uniQure N.V. (QURE) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to the AMT-130 Huntington's Disease Program Update Conference Call. [Operator Instructions] Please be advised today's conference is being recorded. I would now like to hand the conference over to your speaker today, Maria Cantor. Please go ahead.

Good morning, and thank you for joining us. This morning, uniQure announced interim data on AMT-130 in patients with Huntington's disease from our ongoing Phase I/II clinical trial in the United States. This update includes safety and tolerability, clinical and functional, biomarker and imaging data on up to 16 patients treated with AMT-130 across 2 dose cohorts. And 10 patients who received an imitation surgical procedure. . Joining me for this investor event and webcast are Matt Kapusta, our Chief Executive Officer; Dr. Ricardo Dolmetsch, our President of R&D; and Dr. Sarah Tabrizi, professor of Clinical Neurology, Director of the University College London Huntington's Disease Center and Joint Head of the Department of Neurodegenerative Disease at UCL. Please note that we'll be making forward-looking statements during this call. All statements other than statements of historical fact are forward-looking statements. They are based on management's beliefs and assumptions and on information available to management only, as of the date of this conference call. Our actual results could differ materially from those anticipated in these forward-looking statements for many reasons, including without limitation, the factors described in uniQure's quarterly report on Form 10-Q filed on May 9, 2023, and other security filings. Given these risks, you should not place undue reliance on these forward-looking statements, and we assume no obligation to update these statements even if new information becomes available in the future. Now, let me introduce Matt Kapusta, uniQure's CEO.

Thank you, Maria, and good morning, everyone. I'd like to open my remarks by talking briefly, about Huntington's, a truly devastating disease with an estimated 80,000 confirmed cases in North America and Europe, and many more at risk of developing the disease. These numbers make Huntington's one of the largest monogenic disorders across the world. Most people with Huntington's disease begin manifesting symptoms in the prime of their lives, a time normally devoted to family, [indiscernible] children and advancing careers. But once Huntington takes hold and incapacitates instinctive by affecting their ability to move, to think and to behave normally, all the things that make us who we are. As function slowly declines, lives are out of what's most valuable and the casualties can be heartbreaking, lost careers, financial hardship, string friendships and marriages and the suffering of children who cruelly witnessed firsthand, the devastation brought by this awful disease, knowing this may also be their faith. While this marks the 30th anniversary of the discovery of the Huntington gene and protein disappointingly, there is still no disease-modifying therapies available for these desperate patients and their families. At uniQure, we have been tirelessly working to provide much needed hope to the Huntington's community. And today, we're very pleased to share with you encouraging interim data from our ongoing U.S. Phase I/II clinical trial of AMT-130, including up to 2 years of follow-up on 26 patients. Key takeaways from the interim analysis are as follows: First, the onetime administration of AMT-130 continues to be generally well tolerated, with a manageable safety profile at both doses. Second, patients treated with AMT-130 show generally preserve function and early evidence of clinical benefits compared to closely natural history. These positive trends were seen across both doses of AMT-130. Third, NfL trends in low-dose patients indicate a stable to improving neurodegenerative profile at 24 months of follow-up with similar downward trends observed in high-dose patients through 12 months. NfL has served as an important endpoint in recent U.S. approvals for other debilitating neurodegenerative diseases. And we are encouraged that AMT-130 seems to be showing sustained lowering of NfL through 24 months of treatment. And lastly, mutant Huntington levels and low-dose patients continue to provide some support for target engagement of the levels were inconsistent over time, particularly in high-dose patients. As we have discussed previously, this is not unexpected given the assay's high level of variability across the industry. In summary, we believe the promising data from this interim analysis support advancing the clinical development of AMT-130, and we look forward to completing the enrollment of the EU trial in the third quarter and presenting new data in the fourth quarter that will include additional follow-up from the U.S. trial and 12 months follow-up from the low-dose cohort from the EU study. We expect to evaluate those selections at this time, anticipate pursuing regulatory interactions to discuss the data from the U.S. and EU studies, as well as the registrational path forward for AMT-130. We also expect to treat additional patients at both doses in the third cohort of the U.S. study to further evaluate the impact of immunosuppression on the near-term safety profile. We expect to initiate enrollment of the third cohort in the second half of this year. Now let me turn the call over to Ricardo, who will go through the data presentation.

Thank you, Matt. I'd like to start by thanking the [ heroic ] patients and their families that have participated in our study and the physicians who care for them. Huntington's disease is an autosomal dominant disease, which means that there is a 50% chance of parent of the disease passing it on to a child. It progresses from a pre-manifest stage with early psychiatric symptoms to an early motor stage to progressively advance disease in 10 to 15 years. The patients in our study are in early stages of the disease, experiencing some motor symptoms. This is what was previously called Stage 1 or early manifest disease and is now known as HD ISS Stage 2 at an early stage 3. Huntington's disease is caused by an expansion of the trinucleotide CAG in the Huntington gene. The CAG expansion causes the production of a toxic RNA and a toxic protein that together leads to the degeneration of neurons. The first neurons degenerate are the medium spiny neurons of the striatum, which is a retail the brain that controls movements as well as reward and motivations. The disease then spread slowly from the striatum to the rest of the grain, over the course of the disease. AMT-130 is a modified adeno-associated virus 5 viral vector containing a microRNA that targets exon 1 of the Huntington gene. It reduces both the full-length Huntington mRNA and the toxic exon 1 slice isoform. So micro RNAs produced by processing of our proprietary microRNAs [indiscernible] which reduces the potential for toxicity by preventing the generation of the passenger strand and by preventing the overloading of the RNAi machinery. AMT-130 is a onetime delivery gene therapy. It is introduced directly into the striatum of patients, so that it reaches the intended target cells. It's delivered using 6 injections to a stereotactically placed cannula that is the with [indiscernible]. Convection enhancement is used to prevent backlog. The delivery of the gene therapy to its target is monitored using real-time MRI and a gadolinium contrast agent. On the right, you can see the infusion of AMT-130 into the putamen of a patient in our study. We are conducting 2 clinical studies of AMT-130, one in the U.S. and one in the EU and the U.K. In both studies, we are enrolling patients with more than 40 CAG repeats, total functional capacity between 9 and 13 and diagnostic classification level 3 or 4. The patients must be on stable medications. In addition, the patients must have striatal volumes above 2.5 cubic centimeters of the putamen, 2 cubic centimeters for the caudate to allow for safe surgery. The study consists of an initial screening visit followed by the surgery. Followup visits occur at months 1 and 3 and then, every 3 months for 18 months. After 18 months, visits occur every 6 months up to 5 years. During each visit, we collect CSF by lumber puncture to measure neurofilament light chain, which is a biomarker for neural injury and disease progression and to measure of mutant Huntington levels. We also perform MRIs, which we used for safety and to measure full brain volume. Most importantly, in each visit, we perform a series of clinical assessments, including the Total Motor Score, which measures motor symptoms; the total functional capacity, which measures the ability of patients to perform activities of daily life; the Stroop Word Test and the Symbol Digit Modality Test, which measure cognitive ability and speed of intellectual processing. These measures are combined to calculate the cUHDRS, which is a sensitive multimodal measure of disease progression. AMT-130 is being investigated in 2 clinical studies. The U.S. study is called HD-GeneTRX1 and has 3 cohorts. In cohort 1, 6 patients received a low dose of 6e to the 12 vector genomes and four, underwent the sham surgery that included anesthesia and surface incision, but did not include delivery of anything into the brain. In cohort 2, 10 patients received a high dose of 6e to the 13 vector genomes while 6 patients received the control surgery. After 1 year controlled patients in the high-dose cohort were eligible to cross over to the treatment arm. 4 patients have crossed over to receive the drug for a total of 20 patients dosed. We are also conducting a EU and U.K. study that consists of 15 patients, 5 of the low dose and 10 at the high dose. It has no control group, and it will be finished enrolling early in the third quarter. Today, I will be presenting results only from cohorts 1 and 2 of the U.S. study. Overall, the patients in the study were well balanced across the treatment groups in terms of sex, age, time from diagnosis, CAG repeats and disease stage. They had an average total functional capacity of 12 of 13, which means that they were quite functional and relatively early in their disease. Their CAP and PIN scores were relatively well matched, which means that their disease is likely to progress, at approximately the same speed. AMT-130 was generally well tolerated across both cohorts. The treatment of merchant adverse events were transient and were mostly related to the surgery worked on the lumbar puncture. The most common symptoms were procedural headache, procedural complication, post-lumbar puncture syndrome, procedural pain and headache. In addition, 5 severe adverse events were observed in the study. All of them were transient and have resolved. In the control group, 1 patient had a deep vein thrombosis. At the low dose, 1 patient had postsurgical delirium and 1 had suicidal ideation and depression. At the high dose, there were 2 SUSARs, both seem to be related to activation of an immune response to the high dose of AMT-130 that resulted in local edema, headache and some behavioral symptoms. All of the patients with the SUSARs have recovered. Subsequent patients have been dosed with a perioperative steroid regimen and thus far, no acute inflammatory events have been observed in these patients. A key component of gene therapy development in Huntington's disease is the development of a matched natural history cohort. Because Huntington's disease is a slowly progressing and lethal disease, it is not ethical to enroll patients in the placebo arm that lasts for many years. Therefore, the control arm in our study only extends for a single year. To provide a basis for comparison, we have partnered with CHDI to use the TRACK-HD natural history study to create 2 natural history comparison cohorts for our study. One data set shown here in the light orange consists of 105 patients that meet the per protocol clinical inclusion criteria for our study. The progression of these patients is similar to the progression of patients with early manifest Huntington's disease in the literature. Because our patients were also selected on the basis of striatal volume, we developed a second data set that includes only the 31 patients that meet both the inclusion criteria and have highest striatal volumes. The clinical progression of this group shown in some dark orange bars on which slower than the first group. It slightly underestimated the expected rate of progression of the patients in our study because these patients are at a slightly earlier stage of disease than the ones in our study. Therefore, the actual control population is somewhere between these 2 groups. The total motor score is a measure of motor dysfunction in Huntington's disease. It is a potential registration endpoint and is the endpoint that is the least susceptible to placebo effects. Higher numbers of the total motor score correspond towards outcomes. This graph shows total motor score, as a function of time. The green line is the control. The purple line is the low dose. And the blue line is the high dose. The dotted line and the shaded area are the natural history. We observed that in the total motor score, the control patients are deteriorating roughly in line with the natural history at 12 months, which is the last measurement. In contrast, both the low dose and the high dose have relatively preserved function compared to the baseline relative to the natural history. The high dose is do slightly better than the low dose. The total functional capacity is a measure of the capacity of patients to carry out activities at the be living, such as doing their finances and caring for themselves. It ranges from 1 to 13 with higher scores being better. Total functional capacity is also a potential registration endpoint for Huntington's disease. In this measure, patients treated with both the low and high doses have largely preserved their function over the course of the study. They are also doing somewhat better than the natural history. This is encouraging, though we should note that the error bars are large and that the control group didn't decline significantly over the course of 1 year. The Stroop Word Test measures cognitive capacity and particularly the ability to disentangle 2 conflicting stimuli. It is the test in which patients are asked to name the color of word rather than the word itself. In this test, the treated patients across both doses have largely preserved function relative to baseline, as well as improvement relative to natural history. The Symbol Digit Modality's test is a test in which patients are asked to use a key to decode a message composed to symbols. It measures processing speed and cognitive capacity. In this test, patients on the high dose are doing better than both the baseline and the natural history. The patients at the low dose are doing a bit worse in the natural history. This is driven largely by one patient who was quite advanced at the beginning of the study and declined towards the end of the study. The composite UHDRS is our composite clinical measurements that combines the total motor score, the total functional capacity, the Stroop Word Test and the Symbol Digit Modality Test. It is the most sensitive measure of disease progression because it measures multiple functional endpoints. The cUHDRS was generally well preserved across both cohorts of treated patients relative to the baseline and better in treated patients relative to natural history. The control patients did not decline significantly over the course of the year. In summary, we're very encouraged by these interim clinical results. There is an early indication of a potentially positive clinical effectiveness, relatively small group of patients treated with AMT-130 with 1 and 2 years of follow-up. Patients treated with a low dose have generally preserved function in 24 months and maybe showing improvement, over the natural history across most clinical measures. Patients treated with the high dose are trending favorably relative to natural history across all functional measures and are performing slightly better than patients receiving the low dose at 12 to 18 months. The control patients changed very little over the course of 12 months, except for the total motor score, in which they declined in line with the natural history. Neurofilament light chain is a protein that is released from injured neurons and is a sensitive indicator of both neural inflammation and disease progression in Huntington's disease. As we have previously described, we observed an increase in CSF neurofilament light chain immediately after the surgery. The data shows that this increase is not dose dependent and is similar to the increase that has been observed in patients undergoing other surgeries like the implantation of deep brain simulation electrodes. In the dose patients, neurofilament light chain starts to decline immediately after the surgery, consistent with resolving postsurgical inflammation. It is particularly interesting that in patients that received the low dose, neurofilament light chain is below the baseline in 24 months and appears to be declining further suggesting that we may be affecting the course of the disease. These averages are also reflected in data from individual patients, which show that the NfL declines below baseline in 4 of the 5 low-dose patients. NfL in the high-dose patients is slightly higher than in the low-dose patients. This appears to be driven by the 2 patients with a SUSAR that had an elevated neurofilament light chain that return towards baseline more slowly than the other patients as the events resolved. In summary, there is an increase in neurofilament light chain following the surgery that is not dose-dependent and likely reflect inflammation associated with surgery. NfL declines towards baseline over the course of 12 months and patients treated with a low dose NfL levels in the CSF are now below baseline suggesting that there may be an effect on disease progression. In the high-dose patients, NfL has also declined and seems to be returning to baseline by 18 months. Mutant Huntington is a CSF -- and the CSF as a biomarker for target engagement. However, it's found at very low levels in the CSF, particularly in patients at an early stage of their disease. In addition, because AMT-130 is administered directly into the striatum, which is a relatively small region of the brain compared to the total volume of the brain, it's not clear how AMT-130 will affect total Huntington levels. This makes the data from this asset somewhat variable and challenging to interpret. In low dose patients, we observed some indications of target engagement. On average, patients treated at the low dose show a decline relative to baseline and relative to the control. The decline seems to be sustained through 18 months, although there's a coordinated increase at the last time point. It's unclear that this is a real biological effect or an artifact of the assay. At the high dose, we did not observe a decline in the mean relative to the baseline. The individual patient data shown on the right reveals that the mutant Huntington response was highly variable with 4 patients showing the decline from the baseline and the others showing dramatic excursions over time. 3 of 9 evaluable patients in the high-dose cohort had CSF Huntington reduction below baseline as the last management. These dramatic excursions were also observed in the control dates. Importantly, the mutant Huntington data were not correlated with either clinical function, adverse events or with changes in neurofilament light chain. In summary, the mutant Huntington data are complex and confounded by multiple issues, including the reliability of the assay and interpatient variability. We believe that we see some evidence of target engagement in patients treated with a low dose of AMT-130. And patients treated with a high dose of the data are variable, and it appears that there is a decline relative to the baseline in 4 of the patients. In the others, there are excursions that don't seem to be reflected in the NfL measurements or in the clinical measures. It's noteworthy that we don't see significant declines from baseline in the control patients suggesting that any declines from the baseline in the treated patients could be significant. Overall, the interpretation of this assay is still uncertain, particularly because it does not appear to be correlated with either NfL levels or the clinical outcomes. MRI imaging data were used in the study both to assess safety and the potential biomark growth to these progression. We observed a small decrease in total brain volume in all the patients. This slow rate is expected given the early stage of the disease and is slightly above the rate of decrease of [indiscernible] healthy controls. As expected, the other dose of AMT-130 significantly impacted the total brain volume relative to placebo or to natural history. Patients treated with either dose of AMT-130 did have a slightly greater increase in ventricular volume than patients in the control arm. The increase in ventricular volume was not associated with clinical deterioration or symptoms was not dose-dependent and was not related to a loss of brain volume. Volumetric imaging of the striatum was compounded by changes in the structural boundaries of the striatum related to direct infusion into these structures. To summarize, we're very pleased with the results of this interim analysis and believe that the data supports the continued development of AMT-130. The data showed early evidence of clinical benefit in the small group of patients. This is particularly apparent on the total motor sport with the control group deteriorated in line with the natural history and the treated groups retained their function. However, it's also consistent across other domains like the total functional capacity, Stroop Word Test, to Symbol Digit Modality Test and the composite UHDRS. We have also seen decreases in neurofilament light chains that are consistent with resolution of the inflammation caused by the surgery and are excited that NfL is declining below the baseline in the patients that have been followed the longest. This is significant because it suggests that we may be having an effect on disease progression. The mutant Huntington shows some evidence of target engagement, particularly in the low dose and in some patients at the high dose, but the data are highly variable and challenging to interpret. Total brain volume is largely unaffected by AMT-130 compared to natural history. These promising efficacy data are the first results from the study and additional follow-up will be important. Over the next 6 months, we will be adding more patients from the European study to our data set and following these U.S. patients further. These data will be critical for further understanding AMT-130. In the meantime, we're very encouraged by the trends, and we hope that the data provides hope for the HD community. The next steps for this program are completion of enrollment in our Phase I/II study in Europe, initiation of a small third cohort in the U.S. to explore the acute effect of immunosuppression, a clinical update later this year that includes the European patients and a meeting with regulators to discuss the path forward for clinical development of AMT-130. Now it's my pleasure to introduce Dr. Sarah Tabrizi, who has graciously agreed to provide her thoughts on these interim data. Sarah?

Thank you. So I'm Sarah Tabrizi, I'm a professor at UCL. And so I'll start by giving you a disclosure. I'm not being paid directly by uniQure, it goes to UCL my university consulting, and it's not going to me personally. So I'm a physician scientist, I've worked in the HD field for the last 25 years. And I'm here independently because I am dedicated to finding treatments for Huntington's disease. And this program interests me because it targets both full length and exon 1 mutant Huntington, which I think is important. . So I'm just going to comment on a number of the areas briefly. So, in terms of safety and tolerability, I do agree that AMT-130 is generally safe and well tolerated for this sort of gene therapy. And I think importantly, the independent into Safety Monitoring Board is supportive of no changes to the protocol. And as Ricardo mentioned, steroid cover will be tested to see if it mitigates any of the inflammation at both doses going forward. In terms of the clinical measures, these are small numbers of patients. But despite that, I am very encouraged that the clinical measures appear to be going in the right direction compared to the natural history data. And for me, that supports continued clinical development of this program and molecule, in my opinion. In terms of NfL and mutant Huntington biomarkers, so NfL is actually both a safety biomarker and the potential efficacy biomarker in neurodegeneration, and as evidenced by the recent first in ruling by the FDA. Here in this study, in terms of safety, it goes up with the surgery as expected and as the Ricardo showed. And in the low dose, it is back down at baseline at 12 months and below baseline of 24 months, which I think is encouraging. But the high dose, it has the same pattern, but it's slightly higher at baseline, and this is likely due to the higher viral low dose. The CSF mutant Huntington data is complicated, as Ricardo said. The CSF mutant Huntington assay, which I'm very familiar with, is a difficult assay. It has a high coefficient variation, that's actually about 30%, which means the variability within assays and within subjects and between batches. I think, the data does show evidence of target engagement. I think the challenge with mutant Huntington in the CSF is we don't know exactly, where it comes from and exactly what brain regions it comes from. And this therapeutic approach targets the striatum which is actually a small part of the brain overall. It's 20 grams out of a 1,300-gram brain. And so, we don't know how much of that contributes to the mutant Huntington we see in the CSF. In addition, as mentioned, CSF mutant Huntington has been shown to be variable with relation to clinical disease progression, and it's not a clear-cut relationship. So in my view, in the mutant Huntington assay data, these are small numbers, and I think reflect the variability of the assay for mutant Huntington. In terms of the imaging, there's no significant effect on whole brain volumes. The increase in ventricles, I think, is likely to be the neurosurgical procedure and some inflammation. And importantly, there is no associated accelerated whole brain atrophy. No ongoing increases in NfL, which would suggest neuronal damage and no clinical correlates of progression. So for me in conclusion, I think this is very encouraging interim data, which for me, supports the continued clinical development of AMT-130. I think the clinical and NfL trends are going in the right direction, and I look forward to seeing how the clinical program moves forward. Thank you.

Thank you, Dr. Tabrizi. We're now available to answer questions from our research analysts. Operator, please open the line. .

[Operator Instructions] Our first question comes from Debjit Chattopadhyay with Guggenheim.

So the 13% decrease in NfL, does that make a compelling case in front of the FDA that it's reasonably likely to predict clinical benefit? And number two, given the noisy mutant Huntington data, how are you thinking about powering the Phase III study, if the accelerated approval pathways are no go from the FDA?

Okay. Ricardo, do you want to answer that?

Absolutely. So let me just start with the first one. I mean, we're generally encouraged by the fact that NfL is declining below baseline. We haven't had conversations with regulators yet as to what constitutes a significant change. But of course, we'll be having these later -- early next year. When it comes to powering the confirmatory study, again, I think we will need to have conversations with regulators. It's clear that the mutant Huntington assay is very noisy. And to get really reliable data, we're going to have -- need more patients.

[Operator Instructions] Our next question comes from Paul Matteis with Stifel.

I had a couple of questions in relation with Dr. Tabrizi was saying and then just one question for the care team. Dr. Tabrizi points on delivery, there's a lack of clear effect on Huntington and the CSF suggests that you're really just knocking down the protein in the striatum and you're not getting that traveling the cortical areas that are much, much bigger like the animal data for [ deck ]. The second question just related to what Dr. Tabrizi said it. It seems like she alluded to an increase in ventricular volume. I might have missed that, but can you just talk about that? Is that a safety signal here or anything of concern? And then lastly, I think the one other question here is, while the clinical data looks pretty encouraging, does the lack of target engagement on mutant Huntington makes the dialogue with the regulators on a accelerated path forward potentially more at hill because it's a lot to explain?

Sure. Dr. Tabrizi, do you want to answer those first couple of questions?

Yes. So I mentioned the ventricles because -- I was brought up in the slide deck. And I think but I'm not concerned by the increase in ventricular size. I think, it's very likely to be the neurosurgical procedure, which a debris is produced during a neurosurgical procedure and some inflammation, which is causing the ventricles to become slightly bigger. There's no associated accelerated brain atrophy, there's no associated increase in NfL, and no clinical correlative of progression. In terms of the striatum and retrograde transduction to the cortex, the preclinical large animal pig data showed good distribution throughout the large animal brain. And I think, that was -- there was very nice preclinical data in this program in a large animal to support the program. The stratum is critically important in Huntington's disease, and I think the CSF mutant Huntington assay data is not reflecting the fact that this gene therapy is not getting to the cortex. I think, the CSF mutant Huntington data reflects issues with that assay. And so this is the reason why CHDI Foundation of putting a large amount of work into developing a Huntington Pet ligand, which I think will be important for the community. Does that answer...?

Yes. And Paul, maybe just the last part of your question, in terms of accelerating approval, we don't believe that mutant Huntington in and of itself is a surrogate endpoint that is registrable with the FDA. I think, we feel pretty strongly about that, particularly given the fact that in other clinical studies is not demonstrated to be correlative. And even in natural history studies, it has a very weak correlation with progression of disease. And this is less about the fact that it's the appropriate target and more about the fact that it's being measured either in the plasma or the cerebral spinal fluid and precisely what that means from what is going on in the brain is unclear. So our view is that neurofilament light, which has been far better studied and, of course, has been viewed by the FDA as a potential surrogate endpoint in conjunction with supportive trending in clinical data, can serve as the basis of an accelerated approval.

[Operator Instructions] Our next question comes from Joseph Thome with TD Cowen.

Maybe one for Dr. Tabrizi. In terms of we saw several efficacy signals today, UHD are at different measures and brain volume. I guess, when you're looking at a patient, which one of these measures is maybe most important to you when determining sort of the next steps in clinical care? And then, maybe for the company, obviously, there are difficulties in measuring this marker in the CSF, but obviously, with gene therapies as you did with hemogenics, you show kind of robust reduction that's well maintained to kind of elucidate the long-term clinical benefit. So, when you think about this going forward, is there another way to kind of show the predictive durability of knockdown outside of kind of looking at the CSF measure? .

So I'll answer -- thanks, Matt. I'll answer the question about the clinical measures. So, the underpinning of clinical care in Huntington's disease is the United Huntington's rating scale, which is what we do in clinic, which has the total motor score, the total functional capacity and a composite of that is the composite UHDRS, which also reflects cognition. And the Stroop Word and the Symbol Digit reflects cognition. So the measures in the study, the clinical measures in this study are exactly what we measure in clinic and what are important to patients, i.e., their motor neurological function, the daily activities of daily living function, their ability to work and how their thinking is. So the measures that are being used are very closely aligned with what matters to patients, and that's why we use them in the clinic.

Yes. And in terms of maybe the second question, I think, look, in the end, this is going to be driven. The view of the value of this product is going to be driven by the clinical data. Next year, we'll have up to 3 years of clinical data on a subset of patients and likely more than 2 years of follow-up data on more than half the patients in the study. And I think, our view is that if we continue to see meaningful suppression or stabilization of the neurodegenerative profile, as measured by neurofilament light and meaningful potential clinical benefits measured by some of the functional measurements compared to the natural history, that's going to be easier to elucidate, as we get further up in the follow-up. And that's the most important measures, whether it's payers, whether it's patients and whether it's clinicians. It's not going to be long-term suppression. And one of the interesting things we were talking about this earlier is that even with hemophilia, where you have a highly validated surrogate measure and Factor IX activity, what was very clear to matter to the regulators and to the patient community was not whether or not they had Factor IX activity at a certain level, it's whether or not they had controlled bleeding. That's what really we're here to do. And so that's what we're going to be focused on. And in the end, that's what's going to demonstrate durability for us.

[Operator Instructions] Our next question comes from Danielle Brill with Raymond James.

Sorry, can you hear me? Hello?

Yes, we can hear you.

Great. I guess I'm just curious when you look at these data. I know based on prior data generated from Roche, there's a hypothesis about needing to kind of spread the needle and not overshoot on Huntington knockdown because there may be some protected benefits with a wild type. I'm just curious, when you look at these data, your thesis has shifted at all and whether you think the low dose might actually be your optimal dose moving forward?

Yes. I mean, maybe I can answer that. I think, the -- I don't want to talk about another data set, but I think because it's important, the key thing, I think, with the Roche data set or that they were seeing a dose-dependent worsening of patients, right? We are clearly not seeing that, okay? We are clearly not seeing elevated levels of neurofilament light out to 24 months. . So I think that the hypothesis that maybe the higher dose is knocking down wild-type Huntington protein too much, just simply doesn't seem evident. If anything, we look at the clinical data, what is suggested although we only have 1 year of follow-up data is that the higher dose seems to be on a slightly better course clinically through that first year. So I don't think that -- our view on that hypothesis has changed, but I'm happy Sarah, maybe you can discuss your perspectives on wild-type Huntington.

I agree with everything you've said, Matt. I don't have anything to add. I completely agree. I think, the evidence on the clinical measures, which, as I say, despite phone numbers are trending in the right direction and also the NfL goes up with the neurosurgery and then comes down and at 24 months and the low dose group is below baseline. So, I agree with everything you've said. .

[Operator Instructions] Our next question comes from Ellie Merle with UBS.

Can you elaborate a little bit more on what makes the mutant Huntington assay just such a difficult or variable asset just to sort of help us understand some of these data points? And then, just beyond sort of the assay itself, is there maybe a biological explanation or hypothesis for why the higher dose might be increase in mutant Huntington levels? I guess, mutant Huntington to be lower [indiscernible] but higher the CSS and any sort of biological hypotheses around that. And then on durability, just how do you interpret sort of the increases in new Huntington levels out to year 2 at the lower dose? And I guess, how should we think about the durability of the miQURE platform relative to, say, conventional gene therapy in the CNS that sort of expressing a protein and maybe sort of what data you have on sort of the durability and long-term effect of the miQURE approach?

Sure. Maybe on the first question about just the assay and some of the challenges, Dr. Tabrizi, maybe you could address that because you had a lot of experience with it?

Yes. I'm happy to discuss this. So the assay is a Simoa assay, and it depends on a combination of 2 antibodies, 2B7 and MW1. And the assay is challenging for a number of reasons. CSF mutant Huntington is a very low doses in the CSF, femtomolar concentrations. Between batches, there has to be different protein standards used and the assay is a useful tool. However, it has a high coefficient variation which is roughly 30%, which is similar to the coefficient variation that you see in, for example, in a Western blot. So the assay is useful as a tool, but it has its limitations and its variability. And you can see that by the differences and even in the control data set. So these are very small patient numbers and you can look -- and the data is very variable. And these are -- there are between batch effects and interpatient variability. So I think making absolute judgments based on the CSF mutant Huntington assay directions, I think is -- I don't think it is warranted because I think the assay in these very small numbers, has these -- have these issues. And I think that is why, as I mentioned before, that CHDI are putting so much effort into developing a Huntington Pet ligand. The NfL assay and CSF, however, is different. It's -- and you can see that by the size of the error bars, same number of patients. But if you look at the error bars, the error bars are much smaller. And the CFS neurofilament assay is just really performs better, as an assay. And it's -- we know it in the Huntington community about the CSF mutant Huntington assay and a lot of work is ongoing to try and understand where the CSF mutant Huntington is coming from. But at very low levels of femtomolar concentrations, it becomes a challenging assay.

Thanks, Dr. Tabrizi. Maybe, Ricardo, you can talk about the second part of that question in terms of biological activity and durability?

Absolutely. So there are two questions there. One is are there any biological explanations? And we don't exactly understand the dynamics of release of mutant Huntington into CSF, and where it comes from. We are dosing the striatum and probably the cortex and thalamus as well. But there's a whole spinal cord and they're animal cells and they're all kinds of other cells that release mutant Huntington. So in principle, it could be mutant Huntington to being released from other parts of the nervous system that we are not influencing and that are obviously not relevant for clinical function because these patients are doing really quite well. And also there's no increase in neurofilament light chain, which is a key measure of normal injury. In terms of durability, again, I think it's important to not overinterpret that one point. I mean, if you look at it really carefully, you'll see there's only one point where it somehow seems to go back. It happens to be the last time point. We don't exactly know what's going on there. What I can tell you is that in the pigs, we have 5-year data. And in 5 years, we still continue to see suppression because neurons don't divide, we don't really think that there's going to be dilution of the AAV. And in addition to this, we know that we can continue to see the suppression in animals for at least for 5 years, probably for life. So, I hope that helps.

[Operator Instructions] Our next question comes from Joseph Schwartz with Leerink Partners.

I was wondering,, if you have any data on blood levels of mutant Huntington protein, and if you're measuring total Huntington protein in the CSF and blood. And what do these patterns look like across the doses? And have you already -- or will you be analyzing the potential association between the biomarkers of mutant Huntington protein and NfL and functional measures in this data set? And then, I have a follow-up.

Yes. So I can answer all of those. So first of all, Huntington in blood. Of course, our Gene therapy is delivered directly into the brain. So, it doesn't get into the blood. So we wouldn't expect to see any changes in the blood. Usually, people measure Huntington in the blood, they're measuring it in blood cells and -- which is a different assay and it's not completely relevant to our modality. So, I don't think we can do that. . We have measured total Huntington that assay is much less sensitive in Huntington assay. And the last question, the combination -- oh, yes, exactly. So what is the correlation? We have done that, and that's an excellent question. So of course, one of the first things we did when we saw that there was increase in Huntington in the high-dose patients. As we look to see, if there's any correlation between that and NfL or clinical function. And the answer is no. In fact, there may be a slightly negative correlation, which is probably not real. So there is no correlation between those excursions and anything else that we can measure.

[Operator Instructions] Our next question comes from Salveen Richter with Goldman Sachs.

This is Tommy on for Salveen. Wondering, if you could expand on mechanistically and biologically. What is the reason why the CSF NfL levels in the high dose cohort were more variable than the low dose? And why was there that favorable decrease for the low dose and then the increase for the high-dose cohort? .

So Ricardo?

Yes. So first, NfL. So in the high-dose cohort, there were a couple of patients that had 2 SARs and you can actually see that those patients had more prolonged elevations in NfL, which is what you would expect as they had more inflammation, and that took a longer time to come down. And that accounts for the slight difference between NfL -- NfL levels.

[Operator Instructions] Our next question comes from Kristen Kluska with Cantor Fitzgerald.

This is Rick on for Kristen. Just one, given the focus on clinical changes that you've talked about today, can you talk about the expectations around rate of clinical changes that you would predict seeing a natural history cohort for some of these measures? And what would this potentially suggest about how long you may need to follow patients for registrational trial?

Sure, Ricardo, do you want to answer that?

Absolutely. So the rate of clinical changes, we have spent a lot of time trying to collect the best possible natural history. And in Huntington's disease, we're really fortunate because we have among the best natural history cohorts in all of neuroscience. So, we partnered with [indiscernible] a major foundation in our field to develop really good natural history databases. And what we have shown, which is actually shown in our slides, is that we understand that in about 2 years, we see significant changes in this patient population. So, we think that something like a 2-year trial will be required. Of course, with Gene therapy, you have an advantage because once you dose people, you dose them forever. And so, we can continue to follow our patients for 2, 3, 4 or 5 years, which I think will give power to our trial that is not available to people using other modalities.

[Operator Instructions] Our next question comes from Sami Corwin with William Blair.

I notice you began using immunosuppression in your crossover patients and plan to use it in Cohort 3, as well. Could you elaborate a little bit on the rationale? And if, you're seeing any early differences in outcomes or safety with immunosupression and ultimately, you think that might help with durability or the degree of reduction in mutant Huntington?

Sure. Ricardo?

Sure. So why use immunosuppression? So of course, we had a small number of inflammatory SUSAR at the high dose and not at the low dose. And that really prompted us to introduce a corticosteroid regimen perioperatively that is designed to reduce that. We know that steroids are very effective at producing edema. And while we think -- it's probably too early to say how well this works. Thus far, the patients that we have dosed with perioperative steroids have done better. So we -- this is the main reason for really exploring this immunosuppression to see whether we can reduce the incidence of these kind of perioperative events that we saw.

[Operator Instructions] Our next question comes from Patrick Trucchio H.C. Wainwright.

I'm wondering for the third cohort in the second half of 2023, would you need the data from this cohort, in order to advance to the Phase III trial? Or would you be able to have those discussions with the FDA and other regulators before that? And then secondly, maybe for the KOL, just regarding the TMS and ESC, can you discuss what level of improvement would be considered clinically meaningful? And what would need to be demonstrated in the potential Phase III trial for approval?

Yes. So I'll answer the first question and then hand it over to you, Sarah. So the answer -- the short answer to the question about Cohort 3 is, no. We don't believe that we need to have the data from Cohort 3, in order to advance our regulatory discussions or into a registrational study. We do have -- we will have a total of 8 patients in Cohorts 1 and 2 that are utilizing immunosuppression therapy. So we will have, I think, a meaningful amount of experience of immunosuppression therapy, at least at the high dose. Part of why, we're doing Cohort 3 is just to get some additional experience at the low dose. And to the extent that we would desire an alternative immunosuppression regimen. That is another objective that we can potentially need within Cohort 3. The other thing that I want to focus on is that while, of course, when following the treated patients for an extended period of time, really the focus is on the near-term safety profile. So, this is going to be a relatively short-term follow-up for these patients, patients to look at neurofilament light in the adverse event profile. And then, I'll talk to back to you about the functional measures, Dr. Tabrizi.

Thank you. So, the clinical meaningfulness of these measures is a good question actually. So the total functional capacity, as you may know, is a functional score out of 13, and the subjects recruited in this study had a score between 9 and 13, which reflects their ability to work, ability to function, manage their finances. So the TFC has quite -- has a ceiling at 13. And in a study, if you have either a historical control group or a placebo group, in a Phase III, if you actually see slowing of any TFC decline compared to placebo, then -- and clearly significant, then that's meaningful for the patient. The TFC actually is a score that can go up, and it can improve, i.e., people can go back to their original job. But the nature of the total functional capacity means that stabilization tends to be what we look for rather than improvement. Although some people do go back to their original work, which is one of the key questions and regain more independence. In terms of the total motor score, again, compared to placebo, you're looking for slowing -- significant slowing compared to a placebo group or a historical control group with reasonable numbers and a statistically significant effect. But a total motor score delaying or even improvement of 1 to 2 points a year is definitely clinically meaningful. Do you want to go back to you, Matt?

I'm sorry. No, I think we've answered that question. So operator, you can take the next question. .

Okay. Our next question comes from Yanan Zhu with Wells Fargo.

Two questions, one on neurofilament light chain, the other HCC biomarker. For neurofilament light chain, I think the control arm also declined, is that the expected result? And for the HCC, I guess there's the focus on the assay and there's also a focus on whether biologically, maybe it's not -- it could be expected to show any -- to show a decline. I guess, how are you going to determine whether an assay, you need to figure out the assay before moving forward? And on the biological rationale, I was wondering, how do you reconcile what happens here with what happened in animal study, where I think you showed you should be able to show a decline in mutant Huntington?

Ricardo?

Absolutely. So let's just start with NfL. Yes, there was a very small decline in one time point in NfL in the placebo really only at the last 12-month time point. We kind of think that that's kind of noise in the assay. We don't really think that's real. We think that in terms of NfL it was largely unchanged. And it's, of course, less than what we are seeing in the low dose, where the downward trend continues. This is also true for the high dose. So we do think that the effect that we're seeing is the low dose is real. Okay. When it comes to the mutant Huntington assay, we have done an enormous amount of troubleshooting along with CHDI, along with multiple CROs, on this assay. I think, the reality of it is that we're trying to measure effective [indiscernible] and mutant Huntington is an intracellular protein. And therefore, the -- it will always be noisy. So I think, that we will be evaluating with us. This is really valuable going forward because it doesn't seem to really accurately reflect, what we would like it to measure.

Yes. I mean, I'll just add one point. I think that we have a high degree of confidence that where we deposit AMT-130, we're getting target engagement that is consistent with all of the animal studies that we have performed. I think the question is, can we detect that target engagement in the cerebral spinal fluid, right? And I think, as Ricardo mentioned, we've worked with multiple vendors on reading and interpreting the data that we have, we've reanalyzed the data, we've done a tremendous amount of work on this. This is not -- this is -- we don't believe, particularly given some of Sarah's comments, that this is particular to uniQure. This is an industry wide thing, and it might be exacerbated by the fact that we are administering it deep within the brain and we're measuring this outside of the brain. And that might be different from how other sponsors are evaluating their particular product candidates. But to say it again, we have a very high degree of confidence that what is going on in the brain is engagement of the target. And that to some extent, that is borne out in some of the data from the low-dose cohort.

I just want to add to that. I agree with that. Everything that's been I've said and everyone has said about the assay, I think the trends in the clinical data are encouraging because these are very small numbers. And so, the trends in the clinical data are certainly encouraging. And I think, continued follow-up and continued development and more numbers of patients is going to be really important. And I think the NfL data is encouraging as well. So I think the CSF mutant Huntington data for all the reasons discussed, I think, has the issue around the assay. Back to you, Matt.

Thanks, Sarah. Operator, I think we have time for one more question.

Sure. [Operator Instructions] Our last question comes from Debjit Chattopadhyay with Guggenheim.

For letting me back in again. So a couple of clarifications. The low dose data at 12 months, we had a 54% mutant Huntington knockdown. Given the variability in the assay and changes over time, if you were to go back and remeasure that, what would that number be? And I'm wondering, if you have measured that again. And number two, to what Dr. Tabrizi alluded to before, for retrograde transfer, wouldn't the high dose have been better than the lower dose? Wasn't that sort of the underlying principle to go for the high dose, to see if you can get better transfer hence better target engagement?

Yes. So Debjit, we have looked at the -- we have looked at and reanalyzed some of the data. What I would tell you is that generally speaking, the broader trends over time are consistent. So we do continue to see evidence of target engagement. But that when you look at any individual point in time or any individual point in time for a particular patient, you see a high degree of variability, which is why -- I think whether it's, I don't know, an area under the curve analysis or just looking at trends largely over time, that would maybe be the best that you could probably do with this assay trying to over-interpret a particular time point when you see it go, it's way below baseline on one time point and way above baseline in another. We don't think those are there's a high degree of fidelity, when you get to that level of specificity within the data set. And then, I think the second question was intended for you, Dr. Tabrizi.

Could you repeat the question, Luca? .

It wasn't from Luca. It's line -- that actually not in the queue anymore.

I think it was a -- I think the question was about -- in the higher dose that -- yes, in the higher dose, we absolutely did see in animals that you do get border transduction across the brain.

I mean that you did. In the large animals, you did see transduction across the brain in the large animal pig data, and that was published in the science translational medicine. I think the question relating to the high dose, will that get more retrograde transduction. I think, theoretically, yes. And I think that's something, as Matt has said, I think from your data, I believe, Ricardo and Matt, you believe that's the case.

Yes. I think we're -- we do believe that the virus gets structurally be able to transported as well as the micro RNA gets transported between neurons and the decibels. We don't really -- so you ask, well, given that, that happens at the high dose, why don't we see a bigger or a different effect, and we think then of course, we don't really know that. As I said before, it's possible that we in Huntington is coming from a different part of the brain or the things that make them the greatest contribution to Huntington's CSF, of course, of the [indiscernible] cells which line the ventricles as well as the cells that produce CSF, which is the core Plexus. And of course, we don't transduce those cells at all. So any changes there would also mask our effect. But yes, I mean, generally speaking, and in other ways, for example, clinically, the patients on the high dose are doing somewhat better than the patients on the low dose. So there is some dose response.

Okay. So, I think we're done with our questions. So I want to thank everybody for joining the call today. As a reminder, we're very pleased with these encouraging interim data on AMT-130. And look forward to advancing its clinical development and beginning discussions with regulators by early next year. Have a great day. Thanks very much. .

Ladies and gentlemen, that does conclude today's presentation. You may now disconnect, and have a wonderful day.