Do you have a question you'd like answered on a future episode of Translating Proteomics? Send it to translatingproteomics@nautilus.bio!

On this episode of Translating Proteomics, host Andreas Huhmer discusses advances in Alzheimer’s research with special guest and Curie Bio Drug Maker in Residence, Sarah DeVos Ph.D. Their conversation focuses on:

• The impact of molecular diagnostics on Alzheimer’s research
• Recent Alzheimer’s drug approvals
• The future of Alzheimer’s research

*Small edit on Sarah's background - She did her graduate work at Washington University in St. Louis and a Postdoc at Massachusetts General Hospital*

Chapters

00:00 – Introduction

01:54 – Why Sarah began studying Alzheimer’s

03:39 – Current tools and needs for future Alzheimer’s diagnostics

09:52 – Recent drug approvals in the Alzheimer’s space and their relationship to diagnostics

14:26 – Is it possible to develop biomarkers that detect Alzheimer’s at its earliest stages?

16:36 – What is limiting the development of new Alzheimer’s biomarkers?

17:51 – The DIAN trials and learnings from studying dominantly inherited Alzheimer’s

19:33 – The genetics of Alzheimer’s

22:19 – Novel approaches to identifying and understanding Alzheimer’s pathology 

25:54 – Where can proteomics advance Alzheimer’s research?

31:25 – The role of proteomics in Alzheimer’s animal models

34:33 – Sarah’s hopes for the next 10 years of Alzheimer’s research

41:39 - Outro

Resources

Dominant Inherited Alzheimer’s Network (DIAN) trials research updates

o   In the DIAN trials, researchers work with families to study various clinical and basic science aspects of dominantly inherited Alzheimer’s disease.

Amyloid plaque reducing clinical trials:

o   Two Randomized Phase 3 Studies of Aducanumab in Early Alzheimer's Disease (Haeberlein et al. 2022)

o   Donanemab in Early Symptomatic Alzheimer Disease - The TRAILBLAZER-ALZ 2 Randomized Clinical Trial (Sims et al. 2023)

o   Lecanemab in Early Alzheimer’s Disease (Van Duck et al. 2022)

Blood Biomarkers to Detect Alzheimer Disease in Primary Care and Secondary Car (Palmqvist et al. 2024)

o   Clinical research into a new phospo-tau biomarker that can help physicians more effectively diagnose Alzheimer’s disease

Resurrecting the Mysteries of Big Tau (Fischer and Baas 2021)

o   Review covering a potentially neuro-protective form of tau called “Big tau”

Integrated Proteomics to Understand the Role of Neuritin (NRN1) as a Mediator of Cognitive Resilience to Alzheimer’s Disease (Hurst et al. 2023)

o   Paper linking the NRN1 protein to cognitive resilience in Alzheimer’s

o   Nautilus blog post about this paper

On this episode of Translating Proteomics, host Andreas Humer of Nautilus Biotechnology discusses advances in Alzheimer's research with Special guest and CurieBio drug maker in residence, Dr. Sarah DeVos.

Their conversation focuses on the impact molecular diagnostics have had on Alzheimer's research, recent Alzheimer's drug approvals, and the future of Alzheimer's research. Now here is Andreas to introduce Sarah.

Welcome back to Translating Proteomics. In today's episode, I have the great honor of speaking with Special guest Sarah DeVos.

Sarah has over 15 years experience in translational research focused on neurodegenerative diseases. She was a post drug fellow at Washington University in St. Louis as well as the Massachusetts General Hospital before she joined Denali Therapeutics.

And recently she left Denali Therapeutics to join Curie Bio as a drug maker in residence. Sarah shares that one of her driving forces in her career is a desire to do work that has meaningful, positive impact on patient lives.

With that in mind, I'm very excited to have Sarah on the podcast to discuss Alzheimer's biomarkers in the context of clinical practice. In this conversation, I hope we'll cover how biomarkers are used in Alzheimer's diagnosis, treatment, and clinical trials.

What additional information we need from new Alzheimer's biomarkers, and what role does proteomics play in making new ultramarkers and advancing patient care. With that, I welcome you, Sara, to the podcast.

Thank you very much for having me.

So let me start out asking you, how did you get into the field of neurodegenerative diseases? What attracted you to this field of science?

Yeah, so for me, it basically stems from family. So I'm growing up. Everyone in my family gets Alzheimer's disease. And so for me, it was very like, why does that happen?

I grew up in a small farm town in Michigan. It was almost a way of life for my family. Everyone just accepts it. And so for me, growing up, I wanted to see, can I change that, essentially.

So it was fascinating from, like, a science perspective. And so I think when I was in middle school, I decided I wanted to go into research for Alzheimer's disease.

No one in my family had, like, gone into science at all. So it was entirely, like, untreaded. And when I did undergrad, serendipitously found a lab that did tau research for Alzheimer's.

And then I just kind of snowballed and just kept going with it because it was really, I think, looking at tau and its correlation to cognitive decline in Alzheimer's disease, it's such a direct Correlation that, looking at that, then I basically just made my career out of studying that part of Alzheimer's disease.

So the tau pathology piece, and then I did my grad school work in it, my postdoc denali, I worked on TAO because it to me felt very druggable, very directly related to the disease. And Alzheimer's impacts so many families throughout the US throughout the world.

But for me, I wanted to study it in the context of my own personal family.

So for you, this is clearly very personal.

I think, as you already mentioned, the number of people that suffer through neurodegeneration is going to increase dramatically over the next few decades. So given the tools we have today, do you think we have sufficient tools to sort of, first of all, diagnose Alzheimer's?

And how do you see this moving forward?

I think, yeah. So the advent of biomarkers and Alzheimer's disease changed everything, right?

It changed research, basic research, it changed drug discovery, drug development, clinical trials. And so a lot of times I draw the parallel with oncology, right?

So when you think of oncology, you think of being able to image the tumor, to be able to look at it in real time and track it in real time. And that enables the advent of a bunch of therapeutics for oncology. For Alzheimer's disease. I like my blood brain barrier.

I like that it keeps the things out, but it also makes it very difficult to then be able to assess in real time what's happening in your brain. So being able to, for example, with PET pib, right, so Pittsburgh compound B binds to amyloid beta plaques.

That alone changed the entire game of how we think about Alzheimer's disease, clinical trials, being able to properly enroll patients. So for a long time, we're running Alzheimer's trials because dementia is a large umbrella.

Alzheimer's is the largest part of that umbrella, but it's not the whole thing. So if you're going to run a trial, you want to make sure that you have participants that actually have the pathology of interest.

Otherwise it's going to be a moot point. Your data is going to be super messy.

It's going to be very variable because you're not even recruiting people who have Alzheimer's, you're recruiting people who have other types of dementia. And so that single biomarker alone changed the whole game for Alzheimer's disease. And now we're be getting a bit more sophisticated.

We're studying tau. We have a really good tau PET ligand to be able to monitor tau pathology in the brain.

Because for Alzheimer's you have to have both amyloid and tau pathology in order for it to be properly Alzheimer's. But for a long time, and even when I started in the field, the only way to actually diagnose someone with Alzheimer's disease was at autopsy.

And at that point you're obviously too late.

Arguably too late.

Yes.

Sorry.

I recently was at the Association Alzheimer's Association's international meeting and you couldn't but notice that there, you know, it was all about biomarkers.

I mean, a lot of companies have jumped into the space, but double clicking on, you know, the availability of biomarkers a little bit, there's still some very major challenges. I can imagine.

I mean, one of the key diagnosis or key diagnostic tools that you mentioned is the PET scan, but it's expensive and it's certainly not very accessible in some cases. So how have biomarkers, the blood based biomarkers, helped us and maybe what are the limitations for those?

Yeah, I 100% agree. Right. PET imaging is great from like a larger macro scale, it's good to get started. It allows us to look into the brain.

You can see the basically spatial resolution of where the pathology is. But it's incredibly expensive. It is not accessible throughout the world.

And so to be able to do a plasma pull and be able to analyze that, it just becomes way more tractable. Right. So now we have scale, now we have access. And so we're moving in a good direction in terms of both amyloid beta and tau. Right.

So for amyloid beta, perhaps a bit counterintuitive, but you actually see a reduction of a beta in the blood. It's because your brain is a sink.

So it just holds it for tau, I'm sure you saw, right, phospho tau217, huge increase in interest because there is a more direct relationship between tau pathology, neuronal death, cognitive decline. That's a really nice axis, amyloid beta.

You can have people who have a brain full of amyloid beta and they're cognitively normal, so you need both of them. And tau is the one that's a bit closer to the actual neuron dying.

And so for a long time the field has really been searching hard to find a blood based biomarker because you can do CSF poles as well. But that's also very cumbersome. People don't like to do it.

And so it's uncomfortable, not painful per se, but it's a scary thing, right, to have a needle go into the lumbar space and pull csf. So by Making that switch to blood based biomarkers, you can get so much more data, right?

And you can get really nice longitudinal data at like a bunch of different time points. Because with pet, because it's radioactivity, you're also limited in terms of the number you can do a year.

And so it allows us to really carefully understand the dynamics of disease progression for the first set. Right. We do need to be able to go back to correlate what it is in blood versus pathology in the brain.

So you really do need to build a solid data set for us to feel comfortable relying on a blood based biomarker and not having to do the PET imaging. But the field is really, really rapidly moving in that direction.

And when you have something that starts to work, you see people come in, you see people get excited. I got a text from my mother last Sunday and she's like, I'm hearing these things about phosphatau and blood. Is that real?

She's not a scientist, she's an incredible woman. But the fact that it's hitting people like my mom in a rural part of Michigan shows people are getting excited.

And if you have that, then more companies can rely on it. Drug discovery increases, drug development increases, and you just enter into this really exciting phase for things like Alzheimer's.

People can start to feel that it's starting to happen. We're not there. We do not have a cure. But building off of biomarkers, you have a lot more companies entering into the space.

And that's what you want, right? You want more companies to come in, test more ideas so that we have multiple shots on goal.

Yeah.

So just to sort of rephrase what you just said, I think there's the first hurdle we had to be taken in the field where you have to essentially find an easy, accessible biomarker that reflects the pathology in the brain.

And I guess with the tau in combination with amyloid beta, we have a very highly accessible combination of biomarkers that is very good in predicting Alzheimer's or neurodegeneration relatively early on.

The other part that I sort of sensed during the meeting was that people now, with that biomarker available, feel incredibly empowered to do something for the people. There were recently a few drugs approved in that space.

And so let's talk a little bit about the drugs that have been approved and maybe how the biomarkers sort of help in the treatment or even the clinical trials, if that, if that's, you know, correlating as well.

Yeah, right. So there's a series of anti amyloid beta drugs. So these are all monoclonal antibodies.

They come in different flavors, they bind to different parts of amyloid. So you've got Aducanumab or Aduhelm, Lecanemab or Leqimbi and then Donanemab, which I always forget the commercial name.

It was very recently announced. So different flavors that bind at like different properties, different parts of amyloid beta. But they all are doing the same thing.

They're all trying to remove the amyloid beta plaque from the brain. And you can see that very clearly with PET pib. Right.

So that's kind of like the major one for clinical trials that they're monitoring is can you see a reduction in amyloid beta? That's the target engagement. That's what you want to do. And then the question was, if you can do that, then what happens?

In terms of clinical and cognitive decline, it is by no means a cure. However, seeing three different trials begins to build confidence in the biology.

So we just need to understand what is the impact of lowering amyloid beta. What does that mean in the long term? What does that mean for things like tau, for example?

And there's I think some really, you know, with tau pet really powerful data when you do essentially also a post hoc analysis. If you have tau pathology in your brain, an anti amyloid beta is not going to do as much.

So it really begins to help us understand the basic biology of where you are in the disease. Because Alzheimer's is super tricky. It starts 20 years prior to any cognitive decline. Right.

That has made it really difficult to be able to appropriately intervene now, because we're able to slow the disease by 20, 30%. Because we know that, yes, when you do lower plaques, we know this based on biomarkers. We see 27 to 30ish percent slowering.

Then we can say, what does that mean by way of other biomarkers such as tau? What does that mean for the interplay, the timing?

So now we know we need to go earlier and earlier and earlier in the course of disease, such that the earliest you can go, if you have someone who is mild cognitive impairment, the earliest moment for cognitive decline, they will always do better with an amyloid beta therapy versus someone who is mild or moderate, essentially by moderate if you have tau pathology and anti amyloid beta therapies, not going to work for you. But we know it only because of amyloid beta, PET imaging, tau PET imaging and now we're moving to blood based biomarkers.

So the PET imaging allows us to like very crudely understand where we sit, but a blood based biomarker allows us to get way more granular. Right. So then you can say, all right, you come in, you've got early phospho tau217 signal.

We're not seeing a ton in brain yet, but we see a little bit. So we think that an anti amyloid therapy alone is not going to be sufficient for you.

Now we want to shift that person over to something more of like a tau based therapeutic. And there are ones out there, right? There were some tau antibodies for N terminus, did not work very well.

But mid domain those are looking potentially encouraging. There's tau asos which I had worked on before. So then it allows you to redirect a person.

It's almost like a personalized medicine approach built off of blood based biomarkers that allow us to say this therapy or this therapy or this therapy, but again entirely enabled by biomarkers, which is just, it shows you the power of what you can do with that.

It's exciting to see. And I think the other portion that's really interesting here, that most of this happened within a very short period of time.

In fact, if you look at the literature, it's literally the last three, four years the entire field exploded.

Right.

But I wonder whether as we talk about the disease being a disease that start 20 years before you actually notice it in terms of cognitive decline. We have good biomarkers. Can you imagine having additional biomarkers that help us navigate those 20 years when.

You do something like pet pep because it is invasive, because it is expensive, people are not lining up. Medicare is not going to cover you to get a pet PEB at the age of in your late 30s. They're not going to do that.

So in order for us to move in that direction to go earlier and earlier.

Earlier you want something that's a bit cheaper, more accessible so that we can start to build that data set such that it is covered by insurance programs. It is something that people get more and more excited about.

I think drug development in general is quite tricky because you have to start with the disease and then you have to demonstrate some efficacy to get traction to then go earlier. For example, if there was, let's say neuroinflammation is A big part of Alzheimer's disease.

It is one of the spaces that we lack biomarkers in to this day. So we have great biomarkers for the pathology. So if I look at a brain, I see plaques, I see tau tangles, got great biomarkers for that.

When it comes to neuroinflammation, that's a big gap, but is a big part of the disease.

So could there be a blood based biomarker where you see an uptick in gfap, an uptick in a microglial gene way earlier, and then you can hinge on that and be like, okay, so we think that this person might convert to mild cognitive impairment in the next five years. Therefore, we would want to start an anti amyloid beta therapy at this point.

Because if you can get it super early, an anti amyloid might be sufficient. You might not even need to target tau at that point.

It may arrest the disease early enough.

Exactly, exactly. I mean, prevention is always better than reversal. Every indication. This is not neuroscience based.

Like every disease, if you can prevent it, your odds and your chance of success go way, way up.

So do you think that the need for these new biomarkers is currently limited, you know, because we don't have the right technologies, or is it limited because, you know, we're still working through the appropriate clinical trials? Where do you think is, are the opportunities in general there?

Yeah, I think it's definitely multifactorial. I think the sensitivity of the assays is probably one of our biggest limitations.

Okay.

And so, you know, when you do a plasma draw, things like phospho tau217, we haven't been able to see for a long time because our assays weren't sensitive enough for it. We weren't.

You wanna be able to do something that's very targeted in a space of untarget proteomics on target RNA seq, for example, you lose that sensitivity. Right. You have breadth, not depth.

And so being able to find assays that allow us to profile both breadth and depth would be hugely valuable to just start to tease these things out. We also want to rely on genetics.

Being able to go for familial Alzheimer's disease, for example, gives you power in terms of, unfortunately, these people will 100% convert. So then it allows us to track it. Earlier, when I was in grad school at Washu, I was privileged to be able to watch the Diane trials.

So dominantly inherited Alzheimer's network. And there it's kind of twofold. One is understanding therapeutic timing. So if we treat this, what will that have in terms of an impact?

But arguably, I'd say one of the more important in terms of the whole field is understanding the biomarker changes. And so they track these patients over time. So from early there's no pathology, they're totally fine.

And then they just track them every year, two years over time. And that data set has been super powerful.

And you can argue there are going to be some changes between familial Alzheimer's, if you have an APP mutation or a pre senolar mutation versus late onset Alzheimer's, but en masse, they're pretty similar. Right. You have the same pathology.

The time course could slightly change, but being able to rely on those genetic populations to inform the later stage is hugely, hugely helpful. It gives us a starting point, right.

And then we can say, all right, now we've got breadth, now we're going to go into depth on specific biomarkers that we found in genetic cohorts. Now we're going to apply that to a larger population.

But when you go into like something like late onset, more sporadic cases, it's really difficult to motivate people to come in to do lumbar punctures, to get PET imaging. Right. Those are more invasive techniques.

And there is a bit of a chicken and egg where if you don't have a therapy, people tend to, and I don't blame them, say why would.

I do that particularly they're non symptomatic. It's like, why would I go to a doctor and get, get a procedure if I don't have any problems? Right.

You mentioned those genetic populations that have been incredibly helpful because they basically create or limit a lot of the variables that we typically have for sporadic diseases. Right, let's talk about genetics for a little bit there.

So recently there was a proposal, actually in one, I think it was in nature medicine saying we should treat Alzheimer's actually, particularly for patients that have an APOE epsilon 4 mutation as a genetic disease. Because it's very clear that people with that mutation essentially proceed to neurodegenerative diseases much larger, much quicker.

So with this in mind, what else do we see in the Alzheimer field evolving?

Yeah, so APOE is today still the most reliable kind of genetic factor for sporadic Alzheimer's disease. Right. So we know very clearly the odds ratio.

So if you have, you know, APOE 4, 4, so you have two alleles and the odds ratio is like 14 or 15 fold increase chance, it's essentially 100%. Right. Like you could dabble and say like 95 to 100% or 92 to 100%. But it's very clear that, that you're moving in that direction.

The field has been looking for other genetic markers. So there are a lot of microglial genes that can be up a little bit here and there.

None of them come close to apoe like apoe, when you do a Manhattan plot, it blows them all out of the water. But it does help us understand a bit of the biology.

And so it helps us understand what do we think about trem2, what do we think about pill array, some of these genetic aspects that have little increases or decreases here and there. It just helps us understand the biology a bit, which then helps us understand what biomarkers should we be looking for.

Should we be looking for TREM2 levels, for example? Can we do that in CSF? Can we do it in plasma? Should we be understanding a pet imager? Because that was so powerful when it came to the pathologies?

Is there a way to understand inflammation states in the brain with an imaging agent to then help guide us in terms of when's the best time to do a plasma draw or things like that? And so it allows it to, basically you have an anchor point, it points you in some direction. It's more of guide rails.

It's not like absolutely this, but then we can build off of that, understand the biology, and then try and translate.

That back to biomarker.

So clearly, with the ability to measure those mutations, we probably have another really powerful tool that now we can tell people that they may have sporadic Alzheimer's tendencies. I became recently aware of a small company called Synapse DX that actually took a very different approach to diagnosing Alzheimer's.

What they do is basically take a skin cell and then they grow it in a dish for a little bit.

And it turns out because your brain and your skin come from the same development line, when you actually take those fiber plasts from your skin, if you have actually Alzheimer's or you are developing Alzheimer's, the cells have a very different shape. They ineffectually clump up. Whereas if you do not have Alzheimer's or the tendency of Alzheimer's, they actually form nice elongated cells.

And so that's a very simple test that differentiates Alzheimer's clitoris from other neurodegenerative diseases. And so when I heard this, I was like, this is an interesting new approach. I wonder whether we need maybe other approaches in the biomarker space.

And one of them I've been Thinking about is there's clearly literature emerging thinking about neuroprotective functions that the body has.

And so maybe we should also think about biomarkers that says, oh, yeah, you definitely have a tendency to develop Alzheimer's, but you also have a protective function.

Yeah, I mean, when you think of the odds ratio with a lot of these genes, we think about them, usually it was like, you've increased your risk. But there are certain genetic factors that can say, no, no, you are now protective APOE. There is the flip, right?

So 4 increases your chance, 2 reduces your chance. Right. And so there are those opportunities within the genetic space to start to define those.

I still think it basically just leads to the biology of understanding which way you want to move certain things. So should we agonize something, should we antagonize? And that's based off of some of those neuroprotective markers.

So when I think of glial genes, for example, PLC gamma is one where you want to basically, if you have a mutation, it can be protective. It just leads you to the biology. It helps you understand. All right, which way do I want to move certain microglial states?

I think understanding the timing of neuroinflammation, which way you want to move it, that is a huge gap in the field. There's a lot of data, data out there that we really want to understand. The timing part of it.

The skin fibroblast is fascinating in terms of, like, what does that mean? What does that mean for timing? So if you were to pull, you know, so like, I have a punch biopsy scar actually, on my arm.

I did a study when I was a postdoc. You know, you got to. You got to feed yourself somehow. And so it was one of those where, like, if you were able to do that longitudinally, right?

So like, well before early stages, a bit later, and just understand what that looks like, then that becomes really powerful. It is a bit difficult, right, to collect within one snapshot because then you need huge numbers of patients, right?

Alzheimer's disease is just incredibly multifactorial. It is not like, you know, you get it here and then two years later this, and then two years later this. It's just very, very different.

Like, even within my own family, right, I've had grandparents who declined very quickly, and then I had grandparents that, you know, 13, 14 years with the disease. And that's all within the same family, more or less.

And so it's something where longitudinal biomarkers in the same person, just the power behind that should not be underestimated.

IRIS actually started out in the field of proteomics looking at a neurodegenerative problem. I'm coming full circle now.

Back then we developed next generation proteomics technologies, which is just a time when we transitioned from 2D gels to liquid chromatography, mass spectrometry. And so back then we were looking at CSF and we could see beautiful patterns of proteins changing.

The challenge back then was taking out these spots that were detected by silver staining and then characterizing them by mass spectrometry. And back then the resolution was not good enough to actually say this is the specific form of protein that migrates on the gel.

Obviously technologies have improved a lot to the point where you said we use almost assays with single molecule detection sensitivity to pick up these very small amounts of phosphatid in blood. But work in proteomics from your perspective, help in the research as well as in future tools for diagnosis.

Yeah, I think the. So particularly coming from the tau field. Right. Tau is a intrinsically disordered protein.

So it means it's very like, it is very floppy, it doesn't form a crystal structure. Right. So it's been very difficult to nail down. Like what is the form of tau that causes disease?

I would argue to this day we still don't know that question.

Right.

We do not know the form of tau that actually leads to neuronal death. We have a lot of guesses. Right.

But this is where also I think when you think of which protein biomarker using proteomics, understanding and trying to like understand the native state, not applying too many harsh detergents, because you can change the phospho status, you can change. I mean, tau is phosphorylated, it's simulated, it's ubiquitated, it's methylated.

Like anything that's like aided tau has it at some point, it also gets cleaved in different places.

And these all probably mean something over the course of disease, but we've just not been able to really understand that because we didn't have the assays in terms of you are at the whims of your pull down, for example, for a very long time, what antibody, what epitope are you pulling down? That will change your entire readout.

So being able to understand more unbiased what exists in the blood helps us understand, there's the biomarker piece in terms of disease progression, but that also helps us understand. All right, is that the form of tau that we think could be leading to neuronal death. Does that help us understand therapeutics better? Right.

So If I know Phosphatau 217, if that is a major culprit in the pathogenesis of the disease. Now I know, okay, let's build a therapeutic that can basically prevent the phosphorylation of that specific site.

If I think it's just that site, I don't think it's just that site. But it helps then feed back into drug discovery, drug development.

It allows us to approach therapeutics from a totally different angle that perhaps we just had no idea about previously. And then it allows the field to move forward in terms of biology understanding. And ultimately we're doing all of this to try and cure patients. Right.

Like that's the goal is to cure the disease. And so understanding these smaller protein changes, what does that mean? What's coming from your brain? What's within your peripheral compartment?

Right. We talked about big tau, for example, which I love that it's actually called big tau. It's amazing.

I think it was coined by Michelle Godere and it's, it's one of those, right. Where it's like you have a lot of tau in your peripheral compartment as well. How do you separate out brain from periphery?

We can use proteomics to do that. They're very different sizes. Peripheral tau has several more exons. It's huge. It's like almost double in size, Correct?

Yeah.

And so it allows you then to say like, okay, what is actually coming from brain? Because that, that's plagued the field for a pretty long time. Right.

When you think of something like neurofilament light there, it's tricky to understand what is peripherally derived versus brain derived. But we can use, for Alzheimer's we can say what is, you know, amyloid beta is typically just brain centric tau.

We can use proteomics to understand what's peripheral versus cns. We can then say like, what is phosphorylated within the peripheral compartment, what's phosphorylated in the brain?

And it all leads to understanding the biology and then that feeds into therapeutics.

I agree with you. Tau protein is a, the tau protein itself is a very fascinating protein.

When I think about those phosphorylations, I think about literally timestamps that whatever goes on in the brain leaves on that molecule.

Obviously it's healthily filtered by the time it arrives in blood, but it's still sort of like a peek inside what's going on with respect to the biology in Your brain. And so I wonder whether there's a lot of conversation around the fact that Alzheimer's is the symptom, but the causes can be very different.

And in fact there is, I think, a consensus emerging that there may be several driving forces behind it. The death of synapses, maybe part of an immune reaction, there's clearly inflammation involved, as you already mentioned.

So where do you think proteomics needs to be helping and clarifying this? Obviously you're not going to do this on the patient, you're going to be doing this in animal models, for example.

What's the role maybe of proteomics in animal models or any type of other model, alternate model.

Yeah. Animal models are tricky because there is no perfect model there. I think very pointed hypotheses are essential.

And stapling essentially several different models together.

So if you have convergence across IPSC cells, for example, so if you take a fibroblast and then I differentiate them into neurons or a different cell type, microglia, astrocytes, and then I've got a mouse model where I've either knocked something in specifically to address one hypothesis. Can you move up to non human primates, for example, to get larger volumes, to understand more of like a natural state?

Putting all of those together is hugely important, right?

Because if you only isolate and like rely on just one model, like, you are lost, in my opinion, because it's just every model has huge caveats in terms of where I think proteomics can help out a lot.

I think it's, for me a huge gap is still the understanding of what neuroinflammatory signatures change when, over the course of the disease and moving earlier and earlier. So we know my beta plaque starts 20 years prior, what does that mean for microglial signatures? Can we look at plasma?

Can we understand what goes up, what goes down at different time points? Even for hyperexcitability, for example, there is a certain group of people who believe hyperexcitability is involved in early Alzheimer's disease.

Then you move to hypo excitability. So the timing is just so, so important. And I think shifting earlier and earlier and earlier is going to be really, really helpful and important.

Mouse models that we use for Alzheimer's disease, we slam it right? Like, we push it super hard.

So coming up with more gentle models, it's painful from a drug discovery and development perspective because now I got to wait a year and a half for that mouse to age, right? And we don't like to wait a year and a half. We like to go fast. We want a model. It's like in six weeks I'm done. I've got all my pathology.

But I do think that if we're going to go after prodromal Alzheimer's disease, prevent, which I still think is going to be the best course of action, we do need to form those more gentle models, those ones that have a bit more of a slow ramp up and then in humans start to just collect a lot more data from those. And we can use some genetics. Right. We can take the APOE 4 population, we can kind of start to tease all of that out.

But we really need to shift earlier and earlier.

Yeah, I mean, I agree with you that, you know, these models will play a tremendous role studying Alzheimer's disease because you literally have no access to a living brain.

What I think appreciate from your conversation or from our conversation here is that we can use these models in very specific ways to ask very specific questions and eliminate all the other variables that typically confound any type of discovery. And of course, very accessible at that point to, you know, current and future preomics technologies. Yeah.

So thinking about the future, you know, looking forward 10 years, what do you hope to see, you know, in diagnosis and maybe treatment in other areas?

I mean, I would love to see us move to more prodromal studies. So, you know, Biogen, esai, they have their prodromal a trial for Alzheimer's. It's the ahead trial.

So they're taking their lecanumab drug and they're starting treatment in people who are prone to Alzheimer's. Right. So they look at apoe, they look at age, look at family history. I think Donanemab is also doing the same thing.

So Lily's taking Donanemab, they're also doing prodromal trials. So moving in that direction again, shifting earlier and earlier before you have tau pathology, that's really kicking up.

I, I'm super excited to see what those trials look like. People don't know that Alzheimer's starts like that early. Right. It's not really counterintuitive.

It's weird to think you're totally fine and your brain is like filling with plaque. It's just, it's kind of weird. So like for me, like, I know I probably have plaque starting to form in my brain. I feel fine.

And then it's just a slow roll over 20, 25 years. Like there's not a lot of other diseases that have like that like long pathology buildup to get to a point.

And then when you're at that point, then it's so hard to reverse like we are the moment we're in now. The fact that they could see 25, 30% rescue is pretty remarkable. I'm taking a page out of the oncology space. Right.

Go earlier and just trying to mimic Alzheimer's disease after oncology. So can we have, you know, a standard blood based test when you turn 40? If you have a family history of Alzheimer's, maybe We start at 35.

Can we identify amyloid beta? Is there a neuroinflammatory signature that we can pick up? Just trying to get more accurate early on such that, you know, when you are 50? Yep.

We should maybe start an anti amyloid beta therapy and we can bring that cost down over time and then just start to do more of like a prophylactic therapy such that you never even show any symptom. Because trying to reverse something in the brain, so difficult. Right.

You only have a certain number of neurons and when those go, they're not coming back. Right. We can try and play around with like maybe you can get a few more synapses. Right.

If we clear out all the tau pathology, maybe you can reverse a few aspects of the disease, but that has to get borne out. That has not been proven yet. Way easier just to prevent the death from ever happening to begin with.

So I'm very excited to see more prodromal trials kick up on the other side for those who do have mild or moderate Alzheimer's. I'm also very excited to see more tau therapeutics, ones that we think can actually reverse the pathology.

So you think of the ASOs, the antisense oligos out there.

There are some people thinking of zinc finger transcription factors, ways of just lowering the total protein levels, because preclinically it's been shown to reverse tau pathology.

And as of last year in March, adp, Biogen and IONA showed that the tau ASO in their trials, well, given intrathecally, not the ideal route per se, but in human patients was able to reverse tau pathology there. And so it allows us to test the hypothesis. Right.

Just like an amyloid beta therapy, we saw a target engagement, we saw a reduction of amyloid beta plaque. We know that in MCI patients that translates to 25 to 30% improvement.

Now we get to test the hypothesis of if I reverse tau pathology, what does that mean for patients, what does that mean for neuronal death, what does that mean for cognitive decline? And so having both of those pieces will allow us to understand what's the best Place to target amyloid beta or tau.

An advent of neuroinflammation therapeutics rate. We're moving in that direction as well. It's further behind. So I'm curious to see kind of where that nests in.

I also hope that it's going to be probably closer to the end of those 10 years, but combination therapies, Alzheimer's, super complex. I think we are a bit naive to think that one target is going to solve all of it.

And so we have to build the individual ones first, understand what those mean. And then I'm hoping we can start to combine them to like to land on a cure. Not a slowing, but an actual like cure for the disease.

So what I hear you saying is there's lots of avenues we're already pursuing. Some of them are really hopeful.

But what I also hear you saying is that hopefully the treatments we're looking at is not just early but also personalized. So whatever accelerates the particular disease in your brain can be discovered early and then treated in a personalized way.

I can clearly tell you're very passionate about this. And I think as long as we have many passionate people like yourself in the field, we're making the progress you just described.

Yeah, it's, you know, I love diving into the details. The devil's always in the details with these things, right.

And so, I mean, having focused on Tao for so long and working with the ASOs, seeing the power that that shows right now, going back, working with the new group I'm with now, right, Being able to like take some of these newer ideas, spin out little companies, I mean, trying basically throw the kitchen sink at it. And I throw the kitchen sink, I'm throwing the bathroom sink from the garage, I'm throwing all the sinks at it.

Because it's so important that we find a cure for this disorder. It will destroy the health care system. Right. As you had mentioned earlier, the aging population is going up.

It's a burden on like the person themselves. It's a burden on the healthcare, but it's a huge burden on family members, on just society in general.

And as we all move and get older, as we're making huge waves in oncology, right, we're curing just so many different cancers. So population is going to get older and older and older becomes much more pressing. And so I'm excited to keep moving in this space.

I'm very passionate about it. If we cure Alzheimer's, maybe I become an event planner. I get out of sight, like I'm here for that reason.

And so I to be in the moment of watching this wave, watching the exponential growth when it comes to therapeutics, it's so much fun and it really gets you excited. It feels tangible. It really feels in the next five to 10 years, we're going to move in a very good direction.

Perfect. Yeah.

I mean, as someone with a family member who also suffers from neurodegenerative disease, as you said, it's a burden to the family and certainly burden to the society.

If you just imagine a big city like New York has 10,000 people who are lost, just taking care of those people who may be lost somewhere is a huge burden. So I'm with you on the passion side.

I'm on the tool side and hopefully the combination of new tools and clinical trials and new approaches, our business approaches as well, sort of crack that problem.

Yeah, absolutely.

With that. Thank you so much, Sara, for your insightful conversation. I really appreciate it.

Yeah. Thank you for having me and giving me a medium to talk about tau, my passion, Alzheimer's. It's been a ton of fun.

Thank you.

Thanks.

We hope you enjoyed the Translating Proteomics podcast brought to you by Nautilus Biotechnology. To contact us or for further information, please email translatingproteomicsautilus Bio.