Alector, Inc. (ALEC) Earnings Call Transcript & Summary

Good morning, ladies and gentlemen, and welcome to Alector's conference call and webcast highlighting AL002, a novel humanized monoclonal antibody targeting TREM2 for the treatment of Alzheimer's disease. As a reminder, this conference call is being recorded. [Operator Instructions] I would now like to turn the call over to Katie Hogan, Senior Director of Corporate Communications and Investor Relations. Please go ahead.

Thank you, Operator. Hello, everyone, and welcome to our TREM2 event. Before we begin, I will go over just a few reminders. There will be a moderated question-and-answer session following prepared remarks. To submit a written question, please type it into the question and answer panel on the webcast. The webcast replay will be available tomorrow after 1:30 p.m. Eastern Time in the Investor section under Events and Presentations on our website, www.alector.com. I'd like to note that during this presentation we'll be making a number of forward-looking statements, and you can find our disclosure here. Turning now to the agenda. Our Chief Executive Officer, Dr. Arnon Rosenthal, will start by sharing our perspectives on immuno-neurology and TREM2. Next, Dr. Michael Heneka, a distinguished neurodegenerative disease expert, will shed light on why TREM2 holds immense promise in the context of Alzheimer's disease. Following Dr. Heneka's remarks, our Chief Medical Officer, Dr. Gary Romano, will provide a comprehensive overview of our ongoing clinical development efforts centered around AL002, our novel humanized monoclonal antibody candidate targeting TREM2 for the treatment of Alzheimer's disease. Finally, our esteemed guest and Alzheimer's disease expert, Dr. Reisa Sperling, will delve into the landscape of Alzheimer's disease treatment, exploring horizons beyond anti-amyloid beta therapies. With that, I would now like to turn it over to our Chief Executive Officer, Dr. Arnon Rosenthal, to discuss our perspectives on immune-neurology and TREM2. Arnon?

Thank you, Katie, and good day and welcome everyone to our Science Day on immuno-neurology and TREM2. As you hopefully know by now, Alector is completely focused on an alternative therapeutic strategy for dementia in older generation. Instead of targeting the misfolded proteins that typify each neurodegenerative disease, Alector is targeting the brain own immune system as a therapeutic tool to counteract multiple disease pathologies. The scientific rationale for this approach is actually over 120 years old. When Alois Alzheimer first described Alzheimer's disease, he in fact described 3 pathologies, the famous A-beta plaques, the famous tau tangles and the somewhat less famous dysfunctional microglia. Only 2 of these 3 pathologies were targeted as therapeutics for the last 3 decades. And until 10 to 12 years ago when Alector was founded, the third therapeutic target, dysfunctional microglia, were largely ignored. In addition to the original description of dysfunctional glia as a pathology -- as a seminal pathology in Alzheimer's disease, there is more recent human genetics that argue that the microglia plays a seminal role in the disease. And today we know of approximately 100 risk genes for Alzheimer's disease. And the surprising finding was that although Alzheimer's disease is a disease where nerve cells are dying and connections between nerve cells are being destroyed, the majority or a large number of the risk genes for the disease are actually genes that are specifically expressed in the microglia brain immune system and regulate survival, proliferation, migration and multiple aspects of functionality of this immune cell type. So the human genetics really tells us that the immune system in the brain or dysfunctional immune system is a major component of the pathology. In addition to the human genetics, we now know a lot more of the -- about the microglia immune system role in the brain. So in addition to the classical functions of fighting infection, we know that the microglia is the garbage collector of the brain. This is the cell that removes multiple misfolded proteins and, as you will hear from Dr. Heneka, microglia has the tools to remove both extracellular and intracellular misfolded proteins. In addition, the microglia was shown to regulate the oligodendrocytes. These are the cells that form myelin and replacing damaged myelin throughout life and in disease. The microglia is also regulating and replacing damaged synaptic connections, regulate the function of neurons, patching damaged blood vessels and brain tissue and maintains the function of the astrocytes, the largest supposed cell type in the brain that nourish the brain, the neurons and replenish neurotransmitter. I will just show you 2 video to illustrate how active are microglia in maintaining brain health. So one video will show you how microglia, in green here, rushed to a site of injury. And that's marked in X here. This is just a surgical injury in a brain tissue. And the second video will show you similar rushing of microglia to patch injury this time in blood vessels, which is marked in red here. So you'll see the microglia in green really rushed to the site of injury of the brain tissue and really patched the damage. And likewise, if you look at damage in blood vessels, you see the same thing happens, microglia really rushed to the site of injury and patched the damage. So microglia is really surveying every part of the brain every few seconds, sent processes, measure the pH, measure every abnormality and actively respond to it. Despite the fact again that microglia was identified as one of the 3 major pathologies in Alzheimer's disease and the strong human genetics, people avoided targeting microglia as a therapeutic strategy. And one of the reasons for this was that there are a lot of evidence that although microglia can be beneficial, there are many cases that microglia could be damaging or detrimental in the context of Alzheimer's disease. Our understanding is that what happens during disease development is that what's called resting microglia or homeostatic microglia, once disease develops, become beneficial microglia or moderately active microglia. But with time, with aging or maybe as a result of the genetic mutations, the microglia become overwhelmed by the disease. And then the part of their stressed response, they become hyperactive and dysfunctional. So you can see microglia that are active and counteract the disease, but also in some cases as disease progress, you can see microglia that are damaging. So people didn't really know whether to foster microglia activity or kill microglia, and this really slowed down the therapeutic approach sort of targeting microglia. Our understanding today and our hypothesis is that with drugs that target microglia, we can generate many more beneficially active microglia from resting microglia. And this beneficially active microglia will really counteract in multiple disease pathologies. And this is what we aim to do with our TREM2 and other drugs. With this approach of immuno-neurology based on understanding of the microglia biology and human genetics, we developed a portfolio of immuno-neurology drugs. All of these drugs are basically genetic risk -- for targeting genetic risk for different diseases and they are being used as levers to really recruit, restore, repolarize microglia to become effective disease fighting machines. As you will hopefully know by now, we have a Phase III drug targeting frontotemporal dementia with progranulin mutation and we now have 2 Phase II drugs in Alzheimer's disease, the 002 which targets TREM2, which is the focus of this Science Day; and AL101, which is progranulin elevating drug which we just started in Phase II with. In addition, we have a broad portfolio of targets that we will discuss in future science days. So one of the most prominent genetic risks for Alzheimer's disease and a major regulator of microglia is a single transmembrane receptor called TREM2. TREM2 is a damage sensing receptor. It can sense damage in nerve cells membranes. It sends misfolded proteins. And once it senses the damage, it really recruits microglia to the site of injury and repolarize microglia to become a better disease fighting machine. And consistent with the seminal role of TREM2 in microglia function, they've been, I think, over 40 TREM2 mutations been identified. These are all loss of function mutations -- modest loss of function mutations, and they all increase the risk of developing Alzheimer's disease. So mechanistically what you see if you don't have functional TREM2 -- if you, for example, look at A-beta plaques, you see that without a functional TREM2, the A-Beta plaques are diffused, they leach toxic beta amyloid aggregates. And if you look at the nerve cells around the plaques, that are stained in green here, you see that the nerve cells are damaged as a result of the toxic beta amyloid aggregates. If you have functional TREM2, you'll see that the beta amyloid plaques are compact. You can see it in 2 types of staining here. And you can now detect damaged nerve cells around it. The reason for the difference in the nature and the pathology of the plaques is that in the absence of TREM2, the microglia are either dying or cannot detect the beta amyloid plaques. So here on the bottom, you see beta amyloid plaques in green that are much larger than with in the presence of TREM2. And you'll see that the microglia in red do not encompass the beta amyloid plaques and do not contain it. In contrast, if you have normal complement of TREM2, you'll see the microglia in red completely overwhelming the plaques and protecting the surrounding neurons from it. So this is in a mouse model. You see the same thing in human. This is Alzheimer's brain. A-beta plaques in purple surrounded by green microglia that have normal complement of TREM2. Microglia that carry one of the genetic mutations in TREM2 that reduce TREM2 functionality are not able to effectively contain A-beta plaques. And the outcome again is damaged neurons around the plaques that are stained in yellow here. In addition to the loss of functions, there are evidence that gain of functions of TREM2 is protective. So where as loss of function is detrimental for Alzheimer's disease, a gain of function appear to be protective for multiple aspects of disease protection -- of disease progression. So if you measure the level of TREM2 by soluble TREM2 in the CSF, you see that high level of soluble TREM2 is associated with a slower brain tissue loss in Alzheimer's disease compared to low level of soluble TREM2. Likewise high level of soluble TREM2 is associated with slower cognitive decline in Alzheimer's disease, with slower conversion for mild-cognitive inhibition to Alzheimer's disease, with improved survival with Alzheimer's disease, with delayed age of onset and with delayed accumulation of A-beta and tau in Alzheimer's disease. So there are multiple loss of function and gain of function, human genetics data arguing that decreased level in activity of TREM2 is detrimental for Alzheimer's disease, whereas high level is protective. So based on this information, we developed a drug that would really mimic the gain of function of increased level of TREM2. So we developed a drug that really activates the TREM2 receptor. And we made a lot of demands from this drug. We screened over 1,000 antibody drugs to really identify the optimal TREM2 activator. And what we required from the TREM2 activator is as follows. First, we required that it will activate the TREM2 receptor without disrupting the interaction between TREM2 and its natural ligands. We hypothesized that the natural ligands are required to bring microglia to the site of injury. And if you develop antibodies that blind the ligand receptor interaction, you will activate microglia, but the microglia will not know where to go and they will basically not identify the site of the crime. They will just be activated in a random location. And indeed, we have animal model data showing that antibodies that block the ligand receptor interactions are not effective in animal models even though they activate TREM2 signaling. And you see here that we indeed were able to identify such an antibody. You see here, for example, binding of ApoE4 -- lipidated ApoE4 to TREM2. In the absence of our drug, the binding is very modest. In the presence of our drug, there is significant additive effect. Likewise, another ligand for TREM2, phospholipids, which is -- which represents damage membranes, you see without our drug, there is dose-dependent binding, but with our drug, there is significantly enhanced binding. Once we identified a drug that enhances ligand binding, we wanted the drug to activate TREM2 signaling. And indeed, we see that our 002 drug very potently in a dose-dependent manner activate TREM2 signaling as depicted by the TYROBP phosphorylation of the TREM2 coreceptor. Our drug also promotes TREM2 dependent gene expression. It promotes myeloid cell viability, which is proliferation, which are among the critical functions of TREM2. And it also enhanced microglia functionality. This depicted here by elevated level of critical lysosomal enzyme, GKs. So once we had identified a drug that really had all the desired activities in cell culture, we went in vivo. We tested the drug in an animal model for Alzheimer's disease. And as predicted by the human genetics that TREM2 is involved in microglia dependent compaction of beta amyloid, we saw that our drug can facilitate A-beta compaction. You see, again, in this animal model of Alzheimer's disease, you see filamentous toxic A-beta plaques. And in the presence of our drug, these plaques become compact. And as part of these compaction, the plaques become inert. So again here you see A-beta plaques in purple, damage neurons around the plaques in yellow. And if you treat this animal model with our drug, the number of damaged neurons is significantly reduced in conjunction with the compaction of the A-beta plaque. We tested our drug in another model. This is a model of myelin damage. So in this model, you damage the myelin with a chemical toxin and you look at the rate of myelin regeneration. And the first step in myelin regeneration and essential step is removal of the damaged myelin. And you see again, without our drug, the damage myelin resides in the brain for a long time. In the presence of our drug, the damaged myelin is removed much faster. In addition, our drug can facilitate the proliferation of the oligodendrocyte precursor cells, the OPCs. These are the cells that form myelin, and it also induces the maturation of these OPCs to myelin forming cells. So through its action on microglia, our drug is able to facilitate the removal of damaged myelin and facilitate the synthesis of new myelin. As you know, myelin is very famous in the context of multiple sclerosis. But in fact, myelin impairment is a key role -- plays a key component in Alzheimer's disease. There are multiple pieces of data suggesting that abnormalities in myelin contributes to Alzheimer's disease. And we think that this just demonstrates one of the many beneficial activities that AL002 elicit. It will facilitate, again, among -- in addition to removing misfolded proteins, possibly replacing damaged synaptic connections, it will also facilitate the removal and regeneration of damaged myelin, which I think will contribute significantly to the clinical benefit. Since there are at least 2 approved anti A-beta antibody drugs now for Alzheimer's disease, one of the questions that repeatedly arises is, how an immuno-neurology drug or a TREM2 activating drug can fit in, in this field where there are approved anti A-beta drugs? So first, it's worth noting that the anti A-beta drugs have only a modest effect, and we hope that our TREM2 drug, which has a broad activity we have stronger clinical benefit as a stand-alone. But if you consider combination, the only thing that the anti A-beta antibodies do is that they mark the site of the A-beta plaques. They then need to recruit the excavator. The excavators are the microglia. So we have a drug that increased the number of the excavator sales, increase their potency interest, their ability to phagocytosis beta amyloid. So there is a really natural combination between drugs that mark the side of pathology, the side of the A-beta plaques and a drug that enhance the excavators. So we see a lot of potential for combination therapy. And one evidence for this is work that was done by Christian Haass several years ago. Christian Haass measured the efficacy of anti A-beta antibody in microglia or in vivo in cells that either have or are deficient in TREM2. So he measured first the microglia ability to phagocytosis A-beta plaques. Again, this is all in the presence of A-beta antibody. And you see if the microglia don't have TREM2, in the gray bar here, their ability to forego A-beta in anti A-beta antibody dependent manner is significantly reduced in vitro. And likewise, if you look in vivo the plaque density, if you have normal complement of TREM2, the plaque density is significantly lower compared to if you don't have A-beta -- if you don't have TREM2. So there is really sort of clear evidence that TREM2 is really required for effective microglia to facilitate the removal of A-beta amyloid by anti A-beta antibody. And again, supporting the possibility of combination therapy. With all these wells of information, we took the drug to the clinic. And we -- and we went to non-human primate studies. And in nonhuman primates, we saw, again, similar things to what we see in rodents as Gary will tell you, we've seen in human, we saw dose-dependent target engagement following peripheral injection. This is soluble TREM2. You see the level without our drug and with different doses of our drug, we see saturable dose-dependent effect. We saw elevation of multiple critical biomarkers for microglia. Here you see IL-1Ra, this is an anti-inflammatory secreted protein, which is elevated in a dose-dependent manner. Here you see elevation of soluble CSF-1R, CSF-1R is a critical survival and proliferation receptor for microglia. So with some multiple target engagement and what we consider beneficial effects on microglia. And with this -- and with this data, we took the drug to the clinic. As again, you likely know, this program is part of an option agreement with AbbVie, where at the time when the auction -- again, it was executed, we received over $200 million when the auction agreement will be realized, this will be up to AbbVie to decide likely in Q1 2025, we are expected to receive $250 million opt-in payment, additional milestones, and this is a 50-50 profit share agreement worldwide. And because we were by far the first company to develop TREM2 antibodies, as far as we know, we are the only clinical program with TREM2 activating drug in Alzheimer's disease in the clinic. We have a very broad portfolio of patent protection. Just to recap what I've told you, we are developing an alternative therapeutic strategy for dementia and neurodegeneration, which we call immuno-neurology, where we recruit the microglia brand immune system to counteract disease pathology. And conceptually, this approach is similar to immuno-oncology, and we hope that immuno-neurology will have as much impact on neurodegeneration as immuno-oncology is having on cancer therapy. We are expecting and hoping for superior efficacy compared to anti A-beta drugs because we have broad effects on disease pathologies. As I showed you, we think that there will be natural possibility for combination with anti A-beta therapy. We think that there will be clinical benefit at multiple disease stages. This again, something that will have to be tested. But for example, high level of TREM2 several TREM2 as I described to you affect multiple steps of disease progression. So hopefully, our drug will produce that. And because our drug have multiple effects, which are completely independent of A-beta, we think that even if there is a modest effect on A-beta, we do expect significant clinical benefit, for example, to effect on tau, to effect on synaptic connections, effect on the other support sales, effect on blood vessels. So we really have high expectations from this drug. So back to you now, Katie, to introduce Dr. Heneka.

Thank you, Arnon. With that background, I am honored to introduce Dr. Michael Heneka, as a neuroscientist, an expert in the field of neurodegenerative diseases, Dr. Heneka has made significant contributions to understanding the underlying mechanisms of neuroinflammation and its role in neurodegenerative disorders, including Alzheimer's disease. Dr. Heneka currently serves as a member of the Alector Scientific Advisory Board. He is also a Director of the Luxembourg Center for Systems Biomedicine, or LCSB, where his research focuses on the intricate relationship between brain inflammation, the aging process and neurodegeneration. Additionally, Dr. Heneka is an adjunct professor at the University of Massachusetts Chan Medical School. At this time, I'll turn it over to Dr. Heneka to delve further into TREM2. Dr. Heneka?

Thanks, Katie, for this very kind introduction and the opportunity to discuss with you why TREM2 is a promising target in Alzheimer's disease. So as we all know, our societies around the globe will experience a dramatic increase in the numbers of neurodegenerative diseases, and most importantly, of Alzheimer's disease. Around the time Alois Alzheimer has studied his one and only patient, Auguste Deter, in 1910, the German demographical schematic looked like the left side here. In 2015 -- this is our projection. So we will have a dramatic increase in the number of over 65-year old patients, where the disease starts to rise in numbers. Now all these diseases share pathological protein aggregation. In the case of Alzheimer's disease, we find extracellular amyloid beta deposits, as mentioned already by Arnon, and intra neuronal formation of neurofibrillary tangles made of hyper phosphorylated tau. And there are cases these are being used by genetic mutations, autosomal dominantly inherited in APP/PS1, monomer PS2 on the case of tauopathies in the MAP2 gene. These diseases share another feature. They start in one part of the brains right within the brain area and then into other brain regions. And this explains why these diseases have a very long clinically silent prodromal stage. We know from Alzheimer's disease that amyloid beta deposition may start as early as 3 decades prior to the first cognitive deficits. We know for Parkinson's disease that REM sleep disturbance or anosmia starts 20 years prior to the first clinical symptoms. Now the problem is that this prodromal stage may hold very important pathological mechanisms which we hardly know. But we have learned by genome-wide association studies in the past decade some of the pathomechanisms which may have an important impact during these preclinical silent times. And as mentioned by Arnon already, we know around 100 risk variants that increase the likelihood of developing Alzheimer's disease of late onset nature. And these can be classified in roughly 4 different areas: genes that are important for lipid metabolism, same structure and function, lysosomal function and more than 50% of them are important for innate or adaptive immune mechanisms. Now that doesn't tell us yet at which time in which cell these risk variants play their pathogenic role. And this is work by [ Kosa ] and colleagues published last August in Nature Genetics, which nicely combined ATAC sequencing, [ Jivos ] analysis in single-cell RNA sequencing to delineate in which cell type these risk variants are actually playing an active disease-driving role. And it's interesting to see that in Parkinson's disease, for example, there is an important role for risk variants in both in neurons as well as in microglia. In ALS, there is a pivotal role of neuronal cells. The risk variance of ALS seem to primarily play a role in neurons. While in Alzheimer's disease, it's exactly the opposite. So the risk variance seem not to play any role in neurons, but very much so in microglia cells, suggesting that microglia cells and their activation have an important disease driving pathogenic role. Now the question is, why we have an immune reaction in the brain at all? This shows you a Congo red stain from the cortical section from an AD case and MHC Class II positive microglia cells. And long before Congo red was actually being used to stain human brains, it has been developed as something completely different and licensed as a bacterial stain in 1880 by [ Paul Batika ] in Leverkusen in Germany. And it took 100 years now to find out why this is staining bacteria and human brains. And Matthew Chapman then found that bacteria very common to us, including Escherichia coli, Salmonella Typhimurium and Staph aureus carry beta sheet structured amyloids on their surface, very similar to those that are depositing in our brain. That means that myeloid cells, including microglia cells by nature through evolution have been taught for millions of years that beta sheet structured amyloid is nothing else than a pathogen. And nature has equipped myeloid cells with pattern recognition receptors, including the toll-like receptor family and other surface receptors to sense these targets, these amyloids. So upon the ligation of these receptors, the microglia cells undergo fundamental change. They withdraw their processes, they get around, they start an inflammatory reaction characterized by the release of complement factor, cytokines, chemokines, radical oxygen species. At the same time, they stop doing a lot of sensible things for our brain. They reduce synaptic scaling and pruning. They reduce the generation and the release of neurotrophic factors and they have decreased depress clearance. Now immune activated microglia cells remove synapse in excess through complementary. Normally, they contribute to the physiological synaptic turnover. In the case of stimulation by oligomeric amyloid beta, there is an excess uptake of complement 3D CORT synapses, which leads to a reduced synaptic plasticity and functional deficit. Now microglia cells also can support neurons by freeing them from pathological aggregates through export via tunneling nanotubes. And here, you see 2 examples, time-dependent export from alpha-Synuclein aggregates or tau aggregates from neurons through tunneling nanotubes. At the same time, they use these tunneling nanotubes to deliver fresh and functional mitochondria to the neurons. You see the decrease of the mitochondria within microglia and at the same time, the increase in the neurons. And these mitochondria, which are freshly delivered by microglia cells are incorporated and built into the mitochondria network. Now this action, the transfer from pathological aggregates as well as the fresh delivery of functional mitochondria ensures the function and survival of the neurons. Now, this is an example by PET imaging of amyloid beta deposition, inflammation and tau accumulation in the presymptomatic MCI and AD stage. You can see here that very early with amyloid beta deposition, inflammation becomes visible in the -- by this PET ligand. And in the MCI stage, we see for the first time tau, which then stays. And all 3 pathologies can be visualized in the case of full dementia. Now this is important because it suggests that inflammation is a potential driver of tau pathology. And this is exactly what we and others have shown in several seminal papers. So inflammation seems to be a key driver of tau pathology and neurofibrillary tangle formation. Now TREM2 has a complex signaling cascade and function via activation of GSK 3 beta. It drives proliferation of microglia cells. mTOR activation is important for maintaining microglia fitness and microglia survival. DAP10 and SYK are required for microglia response to amyloid beta, and importantly, for the restriction of A-beta pathology. And on the right side, you see a knockout experiment, set of phagocytic amyloid beta uptake. You see that SYK or TREM or DAP12 deficient cells are less capable of removing amyloid beta. Now TREM2 overexpression enhances [indiscernible] receptor expression, enhances the capability to migrate and phagocytose. The attenuation of endogenous TREM2 leads to the reduced phagocytic clearance of apoptotic cells and increased inflammatory status and reduced engulfment of A-beta plaques and it facilitates the spreading and neurotoxicity of beta amyloid. Now below, you see an image where you see what's already introduced by Arnon. There is a reduced presence of microglia cells around the amyloid beta deposit in blue in the TREM2 knockout or TREM2 T66 M mutant. And the green dots actually indicate the result of [Audio Gap] release. So this is a ferroptotic death of microglia through hyperinflammation. So these cells die on the spot if they don't have an intact TREM2 system. Now endogenous TREM2 ligands include ApoE depending on lipidation status, [ ApuJ ], myelin, apoptotic neurons and A-beta itself, as can see below in the schematic. Importantly, the R47H mutant or the R62H mutant have reduced binding to amyloid beta, so possibly indicating why these mutants are less good in clearing amyloid beta from the brain. Now TREM2 insufficiency as a knockout or loss of function variant impairs metabolic fitness of microglia cells. And here you can see that overall TREM2 knockout results in very much reduced ATP levels. There is a reduction of glycolysis, as indicated by reduced ECAR. And there's also a reduced oxygen consumption, right. So there is a reduction of anabolic and energetic metabolism in microglia, very much compromising their action. This work by [ Tyler Owen ] showed that, again, in a mouse model of Alzheimer's disease, the 5xFAD mouse, the knockout of TREM2 results in a massive increase of dead microglia. In this case, the death signal is cleaved Caspase 3. And you see, there is a little death in the 5xFAD-mouse. But when TREM2 is missing, there's a massive increase of microglia death. So interestingly, and I told you that microglia cells can help neurons by exporting pathological protein aggregates via the tunneling nano tube system. In the case of the TREM2 66M or the R47H mutant, this function is massively impaired. They still are capable of delivering mitochondria, but overall there is no change in the levels of radical oxygen species or survival. So TREM2 mutants are less capable of supporting microglia cells by exporting pathological aggregates via tunnelling nanotubes. Now interestingly, soluble TREM, which is a consequence of ligation and sharing is a good biomarker, as we heard before, and increases around the MCI stage in the Alzheimer's disease continuum. So it correlates very well with phospho-tau in the CSF, but less so with amyloid beta 1-42, indicating that it is somewhere in between the early amyloid beta deposition and the late conversion into dementia. The formation of the tau neurofibrils seem to be a consequence of immune activation. And this is also reflected in this analysis. This is the DELCODE study. DELCODE is a study led by the German Center for Neurodegenerative Disease. It's a longitudinal study. And here, you can see when we use the ATN criteria, soluble TREM2 in the CSF become significantly increased in a stage where amyloid beta reduction in the CSF and the tau positive signal has to be present in a late stage. And it's just about depending on the presence of amyloid beta alone. So the pre-stage, if you look below, there is no signal in subjective cognitively impaired or in MCI cases, but in AD cases. So it seems to be around the corner between MCI and conversion into full dementia, A very important moment where drug intervention is urgently needed. And these data are also reflected by more recent pathological analysis that showed that there is a dramatic increase of TREM2 mRNA in AD patient brains in dementia cases, but not yet in MCI. So again, this is a data set from a PET study, which shows that in cognitively unimpaired patients, the soluble TREM indices F is correlating with low Braak stages, while in the MCI cases, there is correlation with higher Braak stages, together suggesting an active role for TREM2 during this stage of disease. Now TREM2 and soluble TREM plays a role at the synapse. There's a positive beneficial effect because microglia removal of amyloid beta induced hyperactive synapses normalized plaque associated in our own hyperactivity. TREM2 binds to C1q and thereby inhibits the classical component activation mediated by -- ultimately by C3, as I showed you before. Over expression of the AR47H TREM2 variant in BB2 cells increased microglia synapse uptake and neuronal loss, and TREM2 fling then supplies variant derived soluble TREM suppresses hippocampal long-term potentiation. Now TREM2 has been shown to interact with tau. Lentiviral induced upregulation of TREM2 protein levels reduced tau hypophosphylation in PPP transgenic mice. The TREM2 R47H loss of function risk variant carrying mice show more neurotic plaque pathology upon AD tau injection as compared to wild type. In general, and summarizing a number of manuscripts TREM2 knockout or AD associated TREM2 risk variance increased tau pathology, tau seating and spreading and neuronal injury in mixed amyloid tau models. In primary tauopathy models, microgliosis and neurodegeneration was reduced. But in primary tauopathy models on an APoE background, there was exacerbated neurodegeneration tauopathy and a decreased then profile. Now TREM2 antibodies are being developed, as you heard, and they have been shown to act on microglia to drive amyloid beta clearance. As shown here, the cortical plaque area is decreased by this antibody, which is called 49. In summary, I showed you that genetic findings link the TREM2 gene variants to an increased risk for neurodegeneration, most prominently Alzheimer's disease. TREM2 function includes removal of cellular debris, restriction of A-beta deposition, sinus protection and maintenance of microglial homeostasis. TREM2 risk variance result in a loss of function, facilitated microglial death, and consequently, an increased disease risk. Soluble TREM to detection in CSF and human [Audio Gap] pathology suggests an activation in the MCI to early AD stage, a key moment in the disease development where intervention is urgently needed. Modulation of TREM2 has proven beneficial in preclinical models and TREM2 antibody ligation represents an efficient way to modulate the receptor. And with this, back to you, Katie.

Thank you, Dr. Heneka. At this time, I'll turn it over to our Chief Medical Officer, Dr. Gary Romano to discuss our clinical development efforts around AL002. Gary?

Thank you, Katie. I'll now walk you through the clinical development of AL002 Alector's first-in-class TREM2 agonistic monoclonal antibody. So it started with a Phase I study of AL002 in healthy volunteers, which is completed in 2020. Single doses ranging from 0.003 to 60 milligrams were assessed in biomarkers were collected to assess target engagement and effects on microglial signaling. And this study had 85 days of safety follow-up. AL002 is well tolerated in the single ascending dose study. There were no drug-related serious adverse events or dose-limiting AEs occurring in the study. A few participants had adverse events that were considered probably related to AL002, we need to withdraw from study drug. One of these had nausea and one of these subjects had nausea and paresthesias. Treatment with single doses of AL002 resulted in dose-dependent target engagement and increased microglial signaling. In the left panel here, you see dose-dependent reductions in soluble TREM2, CSF-soluble TREM2, which reflect AL002 binding and internalization of membrane TREM2. On the right, you see dose-dependent increases in CSF1R. This is a receptor tyrosine kinase that's essential for microglial development and survival. We also saw treatment-related increases in CSF levels of IL-1 RN and SPP1, providing further evidence that AL002 treatment increased TREM2 signaling and activation of microglia. The INVOKE-2 trial is Alector's Phase IIb study of AL002, which is now ongoing in patients with early Alzheimer's disease. This is a randomized, double-blind, placebo-controlled, common close design study of up to 96 weeks of treatment with AL002, in which 381 participants with early Alzheimer's disease were randomized. It includes 3 doses of AL002 that were demonstrated in Phase I to increase microglial signaling. The study is now fully enrolled, and it remains blinded to treatment assignment. We expect to have last patient out in third quarter of 2024, with a study reading out thereafter. Our long-term extension will remain blinded to original treatment assignment, and thus, will provide additional information on long-term safety and treatment effects on biomarkers and clinical outcome assessments. Treatment effects on clinical progression of Alzheimer's disease will be assessed with cognitive and functional assessments. The primary clinical outcome measure is the CDR Sum of Boxes, which, as you know, is the primary endpoint in the recent lecanemab Clarity trial. Secondary clinical and functional outcome measures are listed here, including the ADAS-Cog13 and the ADCS-ADL-MCI, from which we'll derive treatment effects on the iADRS, which was the primary endpoint of the recent donanemab Phase III trial. In our analysis plan, we will use a proportional analysis approach, which will enable us to use all of the data that we collect in this common closed design trial, meaning that we will include data from all participants out to the minimum of 48 weeks, but also include data provided by longer follow-up from those participants who are in the study up to 96 weeks. This trial will also deliver a robust biomarker package, reflecting target engagement as well as treatment effects on microglia activity and on Alzheimer's pathophysiology. Target engagement will be assessed by measuring treatment effects, again, on CSF levels of soluble TREM2. Treatment effects on microglia signaling will be assessed by measuring treatment effects on the levels of CSF1R, SPP1, and IL1RN, which reflect proliferation, survival and phagocytotic activity of microglia. We'll also have our planning exploratory proteomic and transcriptomic assessments of treatment-related changes in microglia subtypes. Treatment effects on Alzheimer's pathophysiology will be assessed with CSF and plasma biomarkers of A-beta and tau, which we're very excited about, as well as both amyloid and tau PET, and we'll also have biomarkers of astrogliosis, neuroinflammation, synaptic health, and neurodegeneration. So as we reported earlier this year at AAIC, a subset of participants in the ongoing INVOKE-2 trial have had treatment-emergent MRI findings that resemble the amyloid-related imaging abnormalities, or ARIA, that has been observed with anti-amyloid antibodies with regard to their MRI features, the incidence, the timing of onset and resolution, and their relatedness to the number of ApoE E4 alleles, as well as the frequency and spectrum of associated clinical manifestations. On the right of this slide, you see an example of MRI images from a 77-year-old female trial participant in which surveillance MRI detected a new zone of treatment-emergent vasogenic edema within the left occipital lobe. The follow-up exam 1 later, the focus of VE was decreased in size and extent when compared to the prior exam. A later surveillance MRI about 4 months later showed complete interval resolution of previously noted VE. However, there was interval development of a single punctate focus of hemosiderin deposition in the area of the resolved VE consistent with a microhemorrhage. As for those of you who are familiar, this is very similar to what we're seeing with the ARIA of anti-amyloid antibodies. Early in the trial, we observed that the most severe clinical and radiographic findings were observed exclusively in ApoE E4 homozygotes, which led Alector to voluntarily discontinue and exclude further enrollment of ApoE E4 homozygotes, continuing the study with the ApoE E4 heterozygotes and non-carriers. And an independent data monitoring committee reviews data from this trial regularly and continues to recommend that the trial proceed. The data on this in the next slide is from our most recent data cut in September. For convenience, I'm going to refer to the MRI changes that we're observing with AL002 as ARIA-E and ARIA-H, although I want to emphasize that given that this is a different mechanism of action, we don't yet know whether the biological mechanism causing these changes is the same as for the ARIA that's been described with the anti-amyloid antibodies. This table compares the incidence and severity of ARIA-E and ARIA-H between ApoE E4 homozygotes and the current study population of so-called non-ApoE E4 homozygotes, which includes, again, the heterozygote carriers and non-carriers. You can see that the overall incidence of ARIA was greatly reduced from 71% to 19% for the ARIA-E and from 71% to 23% for ARIA-H. You can also see in the lower panel that exclusion of ApoE E4 homozygotes resulted in a reduction in the percentage of those with severe ARIA-H. I just want to also note that this study is still blinded to treatment assignment, and so these percentages I'm showing you were calculated conservatively with the assumption that all ARIA occurs in the active treatment arms. In the current trial population, the majority of those with radiographic ARIA have been asymptomatic and clinically serious cases have been uncommon. Here you see that of the 49 participants of the current trial population that had ARIA-E, 88% were asymptomatic and 12% were symptomatic. For most, these symptoms were mild and self-limited. Perhaps most relevant is the incidence of clinically serious ARIA, that is those with ARIA-related SAEs, which is just under 1% of all participants that have been dosed. To recap, we've completed a Phase I study in healthy volunteers that demonstrated dose-related target engagement and dose-related treatment effects on microglia signaling. And our Phase II study has completed randomization of nearly 400 participants. We anticipate it will be reading out in Q4 of 2024 with a robust data package. I want to leave you with some thoughts on what we might expect from the INVOKE-2 trial. Well -- so that includes therapeutic restoration of microglia function that may slow disease progression by enhancing the clearance of misfolded proteins, including amyloid, maybe tau, but also, as you've heard earlier, through the enhancement of other beneficial effects of microglia, including maintenance of synaptic connections, supportive astrocyte, oligodendrocyte function, maintenance, repair of the blood-brain barrier and the vasculature, and preservation of immune tolerance. Given that the agonism of TREM2 has the potential to reduce the brain's vulnerability to neurodegenerative disease through these multiple mechanisms, we might find that the treatment benefits manifest differently from what we've seen in the anti-amyloid antibody trials, such as biomarker responses. For example, lowering the cerebral amyloid PET signal to 24-centiloid threshold, which for anti-amyloid antibodies appears to be a necessary condition for clinical efficacy, may not be relevant to this mechanism of action that goes beyond amyloid clearance. Also, with regard to optimal disease stages for intervention that was mentioned earlier, given the multiple mechanisms by which microglia may decrease vulnerability of the brain to pathogens and disease throughout life, the window for effectiveness may be broader than for treatments targeting specific misfolded proteins like amyloid and tau, for which optimal treatment effects may be limited to those stages of disease cascade in which those specific proteins show the most dynamic change. And finally, with regard to temporal dynamics of treatment effects, some improved microglia functions may manifest earlier in treatment, for example, amyloid clearance or possibly maintenance of synaptic function, where others may be apparent later, such as coming from support of astrocyte and oligodendrocyte function or the repair of the vasculature or blood-brain barrier. So we might not fully appreciate the benefits of microglia therapeutics early in treatment. I want to thank you for your attention and give it back to Katie to introduce Dr. Sperling.

Thank you, Gary. At this time, I am honored to introduce Dr. Reisa Sperling. Dr. Sperling is a neurologist who works on the detection and treatment of Alzheimer's disease at the pre-symptomatic stage. She is a professor in neurology at Harvard Medical School and Director of the Center for Alzheimer Research and Treatment at Brigham and Women's Hospital and Massachusetts General Hospital. Dr. Sperling is also the Co-Principal Investigator of the Harvard Aging Brain Study and the National Institute of Health-funded Alzheimer Clinical Trial Consortium. Today, Dr. Sperling will discuss the Alzheimer's disease treatment landscape beyond anti-amyloid beta therapies. Dr. Sperling?

Thank you so much for having me, and this has been a terrific webinar. I am definitely learning a lot and thrilled to be able to talk to you about how this kind of a program might fit into a really exciting time in terms of Alzheimer's treatment. So let me see if this will let me advance. Here we go. So I'm going to really, again, focus on how we bring these insights into what is a rapidly evolving field of treatment, what the challenges and especially what the opportunities are for a mechanism like going after TREM2. So first, here are my disclosures for companies I've consulted to, including Alector, and also the research funding for some of the data I'll show you. So I won't spend any time on this because we started with this, but I will, again, bring this up that what an incredibly important and expensive devastating disease Alzheimer's disease is. And the good news from my point of view is that we can now detect Alzheimer's disease very easily during life and treat it, going after the pathophysiology during life and I hope even sooner, as I'll talk about, even before symptoms. But the bad news, even though I am a big proponent of the anti-amyloid therapies, is that it's clear that with monotherapy at the early symptomatic stage of Alzheimer's disease, we cannot yet fully arrest progression. So you've seen a couple of different versions of this, but I want to come back to it because, again, the studies done to date and where we have approvals for anti-amyloid antibodies are at the stages of MCI and very mild Alzheimer's dementia. I disagree with the FDA referring to this as early Alzheimer's disease, and I've heard from them since that they realize this is a misnomer because the process of Alzheimer's disease has already been in the brain for at least 10, probably 20 years by that stage. So I'll refer to that as early symptomatic Alzheimer's disease, and I think that is not early pathology. And of course, that's important for which mechanisms we might choose at a given stage of disease. So you've seen lots of pathology pictures today. I won't belabor this except to show you how intricately related amyloid and tau are in the brain, and I'll show you in life how early that actually begins. And again, one reason for me that monotherapy with anti-amyloid, likely by the time people are symptomatic, will not fully arrest progression. So here's some additional amyloid and tau PET imaging. And again, by the time people have even mild dementia, on the far right, they have a head full of amyloid and they already have tau pretty substantially distributed throughout the neocortex, again, often Brock Stage 5 and 6, even by the stage of mild dementia. I will disagree very slightly with Dr. Heneka's excellent presentation because we do see evidence of tau spreading beyond the medial temporal lobe in people who are still cognitively unimpaired. It often heralds the onset of MCI within 2 years in longitudinal data. But in the setting of a lot of amyloid people with tau that's spreading beyond the medial temporal lobe are a very important target, in my opinion, to prevent people from progressing even to MCI and mild AD dementia. So what we have now are trials that are really and approved therapies for what I'll refer to as tertiary prevention and treatment. Again, people who already have substantial pathology in their brains, even though they may be at relatively early stages of symptoms. And I do believe we've had -- the reason we've seen success with anti-amyloid antibodies is moving from dementia into MCI and very mild dementia where there's relatively less pathology. But I think ultimately we want to move even earlier. So let me just show you. And I'm sure you're quite familiar with the data from the recent Phase III studies, but I'm going to point out a couple of very specific things that are, I think, relevant to this mechanism and where we might be able to improve beyond the therapeutic efficacy we have now. So this is, of course, Lecanemab Phase III, which was published just over a year ago with very dramatic reductions in amyloid PET that begin as early as 3 months. And you can see a difference between lecanemab and placebo in rates of decline in the CDR global scale here as well as the CDR-SB, Sum of Boxes. And there's a hint anyway that there's some widening as time goes on. And depending on which analysis method you use, again, that there's a slowing of about 27% to 35% on different cognitive and functional scales. On the far right, I'm showing you the effects on tau PET, which I think were important because we saw this with lecanemab, but not with donanemab. So I've had the opportunity to look at this data more in detail. And here we do see that reducing amyloid dramatically does slow the further accumulation of tau and very particularly in these early tau regions in the medial temporal, meta temporal, and the early areas of temporal neocortex. And that, I think, is a key to the clinical signal we see so far. I had the opportunity at CTAD to present some of the first data looking at the open label extension. So now going through 24 months. And this effect size is largely maintained through the 18 to 24 month period. So the good news is that it's still disease modifying. We don't see these curves coming together. The somewhat bad news is we don't see them widening even further. And I think that it may be that as people progress to a later stage of pathology even if they were on lecanemab, that we may have less of an ability to fully impact that curve. So we did look. I also showed data in the low tau PET subgroup, which will be relevant when we talk about donanemab in a moment. But we also looked at a cut of the data in the lower amyloid group, less than 60 centiloids. And here we could see a bigger effect size, about twice the effect size in the rate of slowing in the overall group, again suggesting that earlier might be better. And the very similar data, in my opinion, was shown by donanemab. Here the 2 -- first, the low to medium tau group versus the overall group. And although they're both significant, you can see that there is a larger effect size in the low to medium tau group. And Mark Minton at CTED showed some similar analyses with cut points on amyloid centiloid, showing again their best effect size is driven by the group who has the lowest amyloid and the lowest tau. And very similarly donanemab is able to really dramatically decrease on amyloid PET, the fibrillar forms of amyloid, in a similar, perhaps even faster time frame that we see in lecanemab. So even though, again, I'm thrilled that we are seeing these signals and I do think these medications will be used, I think we want to get beyond 27% to even 40% or 50% that we can see in the earliest group, I want to get to 100%. So how do we do that? So I think one way -- and I won't talk about that today because it's less relevant to this mechanism, but is to go even earlier with anti-amyloid monotherapy. But I will say I see a role for TREM2 probably at the late stage of preclinical, early prodromal, so you could really, again, protect synapses, be able to protect at that early stage of tau spreading. And this is being tested with lecanemab in the AHEAD study, which I had the honor of leading, and with donanemab in the TRAILBLAZER-ALZ3. We are about to see the first combination trials in anti-amyloid and anti-tau. We've been funded by the NIH for a very large platform trial looking at 2 different tau therapeutics, both as monotherapy and in combination with an anti-amyloid antibody. And that trial will cross preclinical and prodromal disease, so Stage 2 and Stage 3 in the new FDA guidance. But ultimately, I think we actually want to go after multiple proteinopathies because I've talked about mostly amyloid and tau, but TDP-43 and alpha-synuclein are also accumulating in older individuals, and the TDP-43 in particular does seem to have a substantial addition to cognitive decline. So there are a couple of different ways to do this. I work with Howard Weiner, an Immunologist at Brigham, who's been trying adjuvants just to [ zoop ] up the system. That may work, but it's not very elegant. And of course, I do worry about its pro-inflammatory effects beyond the microglia activation. So I was really excited when I heard more from Dr. Romano a couple of years ago, actually, about the TREM2-Alector AL002 for its potential, again, to not just clear A-beta, but to have the synaptic protection and go after other proteinopathies that are accumulating earlier than we think. So I want to say a couple of words about amyloid-related imaging abnormalities. I've been working on ARIA since 2010 because I was very involved in the first BAPI studies, and I'll say that ARIA-E and ARIA-H have been observed with all plaque-lowering anti-amyloid antibodies, but not with antibodies that are only going after soluble A-beta that do not reduce fibrillar forms of amyloid. I think we don't fully know the mechanisms, and it's probably more than 1, but actually, I think the Alector data are very informative here. So initially, one of the ideas was that there was a direct removal of the amyloid from the vascular structures, because the N-terminus of A-beta is sticking out there. The timing of ARIA being this very short, rapid after infusion, and now also seen with sub-Q, suggested that perhaps there was actually mechanical removal, and some nice animal data suggesting you can actually restore smooth muscle function in the vascular structures with direct removal of amyloid. The other alternative is that this is really an inflammatory response to perivascular amyloid that's moving from plaque or the vascular amyloid itself since you do see ARIA [indiscernible] in ApoE E4 carriers even in the absence of exogenous antibody administration. These can be inflammatory as in CAA syndromes. And we do know that the rare symptomatic cases of ARIA do -- are associated with more evidence of inflammation in CSF, even sometimes white cell responses. But again, the timing of ARIA-E didn't perfectly fit with inflammation for me, because typically you see it early and it goes away. You still have more removal, sometimes for 6 or 12 months. You sometimes can have this recurrence. And it wasn't really clear to me it was inflammatory. But then I saw the first data that Dr. Romano showed at AIC, and this really was super informative. So you saw one of these cases that Dr. Romano already showed you. And here's another case. I will say to me, this does very much look like ARIA, both the timing but also, again, the association between ARIA-E and ARIA-H, which is very classically what we see in anti-amyloid therapeutics. But here I got -- I was probably more excited than most people were to see ARIA with an Alector. But this really did suggest to me that this is through a activation of an inflammatory or a microglial pathway, and very exciting from an idea that this is biologically active, exactly how we want to be in these individuals who have amyloid, and potentially biologically active against other proteinopathies. So here we see ARIA without an antibody that is directly binding to A-beta, and again, is very helpful for the mechanism, and I think encouraging for biologic activity, pharmacodynamic activity, and ultimately, hopefully cognitive effects as well. And this is just some of the data. Now, again, I was accused of naming it ARIA because it was a beautiful thing for something that was a worrisome side effect. But I do think that we've learned a lot about ARIA, and of course we want to minimize it, but I actually think there are, it's a sign of a pharmacodynamic treatment response. It's been very clearly associated with greater amyloid production. Way back in 2012, we published this, showing very focal amyloid clearance in the areas where there was first ARIA-E and then later ARIA-H. There's a little bit mixed data between whether ARIA is associated with improvement or worsening of neurodegenerative markers. In general, we do see more atrophy with amyloid clearance and ARIA is associated with more amyloid clearance, but there's a little bit of data suggesting that ARIA may actually be associated with a slowing of tau accumulation, particularly soluble tau markers and CSF. And again, over on the far right is from a recent publication we did trying to, again, map the amount of amyloid production with a clinical response. And not surprisingly, all of those that had significant amyloid reduction were also the antibodies that showed more ARIA. So even though I think we want to minimize it, particularly symptomatic ARIA, I think it's a marker that we are changing the biology in the brain in a way that ultimately has some benefits. So I think there's incredible opportunities for this kind of a program in the setting of already having anti-amyloid therapies. And the first, again, I'll say, I think we may be able to clear beyond amyloid, as you heard from the beautiful presentation. So can we stop the spreading of tau? And ultimately, through microglial activation, I think the question about whether we can impact intracellular or other proteinopathies during the stage that they're going from a soluble transmissible form, early seeding, early spreading, and again, beyond tau, but even into TDP-43. I think the ARIA is something we have to watch for sure. But again, I think it's a sign that we may be reducing amyloid through a different mechanism. And I think it'll be really important in this program to understand the relationship of ARIA to actual amyloid clearance when those data become available. I'm thrilled that the program is looking at tau in tau-PET. I am very interested in the microtubule binding region, both in CSF and actually by the time this reads out, it likely will have a plasma assay. Although the data I saw at CTAD really suggested the MTBR may be marking later stages of tau, mostly Brock Stage 5 and 6. So it'll be interesting to see how early it will mark this. And importantly, something that the program will need to do since, as I showed you, anti-amyloid treatments do affect tau, at least in lecanemab, is to understand what the additional benefit of a TREM2 mechanism will be from the amount of amyloid it lowers, because I suspect it will lower amyloid. The question is, does it go beyond the slowing of tau and slowing of cognition more than you would expect for an anti-amyloid agent? I mentioned TDP-43, which didn't come up a lot here, but I have to say in older people, I'm more and more impressed that this is contributing another proteinopathy that we have a difficult time measuring and have no real targets for therapeutics yet. And so I got interested in this kind of a mechanism that was more ubiquitous to ask whether it potentially would have effects on TDP-43 in addition to A-beta and tau. I do think there will be some challenges in enrolling and retaining participants if we make these anti-amyloid antibodies exclusion areas that become more standard of care. So I very much, I know the team's been thinking a lot about how they'll deal with this and the opportunities for combination therapy, which I think may be a fantastic idea in the future. So I'll end by just acknowledging the great group of people I get to work with and especially the participants who make these studies possible. So I'll turn it over for the wrap-up and question and answers. I look forward to answering any questions that come up. Thank you.

Thank you, Reisa. I hope that we will able to convince you that recruiting microglia and activating TREM2 is a strategy that's worth pursuing for Alzheimer's disease and possibly for other neurodegenerative disorders. Just to recap what we heard today, we are really developing immunoneurology as a broader alternative therapeutic strategy for dementia and neurodegeneration that would either Alector standalone or as you heard from just now in effective combination with anti-A-beta therapeutics, possibly anti-tau therapeutics, maybe Parkinson's disease with anti-alpha-synuclein therapeutics. So we think that recruiting microglia could have broad therapeutic benefit for multiple neurodegenerative diseases. Again, we think that we will have hopefully standalone therapy, but combination therapy is something that we will explore as soon as possible. You heard about the disease stages, and again, this is something that we will have to explore, but the mechanism suggests maybe broader therapeutic windows compared to anti-A-beta antibodies. And as Gary mentioned, we will look at other readouts for clinical efficacy beyond A-beta. As you heard, we may have effect on tau and we should have effect on other disease biomarkers and the clinical benefit. And again, to the best of our knowledge, we are the only clinical program in Alzheimer's disease with a TREM2 activating drug. And we completed Phase II recruitment and expect data in Q4 2024 with, again, data analysis shortly after. And again, this is an option deal with AbbVie and we hope to continue the great collaboration with AbbVie on this program. I will now open the podium for questions and Dr. Grasso will lead the question.

[Operator Instructions] And our first question will come from the line of Greg Harrison of Bank of America.

This is [ Mary Keaton ] for Greg. So the trial, as you've talked about, is in early Alzheimer's. Do you think AL002 could have an effect in those with more established disease? And maybe on the other side of that, how early in the disease progression could this be started to optimize benefit?

Thanks, Mary Kay. We appreciate the question. Maybe we can ask Gary to start and if Dr. Sperling would like to add or Dr. Heneka would like to add as well.

Yes, thank you very much. I think it's a very good question. We -- I think, we believe that earlier intervention is probably going to be more effective across neurodegenerative disease like, it is for any chronic disease. The choice of the MC -- I'll call it symptomatic, early symptomatic Alzheimer's disease or which is really MCI to mild AD. We chose that because it was a combination of, as early as we could go and still feasibly try to get useful data from a phase -- in a Phase II trial. But we would certainly be interested to explore this across the disease spectrum, both earlier and potentially later eventually. As I mentioned earlier, I think given the multiple mechanisms that may be at play here, give us reason to believe that this may not be as constrained as therapeutics that are focusing on one specific misfolded protein. I don't know if Reisa has, I'm sure Risa has some thoughts about this.

Well, I actually very much agree with you, Gary. So of course, I'm interested in early disease. And I do think since we see multiple proteins that are beginning even before symptoms, I actually think this would be a great thing. And we do know we can see that this kind of failure of the immune system begins, even though it may, I agree with accelerate at the stage of MCI, but I think we might want to hit it earlier. But I also agree that considering going to later stages is important because one of the things that really worries me is that people are going to -- there's a narrow slice right now for anti-amyloid therapeutics. And I'd love to see us have something later. And also, as I showed you, as people who start early, but even though we're slowing progression, they're still progressing to moderate dementia over time and we need to stop that. So I would love to see this given in combination or alone in early, but also really look at it in people who have progressed and have a lot of pathology to see if we can have more of an effect.

Dr. Heneka, anything you want to add or should we move to the next question? Okay. Let's move to the next question, operator.

Our next question will come from the line of Tom Shrader of BTIG.

A question on Slide 13, where you have high TREM2 versus low TREM2. Do you think that approximates your treatment effect or is your TREM2 activation significantly more than that? And I had a quick follow-up. The patients that have a lot of plaque, I forget the name of the mutation, but they have a ton of plaque, but they don't decline cognitively. Are there TREM2 levels known?

Maybe we'll ask Arnon to start and Dr. Heneka can add if there's anything you'd like to add.

Yes. Great question. I mean, the difference in the level of soluble TREM2 between patients that decline slower, slowly is pretty modest. I think it's in the range of 20%, 30%. We see elevation of TREM2 signal like multiple folds. So we think that our drug will exceed. And what you see genetically, we think that the genetic effect on the level of soluble TREM2, and TREM2 is modest. It still has clinical benefit, but it's modest changes in levels. We think that our drug will be significantly stronger than that. But again, the clinical trial will tell us. Yes, with regard to your second question, like, there are cases where high level of A-beta are not associated with disease. I mean, so there is some sort of you need elevation of both A-beta and more importantly tau to I think to get clinical symptoms. We don't know what's the level of TREM2 in this type of patients. But as you know, there was recent publications about a protective gene called Reelin well, in the presence of very high level of TREM2 in familial versions of Alzheimer's disease. The disease was delayed by, I think, 20 years. And this was associated with low level of tau. So there was a disconnect. So there are genes that can disconnect the link between A-beta and tau. And if this link is severe, you get protection.

We can move on to the next question, operator.

The next question will come from Ananda Ghosh of H.C. Wainwright & Company.

I think one of the key takeaways concerning AL002 was that the ARIA acting as a [ phospho dimer ] -- the PT biomarker. So I was wondering if Dr. Sperling might elaborate on the role of ARIA on the slowing of the tau, which you mentioned passing by. So that would be helpful to kind of understand also because the TREM2 has been associated both with A-beta and tau. So I was just trying to understand what's the role of ARIA on the tau.

Go ahead, Dr. Sperling.

Yes, I think it's a fascinating question on the tau-PET and one we're looking at. And especially, again, whether this is due to ARIA or actually because we see more anti-amyloid removal in people who have ARIA, at least somewhat more. And again, greater amyloid removal has been associated with greater tau slowing. But I think we really should look at this. And in particular, in people where you saw ARIA related to a rapid decline in amyloid, did it have a direct effect on tau, again, or is that mediated primarily through amyloid? But definitely analysis is ongoing, I think, with all antibodies right now. So it's a great question. And here, let me just say here, I think what's really exciting is you might have a direct effect on tau or other proteinopathies in addition to the anti-amyloid effect. So I'm really hoping for some synergistic removal of tau in the setting of this more direct microglial activation.

Our next question will come from the line of Myles Minter of William Blair & Company.

You've got Sarah on for Miles. So first, we've gotten some questions and would love to get the team's thoughts on a publication from earlier this year from the Holtzman Lab using a mouse model of amyloid and tau deposition testing kind of a similar antibody to AL002, which actually showed the potential to exacerbate tau feeding and spreading. Have you guys done any internal work on this? And if tau is a better predictor of disease progression than amyloid beta, what does this suggest for chronic TRM2 agonism? And I have a quick follow-up as well.

Arnon, do you want to start there?

Yes, and maybe Dr. Heneka can continue. But yes, the animal models, as Dr. Heneka showed, are confusing with regard to tau like, from the same labs or from different labs. You see cases where TREM2 has sort of beneficial effects on tau. In other animal models, in other contexts of A-beta plus tau versus tau alone, you see the detrimental effect of tau on -- of A-beta sorry, of TREM2 on tau. So the animal models are confusing. And as Gary keeps saying, it means that the ultimate test will be in the clinic. It means we think that the human genetic support, beneficial role of TREM2 on tau, means high level of tau, are associated with reduction, sorry, high level of soluble TREM2 are associated with reduction of tau, either directly or indirectly to reduction of A-beta. And we think, again, the genetics support a high level of TREM2 affecting both A-beta and tau the animal models have conflicting data. And the ultimate sort of judgment will be in our clinical trial. That's what we are doing the clinical trial for.

I find this a very important question. And it has to be mentioned that this just occurred with that one antibody and on one specific mouse genetic background. So while it didn't happen in other studies, still, I think it's very interesting. And part of the animal data is, in fact, confusing. But if we split the different mouse models and we look into detail, what and in which timeframe they accumulate pathology, it seems to make sense. And I think it's something which has to be further looked into it. But I don't think it raises too much concerns as it is not a phenomenon that occurs with all the antibodies, but one.

Great. Sarah, you had a follow-up question?

I do. Thank you. Are you guys collecting any information on the TREM2 mutation status of participants in INVOKE-2?

Gary, do you want to take that one?

We are, but we expect it to be a very small percentage of patients in the trial. So we will look, but we didn't stratify based on it. Based on it, it's very low frequency.

Operator, I think we're ready for our next question. Operator?

Once again, the next question will come from Paul Matteis of Stifel.

This is Catherine on for Paul. I guess just on the trial, what's your level of confidence that 1 year is long enough and that the trial is big enough to see separation from placebo? And I guess for the patients in the trial who are on drugs for longer, how much more in effect do you think that you'll be able to see?

Yes. Gary, do you want to comment there?

That's right. So good question. It is a common close design, so we will have data from patients that have minimum of 12, but up to 24 months in the common close design. And I just want to emphasize that our long-term extension, which will be blinded to treatment assignment, will also provide additional longer follow-up on these patients. And it'll be especially useful because the patients who roll over will be happy to titrate up for months on their -- on the drug. Are we confident that this study that has, number one, given that we have we and AbbVie have invested in replacing dropouts and making sure that we have a robust study with enough patients to measure both clinical outcomes and biomarkers. I mean, we're going to have a very, very biomarker-rich study here. And it's even going to be even -- I think, even better than we thought at the outset because of the advances that have taken place in the with the plasma biomarkers, particularly the plasma tau. Reisa mentioned the plasma microtubule binding region, but also a plasma P217 assays, which will help us to assess effects, not only on amyloid, but on tau. And I think for a mechanism like this that has so -- there may be multiple mechanisms that actually at play, contributing to slowing of disease progression, tau biomarkers will be really critical because they track, we know that in every situation we understand that they track very closely with clinical progression. So I think we'll be able to confirm clinical signals with the biomarker, rich biomarker data we'll have.

Thanks, Gary. I think we have time operator for 1 or 2 more if we have additional questions.

And our next question will come from the line of Yao Wang of Mizuho.

This is [ Yao Wang ] for Greg Suvannavejh, Mizuho. A question that we have is what is the bar of success with INVOKE-2 data? And are there any internal benchmarks the company is hoping to see for CDR and other key endpoints?

Yes. Gary, do you want to comment there?

I would say that we -- for us, success will be that we have evidence that we're slowing disease progression and that'll come from a combination of these clinical and functional outcomes as well as the rich biomarker package that we'll have. And this was intended when this study was designed originally by AbbVie and an Alector to do that, to really have a rich biomarker package to complement a substantive but not Phase III size trial.

Great. Thanks, Gary. Operator, do we have any final questions?

Yes. Our next question will be a follow-up from Tom Shrader of BTIG.

I have a question. Do you have any sense of your dosing flexibility yet in your earlier studies? At very high doses, are you driving inflammation in the brain or does it always look safe? Do you look at TNF and interleukin-1? I guess kind of the model is can microglia stay activated if there's nothing left for them to do? I know you're going to say you have to do the trial, but I'm curious if you have any hints.

Yes. Gary, do you want to start there and Arnon, of course, please add?

I mean, maybe I'll start, but I would say in the clinic, we're still blind at the treatment assignments, so we don't really -- can't really -- we don't have any -- we can't yet address dose relationship between ARIA or in other markers of inflammation until we finish the study. That said, I would just say that we've been, besides ARIA and relatively small number of infusion reactions that you expect with monoclonal antibody infusions we really don't have any safety signals that we -- other safety signals that we know of. And of course, our IDMC is looking at unblinded data on a regular basis.

Yes. In non-human primates, we went to very high doses, like 250 mg per kg, and we did not see any evidence of autoimmunity or inflammation. So we do think that -- we don't think that TREM2 is a pro-inflammatory mediator in the classical way. We didn't really see an elevation of any of the cytokines that you mentioned. So we don't, means we don't see evidence that there is a risk of inflammatory pathology in non-human primate. And as Gary said, the limited data that we have in human, we don't see evidence for that either. So targeting the innate immune system means with TREM2 appear to be generally well-tolerated at every dose that we tested.

Thanks, Tom, for the follow-up question. And thank you, everyone, for your participation today. I think that's the time we have. We appreciate it.

This concludes today's conference call. Thank you all for participating. You may now disconnect and have a pleasant day.