Acumen Pharmaceuticals, Inc. (ABOS) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to the Acumen Pharmaceuticals Investor Day conference call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your first speaker today, Alex. Ma'am, please go ahead.

Thanks, Michelle.I'm Alex Braun, Head of Investor Relations. And on behalf of the Acumen team, I'd like to welcome everyone to our virtual Investor Day. Today, we intend to take you on a deeper dive into Acumen's value proposition as we near the Phase II readout for our product candidate for the treatment of early Alzheimer's disease, sabirnetug. As a reminder, we will be making forward-looking statements. These statements are subject to risks and uncertainties that may cause actual results to differ materially from those projected. A description of those risks can be found in our most recent 10-Q filing with the SEC. Any forward-looking statements are only as of today's date, and we assume no obligation to update any forward-looking statements made on today's call. Today's speakers will include Dan O'Connell, our Chief Executive Officer, who will briefly touch on the Alzheimer's landscape and potential of a-beta oligomer therapies to treat Alzheimer's disease. Dr. Jim Doherty, our President and Chief Development Officer, who will explain sabirnetug's mechanism of action and why we believe it could potentially differentiate from other disease-modifying therapies. Dr. Eric Siemers, our Chief Medical Officer, who will review our ongoing Phase II study investigating sabirnetug that reads out in late 2026. and Dr. Paul Chhagreu, VP, Program Lead and Head of Research, who will explain the latest developments in our enhanced Brain Therapy or EBD program, which we believe to be a promising next-generation addition to our pipeline. Following their presentations, there will be a live Q&A discussion. And in addition, there is a question submission box on your screen that will be open throughout the presentation this morning and during the Q&A session to submit written questions. We'll collect those questions throughout the hour and take as many as time allows. And with that, I'll hand this over to Dan.

Good morning. I'm Dan O'Connell, CEO of Acumen. Thank you for joining us today. I'm pleased to provide you with an overview of our opportunity and the setup for Acumen and the value proposition and strategy as we seek to bring better treatment options forward for people impacted by Alzheimer's disease. Let's start with the magnitude of the opportunity and unmet need. Today, an estimated 7.2 million Americans are living with early Alzheimer's disease, a stage characterized by symptoms and confirmed amyloid pathology. Absent a cure or effective preventative treatment, that population is expected to nearly double by 2060 to roughly 13.8 million Americans. For the current treatment landscape, 2 FDA-approved disease-modifying agents are now available and have been shown in clinical studies to slow disease progression. These have ushered in a new era for treatment in Alzheimer's disease. Global sales of the approved anti-amyloid agents are growing and poised to accelerate. The approved agents, LEQEMBI and KONSumLA have recently achieved an annual sales run rate exceeding $1 billion. Sales are projected to reach $2 billion in 2028 and reach upwards of $3 billion to $4 billion by 2030. The early market adoption for these amyloid plaque-directed agents has occurred despite systemic infrastructure challenges and debates about the overall efficacy and safety of these agents, including amyloid-related imaging abnormalities, otherwise referred to as ARS. Nevertheless, we see this market is poised for continued growth and large enough to support multiple multibillion-dollar products given the magnitude of the population. One of the primary drivers of market expansion and continued growth in development is the proliferation of blood-based biomarkers, essentially a blood test to confirm or inform a diagnosis of Alzheimer's disease. The use of blood-based biomarkers has already transformed Alzheimer's drug development. And as an example, we were early adopters of a phospho-tau blood test, plasma test in our ongoing Phase II study. The use of that measure streamlined our screening and enrollment, reduced patient burden and site burden and reduce the overall cost. It was a real way to innovate within the space and achieve efficiencies in our program. Today, we have 4 approved blood tests for Alzheimer's disease. These are all commercially being commercially deployed and we'll continue to, I think, establish and confirm Alzheimer's cases more broadly. Triaging patients in the primary care setting with blood-based biomarkers is expected to allow specialists to really focus their time and attention on treatment options. So we see this as really contributing to the growth overall in the space. As we look at the Alzheimer's market today, we see this as a market that is essentially underdeveloped and poised for continued and significant growth over the next decade. That growth will fundamentally be accelerated through the continued deployment and establishment of the clinical infrastructure to offer treatment options to patients. The adoption of blood-based biomarkers to characterize and potentially confirm more cases in the population. Additionally, other formats of drugs, whether such as the recently approved subcutaneous version of lecanemab for treatment induction is another way to establish convenience and growth within the marketplace. And then I think experience physicians just become more familiar with treatment options, risk assessments in terms of ARIA managing safety considerations for patients. We'll continue to expand the adoption and growth of these agents. Another concept that will continue to fuel the growth going forward is the combination strategies using essentially anti-a-beta approaches as a cornerstone of treatment are expected to be the future state of care. So it's an exciting time in the field and 1 that is poised for continued growth. So in this early phase of AD market development, we see a massive opportunity to improve on the safety and efficacy of current amyloid-targeting approaches. Our goal at Acumen is to develop products with improved benefit to risk profiles with either increased efficacy or increased safety or potentially both. We see a path to better efficacy and safety and a risk-benefit profile improvement based on our approach, which is to target soluble toxic a-beta oligomers Oligomers or the early instigators and persistent drivers of Alzheimer's pathology. Jim will talk more about this mechanism in his talk, but we think it's a path towards differentiation for distinctly within our pipeline. So suffice to say, there is room for improvement here. And we, at Acumen, are laser focused on that opportunity and seeking to have an outsized impact on the disease. So 2026 is a pivotal year for our innovative anti-a-beta oligomers pipeline. We have sabirnetug. Our lead program is positioned to read out clinical proof-of-concept data late this year, really data that we believe will provide the clinical validation of the a-beta hypothesis and user in a new mechanism within the amyloid space. We also have our EBD program, which is enhanced brain delivery. We've reported positive nonclinical data earlier this year and have 2 candidates that we're progressing through IND -- towards an IND in mid-202. So to summarize our value proposition and strategy, we are still in the early days of effective treatment approaches for what is a very large, growing and diverse Alzheimer's population in need of better options. Our novel anti-a-beta oligomer approach is differentiated within the clinically validated analytics space and exciting possibility to differentiate on a benefit-risk basis. sabirnetug has already produced compelling Phase I results in Alzheimer's patients, increasing the probability of success for it and future anti-a-beta oligomers therapies. Our ALTITUDE-AD Phase II trial will read out later this year, which is a hugely exciting milestone for us as a company. And our EBD program gives us future optionality well into the future. Our strategy to expand stakeholder value is based on successful Phase II results, expedite the development of sabirnetug with a partner and seek to advance an EBD candidate through a clinical value inflection point quite rapidly. And with that, I'll turn the call over to Jim.

Thanks, Dan. Good morning, everyone, and thank you for joining us today. As we approach the readout of the ALTITUDE-AD trial, I'm pleased to take this opportunity to talk about some of the key reasons why we believe sabirnetug represents a differentiated opportunity for anti-a-beta disease modifying therapies for the treatment of Alzheimer's disease. I'm going to touch on 3 related topics today: First, the soluble oligomers hypothesis. Sabirnetug the first monoclonal antibody targeting a-beta therapy to rigorously test the oligomers hypothesis in a late phase clinical trial. Critically, as you'll hear later from Eric, the ALTITUDE-AD trial represents a well-powered study focused on the potential benefits of subornation clinical measures and activities of daily living as well as safety in a variety of biomarkers. I'll talk about IgG2, like many other potential therapeutic agents, sabirnetug is a monoclonal antibody of the IgG type. It's important to consider though that subtypes of IgG have different signaling properties in the immune system. Finally, and as Dan mentioned in his introductory remarks, there have been a remarkable expansion on the availability and diversity of fluid-based biomarkers and that will become increasingly critical in the diagnosis of AD as well as in clinical practice. Next slide, please. So thinking about sabirnetug, how does sabirnetug differentiate from other approved antibody-based anti-a-beta disease modifying therapies? Well, like other approved anti-a-beta disease modifying therapies, sabirnetug is an amyloid protein targeting monoclonal antibody. However, unlike the approved anti-a-beta DMTs, a-beta targets soluble a-beta oligomers. Potential those will have potential effects on efficacy, meaning direct removal of what we believe are the most toxic agents that disrupt cortical function. We'll talk more about that as we go along, and potential effects on the safety profile. So with the potential for less interaction with CAA plaques that could adjust the risk for safety effects like ARIA. Also unlike approved anti-a-beta DMTs, sabirnetug, as I was just mentioning, is an IgG2 antibody that has potential effects on the safety profile, potentially a reduction in inflammatory effects and reduced ARIA risk. Next slide, please. So amyloid beta was first recognized as the major protein component in Alzheimer's disease plaques in the mid-1980s, becoming a key hallmark of Alzheimer's disease. This led to the amyloid cascade hypothesis in the 1990s and a key framework for Alzheimer's research that has recently produced 2 anti-amyloid monoclonal antibody-based disease-modifying therapies for the treatment of AD in lecanemab and then more recently, in donanemab. Amyloid precursor protein becomes abnormally processed, leading from amyloid beta peptides that are normally existing in a monomeric form to protein aggregates that can form larger and larger structures with different pathophysiological properties. What you can see on the cartoon is that as you get larger and larger components of a-beta, you do get these structural differences. And it was once thought that this was a fairly linear process going from smaller to larger fragments, but it has been more recently understood that this is more of a dynamic process where you've got cycling between pools of different size and shape amyloid protein. Next slide, please. So why do we care about that? As you can see from this figure from an excellent recent review on the amyloid biology in Alzheimer's disease, you can see that all these different forms of amyloid have been shown to interact with the central nervous system in somewhat different ways. And I'm not going to go through all of the elements here on the slide today, but really what I'm hoping people can see is that a, there's a lot of diverse signaling and pathophysiology that these fragments of amyloid can produce. But more than that, if you notice that specifically the soluble oligomers have a much larger total number of interactions and are interacting with many more functional systems and actively disrupting synaptic function. When you come to all this, this has led to the oligomer hypothesis, a refinement of the original amyloid hypothesis, deposits that these small size clusters of [indiscernible] molecules called oligomers are the main cause of neuronal dysfunction and memory loss in Alzheimer's disease. Okay. Let's talk a little bit more about these a-beta oligomers, if you go to the next slide. Thank you. So why are these a-beta oligomers of particular interest. In addition, as we were just looking on the last slide, there's a diverse set of pathophysiological signals that these small soluble protein fragments caused. They also show up very early in the course of disease. So they are an early elements of the pathophysiology of Alzheimer's disease occurring at the very earliest days far before the presence of larger amyloid plaques and far before the time when cognitive impairment begins to appear. But in addition to being an early component of disease, there are also a persistent component of disease. So these oligomers continue to be present as the disease progresses and other forms of amyloid and tau protein are becoming dysfunctional, contributing, we believe persistently to the pathophysiology in AD. So the next slide, please. So all this leads to this refined a-beta oligomer hypothesis, where plaques may be the visible pathology and the original thinking around amyloid biology, but these small harder-to-oligomers are the most synaptotoxic species and may, therefore, be highly relevant for disease. Next slide, please. So there is something called the Osaka mutation, which is a rare mutation that appears in an extremely small number of Japanese families has been characterized. And the point of this mutation is a mutation in the amyloid precursor protein that leads to the production of soluble oligomers, but no production plaque. You can see in this small number of families is that despite the fact that you don't have plaques of amyloid in the brain, you have robust levels of soluble oligomers, and that is associated with the cognitive impairment of Alzheimer's disease. So this is data that supports that soluble oligomers biology may be sufficient to produce the cognitive impairment of Alzheimer's disease. So a human experiment supporting the soluble oligomers hypothesis. Next slide, please. So all this biology leads to coming up with sabirnetug to target the soluble oligomers. And so sabirnetug represents the first oligomers-selective immunotherapy approach to treating Alzheimer's disease. And it's also the first oligomer-selective antibody that is being tested in a late phase clinical trial. So as you'll hear later from Eric, sabirnetug is currently in a Phase II trial where we're focused on measures of cognitive performance in addition to the biomarkers that have been characterized to date. All right. So why then might you expect to see a difference between sabirnetug and other types of anti-amyloid therapies. We go to the next slide, please. These are a number of studies that show some of the profile differences between sabirnetug, which is targeting the soluble oligomers, and in this case, lecanemab and donanemab to other antibodies that target different sizes and shapes of oligomer. What you can see on the left-hand graph is an SPR experiment looking at relative affinity between [indiscernible], lecanemab and sabirnetug on 2 different protein constructs of amyloid. So at the top, the a-beta [indiscernible] monomer, the so-called normal protein, and as you can see, sabirnetug has lower affinity than either of those 2 agents for monomers. But then if you look at the field symbols at the bottom, there, you're comparing the affinity for soluble oligomers. What you can see is that sabirnetug has a higher affinity for soluble oligomers than either lecanemab and donanemab. And given that monomer exists in high excess in the brain, the relative affinity between monomer and soluble oligomers also contributes to how much antibody is available to target the abnormal lignum proteins. And so we see sabirnetug having a very attractive profile, both in being very potent to targeting soluble oligomers, but also being less potent at targeting monomers giving an even greater relative affinity for sabirnetug for soluble oligomers. On the right-hand side is some recent work from Omar Dalian in the Klein lab at Northwestern University. His team utilized advanced immunoaffinity chromatography using targeted tools like the sabirnetug antibody to pull intact naturally occurring soluble oligomers directly from human tissue. This team reported that sabirnetug preferentially bound to soluble oligomers, whereas RMB-148, which is the mirroring precursor for lecanemab, bound preferentially to [indiscernible] a-beta. So again, additional data from the human brain showing that these antibodies are both recognizing amyloid protein but recognizing different forms of amyloid protein. And as we saw earlier, those different forms of amyloid protein can have very different physiological effects. And these are not small differences. As you can see from the summary on the slide, RMB158 found approximately 64% of the fibular form of a-beta taken from the human brain extracts, and only 36% of the globular or soluble a-beta, whereas sabirnetug, in contrast, was binding about 9% of the globular or soluble form of a-beta and only 1% of the fibular a-beta. So a definite biologically relevant difference between the 2 antibodies. Next slide, please. Sort of furthering the comparison, this is now looking at the relative binding to vascular a-beta in a murine model of vascular CAA, or cerebral amyloid angiopathy. So this is recent work from Martin Granite in Syneos lab at Harvard comparing the binding profile for sabirnetug with the binding profile for lecanemab in head-to-head in a mouse transgenic model of CAA. So in that study, lecanemab exhibited greater plaque and vascular labeling than did sabirnetug, although the authors do caution that immunohistochemistry conditions must be carefully optimized when making direct comparisons between antibodies. These results are consistent with the idea that sabirnetug binds a-beta species less closely associated with vasculature than does lecanemab. This is a careful and comprehensive analysis that we want to have time to discuss fully today, but I encourage you to check out Martin's paper in Alzheimer's and Dementia. Next slide. As I mentioned, sabirnetug uses an IgG2 back [indiscernible]. So that means that both lecanemab and donanemab being IgG1 antibodies signal to the immune system in a slightly different way than the sabirnetug as an IgG2 monoclonal antibody. That antibody subclass influences how strongly antibodies engage immunedeffector functions through Fc receptors and complement activation. So why is IgG2 different? Well, compared to IgG1 antibodies, IgG2 has substantially weaker FC alpha receptor binding, IgG2 gamma activates complement less efficiently, and IgG2 generally induces less antibody-dependent cellular cytotoxicity and less microglial activation. In Alzheimer's disease, that's important because some of the investigators believe that a portion of ARIA and infusion reactions may be related not only to amyloid removal itself, but also to the inflammatory responses generated when antibodies engage microglia and vascular amyloid. Next slide, please. So this is now looking at the current incidence of infusion-related reactions with the approved anti-amyloid beta therapies, lecanemab and donanemab. And I think the point here is that these type of inflammatory reactions do represent a significant impact to patient populations being treated with these agents. And so you can see the numbers here on the screen, both for infusion-related reactions as well as hypersensitivity-related reactions associated with the clinical trials for these 2 programs. Next slide, please. In addition, here are the values for the incidents and severity of ARIA with approved anti-a-beta DMTs. Of course, this is a huge area of investigation in the field in an area where there's an awful lot of effort being placed for the management of patients now that these therapies have achieved the marketplace. And of course, there are changes in practice that have been occurring, most notably with [ donanemab ] where the company was able to go that a change in the protocol for dosing had a reduction in ARIA rates. But you can see that ARIA rates remain a meaningful effect for both molecules across total populations as well across ApoE carrier status. On the right-hand side, you can see that not only is it total ARIA cases, but the number of symptomatic and serious ARIA events, and in this case, ARIAE events specifically are something that needs to be actively managed for patients. Next slide, please. So turning the page a little bit to be thinking about fluid biomarkers. As Dan mentioned in his introduction, and as I'm showing here, there has been a huge evolution of improvement in the use of biomarkers for the diagnosis and treatment of Alzheimer's disease. And this is occurring quite rapidly. So specifically, we're seeing changes from going from originally an autopsy based approach to diagnosis to be the use of amyloid PET as well as other PET imaging agents to do functional measures to diagnose disease. And coupling that with CSF lumbar punctures to measure biochemical biomarkers to allow you to both identify who is suffering from Alzheimer's disease, but also where they are in their time course of disease. So finally, more recently, blood-based biomarkers, there's been a huge surge in both of the diversity and availability of blood-based biomarkers for diagnosis as well as for understanding stage progression in disease and hopefully, in the future to help manage both clinical trials and therapeutics. Next slide, please. So over the last 15 months, there have been multiple approvals in the diagnostic market for new diagnostic tests around [indiscernible] to diagnose Alzheimer's disease mostly in the U.S., although this is moving forward in the rest of the world as well. So we believe that this is going to become an increased opportunity to have better ability to diagnose patients and to also substantially broaden the number of patients who will be diagnosable. Next slide, please. So how are we using these fluid biomarkers in our own studies? And the answer is, in multiple different ways. So because so many different proteins can be analyzed, it really can give, one, a broader picture of the impact and the effect of the agents being tested. So this is a little bit of a cartoon showing the kinds of biomarkers that Acumen has been looking at in our own programs. And you can see we're measuring amyloid-related pathophysiology looking at ABA to 42/40 ratio, so very proximal to the mechanism of action of the antibodies that we are delivering. But also at another level of integration, we're looking at p-tau81 and p-tau217 which, although our tau markers, they are associated with the pace of amyloid change. And then a little further downstream, we're looking at GFAP for measures of [indiscernible] activation, and we're looking at a number of markers of synaptic injury because if our hypothesis holds for how sabirnetug is impacting the pathophysiology of Alzheimer's disease, we should be able to see effects on downstream markers like neurogranin and VAM2 that the antibody does not directly interact with. Next slide, please. So here are some of the results from our Phase I INTERCEPT-AD study in Alzheimer's patients who have been treated with sabirnetug once monthly IV. And you can see these are data from the MAD cohorts. So these subjects by this point in time, had received 3 consecutive injections of sabirnetug and because it was a Phase I study at multiple doses. And what you can see is both a dose-related and persistent change in the ratio of a-beta 40/42 consistent with what you might expect to see if you're able to normalize in amyloid function. And also, we see similarly dose associated changes in p-tau181. And really importantly, we see similar doses aside changes with both [indiscernible] and postsynaptic marker of Synaptic Health as well as VAP2, a presynaptic marker of Synaptic Health. So these data taken together are supportive of the profile of the oligomer targeting sabirnetug as having a meaningful effect on the fluid biomarkers in Alzheimer's patients. So next slide, please. We can also look at these biomarkers in a relative sense. So this is a comparison of results from multiple different existing antibody trials. So these are not cross-study comparisons, and so I won't always have to be careful with interpreting the results. But what you can see is that if you look at sabirnetug in red, what we're seeing is a fairly rapid change in p-tau181 and also in neurogranin, consistent with the hypothesis that we're seeing a marker of early and significant impact on changes associated with Alzheimer's disease with sabirnetug. Next slide, please. So I hope I've convinced you of a couple of things. sabirnetug preferentially binds to soluble oligomers, which is a low abundance, highly toxic form of a-beta that appears during the early phases of disease. This profile offers the opportunity to differentiate from the currently approved DMTs that target other forms of a-beta. In addition, that IgG2 backbone of sabirnetug really offers a different opportunity to interact with the immune system and therefore, impact both the efficacy and tolerability profiles for sabirnetug. And finally, suburban functions rapid and robust effects on multiple fluid biomarkers that are associated with cognitive impairment, both proximally when you think about amyloid protein itself, but also downstream markers like the synaptic markers. So with that, I appreciate your time, and I will turn the call over to Eric, who will talk through the profile for the ALTITUDE-AD trial.

Well, thanks, Jim. And what I'd like to do now is talk a little bit about our ALTITUDE-AD study. It's a well powered Phase II study that we believe is the first study to actually test the oligomer hypothesis with any rigor. So if you want to go to the next slide, we'll come back and talk about this slide in more detail, but what we're going to get out of this study, which is 542 people in a Phase II study, so it's a large Phase II study. We'll look at clinical endpoints. Of course, we'll look at safety. And then we have a number of interesting biomarkers that we looked at in our Phase I study, and of course, now we'll look at it in our Phase II ALTITUDE study. So if we want to go to the next slide. So you might ask yourself, well, how are they going to get all this data? Well, here's how we're going to do it. It's a 3-arm study. So we have 2 different doses of sabirnetug and 1 arm that's placebo. It's an 18-month study with infusions given once every 4 weeks. And after the 18-month period, there is a 1-year open-label extension that's available to people. And we've actually been finding that a very high percentage of people want to go into that open-label extension. But the real readout of the study will be the placebo-controlled portion, which is at the end of the 18 months. So if you want to go to the next slide. So in our Phase I study, we actually obtained a lot of data that were very good in terms of designing and planning this Phase II ALTITUDE study. So if you look at the lower left on this slide, that's our -- what we call our target engagement assay. So what you see on the Y axis is how much sabirnetug there is that bound to an oligomer. The X axis is just the concentration of driving in the spine fluid. So this is a spinal fluid test. The reason why this is so important is, before we did the study, the top dose in our Phase I study was 60 milligrams per kilogram, but we would get the question, well, what if you get up to 60 and you don't see anything, you don't have any safety problems, could you go higher? Well, what this graph shows you on the lower left is that when you get to the upper doses in the Phase I study, you're already getting to the point of diminishing returns. In other words, that curve flattens. It doesn't just continue to go up. And so that tells us that there's really no reason to go above certainly milligrams 60 milligram per kilogram. And really, there's probably no reason to go above 50 milligrams per kilogram. Now on the lower right, you can see some model data. So we took the data from the graph at the lower left. And then we did some modeling with it to choose our doses for ALTITUDE. And so we think the ALTITUDE is well designed in terms of having the right target engagement information. If you look carefully at that at 35 milligrams per kilogram, we actually have quite good target engagement at both peak and trough. But we also wanted to include a 50-milligram per kilogram dose group because we do think that based on our Phase I results, there's more of a chance to see some reduction in plaque with the 50 milligrams per kilogram dose. Whether or not plaque reduction is important for a drug like sabirnetug, sabirnetug that targets oligomers isn't really clear. But just in case that was necessary, we wanted to include that higher 50-milligram per kilogram dose in the study. So if you want to go to the next slide. Now this is a bit small and there's a lot of information on here, and I'll just remind people that the information here is in our corporate deck slides and you can look at it in more detail if you want. But the important thing is, on the left, you're seeing changes in CSF biomarkers, including things like [indiscernible], VAMP2, p-tau181. And it's a very consistent effect in terms of lowering those things. They're not all statistically significant, but directionally, it's very consistent. On the far right is the a-beta 42:40 ratio that tends to go up, which is what you would want to see. So this is after just 3 administrations of sabirnetug in a Phase I study. So these data, at least in my view, were surprisingly good. And obviously, we've taken these types of assays and incorporated them into our Phase II altitude study. On the right-hand side, you can see the plasma biomarkers which generally move the same direction. As probably many of you know, plasma biomarkers in Alzheimer's disease has just been an exploding field -- exploding topic for the field, and we've continued to look at those, obviously, in our Phase II ALTITUDE study. So -- but these were all from just our Phase I study. So if you could go to the next slide. And then the last thing to think about is that we are an IgG2 with the other amyloid-related antibodies being an IgG1. Now IgG2 have less of what's called effector function than IgG1s that conceivably could provide you with better safety. Again, in our Phase I study, those are small studies, but the safety appear to be quite good. We did have 5 cases of ARIAE or about 10% of the patients, but importantly, only 1 of those was symptomatic, and that person's symptoms were very, very subtle. In fact, they were more subjective. You really couldn't pick up anything on exam, and they resolved as we held the drug and the area went away. So Being an IgG2, we think, has the potential for improved safety compared to the other monoclonal antibodies that have been approved or being studied. So if you want to go to the next slide. Well, let's talk a little bit then about the iADRS, our primary endpoint. Some of you maybe more or less familiar with the iADRS as an end point. The [indiscernible] is the other scale that's used commonly in Alzheimer's trials as a primary endpoint. We have it as a secondary endpoint. But if you just want to go ahead and go to the next slide. So let's talk a little bit more about the iADRS scale and the CDR Sum of Boxes just to compare and contrast a bit. Both scales conceptually are similar in that they combine cognitive items and functional items with the functional items being activities of daily living, that sort of thing. There are more items in the iADRS than the CDR [indiscernible] boxes, but still conceptually, it's a composite scale that combines the 2 domains. There are some technical differences in how the scales are administered. So in the iADRS, the cognitive measures are strictly performance-based. In the CER, it's performance based, but it's also based on a structured interview and also radar judgment. It actually takes a minimum of 6 hours of training to be a CDR rater. And there's a certain amount of subjectivity based on people's experience that goes into the rating. So there's a bit of difference in how they're done. For the functional measures, that's a structured interview with the study partner for the iADRS and it's what's called a semi-structured interview with the study partner, plus this added rater judgment piece for the CER Sum of boxes. So the net result of some of these differences, I think, is that if you look at the signal to noise ratio for iADRS, it's better than the signal and noise ratio for the CDR [indiscernible] Boxes. There aren't a lot of studies that have both the iADRS and the CDR [indiscernible] Boxes in them to compare head-to-head. But the EXPEDITION studies, which looked at a drug called solanezumab actually contain both of these scales, and so you can compare them directly. And so in the EXPEDITION study, the effect size was much greater for the iADRS, that was 0.93 compared to the CDR [indiscernible] Boxes, which was 0.006. For the EXPEDITION2 study, the effect size was about twice that of the CDR [indiscernible] boxes. In EXPEDITION3, it was higher, but not by as much for reasons which are clear, but very consistently, you see a higher effect size for the iADRS when you can compare it head-to-head with the CDR [indiscernible] Boxes. And then finally, the last line there is looking at donanemab Phase II studies. And I think this is a good illustration actually, that in the Phase II study of donanemab that had 245 people, the iADRS did reach statistical significance at 0.04, but the CDR [indiscernible] Boxes did not at 0.14. Now when they went on to Phase III with much larger studies, both of those were statistically significant. So it's like a lot of things. If you're underpowered, you just need more patients. But I think it's a good illustration of the fact that you can be positive on the iADRS and negative on the CDR [indiscernible] Boxes. And we just feel that overall, the iADRS is a more sensitive scale and accurate scale, and that's why we chose it as our primary. But again, the CDR [indiscernible] Boxes is one of our key secondary outcomes. So if you want to go to the next slide, One of the things that we did that was really novel at the time in the altitude study was we used a blood test, p-tau217 to screen people for the study. So in this slide, on the left-hand side is data from our Phase I study. On the right-hand side is the data from our Phase II ALTITUDE study. And what you can see is that the primary reason for screen failures in our Phase I study, which did not use this p-tau217 blood test screener was amyloid PET scans. If you look at ALTITUDE, the largest reason for screen failures was the blood test of p-tau217. The overall screen failure rate for both the Phase I and the Phase II is about the same. But the question is, why do you screen fail? And I think everybody agrees that you're much better off screen failing from a blood test than to get all the way to a PET scan and have a PET scan. So at the time we did this, this was really very novel, and we think it actually worked very well. We also think it's something that could be used in clinical practice. In other words, screen with a blood test and then confirm if you think that's necessary with either a PET scan or spinal fluid, but it cuts down considerably the number of PET scans or spinal fluids that you need to obtain. So if you want to go to the next slide. So this is a graph of our enrollment for the study, which was very rapid. We enrolled the study in 10 months. And that was well beyond anybody's expectations at the time. And part of the reason why we think that was the case was that people, the sites and the site PIs have told us that they really like the study design, and they like the drug and they like this p-tau217 screening because, again, if you're going to screen field, it's a lot better to do it with a blood test than with a PET scan. The other thing I might just point out on this graph is that if you noticed like September, October 2024, the rate was a little faster before that, maybe not quite as fast after that, but that's because we were sort of transitioning to our European and U.K. sites, and that always takes a little bit of time. But if we were to just let the U.S. keep running with this, we probably would have even enrolled it a little bit faster. So we are very pleased with this. And what we've heard from the sites is they were very pleased with the protocol, too. So if you want to go to the next slide. And so here's the net result of this. So we -- again, we were doing some very novel things. And at the end of the day, what we want to see is how we compare with clarity the CLARITY study. the CLARITYY study of sabirnetug has a patient population most similar to ours. For those of you who are familiar with the Trailblazer studies of donanemab, they had a requirement for tau, which we did not have, CLARITY do not have. So they have a little bit different patient population. But again, CLARITY did not have the p-tau217 screening because when they designed the study, it wasn't available. And so -- and we did have that p-tau217 screen. But the net result, when you look at the baseline data for ALTITUDE and CLARITY is they are very, very similar. So in other words, we were able to implement this, at the time, novel screening technique with p-tau217, but we ended up with essentially exactly the same patient population as was obtained with the CLARITY study, where they did not have the screen with p-tau217. So we feel like we really accomplished our screening process overall in a way that was very, very good. And -- but we did it in a way that was much easier for patients in the site. So again, we're very pleased with these results. So if you want to go to the next slide. So let me talk that in a little bit more detail about what we are looking at in the study. So I talked already about our primary outcome variable for clinical endpoints is the iADRS scale. But again, the CDR [indiscernible], of course, will be a key secondary. The the ADCS-IADL and the ADAS cog that you see there are actually components that make up the iADRS. So we'll look at those things individually. And of course, as I mentioned and as you heard, we did see some plaque reduction in our Phase I study. And so we'll look at that in ALTITUDE, but whether plaque reduction is really important for efficacy when your drug targets oligomers rather than plaque, it's not really clear whether that's necessary. From a safety standpoint, again, we have an IgG2 rather than an IgG1. We think that has the potential for better safety. And obviously, we're going to look very carefully at aira rates, ARIAE a rates and just adverse events, in general, as you would in any study. Infusion-related reactions, I think, could be an important thing when we do see our data from attitude because, again, potentially with an IgG2, this could be less of an issue than it is with some of the other antibodies being studied currently. And then finally, for biomarkers, again, we were very pleased to see these changes in our Phase I study, and we'll be looking at these again in the Phase II study. One of these, which is particularly, I think, important for us is neurogranin because it's a synaptic biomarker. It's a postsynaptic biomarker in these oligomers are toxic to synapses. And so we're looking at the usual suspects in terms of CSF biomarkers, but the neurogranin is something that we want to demonstrate that and expand on our Phase I results and show an effect in our much larger Phase II study. For plasma biomarkers, GFAP is an interesting one. It actually reflects astrocytes, which are a different type of nerve cell and it's sort of an inflammatory marker. So we saw, again, some Phase I data that look very promising for GFAP. We'll be excited to see the results from a much larger Phase II. And of course, we're going to look at p-tau217. Not -- we use this as a screener, but then we'll also be looking at that as a measure of at least biochemical efficacy. So if you want to go to the next slide. So these are the takeaway messages, and I'm not going to read all of these to you. We just talked about those. But I think, the thing to keep in mind is that we had -- as you heard, we had some very good Phase I results, which led to a Phase II study, 542 people, so not a small Phase II that enrolled very, very quickly, very smoothly. We have a high percentage of patients who elect to roll over into the open-label extension. So they like being in the study. They like being on the drug and certainly the sites have told us the same thing. So we're just very much looking forward to seeing the results from altitude when those unblinded results are available. So thank you very much. And with that, I'll turn it over to Paul.

Thank you, Eric. So in this last segment, I would like to describe our new drug discovery effort developing next-generation antibodies for the treatment of Alzheimer's disease. So as we all know, there are inherent challenges with therapeutic monoclonal antibodies for the treatment of neurodegenerative diseases, including Alzheimer's disease. These antibodies have a very poor penetration across the blood-brain barrier with only about 0.1% to 0.2% of dosed antibody actually reaching the target in the brain. One way to get around this, of course, is to give higher doses of antibody. So companies have increased the dose, and by increasing the dose, you get more antibody that trickles across blood-brain barrier to get to the brain. Another way is to engineer or develop antibodies that have longer half-lives. So if the antibody is around longer in the system, then of course, there's more antibody that eventually gets into the brain. One issue with increasing the amount of drug that you give is that there are safety and tolerability concerns. In the a-beta antibody space, ARIA and ARIAE are the main concerns that we see when we increase drug level. So what happens is we end up with a dose tolerability, safety limit so that we can only give the patients so much drug before we start to see an increase in these safety signals. Another concern with monoclonal antibodies is that they don't distribute throughout the brain equally. So in areas close to large blood vessels into the ventricles of the brain, we see higher concentrations of the antibody and in other brain regions that are called the deep brain regions that are more isolated, we see certainly lower concentrations of antibody. This potentially causing a differential treatment outcome in these brain regions. So one way that companies have tried to get more drug into the brain is the use of the receptor-mediated transcytosis system. So this is a system that the brain uses to selectively shuttle large molecules of interest into the brain, molecules like insulin and transferrin and and other things that the brain needs that are normally kept out by the blood-brain barrier. So a number of decades ago, researchers realized that if you raise antibodies to these receptors on the blood-brain Barrier, that you can use these receptors to carry large molecules such as antibodies into the brain. The most advanced of these is the transferrin receptor. And as we've seen, you can use antibodies or antibodies fragments, attach them to large molecules of interest and use these to shuttle into the brain. This greatly increases the amount of drug that you get to the brain and associated efficacy while at the same time reducing side effects since you can reduce the drug that you're delivering to the patients. So what I'd like to describe to you is our effort to develop bispecific antibodies for the treatment of Alzheimer's disease. So when thinking about developing molecules, we started from scratch. It was a whiteboard exercise for us. We thought about this as 2 major pieces. On the 1 side, the business end of the molecule is the cargo. This is the part that's actually binding to the a-beta species of interest. And on the other end, linked biolinker is the transporter and this is the portion of the molecule that's binding to the transferrin receptor to facilitate the entry of the molecule into the brain. Both sides of these molecules are very important. The cargo determines what species of a-beta you bind to. As we know, there are antibodies that target large portions of a-beta, plaque, fibrils, some target monomer, and our company is unique in developing antibodies that target oligomers. This part of the molecule determines the efficacy. The other end of the molecule, the transporter part, as I mentioned before, can be based on a number of different receptors that are at the blood-brain barrier, CD98, insulin receptor, [indiscernible] receptor are just a few. This part determines the PK of the molecule, how long the molecule was around in the blood and in the brain and also has associated safety risks that need to be kept in mind. So if we focus in and look at these a little bit more carefully, what sets us apart in our view of developing these bispecific antibodies is we believe that a-beta oligomers are the toxic species in Alzheimer's disease. So we've selected 2 antibodies from our portfolio of antibodies to take into this program, ACU193, or sabirnetug, which is currently in Phase II clinical studies that read out later this year and a novel antibody called ACU234, which we developed and has new and unique properties. So at the other end of the molecule is the transporter. So this is the portion of the molecule that helps facilitate entry into the brain. We looked at a number of different companies that had different transporters and different platforms and decided to partner with JCR. JCR is the first company to have an anti-transferrin receptor molecule approved in the world for the treatment of a rare brain disease. And they have a platform of transferrin single chain and BHH transporters that we could use to screen for molecules that work best with our cargo. So this is the approach that we took. So we really were agnostic in how we viewed molecules of interest. So we kept in mind transparent receptor affinity, the architecture of the molecule, the balance of the molecule and selectivity for a-beta oligomers. But we did not go in predetermining that any one of these should be of a certain form. So we looked at a range of transferrin receptor affinity, a variety of architectures, both monovalent and bivalent in a range of a-beta binders. So this slide summarizes about 1.5 years worth of drug discovery work and the leads that emanated from this work, ACU301 and ACU401. And as you can see, these are both bivalent antibodies with a relatively high affinity to the transferrin receptor in the single-digit nanomolar range. And these antibodies we took through a variety of in vitro and in vivo assays to characterize to end up selecting these 2 molecules. You'll notice that these antibodies are at odds with the number of dogmas that are in this field, one being that bivalent antibodies bind too strongly and don't release into the brain, that antibodies with high affinity also don't release and that using a linker will result in clipping of the transporter off of the antibody. In our studies, we've noticed that none of these events happen, that we see very good penetration of the blood-brain barrier and into the brain after subcu dosing, and that antibodies are released into the parenchyma of the brain and are able to engage [indiscernible] species of interest. As you can see at the bottom, we also look to make sure that the antibody still bound a-beta in Alzheimer's brain. So this is a histological study that we did showing that both ACU301 and ACU401 are still able to bind the a-beta species of interest and that this wasn't altered in the construction of these bispecific molecules. So this is just one slide showing some of the in vivo work we did in mice. As you can see, after subcu dosing ACU401, you can see that there's a rapid uptake. The red line into the blood after subcu dosing. You can see that it becomes comparable to the IV dose shown in blue, and that both of these have a good half-life out to a week. When we looked in brain, we see an expected difference in the peaks that we see in the brain. The IV, of course, peaking quicker because it's available immediately to get into the brain. And then the subcu dose, you can see kind of lags a little bit behind the IV dose. Both have a high CMAX, both have a very nice T1 half life and suggest that these antibodies are worth taking forward into additional studies. So our next step was to look at these antibodies and how they perform in primates. So we use synomologous monkeys for these studies. And as you can see, we designed a 2-phase study. So in the first phase, animals were dosed subcu with 5 mg per kg of antibody, and they were dosed with either ACU401 or the monoclonal antibody ACU234. The little blood samples that you see in red were collected over time, over a 35-day period of time. And we also looked at hematology endpoints in a sample that was collected pre-dose and then 24 hours after dosing. Once the blood collections were completed at day 35, all of these animals been rolled into the next phase of the study. So animals were then dosed IV with 2 mgs per kg of the antibody, the same antibody they received before. And then half of the animals, 3 animals were euthanized 3 hours after dosing and then the remaining 3 24 hours after dosing. At the point of [indiscernible], we collected the brain and the CSF as well as the blood samples that you can see in the slide. So when we look at the PK, you can see that both ACU401 and 234 are rapidly taken up. After subcu dosing on the left, you can see kind of a blowup of the first 24 hours. You can see that very rapidly, within 8 to 12 hours, the antibody is getting close to a CMAX and then it's maintained thereafter for a period of time. On the right, you can see where we've tracked the antibody levels out to 2 weeks. And what you can see is that after subcu dosing, ACU401, at the end of 2 weeks, still has a very good half-life and suggest that as we go forward in clinical studies, we'll have optionality on how we want to dose patients with this drug. The other thing worth pointing out is because these antibodies are getting in so rapidly taken up so rapidly that there's less of a concern about half life with these NOLs, with these antibodies as we were with monoclonals that get in poorly because these antibodies seem to be getting in so well, we're less concerned about half life. So when we looked at 3 regions of the brain, the [indiscernible] cortex, the putamen and the hippocampus, we noticed that the trends with ACU401 were very similar in all 3 brain regions. As you can see in the [indiscernible], within 3 hours, you see a pretty remarkable uptake of drug into the brain, about 22 for higher than what you see with ACU. At 24 hours, this increases further to a 4-fold difference between ACU401 and 234, and this is a [indiscernible] that we see in all brain regions, including the putamen, which is 1 of those deep brain regions that's very hard to get drug to. You'll notice in the hippocampus that the ACU levels were higher than expected and higher than what other companies have reported after dosing a monoclonal antibody, they, of course, see similar differences in all brain regions. So we're thinking that this elevated level in the hippocampus is likely due to either a contamination or sampling error. And we're repeating animal studies right now in monkeys and we'll have a chance to look to see if in fact this was some sort of a sampling error. We also looked at CSF levels in this study. I'm not showing that data. But what we saw was that if you look at drug levels in the CSF, there's no difference between ACU234 and ACU401, again, highlighting that the transferrin receptor system is increasing drug levels in the brain but not in the CSF. As I mentioned, we also looked at a panel of hematology endpoints in this study 24 hours after the subcu dosing. We looked at red blood cells, hemoglobin, hematocrit and reticulocyte count. And you can see that either dosing with the monoclonal antibody 234 or with our bispecific antibody that there was no difference observed in any of these endpoints. This certainly is a positive indicator and suggests that there might be a low risk for anemia in patients. So to summarize the key takeaways from this study, we've shown that when we dose with ACU401, we see a robust uptake in the brain with levels as high as 40x higher than a conventional monoclonal antibody. Looking at a panel of hematology endpoints, we see that there's a relatively low risk of anemia based on the markers that we looked at. And after a subcu dosing, we see that the antibody is rapidly taken up into the plasma and that there's a very nice PK profile that's amenable to a variety of dosing options going into the clinic. So when we started this program, we had a lead candidate profile in mind. So what we saw is that the antibody gets seen much better than we had hoped for. We were hoping for around a 20-fold increase in antibody levels with our bispecific antibodies. As you can see, we've got at least 40-fold increased antibody levels. We wanted to make sure that when we combine the transferrin receptor with our antibody that it wasn't compromising binding to the oligomer, and we see that we've maintained high a-beta oligomers specificity versus monomer. We also noticed that we've maintained the ability to find the transferrin receptor at high affinity in the low nanomolar range. And this is important because dosing -- we can dose at much lower levels with this high affinity binding. And we also showed that the high affinity binding doesn't mean that the antibody is stuck to the vessel wall that it actually does release into [indiscernible] the brain. We saw no signals that would suggest that there are going to be issues with anemia. We also looked at the stability of these antibodies and found that these antibodies have good stability. And should be able to be maintained in an auto-injector at 4 degrees for an extended period of time. And finally, we saw that there was a very good uptake of drug after subcu dosing and that this should allow us to move into the clinic with a subcute product. So what are our next steps? Next steps are, of course, we're doing a second monkey study. This will help us determine the dose and the treatment paradigm that we take into the clinic. We're doing additional IND-enabling studies, including supportive CMC work, safety tox work and an in vivo tox study. And then finally, we're developing bioanalytical assays that we'll need for the clinic PK, ADA and other assays, and of course, throughout this process, we have been interacting with the agency for their guidance. So our next goal is to file an IND next summer, mid-2027. And I'm going to stop here and hand it back over to the operator for questions.

[Operator Instructions] Our first question is going to come from the line of Pete Stavropoulos with Cantor.

Thanks for hosting the event First question, when you look at the baseline characteristics of those enrolled in ALTITUDE, how do they sort of compare to the enrolled in the registrational studies for [indiscernible]? And when you look at the baseline PVR [indiscernible] boxes for altitude, it's 2.91%. CLARITY 3.17, I believe for donanemab, it was 3.9%. So sort of help us understand these are similar populations or there's some type of meaningful difference on CDR [indiscernible] Boxes?

Yes. Well, thanks for that question. This is Eric. I'm happy to take that one. So if you look at -- we'll just start off with the CDR [indiscernible] Boxes There's a numeric difference they are a small numeric difference. But especially -- and I know the print's a little small, but if you compare it to the standard deviation, those -- now these aren't head-to-head comparisons, obviously, so you have to be careful. But those differences certainly aren't clinically meaningful, and I don't think, statistically, they're significant either. I mean you get a certain amount of variability in these studies just due to patient or study variability from 1 study to the next. If you look at the [indiscernible] score, that actually is exactly the same in both studies. And if you look across the board at all the baseline characteristics, they are very, very similar. And to my mind, I don't see anything in there that's a meaningful difference in the 2 patient populations. For donanemab, as I mentioned, those patient populations are a little different. And I think that's because they had this requirement for a certain amount of tau positivity. So if you look across all the various measures, the donanemab studies, the patients are a little bit more advanced or a little bit worse. So it is -- that is a little bit different patient population. But for us, for ALTITUDE and for CLARITY, that's about as close as you're ever going to see in terms of baseline characteristics in 2 different studies, I would say.

And as you noted during the call, there's been a lot of progress in the Alzheimer's space in terms of biomarkers, some of which show changes. I don't know if they start to appear for in advance of symptoms as well as some of the underlying pathology like various to tau species. How do these updates, including newer biomarker update? Inform your approach and assumptions about disease and clinical studies? And are there any that stand out to you, especially the newer biomarkers as you look at ALTITUDE or incorporated into the Phase III, including biomarkers not listed in your presentation, like perhaps 243.

Well, Yes. Go ahead, Jim. No. Another great question. And yes, the field is moving so quickly. it's a good problem to have, obviously, but especially p-tau243 is 1 of the ones that is newer and seems to be quite promising. But the first thing that we'll do actually, of course, as I look at our ALTITUDE results and look at the biomarkers where we do -- that we do have in there. So just, for instance, when we designed the ALTITUDE study, there weren't any p-taus that were FDA approved at that point, or there weren't any diagnostic blood-based diagnostic biomarkers approved at that point. So now we have 4 of them. So one of the things we'll go back and look at when we get our ALTITUDE data, is when we do design our Phase III, should we tweak the screening procedure that worked very well previously with the p-tau217 assay that we use. But now that there's 4 different assays that are FDA approved, we'll need to rethink that for what we do in Phase III. So -- but anyway, there's a lot happening in the field, and it's a good problem to have in a way. I mean we're going to look at this very carefully. I don't know, Jim, did you want to add something to that?

Yes, Eric, just to layer a couple of additional thoughts on. And I think, first, taking a step back, Pete, what we see and I think what a lot of people see is that there's just been a continuous development in both the precision of especially fluid-based biomarker analysis, but also the diversity. So we know -- we certainly know that there are additional markers that are being evaluated now, and I think that's only going to continue. You mentioned 243 is one of them. And I think you'll see that, as time goes forward, we're going to have a better ability to both understand where individuals are in their journey in this progressive disorder. And I think that only helps in diagnosis. But I think it will also help in clinical trials. And when you sort of zoom oligomers hypothesis, as we talked about today, oligomers are in early elements of disease. They start showing up fairly early during the time versus the disease well before clinical symptoms have appeared. And as we showed earlier, there's a ton of evidence that there is physical interaction with synaptic circuitry. So it's entirely feasible that there are measurable things happening in the brain that are occurring very early in disease. And so I think future work will be to try to understand, does that mean that there is pathophysiology triggered by oligomers that could be measured early on. So I think there will be multiple ways that the biomarkers are going to be utilized moving forward. And it's just really great to see both the expansion of markers like p-tau217. As Eric said, there are now 4 approved tests out there. And that's only going to help both in terms of trials, but probably more importantly for clinical practice. But then also the diversity of the markets coming forward, being able to understand in more detail what's going on and hopefully how individual treatments are benefiting patients. So I think all that's to come. We've done a fair amount of work analyzing the data from the INTERCEPT study. So there are individual publications out there kind of laying out the data that we talked about today in much more detail. But we certainly think about this in multiple levels. And really, we want to look beyond the proximal amyloid and tau and start incorporating some downstream biomarkers, which we think are going to be really interesting and important for assessing synaptic health and synaptic function.

Just one last question on EBD. Let's say that ALTITUDE reads out positively. Will that impact your decision to sort of bring sabirnetug versus 234 forward? Positive data would just clinically derisk the the binding properties of sabirnetug, why not stick to sabirnetug rather than introduce risk?

So Jim, do you want to grab that one?

Yes, absolutely. I think the way we think about this is, as Paul very well said in his presentation, we've been fairly broad in our thinking around what's the best set of properties to have in a molecule. And we think that 401 and 301, having both of them gives us optionality. Each one offers different opportunities. And it's important to note that 401 is, as Paul described, coming from 234. And though 234 is a distinct molecule from sabirnetug. They are quite similar. So we haven't made major changes. In fact, we still like the selectivity and monomer affinity that we have with 234, and it actually even offers some opportunities different than what we get with sabirnetug. So we think both are really robust candidates, and we'd be comfortable taking either of them forward into the clinic. The work that Paul's team is doing now is going to lay out what's got the overall best set of properties to take forward into Phase I. But I think also the good thing is that what I'd love to see is that both molecules actually do well in that analysis, then we have a tough decision which one to take forward, and we'll still have the other molecules sitting in late preclinical phase that we could bring possibly bring forward in the future. And again, thinking back to what I was just saying a couple of minutes ago, maybe there are different opportunities in either slightly different populations of patients that are identified by biomarkers or in earlier phase of disease. And so we really want to maintain optionality. And so we're interested in profiling both molecules.

Our next question will come from the line of Paul Mattis with Stifel.

This is [indiscernible] on for Paul. Just a couple of quick questions for you. So on the EBD profile, we were wondering what's the [indiscernible] here? So like what dosing frequency would you like to see as well as maybe like how quickly do you think you can get data from in patients following that mid-2027 IND filing.

Yes, so we're open to all of this at the moment. We are currently running another primate study. And in that study, we're looking at different doses and different dosing frequencies. And we're hoping that the output of that study will guide us in how we set up and run our Phase I SAD-MAD study. So certainly weekly, twice monthly, monthly are all under consideration at the moment. And we haven't made any decisions about the dosing frequency. We'll let the data guide us in how we do that. In regards to the design of our SAD-MAD study. Certainly something that we're discussing right now, thinking about the best option for this molecule. And based on data that we get from the ALTITUDE study will certainly directly impact how we think about the studies that we run, the biomarkers we use, the endpoints of interest. So yes, all of that is still under consideration. And as stated, by the middle of next year, we'll be ready to move forward.

Our next question comes from the line of Jason Zemansky with Bank of America.

Congrats on the great progress. Maybe to start, it's probably fair to say a key determinant of success in ALTITUDE will be the placebo arm. So given the use of the p-tau screening, the relatively high proportion of MCI patients and some of the similarities you've highlighted with CLARITY, how should investors think about placebo decline? Is it going to be broadly comparable to prior amyloid studies? Or are there characteristics that could make it meaningfully different?

Well, yes, maybe I can address that. So you're absolutely right. The patient populations -- well, first of all, the patient populations in CLARITY and ALTITUDE are quite similar. And there's more patients who would be classified as MCI then have mild dementia, roughly 80% have MCI. As the disease progresses, actually, the rate of decline, say, in a placebo group actually gets greater. So as you go to earlier stages, you might see a little bit smaller rate of decline. But as it turns out, for your drug to have efficacy, it appears to be that the sweet spot is this patient population that's now being called early AD, which is either MCI or mild dementia due to Alzheimer's pathology. The fact that lecanemab showed a signal in a patient population that's very similar to the patient population we have in ALTITUDE, I think bodes well for us. That seems to be the sweet spot. So even though the placebo decline may not be as great, your drug efficacy ends up being better in that earlier population. So you have to be very careful about comparing studies with different patient populations because you will see different rates of decline in the placebo group. But at the end of the day, what you really want to show is the difference between active treatment and placebo, in other words, the efficacy. So we feel good about the patient population that we've identified. And I would expect the placebo decline will be similar to what we've seen in clarity. We obviously don't know that at this point, but that would be my expectation.

Yes. Makes sense. And then maybe as a quick follow-up for Jim and Dan, I spent a lot of focus on the potential for differentiation through efficacy. But based on your market research and physician work, do you have a sense of how much commercial value could be specifically created through safety differentiation, particularly around ARIA and, I guess, overall tolerability?

Thanks, Jason. So we think there's ample opportunities to differentiate on both efficacy and safety and a clinically meaningful safety benefit, we think, is commercially pertinent. So that is an important underpinning of the value proposition of sabirnetug and really why we think the risk-benefit profile in totality and presumably across maybe a more diverse set of patients, including E4 carriers and homozygous is an important aspect of sort of the overall, the totality of the opportunity for sabirnetug to differentiate.

Our next question is going to come from the line of Geoff Meacham with Citi.