Thanks very much. It's my pleasure to be moderating this panel with Dan O'Connell, CEO of Acumen Pharmaceuticals, as well as Jim Doherty, Chief Development Officer. Maybe I'll kick it over to you, Dan, to just give maybe a three to five-minute overview of the company and where things are at, and then we'll do Q&A.
Great. Thanks, Paul. I'm Dan O'Connell, the Chief Executive Officer of Acumen. Pleased to be here at the Stifel conference again. Acumen is a company that's advancing potentially new treatment options for people living with Alzheimer's disease. Our lead program is sabirnetug, which is a monoclonal antibody that is highly selective for synaptotoxic Aβ oligomers. We're excited to be advancing sabirnetug in a robust phase II study that's now targeted to read out late next year. In addition to IV sabirnetug, we're also advancing a subcutaneous formulation of the product candidate that's completed a phase I study earlier this year. More recently, in the middle part of the year, we have announced a collaboration with a Japanese pharmaceutical company called JCR, which is combining JCR proprietary transferrin-directed carrier technology with Acumen proprietary oligomer-directed antibody cargoes.
We think this fundamental science and hypothesis that we at Acumen have been committed to exploring is the notion that soluble aggregates of the Aβ peptide, principally Aβ oligomers, are a distinct and potentially potent target to slow the progression of Alzheimer's pathology and offer an attractive risk-benefit profile to patients.
Great. Maybe let's talk a little bit about just like the historical oligomer hypothesis, right? Where does it come from? What's the support for it? I know you get this question a lot, but to what degree has this been at all partially tested by other antibodies in the past?
Yeah, sure, Paul. I think that the original oligomer hypothesis sort of co-generated in the late 1990s, early 2000s, in a variety of different labs and with researchers. Really, the Acumen founding scientists were some of the first researchers to characterize the toxicity of these soluble aggregates. What's interesting is that if you take Aβ monomer, which is a normally occurring peptide, the physiological function of which is not yet well understood, the 42 amino acid sequence has a propensity to aggregate.
In some of the original sort of founding science around Acumen, these aggregates, which the founders referred to as amyloid-derived diffusible ligands, had a propensity to bind to neurons in synapses and disrupt long-term potentiation in hippocampal slices and really kind of address the question of, if Alzheimer's is a disease of memory and cognition, how do we tie that biology directly back to synaptic function? We do view Alzheimer's as a disease of synaptic loss. We believe that oligomers play a really deleterious and important role in disrupting neuronal function, synaptic health, and inducing neurodegenerative processes. There is an abundant base of evidence that has been generated over the last couple of decades to really highlight the toxicity associated with these soluble aggregates. I think there are a variety of different forms and configurations that are implicated in that pathology.
With the Acumen programs, we're really trying to neutralize that toxicity and s abirnetug has shown really the ability to bind a range of oligomers in various sizes and configurations and in experimental models, essentially alleviate or stop the toxicity in those experimental setups. I think you asked about the history. The field has learned over the last 15-20 years what not to do. I think we are in the moment where we're positioned to read out ALTITUDE-AD with sabirnetug and really help inform what to do, which is really a preference and a direction towards neutralizing toxic species to open up, unlock greater safety and efficacy for patients.
I think it's interesting to see the evolution too, because Aβ has been known for some time, obviously, to be central to Alzheimer's disease. As we've gotten so much progress in the last couple of years with anti-Aβ type therapies making it into the commercial space, we're starting to think about what else besides Aβ, which is a good discussion. It kind of implies that Aβ is sort of one construct. As Dan was saying, Aβ is actually a pretty complex biology. Thinking about what type of Aβ that you're interacting with, we think, is really important. This is where the data comes in that Dan's referring to on some of the synaptic biology, both functional, immunohistochemical. There's a lot of evidence that these smaller oligomers are actually very toxic to neurons and to synapses.
I think that is a component that is getting a little bit lost in the thinking around plaque binding or other forms of Aβ, too.
Very good. Makes sense, Jim. Thanks. Do you want to talk about the phase I/II two data and the effect you saw in oligomers and plaques at different doses? I think that'll kind of set the stage with the hypothesis testing for this bigger readout coming out.
Yeah, sure, Paul. Let me lead out on that. INTERCEPT-AD was our phase I study, which is conducted at a single ascending dose and multiple ascending dose cohorts exclusively in Alzheimer's patients and early AD patients. That was important to establish aspects around safety, target engagement, and we also looked at biomarker effects. Fortunately, we actually yielded just what we needed in terms of information on the safety profile, which had just five cases of ARIA across 48 patients exposed to drug target engagement, where we used a novel assay to measure the complex of Aβ oligomers bound to sabirnetug in cerebrospinal fluid. That really was a novel innovation on the part of the team, as essentially measuring oligomers in a native fashion has been difficult and unpredictable.
We chose to take a target engagement assay in phase I that showed essentially an increase in signal as we escalated doses, not only in the SAD portion, but then in the MAD. On the biomarker side, we looked at both imaging, so amyloid PET, as well as CSF and plasma biomarkers of significance in the disease. On amyloid PET with only three administrations, this is a relatively short duration study intended to get us to move quickly into phase II with three doses at the high doses of 25 mg/ kg dosing every two weeks and then 60 mg/ kg dosing every four weeks.
We saw about a 20%-25% reduction on the amyloid PET signal, which is really consistent in terms of the magnitude and the slope of the effect with LEQEMBI, for example, which also has some effects on protofibrils and oligomers as well as fibrils and plaques. On the fluid biomarker side, in CSF, we saw a normalization of the Aβ 40- 42 ratio, which is consistent also with the target engagement of oligomers. A shift in the level of p-tau181 in CSF was also observed. We looked at two synaptic markers, both neurogranin and what is called VAMP2, a postsynaptic and a presynaptic marker of synaptic health. Each of those were moving again in the right direction. Interestingly, VAMP2, the presynaptic marker, achieved nominal significance at each of the three MAD dose cohorts.
For us, that was an interesting observation of sort of downstream effects on biology as a consequence of limited duration dosing of sabirnetug. In plasma, the similar analytes were evaluated, both Aβ as well as p-tau217 and GFAP. Those are a little more noisy, just given the time sequence of when those observations or samples were obtained, but still trends. The overall consistency, I think, of both the imaging and the fluid biomarkers post-CSF and plasma were really encouraging to us to see we were in a dose range that had a biological effect, and we had confirmed the target engagement of oligomers with the ELISA assay.
Okay. Great. For the II-B, you've taken forward a couple of doses. How would you kind of frame the hypothesis testing of each of these, right? I think you have a dose that is more robust on plaques and a dose that is going to be more reliant on the oligomer factor, or I guess maybe not totally, maybe not just oligomers, but more biased to it. Can you kind of frame that for people?
Yeah. I mean, I think that is kind of how we framed it. I'll invite Jim to kind of elaborate a little bit just based on some of the modeling and thought that went into the dosing strategy for the phase II.
Yeah. I think it's a case of really being able to apply the target validation assay that Dan was talking about. Target engagement, really, we're able to show across a range of doses in phase I, increasing target engagement with oligomers and up to and achieving a saturation point above 40 or 50 mg/ kg. It really does allow then to set the dose range for phase II to capture that range of high affinity binding to oligomers. The lower dose of 35 mg/ kg will be certainly targeting the low abundance of oligomer species. As you get to the higher dose, it does also give you some interaction with plaque either directly or interacting with oligomers associated with plaque.
It really is a matter of identifying a dose range based on the target engagement assay that was done in phase I.
Okay. Okay. And the high dose, do you think that the expectation is that will be as robust on plaques as lecanemab and donanemab?
As robust on plaques. I mean, lecanemab and LEQEMBI have different.
I guess they're a little bit different.
They are a little bit different, which is interesting.
The clinical effect is the same.
That, of course, Paul, is the crux of the issue, right? Donanemab, which arguably drops plaque signal faster and to a greater degree than LEQEMBI, has equivocal efficacy. So we do not think that really you are going to yield more efficacy by just clearing plaque.
Wet with plaques. I get it.
Yeah. I think that's, and I think if you look at our phase I results in CLARITY-AD for LEQEMBI, they actually had a three-month PET scan. If you look at the Intercept phase I study, the slope and degree of change in these elevated doses at 25 and 60 is pretty consistent. I'd be surprised if we don't have similar effects as LEQEMBI at the high dose, quite frankly.
Okay. Okay. Is the thesis here that the high dose with the same plaque impact plus robust impact on oligomers will drive better clinical efficacy?
I think we've set it up as an experiment for a variety of different outcomes, right? I mean, I think the 35 mg/ kg could have less effect on plaque, potentially a better safety profile vis-à-vis ARIA and all of the clinical efficacy. We think that would be a really.
Right. You could have a safer dose with the same efficacy there.
Correct. Maybe not as much plaque reduction, but maybe really robust effects on these fluid biomarkers, which the field right now, I mean, the innovation around the amyloid PET modality was fundamentally enabling for the field. I think we're shifting now into more of these fluid biomarkers and looking at sort of a constellation of effects, which are more readily obtained through these fluid samples. I could see 35 behaving a little bit different than 50, having all of the equivalent efficacy, but maybe a better safety profile, maybe different effects on some of the p-tau species and the synaptic markers.
Okay. Okay. Okay. How do you think about the competitive positioning over time in a market that is changing, right? I mean, we're going to have one or multiple sub-Qs for initiation, probably approved in the next few years. You've got the trontinemab brain shuttle piece that if that early data are replicated, would be transformative. Obviously, it's like a big if. Where do you see yourselves fitting in over the mid to long term?
I mean, over the mid to long term, I think we have, on the basis of a compelling risk-benefit profile of sabirnetug in any format, IV, sub-Q, or potentially EBD, as being a treatment of choice in that evolving and growing landscape. I think we've had conversations over the years of where is, when is this developing, where is this going? I think we're finally seeing a rate of adoption addressing some of the infrastructure issues, some of the bottlenecks. We have a core belief that anti-Aβ treatment is going to be a cornerstone of addressing amyloid pathology as part of a treatment regimen that's also likely to include other agents and modalities. We think Acumen products directed at toxic oligomers have a distinct and potentially clinically valuable role to play in that armamentum that is likely to develop.
I mean, this is such a huge area of unmet need, both in the early AD space, and we could talk a little bit about preclinical. We think the field is just opening up and excited to think that we have data sets coming out next year to inform EBD development candidates early 2026, and then reading out Altitude later next year.
Yep. Okay.
There's room for improvement and room for other options. The field is kind of moving forward, and we want to have a real impact on that growth and development.
Makes sense. For the sub-Q, can you talk about where your efforts are there? If I remember correctly, the highest dose might be challenging with the sub-Q. Is the right way to think about it that you believe the sub-Q is viable, but it is a little bit more reliant on the low dose results?
A simple answer is lower is easier, for sure.
Of course. Can you actually do that high dose?
Yeah. I think the work that we've done and the ongoing analysis suggests we would have a path to a sub-Q.
Even with the high dose of that.
Potentially with the high dose, correct. Yeah. I mean, they're still subject to lots of other potential issues, but there is a potential plan to address that if, in fact, sub-Q becomes an important or essential element to the program. I mean, we do think IV and sub-Q are options and will coexist and be appropriate in any given situation.
Yep.
It's funny you mentioned trontinemab. Trontinemab is IV only, right? It doesn't have a path to sub-Q.
Totally.
As we think about, if that's really on the horizon, it still sort of keeps the infrastructure of this IV deployment, patient scanning. That paradigm will persist into the future.
Yeah. I guess it just depends what their ARIA rate is, right? If they really, really lower it or not, right?
Right. We don't know. I mean, as they get into a broader base of patients exposed, we'll know.
Totally. You have the transparent question too. Yeah. Okay. Makes sense. Can you talk about this JCR deal you did? I feel like this whole idea of having a brain shuttle went from something that felt like it was this unique differentiator to now like table stakes, like you got to be there. Are all these brain shuttles and transparent contracts, are they all like the same? How should we kind of think about it?
I guess the way I'd make a couple of initial comments and invite Jim to comment, this stuff is non-trivial, right? It's not Legos, and it's not like they're not all.
Can't just like start a transparent company tomorrow?
Actually, you could, but how sustainable and viable that might be.
Yeah. That makes some good slides.
It makes some good slides. Yeah, no, you're right. You could probably start a company and raise money with that stated strategy. I think the devil's in the details. I think there has been a lot of development in this space over the last 15. Again, it's like people have been at this for some time. It's only until recently, particularly with the trontinemab data where you see gantenerumab, a failed Aβ monoclonal, get paired up with a transferrin carrier, and all of a sudden the profile changes pretty substantially and in the right direction, both in terms of safety and the amyloid PET reduction. That proof of principle definitely has fueled a lot of interest. I mean, we've been paying attention for an extended period of time.
JCR, we internally prioritized transferrin as the most clinically validated receptor-mediated transport target for doing anything in this enhanced brain delivery space. Evaluating a variety of options in that particular domain, JCR stood apart as a company that has been in the field for 15 years, has an approved product in Japan, was very motivated to have an opportunity to work in the Alzheimer's space with Acumen. I think we have a very sort of efficient and effective collaboration with JCR, both in terms of the technology synergies as well as some of the financial considerations and the team's ability to work together. It has been a really fruitful effort from our perspective. I think JCR is equally excited about the differentiation that an Acumen cargo or payload brings to the AD space.
As I mentioned, we're intending to have a really sort of a preclinical package early next year to inform a development decision.
Okay. That's great.
Yeah. I would say the receptor-mediated transcytosis has been around for a long time, and there are a lot of different ways to do it. I think as you were anticipating, Paul, there's a lot of complexity too. In many ways, it's a real discovery effort. You're trying to mix properties. It's important that you've got the right carrier. For all the reasons Dan talked about, we've selected transferrin. Having the right cargo is also important as well. Really your final product has got the profile of both components. Lots of people are looking at multiple different carriers. I think as a field, it's sort of, even though it's been on a slow delivery process, it's probably still early days. I think we're likely to see more and more products being engineered this way.
If you think about it, just using as an example, sabirnetug, I mean, when you think about the possibilities, we talked about efficacies of being able to deliver your antibody into areas of the brain that you cannot readily reach without this sort of technology, you could have a different efficacy signal. From a delivery perspective, being able to reduce the volume and the amount of drug that you need to deliver to have your therapeutically relevant effect, you have got a manufacturing opportunity and it gives you some more flexibility. It has turned out from a safety perspective, being able to target transferrin allows you to enter the brain in a slightly different way via the capillary beds where the transferrin receptors are localized. That is one of the key hypotheses for why trontinemab is seeing lower ARIA rates than the parent gantenerumab did.
Yep. Yep. Makes sense.
I think just to kind of reiterate the point, I mean, working with JCR, we've been able to explore a variety of different carrier configurations. There is diversity in the way we're evaluating what options we think are most tenable and advantageous. We are hopeful to have at least two constructs or as many as two, I should say, that would warrant moving into IND-enabling efforts.
Okay. That's great. One more time on previewing the big phase II-B readout. How would you define a clinically significant difference in efficacy? Understanding it's hard to compare across studies. Similarly, what would be a clinically significant difference in the rate of ARIA?
Yeah. So I mean, I think it's fair to characterize the current approved agents have a disease slowing of 25%-30% in the early AD population over an 18-month period. I think we're starting to see the real-world evidence that underpins the adoption of those agents that extends 36, 48 months, which is good for the field and for patients. That 30% threshold, I think, is on the efficacy side, whether it's a iADRS or CDR-SB, some of the boxes seems to be a benchmark out there, quite frankly. Then on the ARIA side, it's perhaps a bit, well, and the other thing is that's aggregate efficacy across the entirety of the study. I mean, if we look at the subsets in some of these other studies, the E4 homozygotes and E4 carriers, they tend not to get the full benefit of these agents, right?
That's either as a consequence of ARIA where there's been dose disruption or what have you. I think particularly in the safety realm, overall rates of ARIA at the 10% threshold is sort of table stakes, I think, right? 10%-12%. We'll see where trontinemab goes. There are other considerations, safety considerations, I think, with trontinemab that may come into play as they get broader later into development. I also think we had kind of an interesting finding or observation, I should say, in phase I with—we had six E4 homozygotes that were exposed to drug at various dose levels, none of which developed ARIA, which really was statistically not anticipated, right?
It's very probable that one of the six subjects would have had some, at a minimum, asymptomatic ARIA, which would be interesting to see kind of how ALTITUDE-AD reads out and whether there are subpopulations that have a higher on that need that actually evidence a better safety profile on sabirnetug.
Yep. Yep. Okay. Great. What do you think about the preclinical studies that are ongoing?
I think it's really exciting for the field. I mean, I think.
Do you think they'll work?
I'm hopeful that we see a benefit. I mean, I think donanemab in that population, given its fixation or its epitope for pyroglutamate Aβ, which may be very low abundant in these low plaque patients, I don't know if that's the best agent, but I think we're going to find out soon. I do know that I think we have a lot of confidence that sabirnetug and/or other related oligomer-directed agents could be uniquely beneficial in that early course of disease, right? I mean, if you look at the Jack curves, there's a growing base of evidence that oligomers are sort of elevated and pronounced just in the early phases of plaque formation. One could think about those, an agent like sabirnetug in that population as being really disease stage appropriate.
It is a little bit beyond the scope of what we can undertake as a small company running a fairly large phase II study today. But as you think about what the future might hold and if, in fact, that population becomes sort of eligible or qualified, there is a validation of intervening in that population, we do think that the Acumen portfolio will be of high interest potentially in the future.
Yep. Yep. If the phase II-B data are great, how much capital do you think you might need to get this to market and get it launch ready? Is that something you could do, or do you think this data is more of a partnership with Catalyst?
Yeah. No, I think, Paul, I think we would envision having a partner for phase III. I think that's just the reality of the scope of the opportunity and our commitment to try to be as expeditious as possible, right? We're a 62-person company. I don't know that we quickly become operationally positioned to optimize the execution of phase III. Team's done a phenomenal job in phase II. I mean, we enrolled it lightning fast, and we're executing for phase II. Really pleased with that execution and sort of the effort put there. Phase III is a little bit of a different animal. The study design is likely to be consistent with what we've done in phase II, but the scale is greater.
I think having a partner in phase III would be an attractive way to expedite the further development of sabirnetug on the great ALTITUDE-AD outcome.
How much of the CMC investments have you kind of saved and gated by this phase II study?
I think it's prudent in drug development to always be stage-gating those types of investments. We're working with Lonza as part of a long-term partnership, and they understand the stage we're at. They've been a good partner for both drug substance and drug product. We'll continue to leverage that relationship as we look for data to underpin the next steps and make sure we minimize the white space between phase II and phase III.
Okay. Great. Understanding that the cocktail hour starts in a minute, anything else you want to keep people here with?
In less than a minute, I just think you're starting to hear some of the enthusiasm and optimism in our space. It's exciting, right? I think it's well-based. I think the field is moving forward. I think there's lots of opportunities ahead for people that are facing a diagnosis of Alzheimer's. We're super excited that ALTITUDE-AD is fully enrolled. We've got the first patients now rolling into the OLE. I think the study is such that it really should provide the evidence in support of this distinct mechanism that we've been in pursuit of for some time. Exciting times ahead and happy to be part of that future.
Yeah. Okay. Best of luck. Thank you very much.
Thanks, Paul.
Appreciate it.
Thanks, Paul.