All right. Here we go. Welcome to the webcast of the Acumen Pharmaceuticals session. Daniel O'Connell, CEO, joining us. Also Jim Doherty, Chief Development Officer, Alex Braun from the IR team. Guys, welcome. Thanks for doing the back-to-school conference.
Thanks for having us.
Pleased to be here.
Awesome. Dan, maybe just kick it off with your vision for Alzheimer's, given the launch of Leqembi and Kisunla today. Maybe how would you grade that and how would you look at this as a long-term opportunity? What differentiation or what rooms for opportunity for improvement do you see from the current standards?
Sure. Thanks, Jeff. I think we are encouraged at some of the progress that's been achieved in the space and the fact that we now have two approved, full traditional approved, disease-modifying treatment options available to patients. I mean, our view, we have been pursuing and developing sabirnetug as an Aβ oligomer-directed antibody. We think that is a distinct way to address disease pathology and potentially offer a better treatment option for patients in respect of efficacy and safety. What's unique, sabirnetug has, from its origins, been intentionally developed to neutralize these toxic species. We do think that our vision is that the field will continue to evolve and grow and that sabirnetug and the oligomer-directed treatment approaches may ultimately be sort of the best approach for patients.
Have you seen anything in terms of the science of how the identification of an oligomer is being deemed the toxic species? Has the literature kind of evolved to where that's becoming more, you know, more apparent? Are you aware of studies ongoing outside of sabirnetug that, you know, could also help support that?
Sure, Jeff. There's a large base of research and evidence in support of an oligomer being a distinctly toxic species of Aβ aggregation. This goes back over 20 years, even at some of the founding science around Acumen. I think that base of evidence has continued to grow. I do think we've also, as a consequence of what's been observed with these first couple of agents, seen a fixation, if you will, on plaque clearance and, in particular, driving up a reduction in the amyloid PET signal as a means of correlating with clinical benefit or, essentially, a treatment benefit for patients.
We think the Aβ oligomer hypothesis has yet to be fully clinically validated, but we at Acumen, with sabirnetug, are well-positioned to provide conclusive evidence, proof of concept evidence, and evidence in support of taking treatment and development strategies towards preferentially targeting this or exploiting the mechanism, if you will.
Okay, that makes sense.
Yeah.
I think in a lot of ways, it's diversifying the opportunities, right? We know that Alzheimer's disease is a complex disorder. You were asking earlier about the drugs that are out there, and it's fantastic for patients. It's a milestone that we've finally gotten to after a tremendous amount of work and investment to treatments that can treat Alzheimer's disease. I think we're also seeing that no one therapy is going to be able to treat all patients at all stages of disease. In a lot of ways, this differentiated mechanism of action by going after the toxic Aβ oligomer is adding a real opportunity overall to the treatment.
Given the data that you have thus far with sabirnetug, the tolerability and the risk-benefit, I think when most people think about this indication, having another mechanism added on would be helpful. What's your view of the combinability as you look forward? Obviously, you have to wait to get the data results next year.
Yeah. Of course, that's right, Jeff. I think, you know, part of our confidence, much of our confidence, I should say, in sabirnetug is predicated on the Phase 1 data, right? The INTERCEPT-AD study we ran in 2022 through 2023 was exclusively in Alzheimer's patients, and it was designed really as a proof of mechanism study. In that study, we confirmed target engagement, so the antibody hitting its intended target being Aβ oligomers, a safety signal consistent with a competitive safety profile in respect of tolerability and including ARIA. We had these biomarker effects, which both on the PET imaging results, we had a modest or sort of comparable reduction in the amyloid PET signal. We also observed a modification of a series of different fluid biomarkers, both in cerebrospinal fluid and plasma.
On a short-duration study and a patient-exclusive study, you know, we think all of that data kind of holds together and puts us in the path to that, you know, optimized product profile in terms of risk-benefit with sabirnetug. That's the evidence we're seeking to generate in ALTITUDE-AD. I think the future state of treatment will probably involve more than one agent, that there are a variety of different sort of points of intervention that physicians and patients will be seeking for optimized treatment. I think layering on, you know, there's a, as I know you're aware of this, there's this debate right now between the two commercial products of chronic treatment versus, you know, treat to clear. We do think that the disease progression, you know, the disease pathophysiology continues even in the absence of amyloid of a PET signal.
Chronic treatment and addressing some of these other biological mechanisms such as Aβ oligomers or tau or inflammatory approaches, you know, as we sit here today, would, you know, I think it's, you know, safe to assume that they will be, you know, there's a future for those types of multi-product combination strategies.
Makes sense.
Dan, you mentioned the blood-based biomarkers. I wanted to ask you on the sort of the diagnosis angle of it. Is it, you know, is it helping, you think, raise the awareness, maybe stage patients? Do you, how do you see it helping you guys? Is it more just helping with enrollment, the speed of enrollment? I just want to get kind of your view of that, the way it could play out clinically and then also commercially.
Sure. I'll take a first stab at that, and then Jim, I'm sure can follow on. I think, you know, right now, the fluid biomarkers are fundamentally enabling, right? They're creating greater awareness, you know, in the population that if you have a suspected case of Alzheimer's disease, there's research that can be done. There are products that you could be, you know, you could be going on or you could be entering a research study such as we had with ALTITUDE-AD. We, you know, employed a pTau217 plasma screening assay as part of the screening process. We do think that that contributed considerably to the sort of speed and efficiency that we were able to enroll the study.
I think we're in the middle of kind of a step change in the precision that we've got around being able to diagnose in Alzheimer's disease. You saw that for the first time with the PET ligands, right? You went from a truly clinical diagnosis of dementia. What we found when the PET ligands came around was that many times those were not truly Alzheimer's cases. It was dementia for another reason. It was that precision and that ability to see more data that made such a difference. The same thing's happening again with the fluid biomarkers. There are a number of them. There are markers that are more associated with the target, so amyloid-related Aβ 42-40 ratios would be the obvious example. There are the tau markers, which are also a core pathophysiology in Alzheimer's disease. We and others are also looking at downstream synaptic markers.
You're getting an awful lot of data on what's happening in the brain for a patient. It really creates the opportunity to, in many ways, stage patients in a way that hasn't been possible before. To understand where they are in the course of disease and perhaps create a situation where you can craft the right treatment for the right patient. We know that Alzheimer's disease is not diverse. It's not the same pathophysiology for every Alzheimer's patient. I think that's where the field's going. There's been an explosion of work in biomarkers, both in the number of biomarkers, but also in the types of studies that are being done to really understand where you're seeing signals, what the signals are correlating with. I think you're going to see a real improvement in staging and diagnosing individual patients. I think that only benefits being able to target new treatments.
I think it's going to be, you know, more patients seeking more treatments. Essentially, if you think about the funnel and the logistics of, you know, the way the commercial products have rolled out, having the ability to prioritize patients, you know, from a blood test is a much more streamlined way to get the right people into the funnel to go through the steps to come on an anti-Aβ or anti-amyloid product.
Have you guys seen anything? I mean, this is genuinely the last real frontier. It has been for decades in terms of unmet needs and the science. Have you seen anything that's maybe noticeable on the AI or machine learning angle to try to put together all these data points, everything from cognition to biomarker data to mechanism data to clinical data, that's even reliably predictive or that could enhance the chance of success?
I think there's a lot of interest in sourcing the data sets that would inform those models. I think that's the real question of the quality of the data set. There's a massive amount of interest. We were just talking about this at lunch, actually, because it's like multimodal, whether it's EEG, it's biomarkers, it's clinical assessments. I think that AI or machine learning will have a contribution to those advances in the future, a lot of interest. Nothing that I can point to today off the top of my head that says like these folks have it figured out.
There's nothing new discovered, like a mechanism or a correlation or anything like that.
The progress of technology, even in the biomarker space, we've seen rapid evolution of a series of different analytes, you know, whether it was pTau181 to 217, there are a couple others that are in the mix. As those diagnostic platforms continue to be able to multiplex across a bunch of different targets, you're going to, you know, you'll extract the things that are most pertinent to, you know, a disease population such as Alzheimer's disease.
Yeah, it makes sense.
In that same vein with the diagnostic test, I guess, what are you hearing from physicians using it? Obviously, you guys have accredited that use of it to accelerate the enrollment process and also get the trial up and running a little bit faster, I guess. What are you hearing from actual kind of physician usage?
I think it's increasing is the short answer. Certainly, our trial is kind of a real-world example of the benefit that having that kind of quickly accessible data can offer. I think that, as I was saying a few minutes ago, we're really in the midst of a lot of studies ongoing looking at how these various markers can be correlated. I think as more of that data comes around, coupling that with the ease of use and the sort of rapidity of getting results back, more and more physicians are going to continue to use the data. You're already seeing that, but I think that that's only going to increase as more of the survey data comes in.
Yeah, we at the AAIC meeting in Toronto earlier this summer, we presented a poster on some of the economies and efficiencies that using the pTau217 plasma screening assay yielded in the ALTITUDE-AD study. I mean, the cost differential between a blood test and a PET scan is a factor of 10 or more. That alone is just a good indication of how the progression of these better diagnostic, more efficient, and yet just as precise measures come to open up the clinical market.
Moving forward, obviously, with the diagnostic testing for pTau217, are there other biomarkers that you guys are looking at behind the scenes, or do you think it's kind of safe to say that pTau217 is kind of like the clear real winner here?
I certainly think pTau217 is the one that's drawn the most attention, and it does seem to be a very sensitive marker. There's a lot of work trying to understand how well it correlates with the drug effects and things like that. There are also a lot of studies looking at, as patients progress in disease, how the pTau217 signals change. Definitely a lot of interest in that marker. I think in the short term, that does mean that's the one that we're going to learn the most about in a short period of time. I do think there will be other markers that get added to that list. One of the real opportunities with a plasma-based system is you don't have to do a single marker. You can get multiple signals from the same sample from a patient.
I think you're going to see, rather than things supplanting pTau217, because I do think it's going to turn out to be one of the most sensitive and most useful markers, you'll see other things added along with it.
Yeah. I mentioned the fluid biomarkers in Phase 1 and the ALTITUDE-AD. I mean, some of the, you know, we looked at, you know, Aβ, some of the tau species, but also some of the synaptic markers, so neurogranin and VAMP2. As we think about the toxicity of oligomers, the evidence over 20 years suggests that they have a propensity to bind to neurons, induce calcium influx, disrupt neuronal signaling, and essentially result in synapse loss. As we see a short-duration study such as INTERCEPT-AD and sabirnetug at the higher dose levels demonstrating non-insignificant changes on a dose dependency on neurogranin and VAMP2, I mean, that's really at least compelling in terms of our mechanism and how ultimately that might yield the differentiated efficacy profile that we hope for for sabirnetug.
Along those same lines, we talked about combinability. I mean, everyone says, you know, beta amyloid and tau are probably the two, you know, toxic species that you want to get rid of sooner than later. It doesn't, so far, it doesn't look like tau, you know, is as effective. I mean, any updated thoughts of whether that, you know, is still a viable strategy in your view or whether, you know, tau is simply more of a biomarker and less of a, you know, less of a disease kind of modifier?
I'd first offer that tau is still a promising therapeutic target in the disease, you know, presumably. I think what we have observed or experienced is in the Aβ space, it's taken us 25 years to get to where we are right now. We're still debating about plaque clearance versus protofibrils versus oligomers. Tau biology is an order of magnitude more complex than Aβ. I think we've learned, if you're tracking tau, we've learned kind of what not to do, which is what we also did in the amyloid space with, you know, monomer targeting agents and so forth, or pan-amyloid targeting agents like Banopapi. There are a couple of studies, there are some ongoing programs that might read out in the next 12 to 18 months in the tau space that would help inform what to do in terms of a tau-directed intervention.
I think the jury's still out, but it is a consequence of learning what not to do and then ultimately focusing in on the thing that is most, you know, therapeutically beneficial.
Yeah.
They talk about, you know, sometimes they talk about Aβ being the trigger and tau being the bullet. The fact that we have an oligomer-directed agent that's had downstream effects on tau species, you know, again, is sort of encouraging. To think that you could be doing both an oligomer-directed agent and then augmenting with a tau-directed product or treatment is, you know, that possibility still very much exists.
Just to follow up, you hit the nail on the head when, I mean, it's just taken a long time to get to where we are with just beta amyloid. Do you think that the tau biomarkers are, from a regulatory standpoint, are they robust enough where they would take maybe shorter-term, you know, kind of changes to pTau217? Like, is there, is there, because I look at like the Hep C angle, right? That went from 48 weeks to basically four, right? As that four-week was predictive of what you'd see, we're not going to go on that speed. Like, is there a three-month or a six-month biomarker trial in Alzheimer's that you could feel confident in using biomarkers that you drug, you know, a drug has an effect?
For drug effects, probably yes. For regulatory purposes, I don't know that I'm ready to, or I would be willing to say that the agency is willing to accept or extend the label. I do think those biomarkers are faster turns on what's working and what's not from a development, de-risking standpoint.
I think it's fair to say we're not there yet. I would also say this is part of the promise, that there's enough data emerging and enough individual markers that, assuming enough data gets generated, you could see making an argument to the regulatory agencies to be able to do something like that. They're certainly paying close attention to the science that's emerging, and they are engaging in conversations. I think they're appropriately going to be pretty conservative, and they're going to want to see a lot of data before they go down a road like that.
On the AI/machine learning piece, is it the companies that don't, why is the data quality not there? I guess there's not maybe a third-party repository of data that everyone could look at. I'm just trying to think of, you know, do people want, the companies want their own access to their own data, and they're not going to share?
There's an interesting dynamic now that the space has gone commercial. I've been in this space for an extended period of time, longer than I might care to admit. It had been very collaborative where data sharing, impetus, and pre-competitive strategies were, in the last 10 years, kind of more the norm. I think there's a little bit more of a proprietary siloing now where people don't want to enable a competitor unnecessarily in the space. I don't know that there's a quality, it's really the data sets, it's quality, but it's also scale, diversity, how well characterized those data sets are. Some companies that have been in the space for an extended period of time may have unique insights applying AI to those data.
Will we learn anything from the commercial if we do follow-on post-market?
Yeah, the real-world evidence, absolutely. We're already starting to see some of that and kind of who's most prone to benefit from, you know, all of that stuff gets back translated, I think, into really kind of stepping closer towards a personalized medicine for like the, you know, getting the right, yeah, you know, getting the right patients on the right drugs.
I know you're not in the prediction business, but if you had a win for ALTITUDE-AD, what would you say that would look like? Is it preliminary, is it numerical, is it, what, I mean, you're probably not going to get too much stats on it, but I don't know, just help us out.
I mean, it's a reasonably powered phase 2, you know, intended to be a proof of concept study. We're prepared for all outcomes, including an outsized success one where that extended dosing of an oligomer-directed antibody yields a pronounced efficacy benefit and does not have the RA liability associated with plaque clearance. I can't predict, but I'm certainly encouraged to think that we have a real possibility of having that outsized success with ALTITUDE-AD.
I can't remember, I think we've talked about this, but if you think about the Lilly strategy of only, you know, having the drug on board and removal of the plaque and then that's it, versus the Biogen strategy of having it on board 18 months and the clinical trials are indefinitely commercially, is there evidence for keeping it on board, you know, while you clear plaque completely?
Yeah, I mean, I think if you look at the Lilly data or the Biogen data, you know, chronic dosing with the right agent seems to be, until a patient progresses out of a particular stage of disease, seems to be appropriate. I mean, the best evidence, I think, in support of chronic dosing is that gap period that Eisai had with Leqembi, which was they stopped dosing, they weren't sure if they were successful. When the study was successful, they put people back on treatment after, I think, on average, about 12 or 18 months. The PET signal increased over that period of time modestly, but other pTau181, some of the Aβ42/40 ratio, all of the other biomarkers sort of came back online, if you will, in the absence of the intervention.
That would suggest that keeping people on treatment while they're not, where they haven't advanced to moderate disease or frank dementia, and where you can keep, and this is where being able to monitor the biomarkers becomes clinically relevant, where you're sort of assessing progression not only on clinical measures, but on the biomarkers and not relying exclusively on an amyloid PET, right? I mean, that's the, I mean, one of the, you know, if you take a step back and you look at donanemab as a pyroglutamate Aβ-directed antibody that's designed to clear plaque, I mean, they deplete targets. It doesn't make sense to actually keep people on donanemab in the absence of plaque. Yet there's other Aβ biology that continues to advance the disease, as supported by the Eisai data.
It could be that you have, you know, say a foundational, you know, oligomer that maybe is not picked up on a PET scan, but constantly being a space clearing it out.
At that level of description, I can assure you that's the case.
Yeah.
I think that the complexity of which, you know, is it protofibrils or larger oligomers or small oligomers? I mean, there's a variety of different species. These things, they do work, you know, kind of in an equilibrium. I do think that, you know, plaque clearance does have a secondary effect on reducing, presumably a secondary effect on reducing oligomer levels. I think most of the evidence that we've seen would support chronic dosing, again, while a patient is kind of within that early AD space. The other area that's of great interest to us in the near future, you know, at some point in the future, is this preclinical Alzheimer's population where their, you know, Aβ oligomer pathology is pronounced.
Like, that's really the initiation phase of disease as aggregation is starting and where you're having presumably like subtle effects on neuronal and synaptic health that don't present as clinical symptomology yet. Yet this, you know, essentially the horse is leaving the barn, you know, sort of the process is started and will continue to advance in the absence of some intervention.
Makes sense.
Changing gears a little bit, over the weekend, Leqembi subcutaneous formulation was approved for maintenance dosing. We know that you guys are developing a subcutaneous formulation yourselves. If there's any color around that, how the company's thinking about the subcutaneous formulation or kind of greater, what do you think it means for the field as a whole?
Yeah, absolutely. As you say, there is now an approval for maintenance with the subcutaneous formulation. I think we, like everyone else, think that one of the major things you gain with a subcutaneous formulation is much more flexibility and ease of use to administer the drug. There are likely to be patients who will want to retain an IV approach, but there are probably a lot of patients who would like to be able to switch over and use a subcutaneous strategy. Some recent studies were just discussed at AIC suggesting that for a number of patients, they're readily able to do that and administer themselves. We definitely feel like that's going to be of value to any treatment, to have that ability.
As you say, we have run a phase 1 study in healthy volunteers with a formulation, a COMEX formulation, that will allow for larger volumes to be delivered. Based on the results from that study, we're continuing forward with our approach with subcutaneous. We really kind of want to see the results from the ALTITUDE-AD before making too many next steps with the subcutaneous approach. What we've been saying is we're likely to complete that study and then, based on results, look at including an arm or a component in our phase 3 study that we would be doing to follow on with the subcutaneous formulation. That's sort of our first approach to subcutaneous. We've also announced recently a collaboration with JCR Pharmaceuticals, which I know you guys know about, for a blood-brain barrier approach. We're calling it enhanced brain delivery or EBD.
In our view, an EBD approach, we already love the profile of sabirnetug, as Dan's been saying, but if we are able to increase the fraction of drug that's being delivered into the CNS, that gives us a lot of flexibility with dose and with formulation and with amount delivered. That would be a second opportunity to go to a subcutaneous administration.
Awesome. Thank you. Also, on your collaboration with JCR Pharmaceuticals, there's some other players looking at transferrin and the transferrin receptor. What do you think differentiates your now new approach as opposed to other players in the field?
Yeah, so we've looked pretty thoroughly at the opportunities that are out there and the strategies that are out there. As you say, a number of people are targeting transferrin as the shuttling mechanism to get into the brain. There are, just as an aside, a number of other proteins that play a similar role, so there would be other possible targets as well. We landed on transferrin because there's the most clinical validation for transferrin in these sorts of products. In the case of JCR Pharmaceuticals, they have clinical experience. They have a marketed product called HisCargo that's approved in Japan for Hunter's syndrome. We like their experience and we like the epitope that their technology binds to. What they've seen with HisCargo is they have not seen signs of anemia with their drug.
That's something that, of course, will need to be played out with the collaboration to see if that holds. We're definitely attracted by their approach to targeting a transferrin receptor and the lack of anemia that they've seen with their HisCargo product.
What's aspirationally using a transport or transferrin or however mechanism you get across the blood-brain barrier, what's your view of where this theoretically could go in terms of the dose? Could you cut the dose half by 90%? I don't know the number. It seems like, I don't know. I think it's less about the, to my opinion, the dose and more about maximizing the effect size, because you can run into, I mean, ARIA, et cetera.
Right.
How do you think that will all kind of shake out from the transport side of things?
Yeah, as Jim is suggesting, I think having that greater fraction of drug in the central compartment and the brain interstitial space through this receptor-mediated transport has advantages. I think, with sabirnetug, we don't, relative to plaque-directed antibodies, we're not running into a dose-limiting ceiling. In the phase one, I think we conclusively showed that. For us, it's really more about these EBD approaches seeming to have a combination of potentially an intrinsic safety benefit just in terms of avoiding all CA plaque because of the capillary delivery, the lower dose requirements to achieve an equivalent efficacy and all of the safety, and thereby just the convenience of doing that in a subcutaneous format.
I think there's the order, the dose pieces, maybe as you look around there, it's sort of like somewhere between a three or bifold reduction in the dose levels that are equivalent to IV is sort of, I think, the benchmark out there.
Okay. What sort of timeframe do you think you can get to, say, a proof of concept?
We got into preclinical candidate designation early 2026. We are actively working with our partners at JCR Pharmaceuticals to profile different carriers and different cargoes or payloads. The JCR transferrin carrier technology has a variety of different configurations on the front end of that. We are looking at sabirnetug and sabirnetug-like payloads or cargoes. Under the terms of the agreement, which I would say is a really capital-efficient way for a small company like ours to get into a validated space, it is a fiscally responsible deal. We think we'll have the preclinical data sets early next year. In a perfect world, those are adequate to help us move towards IND and then getting in the clinic in 2027.
Yep, okay. That makes sense.
From the landscape, I know the recent Prothena data was sort of not optimal. That's another company in the space that people had a lot of hopes for. Any higher-level views of that? I mean, that's the way it is.
I mean, I think putting it in the category of Alzheimer's is a challenging and occasionally humbling space to be active in. I think that, you know, we definitely have a very different epitope and mechanism and feel like we're on the right path with sabirnetug in terms of that risk-benefit profile for patients.
Yeah. What's your view of the, because I know the Denalis and Electors of the world that have had novel mechanisms of action. As we talked about earlier, it's not like this is a completely met medical need. You're adding risk on top of an already high-risk indication if it's not a proven sort of validated mechanism. There is still a need to do something different, right?
Sure.
Is there a mechanism that you think looks interesting that's outside of the beta amyloid kind of TAO axis, you know, either inflammatory? I mean, people are talking about the GLP-1.
Yeah, GLP-1 is going to be, could be very interesting later this year with the Evoke readouts. I mean, that, you know, again, I'm not in the business to make predictions, but if that were positive or directionally, even just directionally positive, I think it would usher in a lot of enthusiasm for complementing Aβ, tau strategies with a GLP-1 approach. I think the TREM2 space is still, you know, still has some potential. I think there's more learnings to be had there, but this whole inflammation component of Alzheimer's pathology is certainly an area that we should continue, you know, the field should continue to pay attention to.
Okay. I'll see if you have any.
Okay. Any final thoughts? Are you good? We covered everything?
I think we covered a lot of grounds. I do think it really is an exciting time for the space. Even in that last question, I think there are so many possibilities when you consider inflammatory approaches, for example. One of the challenges is there are a lot of different possible combinations to look at. When we talked earlier about biomarkers and how that's going to be useful, you can envision efficient trials in the future that are going to allow you to look at combinations in a different way.
Yeah, to me, it seems like, you know, this isn't ripe for sort of AI/machine learning, like, let's try to check that box in the next couple of years to see if we can uncover a novel mechanism or staging or some, you know, some element.
Yeah, and you think about that technology coming on, Jeff, in a situation where the population, the unmet need will persist, right? This isn't going to go away overnight. I don't think anybody's going to sort of steal the bacon here. I think it's sort of continued progress and exciting time. Some of those enabling technologies, whether it be the AI and/or biomarkers or some combination, certainly improves the probability of success in the future and better options.
Thank you, guys.
Thank you very much. Thanks, guys.
Thanks, folks.