sanfilippo, they're actually more challenging to enroll. I mean, they're much more rare. And yet the FDA is saying, "Hey, we need active comparator or placebo control." And actually, the ethics around placebo control is a huge debate in these, especially enzyme replacement therapies, where you know you have a clinical benefit.
We live in this sort of irony right now. The sense I get, and I think probably going to be led by Peter Marks, is that the FDA is going to have this vision that actually biomarkers are going to be the path to approval, especially if they're reasonably likely to predict clinical benefit. But often, as the pioneering programs, you have to establish that relationship. So I think it's a—I mean, I think it's a very important question because in the end, we have people who have different personalities that are making these decisions, and people will change their mind, and then a new pattern will be established with regulatory authorities. I will say that for both parts of the FDA, we have great relationships. It doesn't mean that we agree upon how they view it, but I think very productive and engaging relationships.
Yeah, so just to add the gene therapy perspective, you mentioned Peter Marks, and he's been very vocal and increasingly vocal of really trying to rescue and salvage gene therapy from itself. So I think everyone's very familiar with what's been going on with the Sarepta case, and he's been very adamant that biomarkers just kind of give us something, anything to kind of latch onto. And I think he's really trying to help the industry to get to that approvable endpoint. So for us and how we think about this is, first and foremost, delivery was mentioned. We think getting to the right target tissue has been a really big problem, and we've really focused on that because if you can't get into the right cells and your prospect for benefit is much lower, then your conversation is going to be much more difficult with the regulators.
So first and foremost, get to the right cells, use the right dose. So that's been a big issue in gene therapy, right? If you're underdosing patients, it's going to take you that much longer to actually be able to demonstrate that benefit. And then take your data set and actually have a conversation. And I think what Peter and others are signaling, particularly in the CBER division, is they're open for business, and they want to have that conversation if you have that data. So I actually think it's a great time to be a CNS company as we focus on these kind of core areas. Great. And very similarly along those lines, I mean, Ryan already touched upon it, but how friendly do you feel the FDA is right now to accelerated approval in CNS, especially across the different divisions?
What are some of the key determinants to whether or not an agency will consider the accelerated pathway? Maybe we'll start from the end.
Sure. So a lot of these themes have already been touched on. So we work with CBER, and from everything that Peter Marks has said, I think there is an openness to use accelerated approval. These are potentially groundbreaking therapies, and to get them to patients as quickly as possible, he has talked about that. I think if you combine that with the approval of Tofersen, then there is a nice setup here, not just in CBER, but in CNS overall for using accelerated approval. So we think that portends well for these discussions. Obviously, it's all data-dependent. It depends on how much evidence there is that your biomarker is linked to clinical outcomes. But I think that threshold is definitely lower, it seems, for CBER.
Arne?
Yeah, I think it's not an issue of being friendly or more friendly or less friendly. I think it's an issue of the science is really advancing very rapidly, and there is a lot more information and a lot more confidence in correlations between clinical outcome and biomarkers, for example. So I think the FDA is progressing with the science, and it's really sort of willing to really apply the science. So as Ryan said, I think there will be more willingness to rely on biomarkers because there is more fact-based evidence that biomarkers correlate with clinical outcome like this. A beta, there are multiple examples. The NFL, there are emerging examples. So I think sort of the FDA is going with the facts and sort of modernized the approach.
So I think, yeah, it is a great time to be in neuroscience because I think neuroscience is now where cancer was, I don't know, 20 years ago, and switching to biomarkers, to imaging, to surrogate readouts instead of the very cumbersome and pretty primitive primary readouts that we had until now.
Yeah, I think it depends on the division, and it depends on the experience that the division has had in dealing with the given disease. It can go, I do believe from the comments that Peter Marks has made, that's going in a very positive direction. But I think there are other divisions that may have had accelerated approvals, and either the company hasn't carried through as rapidly on the commitment that they made under accelerated approvals to validate the data or validate the biomarker, and that causes a retrenchment of a given division. I mean, we're in an area and looking at neurogenic orthostatic hypotension and multiple system atrophy in dealing with a division that was burned a little bit by accelerated approval. Therefore, now we have a very cooperative division and a very supportive division with regard to our phase three program.
But we're doing a small phase 3 program to, in fact, achieve full approval, not an accelerated approval, but a full approval. But it is a small study, and it's a small study because we had supportive data in an earlier study and because I think the division believes if you replicate the data of this smaller study, there's going to be a meaningful impact on these patients. But I think with accelerated approval comes with quite and rapid approval generally comes with quite a responsibility. And the responsibility has to be that the biomarker carries through to an effect on disease, period. And if the agency gets its hands burned too many times, then we'll go through some sort of retrenchment as we always do in a bureaucracy or in a regulatory effort, so.
I think we can't forget that between the time of aducanemab approval and then the lecanemab data, there was a lot of backlash around the accelerated approval pathway. I mean, most people are like, "Amyloid reduction does not correlate with clinical benefit." There's congressional hearings. I mean, there was a lot that went into that. I think we're very fortunate that the data played out because I agree completely that what the FDA fears is making a decision that ultimately benefits financially a company but actually doesn't benefit patients. I think that's why they hold a high bar.
And in our field in particular, so again, two separate fields on neurodegeneration with Alzheimer's and Parkinson's, but in, let's say, the lysosomal storage diseases, that higher bar has actually been to our advantage because if you look at Elaprase, for example, and the biomarker that was originally used, which is glycosaminoglycans, just total GAGs, there's like a 75%-80% reduction. And the argument is this drug should be approved. But then actually, as the phase 3 read out, they still had 75% reduction in GAGs, but only a 30% reduction in heparan sulfate, which is thought to be the biomarker that leads to neurodegeneration. And actually, they just recently at World showed that their reduction in NFL was about 10%, and they didn't really see a clinical benefit.
The FDA could have made a decision to approve an IT Elaprase with a suboptimal outcome, and it may or may not benefit patients. I think if given earlier, it could benefit patients. So I think for us, sometimes that higher bar is not a bad thing. You ultimately need to see a clinical benefit related to the biomarkers. But once that's established and you agree upon the assay in fact, and I'm sure many of you are aware, there was a Reagan-Udall Foundation meeting just two weeks ago where this was debated: could heparan sulfate be a marker for accelerated approval? And there were a lot of discussions around the assay itself, but also this relationship between CSF and brain. And why did IT Elaprase show that data, and why are others showing other data? So I think it's a little bit nuanced.
To your direct question, friendly, yes, agreeable, maybe, but led by fear and making a decision that then has to be retracted in some way. I think that that's part of what you see at the FDA.
I mean, just to have a slightly different take on all of this, you mentioned cancer, and I had a similar reaction where 20 years ago, it's like we were all excited if you could get a 1-month survival benefit or a 1-month PFS benefit. And that was enough to get a drug approved. And then we started building off of that and extending it. And now you have patients that have 10 lines of therapy, and they can be on these treatments and alive for a really long period of time. And so if you kind of parallel that, kind of take that into CNS, the kinds of approvals that we have seen, Skyclarys for Friedreich's ataxia, Daybue for Rett syndrome, we talked about Tofersen for ALS, these are not these gigantic effect sizes. And yet you're seeing the FDA approve them. We talked about Sarepta already.
So it seems like the FDA is taking this more kind of oncology mindset, if you will, whether it's accelerated approval or full approval. I think that the merits of accelerated approval have already been talked about. But I think there's this idea of like, just get something to these patients. The risk-benefit is so skewed to the benefit in these CNS disorders, or at least a lot of them. And maybe we can put Alzheimer's in its own bucket because just the sheer number of people affected by that disease, right, it's a little bit more of a lightning rod, and you can have congressional hearings over that, and you're going to break the budget. It's a little bit more akin to the Hep C market and how much care are you really going to charge for a cure?
But I think as you get into these devastating pediatric diseases, I mean, these are some of the words that Peter Marks is using, that just show me something that we can latch onto so we can get treatments to these patients because the unmet need in CNS diseases is just so high. But as long as there's an acceptable safety overall risk-benefit profile, then it makes sense to kind of accelerate these treatments. And I'm not saying that there's not a rigorous scientific support for all of that, but I think there's more of an open dialogue than there was a number of years ago.
So how do you see the path forward and lessons from Aduhelm and lecanemab and also other than don't trigger hearings? But what are the most important learnings? And there's sort of three aspects to this that I'd love you guys to address, and feel free to pick one. But one, the legacy of Billy Dunn. There's new leadership. And is that new leadership as aggressive as he is? Two, we talk about accelerated approval on biomarkers, but there's two ways FDA can grant accelerated approval. There's interim clinical data as well as biomarkers. And we know Peter Marks loves scientific rationale biomarkers. But there is this option for interim clinical data to push accelerated approval. And I don't hear about that at all in all the biomarker conversations, the speeches that FDA gives.
And three, when you talked about how neuroscience is moving almost more towards oncology and FDA's approach, there had been talk a while about an Office of Neuroscience, a Neuroscience Center for Excellence being set up the same way that there was a Center for Oncology Center for Excellence. Where is that?
You can start with me. I have no idea where that Office of Neuroscience is and whether that's going to happen or not. Look, again, I think that this is sort of the big picture case-by-case basis. But I do think that there's a distinction between Alzheimer's and something like Rett syndrome, which is what we're working on. I think there's a willingness and openness to move forward there. And in terms of leadership, I mean, I think one of the big areas we've been focused on is we saw a number of years ago that there were a series of companies running sham-controlled intraventricular administration trials in infants and small children. And you're like, how is it possible that you're going to drill a hole into a kid's brain just to pretend that they have a procedure? That doesn't make a lot of sense.
So in my own personal discussions with Dr. Marks, and it was a sidebar, so I don't want to call it anything more than that, I kind of called out that it's even ethical to have a placebo-controlled ICV trial in these types of indications, right, in children where one of the statutes is like, "Do no harm." So how could you possibly say that the sham is not doing harm? And I felt like his response was like, "Well, there's new leadership, and there's going to be more receptivity there." So that doesn't mean that there's you still have to be able to show that your data is meaningful. So if you just run an open-label study and you can't contextualize it at all, you're going to run into this problem of like, what does this data mean?
So I think the bar is still on the sponsor to figure out how do you set up that external comparator so that there's something to say of like, "Well, this particular improvement is meaningful." So that's how that's at least something that we're thinking about, that the agency may be more open to doing away with sham. But that's going to be administration-dependent, right? So IT lumbar, that might not be the case. That's also CBER, right? We've seen very different outcomes for CBER versus CDER, and there might be some differences there. And I don't want to speculate, but I think we're already seeing some differences on what might be required there. So just something to think about.
I mean, it's interesting. This conversation is all about regulatory path and approval, which 5 years ago or 10 years ago, that's not what we'd be talking about for neuro. There was so much unknown about mechanism and so little progress. But now you're seeing medicines that actually reduce a key biomarker of neurodegeneration like neurofilament. You're seeing actual medicines that are working. So I think it's very exciting that we're having these conversations. I always think it's a little bit humorous. If you work for the center of excellence, I hope you don't everyone else isn't in the center of mediocrity, right? It's like, let's create some center that and you worry about that a little bit. I'm unaware of what's going to happen.
But I think one interesting point, and this is, I mean, part of leading out of the Reagan-Udall Foundation meeting that was basically 13 days ago, is now it seems that Peter is going to lead this broader effort to understand the accelerated approval path across these diseases, recognizing the difficulty of enrollment for rare disease, but also the, I think, very positive experiences, although temporarily painful, in neuroscience. And so anyway, it's a fantastic conversation to be having that we're actually talking about the regulatory path approval in rare neuro and in common neuro.
And I think to tie it back to what Rachel was talking about in comparison to cancer, that's exactly what's going to allow faster approvals is the basic understanding of the biology of these diseases, just like it was in cancer. I mean, you got accelerated approval. You got faster reviews because more was known about the biology of the disease. And one could develop biomarkers and, in fact, understand you were having an impact on disease. And that's where neuroscience is, and that's where it's going to be going to a greater and greater frequency.
Because of the unmet medical need in so many diseases, both large and small, I think it's an extraordinarily exciting time to be in the area and to be able to make an impact at the beginning of this period, which I do think we're at the beginning of the period of neuroscience over the next 15-20 years. I think whether you're talking about Parkinson's disease, whether you're talking about the disease that we work on, multiple system atrophy, or whether you're talking about Alzheimer's or a smaller disease like Rett, it's just terrifically exciting because you could have an impact on these patients. You can give them hope, and you can improve their lives. I think that's, in fact, what's going to lead to faster approvals.
Yes. And you don't need centers of excellence to recognize excellence. And I think that the FDA is recognizing excellence in neuroscience now. For example, we received breakthrough therapy on our drug for frontotemporal dementia, something that was very rare in the neuroscience arena. So I think, in general, the FDA is viewing neurodegeneration sort of more openly and more positively. And even though there was controversy around the first anti-A beta drug approval, I think it did open the field. It sort of changed the philosophy and approach to neurodegeneration. It sort of made it sort of more akin to cancer, give something to the patient, even if it's modest, and we go from there. I think without that, if none of the anti-A beta drugs would have been approved, the whole field would have been stagnated.
I don't think there would have been additional investments or additional interest in the field. So I think that the FDA took a risk, got a lot of heat on this risk, but still did a very good thing for the field. I still think that for patients, because there is clinical benefit, however, modest.
Yeah. I think even as we move more towards earlier accelerated approvals, ultimately, the hurdle isn't just regulatory. It's the clinicians have to believe in differentiation, and the payers in particular have to believe in the amount of differentiation, not just clinical, not just safety, but route and frequency of administration. So ultimately, yes, it's great that we can get approvals earlier, but the metric is going to be similar in the ultimate launch success.
All right. So this is a great high-level interview across the board, touching on a bunch of important topics. So let's maybe dive into some of the specific pipelines that you all are working on here. Well, maybe I'll start with you and then kind of keep going back and forth to keep it interesting. So we recently got some of the first data from your FTD-GRN cohorts. Maybe give us a bit of an update on kind of what the current status of the study is, how many patients you've treated at this point, and really kind of the cadence of upcoming updates from there and kind of what you're watching for.
Sure. Sure. So we've shared data from the first three patients we've treated in our first cohort. We showed increases in CSF progranulin 2-3x the normal levels. We are continuing with cohort one, so there's two more patients in cohort one. Our guidance is that we will initiate dosing in cohort two in the first half of this year, and we are well on track to hit that goal. Then we'll be sharing more data from our cohort one patients in the second half of the year.
I guess this is. I want to ask a number of you are looking at FTD-GRN, and I'm curious to kind of see what, as the field is getting more I don't want to say crowded, but more interesting, let's say, in some of these orphan segments, what your experience has kind of been trying to enroll more patients. I think everybody is inevitably coming up against the reality of this. I guess I'm kind of curious what it's been for you at this point and kind of how you anticipate that over the next year or so.
Sure. Sure. Of course. It's always harder to enroll patients when you haven't shared any data about the program. So since we've shared data, there has been a marked difference in the number of sites and patients who are GRN positive, who are interested. At the same time, a very large, pivotal study stopped enrollment, completed enrollment. And so there is a lot more opportunity for patients who want to get into a trial to come into ours.
Okay. So for this GRN mechanism, again, this is going to be something that I'll also touch on with Arnon and Ryan as well. But there's some speculation that the GRN mechanism might apply to the C9orf72 population, right? And you kind of alluded to the potential to move into that patient population at some point. I guess maybe give us a sense of what the path forward for you is and what you want to see maybe from the GRN patients first and potential timing of how you see that potentially de-risking moving into C9orf72 patients.
Sure. Sure. So I'll just say we're not waiting on anything from the GRN patients to move into C9orf . We've seen enough from preclinical data that if you can get progranulin levels 2-3x, that normal level, there seems to be a benefit for TDP-43 pathology from preclinical models. So for us, it would just be an amendment of our existing protocol to add an arm to include FTD-C9 patients. And we know there was a lot of unmet need out there for trials involving this patient population.
Right. And I guess this maybe ties back into our earlier part of our conversation here. I mean, you've recently shifted focus a little bit to some slightly larger indications. I don't want to quite say broad. We're not talking about Alzheimer's, but away from the ultra, ultra-orphan spaces. And I guess from a strategy perspective, what have kind of been the most important distinctions between where you were focused a couple of years ago and to now and as you've moved into some of these slightly larger spaces? Really, what is that? What's required of that from where you guys stand as you kind of just try to focus there?
Sure. So our initial movements into these indications are still going to be small pilot studies. I would say the biggest difference in how we think about this is longer term, how we think about the scale of manufacturing. So we are an AV gene therapy company. We have our own internal PD capacity. And so we are moving now to a more efficient process to manufacture this. As we get to larger indications, you need to get to that more efficient process before you start any pivotal study. So the good thing is we are already in process of making that move for FTD-GRN.
Yeah. I mean, it's interesting. There's always this conversation about I think it seems to come up particularly for CNS, but I think it's probably applicable to a number of people who operate within the orphan space, how to prioritize which indication. Do you offer validation if it's a novel platform, try to go ultra-orphan or even orphan, and then move into larger spaces? I mean, I guess when you're thinking about that, I mean, what really are the top priorities as a company trying to establish novel gene therapies, novel targets? I mean, how do you decide what to focus on?
Sure. Sure. So we focus first, not just for PBFT02, but the other programs that we aren't necessarily carrying forward, on the shots on goal that are most likely to succeed. So FTD-GRN patients, they're deficient in progranulin. We know that is the underlying cause of their disease for this population. So obviously, that's where we start. We happen to have made a whole lot of investment from a CMC standpoint and a preclinical standpoint on a product that may have benefit in other populations too. So the incremental investment that's required to get into the other indications is not nearly as much. So in that way, we talked about oncology before. It's a little bit like oncology where you have a mechanism, you have the lead cancer that you're going after where you're most likely to work. But we've seen these products get extended to other indications as well.
I think that changes for us because it costs so much to develop a gene therapy. That changes the potential economics of running a gene therapy program.
Okay. And I guess, again, maybe along these similar lines, in terms of interactions with FDA, in terms of accelerated approval, I mean, you touched on upcoming programs in ALS, for example, in an earlier stage, maybe in Huntington's too. I guess in your conversation so far, maybe not even specific to that program, but I'm curious to see what all of you would think about this. Do you expect FDA to stick by the precedent they set with Qalsody? Do you expect that, number one, what we saw with the reduction in NFL will actually pan out in some kind of clinical benefit? But if you move into ALS specifically, or even something like Huntington's, how applicable it would be and how firmly they would stick by what they had already decided, or would they change it based on what happened with Alzheimer's?
Yeah. So obviously, that would be a dialogue. But my preexisting, what I would go in with would be if they have done it in one indication already, such as for Tofersen, then I would assume that a similar approach would work for other types of products as well, especially with CBER.
Okay. And then I guess if that's applicable to ALS, would you expect that to be true for something like Huntington's where it's obviously a notably bigger market, but still overly available therapies at that point?
I would.
Yeah. Interesting. Okay. Maybe we can move along to Arnon in a little bit. I think it's just maybe a half step left, different modality, but kind of similar vein here. Obviously, we got some phase 2 data with latozinemab and FTD-GRN as well, really showed very promising kind of slowing of clinical symptoms, progression of disease over time. I guess one thing that we get a lot of questions about from the phase 2 data is really this natural history comparator. So I guess maybe just walk us through how you selected those patients, in what context does that make sense, and maybe what FDA's reaction to that as a comparator moving forward in other studies would potentially be, you think?
Yeah. So we have shown in an open-label phase 2 study with FTD that carried the progranulin mutations that our drug appeared to slow down cognitive decline by 48% over 12 months treatment, normalized multiple disease biomarkers, including GFAP, appeared to slow down brain tissue loss. But it was, as you said, an open small open-label study where we took historical comparators, means we tried to match the historical control to our patient population based on the baseline cognitive deficits, age, level of neurofilament, gender. And we did it blinded. So we think that we did a pretty rigorous process. And we just talked about rare disease that it's possibly immoral to have a placebo-controlled group in ultra-rare diseases. So I think we did as rigorous a process as possible, given the number of patients that were available and the data, all the data.
We will validate the data in our phase 3. The FDA used this open-label study data to give us breakthrough therapies. The FDA sort of sold validity to this data.
Okay. So I guess maybe help us understand how the patients and the group, the cohort that you selected for the natural history comparator, when we're looking at the patients in the phase 3 study, which I think we should get data from next year, potentially, from a baseline characteristics perspective, how did those two populations kind of compare? Are they roughly comparable? Would you expect similar kind of natural history, quote unquote, to say, in the actual phase 3 study population?
Yeah. They are very comparable, actually. Patients from the historical cohort went to participate in our actual trial. So the patient populations are very comparable. It's the same source of patients.
Okay. So I guess then to that extent, we also get some questions about kind of powering for the phase 3 study, how you took what did you do with the phase 2 data that helped kind of guide some of the assumptions for the phase 3? What was kind of the thinking along? How did you power it? And I think it's powered for, I think, 40% treatment effect size. Why does that make sense? What does FDA think about it?
Yeah. So initially, based on the, again, historical control, there are cohorts that have been followed both in the U.S., in Canada, and Europe over time. And you know the rate of disease progression and the level of variability between patients. So based on this, we estimated how many patients we will need in order to achieve certain effect size. And initially, yes, we calculated based on the historical cohorts that to achieve an effect size of 40%, you need over 100 patients. We now recalculated the effect size using our own patients. And we also decided to remove at-risk population from our trial. The initial trial was a combination of at-risk patients, patients that carry the genetic mutations but don't have symptoms yet, as well as patients that have symptoms.
And we found out based on our own analysis that patients that carry the mutations and don't have symptoms, their conversion rate is very variable. So they added significant variability to the trial. So if you remove them, the variability between patients is significantly reduced. So now with the current phase 3, which we completed recruitment for recently, we think that we can detect effect size of 25%, which we consider the minimally clinically beneficial, meaningful effect. So with our current phase 3, which is ongoing, again, we completed recruitment. We recruited over 100 patients. We think that we will be able to detect over the trial duration an effect of 25% slowdown in cognitive decline.
Okay. So I guess that kind of gets to my next question. So it's not just necessarily statistics for the phase 2, right? You really want to see 25% or above difference between the two arms? Is that kind of how you see it? And that's based on analysis of the phase 2, but also feedback from the physician community and FDA?
Yeah. Absolutely. Means this is considered by KOLs to be a clinically meaningful benefit. This is more or less the benefit that you see with the anti-amyloid beta therapeutics.
Right. Okay. And I guess, look, obviously, this is an IV-administered antibody, right? There's a few different approaches that are being used, even just in FTD-GRN alone, but of course, which is applicable here. Do you have any interest, or have you thought about different dosing windows after the phase 3 reads out, potentially extending it to q6 week, q8 week, or even trying some kind of subcutaneous formulation?
Yes. We will do all of it. The initial dosing was determined by our pharmacokinetic analysis. We want to keep progranulin at physiological levels throughout the treatment. So the current dosing and dosing intervals were determined to basically restore this single missing protein from 50% or less of the normal level back to completely physiological level continuously. But this may not be required for full therapeutic benefit. Maybe intermittent level of this missing protein is sufficient. So we will absolutely, if we see clinical efficacy, we will absolutely try to switch to subcutaneous and less frequent dosing intervals. And then yes, we'll maximize the benefit.
All right. I did also want to ask quickly about TREM2 because we're going to get some data from this Q4 of this year. We know you updated the safety protocol for the trial after some ARIA signals, which I also kind of want to touch on. Everybody in Alzheimer's here, I want to get your thoughts on kind of this potential connection, not just with TREM2, but kind of ARIA and the amyloid hypothesis, as it stands today. But I think it kind of came down to excluding the APOE4E4 homozygotes from the study. I guess why was that the right move? And have you seen any other safety signals since that update?
So just to recap, we developed an activator and antibody activator of TREM2. TREM2 is an immune checkpoint receptor for the immune cells in the brain, a cell type called microglia. It's a prominent risk gene for neurodegeneration. If you don't have TREM2 at all, you develop dementia by the age of 40 at 100% penetrance. If you have one good and one partially bad copy, you have increased risk of Alzheimer's disease, which could go up by 3- to 4-fold. And conversely, people that have higher level of TREM2, as measured by soluble TREM2 in the CSF, are partially protected from Alzheimer's disease. They have slowdown in cognitive decline, slowdown in brain tissue loss, delayed age of onset, better survival, and delayed buildup of A beta and tau.
So there is very strong scientific rationale to activate this immune check receptor and to recruit the immune cells in the brain to counteract the disease. We completed recruitment in phase 2. Almost 400 patients were recruited. We saw incidents of ARIA. ARIA is an imaging abnormality that has been seen on MRI. Similar imaging abnormality was observed with anti-amyloid beta antibodies. We see the same prevalence, the same appearance as anti-amyloid beta antibodies, the same time of appearance, the same dependence on APOE4 copy number, the same reversibility, the same ability to recover from the ARIA. So the ARIA that we see with our TREM2 activator that recruits the immune cells to counteract the disease seem to be indistinguishable from the ARIA that you see with anti-amyloid beta antibodies that showed clinical benefit.
So because the ARIA, which again is just an imaging abnormality, the actual ARIA with clinical symptoms, severe clinical symptoms that we see is less than 1%. We saw 2 patients from over 300 patients that were treated. But because the ARIA was APOE copy number dependent, we voluntarily decided to remove APOE4 homozygotes from the trial. This is about 12%-15% of the Alzheimer's population. And since we did that, the incidence of ARIA was significantly lower. We don't see any ARIA with sort of severe symptoms. So it was, I think, a very good move. Again, it was a voluntary move. We didn't want to get back down. We wanted to complete the trial and see if there is efficacy. And if there is efficacy, we will find ways to bring back the APOE4 homozygotes into the trial.
Edgar, one last question, and then I'll pass it to Eva. Just because we've gotten this trying to understand really the function of TREM2, right? So when we get the phase 2 data in Q4, I mean, CDR sum of boxes is the primary endpoint, but you're looking at a number of biomarkers, right, some of which got measured in some of these other Alzheimer's studies, like lecanemab and all of these. I guess my question really is, let's say if you see statistics on CDR sum of boxes, but you don't really see any notable changes in any of those biomarkers, or even vice versa, what does that tell you? I mean, how much confidence does that give you to move into a phase 3?
Yeah. We expect to see at the minimum positive trends in the cognitive measurements, the CDR sum of boxes, or any of the other four activity of daily living biomarkers or cognitive markers, as well as effect on biomarkers. I think that it's very unlikely that we will see positivity in the cognitive benefit and no effect on any biomarkers. For example, on tau, we think that tau very tightly correlates with clinical benefit. It's possible that we will not see, for example, a strong effect on A beta and still see cognitive benefit and tau benefit. Means A beta is less directly connected with clinical benefit than the other biomarkers. But I think it's very unlikely that we'll see complete disconnect between any of the biomarkers that we look at. We will end up looking actually at 5,000 biomarkers through proteomics.
I think it's very unlikely that we will not see effect on any of the traditional neurodegeneration biomarkers and still see cognitive benefit. They should be linked mechanistically. Otherwise, it will be suspicious.
Okay. So moving on to Rick. Can you start by giving a brief overview of multiple system atrophy? What are the unmet needs in this disease, and how big is the market?
Yes. So we're working in multiple system atrophy, specifically in neurogenic orthostatic hypotension and multiple system atrophy patients. MSA is a terrible, terrible neurodegenerative disease, 100% fatality of the disease. There's about 50,000 patients in the United States that have multiple system atrophy and about 40,000 patients who have neurogenic orthostatic hypotension. It's a progressive disease. It generally takes patients about 2.5 years in order to get an adequate diagnosis of MSA. And almost all patients eventually have neurogenic orthostatic hypotension that, in fact, causes vision weakness, dizziness, pain, coat hanger pain, like what you would see in Parkinson's patients, as well as other inability to stand for a very long time that's strictly due to the NOH symptoms. So what we're trying to do is simply to treat the NOH by elevating norepinephrine levels in these patients in a very predictable way.
We've done an earlier study where we actually looked at Parkinson's patients. We looked at MSA patients. We looked at primary autonomic failure patients and, in fact, found that our drug Ampreloxetine had an effect on multiple system atrophy patients. There was biological rationale for that because of largely intact peripheral nerves, such that the increased levels of norepinephrine could, in fact, increase that blood pressure and achieve tissue perfusion, which is the key to the relief of these symptoms. We got that data. We went back into the FDA, agreed with them on a design of a 12-week open-label study and an 8-week randomized withdrawal study looking at a composite endpoint, patient-reported outcome, composite endpoint to really measure that we were having an impact on these patients' disease and their ability to interact and, in fact, live a better life.
We also followed that with an analysis that was presented last fall where we looked at, well, what sort of a movement in this OHSA composite score, in fact, makes a difference? How big does this move need to be? And we showed that, in fact, a move of around 1 point, in fact, shows that there's a significant worsening or, in the other direction, an improvement in patients' overall well-being. So I think overall, we're set up quite well from a regulatory perspective that if we deliver on our phase 3 program that's underway, the Cypress study in which we hope to enroll about 100 patients into the open-label portion, and then to get 60 evaluable patients out of the randomized withdrawal portion, should we see success there, then I think we've laid the groundwork quite well for an FDA approval.
There really isn't anything that works in these patients. It's an absolutely terrible condition. I think there are disease-modifying trials that are underway that are primarily alpha-synuclein antibodies. But the complexity of the alpha-synuclein deposits in these patients is extremely high. And so we look hopefully some of those disease-modifying therapies will work. But I think in the meantime, that we'll be able to offer the ability of patients to engage in their normal life during the process of MSA by treating their neurogenic orthostatic hypotension.
In Ampreloxetine's phase 2, the efficacy was more pronounced in MSA compared to Parkinson's or pure autonomic failure. Why is that?
So, I think that the reason it worked in MSA patients was intact peripheral nerves, that largely MSA patients have intact peripheral nerves, which really enable the norepinephrine to exert the pressure on blood vessels and achieve perfusion in the tissues that, in fact, relieve the symptoms.
All right. Great. Let's move to Ryan. So a lot going on in Denali these days. I guess first and foremost, I want to ask you, I mean, you have a number of different modalities, right? You have the enzyme-conjugated TVs, and you also just announced the oligonucleotide-conjugated TVs. Maybe first off, timing to the first potential ASO TVs, and how are you kind of deciding which modality to take into which indication? Because I feel like that's kind of the question on a lot of people's minds when it comes to these. But how do you guys decide this?
So I think the way that we're looking at Denali today, and the way we've framed it, is two peaks. We have peak one and peak two. Peak one includes all the clinical readouts that are happening in the next, let's say, 9-24 months. There are 7 programs there, including 4 of our small molecule programs, but also our 3 most advanced transport vehicle-enabled programs. So the enzymes are, we believe, a franchise opportunity with Hunter leading the way. We're dosing in Sanfilippo as well, and we're going to expand the enzyme franchise. The next peak, or peak two, we see a shift. We still continue with the enzyme franchise but go now towards Alzheimer's and Parkinson's.
We made the decision and announced it that our two most advanced ASO programs are going to be for MAPT and SNCA, so for tau and synuclein. I think part of this is that we see the foundation of Denali is engineering brain delivery, so using the transport vehicle technology to get biotherapeutics across the blood-brain barrier, an area that's now become very hot. I mean, many people have their own blood-brain barrier platforms or just started their own blood-brain barrier platforms. We think that's fantastic. I think the data in Hunter and then the Roche brain shuttle data, our data in Hunter with both biomarker correction and robust NFL reduction is really validating for these types of approaches. We expect everyone will be quite interested in that.
We actually believe that our ultimate goal is to solve Alzheimer's and Parkinson's, but it was crossing the blood-brain barrier was step one, and hence the enzyme transport vehicle to be able to do that.
Okay. So I mean, you touched on MPS too, right? So based on the hearing labs I don't even want to call them here, based on the meeting last week, based on your conversations with FDA, but you're going to need full cognitive endpoints here. And I guess more importantly, looking forward, how does this kind of impact some of these lysosomal storage disorders at large? Do you see FDA trying to basically set a precedent here with the biomarkers accelerated approval? And do you think it would have application to like FDA 3A for you guys?
Yeah. So that's a great question. Obviously, our desired path would be accelerated approval, especially with the data package that we have now. So we have over 40 patients enrolled in the phase 1/2, which is a very rare disease. So it's a very large phase 1/2. We see robust biomarker correction of the primary substrate, heparan sulfate , as I mentioned before, NFL, but also improvement in hearing, improvement in behavior, improvement in cognition. And there's enough natural history data, especially from the intrathecal Elaprase study, to know that we can actually improve beyond what you'd expect as declining during that period of time. So we hope that would be enough. But I mentioned before that that higher bar is actually a good thing for us because otherwise, we probably would be competing with the intrathecal Elaprase right now if it was simply a biomarker for approval.
And so it may be the base case may be readout Compass, which will fully enroll this year. But the accelerated case is something that we continue to really focus on in addition to completing the Compass study. Now, it doesn't mean that subsequent enzymes would have to go through the same full approval path. It could be that we establish the relationship between heparan sulfate and clinical benefit. And then going forward, now there's confidence that that and neurofilament are predictors of clinical benefit, so.
Great. Rachel, we'll move to you and your NGN-401 is your lead compound in an open-label phase 1/2 pediatric study for Rett and was designed with your EXACT gene therapy platform to deliver precision-dosed, functional MECP2 to brain cells. First, can you walk us through the EXACT platform and the elements that allow for precision dosing, which is not really something seen generally with AAV gene therapies?
Yeah. So taking a huge step back, it's really important with certain genes to be able to control the levels on a cell-by-cell basis. So some of the things we've touched on today, lysosomal storage diseases, you can express an enzyme at super high levels and just going to chew up a substrate. So if you have really high levels, it doesn't matter. It's fine. It's safe. But in a disease like Rett syndrome, where MECP2, the transgene that causes disease, is a transcription factor that's responsible for the control of thousands of genes, it's not then surprising that if you have too high of a level, it can be really toxic. So that's the problem statement and why EXACT was ultimately developed. But there are other complex disorders that are also call it dosage-sensitive, meaning you have to get transgene levels within a certain range.
Because if you don't, then you can have these downstream side effects that are bad. So EXACT was developed in order to control protein levels from the transgene on a cell-by-cell basis. And the way that we do that, it's a very elegant scientific concept. We're literally playing off of nature. And how does nature regulate transgenes? Well, or just endogenous genes, nature uses microRNAs. And they use recognition sites for that microRNA to target that expression and make sure that things that are too high go down. So we're playing off of that, except what's special about EXACT is that we use a microRNA that has no homology to the human transcripts at all. So that way, it's only regulating the transgene. And then it sets up something called a negative feedback loop.
So the more vector copies that go into a cell, the more microRNA gets produced, the more those transcripts get destroyed. And so basically, the more you have, the more it shuts itself off. And in this way, we've been able to show on a cell-by-cell basis a very narrow level of protein distribution, which has really enabled us to basically push the dose to get into as many cells as possible with AAV9. We haven't talked a lot about it, but there are limitations, as I think most folks know. It's not the most efficient vector getting into neurons, but it's the best we have today. But if you can give higher doses, you get more cells penetrated. But exact helps you limit the amount of overexpression, right, to keep it in very tight levels.
Right now, that's really set the stage to enable us to go directly into a pediatric population, I think you mentioned. We're in the clinic right now. We had a big announcement yesterday that we can touch on, but very excited about that program.
Great. Before we get to the protocol amendments and the dose expansion, cohort expansion, can you talk to the preclinical models in Rett and the preclinical studies that you conducted that give you confidence in that safety and efficacy range?
Yeah. So the first thing to mention is that Rett, unlike a lot of the diseases we've been talking about today, is not a neurodegenerative indication. And the reason why I'm going to tell you that is that when you're in a neurodegenerative indication, your neurons are blowing up, right? They're being destroyed on a regular basis. And so you're in a race to get your drug to work before the neurons are destroyed. With Rett, the neurons are there. They're not functioning properly, but they're at least there. And so the hope then is that when you get to deliver MECP2, that there's a wider window.
The reason why I'm mentioning all of this is that in preclinical models, we or others have shown, and just starting with the big picture, that if you take away MECP2 later in life, at any point in life, the animals actually develop Rett syndrome. This is a protein that is critical for neural maintenance throughout life. It's not just expressed in a tiny little window of development. On the flip side, if you restore MECP2, you can reverse the symptoms of Rett in animal models. Putting that together, and then when you look at the knockout model, if you have clinical severity mutations from human patients and introduce those into mice, the survival of those mice is correlated with the clinical severity, meaning a very severe mutation leads to a much shorter survival, and a more mild mutation leads to longer survival.
So the totality of that sort of backbone of what is available in the Rett space is very exciting in the context of this not being neurodegenerative, suggesting that there is potential for disease reversibility in Rett syndrome with a gene therapy if you can get it at the right dose to the right target cells. On top of all of that, we also, of course, have looked in non-human primates to make sure that we're getting expression in relevant regions of the brain that underlie the cardinal symptoms of Rett syndrome. So, for example, communication, motor function, whether it's gross motor or hand function, breathing, we're able to get to these key domains, and that's really critical. If you take out that microRNA mechanism, that safety valve that we talked about that really prevents the overexpression, these animals experience severe toxicity.
So mice are dead in 2-3 weeks. NHPs, just in a short 3-day study, are already experiencing loss of nerve conduction. So I think the totality of that data suggests we have a wide therapeutic window. An unregulated conventional gene therapy is never going to have that therapeutic window to dose. We've kind of set up the clinical trial to pick doses and delivery that really maximize the therapeutic potential.
Which brings us to the protocol amendments. So on Monday, you allow it. So on Monday, you announced amendments that allow cohort one expansion for the low dose and the addition of cohort two, the higher dose. Can you walk us through those doses, the most recent amendments, and the data detail that we can expect now in your first entry readout?
Yeah. I'll just try to be brief about this. We opened an IND with 5 patients in a single dose. What we announced yesterday was an expansion of that dose from 5 patients to 8 patients, doing away with the stagger. Initially, the FDA required a stagger of all 5 patients. That has now been taken away. We've dosed 3 patients, and we'll continue with that dosing. In addition, we added a dose escalation cohort to really push it. That's our top dose. There's no doses after this, but to really kind of go from bracketed between the two efficacious doses in a mouse to even push beyond that most efficacious dose to, again, push and get into as many cells as possible. That's also 8 patients.
Data in the fourth quarter is unchanged, but that would be interim data from cohort one, from the first few patients of cohort one. Well, we'll have more follow-up. So the first few patients, the first patient was dosed in the third quarter of 2023, the fourth quarter of 2023, and then just this quarter. So when you think about it, those 3 patients are going to have the most duration. But we're continuing to dose in cohort one. So the amount of data that will be available, obviously, we'll have less follow-up on patients 4 and patients 5.
But we'll put together the point of giving that guidance, which was issued in July of last year when we announced our reverse merger, was to we knew what our operational plan was, and we wanted to be able to provide a set of data that could at least have the prospect of demonstrating reproducibility, durability, and being interpretable against not only the patient's baseline but also the natural history of disease. So that's how we're thinking about that and why we're not releasing one patient here and one patient there.
Which brings me to the final question, of course. Not only are they going to compare it to baseline, but they will compare it to your competitors, specifically Taysha with adult data. So how do you do that compare and contrast within Rett?
I don't know that adult is really the fair comparator. Taysha's running a pediatric study. They've said they're going to have data mid-year. It will have, obviously, a lot less follow-up because it was started later than our data. But I think it's tough to kind of compare a 20- and 21-year-old to a four-year-old or a six-year-old. So I think the more relevant comparison is going to be pediatric data. And obviously, we believe we have a best-in-class program based on a number of the things we talked about, but full-length gene, maximizing delivery, and then also our EXACT technology that allows us to really push the dose and have this robust regulation. So we're excited, obviously, for patients to have as many options as possible. But we do think that we have a very nice setup for having a good comparator data.
Great. Thank you, Rachel. I think we're at time now. Thank you, everyone, for joining us, and thanks to our panel.