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

My name is Michelle, and I will be the operator for today's call. [Operator Instructions] Please note that this conference is being recorded. I'll now turn the call over to Alex Braun, Head of Investor Relations. Alex, you may begin.

Thank you, and welcome to the Acumen conference call to discuss the top line results from our first in-human Phase I INTERCEPT-AD trial, investigating ACU193, in early Alzheimer's disease. The majority of these data were presented yesterday, at the Alzheimer's Association International Conference, or AIC, in Amsterdam. Please note that during today's call, we'll reference slides that are available on the webcast. If you've not already done so, please go to the Investors section of our website to access the webcast player. A PDF of the slides has been made available on our website as well. During this presentation, we'll be making forward-looking statements based on our current expectations relating to the clinical development of ACU193, and the potential therapeutic value of ACU193. These forward-looking statements are neither promises nor guarantees and are subject to uncertainty and risk. Please see Slide 2 of the investor presentation accompanying this webcast. Our press release issued yesterday and our most recent annual and quarterly reports filed with the SEC for important risk factors that could cause our actual performance and results to differ materially from those expressed or implied in the forward-looking statements. We undertake no obligation to update or revise the statements provided on this call as a result of new information or future results or developments. The data presented today may not be indicative of future data or results of future clinical trials or real-world results. We'll begin with the presentation of the top line results, followed by a moderated Q&A. And finally, we'll take questions from our analysts and the audience. Now I'll turn the call over to Dan O'Connell. Chief Executive Officer of Acumen.

Thank you, Alex. Good morning to those of you back in the States, and thank you for joining us. Our team has been at the Alzheimer's Association International Conference in Amsterdam, all weekend, and had the distinct privilege of presenting our top line results from our INTERCEPT-AD clinical trial to the medical community here yesterday morning. Today, we're pleased to be discussing those results with you. With me today from Acumen are Dr. Eric Siemers, our Chief Medical Officer, who will take you through the data; and Matt Zuga, our Chief Financial and Business Officer. We're delighted that following the data presentation, we'll be joined by special guests, Dr. Steven DeKosky and Dr. Lawrence Honig for a discussion related to the data set. Dr. DeKosky is Deputy Director of the McKnight Brain Institute at the University of Florida and is a member of Acumen Scientific Advisory Board; and Dr. Honig is the Director of the New York State Center of Excellence for Alzheimer's disease at Columbia University and was also an INTERCEPT-AD trial investigator. I'd like to thank both of them for taking the time to be with us today and look forward to hearing their clinical perspectives on ACU193 INTERCEPT-AD data set. I'd like to start off the call with a brief review of Acumen’s antibody ACU193, a high-level summary of our Phase I results. Acumen was founded on the scientific premise that Aß oligomers are distinct and important triggers for the initiation and propagation of Alzheimer's disease. Scientific consensus asserts that Aß oligomers are the most toxic form of Aß and once oligomers bind to neurons, they inhibit synaptic function and induce neurodegeneration. Our product candidate, ACU193, an affinity matured humanized monoclonal antibody is differentiated from other monoclonal antibody studied in Alzheimer's disease based on its high selectivity for Aß oligomers, more than 500-fold selective for oligomers over Aß monomers and more than 85-fold selective for oligomers over fibrils, which has been theorized to lead to greater efficacy. ACU193 was also designed with reduced effector function, which is theorized to produce lower rates of ARIA-E compared to plaque targeting antibodies. Notably, our asset was developed was originally discovered in partnership with Merck & Company and -- which completed extensive preclinical studies to confirm the attributes of the antibody. And at present, ACU193 is being developed by Acumen team, several members of which spent many years during early phase and late phase global Alzheimer's clinical trials at Eli Lilly. Yesterday we reported positive results from a first-in-human study of ACU193, INTERCEPT-AD. Before I turn to our convincing proof of mechanism results, we were highly encouraged by the rapid dose-related statistically significant amyloid plaque reduction observed at the higher doses study, 60 milligrams per kilogram every 4 weeks and 25 milligrams per kilogram every 2 weeks. Although plaque is not ACU193's intended target, it appears 193 reduces amyloid at a rate and level comparable to currently approved monoclonal antibodies at a similar time point, in this case, around 3 months. Eric will speak to the possible mechanistic explanations for this finding, but we believe it demonstrates further evidence of 193's activity in the brain and is a promising development given the established correlation between plaque reduction and reduction in cognitive decline. You will see in the data we present that 193 also demonstrated robust dose-dependent target engagement of Aß oligomers as measured by a novel assay developed by Acumen and which exceeded our expectations. In fact, we observed our higher doses approaching maximum target engagement. This finding we are particularly excited about, given this is the first time an oligomer targeted antibody has demonstrated direct target engagement. ACU193 was well tolerated with no drug-related SAEs and low rates of ARIA-E, and based on the pharmacokinetic profile observed, monthly dosing is supported. To dig into the plaque reduction observed, we plotted the mean reduction in amyloid plaque in centiloids for higher doses 60 milligrams per kilogram dosed every 4 weeks and 25 milligrams per kilogram dosed every 2 weeks for total of 3 doses in each cohort against the 3-month plaque reduction observed in lecanemab's Phase III CLARITY AD study. We chose to compare to lecanemab because the design of this antibody is most closely aligned to 193 and that they both target soluble toxic amyloid aggregates, but also reduce plaque. The graph on the left indexes the baseline values to 0, while the graph on the right shows the absolute values. As you can see at a similar time point 193 reduced plaque to a comparable level as lecanemab. Our 60 milligram per kilogram dose once every 4 weeks and our 25 milligram per kilogram dose every 2 weeks, both demonstrated statistically significant reductions within each cohort. It is worth noting that our 193 cohorts did have lower baseline levels of plaque. However, the slopes did not appear -- do appear comparable despite this fact. We did not show them here, but earlier generation antibodies, which have failed to show cognitive benefit in Phase III trials will not be able to achieve this rate or level of plaque reduction at this early time point. Typically, with faster and larger reduction in plaque, the rate of ARIA-E tends to increase as well. Lecanemab, however, exhibited approximately half the amount of ARIA-E in their Phase III trials compared with plaque targeting antibodies. ACU193 also exhibited low rates of ARIA-E in our Phase I study, even at elevated doses. As seen here when comparing after a single dose Aducanumab saw 100% of patients experienced symptomatic ARIA-E at 60 milligrams per kilogram dose in their Phase I study. And Donanemab saw 50% of participants exhibit ARIA-E at their 40-milligram per kilogram dose. While ACU193 maintain a low rate of ARIA-E at 60 milligrams per kilogram. In this table, we wanted to compare only ARIA-E after single dose, due to the timing of the MRIs in our study, we're showing 6 patients from the SAD who received 25 milligrams per kilogram because the MRIs for that cohort were taken after 2 administrations of drug. Also note that these are not direct comparisons as no head-to-head trials have been run between these products. As you'll hear from Eric in more detail, our overall rate of ARIA-E in the INTERCEPT-AD trial was 10.4% with 2.1% or 1 patient being symptomatic. We highlight this to make the point that 193 has the ability to dose quite high on a comparable basis to other monoclonal antibodies while maintaining a low rate of ARIA-E, and by doing so, appears to be engaging more of the toxic oligomer target in the brain as seen with our near maximal target engagement. This leads us to believe that 193 could have a broad therapeutic index. Here's a full view of how we think about 193's proof of mechanism as demonstrated in the INTERCEPT-AD trial incorporating low levels of ARIA-E dose-related pharmacokinetics that demonstrated concentrations of 193 orders of magnitude higher than reported Aß oligomers and CSF and dose-related statistically significant targeting engagement. Taking together with the observed rapid plaque reduction, all these elements will guide dose selection for our next study. Taking a step back, I'm very pleased with the positive results of INTERCEPT-AD, and I'd like to thank our team, the patients, their families and our investigators for all of their hard work that made this such a well-run trial. The success of our novel target engagement assay robust study design and committed execution allowed us to demonstrate proof of mechanism for ACU193 and establish dosing optionality for our next phase of development, given the broad therapeutic index. And with that, I'd like to turn the call over to Eric to present the INTERCEPT-AD data.

This diagram outlining the design of the trial and in some respects, it's a fairly typical SAD MAD first-in-human study. There are a few differences mainly that we did not wait to complete all of the SAD cohorts before starting the MAD cohorts. Now there were numerous safety measures that were built into this trial. It was, again, a first-in-human trial. And so sentinel dosing was used for the SAD cohorts. And then after each cohort completed, there was a safety review before the next cohorts were started. So again, in this design that we have to look at Cohort 2 after that cohort was finished and had a safety review that actually triggered the initiation of Cohort 3 and also the initiation of Cohort 5, which was also a 10 milligrams per kilogram. But in many respects, a fairly typical SAD MAD study which was done in patients with early Alzheimer's disease, in other words, MCI or mild dementia with evidence of being amyloid positive. So if you go to the next slide. So these are the baseline characteristics of the patients that we had in the study. I think these are fairly typical for what you would see in a population of early AD again MCI or mild dementia and we'll talk about some of the specifics in here, but overall, these are generally similar to other trials using similar patient populations. So if you go to the next slide. So to go to the safety, which again is always the primary outcome in the Phase I study. These are the serious adverse events, and we actually had 3 SAEs that actually occurred in just 2 patients. One of the patients had the lower 2 SAEs. But generally, as you can see here, there's nothing that would suggest that it's related to the drug and nothing that was problematic in terms of administration of the drug. So if you want to go to the next slide. So here's a little bit more detail. Now this is -- these are the treatment emergent AEs and these are ones that are greater than 3% and also have ACU193 with a greater number than placebo. And again, this is a relatively small Phase I study as they always are. And so the numbers are small. So we did have 3 cases of COVID in the ACU193 group compared to 0 in placebo, a couple of cases of bronchitis in the ACU group compared to 0 in placebo. But generally, not anything there that in a Phase I study would raise significant concerns. Now the ARIA-E, as you can see, we'll talk about that more extensively. That obviously does have some relationship to treatment group as well as ARIA-H. So if you want to go to the next slide. So this slide is a little complicated but I think it's worth taking a minute to walk you through it. At the top of the slide are the SAD cohorts, the single ascending dose cohorts and you can see the various dose groups. At the bottom of the slide are the MAD cohorts with these arranged by the actual doses. The color coding here is, green is no ARIA-E yellow is asymptomatic and red is our single case of symptomatic. We went ahead and left in the placebo-treated patients here because people on placebo or no treatment can rarely have RAE. But obviously that is not expected and we didn't see it in this study. And we also have included here their APOE status and I'll come back to that in just a second. But in terms of RAE cases that we had, we had one case at 10 milligrams per kilogram in the MAD cohort. Now that occurred after three doses of the MAD cohort. The each person had 3 doses. This occurred after the third dose was asymptomatic and radiographically was actually very mild. It was actually missed on that the local read, picked up on our central read and then resolved at the end of the study. At the 25 mg per kg dose group, we'd have one person again in the MAD cohort that developed APOE and that developed ARIA-E. And that was actually first picked up on the day 70 MRI after the second dose. But the day 28 MRI was reread, which was after the first dose. And actually, in hindsight, there were some very subtle ARIA-E that was picked by our central readers there. Now that patient radiographically is improving. Again, this is all asymptomatic, but radiographically is improving. It hasn't quite resolved yet, and we're continuing to follow that patient. And then at 60 mgs per kg, we had 3 cases of ARIA-E. In the SAD cohort, this is 21 days then after a single administration of 60 mgs per kg, we did have a case of asymptomatic ARIA-E, picked up on the scheduled MRI, which then resolved by the end of the study. And then in the MAD cohort at the lower right, you can see this as our one symptomatic case of ARIA. This case actually occurred after the first dose. And the patient wasn't dosed again. So this really occurred after a single dose. It was picked up on the scheduled MRI, and then the person was brought back in for further clinical evaluation. The reason why we called this symptomatic is that the person complained of the right leg "giving out" without really focal findings on the exam, but the ARIA-E was in the left parietal region of the brain, and we elected to call that symptomatic. And as it turns out, the ARIA resolved radiographically fairly quickly, and the symptoms resolved along with that. So the patient did have what we think to be symptomatic ARIA but it resolved quite quickly. And then the other case was in the MAD cohort was asymptomatic. That was day 63, picked up after the third scheduled dose on the scheduled MRI and again had resolved by the end of the study. The one other thing that's, I think, important to point out on this slide is that we had a total of 6 people who were APOE for homozygous who were treated with ACU193. Of those 6 APOE4 homozygos, none of them had ARIA, which is quite different than other antibodies where you see rates of 30% to 40% of people who are E4-homozygos having ARIA. Now in a small study like this, it could be due to chance, but we will continue to watch this very carefully because that could certainly be a differentiating factor compared to other antibodies. So if you want to go to the next slide. So this is just a bit more detail about the ARIA cases that we had. We were only very recently unblinded. And so we're continuing to dig through the details of this data, especially at the individual patient level. But just to give you an example, the third line there was our patient with symptomatic ARIA. That person had the biggest decline in their plaque load based on PET, which is obviously of some interest. And so we are going to continue to analyze data based on beginning plaque load, the reduction in plaque load, that sort of thing. The other interesting thing, and this is something just to continue to follow moving ahead is that all the cases of ARIA-E were female. I don't think there is any obvious explanation for that. Again, it could just be by chance. But that something that we all continue to watch quite carefully. So if you want to go to the next slide. So here are the changes in plaque load based on florbetapir PET. And on the left-hand side of the slide, you can see this is from the SAD cohorts. The red dotted line are the pooled placebo patients. And you can see that they are really quite flat. I mean there's no change, as you would expect, in their plaque load between baseline and endpoint. And then in the treated group, you can see with the various doses, there's no consistent change. And so again, with a single administration of drug, not surprising that we don't see any appreciable change in plaque load. On the right-hand side of the slide, a situation is -- this is in the MAD cohort. The situation is a little bit more interesting. At 10 mgs per kg, we don't see any difference between baseline and endpoint in plaque load. But at the higher doses, both 25 mgs per kg and 60 mgs per kg, we do see a decrease in plaque load. And that actually, if you look at just the difference between baseline and endpoint, in both of those cohorts that change was statistically significant with a p-value of 0.01. So there did appear to be a change in plaque load in those cohorts at those higher doses. So if you want to go to the next slide. And so here's the individual patient. And again, we're still analyzing these data, but I wanted to show them to you today. And so on the left-hand side is the 60 mgs per kg group, and you can see there's most of the people in that group did seem to have some decline in their plaque load, not all the patients. The patient with the blue triangle towards the bottom was relatively flat. But most had some degree of decline. Now the green triangle was the patient who had symptomatic ARIA. So you can see that the endpoint plaque load wasn't particularly low, but that was the biggest decline. And so it's this kind of data analysis that we're continuing to do. If you look at the right-hand side of that slide, this is the 25 mg per kg dose group. And there are certainly some patients there who have some decline, not all of them. I think to my eye, it looks like there's a little bit more stability, even though some of them do drop. And I think that's an important thing that we'll look at further because there is this question of C max and whether or not some of the ARIA is driven by C max and maybe also the plaque reduction. So we're going to continue to analyze these data to try to understand it better. You can see that there's the 1 patient in this cohort that had ARIA, which was asymptomatic. It's kind of buried in the middle there, but there was a decrease in plaque load, but not anything that was really different than other patients in that cohort. So we'll continue to analyze these data and to understand it better because this question about ARIA, what's the risk factors to C max make a difference. That's not just a question for us, it's a question for the field. So if you go to the next slide. So this gets into the PK data. This is just serum PK. This is fairly well behaved actually. I mean it looks good. On the left-hand side, you can see after a single dose, it's a very nice dose-dependent effect in the PK. On the right-hand side, those are the multiple dose cohorts. And you can see with multiple doses, you don't have any accumulation of drug, which is always a good thing. So the serum PK looked really quite good. If you want to go to the next slide. For immunogenicity, we -- those data are still coming in. We have a handful of cases. And in each of those cases, thus far the titers are quite low. There doesn't seem to be a relationship with PK. And so we're continuing to follow that. But at this early stage of our analysis, we don't see any evidence of immunogenicity that would be problematic. So if you want to go to the next slide. So here's the CSF PK, which I think is really actually of more interest. And so on the left-hand side of the slide, you can see for the single-dose cohorts. There's this very nice dose-dependent increase in CSF concentration of drug, 21 days after the single administration. And then on the right-hand side of this slide, you can see in the multiple dose cohorts there is also a very nice dose-dependent increase in CSF drug concentrations. The dotted blue line there is just to bring your attention to the fact that the y-axis on the 2 plots is not the same. And so after multiple doses, you do have substantially higher concentrations of drug in spinal fluid than after a single dose, which obviously is what you would expect. So the CSF exposure data looks actually quite good. So if you want to go to the next slide. So one of the really important aspects of this trial for us was to show evidence of target engagement. Now target engagement for us means showing that you have antibody bound to an oligomer. If you have an antibody that targets plaque, you can just get a PET scan looking at plaque load and show target engagement, but we don't target plaque, so that wasn't our target engagement. We did have some changes in PET, which we just talked about, but that's not our target. So this very interesting assay was developed that's not been done before. And so our group at Acumen developed this assay. It's an immunoassay. The capture was with an anti-idiotype antibody, the detection antibody binds to oligomers. But the point is, is the assay only gives you a signal when you have ACU193 bound to an oligomer. So it's actually very technically difficult to do. But if you want to go to the next slide. So but what we found in study is that on the left hand side for the single dose cohort, at 2 mgs per kg it was below sensitivity the assay. But at 10, 25 and mgs per kg, you see this very nice dose dependent increase of the complex of ACU193 bound to oligomers. And then in the normal dose cohort on the right hand side, you see the same thing. You see this very nice dose-dependent increase in the complex of ACU193 bound to oligomers and in fact, even in this very small Phase I study, we reached a statistical significance on some of these differences as you can see on the slide. So that the assay actually performed very well the study. So if you want to go to the next slide. So this is a slide that was really at least it might be very important in terms of our drug development program. And so what you see here is on the X-axis, this is concentrations of antibody of ACU193 in CSF, so drug concentration in CSF. And on the Y axis, what you see is the complex of ACU193 bound to oligomers. And these are individual patients then in the various cohorts that you see plotted here. So what you can see is that the lower doses, you have that the amounts coming up, but then you start to reach a plateau. And so when you start looking at the 60 mgs per kg every 4 weeks dose group and the 25 mgs per kg every 2 weeks dose group, you're really on the flat part of the curve. And the reason why that's so important is that tells you that you're at the point of diminishing returns in terms of target engagement. And that means that there's no reason to think that you would want to look at doses above 60 milligrams per kilogram. Whenever you're designing a Phase I first-in-human study, dose selection is very difficult. But in this case, the doses that we selected actually bracketed exactly what you would want. And so the beginning of target engagement and then getting to the point where you were near Emax or at the point of diminishing returns. So we have a little bit more work to do to pick the doses for the next study. One of the things we'll need to do is model what we think the CSF concentration of drug would be with 25 milligrams every 4 weeks. But that can be pretty easily done from these data. And so we haven't made any final decisions in terms of the doses for the next study. But based on these data, we clearly will be able to make those choices and will be able to choose doses for the study. And we know it doesn't have to be above 60 milligrams per kilogram. So if you go to the next slide, I guess -- and with this, I'll just turn it back over to Dan to kind of summarize for us.

Thank you, Eric. As you've just heard from -- in detail from Eric, the results from our Phase I study support advancement in the next phase of development for ACU193, which we designed as a Phase II/III study. Rapid dose related amyloid plaque reduction in our high-dose cohorts is a welcome development further supporting the convincing proof of mechanism data we have generated. ACU193 was well tolerated in patients with early AD and near maximum target engagement was demonstrated, establishing broad therapeutic index and path to monthly dosing. We did include exploratory analysis and INTERCEPT-AD including computerized cognitive testing and MRI arterial spin labeling pulse sequencing that did not show effects due to the short duration of the study and the small end, which was expected. Fluid biomarker data is expected late third quarter, which will help further inform the design and dosing strategy for our next trial moving forward. I'll round up today's call with a preview of our future strategic plans. We anticipate an end of Phase II meeting with the FDA in the fourth quarter to guide our proposed Phase II/III study design. This includes starting as a Phase II and following an interim analysis expanding to a Phase III study should it be warranted. We also are actively investigating the development of the subcutaneous administration of 193, which now has a benefit to be better informed based on our Phase I results. Evaluation of potential next-generation product opportunities and exploration of advantageous partnerships for visual initiatives we are pursuing with the best interest of shareholders in mind. Now with that, I'd like to welcome Dr. DeKosky and Honig to discuss ACU193, and I'll turn the call back over to Alex.

Thanks, Dan. Before we open the call more broadly for Q&A, I'd like to invite Dr. DeKosky and Dr. Honig to share their overall impression of the Phase I results presented today. Eric should also be included here in case there are any questions regarding the results. So Dr. DeKosky if I can start with you. Given your experience in Alzheimer's disease, what is your overall impression of this Phase I data? Do you believe the safety profile on the PK and the demonstrated target engagement suitably address proof of mechanism for ACU193? So Dr. DeKosky, I think you might be muted. There you go. Okay.

I certainly believe it presents proof of the mechanism, the -- especially some of the serial dilutions and serial changes with increasing dose speak to that issue. I was also very encouraged because this is primarily initially this for safety that although there are 1 or 2 cases that have shown, at least in my view that the drug is working because of the nature of the side effects, that side effect profile is benign in my view, quite benign. So -- and no specific adverse events, serious adverse events that can be directly traced to the medication itself. So overall, I'm very encouraged by this, seeing a lot of these kinds of preliminary studies over the past several decades as Dr. Honig and Dr. Siemers has seen probably 10x what both of us have seen, but this is as positive an early study as I can hope for. We know a great deal about how the brain responds to these various antibodies. There was an article this morning came out in the New England Journal talking about solanezumab, which unfortunately did not hit its targets. So we have an immense amount of antibody data in Alzheimer's disease. Some of these have never made it into Phase II trials because their Phase I safety was not anywhere near what it should have been. But overall, I'm very encouraged by this. Too soon I think, to make much of the changes in cognition, but certainly, I prefer to see stability to slight declines in rate of change than a worsening of cognition, which we had seen in a couple of other trials.

Dr. Honig, any -- do you agree, I suppose? And any other feedback?

Yes. So I mean, I certainly concur that with the Steve, and as an investigator the overall results in this 60-plus patients concur with my personal experience with the patients they had enrolled in my side and seems like a safe drug to administer without much of the way of side effects than a low rate of ARIA. And I think the safety, which is clearly the major concern in the Phase I study was demonstrated and the other would be part of the engagement and preliminary efficacy measures and I would concur that certainly the antibody is getting into the CSF, but the antibody is binding oligomers and there is, I think, reasonably clear evidence that at the higher doses, there is some removal of plaque as we expect for most of these antibodies and interact with forms of amyloid that are larger than [indiscernible] or plaque. So I don't think it's surprising that the oligomeric than antibody that binds oligomers will remove plaque. I think in fact that it's an encouraging indication that in addition to fully interactive with oligomers does remove plaque from the brain at the higher doses. So safe, and meeting its target and a little bit more in terms of the plaque engagement.

Great. Yes I'd like to dig a little bit further into the plaque reduction that we saw. So I'll turn it back over to Dr. DeKosky to talk about your take on the fact that amyloid plaque was reduced in those higher doses and the 60 mgs per kg every 4 weeks and the 25 mgs per kg every 2 weeks. Despite this not being an objective of the study?

Well, one of the things that we do know is that things that happen in the brain are necessarily exactly reflective of the way they react in a test tube or a [Audio Gap]. And some of those things we can predict some we can't. I have been very focused on the issues of equilibrium and I think to the extent that it targets predominantly oligomers. The oligomers we're going to do 1 of 2 things. They were either going to in terms of toxicity. They were either going to bind to the synapse where we know they cause either dysfunction or disruption, or they're available for adding to the plaque to the extent that you remove concentration of ABOs from the oligomers from the extracellular space. Equilibrium suggests that you may well get things that come off the amyloid plaque and are taken upo by antibody. And of course, the antibody is also in its being removed by the fact that its attached to the beta, which lowers the amount -- that concentration in extracellular space. So it wouldn't surprise me if any of the things that we have attacking anything that disrupts, a beneficial thing in my view, since we know the plaque removal does currently with cognitive changes.

Great. Anything to add Dr. Honig?

No. I think the -- as we all understand, it's hard to know for certain what the mechanism of the plaque reduction is, whether it's additional body of the plaque or whether it's as Steve mentioned that binding oligomers thereby causing facts to disassemble. But either way, clearly, there is evidence of binging oligomers and it's clearly evidence at the highest doses of decrement plaque, and plaque load. So I think both of those are encouraging that the presumed primary method of action is a dividing the oligomers and we believe that's where the major toxicity in [indiscernible] . Also we know that plaque reduction in the whole of spectrum of the monoclonal antibodies that gets amyloid. And for that matter, other anti-amyloid treatments, that plaque reduction seems to correlate with efficacy to the extent that we have data on a variety of other agents. And so to that extent, it's certainly encouraging that this is also demonstrated plaque removal at the higher doses in just a few doses. We're talking about a very small number of doses, short period of treatment. And nonetheless, we're already seeing plaque reduction.

Right. I'd like to talk about the target engagement assay given it was novel. And if you think that assay and the results and along with the PK and the safety and the plaque reduction. If all of those together allow for the dose selection in Phase II/III and also if from a clinician's point of view, if you think monthly dosing is compelling from a product profile point of view for this product. Either one can start.

Certainly, I would say that monthly dosing, certainly is more favorable for our patients and for that matter, for practitioners. So we certainly would prefer to see the lower burden of monthly dosing. And in this case, of course, the actual dose per month between 25 milligrams biweekly and 60 milligrams monthly does not differ by that large. It seems like both were at a -- were reasonably equivalent. And certainly is promising the idea of having possibly a monthly dosing schedule rather than a biweekly schedule.

Well, there's both convenience for the patient and family and there's also a large cost difference for whoever will have to bear It. If you drop the infusion by 50% of total, like an expected total. It certainly is easier in a month if it gets to subcutaneous, which we're exploring now. It will be even better, even if it has to be at 1 month's intervals. We certainly have people give small doses of various medications to themselves, diabetes getting insulin, now pretty much everyone getting Ozempic for weight control in the last be injected. But I think it is much easier to do it once a month and twice a month in any of our patients and their families certainly would tell us that because they don't all come from right near the clinic where it's being done.

Yes. Quickly turning to the ARIA-E. I know you both mentioned that you were satisfied with what you saw with the rate of ARIA-E in this study. But related to that, should clinicians be focused on symptomatic or asymptomatic ARIA-E rates in our opinion?

Well, I think it's often a hard question. There's even different forms of symptomatic in the sense of the people who report symptoms then you notice they have ARIA-E. The people who you noticed they have ARIA, and you ask them very clear where you can question deeply about they have any symptoms, and say, oh, yes, maybe there's some little symptom that may or may not relate to the ARIA. So I think -- but I think we would all agree that asymptomatic ARIA is less scary than symptomatic ARIA. Symptomatic ARIA is less scary than serious symptomatic ARIA. Clearly, in this Phase I, we did not have any serious ARIA but there was a little bit of symptomatic ARIA and a little bit more ARIA as a whole. Again, ARIA, in a way, might almost be now viewed kind of by plaque reduction as an evidence of target engagement in some sense. So I think the neurologic community as a whole and the dementia specialist community is less, finds ARIA less fearsome than they did initially. The idea of having brain edema or a brain swelling certainly wasn't something that people looked as a favorable event. But on the other hand, I think we realized that it isn't quite as scary as it might seem, and that, in particular, if 90% of it there's always asymptomatic that makes one less feared of the side effect.

I'd just like to add that it's kind of like if a tree falls in the forest, does it make a noise. We would not have found the best majority of the cases of ARIA in any of the clinical trials that have been done so far. If we were not doing schedule our scans and then to our -- I want to say to our surprise, but one of the reasons we're doing them was to see what happens. And in fact, what we noticed was a great many asymptomatic for the most part, evidences of cerebral edema. And therefore, if these studies have been done without such frequent data, we'd be talking about the below 10% incidence of symptomatic abnormalities. Now I noticed that the FDA wants them at certain points. If you look at the instructions. We're giving other antibodies. But I think Larry's point is emphasized by the fact that we wouldn't even know about this if we had not scheduled regular findings -- or sorry, regular scans to know that things were quietly going on beneath the surface. And if they don't cause any symptoms that may go away vastly by 3 months, that's acceptable to me. We have a lot worse side effects of our medications than this.

Great. Last question before we open it up to the analysts to pepper us all with questions. But with the investigators point of view, Dr. Honig what were your impressions of ACU193 in the clinic? And was there enthusiasm around the trial and the prospects for 193 by patients and caregivers?

I think seeing as the current therapeutic spectrum for Alzheimer's disease is confined to in terms of efficacy signals to the agents that remove beta amyloid from the brain, there was a fair bit of enthusiasm, but by this trial as another antibody that might have more specialized mechanism. But that will remove amyloid from the brain. I think it's a wide appreciation in patients and their caregivers that amyloid is the principal evader, if you will, in the Alzheimer's cascade, and that removing amyloid seems like a good thing. It's sort of something that intrinsically appeals to patients. And certainly, the evidence from this agent and other agents at this stage now that we know so far is looking very encouraging and patients perceive it that way as a useful thing to do to move something that's causing entry to the brain. And intravenous treatments are never quite as eagerly anticipated as subcutaneous or oral, but on the other hand -- or other methods of administration, but there doesn't seem to be a large burden with this intravenous treatment relatively painless and relatively acute side effect free treatment.

Excellent. And thank you both so much for providing your time and your perspectives on ACU193 and the results from the study. And at this point, I would like to invite the Acumen executives back and open the call for Q&A from the queue, the audio queue. So -- also, if we do have time at the end, we'll take some questions from the chat box if you have submitted any. So with that, I'll turn it over to the operator to open up that queue.

[Operator Instructions] The first question comes from Paul Matteis with Stifel.

Great. And congratulations on the data. I have three, if you don't mind. So I guess, as you sort of take a step back now, what do you see as the clearest opportunity for differentiation? Is it on efficacy because you're lowering plaques and also lowering oligomers and there may be an adverse effect, is it on safety, dosing, dosing or something else? Second, I know you still have to do some work on modeling for once monthly. But I think, Eric, you've talked about some CSF data from the SAD that has underlined your optimism that once monthly for 25 is viable. Can you speak to that? And what would you say your level of confidence is now? And then lastly, where are you on a subcu? And I guess, what else needs to be done to get that ready to enter the clinic?

Thanks, Paul. Can I take the first and last part of that question? And maybe, Eric, you can -- it's like a sandwich, you can pick up the middle. I think, firstly, we're still very much in pursuit of a best-in-class efficacy signal with our mechanism with this agent. I mean this was a Phase I first-in-human study. So we're very interested in seeing what the next I can produce in terms of efficacy and proof of concept potentially on the way to potential registration. So I think it is interesting. I think the safety profile is encouraging, but still needs more chronic dosing and exposures over time to sort of see where that settles in. But as we've noted on this -- in this discussion, I think we've got sort of this duality of a high dose that clearly can clear plaque at a rate similar to the first approved product. And a big dose that looks like it's in that zone, but may be a bit safer and ultimately may be more efficacious. And that's just -- we don't have that evidence just yet, but we do think that a robust Phase II/III study should bring some clarity into that risk benefit and the efficacy potential of 193. And then on the subcu, this is -- all this data is just fresh off the presses, and I think we still have some time to look at PK modeling and other elements as to how we get to a subcutaneous formulation. We had initiated some assessment of those options just on a technology basis earlier this year. I think those will now be more informed on the basis of the study PK. And I think probably we'll have an update by the end of the year as to what the next steps really need to be for subcu.

Okay. And Dan, just on the next clinical steps. You've talked about a Phase II/III. I think you've also talked about the possibility of some sort of concurrent biomarker study. Is that still on the table? Or how are you thinking about the different possibilities?

Yes. Thanks, Paul. So I think that is apparently now on the table, just given the biomarker effects observed in this study that one possibility would be to explore some sort of parallel study for the purposes of the biomarker. That is not -- you're hearing that from me. I think the team has rarely had a chance to digest this data, and we've got a lot more information to think about as to what our -- the broader development plan should be going forward. But certainly, with these data, it would appear that option exists. So you'll have to stay tuned. And maybe Eric do you want to comment on the concentrations in the CSF?

Yes, right. So as far as the 25 mg per kg dose. Yes, the data that we do have from the SAD, so this is after one administration of 25 mg per kg, the spinal fluid was obtained 21 days later. And it was -- the concentrations were well in excess of reported oligomer concentrations in spinal fluid. Now what some of our data are showing us now, especially that target engagement curve is that the antibody apparently is picking up oligomers in brain, interstitial fluid where the concentrations of oligomers are higher and then we're measuring it, of course, in CSF. So I think what's really going to be important about that 25 mg per kg every 4 weeks dose is we'll plot out where that our 3-week data are, and then we'll also have our modeled every 4-week data, but we want to know where we fall on that curve. And so -- and maybe just 1 quick comment about biomarker studies. I've also seen that there's a suggestion of what we do a sort of separate Phase II before you do a Phase II/III and that would be sort of a traditional sort of a drug development approach. I think we have ways in our Phase II/III that we don't have to create an extra study in between. But the team will need a lot of time to talk about that. And again, partly getting back to the dose, when we started the study, we didn't know whether we would have any plaque reduction, and we had some people say, well, why do you bother getting a PET scan after the baseline since you don't target plaque. And the answer is, well, that's the theory. Let's see what happens in the clinic. And then we have our answer. And so we've got both. We've got target engagement in terms of oligomers and we also have this effect on plaque but the effect on plaque was sort of less certain, but now we know we've got it.

Can I ask one last quick follow-up, I'm sorry. I just I forgot to ask this, and I think it's important. Just on the clinical endpoints, I know you said they're not powered, but just can you just offer any insight? Was anything trending in favor of the drug given that I do think, at least on the curves for some of the other drugs like lecanemab at an early time point, there may be some numerical separation, like -- what did you see on CDR? Or is it just too small of a sample to really speak to this?

Yes. Well, let me -- without getting too much into the weeds of drug development. So for like in the CLARITY study, you can start to see things on some of the clinical measures like the CDR fairly early, but that's a study of 1,800 patients. We have a study of 8 or 10 people per cohort, and 2 of those were on placebo. So the sample sizes are so small that I think you really do have to be careful about not trying to read the tea leaves too carefully. And Steve and Larry may want to comment on that, too. So what I will say is that with this computerized cognitive testing that we did in collaboration with Cogstate, we were looking for kind of a symptomatic bump and improvement with only 2 or 3 doses of drug. And we didn't find any convincing evidence of that. But the other thing to keep in mind is we also didn't see any indication that people became worse, and there have been drugs developed for Alzheimer's disease. It actually made things go the wrong direction, and we didn't see any evidence of that at all. So this is a little bit of a cautious answer to your question because I could pull out data to say this looks good or that looks good. But the reality is in a study like this, you have to kind of humble in terms of not trying to overread the data and things like the CDR some of boxes.

Our next question comes from Tom Shrader with BTIG.

Congratulations really remarkably rich Phase I study. A couple of dosing questions for Eric and then maybe a mechanistic follow-up. Eric, you've talked about titration dosing as it's reduced ARIA in the past. Are you still thinking about that given that it looks like you have to go to quite high antibody doses to really see plaque removal at all? Or does that make less sense? And then a follow-up to the last question that I think you talked to, but is modeling good enough now that you would launch into a 25 mg monthly cohort in a bigger trial? Or do you think you'd do some sort of smaller experiments first?

Yes. So as far as the titration question, as you know, that strategy was used with Aducanumab and for me, maybe a little surprisingly, but it seemed to work. And so I think that's something that as we go to 60 mg per kg, there's no reason not to do that, and it would be something like 45 mg per kg for 3x and then jump to 60 mgs per kg. And so yes, we're -- again, we haven't finalized any of these decisions yet. But I don't think there's a lot to lose to do that unless you would do a very prolonged titration period when you're kind of under-dosing for a long time, but I don't think we would do that either so. So that's one of the things that we would do. In terms of your question about launching into a study using model data, again you could do a more traditional sort of Phase II at this point. But the way our Phase II/III study is being designed is that you will get to an interim analysis. And the interim analysis will look at a number of things. It will look at clinical measures that will look at biomarkers and a number of things to make a decision whether to scale up to full-size Phase III or to just let the study complete as a Phase II, which we may end up doing. But the DMB who will be looking at those data may also recommend dropping a dose. So if it turns out that at the lower dose, whatever it ends up being, you don't really see anything in biomarkers or clinical, maybe you drop that. On the other hand, maybe at the lower dose, you have really good efficacy, but a lot better safety and maybe you drop the higher dose. So the way that adaptive design is being done is it allows you to do some of the things you would do with a stand-alone Phase II study. But at the same time, not do that stand-alone Phase II study and create that extra white space and basically delay the antibody becoming available to patients.

Got it. If I can ask a quick mechanistic follow-up. The patients that had the maximum plaque removal, did that patient essentially clear plaque? Or do you have a sense of the percent plaque reduction? Or it looks like probably just that 1 patient that's relevant.

Yes. No, thank you for asking that. I should have clarified that. None of these patients had their plaque lowered where we would consider them amyloid negative. So again, it's a short study and whatnot. But generally, somewhere between 25 and 30 centiloids if you get below that, you would be considered amyloid negative, and none of our patients dropped below 25 or 30 centiloids. Now again, remember, that's with 3 administrations a drug in 3 months. Now what might happen after 18 months, we'll have to find out. But at least in this short study, we didn't see anybody, even though they declined, they didn't decline to the point where we would consider them amyloid negative.

The next question comes from Colin Bristow with UBS.

Congrats again on the data. Maybe just a follow-on on Phase II dose selection. So your [indiscernible] curve suggests maximum target engagement from really the 25 milligrams and at 2 weeks and the 60 milligrams at 4 weeks. I'm just curious, what is your thinking around evaluating a potential intermediate dose? And I'm guessing at 4 weeks. And then second, can you give us more details on the ARIA-H that you observed in the trial, I think there were 4 cases. Was there any cerebral hemorrhage here? And can you comment on that with regard to APOE4 status? And then maybe lastly, can you just remind us if you measured NFL in the trial? And if you did, is there anything you can share at this point?

Yes. So again, thanks, all good questions. We actually just had this conversation earlier today is while we're modeling data, say, for 25 mg per kg every 4 weeks, we could ask the model what if we did 35 milligrams every 4 weeks or 40 milligrams every 4 weeks. And I think that's one of the nice things about modeling data is you can make those kind of estimates in terms of adjusting doses. And so I think that's one of the advantages we have with this rich data set, but now we have something to model with. In terms of ARIA-H, yes, we did have 4 cases. These were people who just either went from 0 to 1 or maybe 2 to 3. Anyway, there is only an increase in one ARIA-H. I can't off the top of my head, tell you what their APOE status was, I'd have to look that up, but there wasn't anything remarkable about that I know. I think, the most important thing about ARIA-H is I think it's becoming kind of apparent that when you talk about hemorrhage, that sounds like a pretty scary thing to a lot of people. But for ARIA-H, the vast, vast majority of those are microhemorrhages. So those microhemorrhages happen in people with Alzheimer's disease as part of the disease course. And so there -- microhemorrhage is virtually always asymptomatic. Now if you look at observational studies from a statistical standpoint, if you have more microhemorrhages, you may have a little bit more disease progression. But certainly, you don't see anything within an individual patient. Now the macrohemorrhage, which would really in the absence of anticoagulants, be a very rare event, that obviously would be worrisome. But 99% of what we're talking about here in the absence of anticoagulants would be these microhemorrhages, which are really asymptomatic, you see them with the disease. It's in most patients with Alzheimer's disease after a period of time will have microhemorrhages. So I think that's created maybe just a little bit of confusion. And the word hemorrhage is little frightening, but these are microhemorrhages. And then finally, as far as NFL goes, recent studies for Alzheimer's disease have really shown that NFL is necessarily the best biomarker. But we will be setting out fluid biomarkers, both plasma and CSF, we're kind of in the final stages of deciding exactly which biomarkers will send out. Certainly, it will be one of the PTAs probably GFAP that sort of thing. But the one thing I would caution people on is, this is a small study where people got at most 3 administrations of the drug. So I would not expect to see any change in those downstream biomarkers. I think our reason for wanting to run those is to get some familiarity with the labs, with the variability, that sort of thing. So that when we really do need to rely on the results like for our interim analysis in the Phase II/III we'll have a really good understanding of what we get from these labs with these biomarkers. But I personally wouldn't really expect to see a change in those in this small, short study.

If I can just jump in really quickly since we have Dr. DeKosky and Dr. Honig on the call, do you agree with Eric in his assessment of ARIA-H, and would you have anything to add to that?

I think the major thing I'd add is having done this for 30 years. You see people who have microhemorrhages, which basically is a spot around a vessel where in all likelihood amyloid in the vessel wall has caused weakness and hemorrhages or blood has come through, and those spots are asymptomatic. At the same time, because of the amyloid, it builds up in the blood vessels of the vast majority of people with the disease, some more than others. There is a spontaneous hemorrhage, not just a bit of blood around a small blood vessel but actually a mass hemorrhage that occurs in people with AD who have a significant amount of angiopathy that is deposition of amyloid in the blood vessels. And those occur spontaneously in people with the disease. If we had been unlucky enough to have a hemorrhage occur in this relatively small state, we'd be debating forever, whether this was one of the spontaneous hemorrhages or whether it was caused by the medication. So there are these -- not unlike the appearance and disappearance of ARIA, there are these things that occur in the brains of AD patients. And when you look at a significant number over a significant amount of time, you see them. This has been, as far I'm concerned, a pretty clean study so far. There were no big flags here that say something about this drug is not only toxic, but will be toxic if you continue these study. So I would -- like Eric, I would minimize the thought that there was a significant meaning to the small hemorrhages. I call them blood dots. And as long as they're asymptomatic, I would say it would be perfectly reasonable to continue. Larry, I'm curious about your response to this one as well. I don't want to put you on the spot.

Well, no, I think it's why they recognize that microhemorrhages like Eric and you have mentioned that are asymptomatic in the vast majority of the cases. In some cases, there may be brief transient neurological events associated with them that may or may not be associated by the patients or their practitioners. But microhemorrhages is symptomatic. I said microhemorrhages generally with that the drugs that have more extensive data in this past broad class of amyloid agents have not been particularly associated with the drug itself, but rather just simply with [indiscernible] of time. As Eric alluded to, the major concern in [indiscernible] of course, is always the macrohemorrhages are larger hemorrhages which often can be symptomatic. And again, it's not clear that they're in the drugs and the other drugs which they have more extensive data, it's not to the drug related. And again, it does seem like edema or swelling even though it doesn't sound as bad as bleeding is the greater concern with drugs of this class. So obviously, in patient perspective, the idea of bleeding in the brain sounds very fearsome.

I just wanted to briefly give you a anecdote. I saw a patient on a Thursday, women who had fairly obvious. I'd say, moderate Alzheimer's disease. And when she came in on Monday the following week for an MRI scan, I got a frantic call from the resident on call because he's been called to radiology because this lady had a rare parietal superficial hemorrhage of some significant in size, although she walked in. Resident Senior found all sorts of left-sided abnormalities, including a little bit of neglect and looked at my evaluation from 3 days before and said, did you miss all this? I said, "Well, no, she [indiscernible] over the weekend. And of course, this is totally spontaneous is was in the days before. She was not in a trial. It wasn't very much going on except in [indiscernible]. These things occur spontaneously. And as I said, over the course of, let's say, an 18-month trial, it would not be unexpected to see a few of these spontaneous ones. You have to hope for is that your drug does not precipitate any and so far, we look like we're in very good shape within that record.

That's great. Thank you. Thank you for -- thank you, both. I'll turn it back over to the operator to continue the queue. Apologies.

The next question comes from Pete [indiscernible] with Cantor Fitzgerald.

Dan and team, congratulations on the DAC presentation for 193. My first question is about next steps and how you're thinking about the next study. I know the data is fresh and you're going through it. But is there a route forward where you can try and tease out potential therapeutic benefit not associated with the plaque reduction but perhaps associated with oligomer removal those where -- there could be no or really low ARIA rates but close to maximum target engagement. And then a dose you expect there to be a low ARIA and plaque reduction, say, using lecanemab as a benchmark? Or is your current thinking that you just go for doses that will likely show plaque reduction since there's an established regulatory path forward for accelerated approval?

Great question, Pete. I'll just -- I'll lead up, maybe Eric can provide some comment. I mean I think we've established that 60 mgs per kg every 4 weeks is sort of the highest rational dose given the target engagement and given the plaque observations. I think that a mid-dose that generally has been described on this call and maybe referred to as the 25 -- like a mid-dose at 25 mg per kg every 4 weeks. Presumably is not going to have that rate of ARIA as may achieve sufficient oligomer target engagement to sort of demonstrate that unique benefit of that mechanism. So as we said, I don't think we have dialed in precisely the doses that get carried forward here. But I do think that with high and mid doses, we should have sufficient target engagement and be able to pursue clarity on the best risk benefit profile in terms of safety and efficacy for patients.

Yes. And maybe just to add a bit to that, and again, I kind of mentioned this before, is that these clinical measures are -- have a lot of variability and really are very noisy. And you can look at large trials like the CLARITY trial or the ENGAGE -- well, the EMERGE trial. And look at group differences. But when you try to look at individual subjects, I haven't done this for a long time, you can sort of kid yourself and talk yourself into things that may not really be there. So I think what we really rely on is things where there's less variability. And so our target engagement assay is one of those things. The PET scan is another thing. You don't get placebo responses in as much. Well, the test retest, let's just say in a PET scan is much better than something like a CDR sum of boxes. So we'll rely on those things to guide our dose selection. But again, for the next study is a Phase II/III, then when you get up to bigger sample sizes, part of that interim analysis will include those clinical measures.

All right. And for Dr. DeKosky and Honig, the symptomatic ARIA that was observed in your experience or from your discussions with others who treated patients, with anti-amyloid antibodies is leg dysfunction commonly observed for those experience in a symptomatic ARIA? In other words, would you be confident that leg dysfunction was a result of ARIA? Or would you be on the fact..

So I mean, ARIA is a team of the brain. So typically, the symptoms related to the location of the edema. And sometimes scenarios that's remarkably silent. Sometimes it's just accompanied by a nonspecific symptoms such as headache, sometimes by an electrical irritability, causing a seizure. But certainly, in this case, it was viewed as related because, in fact, the leg dysfunction was in the -- an anatomically correlated to the area of the brain that was affected by the ARIA, they happen synchronously. So I think the overwhelming odds in this case, although one can never be certain, given that there was unexplained leg dysfunction that related to the area of the brain was that it was symptomatic. But again, it was entirely reversible, radiographically and clinically as the vast majority of ARIA-E is...

So if I remember correctly, Larry this hurt, edema hurt ARIA was in the right parietal lobe and if so, if it's posterior enough that it doesn't cause primarily weakness, it might well cause some kinds of either neglect or alteration of [indiscernible] of her leg which makes her feel like something nonspecific that isn't as clear as we'd like to have it. But given the fact that it occurred in the region, as you said, it's referable to right leg function probably was due to my view would be that it was probably related to the ARIA. On the other hand, she was walking and she just had an auditory complaint. So it speaks again to the fact that ARIA can be might be mild and in many cases, never noticed. It takes a lot of ARIA, sometimes a few we've seen that have been diffuse that is in multiple places in the brain before people actually are confused or have some kind of cognitive disturbance above that of the dementia itself. In her case, she was describing things, so it obviously did not close any kind of global problem.

The next question comes from Judah Frommer with Credit Suisse.

Yes. Congrats again on the data guys. Just a couple more from us. Maybe first, just as you continue to think about modeling Phase II dose, can you give us a little more detail on the immunogenicity that you saw in this trial and how that might factor into dosing in the Phase II/III? Additionally, can you give us a little more color potentially, Dan, on what the subQ update might look like at the end of this year and then lastly may be for Eric. Just I guess as you think about the data and obviously, you are still going through it. But given the ARIA, given the plaque reduction, any evolution in the oligomer hypothesis that you're considering at this point? Or do we kind of need to wait for functional demonstration to start altering that hypothesis?

So talking about Dan, do you want me to take the first and the third one, and then I'll let you comment on the subcu.

Sure.

Yes. So thanks. So yes, really interesting questions actually. So in terms of the modeling the dose versus immunogenicity. I mean that actually a really good illustration of the fact that early on in analyzing the data. And we certainly would want to -- if we did, and so far, we haven't. But if we did come up with an individual who had an antidrug antibody and then that was neutralizing and so they had a reduction in drug. Yes, we would figure that into the modeling because you probably wouldn't model in a person who had that. Again, we haven't seen that so far, but we don't have all the data yet. So yes, we would definitely figure that in. As far as your broader question in terms of sort of the oligomer hypothesis and does the ARIA and the plaque reduction change that? That's something that will probably be discussed for a long time. I think as far as right now, this minute, I would say that our oligomer hypothesis by itself hasn't really changed. And our feeling about plaque reduction hasn't really changed. And I think it is important to note that for the drugs that reduce plaque that have been successful they do reduce plaque down to essentially 0 by the end of 18 months. And so this is very analogous to lecanemab actually, where they target protofibrils, but they get a great deal of plaque reduction to the point where it does get down to essentially 0. And so what accounts for the efficacy there, is it the plaque reduction? Or is it the targeting protofibrils? I don't know that there's any way to answer that. And I think we could end up in the same situation. We'll see what we have in terms of plaque reduction at 18 months but we can end up in a similar sort of situation. But I just saying, I am a clinical trialist and so even though these mechanism questions are really interesting, and we talk about them a lot. At the end of the day, what we really need to see is how much does it slow disease progression and what's the safety profile. And those are the 2 things that will really allow you to make a decision -- well, it will allow regulators to make a decision to approve it, but I think it will allow clinicians like the two folks we have on today to make a decision, do I want to treat my patients with this drug. So the mechanism questions are really interesting. But at the end of the day is what does it do for people clinically?

Thanks, Eric. And Judah, I think in terms of subcu, I would expect by year-end to have kind of the game plan in terms of time lines and format and sort of clarity around precisely how we're going to move forward.

The next question comes from Charlie Yang with Bank of America.

Congrats on the data. This is Charlie Yang on for Jeff Mitcham. So I have three questions. Number one, mechanistically, can you just maybe provide more clarity on what's the reason for the lack of ARIA-E observed in the homozygous patients? Is there any mechanistic rationale for that? Number two, in terms of the trial design, and I believe CLARITY AD, lecanemab, they suspend dosing for patients who experienced moderate even for patients who only experienced moderate asymptomatic ARIA-E. I'm wondering if that was done here and perhaps maybe going forward, what's the intention in terms of the dosing from that perspective? And number three, based on the [indiscernible] test results that you are seeing on Phase I, do you intend to continue to use that test in a later stage of study? Or are there any commodification that you intend to do based on what you observed?

I'm sorry, the last [indiscernible] test. Yes, we will include that in the next study. I think those are to you...

Yes. Okay. So yes, as far as the lack of ARIA in the APOE4 homozygous, I can't tell you today. I really don't have a mechanistic explanations to that. And I mean the fact that it's the opposite of what's seen with other antibodies. And again, you have to preface all of this was saying, this is a very small trial and It's very preliminary, and we'll have to see how all this plays out. But since it's -- if this does hold up, and it's essentially the opposite of other antibodies, we have higher rates in homozygous, then I think maybe we have to sort of consider that and what our theories are of what really causes the ARIA. And that's not entirely clear exactly how that works. So we'll have to think through that. But I have to admit, right now, I can't come up with a really good explanation for that. But it's a finding and we'll continue to follow it. In terms of what we would do in a Phase II/III in terms of dosing, if a person has ARIA. I think that's a really good question. And we haven't made any final decisions about that. I think as the field actually has come to be more comfortable with, especially asymptomatic ARIA. I think, could consider for people who had asymptomatic ARIA that was radiographically mild that we would continue to dose those people. It turns out in early studies of bapineuzumab before this ARIA was very well understood or described that they dosed through ARIA a lot because they didn't know any better at the time. And actually, it was -- it didn't cause big problems. So I think we'll need to have some criteria for that. Obviously, if somebody is symptomatic, you would hold the dosing. I think even if they were asymptomatic, if radiographically, it was, say, moderate, you would hold the dosing but we need to make final decisions about that. And then as far as the Cogstate testing, yes, so I'm glad you asked that question. That Cogstate battery that we showed the results on. There's actually 5 different things that go into the battery, and we've only seen the results of the composite of all 5 of those things. Now, I would guess that some of those pieces of that battery will work better than others. In fact, I had a conversation with someone from Cogstate earlier today saying one of the next steps that we need to do is to dissect out that battery and see which things seem to be useful, which things are not so useful and then decide what to incorporate into the Phase II/III. I mean the whole battery is nearly half an hour as it is. So it's not particularly burdensome. But if you can make it shorter and make it better, that would be a good thing.

I show no further questions at this time. I would now like to turn the call back over to Alex.

Great. Thank you so much for everyone who joined today. We don't have time for questions in the chat box, but we will get back to you if you did submit one. Thank you again to Dr. Honig and Dr. DeKosky for their time. And I encourage if you have questions to anyone in the audience to reach out to us here at Acumen. Thank you again, and have a great day.