Good afternoon, and welcome to part five of the Arrowhead Pharmaceuticals 2024 summer series of R&D webinars. At this time, all attendees are in a listen-only mode. A question-and-answer session will follow the formal presentations. If you'd like to submit a question, you may do so by using the Q&A text box at the bottom of the webcast player. As a reminder, this call is being recorded, and a replay will be made available on the Arrowhead website following the conclusion of the event. I'd now like to turn the call over to Vince Anzalone, president, vice president of Finance and Investor Relations at Arrowhead. Please go ahead, Vince.
Thanks so much, Tara, and thanks everyone out there for joining us today. Before we start, just wanna make sure everybody knows we will be making forward-looking statements, so refer to our SEC filings for the risk factors. Okay, so, as Tara mentioned, this is part five of our summer series of R&D webinars. It's, you know, I realized I walked out of the office today and realized it's no longer summer. It's still... It's fall now, but here on the West Coast, it feels like middle of summer. It's very, very hot here. But we wanna thank everybody for joining all five of these sessions. They've been really informative, and we appreciate all the work from the Arrowhead folks, and we appreciate all the participation from investors.
So the goals of the summer of R&D series of webinars were to provide some focused time, to talk about underappreciated parts of our pipeline, to talk about advances in the TRiM platform. And these first two are really important today 'cause that's exactly what we're talking about. We haven't talked a lot publicly about our CNS platform, and the multiple pipeline candidates that have come out of that, and we've really had some important, and some exciting breakthroughs recently, that we wanna talk to you about today. Third, you get to hear directly from the Arrowhead scientists and the clinical development folks and clinicians that have worked on these programs, that continue to work on the programs.
And also, importantly, you get an external physician perspective on the disease areas that we're trying to act. So today, here's what we'll talk about. Christi Esau will talk about the CNS portfolio broadly, and the intrathecal administered part of the platform. James Hamilton will talk about our first candidate using that platform, ARO-ataxin-2. Tao Pei will talk about some advances in a subcutaneous delivery method for delivery to CNS that's, again, subcutaneously administered, that crosses the blood-brain barrier. Some really exciting stuff here. Christi will talk again about targeting tau, specifically, for neurodegenerative diseases, and also a couple early pipeline programs that we haven't talked about publicly.
Dr. Jose Soria, who I'll introduce in just a moment, will talk about where the unmet need lies in Alzheimer's disease, and then I'll give you kinda some key takeaways for the program. So before we move on, I just wanna introduce Dr. Soria. We're very lucky to have him with us today. He's the director of clinical research at The Neuron Clinic, and he's also an assistant clinical professor of neurosciences and an attending neurologist at UC San Diego. He knows the space very, very well, so we appreciate his joining us today. And feel free to ask questions that he'll be able to answer. That's what he's here for. So now I'll turn the call over to Christi.
All right. Terrific. Thank you, Vince. So neurodegenerative diseases affect more than 50 million people worldwide, and they represent the leading cause of disability, but there's almost no disease-modifying therapies available. On this slide, we're highlighting some of the most prominent examples of neurodegenerative disease, from prevalent diseases like Alzheimer's and Parkinson's, to rare diseases like Huntington's, ALS, and the ataxias. I'm sure almost all of us have a family or friend who's been affected by one of these devastating diseases. One thing that they have in common is that they're caused by abnormal protein aggregation in the central nervous system, which causes toxicity to neurons. So this has been a difficult mechanism to drug with traditional therapeutic modalities, but with RNAi, we can knock down the root cause of the disease, the toxic protein, and this has potential to be disease-modifying.
So with the advancements we've seen in RNAi design over the past decade, the potential for robust delivery to tissues beyond the liver has become a reality, and Arrowhead has been aggressive in moving into these new tissue types, as you've heard these past few months from the summer series. The advancement of RNAi into the CNS is also coming at a time of rapid progress in our understanding of the genetics of these disorders and development of biomarkers to monitor disease progression, which is enabling more effective clinical development and increasing the probability of success. So all of these advances are laying the groundwork for a new generation of disease-modifying therapies in these historically intractable diseases. So today, we're disclosing our first CNS-targeted program moving into the clinic, which is ARO-ataxin-2 for spinocerebellar ataxia.
This program is coupled with what we're now calling our first-generation platform, which is intrathecal delivery to the CNS. So as we've shared previously, it's a simple lipid conjugate design, which, after intrathecal administration, leads to potent target RNA inhibition throughout the brain and the spinal cord, and in all relevant cell types. Just as we've seen for the liver and lung TRiM platforms, it has a long duration of action that should enable quarterly or potentially even less frequent dosing, and safety so far has looked good. We've completed GLP tox for our first clinical program, and there were no serious adverse findings, so I wanna remind you of some of the preclinical data for this platform that we previously shared, starting with here's some data from rodent models.
On the left, we're looking at target RNA knockdown in the spinal cord of rats four weeks after a single intrathecal administration, which here means lumbar puncture. This is a human SOD1-targeting siRNA, and ED 50 in the spinal cord is calculated at 33 mcg, which compares favorably to the published ED 50 of the SOD1-targeting ASO, tofersen, which is FDA-approved for SOD1 ALS. So we're able to achieve maximal reduction of 95%, so good efficacy in addition to potency. On the right, we're looking at SOD1 RNA knockdown in mouse two weeks after single intrathecal administration, which here is directly into the cerebral ventricle, or ICV. And again, the data show maximal reduction of 95% and an ED 50 of 13 mcg, and this is about four times lower than the published tofersen ED 50 of 64 mcg.
How does this translate into the non-human primate? In this graph on the left, we're looking at target RNA knockdown after a single 45 mg intrathecal lumbar puncture administration at one month post-dose. You can see on the left, spinal cord and cortex regions, we're achieving 90%-95% reduction of SOD1 RNA in those brain regions. Throughout most of the brain, though, we're able to achieve 70%-80% reduction at the RNA level. Even in the deeper brain regions on the far right, caudate putamen, which are known to be difficult to reach with intrathecal administration, we're able to achieve 50% knockdown. This broad target knockdown throughout the brain means that this platform can be applied to a range of neurodegenerative diseases affecting different areas of the CNS.
Now, on the right is a section of the cortex in the treated monkey, where we're detecting the siRNA by in situ hybridization. You can clearly see the pink staining of the siRNA in the large neurons. You also see it, though, in the yellow-stained astrocytes and the blue-stained microglia. So this is just one piece of representative data that demonstrates we can target multiple cell types in the brain, which is important because most of all of these cell types do contribute to disease. So we've also looked at duration of action in the non-human primate, and here we're looking at SOD1 protein levels in the spinal cord and cortex regions at one month, three months, and six months after a single 45 mg dose administered intrathecally.
SOD1 protein is known to have a long half-life, so the protein levels at one month don't fully reflect that RNA knockdown I showed you on the previous slide. At three months, we see 70%-90% reduction in SOD1 protein across these brain regions, and they're sustained at the six-month time point. This duration of action should be sufficient to support quarterly dosing regimen at a minimum, and half-yearly or longer is possible. On the right, you can see that the SOD1 protein levels in the CSF track the levels in the tissue very well, with maximal reduction of 70% around two months post-dose, and this is sustained out to the six-month time point. Target protein levels in the CSF have been a useful biomarker in the clinic to demonstrate target engagement in the brain for a number of targets.
This level of reduction we're seeing here compares favorably to the 50% reduction reported in non-human primate for tofersen after a more aggressive dosing regimen. So highlighting the potential for siRNA to achieve better potency and efficacy compared to ASOs. After that introduction to our intrathecal platform, James will share more about the first program moving into clinical development.
Thanks, Christi. So I'd like to introduce everyone to the ARO-ATXN2 program for SCA2, or spinocerebellar ataxia type 2. As Christi mentioned, this is our first program using the IT route of administration. And just a brief introduction to the condition. So SCA2 is a dominantly inherited repeat expansion disorder with a prevalence of about five per hundred thousand. It's caused by variable lengths of a repeat expansion, leading to a mutant expanded polyglutamine tract in the ataxin-2 protein that tends to aggregate and form clumps in the Golgi or in the endoplasmic reticulum that as it accumulates leads to neuronal and cellular dysfunction.
This dysfunction, over time, leads to a progressive cerebellar ataxia, so these patients have difficulty with ambulation, an unsteady gait, difficulty with speech and with swallowing, as well as some other either Parkinsonian or ALS-like symptoms. Symptoms tend to develop in the 20 year to 30 year age range, and patients may eventually need the aid of a walker or a wheelchair within about 10 years after symptom onset. The patients do have a shortened survival with a median survival after onset of symptoms of about 10 years to 20 years, and the care right now is purely supportive. There's no disease-modifying therapy available, and this kind of a disease where there's an accumulation of a mutant or a toxic protein is ideal for an RNAi therapeutic.
Here we're showing some data that we developed in a transgenic mouse model. On the left, we're showing knockdown of the human mRNA. So we've got a nice dose response in the mice with IT administration, and then the human protein in the middle. Then on the far right, we're comparing our ARO-ATXN2 molecule in the same model with the competing ASO against the same target, where we think we have a potency advantage. We've also done work in the non-human primate with ARO-ATXN2, and here we're showing post-IT dose day 29 knockdown in various regions of the CNS. You can see particularly at the top dose in the cerebellum, which is the critical area here for spinocerebellar ataxia, we're getting up to 83% knockdown of ataxin-2.
This effect progresses or is sustained rather out through around six months. Again, focusing on the blue bar, representing the cerebellar tissue, we can maintain knockdown better than 50% after a single dose out through about six months. We've filed to initiate a clinical trial in patients with SCA2. This will be a single escalating dose study in SCA2 patients, starting at 10 mg single dose, and dose escalating through 100 mg. We'll be evaluating, of course, safety and tolerability as well as PK. Importantly, we'll also be looking at key biomarkers like ataxin-2 protein levels in the CSF and levels of NfL in the plasma, as well as in the CSF, which, similar to ALS, has been correlated with disease severity and disease progression in ataxin-2.
And then there are several functional rating scores that we'll be evaluating in this study, as well as MRI-based imaging. The study will be conducted in the Asia-Pac region, New Zealand, Australia, Taiwan, Canada, and also parts of the EU. So we can take questions on that program later, but now I'd like to turn things over to Tao Pei, who will walk us through the subcutaneous platform for siRNA delivery to the CNS. Tao?
Thank you, James. So we'll cover Arrowhead's new CNS-targeting TRiM platform, as James mentioned, using a systemic delivery approach by subcutaneous dosing, in contrast to the local intrathecal administration. Here on the left is a cartoon of the new TRiM platform. The siRNA duplex is covalently linked to a ligand that targets TFR1, or transferrin receptor 1, for delivery of the siRNA across the blood-brain barrier and into the central nervous system. This ligand-driven approach enables non-invasive BBB penetration, followed by productive cellular uptake in brain tissue. Using this new delivery platform, we have demonstrated deep and durable knockdown of multiple therapeutically relevant genes in CNS. In addition, this platform is comparable with subcutaneous administration, with potential monthly to quarterly dosing. Furthermore, we have demonstrated a favorable safety profile in both rodents and non-human primate.
In this non-GLP exploratory tox studies, we have seen a minimum of ten-fold safety margin over the efficacious dose. Transferrin receptor 1 is enriched within the endothelium of the blood-brain barrier and has been investigated as a target receptor for drug delivery into the CNS. The primary function of the transferrin receptor is to deliver iron across the blood-brain barrier through a process known as receptor-mediated transcytosis. In this pathway, the receptor undergoes rapid kinetics of internalization and recycling. When we take advantage of this endogenous process, we're able to non-invasively transport our TFR1 ligand conjugated siRNA across the BBB and assess the brain parenchyma. To avoid competing with endogenous transferrin in blood circulation, our ligand is designed to bind to the apical domain of transferrin receptor 1 away from the endogenous transferrin ligand binding site.
This design allows us to not only leverage TfR1-mediated transcytosis to cross the BBB, but also to circumvent potential interference with the endogenous ion trafficking pathway. Here we show we can achieve BBB penetration in mouse using our TRiM CNS subQ platform. In these studies, we compare siRNA delivery in wild-type mouse brain and TfR1 transgenic mouse brain, the latter which expresses the human transferrin receptor 1 gene. We assess ligand-driven delivery of siRNA by tissue stain on the left and SI quantitation on the right. In the tissue stain on the left, we visualize siRNA in mouse hippocampi. The big blue dots are the nuclei of brain cells. The green dots are the endothelial cells in mouse BBB, and the small red dots are siRNAs.
As you can see, first, the red colored siRNA is in the brain, and second, majority of the siRNA is co-localized with the blue-colored brain cells instead of green-colored endothelium cells, indicating our TRiM conjugate has reached inside the brain. Moving from cellular level to tissue level on the right, we also see the power of ligand receptor pairing utilized in our TRiM CNS subQ platform, which allow us to achieve 56-fold siRNA concentration in the brain of TFR transgenic mouse, compared to the siRNA delivered to the brain of wild-type mouse. Comparing the TRiM CNS intravenous platform, colored in green on the left, with the same subQ platform, colored in blue on the right in non-human primate, we focus our comparison in four regions. The thoracic spinal cord region represents the spinal cord of CNS.
The temporal cortex represents the cortex region of the brain, and caudate and putamen represents the deep brain regions. Here we found the CNS intrathecal platform delivers much more siRNA to the spinal cord and cortex than the deep brain region by as much as 22-fold. In contrast, using our CNS subQ platform on the right, we see relatively even siRNA distribution across all brain regions. The siRNA concentration differences between the temporal cortex and caudate or putamen are within only 2-fold. Moving from tissue PK to PD in these four CNS regions in non-human primate, we see good correlations. Using MAPT as an example here, with our TRiM CNS intrathecal platform on the left, we can achieve very good knockdown in spinal cord and cortex, even at three months after a single dose of 50 mg of TRiM conjugate.
However, minimum knockdown of MAPT mRNA is observed in caudate or putamen. In contrast, with TRiM CNS subQ platform on the right, we see over 50, up to 65% knockdown of MAPT transcript across different brain regions, from spinal cord and cortex to deep brain in caudate and putamen. That's at three, three months after a total 9 PK dose by subcutaneous administration. When we zoom out and examine early time points in the same CNS region, we see already 60s up to 80% knockdown as early as day 29, and about 70% knockdown in all four CNS regions at day 43, and again, up to 65% knockdown on MAPT mRNA at day 99. That's three months post subcutaneous dosing. The knockdown duration data in non-human primate shown here supports a monthly to even quarterly dosing regimen.
Finally, our current formulation can deliver 150 mg siRNA-based TRiM conjugate in less than 4 milliliters total volume. This will support subcutaneous administration in upcoming clinical trials. Next, I will hand over to Christine to discuss our CNS programs using this new CNS TRiM platform.
Thank you, Tao. Okay, so the first program moving forward with the systemic CNS delivery platform will be ARO-MAPT subQ, as we're calling it, for Alzheimer's disease and tauopathies. Tau protein is encoded by the MAPT gene, and it's abundant in neurons, where it promotes stabilization of microtubules. However, it's intrinsically disordered, which makes it susceptible to aggregation, particularly in the context of certain post-translational modifications. Hyperphosphorylated tau can assemble into structures called neurofibrillary tangles inside cells, which are correlated with neurodegeneration in Alzheimer's disease, and these structures can be visualized using PET imaging. What is the role of tau in Alzheimer's disease? Alzheimer's, of course, is characterized by amyloid plaques made up of amyloid beta.
As most are probably aware, we've seen the first kind of glimmer of hope for Alzheimer's therapy with the approval of the anti-amyloid antibodies that can clear the plaque and slow the disease, but amyloid plaque can accumulate in the brain for decades before the onset of clinical symptoms. It's the red curve on the left is the amyloid plaque, whereas tau comes up later. That's the blue line. And it's actually the downstream accumulation of tau neurofibrillary tangles that's correlated with the clinical decline. How amyloid leads to accumulation of tau is not clear, but most evidence points to amyloid formation as necessary, but not sufficient for clinical manifestation of Alzheimer's disease.
So it makes sense that anti-amyloid therapies would have limited impact on the established tau pathology in people with clinical symptoms, and it would be less effective in patients with a higher burden of tau in the brain, and therefore, later in the disease progression. So the anti-amyloid antibodies have struggled with serious side effects as well, which have limited their use, and, most especially, in the patients who need it the most. So with tau, the hypothesis is that clearing the toxic neurofibrillary tangles can preserve cognitive function. There's been many approaches to reduce tau tangles that have been tried, but most have not been successful. Immunotherapy has been difficult due to the intracellular localization of tau. So RNA targeting may be the optimal approach to prevent tau protein production, and over time, clear tau tangles.
And the first evidence in support of this hypothesis comes from the Biogen MAPT-ASO program, which last year they reported reduced tau PET signal in the brain of patients starting at six months after inhibition of tau with the ASO, and increasing over the course of two years. Now, the next step is to link that reduction to clinical benefit, which is ongoing in a larger clinical study. So notably, the Biogen program is an ASO delivered intrathecally, which is going to be challenging to scale to meet the need of the large patient population of Alzheimer's disease. So a systemically delivered siRNA can make that possible.
So I wanna also mention that there are other neurodegenerative diseases caused by tau tangles that don't involve beta amyloid as an initiating step, and these are collectively called tauopathies, and each affects different regions of the brain, and consequently, has different clinical manifestations. They're also caused by a diversity of tau isoforms and aggregate structures that are responsible. So the ARO-MAPT subQ is designed to target all of these isoforms at the RNA level, so it's agnostic as to tau protein structure and can be applied to all of these tauopathies. So let's take a look at some of the preclinical data we have for the ARO-MAPT subQ candidate moving into clinical development. So the transferrin ligand shows similar affinity for the transferrin receptor in non-human primate, compared to human, and so we expect the activity in the monkey to be translationally relevant.
So in this experiment, we've given three weekly 3 mg per kg subcutaneous doses, and the 3 mg per kg represents the siRNA only. So at two weeks after the last dose, we looked at MAPT mRNA in a panel of CNS tissues. So as Tao, the chemist, highlighted, we see 60%-80% reduction in target RNA across a broad range of brain and spinal cord regions, and this includes the Alzheimer's disease-relevant cortex and hippocampus. Now, on the right, we're looking at the MAPT mRNA in the hippocampal neurons of these monkeys using RNAscope.
So in the control group on the top, you can see a lot of red spots in the hippocampal neurons, but on the bottom, these are largely wiped out, suggesting, again, we're getting the siRNA out of the vasculature into the brain tissue and the relevant target cell type to reduce the target RNA. And looking at duration of action in the monkey, now here we're zeroing in on those disease-relevant brain regions, the hippocampus and the frontal cortex. And this is the same dosing regimen where we have three weekly subcutaneous doses of 3 mg per kg, and then we look at two weeks, four weeks, and 12 weeks after the last dose. So the mRNA level for the two tissues has solid lines, and the protein has the dotted lines.
And you can see the RNA reduction has reached a nadir already at two weeks post-dose, and it's gradually rebounding over three months, but still maintaining 50% reduction in target RNA. And the tau protein fall is a little delayed. It has a long half-life, but so the nadir is more a month after the last dose, but there's a similar magnitude of effect and durability. Now, we've taken the data from this study to map out- to model out a projected clinical dosing regimen. Now, on the left is the PK/PD relationship we've seen for the MAPT RNA versus tissue concentration in a panel of CNS tissues. It's a very well-behaved relationship, with tissue concentration increasing with additional doses and rebound of the MAPT mRNA level as the tissue levels rebound over time.
So from this, we calculated an IC 50 in CNS tissue of 270 ng per gram and the tissue half-life around five weeks. Now, for modeling, translating this into human, we expect a longer half-life, as we've seen for siRNA and other target tissues. And so when we incorporate that allometric scaling, we arrive at a quarterly projected dosing regimen, which should be able to maintain RNA knockdown in the range of 50%-70%. And that's what's shown on the right side graph, where the predicted tissue concentration is plotted in blue and mapped to the left side axis, and the corresponding MAPT mRNA level is in red and mapped to the right side axis. And the blue zone that's highlighted is the optimal target knockdown level, 50%-80% for MAPT RNA.
You can see the red line is staying in that zone. This program's currently in preclinical studies to enable a CTA filing in the second half of 2025. We've done exploratory tox in non-human primate and transferrin transgenic mice at up to 10x an efficacious dose, which is supportive of further development. The data we've shared show that the potent and long-lasting activity of this compound supports a clinically relevant dosing regimen, and the formulation volume is also clinically feasible. ARO-MAPT subQ has potential to be a disease-modifying therapy for Alzheimer's and other tauopathies, with potential for broad uptake in this large patient population due to the convenient administration. Now I have a couple of additional programs that we're excited about using this systemic platform. First, Huntington's disease.
Huntington's disease is the most common monogenic neurodegenerative disease with in the world, with a prevalence of about one in ten thousand in the U.S. and Europe. The symptoms include motor dysfunction, cognitive impairment, and neuropsychiatric problems. The median survival after onset of motor symptoms is fifteen years. Typically, onset is in middle age, although there's a number of genetic modifiers that can influence that, and there are no disease-modifying therapies available. Now, the molecular cause of the disease is an expanded CAG repeat in exon 1 of the Huntington or HTT gene. This results in production of a protein with an expanded polyglutamine tract, which can aggregate and cause neurotoxicity. This neurotoxicity starts in the striatum, which is a region located deep in the brain, which has been difficult to reach with intrathecal delivery.
So huntingtin protein-lowering approaches have been tried with intrathecally delivered ASOs. But as I'm saying that they likely have low delivery to these deep brain regions, at least at the dose levels tested clinically. So at dose levels required to reach meaningful tissue concentrations in the striatum after lumbar puncture, there's significant drug accumulation in the spinal cord, which increases potential for safety problems. But with the CNS TRiM subQ platform, on the other hand, as Tao is pointing out, we don't have that challenge. We have this, with this more uniform brain distribution that we're able to achieve, we're able to fine-tune the optimal amount of knockdown throughout the CNS without overloading the spinal cord with drug. Now, some of the preclinical data we've generated first in the mouse.
On the left, we're looking at a Huntington's disease mouse model with our lead candidate, siRNA. The YAC128 Huntington's disease mouse model has a human Huntington transgene, which has 120 CAG repeats. In this study, the compounds were administered intrathecally with our first-gen platform, the lipid conjugate, just to look at the inherent potency of the compound, the siRNA sequence. We're measuring human mutant huntingtin protein in the striatum, and you can see with the lead candidate in blue, we have a nice dose-response with maximal knockdown of 75%. In the middle, in green, we have an exon 1-targeting siRNA, which you can see shows limited efficacy against the mutant Huntington RNA.
You may remember that the CAG repeat expansion is in exon 1 of the HTT mRNA, which might affect its structure and accessibility to exon one targeting siRNAs. So we see some challenges with the exon one targeting strategy, as they might show less efficacy as the repeat sequence expands, which is the biological context when you need it the most. And on the right in red, we've compared an ASO with the sequence and chemical modification pattern of tominersen, which has been tested clinically for Huntington's, and you can see that it is also less potent compared to our lead siRNA sequence. Now, on the right, we're looking at that same lead siRNA using a systemic delivery platform.
First, on the left side, we're using a surrogate mouse transferrin ligand in the YAC128 mice, and the dosing regimen is four daily doses of 3 mg per kg, subcutaneously administered. One month post-dose, you can see we have about 50% protein reduction in the mutant human huntingtin protein in these representative brain regions or CNS regions, cortex, striatum, and spinal cord. Now, on the right, we're using the human transferrin ligand. This is the actual clinical candidate with the transferrin ligand that will move into the clinic. Here, looking at activity against normal endogenous mouse huntingtin in the human transferrin receptor transgenic mouse. The siRNA is cross-reactive with the mouse gene, and we see a similar or better level of protein knockdown in this experiment.
So we're breeding these transferrin ligand, or sorry, transferrin transgenic mice with the YAC128, so we can do the disease model efficacy study with the clinical candidate. But that's not ready yet. So we've also performed a dose response in non-human primate, where we gave three weekly doses subcutaneously of point three, one, or three mg per kg. And we're looking here at day 43, which is one month after the last dose. We're looking at huntingtin protein levels, so monkey huntingtin protein. And in this set of tissues, cortex and striatum, putamen and caudate are the disease-relevant deep brain regions. We see dose-dependent reduction of huntingtin protein, up to 80%, 75%-80%. So this program is also moving into development. We're looking forward to a CTA filing in the second half of 2025.
It has the potential to address the root cause of Huntington's disease and achieve best-in-class levels of target engagement in the disease-relevant brain regions. Finally, we're also moving forward with an alpha-synuclein targeting therapy with this platform. Toxic aggregation of misfolded alpha-synuclein, which is an abundant protein in several cell types encoded by the SNCA gene, is at the root cause of a collection of neurodegenerative diseases. That includes the prevalent, very prevalent Parkinson's disease and Lewy body dementia, as well as the rare disorder, multiple system atrophy. As the accumulation is intracellular, just like tau, it has been difficult to go after alpha-synuclein with immunotherapies. But reduction in alpha-synuclein RNA to reduce protein production has potential to be disease-modifying.
And also with different conformations of alpha-synuclein associated with different synucleinopathies, again, by targeting the RNA, we'll be able to treat all potential synucleinopathies with one siRNA. This program's at an earlier stage. We're just in the process of identifying the final candidate now, but this is just data showing proof of concept for alpha-synuclein targeting in the mouse with the systemic delivery platform with one of our lead candidates. After four daily 2.5 mg per kg subcutaneous doses, we looked at different time points post-dose to find the nadir and monitor the rebound of SNCA target RNA. In the cortex and brainstem, we see maximal 85% reduction in SNCA at one month post-dose, which returns to baseline over the course of four months.
This program's progressing well, and we expect to select the candidate by the end of the year. So to summarize, we're very excited about the profile of the CNS subQ TRiM platform, which we've shown can achieve deep knockdown of multiple targets in brain tissues at clinically relevant dose levels. RNAi is the ideal therapeutic modality to treat many neurodegenerative diseases, and this platform advance is critical to enable broad delivery to patients. The platform also enables us to reach areas of the brain that weren't feasible with intrathecal delivery. So we've shared just the first couple of programs moving into clinical development, and we're looking forward to seeing these programs reach patients at the end of next year. So next, we're going to hear from Dr. Soria about the unmet need in Alzheimer's disease and how the tau-targeting therapies can fit into the landscape.
So, Dr. Soria, thank you so much. The floor is yours.
Thank you, and thank you for the opportunity to discuss this topic with the audience. I'm a practicing neurologist and director of clinical research, and the topic is very relevant to the patients we see on a daily basis in clinic. I work directly with in the field in a number of companies trying to bring to development disease-modifying therapies in the field of Alzheimer's disease, and also receive funding to conduct clinical trials to bring these therapies to the clinic. We'll discuss today the definition of the Alzheimer's disease as it is evaluated in clinic. We'll go over some aspects of care and treatment, and then the significant unmet need when it comes to targeting tau clinically and bringing this therapy to clinic, and the challenges and opportunities as the field is evolving rapidly.
So we'll first go over the definition of dementia. I think this is important to interpret the data that we see in clinic. Dementia refers to a decline in cognition that is subsequently accompanied by direct associated decline in function. And to meet the definition for dementia, there has to be at least two cognitive domains that are affected, and that can include memory, language, and other domains. The definition of Alzheimer's disease dementia, you'll see frequently the term probable AD dementia or probable Alzheimer's disease dementia, and that refers to a kind of dementia that has an insidious, slow onset with a reported decline in both cognition and function. And the most common manifestations include short-term memory loss. Often, there can be other manifestations present.
You might also see at times the definitions, clinical definitions that include possible AD dementia, and that refers to other presentations that are not so typical. They can be more rapid in onset and decline, or there can be other significant comorbidities that affect the diagnosis. So typically, you'll hear and you'll see clinically the word possible AD dementia. Alzheimer's disease is the most common form of dementia in patients seen in clinic, accounting for close to 60%-80% of the cases. And in over 50% of the cases of Alzheimer's disease dementia, there's more than one pathology. So you'll typically see Alzheimer's disease plus vascular disease or Alzheimer's disease plus synucleinopathy, as we just heard, and that includes Lewy body dementia and other forms of Parkinson's.
Other less prevalent comorbidities include hippocampal sclerosis and chronic traumatic encephalopathy. So typically, comorbidity and mixed pathology is the rule. The two main pathologies underlying the diagnosis of Alzheimer's disease include beta amyloidosis, so the presence of amyloid beta, and the spread of neurofibrillary tangles. And on the left of your screen, you'll see the figures indicating phase one, two, three, four, and five, and this is a description of how amyloid is deposited in the brain according to what's called the Thal phases. And it's simply meant to represent and show how amyloid first appears in the cortex and then spreads other adjacent areas in the cortex and into other subsequent subcortical and even brainstem regions over time. This happens close to fifteen to twenty years before the onset of clinical decline.
In the middle, you'll see the neurofibrillary tangles or the Braak stages of how tau spreads throughout the brain. This is more significant, more relevant to the clinic because this directly correlates with the cognitive, the onset and the progressive cognitive decline. These are cross-sectional. These are coronal slices, post-mortem slices, looking at tau, phosphorylated forms of tau. It's meant to represent and show the onset of tau pathology, initially in the medial temporal lobe, in the entorhinal cortex, and then, in the subsequent spread throughout the temporal lobe and into lateral and superior temporal lobe regions.
The important point about this slide is that tau pathology spreads in such a way that by the time they're spreading into the temporal lobe laterally, and as the tau pathology is about to leave the temporal lobe, that correlates with decline, cognitive decline, particularly functional decline, and we'll see that soon in other slides as well. By the time a patient has severe dementia, significantly impaired in severe stages of the disease, there can be quite a spread of tau throughout the entire cortex. We want to intervene early at a time where tau is still within the temporal lobe, ideally, and if it's spread to other regions, still be able to intervene and stop the spread throughout the cortex, the cortical regions.
These neuropathological diagnoses have been updated over the years for Alzheimer's disease, and it's important to mention that it does now include descriptions about how to evaluate other comorbid pathologies, as many patients are typically found to have other comorbidities, including synucleinopathies and vascular disease as well. In the left side of the screen, you'll see the clinical stages of Alzheimer's disease, starting with a preclinical stage of the disease, where patients can show biological changes of Alzheimer's disease, the presence of amyloid changes and tau changes, but they do not show clinical symptoms, and then the first clinical stage of the disease, it's referred to as mild cognitive impairment, and that's a stage where patients have cognitive decline, but they do not have significant functional changes. Subsequently, after that, you'll see more mild or moderate dementia stages of the disease.
And in this context, dementia refers to the loss of independence, the decline in function. And tau is more directly correlated with these transition points of mild cognitive impairment to mild dementia, and then especially from mild dementia to more moderate stages of dementia, for which we do not have direct disease-modifying therapies at this time. On the right side of the screen, you'll see the biological correlate of these transition points, starting with a blue line, referring to the changes in amyloid that starts in spinal fluid about fifteen to twenty years before the onset of cognitive decline, and then subsequently, tau changes in the green line.
This refers to the ongoing tau changes that more closely correlate to the decline that one sees in clinic, to the actual clinical changes that the patient shows. It's important to know the recent revisions of the definition of Alzheimer's disease. It was recently updated by the Alzheimer's Association Working Group just this year. And there are two reasons why this happened, and that one is the need to classify and define Alzheimer's disease as a biological process defined by the onset and change in amyloid and tau. And there were two practical reasons why that happened. One is the existence of blood-based biomarkers in clinic, which we now have, for which we now have commercially available assays, and the existence of disease-modifying therapy that are now making it into clinic.
One last component of the criteria, and it's very relevant to our talk today, is the setting up the framework for biological staging of Alzheimer's disease, and that is, if you look to the table at the bottom, that is beginning to stage patients biologically according to their tau burden, especially how much tau they have in the temporal lobe regions and the spreading of the tau outside of the temporal lobe region into neocortical regions, and this, the proposal is that this could be done with both tau PET, and they can be done also with fluid biomarkers as well. Those tau-specific fluid biomarkers are slowly... I suspect that they would also slowly make it into clinic as well.
The reason for this is, one, would one have a very clear biological window for intervention that would clearly correlate to those transition points in clinical decline for patients. This is another table in the definition that tries to correlate clinical and biological decline. And for the purposes of this talk, clinical stage III refers to mild cognitive impairment, and clinical stage IV, V, and VI is dementia. And on your left, you see the biological stages, early, intermediate, and advanced. So essentially, a big significant unmet need is that being able, needing to and being able to stop transition points from IIIc to 460 .
So being able to slow down disease and stop it as it is transitioning from a mild cognitive impairment stage into mild dementia, but then especially later on, transitioning into mild dementia to more moderate and dementia stages. As you, as we know, and as has been mentioned, Alzheimer's disease is a chronic, indolent fatal disease. We have close to seven million patients affected in the United States. It's likely closer to ten million if one includes patients with mild cognitive impairment and other even preclinical stages of the disease.
In the world, there are close to 55 million patients affected with the disease, but I suspect these numbers are likely low compared to the reality on the ground, as the diagnosis is still very much dependent on neurological access to healthcare and diagnosis by clinicians. In clinic, as of now, in clinical practice, patients are still diagnosed clinically based on an interview and cognitive testing. The clinical diagnosis itself doesn't directly require biomarker testing. It is made again, based on that interview in clinic. There's a stereotypical pattern to the cognitive profile that patients show. Patients typically have a short-term memory loss and associated difficulty naming and diminished semantic fluency. A lot of this happens in clinic on frequent visits, and the clinical diagnosis is made like this.
However, with the existence of biomarkers, this has now been incorporated, at least there have been attempts on several guidelines to incorporate this into the clinical practice and slowly make it into clinic. Now, my patients do undergo the clinical evaluation, and they're given a clinical diagnosis, but then they're subsequently confirmed to have Alzheimer's based on biomarkers. And as of now, what we have available and we've had available is there are PET imaging that's done, spinal fluid analysis, but also blood-based biomarkers, which are now, again, present in clinic with significant quality and now in accuracy. And that there's a lot of conversations about what's the level of sensitivity and specificity that would be required for scaling of this biomarkers in clinic.
One last note here is that as of now, there's still biomarker testing in individuals who are unimpaired is not recommended. So a lot of the guidelines are going into diagnosing and confirming patients with essentially clinical stages of the disease, specifically those, and ideally those also transitioning through the different dementia stages for which we have a high unmet need. There have been challenges bringing some of the new therapies to clinic. We in Southern California in the U.S. have a large site that have been bringing the most recent therapies, and we've been learning along the way how this is received by the public, by patients, and by families. And we've just recently also published some of this as of today in our Neurology Live magazine.
A lot of the times when we try to give the education to neurologists, to primary care practitioners, about this, we have to essentially go back to the basics, go back to the clinical symptoms. What is the value added to patients? What is the stages of disease that we wanna target? And that's been the part that's been most important. I mention this because most of the patients as of now that we see in clinic are in a mild to moderate dementia stage of disease. And those are not the patients necessarily receiving the newly available anti-monoclonal, anti-amyloid antibodies.
We still have a large population of patients, I would say over 50% of the patients that we see overall, who would not be a candidate for anti-amyloid therapies just based on disease stage alone. This is a study back in 2018 by my mentor, Dr. Doug Galasko, but it's essentially trying to point attention to the idea of the right patient, the right time, the right therapy. And as you can see in this graph, amyloid therapies are meant to be directed before the onset of cognitive decline. And again, this is not by definition most of the patients that we see in clinic. By the time we see patients in clinic, by definition, most of the times they're symptomatic.
As you can see in the graph, it is blocking extracellular tau spread and tau clearings that is the more relevant to clinical symptoms as these are the time window, the biological time window that can be targeted by tau therapies. Of course, there'd be other needs to also target cellular dysfunction, metabolic abnormalities and inflammation that comes along with tau as well. Referencing the monoclonal antibodies recently and how they relate to tau pathology, looking at the Clarity AD study, the lecanemab study over 50% of participants in the lecanemab study especially those that had low tau or no tau had no decline in measures of cognition and function at three years. This is part of their open label data on the anti-monoclonal antibody lecanemab.
What essentially this says is that the therapies, the anti-amyloid therapies, work mostly in a very early stage of disease when very little tau is present, if any. There's also similar data coming through in the TRAILBLAZER-ALZ 2 phase III study for donanemab, which is the second antibody that received traditional approval. It's likely gonna be the case again, that those patients with less tau tend to perform well better compared to those at higher burdens of tau. We still have a high unmet need for those patients that have more higher tau burden or more advanced in the disease process, particularly in the mild to moderate stage of the dementia as well. Just a brief note that anti-amyloid antibodies do have some effect, a modest effect on tau.
For example, lecanemab show a slowing of a spread of tau in the temporal lobe regions over 18 months. But again, these are modest effects, and donanemab similarly had effects of slowing of tau in the temporal parietal and frontal lobe regions, but to a modest extent. We need to be able to do this to a significant extent in a shorter timeframe to be able to help patients. So how are tau therapies expected to be different than the current treatment? They're again like more likely to be efficacious once the cognitive decline begins. Tau targeting tau pathology, and once it's done effectively, it's expected to shorten the time needed to reach the primary endpoints in the clinical trial.
Ideally, the clinical trials are gonna be shorter, and the participant enrollment requirements are likely to be less. That's my prediction, even looking at the recent phase II studies on tau. Once again, but we'll need an effective tau therapy to make this a reality. And lastly, tau therapies are not expected to be associated with the amyloid-related imaging abnormalities, that is, brain bleeding or brain swelling, that can often be seen in patients receiving anti-amyloid therapy. The fundamental difference about tau in clinical practice is that it ideally will be used in patients who are more advanced in the disease process, and this is very significant, very relevant for daily practice. And again, this is the largest majority of patients we see with Alzheimer's disease in clinic.
It would require leveraging, identifying the biological window, so especially tau PET imaging will be needed going forward and other fluid-specific measures of tau. It would also allow treatment of patients who do not currently have access to anti-amyloid therapies. For example, patients with high vascular burden of disease and cerebrovascular disease. And also open treatment options for patients who are APOE4 homozygotes and receiving anticoagulants, who are not often included in clinical practice in the anti-amyloid therapies. And then it's likely that tau would play a role as a follow-up sequential therapy for many patients who have already received anti-amyloid therapy. Those patients are called TRAC or treatment-related amyloid clearance. And as the years go forward, you'll see many, many of those patients finishing amyloid therapy and then needing to continue as a kind of maintenance therapy.
So the idea would be induction therapy and then maintenance therapies for those patients coming early. For those, coming early into the clinical system, for those that come in in the middle, in mild dementia to moderate dementia, they'll likely go on to receive anti-tau directed therapy directly. And so, in summary, the Alzheimer's disease definitions have evolved to include the use of biomarkers and identifying patients early. The early clinical detection starts with a mild cognitive impairment stage, and tau correlates with a decline that one sees in clinic, is a most directly correlated. Treatment options are needed for patients to mild or moderate, dementia, those transitioning to those stages, especially when tau is in the temporal lobe and about to leave the temporal lobe into neocortical regions.
And considerations in clinic will include the right patient, the right time, looking at the biological window, and then the mechanism of action, the route of administration, and the safety, the risk, benefit-safety profile of combination therapy going forward. Thank you, for your time, and I'm gonna pass it back to Vince.
Thank you so much.
Okay, so there's a lot in this, and I know that this is the first time that we've spoken publicly to any degree on the CNS platform, really, and certainly the systemic delivery to the CNS. So there's a lot to digest here. You know, in my mind, these are kind of the key takeaways. These are early programs, but we think that they're really promising and really important. We now have the ability to administer the CNS with two different routes of administration. Intrathecal, which as Christi showed, gets broad distribution to the cord and to key regions in the brain.
The subcutaneous or systemically administered program, we think gets better distributions to the deep brain regions, which is critical to some of the targets that we've talked about today. We think Arrowhead has a growing pipeline in space. There's lots of additional interesting targets, and to date, our data continue to be compelling across the board. The targets that we're going after have historically been difficult to target, for one reason or another. And so we think that they're ripe for innovation. And we see Arrowhead having a first-mover advantage, particularly with the BBB systemically administered platform. And finally, we see siRNA as a really promising mechanism for many CNS targets. siRNA is...
The hallmarks are high specificity and long durability of response, which is critical for a lot of these targets. So where are we? We talked about four individual programs, and these are roughly in order, you know, of how far they are along. The first program, ARO-ATXN2, as James mentioned, we've already filed the CTA, and we intend to dose the first patient in the first quarter of 2025 . I believe screening actually starts next month. ARO-MAPT and ARO-HTT, we're on schedule to have CTA filings in the second half of next year, and then likely start clinical studies thereafter.
And then ARO-SNCA is the newest program that we're looking to nominate the candidate before the end of this year, and then we'll provide a little bit more guidance later on timing for that. But where our pipeline is, again, this is the first time we've talked about it, but it's already relatively broad in critical diseases with substantial unmet need, where legacy therapies have not been effective, and have not moved the needle. And so we're really trying to do something critical here for so many patients with neurodegenerative disease, with really interesting technology that's well suited for this purpose. So that's how we see our CNS portfolio. Now I'm gonna open the call to some questions.
Let me take a moment just to compile. Now, we only have about fifteen minutes for questions, so I won't be able to get to all of them, unfortunately, but give me a moment to compile some questions. Okay, our first question is from Patrick Trucchio at H.C. Wainwright, and his question is: "Can you discuss the level of confidence that the target engagement you are seeing in the MAPT program in NHPs should translate to humans, and achieve a 50%-80% level of knockdown cited during the presentation?" So I guess I will ask Christine and James on that one.
Yeah, I can, I can take that one.
Christine.
Yeah, I think, as I mentioned, the affinity of our transferrin ligand in the monkey is similar to human, and so we're making some assumption that that's gonna correlate and be translationally relevant. I think the translatability of other transferrin targeting approaches from companies like Avidity and Dyne, they have translated well, so I think we have good reason to expect that.
Thank you. Next question comes from Andrea Tan at Goldman Sachs. The question is: "Given tau aggregates are found in the extracellular space, do you think targeting intracellular tau is sufficient to attenuate tau burden?" Oops, sorry, it just switched. Okay, that's probably enough of the question together.
Yeah, yeah.
Maybe I'll ask Christine and also Dr. Soria to opine on it.
Yeah, I would just say that the neurofibrillary tangles are actually intracellular, and there is that transmission throughout the brain that a lot of the immunotherapies have been aimed at, trying to catch it as it's traveling from cell to cell. But the tangles themselves are intracellular, and so it could be the reason why these immunotherapy approaches haven't been successful.
So similar, following along with that, is the one benefit of targeting the intracellular neurofibrillary tangles. Up to this point, there was a belief in the field that this could not be done, and there've been an assumption that it was not gonna be possible. But as it was mentioned earlier, we're getting a glimpse that it might be possible to not only stop the spread of tau transsynaptically, but also reduce the intracellular burden of neurofibrillary tangles. And this would be a very significant to be an understatement to be able to do this at large scale, because it would-...
Clinically, one would expect that there would be. It would open up the field to be able to do other things as well, once the neurofibrillary tangles burden can be reduced as well, and assuming that it can be done at a time where damage has not been completed, but I think just with what we know so far and the data that we've seen from the trials that have been done in human data, it clinically would be very significant to reduce the intracellular neurofibrillary tangles.
Thank you.
Next question is from Maury Raycroft at Jefferies, and this is also for Dr. Soria. "There will be some tau clinical data reported at the upcoming CTAD Medical Conference. Is there anything in particular you will be looking for in the data updates? Also, what are the trade-offs of an antibody approach that targets unique regions of tau versus an RNAi-based approach?
I'll take that question. I'll start. What I would be looking for in the CTAD conference is the effects translating some of the reduction in neurofibrillary tangle effects that have been seen in tau PET, translating that to clinical data, as it was mentioned earlier, to see how much of the reduction in tau PET translates into clinical benefit. Knowing that patients once they're having neurofibrillary tangle spread, they're already in a mild to moderate dementia stage of disease, and you would wanna confirm the extent to which reducing those intracellular neurofibrillary tangles correlated to a slowing of decline. Ideally, that would be something that should be appreciated, given how fast patients change in that stage of disease.
So my hope is that we'll see data regarding that. My hope is that we'll also be able to answer some questions about when is the right time for tau, which is still to be decided. Perhaps we're still a little bit too early, and even the therapies could be used slightly later in the disease process. With regards to the intracellular targeting versus the immune therapy, antibody-mediated therapies, the trade-offs and the benefits is that you would wanna not only be able to inhibit tau spread, but you would want a way to also melt away the neurofibrillary tangles inside the cells.
If siRNA therapy has a way to get us there, it would essentially let us go further, deeper into the clinical stages of disease, in more moderate and even severe stages of the disease. Whereas passive immunotherapy, antibody-mediated therapy, might only be able to slow us down or block us where we are. So, you know, the answers there are still to be decided. Also, the dosing as well, and, you know, the extended dosing of the siRNA therapy compared to immune therapy as well will be a difference in clinic, as is expected to be less frequent with siRNA compared to antibody therapy.
Thank you. Christi, you have anything to add on mechanism of antibody versus siRNA?
Yeah. No, I think Dr. Soria described that very well. I think by accessing the intracellular tau and reducing the tangles, we have so much more potential to impact disease than the antibodies.
Thank you. Next question is from Mayank Mamtani at B. Riley Securities, and this is regarding the BBB platform. "What is the ratio of CNS versus peripheral penetration, and do you see greater parenchymal distribution? And could you speak to the extent that deep brain structures are reached?
Yeah, I think the, I'll see if I can remember all those questions. But in terms of the deep brain structures, you know, in tissue accumulation and knockdown, you can see in the data that we shared, that it's pretty similar across the brain, which is, you know, coming from the blood side, the vascular side. That makes sense. What was the first question again?
How much gets into the CNS and into the brain?
Oh, this is the peripheral tissue.
... and how much is it peripheral? Right.
Yeah. I mean, the usual organs of accumulation in the periphery are still accumulating. But so there, there's no surprises in the peripheral accumulation. You still get some compound to the liver.
Yeah, I want to add a couple points, that is, yeah, I agree. I mean, we have seen a peripheral tissue accumulation in liver is most. But the liver doesn't have much of a transferrin receptor one. Instead, it has transferrin receptor 2. So even accumulate there, we actually don't see much RNA silencing effect in liver.
Thank you. Next question is Patrick Trucchio at H.C. Wainwright. And this is with regards to ARO-MAPT. "With half-life of five weeks, what data support quarterly dosing? And is 50% mRNA knockdown enough to achieve benefit in symptoms?
... Yeah, so I if you just model it in the monkey with no allometric scaling, monthly dosing is supported. So I would say monthly is the minimum profile that we would expect, which for a subcutaneous administration, is completely acceptable. With the expected extension of half-life in human, that puts us potentially a quarterly, but monthly is also fine. And what was the second question? Sorry.
and is 50% reduction-
Oh, yeah.
Would that be a clinically meaningful level?
That's an open question. Biogen, I mean, Biogen's doing the study to tell us that, but that's about what they've achieved, is 50% reduction in tau in the CSF at least. You have to take into account that's an intrathecal therapy, and so the different brain regions will have different distribution compared to what we would be able to achieve, which would be a little more evenly distributed across the different brain regions. But that's an open question. In the preclinical models, certainly it is enough.
Thank you. Next question is Jason Gerberry from Bank of America: "Can you detail the market opportunity for ARO-ATXN2, based on how prevalent the SCA2 genotype is in the all-comer SCA population?
Yeah, I think, so this would be certainly a rare disease. I think we described the prevalence in the slide. And the therapy that we're developing should be applicable to all SCA2 patients, not any of the other SCAs, but at least it's relevant to all SCA2 patients. So, you know, we're still early in terms of the commercial assessment, but if you consider rare disease orphan pricing for a population, let's say, you know, four to five thousand in the major pharmaceutical markets, you know, you can sort of back out the overall market opportunity for that.
Thank you. And we have time for one last question. Comes from Ellie Merle at UBS: "What are your considerations in selecting a lead candidate for the ARO-SNCA, SNCA-2, I'm sorry, SNCA mRNA, and have you determined how much numerical reduction is correlated with clinical improvement?
Yeah, so we have several candidates being put head-to-head in a monkey study right now, which is just reading out, so that will be the final basis of the decision. In terms of the level of knockdown required, it's the same, you know, 50% reduction, certainly in a preclinical model is enough. How it's gonna translate into clinical practice is to be determined, because no one's ever had a tool that could knock it down the way that we're going to be able to do that, so this is going to be really exciting, cutting-edge clinical research.
I know it's not satisfying, but I think that that's the answer or that's the question we get all the time for every new novel target: How much do you need? And I think that, well, the clinical data will show that. The good news is that, you know, we feel like we have some room here. We feel like we're getting a better level of reduction of the target than other methods. And Christi showed, you know, plenty of the ASO data on similar targets, and we feel like we're exceeding that. And so we feel like we've got a pretty good chance here. But the bottom line is we need to do the clinical studies and do the experiment to get the answer there.
So that's all we have time for, so I wanna say thank you to Dr. Soria, and to Christi, and James, and Tao, and to everybody who watched it today. Thanks so much for taking the time.