Welcome to the Morgan Stanley Global Healthcare Conference. I'm Jeff Hung, one of the biotech analysts. For important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. For this session, we have Alector with CEO and co-founder, Arnon Rosenthal, and CFO, Marc Grasso. Welcome. Great to be here.
Welcome.
Thanks, Jeff. So for those who may not be as familiar with Alector, can you just provide a brief introduction?
Sure. Alector was created with the idea of recruiting the brain immune system to counteract multiple disease pathologies. So whereas most other companies are going after the typical misfolded proteins, as a therapeutic strategy, Alector is taking a chapter from the playbook of immuno-oncology. And instead of going after misfolded proteins directly, we are recruiting the brain immune system to counteract not just the misfolded proteins, but multiple other pathologies that emerge in the brain. And the scientific rationale for this approach is human genetics. In the last decades, multiple risk genes have been identified for Alzheimer's disease, Parkinson's disease, and other major neurodegenerative disorders. And it turned out that the majority of the risk genes are actually either immune checkpoints or immune stimulatory molecules.
Although Alzheimer's disease is thought to be a disease where neurons are dying and synaptic connections between neurons are being destroyed, it turned out, based on the human genetics, that the immune system plays a key role in facilitating and not preventing the disease. So all our drugs are validated by human genetics, and they are all forms of either immune checkpoints or immune stimulatory molecules that harness the immune system to counteract the disease.
Great. Well, neurodegeneration is a crowded and difficult space, but there have been a string of successes in Alzheimer's. Can you just touch upon the insights that you've gained from these recent results?
Yes, there have been recent successes, not just in Alzheimer's disease, in migraine, the CGRP blocking drugs, in spinal muscular atrophy. So there've been actually... Even though neurodegeneration appeared non-tractable for a long time, there are quite a few successes recently. And yes, so Alzheimer's, the most notable are two or three anti-A-beta antibodies that's been approved. What it told us first is that Alzheimer's is a tractable disease, meaning it's not like, you know, a stroke, that there's been more than a thousand drugs that failed. So Alzheimer's is a tractable disease. On the other end, the drugs that were approved showed very modest efficacy, so there is room for either novel drugs or drugs that act in combination with the anti-A-beta therapy.
The third thing that it told us again, that the technology and the knowledge in how to run clinical trials in Alzheimer's disease is now matured. It means we now have multiple imaging tools, PET imaging for amyloid-beta and tau. We have multiple serum and CSF assays that represent disease progression and severity. We know how to identify Alzheimer's patients. The tools to run a good, conclusive clinical trials are there now. I think the field is now mature to come up with really new revolutionary drugs.
So how is Alector's pipeline differentiated from these programs? And what gives you confidence that ineffective and damaged microglia could be a common pathology underlying many neurodegenerative diseases?
As I mentioned, we are the first company that's really recruiting the brain immune system as a therapeutic strategy. And we know that in the brain, the immune system acts like the National Guard, the police force, the healthcare system, the garbage collectors. This is the entity that really makes sure that the brain functions. Imagine in New York, if the police force, the National Guard, and garbage collection are stopping to function, what sort of mayhem will occur? And that's what happens in the brain when the immune system stops functioning, either because of human genetics or because of aging. And we are ensuring sort of microglia, the immune cell's health and ability to counteract sort of multiple disease pathologies.
In this respect, we are fundamentally different than all the other approaches that really target one pathology. We know that Alzheimer's disease is associated with multiple pathologies. There's amyloid-beta being accumulated, there is tau being accumulated, synaptic, the connections between neurons are being destroyed. Half of the connections between neurons are being destroyed in Alzheimer's disease. The myelin wrapping on nerve cells is being destroyed. The support cells in the brain, the astrocytes, are dysfunctional. These are the cells that nourish nerve cells, and these are the cells that recycle neurotransmitters. They are disrupted. The nerve cells themselves don't function well. They are hyperactive, actually. The immune system in the brain control all these aspects of brain health. Our therapeutic approach is significantly broader than going after a single misfolded protein.
We think that because of the broad therapeutic approach, ultimately, our microglia modulating drugs will lead to better clinical efficacy.
Great. Well, let's start with your TREM2 program, the AL002. Data from the phase II INVOKE-2 study in Alzheimer's patients are expected in the fourth quarter. Can you just talk about the trial design and, and what's needed to complete the phase II?
Yeah, sure, Jeff. So the, you know, we're on track for the fourth quarter, and the trial is designed to really show what Arnon was mentioning in terms of the breadth of the activity of AL002. So the primary endpoint is measuring the slowing of disease progression as measured by CDR Sum of Boxes, so kind of standard that we've seen for other Alzheimer's therapies, and we're measuring that over a time period of between 48 and 96 weeks. So it's a common close design study, and that means that the last patient is treated for 48 weeks, and then you gather data up to 96 weeks for all of the additional patients. There's three active dose arms, and there's a placebo group, and we'll be looking at the totality of the data.
From our standpoint, it's going to be very important to see the picture in the context of not just the slowing of the disease progression from the standpoint of the clinical, the cognitive, and functional measures, but also the biomarker evidence of this broader mechanism of action, including looking at amyloid, tau, and other measures of microglial health.
You saw ARIA in a subset of APOE4 homozygous patients. Can you just talk about the ARIA that was observed, how serious were they, and what modifications were made to the study for mitigating ARIA?
Sure. So first, just to recap, we have AL002 is a drug that activates TREM2. TREM2 is a prime immune stimulatory receptor for the microglia. It's a major genetic risk for Alzheimer's disease and other neurodegenerative disorders. So really, it's a prime example of immune checkpoint, immune stimulatory molecule for the brain, and we have a drug that activates it based on the human genetics. And, as Jeff noted, we are completing phase two data. We'll have data by the end of the year, and one of the things that we have seen in the phase two are incidences of ARIA. ARIA are MRI imaging abnormalities that were reported with the anti-A beta antibodies. They are thought to be related to removal of beta amyloid from the vasculature.
So once you remove beta amyloid plaques from the vasculature, you create small holes in the blood vessels, leading to leakage of blood products from the vasculature to the brain, and this is basically what cause the ARIA. So the ARIA in with the anti-A beta antibodies were shown to occur early. They are reversible. They occur largely once. They are dependent on the copy number of APOE4. And we saw incidences of ARIA, which are indistinguishable from the ARIA that was reported for beta amyloid. So first, this suggests that we have an active drug, maybe possibly even a drug that removes beta amyloid, similar to the anti-A beta drug. And it's important to know that the ARIA is a very manageable phenomena. Again, it's an MRI abnormality.
It rarely leads to clinical symptoms, means the percentage of clinically severe symptoms that we see is less than 1%. We identified one, two patients out of over 300 patients that were treated, and even in the severe cases, the ARIA resolve, and the patients are generally fine. So in order to manage the ARIA, what one does now is doing a slower dose escalation. So you start with slow dose, and every month or every eight weeks, you escalate the dose until you reach the targeted dose. And when we are doing that, we see sort of, again, very mild cases of ARIA that are largely non-symptomatic at all.
In less than 10%, they have mild symptoms like headaches and nausea that is rapidly reversible, and in very rare cases, we see more severe symptoms, again, less than 1% in our clinical trial.
And you recently reported baseline characteristics. So, you know, given heterogeneity with dementia and cognitive impairment, what gives you confidence that you're studying the intended early AD population?
Yeah, we means, like in any disease, prevention is best, like starting early in the treatment is better. And we are, we elected to conduct a phase two clinical trial with people that either have what's called mild cognitive impairment or very early Alzheimer's disease, to really capture the disease before it becomes too severe and possibly irreversible. So we presented the baseline characteristics of our patients, and they are consistent with the patient population that we intended to treat. Based on the CDR, this is the clinical readout, we see that our patients are indeed at a very early stage of the disease, they are basically have very mild Alzheimer's, clinical Alzheimer's symptoms, and they would be amenable for treatment that would slow down, hopefully even stop the disease progression.
So the clinical patients that we are treating is very similar to the clinical patients that was used for the anti-A beta antibodies. And again, so, it will give some perspective of our clinical trial.
On the baseline data, you saw a mean amyloid centiloid of around 100. Can you just talk about your expectations for what level of amyloid clearance you would expect in INVOKE? And maybe just more broadly, you guys touched upon this a little bit ago, but you know what would you want to see in this study? What's kind of the bar for success?
One of the critical criteria for recruitment into Alzheimer's disease patients now is PET imaging that measures that patients actually have a high level of A- beta or beta- amyloid plaques. In the past, before the emergence of PET imaging, about 30% of people with dementia turn out to be not to have Alzheimer's disease, as defined by beta amyloid plaques. All our patients have to go PET imaging for A- beta or serum or CSF testing for beta amyloid to ensure that they have actually Alzheimer's disease. For people that have even mild symptoms, Alzheimer's A- beta plaques are fully loaded in the brain. That means if you measure it, it's called 100 centiloid. That's the maximal level of A- beta plaque that you can measure with this PET imaging.
With A- beta amyloid antibody, anti-A beta antibodies, what was observed is that the antibodies can reduce beta amyloid very significantly. They require to reduce beta amyloid by 80% to 90% to elicit some clinical benefit. So we, again, measuring the beta amyloid with PET imaging. Because of the broad mechanism that we have, we don't think that we are dependent on complete removal of beta amyloid for clinical efficacy. We think that we can elicit clinical efficacy even with modest reduction of beta amyloid. So we are looking forward to see some effect on beta amyloid, but we don't think that the trial and the clinical efficacy of this drug are really dependent on complete removal of beta amyloid, which is different than what you see for anti-A beta drugs.
Now, for TREM2, you know, some of your competitors have established other modalities, like small molecules, with some data to support differences in target profile. Do you have data as to what extent the monoclonal antibody binds soluble TREM2 versus membrane-bound TREM2? And what are your thoughts on how this could impact efficacy?
In our view, and I think it's still the prevalent view in the field, what's important is activation of the membrane TREM2. This is the receptor that's expressed on microglia and transduce signaling that stimulate microglia to proliferate, to survive, to migrate to the site of injury, to enhance its ability to phagocytose misfolded protein, enhance its ability to protect the surrounding neurons. TREM2, as part of its regulation, is being cleaved by a membrane protease. You see if you measure in the CSF of human, for example, you see some level of soluble TREM2, and there's debate in the field whether soluble TREM2 is just a by-product of negative regulation of the receptor, just cleavage, or whether it has some independent activity.
We think that what the soluble TREM2 does, it just prevent the system from activated prematurely. If there is normal wear and tear in the brain, there is some release of damaged membrane fragment, or there is some low level of beta amyloid that is not sufficient for immune response, you want to capture and prevent that. So the soluble TREM2 just suck low levels of ligands and prevent it from reaching the membrane TREM2 and activate the system. Once there is a lot of damage or stress ligands, it overwhelms the soluble receptor, reaches the membrane TREM2, and then the immune system can be recruited and deal with the disease. So we think that soluble TREM2 has no independent function outside of modulating the membrane receptor.
So our drug is really designed to bind the membrane TREM2 to activate the receptor on the microglia cell surface. It does it both as a standalone drug, and it does it also together with the natural ligand. It actually enhances the ability of the natural ligand to activate TREM2. So we think that, again, our drug is focused. It binds in what's called the stalk region of the receptor, in a region that is different than the ligand binding region. And it basically primarily activates the membrane receptor. It doesn't really impact the soluble receptor.
Okay, great.
Mm-hmm.
Let's move on to latozinemab. Could you just talk, you know, give a brief overview of the program, how FTD differs from late onset dementias, and what you and your partner, GSK, hope AL001 can achieve?
The second disease that we are going after, besides Alzheimer's disease, is frontotemporal dementia. It's an early onset form of dementia. It's the largest type of dementia for that hits people under the age of 60. It's a very rapidly aggressive form of dementia. It progresses at least three times faster than Alzheimer's disease. People lose brain tissue very rapidly, again, three to five times faster than Alzheimer's disease. Clinically, the disease appears different than Alzheimer's dementia. Alzheimer's dementia is primarily typified by loss of short-term memory, recall memory. Frontotemporal dementia is typified by behavioral disinhibition, speech pathology. And again, it's very lethal. Once you get diagnosed, you die within seven to 10 years.
Because it's a different disease, I mean, again, there are no drugs for this disease, either disease-modifying or even symptomatic. Because it's a different disease, we developed a different drug for this disease. The disease has genetic underpinning, and one of the genetic causes of the disease are mutations in an immune regulatory protein called progranulin. Basically, patients that have one good and one bad copy of progranulin invariably develop this frontotemporal dementia. We develop a drug that elevates progranulin back to normal level in these patients by blocking degradation cascade of progranulin.
Now, Alector is presenting INFRONT-3 baseline characteristics in the next few weeks. What should investors focus on in the baseline characteristics?
Yeah, I think the thing to look for is, you know, have we enrolled the representative population for studying this medicine? And I think what you'll see is that we have. And we'll be presenting this at a conference, ISFTD, which is the Frontotemporal Dementia Conference, coming up in September in Amsterdam. And you know, similar in some ways, you know, in the Alzheimer's studies, you know, you want to see that, you know, we're enrolling the target population. There's a significant number of patients in our phase 3 who are symptomatic frontotemporal dementia patients. We enrolled a hundred and three symptomatic patients. That's ahead of our target of 90 to 100 . And again, you know, you'd look to see that it's representative of that population.
Now, if I'm not mistaken, for INFRONT-3, enrollment completed last October.
Correct.
The study is expected to go 96 weeks.
Yeah.
Can you just talk a little bit more about the trial design and, you know, when you would expect to start reporting results?
Yeah, absolutely. That's correct, so we completed enrollment in the fourth quarter of last year, and again, ahead of our target in the symptomatic population. We also have some pre-symptomatic patients as well. It's comparing treatment arms, so 60 mg IV once monthly as compared to placebo, and as you mentioned, it's over a treatment period of 96 weeks. We're looking at an adapted endpoint similar to CDR-Sum of Boxes for Alzheimer's, but it's made specific with two additional domains, and as Arnon mentioned, in frontotemporal dementia, there's a differentiated impact on the frontotemporal cortex as compared to the hippocampus in Alzheimer's, and that affects behavior, and it affects language, so those are two additional domains that are included in this CCDR-Sum of Boxes .
It's an adaptive CDR Sum of Boxes as the primary endpoint. Then also similar to INFRONT-2, we're looking at a range of biomarkers that speak to the, again, the health and functioning of the microglia. In this case, we're, you know, restoring progranulin back to normal levels, and you wanna see that that's enhancing the activity of the lysosomal activity and the microglia and other measures. And that's also in the context of what we saw in the phase II. One of the things that we saw, in addition to improvement in the biomarkers and slowing of disease progression, is also some improvement based on brain imaging and volumetric MRI, and that will also be a component of what we're looking for in the phase III.
Timing of data is gonna be towards the end of 2025 , early 2026 .
Now, you just talked about the high degree of variability with at-risk versus symptomatic patients. So what drives this variability? Like, how does it manifest and, you know, how is INFRONT-3 being conducted to mitigate this?
So first, just to add a few things. Meanwhile, we received breakthrough therapy for this drug based on our extensive open label phase II studies, which showed, in addition to Marc mention, also slowdown in cognitive decline by 48% over 12 months, which is very profound if it sort of stays in the phase III. So initially, we, the clinical trial was designed to include both symptomatic and pre-symptomatic patients. We did blinded sample size analysis of our phase III trial, and we saw that the pre-symptomatic component of the trial shows very high variability. The variability is because it's hard to predict when the pre-symptomatic will convert to symptomatic.
I mean, we initially, based on historical data, thought that people that have the genetic mutations and have high level of the neurofilament biomarkers will convert within one to two years to symptomatic. It turned out that the conversion is much less predictable. It could be three, it could be five years. So the variability is really based on the inability to predict the conversion timing and rate. And if you remove the pre-symptomatic patients from the trial, once you go to only symptomatic patients, the progression rate is much more predictable; it's much less variable.
So this enables us to have sort of a meaningful, sort of a decent-sized trial, and we are actually if we remove the symptomatic, we were able now to detect effect size of in the mid-20s, like 25% slowdown in cognitive decline, whereas before we were only able to detect slowdown in cognitive decline around 40%. So the trial, as it's designed now, is able to detect up to 25% slowdown in cognitive decline. It's focused on symptomatic patients. The phase III, it's a pivotal phase III the data justify that we will file a BLA based on this. We agreed with the FDA about both the primary and the secondary endpoints.
As Marc said, we have a cohort of pre-symptomatic patients, which we are still following, and eventually we would like to include pre-symptomatic in the label, and then we'll see how we best do that.
You've mentioned that the FDA is open to evaluating plasma and CSF progranulin as supportive evidence of clinical efficacy for the BLA review. Can you just walk us through some of these discussions?
Usually with clinical trials, you have to run through two, phase III, validating phase III's, in order to get approval. That means we, in a way, agreed with the FDA that one clinical trial will be sufficient. And instead of, in a way, running a second clinical trial, we will be able to show that progranulin, the missing protein that causes the disease, is normalized with our drug. So if we have one positive clinical trial, and we show that the progranulin is elevated back to normal level, it should be sufficient for the BLA filing.
Could you just walk us through your current thinking on, you know, commercial launch preparation and education that's required?
Yeah. Yeah, absolutely. So, you know, as Arnon was mentioning, you know, this is a form of dementia that unfortunately is underserved. There's no current therapies, and it's also often misdiagnosed, and it's underdiagnosed. There's a familial component, strong familial component. So one of the things, you know, that we're active in right now is raising awareness for this disease. This form of disease that, for example, Bruce Willis has, well, Wendy Williams has, and, you know, it, it's really devastating for the patients, for their families, and it's very difficult for the caregivers. We've spent a significant amount of time already with these families and caregivers across the world.
and, you know, having now fully enrolled this pivotal phase III study, we, you know, we're, I think, in a, you know, a much more advanced position as it relates to this, disease and, the first disease-modifying therapy for this disease to really, you know, make a substantial change. And I think a lot of that groundwork is really important. As I mentioned, we're, you know, at this conference in Amsterdam, later this month, you know, that's from the frontotemporal dementia conference. There's patient advocacy groups that are part of that. There's longitudinal cohorts of patients that are part of that. And there's gonna be continued efforts to, you know, raise awareness for this disease over the coming months, going into commercial launch.
Great. Maybe a couple questions on your BBB platform technology. You know, we've seen several advances in BBB delivery. Could you just talk about your proprietary platform and how it compares with the current landscape?
Since we are completely focused on brain disorders, one of the things that we are focusing on is developing technologies that enhance drug delivery to the brain. We spent quite a few years sort of maturing what we call our blood-brain barrier transport technology or Alector Brain Carrier. Alector Brain Carrier is really dependent on the Trojan horse approach. As you know, the brain is generally shielded from the serum, means that blood product cannot really enter the brain. But there are receptors on the brain endothelial cells that transport specific nutrients that the brain require, like transferrin or insulin or insulin-like growth factors or amino acids. You can hitchhike on this receptor, and they internalize their natural ligands. You can really...
If you bind something to the same receptors, they internalize your drug together with the natural ligands. So we are using two receptors to transport either antibodies or proteins, enzymes, or eventually ASOs from the periphery to the brain. One of the technologies is the transferrin receptor that's expressed, again, on the brain endothelial cells and transports transferrin through that iron to the brain, and the second technology is an amino acid transporter called CD98, that again, transports amino acids from the plasma to the brain, and each of these technologies have different cons and pros. Other companies are developing similar technologies. What's unique about our technologies is the versatility and ability to tailor to the specific target. Means, the short side of these technologies is that there are issues with adverse effects.
Sometimes, like, the drug doesn't only go to the brain. It goes to peripheral tissues, like reticulocytes, that express high level of the target receptors, such as transferrin, and it leads to anemia in many cases. You really need to tailor the technology to optimize the safety and efficacy profile. We developed a technology that really can be optimized for each drug to maximize the safety and the efficacy. In this way, we are much more versatile than other technologies that are currently being used.
And you talked about transferrin receptor and CD98. So to what extent are these pan-cellular, you know, versus achieving cell type specificity?
So to a degree, they are both pan-cellular. Like, the transferrin receptor is, in the brain, highly expressed on nerve cells and less expressed on support cells, like, again, microglia, astrocytes, oligodendrocytes. CD98 is, to a degree, the opposite. It expressed at low level of neurons and at higher level on microglia. And also in the periphery, they express in different tissues. Transferrin is expressed at very high level on reticulocyte precursors. CD98 is expressed in other peripheral tissues. So again, depends on the target in the brain, depends on what side effects you try to avoid. You can choose what which of these technologies to use. But so again, if you want to transport something that will focus on microglia, maybe CD98 is better.
If you want something for neurons, maybe transferrin is better. But generally, they mainly, they go into the brain and they will distribute pretty broadly.
Great. Maybe one last question. Can you just remind us how much cash you have in the runway?
Sure, Jeff. So, as of the end of the second quarter, we had just over $500 million in cash and no debt. That's runway through 2026. That's a full two years beyond the expected INVOKE-2 data for AL002 later this year, and approximately a full year after the phase III pivotal data for latozinemab towards the end of 2025, early 2026. That does not assume any milestones from our partners, including the $250 million payment from AbbVie, if they choose to opt in on the AL002 program, or from our partner, GSK, on the progranulin programs.
Great. Let's leave it there. Thanks so much for your time.
Thanks, Jeff.
Thank you.