Great. Thanks very much, everybody. It's my pleasure to be moderating this chat with the Acumen team. Maybe we'll have Dan O'Connell, founder and CEO, talk for a bit about just the background of the company and maybe set up, Dan, 2026 is, I guess, sort of a big year for you guys. Then, yeah, we can do more specifics. That sound good?
Yeah, that sounds great, Paul. Thanks, for having us. So, yeah, I can lead out a little bit on kind of, Acumen and where we are this year. It is a massive year for us that we've been building towards for some time now. Acumen is a biotech company working to advance better treatment options for people afflicted by Alzheimer's disease. Our lead program, sabirnetug, is a humanized monoclonal antibody, an IgG2 monoclonal that is highly selective for toxic soluble Abeta oligomers, and we think that's a distinction that will serve it well in the clinic. We are currently running a phase II study, which we refer to as ALTITUDE-AD, which is a sizable, two active doses versus placebo. We can talk a little bit more about ALTITUDE.
We're also working on a subQ format for sabirnetug. Then thirdly, as part of the portfolio, we've had some recent news to support our enhanced brain delivery program, which is a program that employs a transferrin receptor-mediated transport of Abeta oligomer antibodies such as sabirnetug and sabirnetug-like products. One of the things I think that a distinction for Acumen is our persistent pursuit of innovation and our ability to execute. This is really a year of last year was a big year of execution. This is a year of data and catalysts. We have the EBD data. I'm sure we'll have a chance to talk a little bit about later on the call.
More importantly for that phase II ALTITUDE-AD study, we'll read the 18-month primary outcome and other related outcomes late 2026. That's intended and positioned as really the first deliberate, intentional validation of the Abeta oligomer hypothesis clinically. Massive year for us and excited to see the progress we're making.
Great. Well, Dan, maybe what we can dig back into the whole scientific side of oligomers and some of the phase I data. You know, now that we're pretty close to this to-be readout, how are you setting up what constitutes success? Like, what actually would be validating of the oligomer hypothesis in your mind?
Yeah. Thanks, Paul. I think it really comes down to an objective better risk-benefit profile for patients. This is sort of the ratio of treatment efficacy relative to the risks associated with administration. That efficacy safety sort of ratio, we think sabirnetug can potentially deliver on greater efficacy, so 30% or better, closer to 40%. You pick a number well above 30% as an efficacy measure at an 18-month endpoint. Then safety in terms of the ARIA risk, not only you know across all genotypes, in addition to relative to the current approved products.
Makes sense. Do you feel like, I guess, I'm not trying to pin you down too much, but as it relates to, like, the effect size or an ARIA rate, like, are there thresholds in your mind that kind of reflect differentiation from standard care beyond the signal-to-noise ratio?
Yeah. Well, I think the convention sort of on the street is to think of 30% slowing over 18 months as sort of a benchmark for the current two agents. I don't think there's much of a distinction between the two, quite frankly, on efficacy. I think from a safety perspective, you know, you're in the 10%-12% rate of ARIAs. It would be a benchmark, certainly for the approved agents.
Right. Yeah. Okay. Maybe talk a little bit about the study itself and the two doses you selected.
Sure.
You know, you have a dose that is more competitive on plaque busting and then a dose that still lowers plaques, but maybe not as robustly, but also still seems to have this oligomer effect. Maybe just go back to the one B, give the context around those doses, and then, you know, again, like, how are those doses testing different hypotheses in your mind? Like, how would you frame it?
Yeah. Thanks, Paul. Let me lead out and I'll invite James Doherty, our Chief Development Officer, to provide additional color, particularly on the modeling employed to precisely hone in on the doses for phase II. I think in phase I, excuse me, for INTERCEPT-AD, we did see both of the high-dose cohorts in the MAD portion of study, which involved three administrations of drug. So essentially a three-month time point for the imaging biomarker outcomes. At 60 mg per kg and at 25 mg per kg dosed every two weeks, we see plaque reduction as measured on PET of 21%-22%, roughly.
Essentially a little bit higher, a little bit more rapid in the 60 mg per kg. Both essentially achieving that kind of the magnitude and slope of reduction as, say, Leqembi, which had a three-month time point in Clarity AD. We embraced the amyloid-lowering piece of that. I think the other findings in INTERCEPT were the safety profile. We had a total of five cases of ARIA E, three each at the high dose of 60 mg per kg dosed three times, and then the one each at 10 and 25 mg per kg. We, you know, objectively, it looked like there was a dose relationship between the ARIA E and the higher dose cohorts.
Coming out of the INTERCEPT-AD, we also read out a series of fluid biomarkers, which again, were supportive of the pharmacodynamic effects that sabirnetug is generating in patients with just three administrations and sort of factored that into some modeling and concluded with two doses for the phase II. Maybe Jim, you wanna elaborate on some of the methodology there and kinda how we're thinking about, you know, the 2 doses and whether they're distinctly different or testing which of the hypotheses in the clinic.
Yeah, happy to, Dan. You know, I think, first, Paul, thanks for having us here today, and thanks for having me. You know, I think the way we think about it is, you know, this all comes down to A-beta that's obviously the target and many of the competitor antibodies referenced are also going after the same protein. But of course, we all know the protein's pretty complex, and it tends to be sticky, and it tends to oligomerize and form larger and larger structures. When you think about individual antibodies, it really does come down to the epitopes that you're targeting and what you're binding to.
We know from our phase I studies, as Dan said, he gave you a lot of the outcome data, but we also had a pharmacodynamic assay included in this, where we're actually able to take CSF samples from patients and measure the bound complex of sabirnetug, as well as the soluble oligomers coming out of the brain. Obviously, that's tremendously valuable for any phase I study of a pharmacodynamic assay that gives you a sense of how your various doses are performing. In this case, it also really gives us good target engagement data because this is amyloid, soluble amyloid in the brains of Alzheimer's patients that you're complexing out and collecting in CSF. The doses that we selected came out of the effort from phase I to plot the concentration of this bound complex of oligomer plus antibody that we're collecting.
You can see a nice asymptotic relationship as you go to the higher concentrations tested in phase I. Short, long story short, the two doses for phase II , 35 and 50, nicely bracket the steep part of the curve, for the lower dose and the more asymptotic part of the curve at the top end. We think we've got it nicely bracketed between the component where you're binding oligomers with higher affinity, and then at higher concentration, you still have that oligomer binding, but you're starting to pick up some of that lower affinity binding to plaque that you're talking about.
It's not really testing two hypotheses per se, but it is giving us a nice separation in dose range for a more selective binding effect to oligomers only, and then at higher concentrations of antibody, that same effect on oligomers plus more binding to some other larger species like plaque.
Yeah. Makes sense. Thank you, Jim. Maybe going back to the whole scientific piece behind this oligomer hypothesis. I'm sure you could spend an hour talking about this, but in a more succinct way, like what's the history and what has made this hypothesis so appealing? Then what would you say to someone who says, you know, Biogen had talked about aducanumab as having, you know, potency in oligomers or lecanemab as well. I mean, you can go back to some of the 2015 transcripts, but they really actually, Al Sandrock's talking a lot about oligomers. I guess maybe talk about, one, you know, the hypothesis itself and why you feel like it's well supported. Then two, like, how do we know that it hasn't really been tested yet already with some of these other antibodies?
Yes, sure, Paul, of course. I think what's distinct about sabirnetug is, you know, it originated and was raised against oligomers. These soluble... It's not just soluble Abeta, but soluble Abeta aggregates that have been shown to be potent toxins to neurons binding to synapses and inducing tau phosphorylation. There's an abundance of evidence. I mean, even today, you know, we're at the ADPD meeting, and the experimental setups to evaluate Abeta toxicity, you know, routinely and almost exclusively involve the use of Abeta oligomers as sort of the test of Abeta toxicity in any preclinical workup.
I think what we've been encouraged, particularly, you know, given that origin story for sabirnetug and its desire to preferentially and selectively clear, remove, and neutralize toxic Abeta oligomers, is that we have phase I data that confirm the safety profile, the target engagement of oligomers, and then consistent effects on biomarkers. You know, that was a fairly pronounced effect for a short duration study and really supported our ability to move quickly into phase II and really also serves as the foundation for why we think phase II is, you know, has attractive risk essentially probability of success in terms of the outcomes later this year.
Yeah. Okay.
I think.
Okay. Great.
You know, the other thing I would add, Paul, is that all these antibodies are targeting Abeta, and so we tend to say Abeta like it's all one thing. Of course, as I was talking about, there's this sort of rich complexity of different size species, all of which have different epitopes. I think of all these antibodies as having a preferring conformation, and then at higher concentrations, lower affinity, other effects as well. That's not simply the case for our antibody. That's the case for lecanemab, donanemab, any one that you wanna talk about. I think one thing that gets overlooked often, and we do talk about it vis-à-vis sabirnetug, is monomer binding. Monomer, which is, quote-unquote, "the normal protein," is expressed in really high concentrations in the brain.
If you have a high monomer binding potential, then despite your other affinities for the different species, you're gonna end up losing much of your antibody to monomer binding. We see one of the properties of sabirnetug that's really important is not only the oligomer preference, but a lack of preference for monomer binding. You got a nice functional selectivity between the two, and that's often underappreciated for some other antibodies that have high monomer binding.
Yeah. Yeah. Okay. Okay. Maybe let's switch gears in the last kinda eight or so minutes or 10 minutes here and talk a little bit about your guys' efforts towards the subQ, and also the shuttle side. But maybe on the subQ piece, maybe talk about, like, the work you've done there so far, the data you've generated. Then, you know, I've asked you this before, but should we think about both doses in this 2 B as being viable from kind of a subQ that can replicate that pharmacology, or is it, or is it just a low dose? Like, how much kinda work have you done there to really fully de-risk this?
Paul, I think what we're looking to the ALTITUDE-AD results to inform the next step with subQ. I think we've gotten to a nice place with what we've done so far. I don't think we have much of an update, quite frankly. You know, we completed the phase I healthy volunteer study, you know, about a year ago, and we have that information and continue to you know, map out a strategy going forward. I do think it'll depend on which doses is, essentially has that best risk benefit profile coming out of ALTITUDE-AD.
Okay. Makes sense. If the highest dose looks clearly the best, that's doable from a subQ perspective.
Based on the work we've done, yes. I mean, there's a path towards that.
Okay.
Yeah. I mean,
Okay.
That of course would necessitate, you know, a differentiation that would warrant, you know, pursuit of that particular format. We, you know, we do think.
Right
IV is still part of the potential regimen for sabirnetug.
Yeah.
In addition to subQ.
Yep. Yeah. Okay. Makes sense.
I think in the time that's left, I mean, but let's just quickly hit on ALTITUDE-AD too. I mean, I think we talked about outcomes, right? I do wanna just emphasize sort of the execution quotient of, you know, standing up that study as a small organization working in the Alzheimer's space. I mean, we have a team that's very dedicated to pursuing innovation and execution, and I think we launched the study May 2024. We completed enrollment of 542 subjects in roughly 10 months in March 2025. You know, now the study's being conducted as a registration quality study. As we've established, we have these two doses, each of which could be safe and efficacious to patients.
It's exciting to be thinking about reading out the outcomes later this year. The other potential point to make on the outcomes is we are using the iADRS as our primary, and I know there's some debate in the field about kinda the preference towards CDR Sum of Boxes. That is a key secondary endpoint. But based on you know, our insights, you know, we chose iADRS because it tends to be a bit more sensitive in this early AD population that are employed and enrolled in the study.
You'll have CDR, ADAS-Cog, ADL data, correct?
Great. Yeah. I mean.
That's great.
You can't get to the composite without ADAS and.
Exactly
You will get there.
Yep. Okay. All right, great. No, all fair. That is good context, Dan. Thank you. Yeah, maybe let's kinda cover the JCR collaboration and some of the updates there you recently provided.
Yeah. I think there's a lot of enthusiasm about blood-brain barrier transport mechanisms. You know, we sort of started to landscape the space in 2024, as some of the early Roche trontinemab data started to serve as an interesting proof of principle, taking gantenerumab, an antibody that, you know, didn't have late phase success, and fundamentally changing the profile. I think that was a sort of a bright line event for the field, and one which, you know, was a call to action for us as we think about future directions and expansion of our portfolio.
We looked at a number of different receptors and a number of different platforms and a number of different partners and arrived at a conclusion to collaborate with JCR Pharmaceuticals, which has been in this space for almost 20 years, working on transferrin in particular, and developing, I think, unique insights and expertise around the carrier technology associated with their IZCARGO and J-Brain Cargo. We're happy to sort of step into that collaboration in partnership with JCR late 2024, and then formally announce it in the middle of last year in July, which is when we publicly disclosed the option and license agreement that we have with them, and guided towards having development candidates late 2025, early 2026.
That is really most recently, earlier this week, we announced the results from our non-human primate study that was evaluating three different enhanced EBD or enhanced brain delivery Acumen constructs that hit the mark in terms of our objectives for safety as well as enhanced brain exposures. Jim, I don't know if you wanna provide any other color more on the details of those data.
Yeah, absolutely, Dan. Well, you know, look, as Dan says, we're hugely excited, as are many people, about the potential for this technology for delivering macromolecules into the brain. This is one of the big challenges in CNS, is getting large molecules into the brain well. You know, being able to do that is fantastic, but being able to do that without disrupting the profile of the cargo you're trying to deliver into the brain, also critically important. When Dan says we've sort of hit the mark with what we're trying to achieve with our constructs with JCR, that's what we're excited about. We think we've been able to do that.
We're getting in the primate brain, as we've reported, somewhere between 50 and 40-fold improvements in brain concentrations relative to an un-EBD'd antibody. In addition to that, we're not seeing in the primate study so far any significant reduction in hematological parameters associated with anemia.
Yeah
That really gives us a nice combination. That was part of the reason why we chose JCR, is because of their clinical and ultimately commercial experience with IZCARGO for Hunter syndrome, where they're not seeing a significant risk for anemia. That, that's an important factor as well.
Yeah. Like, how predictive is the NHP work on de-risking this anemia issue? Part of the reason why I ask is, like, you know, I think Denali, right, has this mutation that they leverage cis-La a lot, right, to sort of try to make silencing at the TFR, but not in the context of the brain. I mean, can you have it both ways where you can sort of avoid anemia but also be just as robust for, like, amyloid lowering? It feels like that's still, like, a little bit of an open question in the space and maybe a different question in Alzheimer's versus Hunter syndrome. Maybe I'm overcomplicating it. What would be your response there, Jim?
Well, you know, I think these are all questions for the space. Like I said, that's part of the reason why we are attracted to JCR, is because with some other technologies, you're seeing this as a larger risk. You know, I think there's a bunch of factors that goes into it. We spent well over a year's time working with JCR. It's not as simple as you just take your antibody and you take the technology and click 'em together and you're done. We've learned quite a lot and there's not enough time to go into all the detail here, but a bunch of different things matter. Affinity to transferrin matters, and there's sort of a sweet spot.
Like effector function too, right? Like, you need effector function in Alzheimer's, or do you think you don't need it?
Well, recall that our solution to this approach is to use an IgG2. That's sort of inherently built into the functionality of sabirnetug. You know, that has reduced effector function. A lot of the sort of more immunogenicity work that Denali and others have done is essentially taking an IgG1 and turning down that immunogenicity or that effector function activation. You know, we are achieving something similar with a different approach by going for just a natural IgG2. Yeah, I do think that a couple of these factors are important variables when it comes to getting this right.
Yeah. Okay.
I think obviously the payload cargo distinction is important as well. I mean, I think there's a lot of programs that are targeting pyroglutamate Abeta. You know, that's really kinda where the crowd is at. I think if we, you know, with sabirnetug validate and kind of provide further support for the oligomer hypothesis, a supercharged sabirnetug in an EBD format becomes something that looks, you know, quite compelling from a development perspective.
Okay. Very interesting. What could be the timelines there, assuming that the 2B data later this year validates, the sabirnetug hypothesis more broadly?
Yeah. Our goal is to be in the clinic in 2027. I think we've guided to mid-2027 for an IND.
Got it. Makes sense.
Recall, I think we might've mentioned this the other day. I mean, we have the primate study in particular evaluated three candidates. There are actually four finalist candidates. Under the terms of the JCR agreement, we have the ability to license for the same upfront payments or same economics under the same umbrella that the agreement two candidates. I think we'll continue to proceed with a couple of candidates, you know, towards a final designation of a clinical candidate, you know, late this year, early next.
Yeah. Okay. Okay. When the 2B reads out, like, do you think? Like, would you seek to kinda develop all of this independently? Like sabirnetug in another phase III, move this forward to prove a concept, or would you try to partner? Like, what's the thought process there, Dan?
Yeah. Well, I really think we need to see the data in ALTITUDE-AD, right? I think, I mean, there's a whole host of potential scenarios that would be attractive for Acumen to accelerate the development of sabirnetug as a treatment option for patients. That's our primary commitment, is how do we get this, you know, into the clinic as a medicine if we are successful, not only in phase II, but if it necessitates a phase III, which is our current plan. We could conceive doing that independently. We could certainly conceive doing that with a partner. These are, you know, large and expensive studies, so I think having access to resources is an imperative.
Having a collaborative approach to further development would be, you know, one potential option that could be attractive.
Yep. Okay. Great. All right. I think we're out of time, basically. Anything else you'd like to add, Dan?
No. Thanks, Paul. It's obviously a very exciting year for us and we're encouraged to think about all the progress that's being made in this space, and we're particularly hopeful for the impact that Acumen is gonna have on people afflicted with Alzheimer's.
Awesome. All right. Best of luck. Thank you guys very much.
Thanks, Paul.
All right. Thanks very much.
All right. Bye.