Amazing. Hello everyone and welcome to the Cantor Global Healthcare Conference. I'm Samantha Schaeffer. I'm with the Cantor Biotech team. Today we have with us Acumen Pharmaceuticals, a company that Pete Stavropoulos covers, and I'm pleased to introduce Dan O'Connell, the CEO, and Jim Doherty, the Chief Development Officer. Let's get started off with an introduction of yourselves, followed by a description of the company, and why and how Acumen was founded. Yeah, great. Thanks, Samantha. Thanks to you and the team at Cantor for having us here. I'm Dan O'Connell. I'm the Chief Executive Officer of Acumen. I've been in the life sciences industry for over 25 years as an early-stage investor and, more recently, as the Chief Executive of Acumen. I've been focused on working with neuroscience companies for an extended period of time, principally on the business and leadership side. I'm really excited to share some updates on Acumen today. Fantastic. I'm Jim Doherty, the Chief Development Officer for Acumen. I'm a neuroscientist by training. I've worked in pharmaceutical R&D for my whole career, 25+ years now, in both large companies and small ones, and been with Acumen for a little over a year and a half now. Perfect. Thanks, Jim. Let's just jump into it. Your lead compound is sabirnetug. It's an anti-amyloid-beta antibody with a unique selectivity profile. Can you take us through the history of the molecule, how it was developed, and how it came into use? Sure. Sabirnetug is a humanized monoclonal antibody that has high selectivity for a soluble amyloid-beta aggregate, which is referred to as oligomers. Sabirnetug actually goes back to the founding origins of Acumen, which was started really as an academic spin-off from Northwestern and USC, where some of the scientific founders of the company had been investigating and interrogating the toxic properties associated with these oligomers or ADDLs at the time that they were working on it. That early phase research and discovery effort turned into a research and development collaboration between Acumen and Merck. In fact, Merck was really the primary lead on inventing sabirnetug, which previously was referred to as ACU193. The unique properties of sabirnetug are really based on its selectivity profile that it has a high affinity and preference for soluble aggregates that have been shown to be toxic, which is different than amyloid-beta monomers or amyloid plaques, the amyloid plaques. Over a course of efforts, internally to Merck, they made a strategic decision to advance a different Alzheimer's candidate in the 2010 timeframe. Acumen was able to reacquire the entire data package and worldwide exclusive rights. Sabirnetug, as it sits here advancing through a sizable phase II s tudy, and we'll talk about, is a wholly owned asset of Acumen. We're really excited about some of the phase I data that was generated a couple of years back and now how quickly the phase II has been executed. You know, when we hear about amyloid-beta antibodies for Alzheimer's disease, we think of the producers of the approved drugs like lecanemab and donanemab. However, the therapeutic hypothesis for sabirnetug is quite differentiated from lecanemab and donanemab. Could you give us a synopsis on the oligomeric hypothesis of sabirnetug? Sure. The oligomer hypothesis is one that holds these soluble aggregates of the Aβ peptides. You know, Aβ is cleaved in a couple of different places and can produce a peptide of varying lengths, Aβ42 being the one that's most prone to aggregation and so forth. Oligomers are a consequence of one or more of those Aβ peptides coming together. The underpinnings of the oligomer hypothesis hold that these oligomers, as evidenced in labs, activity of founding labs and labs elsewhere, have shown that these aggregates have a propensity to bind to neurons at synapses and cause deleterious and disruptive effects, including calcium influx into cells, disruption of long-term potentiation, which is a surrogate for cognitive local membrane processes in hippocampal slices, as well as induced tau hyperphosphorylation. The strongest linkage between Aβ and tau, we believe, is mediated through oligomer toxicity. Could you go into some of the specific molecular properties about affinity and selectivity for the Aβ oligomers over Aβ monomers? Sure. At a minimum, we really don't, we have high selectivity versus monomer, which is really the important element to maintain, you know, target functional target selectivity and potency of the antibody. Monomer is orders of magnitude more abundant. The extent that your agent, your antibody, or other agent interferes with the monomers, we can sort of characterize that as target distraction, right? I think our selectivity for oligomers over monomer is 10,000-fold. It's very, very, very high in various different, you know, experimental setups. For fibrils or plaques, it's more in the order of 90-fold selective for oligomers over those other fibril/plaques areas. That really is, you know, we're testing the oligomer hypothesis with sabirnetug. I mean, it was originated as an oligomer-directed antibody, and we had great phase I results, which we talk a little bit more about. That has really helped facilitate the specificity and liquidity of the oligomer in phase II. We will jump into those data points. Very quickly, what were some of the key scientific data that enhanced your conviction in the oligomeric hypothesis? It's 30 years of research that really continue to call out and validate the toxicity associated with these different species. There are, you know, there's a broad spectrum of these species. I think it's to be determined which in particular conformations are the "most toxic of the toxic." These are things I think we're very confident in sabirnetug's profile for hitting a broad spectrum of aggregates from as few as a dimer, like low molecular weight to mid-molecular weight oligomers. Having that diversity of mechanism and coverage, we think, is likely to be beneficial and a negative, which yet to be elucidated precisely, which, if there is one or more, I mean, I would sort of take the view as it's unlikely, but probably a variety that function in numerous pathologics. Maybe just to add a little bit to what Dan is saying, obviously what's so successful about monoclonal antibodies as medicines is their specificity. You'll often hear people talking about amyloid targeting. Not that that's wrong, but it's probably way too broad because, as Dan's been explaining, amyloid biology, these proteins are very sticky. They tend to go out together and they have multiple, multiple forms. Different antibodies are recognizing individual forms. To answer your question around the oligomers, what caught my eye in the data, and as Dan was saying, there's 25+ years' worth of studies that are out there. It's a lot to summarize, but I would say there's both binding to synapses. Now you're talking about the sort of business units of cortical function, and these toxic oligomers are actually intercalating into the synapses and disrupting synaptic function. You're seeing consequences when it comes to functional measures. Things like synaptic plasticity is no longer as efficient, and even up through into the modeling, the disruption of ongoing cortical activity. I think that's really what is the differentiating piece for the hypothesis, because by targeting those forms specifically, we think we're going to have a differentiated effect on the pathophysiology of the oligomer. It's very fascinating. Okay, let's jump into some data. In July 2023 at AAIC, we had the first look at the phase I-B program for sabirnetug in early Alzheimer's patients, and then some patient-level data at CTAD a couple months later. Could you give us a brief overview of the study design, including the dosing levels of the drug? Sure. Jim, you want to take us through some of that? Yeah, for the INTERCEPT study, it is a phase I study, but it was a phase 1 study conducted in Alzheimer's patients. The population that we're studying with sabirnetug is early Alzheimer's disease, either MCI or mild dementia. That's measured through a number of different clinical scales. The important point is, even in the phase I study, we're looking at patients, and that really gives you the opportunity to do a number of things. In addition to the tolerability and pharmacokinetics, which are standard for phase I, the team was able to include a number of metrics that look at function along the way. From a pharmacokinetic perspective, the drug is well behaved and consistent with once-monthly dosing. That's why we've continued with that into the later study. From a safety perspective first, generally well tolerated. The issue for any amyloid-related therapeutic is ARIA and ARIA-E specifically. We did see a few cases of ARIA-E in the phase I study, a total of five across all the dose groups. Multiple dose groups, I'll just talk about the MAD cohorts, so a low dose of 10 mg, 15 mg, and 60 mg per kilogram in the phase I. A good dose range. We saw a total of five cases of ARIA, which works out to about 10%. Although it's a small sample size, that compares nicely with the available data sets for the marketed drugs. On the end, it's worth talking about the efficacy side of things since we were talking a bit about the mechanism of action. A number of the measures were designed to really give a feel for the effect that sabirnetug may be having on amyloid levels. We did look at amyloid plaque distribution by PET scans. Although that's not the primary effect for sabirnetug, we did see a nice reduction in plaque size over the three-month dosing period for the MAD cohort, about 20% reduction at the top dose, which is nicely in line with what was seen at a similar time point in the clinical program for lecanemab. Probably more interesting to us is we looked at a number of soluble biomarkers, and those are biomarkers directly of disease processes. Looking at Aβ 42 to Aβ40 ratios, sort of the next step downstream, looking at tau hyperphosphorylation, but then also going to the next level of integration and looking at synaptic markers. The consistent story is that you see a reversal of the aberrant levels of these markers, which is exactly what you predict if you're having a beneficial effect on amyloid levels. We're very encouraged by this data, and it really is a nice launching pad to take us into the current outcome data. The only point I'd add is the target engagement in phase I too. I mean, the safety, the biomarker activity was robust, and target engagement was really, I mean, given that we're targeting oligomers, there have not uniformly been established oligomer assays. The team developed a novel ELISA-based method of confirming target engagement of sabirnetug bound to oligomers as a complex. We see those related effects, and we also see sort of maximal target engagement at the 25 mg/kg and 60 mg/kg in the AD dose cohort. That was really another element to the totality of the phase I data that helped inform and underpin the phase II change. I do want to spend a minute on ARIA, which, you know, is a concern of the approved lecanemab and donanemab. As you pointed out, Jim, there were some observations of ARIA, but it was at a much lower rate than lecanemab and donanemab. Do you think that the lower rates were due to lower affinity for certain amyloid species, or possibly driven by the fact that sabirnetug is an IgG2 molecule with reduced effector function? I think it's both possibilities. I can tell you a little bit about, yeah, the IgG2 isotype was a development decision actually taken back in the Merck phase. That was intentional to sort of, you know, minimize microglial sort of effector function and response to essentially the inflammatory elements. I do think that given the preference for oligomers over plaques, it's equally likely that sabirnetug doesn't see as much plaque as some of these other plaque-directed agents. It's also important to establish that, you know, with this dose range in phase I, I mean, we're delivering a lot more sabirnetug drug relative to the other plaque-directed agents that have, you know, sort of an ARIA ceiling. In fact, when the phase I was designed, that 60 mg/kg SAD cohort was very much intentional, given that at the time lecanemab was being developed in phase I, their SAD study had a 60 mg/kg dose cohort, and three out of eight of the subjects that were dosed at that level experienced symptomatic ARIA hospitalizations. They sort of can't push the dose higher for some of those other agents. As we found, particularly with peripherally delivered antibodies, the higher dose equals greater central exposure. This dosing kind of window that we have for sabirnetug is safe and demonstrates our engagement. Do you think these ARIA observations affect your view of the oligomer hypothesis, or do you need to wait for some functional data? I think, in terms of safety overall, there are other non-Aβ agents that have elicited ARIA, right, M2 and so forth. I think the biology is probably pretty complex. It's hard to pin down precisely how these mechanisms play. As Jim mentioned, we did move the PET signal 20%-25% at the higher dose level. There could be some elements of that, although there's no correlation between the plaque reduction and the ARIA, which is another question to be answered in time. No, it's impossible. Okay, moving on to the Phase II ALTITUDE-AD study. In March, you announced enrollment completion in the study with a readout of 4Q 2026. Could you talk about the Phase II design and how the INTERCEPT data informed study design? Sure. Maybe I'll start, and Jim, you can comment. The phase II is a fairly, you know, well-powered phase II in Alzheimer's patients. This is the early AD population we employed in phase I, and this is kind of the standard population. This is mild cognitive impairment, and then dementia due to the early Alzheimer's disease. The study was designed with two active dose levels, one at 35 mg/kg dosing every four weeks with Q4W, and then a higher dose at 50 mg/kg dosing Q4W. It's an 18-month study, which is standard for a primary clinical outcome. We're using the IDRS as that primary outcome measure, and then we'll have CDR sum of boxes, ADAS, and the components of the IDRS, as well as all of the fluid and imaging biomarkers, both inwards PET as well as a PET substudy for the study. We did move quickly on the study. We have a first patient dose that was announced in May of 2024. We thought we might have to, you know, we exceeded our expectations in terms of our cohort. We We enrolled about 532 subjects, so roughly 180 for the cohort in roughly 10 months. As a small biopharma kind of in a space that's traditionally kind of the larger players' base, I think we're really encouraged. They did a remarkable job. I think it's a combination of the quality of the team, the ability to execute, as well as the value proposition of, you know, the suburban research based on phase I data. I think it's probably worth saying, I mean, as Dan said, the study enrolled in 10 months, so there was obviously a lot of interest in the study. That was certainly gratifying to us because the study is being conducted in the U.S., in Canada, in Germany, and in Spain. It's across a number of different countries. What I would say is that the level of interest that we saw was matched pretty much in all the geographies, which is great. There are certainly approved options out there now. That was a question that we often got, was that going to be an issue for us? Clearly it wasn't. We were able to deliver 532 subjects in less than a year. How about baseline characteristics? Do you see this cohort as being similar to the lecanemab and donanemab populations? The lecanemab and donanemab populations are actually a little bit different. If you look at Clarity and TRAILBLAZER 2, because Lilly employed the tau measure as part of their stratification in that study, those population things are a little bit more advanced than y AD. I think Clarity AD is one that is pretty consistent with what we saw in phase I and what we would anticipate from phase II. You mentioned, Dan, the IDRS as the primary efficacy. How is this scale different from CDR sum of boxes? Is this a strategic choice? It is a strategic choice, and it's one based on statistics. The IDRS is a composite scale that takes elements of the ADAS-Cog and the ACDS ADL or activities of daily living measure. It's basically been established as a more sensitive measure for this early AD population. It was developed at Eli Lilly. Our Chief Medical Officer is a former medical lead on the solanezumab program at Eli Lilly, so he's very familiar with the measure. If you look at donanemab, which is Lilly's drug, in the phase II that they ran, the TRAILBLAZER primary, they had iADRS and CDR sum of boxes. They hit on the iADRS. They didn't hit on CDR sum of boxes. In the phase III, they maintained the iADRS as the primary. They hit on both the iADRS and the CDR when you do a larger study. We think that just on the basis of reported outcomes and statistics, the iADRS is an appropriate one for the stage that we're in. I think it's the right choice. The doses you've chosen for ALTITUDE are 50 mg/ kg and 35 mg/ kg. Correct. Every four weeks. That's IV dosing every four weeks. Yes. For that low dose, 35 mg/ kg, what are your expectations in terms of ARIA and plaque reduction? I think in the phase I, we had the total of five phases. Three were at 60 mg/kg. We had one at 10 mg/kg and 25 mg/kg. I do think that there's the potential for the 35 mg/kg to differentiate from the 50 mg/kg dose cohort on ARIA. Maybe we'll have to see the data tell, but I think for the dose sort of strategy in phase II, we employed a good degree of dose modeling and leveraging a lot of that target engagement data to support the notion that at these levels, we should have, even at trough levels of drug, adequate coverage target, such that we're essentially the optimized treatment effect. For patient screening, you implemented a very unique strategy. It's a two-step process that uses plasma p-tau217 biomarker testing. Could you describe this process and the outcomes? Jim, you want to take that? Yeah, absolutely. As you know, there's been a tremendous explosion in the use and expansion of plasma-based biomarkers. Eric and the team did a great job in recognizing that, and in particular, p-tau217, which has turned out to be quite a sensitive marker for amyloid levels, interestingly, in AD patients. One of the challenges historically in Alzheimer's clinical trials has been, if you use clinical diagnoses alone to patients, what people found, and I'm sure everybody knows these data quite well, is that a number of subjects that were enrolled in earlier phase Alzheimer's trials probably didn't have Alzheimer's disease. They certainly had dementia, but they had dementia for other reasons, and the amyloid levels were showing that those folks were inappropriate. Of course, that's the last thing you want to do in a trial is have a significant population of folks who are not appropriate for the therapy. Certainly bringing in PET as a screening pool was beneficial. PET also has its own limitations. There is radiation exposure. It's much more time-consuming, invasive, and all those sorts of things. What you can see with the plasma-based biomarkers is they offer the opportunity for much simpler, much more rapid, and frankly, much more cost-effective screening. Eric and the team did was put effectively a prescreen in place where during the first screening visit, subjects were tested for p-tau217 levels. We established an internal cutoff, and you only move forward to later screening if you had a p-tau level above that screening cutoff. The effect of that was that in our phase I study, consistent with the earlier studies, a substantial fraction of subjects who went for a PET scan turned out to be amyloid negative. That changed quite a bit in the presence of the p-tau217 screen. You go from double-digit% without the p-tau217 screen to a 15%, 16% failure rate in the presence of p-tau. That had major consequences for efficiency in the study. It means that many more of the patients who are going for the final screen of either PET or CSF markers were positive for the study. It saved money, it streamlined, and it did a number of positive things. It really just is an interesting highlight for, I think, what will be a much larger use for plasma-based biomarkers in the future. Yes. You entered into a license and collaboration agreement with Halozyme to develop a subcutaneous formulation with its ENHANZE technology. Thank you. What's the overall strategy and development plan for this program, and when can we expect some updates? We think there's a lot of value and flexibility for patients for having a subcu formulation for sabirnetug. As you say, we have the data from the phase I study, which is our initial look at the pharmacokinetics of delivering with the enhanced technology. We use those data to plan for the future. We were able to generate the PK data we wanted to be able to see to continue forward. I think at this point, it really comes down to something we've already been talking about, which is what's the appropriate dose level of sabirnetug to target for a subcutaneous formulation. Given how much work there is involved in delivery devices and that part of the process, we don't want to move forward with next steps until we have a better sense of the appropriate dose to be targeting for subcutaneous. Having said that, the team is very much already working on plans and what the pathway would look like and what devices we would want to qualify for the purposes. We think it's going to be an important part of the program, but timing is everything. Do you think you could possibly switch a proportion of patients from the ALTITUDE study to a subcu version? Much more likely that we would do something like that as part of a phase III confirmatory study than the current active study. Again, since we enrolled so quickly, the study is moving very quickly, and we really want to get those data on the doses at 35 mg/kg and 50 mg/kg before necessary. Is there anything else you'd like to mention about phase II before we move on to? No, we've got it to top line results late 2026. Yes. The study is going really well. Retention is good. We're really encouraged with the progress of the study. Perfect. The last thing we want to touch on is something that we're really excited about, is a recent agreement you entered into with JCR Pharmaceuticals for their blood-brain barrier crossing technology. This is a technology we believe will be very disruptive to the neuroscience field for CNS drug discovery. Can you tell us about this tech and how you plan to leverage it for sabirnetug and possibly any other antibodies? Sure. Yeah, we agree with you that this technology is really fascinating and very, potentially very, very interesting for neuroscience. As fantastic as monoclonal antibodies have been in the treatment of disease, one of the key challenges for brain disorders is that only a very small fraction of the systemic dose actually gets into the brain. Just for context, the general number that gets tossed around is about 0.1% of the systemic dose actually gets into the brain. Anything that one can do to meaningfully increase the fraction that gets into the brain, you would predict is going to have pretty meaningful effects on programs overall. With that kind of thinking, we started doing a scan of the landscape of looking at the various technologies that are out there. In addition, we've chosen, as you say, to work with JCR around their IZCARGO transferrin-based system. There are other systems out there that people are interested in. For us, the decision was relatively straightforward. There is the most available clinical validation for transferrin as the appropriate carrier into the brain. More than that, JCR has a marketed drug that uses this technology for Hunter's syndrome. For us, that was very attractive as a clinical level of validation for this. Your technology in collaboration with JCR is the so-called enhanced brain delivery tech? Yes, we've characterized the terminology. terminology. Our terminology is enhanced brain delivery or EBD. As Jim just described, we're using the JCR carrier technology with Acumen proprietary Cargo, sabirnetug and sabirnetug-like monoclonals that again have kind of a high selectivity towards our favorite target being oligomers. The focus is going to be oligomers. Yes, it will continue to be oligomers. We do like the way the arrangement works with JCR. We do have the ability to advance as many as two candidates under the option target. All right, in the very last minute, our last question is, you know, if we're sitting here 12 months from now, what are the accomplishments that you'd love to tell me about from this past year? Sure, the next look out over the next 12 months, it's going to be an amazing run here. We'll have preclinical candidates out of the EBD program early 2026. I think characterizing kind of that strategy, as we do, again agree with your assessment that that technology long-term is going to be really compelling in the space. We'll be, in September 2026, sort of working on final patient visits and other elements of closing out the study and logging the database and getting very close to data. It'll be an exciting time. Amazing. Thank you so much, Acumen team, for joining our Cantor Healthcare Conference, and thank you to our audience. Thank you, Samantha. Thanks, Samantha.