Wave Life Sciences Ltd. (WVE) Earnings Call Transcript & Summary

Good morning, and welcome to Wave Life Sciences SELECT-HD Clinical Trial Results Conference Call. [Operator Instructions] As a reminder, this call is being recorded and webcast. I'll now turn the call over to Kate Rausch, Head of Investor Relations at Wave Life Sciences. Please go ahead.

Thank you, operator. Good morning, and thank you for joining us. With me here today are Dr. Paul Bolno, President and CEO of Wave Life Sciences; and Anne-Marie Li-Kwai-Cheung, Chief Development Officer. This morning, we issued a news release announcing results from our SELECT-HD trial WVE-003. This release can be found in the Investors section of our website, www.ir.wavelifesciences.com. Before we begin, I would like to remind you that we'll be making certain forward-looking statements related to our current plans and expectations on today's call, which are subject to certain risks and uncertainties. Actual results may differ materially due to various factors, including those described in the Risk Factors section of our most recent Form 10-Q, 10-K and other SEC filings. These forward-looking statements represent our views as of this call and should not be relied on as representing our views as of any subsequent date. We undertake no obligation to publicly update or revise any of these statements. On today's call, Dr. Paul Bolno, President and CEO, will begin with opening remarks. Afterwards, Anne-Marie will review the SELECT-HD clinical trial results, and Paul will then close with a review of Wave's lead pipeline programs and near-term milestones. I'd now like to turn the call over to Paul.

Thank you, Kate. Good morning, everyone. We are excited to be speaking with you today to share the positive results from our SELECT-HD trial of WVE-003. WVE-003 is our first-in-class, allele-selective antisense oligonucleotide, which we are developing in a disease-modifying treatment for Huntington's disease. These results are historic as they represent the first clinical demonstration of allele-selective silencing of any disease target. This is a significant moment for Wave for the oligonucleotide field and for patients. In HD, the results represent the opportunity to bring forward a novel therapeutic approach for a devastating disease that has no disease-modifying treatments despite the cause of the disease being well-understood. Anne-Marie will speak about the disease etiology, our clinical trial results and why WVE-003 is best positioned within the field to change the trajectory of this disease. Today's positive data are also reflective of more than 10 years of platform innovation pioneered by Wave. Our proprietary chemistry, including stereochemical control and PN chemistry has enabled Wave to design highly selected candidates, which in HD has enabled robust engagement of a single SNP on a single allele. The potency, durability and selectivity of mutant HTT reduction observed in the study after just 3 doses is impressive and sets WVE-003 apart from others. The results provide further evidence of the strong preclinical-to-clinical translation we are seeing in our studies as well as the value gained through robust preclinical models and our leadership in translational medicine. To briefly highlight today's data, we have reserved potent, durable, and selective reductions of mutant huntingtin of up to 46%, which well exceeds the targeted 30% threshold expected to impact clinical outcomes. Multi-dosing with WVE-003 was also generally safe and well-tolerated. We confirm the allele selectivity of WVE-003 with our data showing preservation of healthy wild-type protein. Ventricular volume was in line with natural history, an important finding as pan-silencing oligonucleotides have accelerated ventricular volume expansion in prior studies. We also observed early signals of positive clinical effects, which was not expected at this early time point and provides us with added confidence in our allele-selective approach. Specifically, we saw a statistically significant correlation between mutant huntingtin lowering and slowing of caudate atrophy, which is an imaging biomarker predictive of clinical outcomes in HD. We also saw point estimates favoring WVE-003 on total motor score, another encouraging signal that we did not expect at this early stage. Now I'll turn the call over to Anne-Marie to provide further details on the study results and next steps to the program. Anne-Marie?

Thank you, Paul. HD is a devastating disease, and patients today have no disease-modifying therapeutic options. HD is a monogenic autosomal dominant disease that is fully penetrant and affects the entire brain. In the U.S. and Europe alone, there are approximately 160,000 presymptomatic and 65,000 symptomatic patients. HD is driven by both a toxic gain of function by mutant huntingtin protein and loss of function of healthy wild-type huntingtin protein. Wild-type huntingtin is essential for the proper function of the central nervous system. As you can see on the chart on the right, the burden of disease is detectable more than 20 years before clinical diagnosis, with patients losing around 2% to 4% of caudate volume per year. When clinical symptoms emerge, they often start with subtle changes in cognition and motor function, which can be measured by outcomes such as Symbol Digit Modalities Test, or SDMT, and Total Motor Score, or TMS. With allele-selective knockdown of mutant huntingtin comes a unique opportunity to intervene earlier in the disease course to delay disease progression. Wild-type huntingtin is a scaffold protein that sees the formation of numerous protein complexes, including those that support protein and organelle trafficking, as well as transport of vesicles containing brain-derived neurotrophic factor, BDNF. It plays a critical role in multiple other processes fundamental to neuronal homeostasis and health. But patients with HD have decreased levels of wild-type huntingtin because of the CAG repeat expansion mutation. In addition, mutant huntingtin has detrimental effects on wild-type huntingtin at the protein level. So there is further decrease in the amount of functional wild-type protein in neurons. The selective lowering of mutant huntingtin is expected to have a dual benefit. First, it will decrease the toxic effects of mutant huntingtin; and second, it will release sequesterd wild-type huntingtin from aggregates, as well as from functional interference by mutant huntingtin at the protein level, thereby increasing functional levels of wild-type huntingtin protein. Cilia in the CNS illustrate this point. Wild-type huntingtin plays a critical role in the formation and function of celiac, which control the flow of CSF and help maintain homeostasis in the CNS. In Huntington's disease, apparent protein interactions with mutant huntingtin results in long disorganized cilia that can compromise CSF flow and neurogenesis. Depletion of mutant huntingtin has the potential to improve these phenotypes. In contrast, further depletion of wild-type huntingtin through pan-silencing has the potential to exacerbate it, further disrupting CSF flow, causing an increase in ventricular volume and hydrocephalus. 003 is a first-in-class allele-selective oligonucleotide. To selectively and specifically target the mutant huntingtin target, 003 targets a single-base difference, which we call SNP3, that exists on the mutant huntingtin allele, co-segregating with the disease-causing CAG repeats. Such precision, along with the potency and durability, is enabled by our unique and proprietary chemistry, including our stereochemistry and PN chemistry. We have leveraged our preclinical data to predict a therapeutic dose and will show this translated in the clinic successfully. I'll now turn to the results of our SELECT-HD clinical trial. I'll start with a recap of SELECT-HD, which was designed to demonstrate a favorable safety and tolerability profile for WVE-003, pharmacokinetics and CSF, potent selective and durable mutant huntingtin reductions of greater than or equal to 30% in CSF, and to confirm allele selectivity with preservation of wild-type huntingtin. While the study was not powered for clinical effects, we included several exploratory assessments. These include MRI measurement of caudate tissue to assess the rate of caudate atrophy and known imaging biomarker that is predictive for clinical outcomes and multiple clinical measures, including TMS, an important measure that changes early in the disease process. The SELECT-HD clinical trial included 2 phases, a single-dose phase and a multi-dose phase. The trial included a single ascending dose portion that evaluated doses of 30, 60, and 90 milligrams, and a multi-dose phase, which evaluated 3 30 milligram doses of WAVE-003 administered every 8 weeks. Biomarker CSF samples as well as MRI and clinical assessments were performed at day 169, or 8 weeks post last dose. We also included a follow-up time point at day 196 to assess the durability of mutant huntingtin knockdown at 12 weeks post last dose. I'll begin with an overview of baseline characteristics. Baseline characteristics including age, gender and CAG length were generally balanced across cohorts for both the single- and multi-dose portions of the study. Cohorts were also well-balanced on the ISS disease stage at baseline. Now moving on to safety. In the single-dose portion of the study, AEs were well balanced across the 4 cohorts: 3 active and 1 placebo. The most common adverse events were headache and back pain. There was 1 severe and 1 serious AE in the placebo cohort and 1 serious AE in the 60-milligram cohort, and 1 severe AE is eating to withdrawal in the 90-milligram cohort. In the multi-dose portion of the study, adverse events were also balanced between the placebo and the 30-milligram cohort and all mild to moderate in intensity. The most common events were headache, falls, and fatigue, which were balanced between the treatment arms. There was an imbalance in treatment-related adverse events that were also mild to moderate in severity. Mild events included hyporeflexia and tingling, and moderate events included CSF abnormalities, including protein elevations. As mentioned earlier, ventricular volume is an important safety measure as ventricular enlargement and hydrocephalus have been seen with pan-silencing on the oligonucleotides. Importantly, ventricular volume was consistent with natural history in our study. There were no cases of hydrocephalus. I'll turn now to the target engagement markets, including mutant huntingtin and wild-type huntingtin levels as measured in CSF. Shown here are the results from the single-dose phase of the study. With a single dose of WVE-003, we saw robust mutant huntingtin silencing, as shown on the left, which persisted at 12 weeks, with 41% silencing relative to placebo at the 30-milligram dose. Notably, although we did not see a dose response, we saw consistent silencing across all treatment cohorts. Across all doses, we saw consistent preservation of wild-type huntingtin, as shown on the right. Turning to the multi-dose portion of the study, we see that mutant huntingtin reductions continue to deepen the repeat dosing. And we observed peak mutant huntingtin reductions or 46% versus placebo at day 169, or 8 weeks following the last dose. Reductions were durable, with mutant huntingtin reductions of 44% versus placebo out to day 197, or 12 weeks past last dose, supporting the potential for quarterly dosing. As we look to the selectivity in the multi-dose portion of the study, we again saw consistent preservation of wild-type throughout. Interestingly, we see some statistically significant increases in wild-type huntingtin protein, which, as you recall, is associated with neuroprotection. Consistent with the literature on natural progression of neurofilament light chain, or NfL, in HD, we also saw that patients in our study experienced increases in NfL over time. On this slide, we depict the patient level NfL data for WVE-03 multi-dose cohort. The WVE-003 treated patients are depicted in each line, while the gray area shows the 5th to 95th percentiles. As you will see, the majority of 003 participants trended with placebo, there were 5 participants who had elevations in NfL that exceeded the placebo range, 2 of which was still elevated at the end of the observation period. We recognize that quarterly dosing may mitigate NfL increases in future studies. With robust and allele-selective reductions of mutant huntingtin, we would expect over time a slowing of caudate atrophy, an imaging marker predictive of clinical outcomes. Encouragingly, at the week 24 time point, volumetric MRI assessment of the rate of caudate atrophy appears to favor treatment with WVE-003, with patients experiencing 4.68% loss versus 5.1% with placebo. While this study was not powered to detect a significant sequence for this measure, and it was not statistically significant, a significant correlation was observed between allele-selective mutant huntingtin reduction and the rate of caudate atrophy, which is the first demonstration of such correlation in the clinic. This is very exciting and increases our confidence that WAVE-003 treatment can translate to clinical benefit. Turning to clinical assessments. As planned, we evaluated several clinical outcome measures. Again, we were not proud to detect change with this study. As such, we were encouraged to see a favorable point estimate in the Total Motor Score. While this was not statistically significant, it was intriguing given that TMS is a sensitive measure to detect early functional benefits. There were no detrimental effects on any clinical measure. This exciting clinical trial results have not only demonstrated the potential of our platform to achieve selective mutant huntingtin silencing, but also confirmed a favorable safety profile, including ventricular volume, consistent with natural history. We are also incredibly encouraged that target engagement correlated with slowing of caudate atrophy. Our robust statistical analysis of natural history data supported by the literature bolsters the case for caudate atrophy as an early predictor of clinical outcomes, opening the door to using this imaging biomarker to enable an accelerated approval. We have plans to engage regulators before year-end 2024. I'll now turn it back over to Paul.

Thank you, Ann-Marie. It is incredibly exciting to see the continued translation of our platform and preclinical data in the clinic. In addition to the potent and durable allele-selective silencing achieved after 3 doses of WVE-003, we have also previously demonstrated high-muscle concentrations and the highest amount of exon skipping ever reported in DMD. Like in HD, these data were also reported after just 3 consecutive doses. This compelling clinical evidence grows our confidence for our broader pipeline ahead of multiple upcoming milestones. In the third quarter of this year, we expect 24-week dystrophin data from the FORWARD-53 study of WVE-N531. If positive, these data would enable an accelerated path to approval in DMD. We also expect proof of mechanism data for WVE-006 and AATD this year. This would be the first-ever clinical data for RNA editing and will be a meaningful for derisking the program and our broader RNA editing pipeline. We also are rapidly advancing our novel siRNA program targeting INHBE obesity, which is now called WVE-007, and we continue to be on track to initiate our clinical trial in the first quarter of 2025. Together, our lead programs span multiple mechanisms, cell types and disease areas, giving way to multitude of opportunities to bring value to our shareholders as well as to patients. With the continued translation of our lead programs, we expect to address more than 50 million patients in the U.S. and Europe alone. Success in any of these programs would unlock additional pipeline opportunities, including additional exons in DMD, SNPs and HD, hepatic and extrahepatic RNA editing in AMR programs, and hepatic siRNA programs. We are also actively planning for R&D Day this fall, where we expect to provide updates on our growing wholly-owned pipeline. To close, this is an incredibly exciting time for Wave as we continue to advance our mission to unlock the broad potential of RNA medicines. We look forward to sharing more updates with you very soon. And with that, we can open up the call for questions. Operator?

[Operator Instructions] Our first question comes from Joon Lee with Truist.

Congrats on the strong data. When do you plan to have a meeting with the FDA? And would you be asking for caudate imaging as a surrogate endpoint? And if so, how confident are you that the FDA would be willing to accept imaging as a surrogate endpoint? And I have a follow-up question.

Thank you, June. I'll let Anne-Marie take the first question.

We plan to align with the regulators by the end of this year. As you've heard today, caudate atrophy is an imaging endpoint, which is predictive of clinical outcomes, and accelerated approval is contingent on endpoints, which are likely to predict clinical benefit. It isn't required to be a surrogate, but likely to predict clinical benefit. And caudate atrophy, we believe, meets that threshold. So we're very encouraged, and we're looking forward to discussing these data with regulators.

Yes. I think just to add on to that, I mean, as we've talked about before, Joon, and with others, I think the community at large, patients and clinicians, are looking for a clinical surrogate endpoint and has been recently shared where there has been data looking at as Anne-Marie referred to earlier in natural history that there has been looking at imaging data from Track-HD and others, that you can correlate caudate nucleus change with clinical outcome measurements like TMS, CHD, URS. So I think the convergence at this moment in time where in the absence of disease-modifying therapies and a desire to find alternative endpoints using clinical biomarkers and imaging biomarkers, I think we're in an interesting time with imaging and clinical endpoints crossing, and I think it creates a very unique opportunity for us to bring that forward. I think what the agency is looking for is in medicine with which to bring it forward. So I think we've got a great profile.

Great. And looking forward to the updates on that. And then yesterday, COO of Alnylam, Yvonne Greenstreet, stated at the end of her prepared remarks that they're emphasizing on obesity, diabetes, and metabolic disorders as their focus areas. I can only imagine she's referring to their INHBE program. How does today's data from you guys help derisk your plans for the INHBE program? And where are you with your IND-enabling studies that?

No, it's great to hear. I think it's an exciting time in metabolic diseases and obesity, and particularly the genetics related to that. So I think what today represents is the fact that the chemistry is translating. We're seeing that in the clinic, we saw that earlier with our DMD data from the initial Part A portion of the study, and we're excited for the 6-month dystrophin data to continue to translate. But I think we can look back today and say chemistry is translating. As we apply that forward to a GalNAc-conjugated siRNA, where we have already shown potent durable reduction of the protein, we see weight loss similar to supaglutide, we're preventing the rebound weight gain following withdraw. I think we're excited about that program and accelerating it rapidly. Currently, the preclinical IND-enabling studies are on track. And as we said on the call, we anticipate initiating the study in the first quarter of next year.

All right. And I'll hop back on the queue.

Our next question comes from Tiago Fauth with Wells Fargo.

Just wanted to follow up on the NfL. Wondering if there were any associated increases in white blood cells or proteins or any other size of information that correlated with that. You also mentioned that quarterly building might prevent me some of that NfL increase. Is that something that you would potentially explore on the registrational trial or during the open-label expansion? Wondering how you're thinking about dosing optimization going forward.

Anne-Marie, do you want to start?

Yes. I think it's difficult to, at this stage, show that NfL is a safety biomarker. So for instance, in our study, we had a patient with high NfL, but had the best outcomes on TMS. However, I think your point is well-taken. From the PK/PD data, we can see that quarterly or less frequent dosing would be possible, and that's something that we will consider moving forward to mitigate risk of NfL, if that's necessary.

Completely agree. I mean I think there's additional data, too, which is intriguing that NfL a sequester protein inside the huntingtin aggregate. So just as we saw corresponding increases in wild-type huntingtin as Anne-Marie said, there's also a very distinct possibility that NfL over time is built up in aggregates and therefore, released. So I think as Anne-Marie noted that NfL is not a straightforward predictive safety biomarker. And I think what's more encouraging to us is to see clinical improvements in patients and particularly looking at Total Motor Score to look at caudate nucleus slowing in atrophy and so real measurements of progress of patients.

Next question comes from Steve Seedhouse with Raymond James.

Congratulations on the very encouraging results. First of all, what I wanted to ask about was I just -- I thought there was some commentary on the recent Huntington's conference actually about accelerated approval pathways and rethinking that. And so I think you're approaching this with some existing understanding already of what a path might be even before you go to FDA. Can you just talk about that, what you already know or what you heard about just the evolution of this in Huntington and the availability of the path?

Yes. Anne-Marie, do you want to start?

Sure. Yes, I think you're absolutely right. There's a real convergence of opinion in the KOL and patient community that it's absolutely imperative that we are now designing efficient and smaller and faster trials using sensitive markers of disease. And in general, there's a lot of coalescence around the concept of the totality of data. So seeing things like knockdown of mutant huntingtin and in our case, also preservation of wild-type and slowing of caudate atrophy together could be an extremely compelling package. This is something that CPAP and a group called HD Risk, which is a consortium including industry, have been working on together. And while there's lots of different endpoints available, this combination does look to be very promising.

Okay. And just a couple of questions on the data, or 2 questions on the data. So the TMS increase that you're seeing in the placebo arm, it seems like it might be tracking a bit higher than natural history based on available data, at least an old JML paper we could find. So I'm curious if that's a reflection you think of the eligibility criteria for this study or was sort of rapidity of the TMS increase in the placebo arm a bit surprising?

I mean it's a somewhat small study. And so there can be fluctuations compared to the natural history given it was 24 patients in -- or, 23 actually, treated in the multi-dose phase. So whereas it might be a bit higher than natural history. I think it's more or less reflective of what we would expect to see of patients with this stage of disease. And the comparison with placebo, but the difference between the 2, I think, is what's really interesting. Now as we said, it's not statistically significant. But in combination with what we've seen on caudate atrophy and the statistically significant correlation between caudate atrophy and mutant huntingtin knockdown, these are really exciting results.

Yes. It does look like a pretty large delta. The last question I just had, just on the NfL increase I just wanted to pin this down, your hypothesis on the possible mechanism there for the increases transient or otherwise. Do you think -- is this like an inflammatory response to the oligo you think or something else?

Yes, I think it's hard. There's not really a correlation where everybody gets a spike has increases in protein. So I think this idea that it's purely related to an inflammatory approach, I don't think is necessarily the case. I think everybody in the field has been watching it for a while. So I think we can't ignore it. We watch it. We follow it. But I think this idea around NfL, where there have been papers showing NfL trapped inside a sequester in these aggregates is also a hypothesis. So I think we've got a multitude of competing hypotheses around it. I think we went back and looked at the publication following the GENERATION HD study. That study showed no correlation between NfL and worsening in the clinical progression. So I think that was a very large study. So I think we just have to continue to watch it. But I think as you saw today, it's not getting in a way of clinical efficacy.

Congratulations on the really promising results.

Thank you.

Our next question comes from Salim Syed with Mizuho.

Great. Congratulations on the data. I guess a few from us, Paul. First, can you just answer the question whether -- have you ruled out -- can you file on this particular data set, or is it definitive that you will need to run another clinical study even for an accelerated approval, like maybe just answer that question all have some of the data?

Yes. I think our next step is to engage regulators and have that discussion. I think these data are obviously encouraging, and I think are supportive of the study, but I do think there's a discussion around the imaging-based study to see that. I think what's great about an imaging-based study based on this natural history is it's a relatively short time frame to unlock that potential, as Anne-Marie shared, within 12 to 18 months, you can see changes in our MRI imaging.. So I think we need to engage regulators, discuss data and discuss next steps.

Okay. And has there been any discussions previously on how you would think about potential size of the study or a safety package?

I think that's what we would engage. So the key is making sure that, that study is powered to deliver. And as we've said, there's about 77 patients in the treatment arm. So you're looking at a study that's a lot smaller than having to run a clinical outcome study. But again, those data and that size and that design needs to be agreed and aligned with regulators, which is the next step of the program.

Okay. Cool. And then just on the caudate atrophy piece here, just given that you guys have noted that you're doing work on is predicting clinical outcomes, just could you maybe quantify for us exactly what is the relationship that you guys are seeing? How you draw the relationship between atrophy and the slowing of decline in TMS?

So at the sort of natural history level, it can be seen that baseline caudate atrophy correlates both with TFC disease progression and things like SDMT and TMS. So through the work of the community and the KOLs who created the new staging system, it can be very clearly seen that caudate atrophy changes early in the disease course, 20 years before clinical diagnosis and continues through the disease stage. And what we have seen in our study is an apparent slowing of caudate atrophy compared to placebo, again, not statistically significant, but the correlation with the mutant huntingtin knockdown, which was statistically significant, I think, really provides evidence that the knockdown of mutant huntingtin with the sparing of wild-type has this opportunity to slow this worsening that occurs during the disease course.

Just to add to that, I mean, we talked about in the Huntington's field for a long time. This correlation that increased levels of huntingtin are drivers of progression of disease. So I think the correlate was always something that -- that's why the therapeutic target of mutant huntingtin reduction we is so important. And I think these data are still important. I think the first time we're showing, to Anne-Marie's point, this correlation that now reducing mutant protein actually correlates biologically the slowing of caudate atrophy, and then the signal in TMS. So I think we actually now have a very linear way to think about the disease not just in the way of how it progresses, but actually in how do you slow it and you ultimately reverse it. And I think the opportunity that we have, I think, uniquely to preserving wild-type may be the way and why we actually are able to discern mutant protein reduction and these changes despite others having demonstrated mutant protein reduction. And I think what's compelling as we think about paths forward into the future is all of these studies right now are what used to be called the manifest setting or this later-stage Stage II and III setting. And the opportunity that's really in front of us for an allele-specific therapy is being able to really address the disease at the beginning, stage 0 and 1, and therefore, really have an impact before neurodegeneration kicks in.

Okay. And just lastly, Paul, just can you remind us on manufacturing where you guys are to the extent that you get a path here that's relatively quick, just where you are in manufacturing for 003 and perhaps just generally?

Yes. I mean as we've said, across multiple programs, one of the investments we made years ago at Wave was building our GMP manufacturing capability. So we've been able to manufacture supply for our studies. Again, this is small intrathecal administration. And I think the potency is obviously -- is wonderful to think about the doses and volumes needed. So we've been able to support studies. We'll continue to do that. And I guess the key right now is next step is engagement with regulators around the plans for next steps, but we're able to produce medicines internally.

But regarding commercial supply, is that something that's on the radar?

Commercial, it's -- we have the opportunity and partnerships to be able to scale. We've done that as one remembers a while ago where we had a program, Suvodirsen, where we were able to build and scale for DMD at large scale and quality. So we are able to transfer our processes. I think the step right now is take these data which are exciting and engage regulators on the plans and paths for the next step.

Congrats, again.

Thank you.

Our next question comes from Luca Issi with RBC.

Great. Maybe the first on regulatory. Again, I appreciate you have a strong case here for pitching the FDA potential for accelerated approval based on imaging biomarkers here. But what is the plan in a scenario where the FDA does not let you do that and actually ask you to run a trial with UHDRS as the primary endpoint as they did with Roche? Would you indicate a still move-forward in that scenario? Again, any color much appreciated. And then maybe quickly on NfL. Can you just talk about the kinetics of the elevation of NfL? It looks to me that the numbers trend upwards over time and you only obviously dose 3 times while I'm assuming you will need those much longer for the Phase III. So just wondering how I should think about that? How confident are you that quarterly dosing will spare that signal? Any color much appreciate.

Yes. No, I mean, to take on your first question, very clearly, our plans are focused on bringing forward a potential accelerated registration pathway. And based on the conversations that are emerging in the field, we think that's very much in line with the discussions that have been had. I think as it relates to running larger studies, we're taking a disciplined approach to that accelerated registration pathway. Obviously, as you mentioned, we also have a collaboration with Takeda. And so the next step, and it's an important next step, is submitting the opt-in package to Takeda, which we're on track to do following these data sets. So I think we will holistically look at that, but our focus right now is assembling these data, engaging regulators, and I think aligning on a productive discussion about an accelerated pathway. As it relates to NfL, and your comment about trending off, I think it is important that when we also looked and followed those placebo patients, as Anne-Marie said earlier, they trended in the same way. So they're following this natural history trend up. There were those few outliers, I think it was 4 or 5 that came up. There's not really as you could see any characteristics that were unique across those, including some coming down despite dosing. And I think that example of patients with, and this has been seen with other medicines. As you dose, people can dose through NfL and see it come down. So I think at this point, there hasn't been a clear consistent thesis on why how to correlate NfL and dosing.

Our next question comes from Joseph Schwartz with Leerink Partners.

Congrats on the exciting results, Wave team. I was wondering how did the patients treated with WVE-033 -- or, 003, perform on other functional measures such as UHDRS, Group and TC? Can you remind us which of these functional measures have been associated the most with changes in caudate atrophy in patients with different stages of HD in natural history?

So on the other clinical outcomes, patients did no worse than placebo, which I think is really important because in other programs there have been early emerging signals of patients starting to do worse on these kinds of outcomes. So that's extremely encouraging. As I mentioned earlier, for sure, in the natural history, correlations with caudate atrophy with things such as Total Functional Score, which is the endpoint that FDA really anchors on overall and things like SDMT and TMS that are early markers of clinical progression for patients once they enter the clinical phase of disease.

I think if you look back, and there is a graphic that Anne-Marie shared about the kind of where early signals come from. So again, sensitivity measures. TMS and caudate atrophy, both align with that early transition from early stages of disease to later. So those would be the ones most expected with early signal. So at 6 months after 3 doses, it would have been unanticipated, as we said at the beginning, to see any reads on those endpoints and the fact that we could see those early sensitive measures, which are correlated. Caudate atrophy and TMS, both aligning, I think is extremely encouraging as the study.

Right. Yes, that makes sense. So it looks like you are planning to study 003 in Stage I and II patients in a registrational study. I was just wondering if you could expand on those thoughts relative to the predominantly Stage II and III patients studied in SELECT-HD? And then also discuss your thoughts around a 12- to 18-month time frame. Can you lay out your thinking about how that's likely to be sufficient? And are you thinking about a fixed duration, or could you employ an interim analysis?

So in neurodegenerative disease, it's generally accepted that intervening earlier in the disease course is beneficial to try and slow and halt the disease before there's irrevocable damage and loss of neurons. So what we're seeing today in SELECT-HD, which was predominantly Stage II, III, I think it's actually really encouraging because these patients are a bit further in their disease course. And the beauty of WVE-003 is that we have the ability to go earlier in the disease course because of the allele-selectivity, patients who are in Stage I and II, particularly I, are yet to have any clinical signs of disease. And so given that wild-type huntingtin is critical in neuronal homeostasis, the ability to have this allele selectivity gives you that opportunity to go earlier. So I think this is something that's very unique to us. But in general, the field is trying to move into earlier studies. And these are also based on learnings from other disease areas such as Alzheimer's disease. The 12- to 18-month time duration of the study is really driven by the sensitivity of the endpoint to change. So because caudate atrophy is a sensitive measure of disease progression, this gives you the ability to design a study which is smaller and shorter to enable you to detect a treatment effect. You might have to repeat your third question.

Yes, I was just curious if you're thinking about a fixed duration for maybe using an interim analysis?

Yes. So with the design of this study with looking at caudate atrophy as the primary endpoint, the 12- to 18-month study would allow you to detect statistically significant outcomes based on caudate atrophy. And we believe that, that in itself would be fileable. Of course, with accelerated approval, you do have to do a confirmatory study, which would look at longer clinical outcomes. And that we will align with regulators on that pathway.

Our next question comes from Joon Lee with Truist.

Paul, is preservation of caudate size and elevation in neurofilament light chain compatible? I mean if neurofilament light chain is indeed elevated, and shouldn't that lead to more progressive volume loss, and just trying to understand what is real here in the data? And then the follow-up question -- the second question is, you mentioned that the ventricular volume was in line with natural history, seems like a very important safety data point. Can you elaborate a little bit on that and what you've seen in other studies in Huntington's disease with regards to the volume [indiscernible].

Yes. I think what's ultimately important is that there isn't yet a firm understanding of NfL. And I think that's key. I mean, I think the NfL can -- we've seen NfL come up and down in other studies, but clinical outcomes do worse, right? So I think the most important thing and what's most encouraging about our data is the directionality of all the data sets. We're removing the toxic protein that actually is responsible for killing neurons and doing that substantially. We're increasing neuroprotective protein and wild-type, and we're seeing that translate those biomarkers into, as you said, not just correlation with slowing caudate atrophy, but now starting to move clinical measurements in a favorable direction. So when we put that totality of data together, I think it really does suggest the driver is mutant huntingtin that ultimately can drive disease progression. I mean as Anne-Marie mentioned as well, we have a setting of NfL going up and patients actually doing better on Total Motor Score. And we also know this idea sequestration of wild-type protein in the aggregates is a compounding variable. So I think holistically, I think the field will continue to watch NfL. But I think what we're going to continue to watch are really sensitive physical measurements like caudate nucleus volume on MRI, which is sensitive and you can watch that and continue to see that improve, clinical outcome measurements, and ultimately an impact on the toxic species, the mutant huntingtin aggregate and enhancing wild-type. As it relates to ventricular volume, I mean one of the challenges you need large sample size is to look at changes. So I think, again, what's encouraging is we can compare ventricle volume to natural history and show that there's no change. So I think it's important to us that, that's something that we do see distinctly different than what's been seen in pan-silencing. So again, you put all of the biomarkers together, mutant reductions, but really critically, wild-type sparing. And I think we, today, have really demonstrated that people need to be focused on understanding the wild-type protein because it is, we believe, helping to continue to drive the benefits that they've seen. It's not just a toxic gain of function disease, but the toxic loss of function, we think, is an important contributor, ultimately, in driving a neuroprotective effect.

Great. Well, thanks for the clarification and congrats on the data, again. Very strong.

Thank you so much.

We'll take our last question from Tiago Fauth of Wells Fargo.

Just a quick one on the Takeda opt-in. Can you just remind us on the timing or structure once you submit the opt-in package? Is there a clock time when they need to make a decision? And would you expect it to be conditional on some regulatory input or alignment? I'm curious how that might play out over the next few months.

It's great. And I will say, I think Takeda was incredibly excited to share the data. As clinicians and scientists when they saw, I think the field recognizes these were important data sets. So I think what's important is we will submit the package these data to meet the criteria for us, submitting our package for a potential opt-in. As you pointed out, there is a clock -- I can't disclose the timing on that clock, but there is a period with which they have to review. That's a big time period. There's no condition on a regulatory approval or alignment, that's independent. What's important in the agreement, even when they opt in, is then this collaboration shifts to a 50-50 R&D split, profit split. And the opt-in payment importantly offsets our contributions to that 50% of our obligations on R&D expenses. And then there continues to be subsequent milestones during the progression. Also importantly, we control the development. We're the development leader for this program through final registration with FDA. So in terms of communication and time line and progression, we would retain that. And we've retained in the U.S. commercial and scientific MSLs and the like. So it's a substantial collaboration on the other side of a potential opt-in.

I'd now like to turn the call back over to Paul Bolno for any closing remarks.

Thank you, everyone, for joining us today, and thank you again to the Huntington's disease community for your trust, support, and collaboration, as well as to our team at Wave for your fierce commitment to reimagine possible for patients. We look forward to speaking with some of you after the call. Take care. Have a great day.

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