Ladies and gentlemen, the program is about to begin at this time, and it's my pleasure to turn the program over to your host, Alex Shanahan. All right, great. Thank you, Operator, and thanks everyone for joining the session with Alector Therapeutics, part of our B of A 2025 CNS Conference. My name is Alex Shanahan. I am a Senior Biotech Analyst covering SMID-cap and Alector at Bank of America. And I have the pleasure of being joined today by the Alector management team, including Arnon Rosenthal, Co-founder, Chief Executive Officer and Director; Giacomo Salvadore, Chief Medical Officer; and Neil Berkley, Chief Business Officer and Interim Chief Financial Officer. Thanks, guys, for being here. Thank you for inviting us. Great, great. Well, let's jump right in. I mean, there's a lot to unpack from the pipeline. But Arnon, maybe if you want to give a quick overview of the company, clinical programs, and upcoming catalysts. Yes, absolutely, so Alector is a neurodegeneration-focused company. We are committed to finding therapeutics for Alzheimer's disease, Parkinson's disease, and other neurodegenerative disorders. We have so far taken five drugs to the clinic. We have an ongoing Progranulin-elevating drug in Alzheimer's disease. Progranulin is a risk. Loss of function is a risk for Alzheimer's disease, and progranulin is actually a universal risk gene for neurodegenerations. Loss of function of progranulin is associated with frontotemporal dementia, with Alzheimer's disease, with Parkinson's disease, with ALS, with LATE, which is another type of late-onset dementia, and we developed a Progranulin-elevating drug, and together with GSK, we are now testing it in Alzheimer's disease, and we are expecting to have an interim analysis of our phase II study in the first half of next year. In addition to the clinical programs, we have multiple preclinical programs that are propelled by our Blood-Brain Barrier technology. We have an anti-amyloid-β antibody drug that's targeted to be in the clinic next year that's driven by our unique proprietary Blood-Brain Barrier technology. We have a GCase enzyme replacement therapy that's propelled by our Blood-Brain Barrier technology for Parkinson's disease and eventually Lewy body dementia. More than 10% of Parkinson's patients are associated with loss of function mutation in this lysosomal enzyme. And up to 30% of Lewy body dementia is associated with loss of function mutation in this lysosomal enzyme. And we engineer the enzymes and optimize Blood-Brain Barrier technology that enabled effective delivery to the brain as an enzyme replacement therapy for these brain disorders. In addition, we have a whole portfolio of Blood-Brain Barrier-enabled siRNA programs, including Tau siRNA, α-synuclein siRNA, NLRP3 siRNA. We think that with peripheral delivery of siRNA, you can really expand the usage of siRNA, develop a safer siRNA and more effective siRNA that can distribute homogeneously throughout the brain. We are, again, focusing on multiple types of neurodegenerative disorders with multiple drug modalities. We have resources to really bring these programs to value-creating points. Okay, great. I think that's a great introduction. And maybe we can sort of piggyback off the R&D day that you guys hosted back in September and starting on the ABC platform, since that was a major focus of that update. And I think most investors are familiar with sort of the limitations of prior approaches to delivering drugs to the brain. How is the ABC platform maybe helping to solve for this or help broaden the therapeutic window? And how are you thinking about building out the pipeline in terms of target selection, optimization, et cetera? Conceptually, the sort of Blood-Brain Barrier technologies enable the delivery of large molecules like antibodies, enzymes, and nucleic acid to the brain at 10- 50 fold higher concentrations. And this enables the usage of sort of lower dose with peripheral injections. It enables higher safety. And in many cases, like in enzyme replacement therapy and in siRNA, there is really actually no possibility with peripheral injections to get to the brain. With antibodies, which have a longer half-life and they are more stable, you can get a very small percentage of the drugs to the brain after peripheral injections. And the approved anti-amyloid beta antibodies, for example, are naked antibodies. They are injected peripherally, and they still get at enough doses to the brain to show therapeutic benefit. But with Blood-Brain Barrier technologies, you can use up to, again, 10x lower dose. And because the drug enters the brain through a different route, you increase the safety. For example, the naked anti-Aβ antibodies show a higher degree of area of inflammation and blood vessel leakage because they bind, likely because they bind to Aβ plaques on the large blood vessels. With Blood-Brain Barrier technology, at least the Roche antibody has shown that you reduce and practically eliminate these side effects. So again, Blood-Brain Barrier technologies increase the amount of drug to the brain, increase the efficacy because you distribute the drug homogeneously throughout the brain, even in deep regions that naked drugs don't get into. And in many cases, it increases the safety. So I think it will revolutionize sort of drug treatment for brain disorders and enables, again, drug modalities like enzymes and siRNA that were completely not accessible to the brain before with peripheral injections. Yes. And maybe we can go a little deeper there on sort of around the technological progress that you guys have made, just around the technology and how you're sort of achieving the Blood-Brain Barrier penetrance that you're seeing in your preclinical models. Yeah, so we, as quite a few other companies, are using the transferrin technology as a Trojan horse to transport antibodies, enzymes, and nucleic acid to the brain. So the Blood-Brain Barrier has transferrin receptors normally to deliver iron to the brain, which is a required nutrient. And basically, this technology hitchhikes or just uses this receptor and enables the delivery of, again, enzymes, antibodies, and nucleic acid to the brain. Multiple companies are using these technologies, but there are significant subtleties that make some technologies better than others. So the main differences between technologies are sort of the range of transferrin affinity binding that is being used, the epitope on the transferrin receptor that is being used, and the drug configuration, whether you use bivalent binding or single-valency, what do you use to bind the transferrin receptor. These differences have an impact on the level of access of drugs to the brain. They have impacts on the durability of the drug in the serum, which impacts the dosing intervals. And it impacts the safety. Unfortunately, although TfR is very effective in getting drugs to the brain, there are 10x more transferrin receptors on red blood cells than on the Blood-Brain Barrier. So a lot of the drug actually goes to reticulocytes instead of to the Blood-Brain Barrier. And if there is a lot of drug binding reticulocytes, it depends on the drug configuration, whether it has an active effect or function or not. But it can cause damage to reticulocytes, and that causes anemia. So anemia is really part of the sort of target-mediated adverse effects of this technology. So again, you have to find an epitope on the transferrin receptors that reduces the anemia. You have to tailor the affinity. Enough drug gets into the brains, but the drug doesn't have enough time to bind reticulocytes to facilitate immune response against reticulocytes and damage. We were able to really identify what we think is a good unique epitope that reduces the ability to induce damage to reticulocytes. We have like a thousand-fold range of affinities that we can optimize different drug modalities to maximize brain penetrations and minimize adverse effects. For example, with our anti-amyloid beta drug, we think that we are able to bring drugs to the brain at up to like 12-fold higher concentrations than competitors. With siRNA, based on what we see from the literature, we can bring siRNA to the brain at more than 10x higher concentrations than competitors. We think that we have significantly reduced anemia because of the epitope that we are using. We think that, again, not all transferrin-mediated Blood-Brain Barrier technologies were created the same. We think that time will tell in the clinic, but we think that we have a really exceptional technology that enables very good delivery of multiple drug modalities to the brain with very manageable safety profile. Yeah, no, that's helpful. And I do want to sort of ground the discussion in the different assets you've nominated so far, but maybe just to sort of frame sort of the breadth of the approaches that you're pursuing. And I think one interesting thing that came out of the R&D day was just around the adaptability of the ABC platform. You've got the brain carrier, right, that allows for crossing the Blood-Brain barrier, but you're able to adapt the antibodies with multiple therapeutic arms or an enzyme cargo or a nucleic acid cargo. I guess maybe just quickly touch around the sort of medicinal chemistry piece that got solved for the platform. Yes. So it means the way we link the transferrin moiety to the drug, basically to create the shuttle combined with the cargo, we have a lot of flexibility in this, like where we bind the transferrin, like where we integrate the transferrin moiety with the cargo. And for each drug modality, we do it differently. Like for antibodies, it means the transferrin moiety is at the sort of C-terminal of the antibodies away from the binding domain, like from the action of the antibodies. For enzyme replacement therapy, it's part of the arms of the antibodies. And for siRNA, it's in a third place. So we have enough flexibility. We optimize the drug configuration for each drug modality. And that's really important because we actually tested five different drug modalities, five different configurations for each drug modality. It makes a really big difference on the ability to enter the brain based on the drug modality and also on the half-life in the serum and on the safety. So the drug configuration is really part of the critical components of a safe and effective transferrin-mediated technology. So again, we have a thousand-fold range of transferrin affinity that we are optimizing. We have multiple transferrin epitope binding domains that we are optimizing. And we have multiple drug configurations that we are optimizing. And the three things together really enable us to create an optimized drug that's really dependent on the requirement. And again, different drug modalities have different requirements. Like enzymes and siRNA have shorter natural half-life in the serum. So you want them to move to the brain faster. So you need a higher affinity compared to antibodies. Antibodies like A beta or tau likely require the full effector function because you want to recruit the immune system to remove the A beta plaques or the tau aggregates. So if you need a fully functional effector function, you increase the safety risk for reticulocytes. So you need to find the TfR epitope that reduces the adverse effects. So again, it really depends on what the drug modality is and what the drug requires, whether it requires. Sort of, the history around A-beta targeting sort of fed into your design for your molecule. Yes. So anti-Aβ drugs for some have been around for like 25 years now, I think. And the first generation was completely ineffective because in some cases, people used like an inert effector function. In other cases, people used the wrong binding epitope to Aβ. But as you know, like more recently, there have been three anti-Aβ drugs that were approved. Two of them are on the market. And both of them are what I call naked antibodies. They are just antibodies without Blood-Brain Barrier technology. And these are sort of lecanemab and aducanumab. And they both show significant removal of Aβ plaques over a six-month period and modest clinical benefit of 25%-30% slowdown in cognitive decline over 18 months. The main liability of these drugs outside of the modest clinical benefit is that they are associated with a higher level of what's called ARIA, like meningoencephalitis and blood vessel leakage, especially in APOE4-positive Alzheimer's patients, which are the 70% of the Alzheimer's population. So the adverse effects are really a major issue for this first-generation drug. So as you know, like Roche recently came up with an anti-amyloid beta drug with the Blood-Brain Barrier technology, transferrin-mediated Blood-Brain Barrier technologies. And so far, they were able to show that this reduces the ARIA-related adverse effects. It seems to be like to zero, like the level of ARIA that they report is indistinguishable from the level you've seen in untreated Alzheimer's patients. And they show a more rapid and extensive amyloid beta removal, which could translate to better clinical benefit. So it's pretty clear that second-generation anti-Aβ drugs with Blood-Brain Barrier technology will displace the naked antibodies. And then among the anti-Aβ drugs with Blood-Brain Barrier technology, we will have to see what stands out, like whether the level of antibodies that you get into the brain, whether the level of anemia, and in the case of the Roche antibody, whether the level of the infusion reaction that requires steroid treatment will really make an impact on the drugs. So again, we think that we can really excel with our anti-Aβ antibody. We see that we can deliver very high concentrations of antibodies to the brain, 10- 12 fold higher than what was reported by competitors. So this will enable significantly lower dose and likely subcutaneous delivery. The lower dose will also be associated with lower safety issues, like infusion reaction, for example. We think that our drug would enable subcutaneous delivery, which would be transformative for this class of drugs because having to go to infusion reactions every month is not convenient for patients. There are not enough infusion centers for all the Alzheimer's populations. The compliance is low because of the requirement for infusion. We are really targeting sort of subcutaneous delivery at much lower dose, which will facilitate safety. Hopefully, our drug will not, because of the low dose and the configuration of the drug, will not elicit infusion reactions, which is a major liability for some of our competitors. Again, the anemia, like Roche seems to manage the anemia. Again, we think that the epitope that we have chosen would have lower anemia than the epitope that competitors have chosen. Finally, we are like to bind the Aβ plaques. We have used the pyroglutamate domain, which is different than what sort of the Roche antibody is binding. The Roche Trontinemab antibody used Gantenerumab, which as a naked antibody was not effective clinically. It binds generally the N-terminal of A beta. And we think that our epitope, as Lilly has shown, is the strongest in removing A beta as a naked antibody. So we think it will be even stronger in conjunction with Blood-Brain Barrier technology. So we think that we have a really good combination of an optimal anti-A beta epitope, a fully functional effector function that would enable very profound removal of A beta and transferrin binding epitope that has reduced anemia and can deliver drug to the brain at very high concentrations. So the combination of these features, I think, could hopefully make it a best-in-class drug. Okay. That's super helpful. Thanks for the really detailed overview of the program. You also have AL137. Maybe you could just speak to the optimizations that have gone into this antibody and where you'd sort of direct investor attention as you're targeting for some of these studies next year. Yeah. So we have a lead antibody, which is AL137, and a backup antibody. They have somewhat different features. Both of them have really good brain penetration, like 4-12-fold higher than what we think competitors can deliver. One of them has slightly higher risk of reticulocyte damage. So we have like balance between efficacy and safety. We are going for maximal efficacy because we want to move to subcutaneous and be able to use really low dose. So the drug that enters the brain best is our lead. And we'll see how it works in the clinic. And we have right behind it, we have a backup antibody in case we need to optimize the safety features. But we are, again, targeting to have our lead in the clinic in 2026. Giacomo can describe the clinical plan, but our goal is to go to patients and to show amyloid beta removal and minimal ARIA and no manageable reticulocyte damage and hopefully no requirement, like no issues with infusion reaction as quickly as possible. Yeah. I guess on that, do you think the study and maybe Giacomo can weigh in? Do you think the study will be designed in a way where you could see dosimetry, maybe on ARIA, just to get a sense of sort of the, because it is an on-target, right? If you can get a sense of sort of the biologic effect of AL137, I guess just what is the information that you think we'll be gaining from the study? Presumably, efficacy is maybe a secondary there. Yeah. I think thanks to the implementation and standardization of amyloid PET imaging, which is the gold standard pharmacodynamic measure for anyone studying anti-amyloid treatments in the clinic, we are able to show to investigate the effect of the drug on amyloid clearance relatively quickly and in a small number of patients. 10-15 patients are enough to have a good estimate of the amount of amyloid clearance. 10-15 patients, and this is an ideal number to make it suitable for multiple ascending dose studies, so as early as in phase I and to gather this very important information. You asked about ARIA. For ARIA, I think it's a similar reasoning because there is a background ARIA rate. There is 5%-10% in patients with AD. However, the anti-amyloid treatments, the first-generation ones, Aducanumab, Lecanemab, and Donanemab, they were showing much higher ARIA rate. These figures that I talked about, that is 5%-10% natural occurrence of ARIA, 20%-40% ARIA rate observed with anti-amyloid treatments allow us to have a good grasp on the liability on ARIA of AL137 as part of the multiple ascending dose study as well. A cohort of 10-15 patients, those with the drug, already give us a good idea about the ARIA risk. Of course, the cohorts in the MAD can be expanded and one can get better and more precise point estimates to really allow us to choose the optimal dose to move forward. In summary, I think looking at considering amyloid clearance and ARIA rate, we can get meaningful answers in phase I with multiple doses. Early in the development program without waiting for a proper phase II study. So ARIA occurs very early in the treatment after the first or second injection usually. So you can very quickly see if there is ARIA risk or not. Okay. Okay. That's helpful. Maybe for the sake of time, we can move on to AL50. This is another ABC, a GCase-ERT ABC for Parkinson's. Maybe to start, do you think the longer plasma residence and activity is a differentiating characteristic of this asset? Maybe you should put that into context of competing therapies. Yeah. So just to recap, over 10% of Parkinson's patients, up to 30% of Lewy body dementia patients are associated with sort of loss of function mutations in these lysosomal enzymes, GCase. Also, all Gaucher disease patients are sort of caused by loss of function mutations in this enzyme. For the peripheral pathology of GCase, Gaucher disease, there is enzyme replacement therapy currently that works very effectively. But because the enzyme is very short-lived, like minutes in the serum, it doesn't enter the brain. So the Parkinson's pathologies or Lewy body dementia pathologies cannot be treated with current enzyme replacement therapy. So we did two things. First, we engineered the enzyme itself to increase the half-life in the serum, as you mentioned, and activity by 10- 40 fold. And we integrated with our Blood-Brain Barrier technology to enable to bring it to the brain. Yes, I think that the longer residency in the serum enables the drug more time to enter the brain. I think it's part of the mechanism of action. I think there are two components, like l onger time in the serum that gives it time to enter the brain through the Blood-Brain Barrier technology and higher resilience in general that enables the enzyme to transcytose to the Blood-Brain Barrier and then to go into the lysosome without losing activity. We tested thousands of enzymatic mutations to really have an enzyme that's resilient enough to stay in the serum like hours instead of minutes to be able to enter the brain through the Blood-Brain Barrier. And then not only that, but also enter the lysosome, which is the natural site of action, without losing activity. And for this, we have to optimize both the enzyme itself and the Blood-Brain Barrier technologies. And now we were able to show in non-human primates that we are able to, with peripheral injections, bring the drug to the brain, to the lysosomes and retain activity. And we think that the non-human primates are very predictive for humans. So we think that we would have a drug for Parkinson's disease and other indications that brain deficiency and GCase are pathological. Okay. Yeah. And I think you've said that you're targeting first-in-human studies in 2027 for this asset. Obviously, you mentioned opportunity in Parkinson's, but also Gaucher disease and Lewy body dementia. I guess, how do you sort of prioritize the indications here? And I guess, what are sort of the data points you'll be looking for to make those determinations? Yeah. I can take this one. Parkinson's disease is the lead indication. It's a disease where there are only symptomatic treatments being approved. And GCase remains a very interesting target given all the genetic links that Arnon just mentioned. I mean, we will start the phase I program in healthy volunteers and then move early to investigate the effect of the drugs in patients with Parkinson's disease and GBA1 mutation. We can definitely have a look into pharmacodynamics, measuring the effect on GCase where we expect to see an increase. And then we also plan to look at a marker of disease progression. These are more exploratory. But in the end, I think we can get a good grasp on the dose to be moved forward through the phase I program, both in healthy volunteer and patients as it pertains in pharmacokinetics and pharmacodynamic results. One of the issues with Parkinson's disease, of course, is the fact that the endpoints are very noisy. So in the end, moving towards the clinic, we will have to think about studies that will need to have sufficient sample size to look at the effect of the drug as compared to placebo and disease progression and make the decision on the subsequent steps. But the main point that I would like to leave you with is the fact that there are clinics that have already patients with PD and GBA1 mutation, non-mutations that typically have databases and are ready for being enrolled in clinical trials. So we can have early data in patients, which are always valuable. I think there is a shift towards having data in patients as early as possible. And that's the strategy that we are following. Okay. Very helpful. Yeah. It's worth noting just quickly that even though it makes sense to start with Parkinson's disease with the genetic mutations in GBA, there are biochemical data suggesting that elevating GCase will be beneficial also for the sporadic form of the disease. Parkinson's patients, even without the mutations that have low level of GCase or have high level of the toxic lipids that accumulate if you don't have GCase, have much faster disease progression. So we think that even sporadic Parkinson's patients will eventually benefit from this drug and the same for sporadic Lewy body dementia patients. So we think that there are already, I mean, it's over, I think, 100,000 patients with just in the U.S., Parkinson's patients with the GBA mutations, but we could go beyond that to the sporadic version. Okay. And then last question I have on the ABC platform, and then we can turn to 101. You've got a few siRNA ABCs as well against fairly well-known targets like Tau and alpha-synuclein for Alzheimer's and Parkinson's, respectively. I guess, why siRNA versus an antibody? And is this hedging for the other two programs that we talked about, or could they actually be used synergistically as an end goal? Yes. Sort of for both Tau and alpha-synuclein, there is sort of good rationale for siRNA. The pathology of both Tau and alpha-synuclein is largely intracellular. Basically, the Tau aggregates are intracellular and the alpha-synuclein aggregates are intracellular. So antibodies do not have access to the intracellular aggregates. The idea for antibodies is that maybe they'll capture the pathological versions of Tau and alpha-synuclein when they sort of spread from one cell to another. If they go to an extracellular phase, it's not clear that that's really happening. It could be that even the spreading has happened through vesicles that are protected from antibodies. So it's not at all clear whether antibodies that only capture the extracellular versions of Tau and alpha-synuclein will be effective. So far, there were several anti-Tau antibodies that did fail. It could be that they just didn't use the right epitope, but it's very possible that siRNA or nucleic acid in general will be more effective for these indications, and again, linking siRNA for Tau or alpha-synuclein with our Blood-Brain Barrier technologies will enable peripheral delivery, will hopefully ultimately enable subcutaneous delivery, and that will really transform the accessibility to patients. Currently, the ASOs that are being tested for Tau, for example, are having to be injected intrathecally. I think it's hard to scale up intrathecal injection to millions of Alzheimer's patients. The distribution in the brain is not homogeneous when you do intrathecal injection compared to peripheral injections, so both with regard to safety, convenience, and efficacy, linking siRNA to Tau or alpha-synuclein with Blood-Brain Barrier technology will be significantly superior, and I think siRNA will be superior to antibodies for these targets. For Aβ, it's not the same because Aβ is largely extracellular, so you can access it with antibodies, but these two targets are intracellular, so siRNA makes more sense. Okay. Okay. That makes sense. Maybe we can shift gears and talk about AL101. Enrollment completed in this study in April of this year. I think you guys have guided to a first half readout next year. Maybe taking a step back, you have experience developing and running a large pivotal study for Alzheimer's. You have experience developing a drug for a different indication, but same target, the PGRN target. I guess, what are sort of the learnings that are being applied to the 101 program? And then we can get sort of into the design. The AL101 program is run by our partner, GSK, and we were able to leverage some of the learnings from the study that we did in Alzheimer's with AL002. The trial completed the enrollment ahead of schedule. We knew where to go in terms of countries, in terms of size. We already had experience on how to make the implementation of amyloid beta and Tau PETs easier to implement in the context of a global phase II study. From an operational standpoint, I think it was important to have this experience before. This enabled a relatively quick enrollment. There is also quality aspects that we learned. I mean, we implemented very careful oversight of the quality of the clinical endpoints, and this has been clearly also done here in the phase II study with AL101, the phase II study, which is currently ongoing. Other learnings are not directly related to our experience, but I think one of the pivotal moments for AD is when the data started coming out showing that a trial duration of one and a half years, 76 or 78 weeks, is enough to see an effect on key biomarkers of disease pathophysiology in AD. Even though these learnings were from Aβ molecules, and here we are testing AL101, which is not an amyloid removal drug. However, I think some of these things related to study design and duration are highly transferable. Of course, there is all the progress made with biomarkers, fluid biomarkers, all the p-tau species, p-tau 217, 181, that we understand more and more about the relevance, the relationship with amyloid PET signal and Tau PET, and we are better able to interpret what effects on those markers mean. I think there were a lot of lessons that we implemented, lessons learned that we implemented in AL101 phase II study, and we're looking forward to the results of the interim analysis in the first half of 2026. Yeah. Yeah. So were we. And I guess maybe we can walk through sort of the primaries and the secondaries. I think CDR-SB is being used again. This is what you've used in prior studies too. It's, I guess, a fairly well-regarded endpoint, maybe. Is there a particular time point that you think is maybe the most impactful for understanding the potential for the program on CDR-SB specifically? And then, I guess, which of the secondary endpoints are you most focused on sort of for the potential in a phase III study? Yeah, sure. I think previous studies have shown that for drugs that work in AD, there are initial effects that may be apparent as early as after six months, but then the drug-placebo separation becomes more apparent at later time points, namely 12 months and 18 months. The study duration of AL101 is 18 months. So when we are looking forward to see the effects of the drug at those time points that I mentioned. CDR-SB is the primary outcome measure. We also have other composite endpoints that are used in this trial, such as the iADRS, for example, and the outcomes, and we're going to look at the effects on all those measures, knowing that all these CDR-SB and the other secondary endpoints that I mentioned are not independent. They show some high correlation. So if you think about the results with Lecanemab and Donanemab, they show somehow similar effect on these different endpoints. Besides clinical outcome measures, I also would like to draw your attention to the biomarkers that we implement in the study. We have Amyloid PET, we have Tau PET, we have fluid biomarker, the p-Tau species that I mentioned. So those biomarkers will be important besides the clinical endpoints to give us a more precise idea about the overall effect of the drug on the clinical progression of Alzheimer's disease and the effect on the biology of the disease through the biomarkers. Okay. That's super helpful. I know we've got a couple of minutes left here, and I didn't want to finish without giving Neil a chance to talk about capital allocation. You guys have to recoup with about $300 million in cash, I guess. How do you sort of see your current clinical programs and also the ABC studies being supported by this, and where does this sort of get you in terms of late-stage trials? I think you might be on mute or double muted? So yeah, maybe I'll start answering that. I mean, so. Okay. I I think it may be working now. No, no. Is it working now? Yes. Yeah. Sorry. You're coming through. Okay. Somehow my mic muted. Sorry about that. Yeah. No, we do have runway through 2027 as guided, and we feel we're very well capitalized to advance multiple programs from our ABC platform as well as complete and execute on the 101 trial. We believe we will be able to achieve multiple value-creating milestones within the runway, including moving into patients and getting patient data with 137 as a key example, and being able to move other programs to IND. So we feel good about our position. We have the runway to execute, and we believe that we have multiple shots on goal. Okay. Very good. Well, I think with that, we're at time, so we'll probably have to leave it there, but really appreciate the great overview of all the exciting things going on at Alector, and thanks for the team participating in our conference this year. Really appreciate it. Thank you so much for the opportunity. Thank you. Thank you. Thank you. Thank you.