Great. Thanks, everyone. Let's get started. My name is Yanan Zhu, and I'm a biotech analyst here at Wells Fargo. We're fortunate to have uniQure management team with us here today. And for this fireside chat, we're fortunate to have Matt Kapusta, CEO of the company. Thank you, Matt, for being here.
Thanks for having me, Yan.
Great. So, just to kick us off, Matt, could you give us a brief overview of the company's AAV gene therapy platform and the initiatives?
Yeah, sure. So uniQure is, you know, one of the pioneers in the gene therapy space, founded more than 25 years ago. Not only did we get the first AAV approved product in the world back in 2012 , but we did it again back in 2021 with Hemgenix. So we have a very experienced group of people that have a deep domain expertise in understanding AAV gene therapy, how to design constructs, how to develop them, how to produce them, of course, and manufacture them. And we have a portfolio of a number of gene therapy assets, really focused on CNS disorders, as well as liver-directed therapies. We have four programs that are currently in the clinic right now with a lead program in Huntington's disease.
Great. Thanks. And, maybe on the corporate strategy front, you recently had the announcement of sale of the manufacturing facility and reduction of headcount. Can you talk about how these measures support your goal of conserving capital and streamlining operations?
Yeah, our I talk all the time inside the company about my passion. Our mission is delivering therapies to patients, and at the same time, doing it in as efficiently as a way as we possibly can, so you know, we have a fairly high cost of capital now. We haven't done a capital raise or traditional equity raise since 2019. We're very dilution-oriented, and you know, we want to ensure we have the capital to get to value creation across our pipeline, and that is obviously focused on our four clinical programs.
At the same time, we want to have as great a flexibility as possible on our lead program in Huntington's disease, that should we be able to achieve alignment on an expedited development and approval pathway, that we can take this forward as far as we can ourselves.
Got it. Let's talk about your Huntington's program. Congrats on the phase I/ II data update in July. In a presentation, you made a comparison of AMT-130 patients to propensity-matched external controls. What was the rationale for using this method, and how does regulators feel about propensity-matched controls in general?
Yeah. So a propensity-matched control is a fancy name for an analysis, a statistical analysis, that essentially ensures that you have as closely matched control group compared to your treated group. It is a tried-and-true methodology that has been around for many years. There's been products approved on the basis, and I think there were 40 companies in the prior years that have used similar analyses, when deploying an external control. The FDA doesn't come out and say, you know, we support one method or the other, but the FDA, staff on the FDA have actually presented posters in the past on why propensity matching is, you know, actually a more robust analysis. We also know of situations where they've asked companies or have actually conducted themselves, their own propensity matching analyses.
So what our view is, we know for Huntington's that there are prognostic factors that impact the disease course or progression of the disease. We selected eight such factors and wanted to ensure that we had balanced attributes in our natural history comparator group, such that when we look at the clinical outcomes associated with the external control and compare those to the treated patients, that we're, you know, eliminating biases and variability associated with imbalances in those criteria.
Got it. Got it. You mentioned there are examples of approvals based on such controls, right? So could you give some examples of that?
I mean, look, I mean, we know that AveXis looked at propensity weighting. We know that Sarepta utilized propensity weighting. It's analyses that have been done. Now, whether or not it was the central component of the approval, that's information we don't know. But this is an analyses that has been used numerous times, and we know that the FDA is well versed on propensity weighting, and that, when we speak to statistical experts, that they view it as highly robust when looking at external control groups. I mean, because if you, I mean, it just makes sense, right?
If you, if I were to have, in my external control group, a CAG repeat that was substantially lower than those that were in my treated group, right, you would expect to see different clinical outcomes and different progression of disease. And that makes it difficult to interpret therapeutic benefit when you're comparing two groups such as that.
Got it. Makes sense. When will you be discussing the regulatory path with FDA? Are you requesting an end of phase II meeting, or could this be part of the RMAT kickoff meeting following the recent designation?
It's definitely gonna be part of the RMAT process. I mean, the way we view the RMAT process is this really affords us an opportunity to interact earlier and more frequently with the agency than otherwise normally would be the case. So I think normally we would probably consider having an end of phase II meeting at some point next year. Now, under RMAT, we can have the interaction with the various disciplines in the FDA to specifically discuss what you just said, right? I mean, the whole purpose of our interaction is really to understand what our path forward is. What are the hot buttons or the issues? How did the FDA view therapeutic benefit? What's the potential to utilize natural history? What's the potential? How does the FDA view long-term clinical outcomes from our phase I, II study?
I mean, these are the questions and the answers that we hope to get in this interaction. It is, and it's very standard to not get every answer in a singular meeting. We do have this RMAT process, where we'll be meeting with the multiple disciplines in the FDA at the kickoff meeting, and also laying out a plan for interacting more closely with those sub-disciplines over subsequent meetings.
Got it. And the kickoff meeting, do you have a sense of the time for that meeting?
So the way that it works is that we would normally put in a meeting request, that would typically be done about two months prior to the meeting. About a month prior to the meeting, you would submit a briefing book that would provide positioning statements around the specific questions you want to answer. You would get in advance of the meeting written responses from the FDA, and then tailor your meeting based on those written responses. We've guided that we would expect to have that meeting in the second half of the year. We continue to communicate that guidance, and nothing has changed.
Great. Great. How should we think about a potential accelerated approval pathway with biomarkers and supportive endpoints?
Yeah. So I think when you think about Accelerated Approval, you have to step back and think about what is, what data are we producing in our program. So traditionally, Accelerated Approval have typically been focused on data sets where you are able to secure therapeutic biomarker, potentially, therapeutic biomarkers that are considered surrogate proxies for clinical benefit. And you're able to demonstrate with the FDA a correlation between those surrogate biomarkers and clinical outcomes. And that's always been the tricky part, right? Because how do you demonstrate, you know, correlation between the biomarkers and clinical outcomes without actually having clinical outcomes data?
So, I think that the difficulty in Huntington's is that the biomarkers are, there are biomarkers, but they are not nearly as validated as they are in other indications. So the benefit that we have at uniQure with AMT-130 is, because we're a one-time administered therapy, we can actually accumulate long-term clinical outcomes data associated with our gene therapy in early clinical trials. So, we believe that the FDA does not absolutely require a validated surrogate biomarker in order to assess accelerated approval, and that they, if there is evidence of clinical benefit that come out of or derived from clinical outcomes, that would hold significant weight in evaluating this. So I think we're gonna be looking at both the composite Unified Huntington's Disease Rating Scale.
That's a key endpoint that looks at outcomes data across cognitive, motor, behavioral, and psychiatric factors. We'll be looking at Total Functional Capacity, which we know is an area of focus with the FDA, and we'll also be looking at neurofilament light. Neurofilament light is a key biomarker that we think is supportive, but doesn't necessarily have to be a primary surrogate endpoint. I think we're gonna be looking at a totality of evidence that is likely driven by propensity-weighted or propensity-matched clinical outcomes data and supported by neurofilament light, which is probably as validated of a measure of neurodegeneration as there is within the CNS space.
If you, I wanted to follow up on a point about neurofilament light chain may not necessarily be the surrogate, could be a supportive endpoint. If it is not the surrogate, what biomarker do you plan to use as the surrogate?
Let me repeat it again. We do not believe you must have, if you have long-term clinical. Just think about what we're asking.
Right.
Right? What is it a surrogate to? To surrogate to clinical benefit, clinical outcomes. If you have long-term clinical outcomes data, then the question is, what is a surrogate actually required?
Right.
So, you know, next year, we're gonna have three years plus of clinical outcomes data from, you know, close to 30 patients that have been treated with AMT-130. You know, our view is that if we can show a statistical powering across approvable endpoints such as cUHDRS and potentially total functional capacity, and then have neurofilament light as a supportive evidence in the totality of data picture, is that, you know, how does the FDA view that long-term clinical outcomes data as it relates to demonstrating potential therapeutic benefit? That's gonna be the core question that we wanna talk about. We're not as focused with well, what is our surrogate biomarker? Right, we need to have a surrogate biomarker. What do you mean?
We have multi years of long-term outcomes data, and our view is that that is going to be, weighted importantly by the agency.
I see. I see. But, I think the data you have, would you consider that to be appropriate for full approval path or still an Accelerated Approval path?
No. Well, yeah, I think it would probably be aggressive because, you know, the study was not designed prospectively, and really statistically powered in a way that I think, you know, I would consider gold standard. So I think, that, you know, looking to get a traditional approval, while you can maybe theoretically argue that you have clinical outcomes that could support it, I think that might be, aggressive and maybe a bridge too far. But if you have statistically compelling evidence based on post hoc analyses with long-term outcomes that support clinical benefit, can you make an argument for conditional approval? I think that one might argue, actually, that's more persuasive than even a surrogate biomarker that hasn't really been validated and correlated with therapeutic benefit.
Yeah.
Such as mutant huntingtin in the CSF, right?
Right. Yeah. So that should make a very interesting interaction with the agency. So definitely looking forward to hearing update from that RMAT kickoff meeting, second half. But maybe a related, somewhat related question: Do you think the RMAT acceptance makes you more confident that FDA may accept an external control as a comparator arm?
It definitely makes me more confident that, when the FDA is evaluating potential therapeutic benefit, that they are willing to utilize a natural history comparator. Now, evaluating potential therapeutic benefit in the context of an RMAT application is no doubt different than, you know, determining that or evaluating that for registration.
I think I was pleasantly surprised, because quite frankly, you know, we designed our phase I/II study with actually a properly constructed internal control, and I think it was very possible that the FDA would say, "Look, that's what we need to use." I think this makes me optimistic that the FDA, you know, is gonna look at therapeutic benefit that way, but we obviously need to have the conversations with them when we think about expedited approval, utilizing an external comparator. We need to have those conversations in order to understand really how they're thinking about registration.
Got it. Yeah, have you had any direct interaction with the FDA following the RMAT designation?
We have not had any, you know, other than our typical reporting requirements, I mean, we have not... You know, we announced that back in May, right? So it's only a handful of months later. Right now, we're, we've not had any formal interactions with the FDA that would be minuted in such way.
Right. I think it would be helpful if you could lay out the path that you will be suggesting to FDA and what are the other potential alternatives. I think you articulated pretty clearly what you think the path should be. What might be the alternative? Is that a fair question?
Yeah. So there's, you know, there's bookends, and then there's, you know, there's different permutations in between. But I think if you look at the bookends, you know, we believe, given the extraordinary high unmet need in HD and lack of any disease-modifying treatment options for these patients, that the ability to utilize and leverage long-term outcomes from a phase I/II study is a compelling basis for expedited approval. That's one bookend we talked about. The other bookend would be that you know, you need to conduct a proper phase III study with a you know, clinical outcome measure as a primary endpoint and an internal control group. And this would be you know, potentially akin to what you know, what Roche constructed and executed with tominersen, right?
And you know, what you can see is that, you know, if in fact it is a slow progressing disease, there is patient variability, we're looking to slow progression. You know, if you're, you know, if you're required to have to do very large studies with two- to three-year endpoints, you know, you're looking at significant time and substantial dollars, right? And then there's stuff that's in between, right? Do you need an internal control? If you are required to do a phase III study, can you still utilize an external control? Does the FDA come back and say, "Look, you know, we're comfortable potentially looking at a phase I/ II long-term outcomes data, but we want you to supplement that with some additional patients focused on ABC." There's things in between, but that's kind of the bookends of the path forward.
Great. Thank you. So, I think you can share anything that you learned on neurofilament light chain from cohort three, you know, evaluating the perioperative immunosuppression? Any initial signal that the initial spike of neurofilament light chain could be mitigated, you know, if to any degree by this approach?
Yeah, we haven't shared anything yet. I think we've guided that that will be Q1 of next year, most likely, where we would provide some commentary on that. But I do want to clarify, you know, the. We look at really two - when we think about NfL post-operatively, there are really two pieces of this. One is the initial spike, and the other is the time that is required to return to baseline. When we look at the initial spike, we believe that is predominantly driven by the procedure itself. We saw that this was completely anticipated. We saw this in large animal studies, even in animals where we administered saline. Simply just the intraparenchymal nature and the mechanical trauma of inserting catheters, right, is going to cause some level of scarring. It's natural.
We've seen this in deep brain stimulation. We've seen this in, I think, chemotherapy infusion, and that, in our data, is not dose-dependent. Actually, if you look closely, the spike of the lower dose is actually higher than the spike of the higher dose. What we do see is that the time that it requires to return to baseline does appear to be somewhat impacted by the dose. And so patients at the lower dose return to baseline approximately one year after administration, which is also not unexpected. Patients at the higher dose took a little bit longer, took about six months longer, and that might be, to some extent, related to inflammatory processes associated with, you know, administering a highly concentrated dose of a vector within a tight compartment within the brain.
I really look at NfL across two of those things. I think it's possible. I don't think that immunosuppression would likely impact the peak because it, you know, there's nothing changing in terms of the mechanical process itself. I think one thing that we'll look at is, you know, does it impact the time that is required for returning to baseline? We'll see and discuss those aspects next year.
Got it. Very helpful. And maybe to close the loop on the regulatory interaction front, you know, you will do this with you expect to have this meeting second half 2024. When could the street get a sense of what you learned from those meetings?
Yeah, I think what we would plan to do is, at the very least, provide after the meeting a procedural update and comment about what the next, you know, steps are with the FDA process. We will not talk about the specifics or the content around the path forward until we formally align with the FDA, and what we've been guiding to is, you know, that is unlikely to happen after one singular meeting. It's likely gonna require multiple meetings, but at the very least, we'll try to give as much visibility as we can about the process and the timelines going forward.
Got it. Appreciate that. I appreciate that. Let's talk about your temporal lobe epilepsy program, AMT-260. You know, I think you look to first patient dosing this quarter, if I remember it correctly. Is that? Are you still on track to do that?
Yeah. I mean, we are actively screening patients across all of our new studies, and there's a lot of progress going on in that regard. For temporal lobe epilepsy, the key thing is gonna be identifying those first couple of patients. They're more like sentinel patients. And there is a requirement that those patients have non-dominant lesions. And so those patients are gonna be a little rarer, but we're actively screening patients and putting as much effort as we possibly can to treat the first patient as soon as possible. We'll likely provide the next update at the very least or the very latest for Q3. And we're also similarly making excellent progress for our ALS study as well.
Do you think you're also on track to dose the first patient in the SOD1 ALS study this quarter?
Yeah, I mean, I think our guidance is third quarter. We're not going to change our guidance until we revisit this after the quarter. But as I said, we're actively screening patients, and we're making excellent progress.
Got it. I see. For the ALS study, in which region is it re-enrolling, U.S. or ex-U.S?
It's U.S.
U.S.
U.S. sites. Yep.
Do you expect Qalsody's availability impacts finding patients?
There's no doubt having a commercial product and, you know, is different than having no therapeutic alternatives, but we're actually pleasantly surprised at the level of interest in the study, despite the fact that there is a commercial product available. I think there are patients that, for one reason or the other, are not candidates for Qalsody, and other patients that are very concerned with chronic intrathecal infusions of the product and the tolerability associated with that, and find it very attractive to have a one-time administered therapy that is targeting the same, you know, fundamental biology, if you will.
Got it. Maybe a technical question about the serotype of the vector. The Huntington's program used the AAV5. The temporal lobe epilepsy program used AAV9, whereas the SOD1 ALS used AAVrh10. Can you talk about these differences?
Yeah. I mean, let's be very clear. There is not one mousetrap. I mean, AAV is a bus, right? It's a bus that carries passengers, right, that get delivered into cells. There's not one Greyhound bus that can do that. But we do look specifically at, you know, a number of factors, including the specific cells that we're targeting, the manufacturability associated with the AAV construct, as well as immunogenicity. For Huntington's disease, which was an internally developed program, you know, we had done a significant amount of work looking at AAV intraparenchymally delivered AAV, compared to other vectors, and believe AAV is as effective of a vehicle to deliver a genetic cargo. And obviously was, quote, unquote, "on platform for uniQure." So that is the basis of HD.
AAV9 is, you know, well-characterized and understood as a CNS vector, and that was the underlying construct that was being developed by Corlieve, which was the company we acquired to get access to that program. And then rh10, in a similar way, was also being utilized for ALS. And you know, each of these vectors has substantial data in looking at the specific cell type and the delivery, administration procedure to being an effective vector for the genetic cargo of choice.
Great. And maybe let's also touch on the route of injection for the TLE program. How similar or different is it compared with the Huntington's program's route? Have those route that you are going to use for TLE being used in any other clinical studies in the past?
Yeah, it is. So the great thing about this is that it leverages a lot of the same know-how, techniques, and componentry associated with HD. It is an intraparenchymal delivery. It is also utilizing the same ClearPoint system with convection-enhanced delivery and under contrast-enhanced MRI. So you can, you know, real time confirm that you're delivering it to the key areas that you need to be delivering it to. So that we can leverage, and I do think we've demonstrated across numerous patients in HD, that this is a generally well-tolerated procedure. At the same time, it is a much more simpler procedure, because where in HD, we're putting in three catheter infusions per hemisphere to ensure that we get adequate coverage of the putamen and the caudate.
For mesial temporal lobe epilepsy, we're really focused on hippocampal lesions. It's a much smaller geography of the brain and can be addressed with just one catheter infusion. So, you know, it'll be a more streamlined procedure with shorter anesthesia and, you know, and I think, you know, likely would, you know, could be done even on almost an outpatient basis, quite frankly.
Great. Thanks. Thanks for that color. For the TLE study, what is the most relevant clinical endpoint, and what might be considered a meaningful change?
Yeah, I think that the, you know, the, this is a totally different, you know, kind of, indication and, evidentiary, data package in defining therapeutic benefit than, let's say, in HD, right? Which is a more neurocognitive or neurodegenerative disease. So for epilepsy, you know, you're really looking at seizure frequency. So these patients come in, and particularly initially, these are gonna be patients that are you know, really uncontrolled by available anti-seizure medication. And there are quite a number of patients that are just not able to be controlled. And, you know, they, as by design, will be experiencing at least two seizures a month. And, so we'll have an understanding of what their seizure frequency is coming into the study, and then we'll be evaluating that on the back end, after the therapy.
So there's no doubt that that is gonna be, when we think about understanding the clinical potential of AMT-260, gonna be the most important thing. But we also can look at other markers, and probably the most interesting one is gonna be looking at epileptic discharges, which is essentially, you can look at these, you know, under electrical imaging to see and get a sense of, is there still, excitatory firing of neurons in the emanating from the key areas of, or the epicenter of the seizure activity, and that can also be an indicator that you're having therapeutic benefit as well.
Great. Great. Looking forward to further update on this program. Maybe in the remainder of the time, if we could touch on the Fabry's-
Sure.
program. Congrats on dosing your first patient recently.
Thank you.
Like, the question here is mainly, how do you think your Fabry's program is differentiated from Sangamo's and 4D MT program that are, you know, in a clinic earlier?
Yeah. So there's really two elements of key differentiation that relate to the construct that we've developed with our Fabry disease gene therapy. The most important is that we are leveraging the delivery vector that was utilized in our Hemgenix program. So AAV5, as a liver-directed vehicle, in my mind has probably the longest term data set that demonstrates safety and lower immunogenicity. We treated probably nearly 80 patients across multiple programs with long-term data package that has demonstrated that you know in terms of liver enzyme increases and cellular immune responses that we just see less of that with AAV5. So that's a core principle. And in my mind, for an indication like Fabry, demonstrating safety, the highest bar for safety is gonna be most important. We've seen this with hemophilia.
And so we think that, leveraging a validated, delivery vehicle that is being used in a commercially available product is gonna be core. That's number one. Number two is that we have demonstrated with AAV5, in a large pivotal study in hemophilia, that we can have a therapeutic benefit in patients irrespective of preexisting neutralizing antibodies. So we believe that we can develop a therapy that has the potential to provide access to nearly all patients that have Fabry disease. And then lastly, we're utilizing a promoter that is highly potent and liver specific, and it's a proprietary promoter that we think has the potential to lead to super physiological levels of GLA activity. So that's the areas of differentiation that we're gonna be focused on.
Great. Thank you. That, that's super helpful. I think we are out of time, and I wanna thank Matt for being here and providing all these insights, and thanks, everyone, for attending.
Yeah. Great questions, and thank you.