Changing the course of dementia depends entirely on where you are standing, and this panel is standing in four different places.

Recorded in front of a live audience at the Alzheimer's Research UK Thames Valley Network Dementia Research Day 2026, Professor Emma Mead, Chief Scientific Officer at the ARUK Oxford Drug Discovery Institute, puts the question to four people who see the problem from very different ends: Associate Professor Sanjay Manohar, computational neuroscientist and lead of the Cognitive Disorders Clinic at the John Radcliffe; Professor Michele Hu, consultant neurologist at Oxford, who runs the 1,600 person Oxford Discovery Parkinson's cohort; Associate Professor Laura Winchester, bioinformatician in Oxford's Department of Psychiatry; and Peter Johnson, head of service at Dementia Oxfordshire. They cover precision diagnosis, adaptive trial design, blood biomarkers, exercise and diet, digital monitoring at home, and what patients say frightens them most, plus audience questions on genetic testing and the biomarker gap. It closes with a quick fire round where four specialists name four different priorities, and not one of them is a drug.

Key takeaways

• Of roughly 500 patients seen with Alzheimer's, only about 10 have typical Alzheimer's and nothing else. Everyone else carries a different combination, and that fingerprint matters now that treatments target specific molecules.
• Until this year, getting a biological diagnosis meant a lumbar puncture, so around 70% of patients never had one. Blood markers change who gets tested, and raise the harder question of what you tell people afterwards.
• Multi-arm multi-stage designs, borrowed from prostate cancer, run several treatment arms against one placebo group and force go or no go calls at 6 to 12 months. ACT-PD is doing it for Parkinson's.
• In the exenatide trial the placebo group improved over two years simply from being in a trial. Regular contact and something to aim at has a measurable effect you have to design around.
• Once the survival figure and the monitoring burden are explained, most patients conclude the new Alzheimer's therapies are not yet for them. What they ask for is quality of life, not longer decline.
• Asked what would make the biggest impact, the panel chose home digital monitoring, opening up trial data after studies close, AI as a support for thinking, and joined up hospital and community care. Not one of them named a drug.

--

A transcript of this show, links and show notes and profile on all our guests are available on our website at https://www.dementiaresearcher.nihr.ac.uk.

If you prefer to watch rather than listen, you will find a video version of this podcast on Apple Podcasts, YouTube, and on our website.

Leave us a tip:

https://dementia-researcher.captivate.fm/support

Follow us on social media:

• https://www.instagram.com/dementia_researcher/
• https://www.facebook.com/Dementia.Researcher/
• https://www.twitter.com/demrescommunity
• https://www.linkedin.com/company/dementia-researcher
• https://bsky.app/profile/dementiaresearcher.bsky.social

Download and Register with our Community App:

https://www.onelink.to/dementiaresearcher

We gratefully acknowledge the support of our funders: Alzheimer’s Association, Race Against Dementia, Alzheimer’s Research UK, Alzheimer’s Society, and the National Institute for Health and Care Research.

The views and opinions expressed by guests in this podcast are their own and do not necessarily reflect those of the producers, funders, or sponsors.

Subscribe to our sister show 'Dementia Researcher The Blogs':

https://podfollow.com/dementia-researcher-blogs

- [Announcer] The "Dementia

Researcher Podcast,"

talking careers and research,

sharing conference

highlights, and so much more.

- Hello and welcome.

I'm Emma Mead.

And what you're about to hear was recorded

at the Alzheimer's Research

UK Thames Valley Network

Dementia Research Day 2026.

We spent the day hearing

from great speakers,

sharing their research,

and looking at where science is heading.

And we closed with a panel on the question

that the whole field keeps coming back to:

What's actually going to

make the biggest difference

changing the course of disease?

I was joined by four people

who come at this from

very different angles.

Professor Sanjay Manohar,

professor Michele Hu, Peter Johnson,

and associate professor Laura Winchester.

Here's the conversation.

(bright music)

Hey, good afternoon, everybody,

and I'd like to extend a

warm welcome to everybody

who's listening on the

"Dementia Researcher Podcast"

this afternoon.

Today, we have a panel

and we are going to discuss

changing the course of dementia

and what will make the biggest difference.

And I'm delighted to be joined

here by an expert panel.

And I'd like to hand over

and allow our panel to

introduce themselves.

We'll start with Michele

at the end, please.

- Thank you. My name's Michele Hu.

I'm a neurology consultant

of nearly 20 years,

over 20 years experience,

and I focus on Parkinson's

and atypical Parkinson's

and therapies for this.

But I'm also on a dual role

of professor of clinical neuroscience

in Oxford University, Nuffield Department.

- Hello, I'm Peter Johnson.

I'm head of service for

Dementia Oxfordshire.

That's a service delivered

by Age UK Oxfordshire.

And we support everybody

in the community living

with the dementia diagnosis

and their families across Oxfordshire.

- Hi, I'm Laura Winchester.

I'm based in the Department

of Psychiatry at Oxford.

My interest is in bioinformatics

and looking at data

to understand Alzheimer's and Parkinson's.

- Hi, I'm Sanjay Manohar.

I'm a computational neuroscientists,

but I also run the

Cognitive Disorders Clinic

at the John Radcliffe,

which sees a variety of dementia patients.

So unlike the psychiatrist who

see the older age patients,

we see all the younger dementias

people with Alzheimer's

and slightly unusual dementias as well.

And my area of interest is

kind of more neuropsychiatric

and cognitive features of this disease

and how we explain them.

- Thank you.

It's great to have such a

range of perspectives here

on the panel today,

so I think we're gonna

have a great discussion.

So before we look ahead,

to what we think will make

the biggest difference

to people living with dementia,

I'd like to kind of take a step back

and let us think about what

the current challenges are

in the field.

So I'd maybe like to start

with Peter, if I could,

and hear from your perspective

from the Dementia Oxfordshire community

and think about what the

people that you interact with

really see as some of the

biggest challenges at the moment.

- So dealing with the

dementia diagnosis is a hard

and often desperate journey,

both for the person with the

diagnosis and their families.

And probably one of the

biggest challenges they face

is dealing with the stigma.

I mean, we've heard it

alluded to earlier today

that you could suffer.

You get socially isolated very easily

that some communities have

no words for dementia.

So having the conversation

with your community

can be very difficult

and you can be inadvertently

isolated and made very lonely.

And I think that culturally,

we inadvertently isolate

and disempower the people

by, for instance, if

when you get a diagnosis.

For the person with the diagnosis,

the message sometimes

they take out of the room

is my life is over.

I need to prepare for the end.

And for the partner,

they will go into the room as their spouse

and leave as the carer.

They're told that they're a carer

and they're expected to care

for this person they've

lived with as a partner

for 30 years.

- Yeah, I can see

that that could be very

distressing and difficult.

I mean, Michele, is that

something that you see

in the clinical setting as well,

or are there any other

challenges that you come across?

- So yes, so when you're

giving somebody a diagnosis

of early Parkinson's,

often what people most fear

is what will be my risk of

developing future dementia?

And also it's increasingly

a condition of ageing,

both dementia and Parkinson's.

So because there are populations

across the western world

are globally increasing

in terms of how people are living longer,

we are getting really a pandemic

of Alzheimer's and Parkinson's.

So one of my concerns is, are

we gonna have enough people

to diagnose these conditions

and also to effectively treat

and support in the community?

- Yeah, it's really important.

Sanjay, maybe from your perspective,

it sounds like you interact

with younger people

who are given sort of diagnoses.

Do you see similar challenges

or are there differences between

sort of older populations

and younger individuals?

- Yeah, I think the

challenges are very similar.

There's a massive shortage

of community care for these people.

And I mean, Dementia Oxfordshire did

a massively important job in this

where the NHS and social services

can't quite stretch that far.

One thing I'd say about this,

this carer business is really critical.

And half of the patients with Alzheimer's

who we send out of the room

promptly forget they've got Alzheimer's.

And the other half,

you know, they're very distressed by it.

But in all cases,

the carer is the one who

bears the responsibility.

There's always a proportion of

people who don't have a carer

and or are already kind of,

you know, being cared for

in some way.

And they're a difficult

population too. Yeah.

- I mean, Peter, maybe, do you

have any perspective on that

where Sanjay says there are some people

that don't have carers, you know?

Is that something that

you come across often

in the Dementia Oxfordshire group?

- I think it's surprising

the number of people

that are living alone with dementia,

either because they have no family

or because their family

lives a long way away,

often out of county.

And that can be very challenging

and make it much harder

to support that person.

- Yeah, I can imagine that

would be really difficult.

And it's good that there are

organisations like yourselves

to provide that support

and groups that can

really kind of help people

who might be feeling isolated and alone.

So it's wonderful to see that.

I wondered maybe if we can move a bit

towards that more sort

of clinical perspective

and think about sort

of gaps and challenges

in terms of how we are treating dementias

and other sort of

neurogenerative conditions.

And maybe, Michele, if I

could come to you first,

what do you see as the major challenges

in the clinical space at the moment?

- I think who diagnoses dementia,

whether it should be, for

example, a neurologist,

a general physician, or a psychiatrist,

is something that we haven't

really coordinated very well.

I personally don't think it matters

as long as you actually

have a good understanding

and a good rapport with the patient.

So I think we need to work better

with our community services

in a more joined up way.

So for example,

we've recently met with our

community psychiatry team.

Our older age psychiatrist said,

and they've actually said,

"We think you neurologists are better

at diagnosing Parkinson's dementia,

but we're better at

supporting in the community.

So could you do that front

aspect and initiate treatment?

And we're happy to continue it

and provide community support."

I think that actually

is a way going forward

that optimistically may make

a better use of our resources

and be more effective.

- That's great. Thank you.

And sort of thinking also about

how we understand disease,

and maybe I'll turn to Laura

and ask, you know, what

are the main challenges

when we think about, you know...

So bringing together our

disease understanding

and combining that with

our clinical practise,

you know, to me, that

feels like there's a bit

of a gap there. I don't know

if you have any thoughts

about what the challenges are

that we can, you know, overcome

to address that question.

- So I think there's lots of gaps there,

and I think there's lots

of people in the room

are probably working on

different bits of those gaps.

I think there is a lot of work now going

into not just looking at maybe cohort,

but real-world data as well.

So there is people researching both.

And I think it's all about

what data is available

and as people maybe thinking

about people coming to clinics

and collecting data from those people.

So that's available for us to look at

in addition to what we are

looking at sort of collecting

and designing.

So I think it's joining

up those different groups

and collecting as much

information as possible

so that we can continue

to study that as well.

- That sounds like a very important step.

I mean is that feasible

in a clinical setting?

I don't know if, Sanjay, I can come to you

and, you know, think

about how we can tie up,

you know, get collecting the right data

for researchers like Laura

to use it going forward

to address some of these key questions.

- Yeah, and I think

coming to a day like this,

you see so much great research,

and then maybe something

that will surprise

quite a few people

is that out of, let's say the 500 patients

with dementia, you know,

several years with Alzheimer's disease,

about 10 of them are typical Alzheimer's.

The rest all have something else as well.

We've heard a little bit about copathology

and you know, how you might have amyloid,

how you might have Lewy body pathology,

corticobasal pathology,

all these other things.

But also they have vascular disease,

they have epilepsy,

they have head injuries,

they have, you know, everything else,

they have comorbidities

elsewhere in their body,

diabetes, and so on.

So each patient we see

has kind of a unique fingerprint profile

in a very high dimensional space.

And each one is unique,

right? Each patient.

That's the key.

And when you see research

that you know, ultimately

it's like labels people

with having Alzheimer's or not,

well, we give Alzheimer's

labels on a whim, let's say.

The process of diagnosing the disease

is partially biological, partially social.

It's partly to do with what's important

and needed by that patient at the moment.

And that's a bit old-fashioned, right?

Because this was conceived in a day

where we didn't have any

disease-modifying therapies.

Now if you say we're gonna give treatments

that target a molecule,

and it becomes so much more important

to get precision diagnosis

to fingerprint each patient

with exactly what's going on biologically.

And then we've made very

early steps into this

by now having biomarkers

that we can test promptly and quickly.

Until a few years ago, until

like this year, in fact,

we had to get cerebral spinal fluid

with a lumbar puncture on every patient

if we really wanted to know

the biological diagnosis,

which meant that about 70%

of patients we didn't bother

to get the biological diagnosis, right?

It's expensive, painful, and inconvenient.

But I think that is the

challenge I would highlight

is that everyone needs to understand

the precision with which

we know these labels

and also think of them

as inhabiting this high-dimensional space.

And I think that once that

challenge is addressed,

we'll have much better avenues

into treating this disease.

- Yeah, that's so interesting.

I'd like to maybe pick

up at the point you made

about the different sort of samples

that we can gather from the clinic now,

and maybe ask Laura sort of, you know,

have you got some examples

of how that, you know,

huge amount of data now

that we can gather from

patients clinically can be used?

You know, are there any

examples from the work

that you're doing for that?

- So I think, actually, you

touched on a couple of things

that we're thinking about.

So looking at those

datasets and pulling out...

So there is a biomarker

that's working from blood now,

but that's one, and maybe

there are more out there.

So p-tau is looking for amyloid.

We might be interested in markers

for some of the other pathologies

that we're thinking about.

So when you're thinking about dementia,

it's a complex set of pathologies in there

and we can use the datasets

that we have to perhaps...

Now we have the bigger datasets,

you might have power to pull

apart these complex problems.

So thinking about the pathology level,

but also thinking about

at the clinical level

and looking to see whether

there's a lineup there.

In some cases, there isn't,

and there's a very interesting question

with sort of resilience.

So whether you have a buildup of amyloid,

but these patients

are not showing the clinical symptoms yet.

And so that group are

interesting to study.

And so as we get more

data on these patients,

we'll be able to pull out

more information about that.

And it's about understanding,

getting a bigger picture.

And I think these datasets, as they build,

become really important.

- That's really interesting. Thank you.

And I mean, obviously,

to do these analysis,

we need people to donate samples.

And maybe, Peter, from

your experience, you know,

is there a great sort of

appetite in the dementia research

or the dementia community to

support dementia research?

And do people really know very much

about how they can go about donating blood

or any other sort of samples?

- I think the reality is

that people with carers

and people with a diagnosis are time-poor.

So the difficulty is finding

a way to engage with people

that are overwhelmed by caring

and the condition they're managing.

So I do think that they do want to engage,

it's the practicalities of how you engage.

And you kind of need,

I mean we've talked about this previously,

but you kind of need to

meet them where they're at.

So if you engage with them,

so we have an experts group

that's often used to engage with people,

but any of our groups that meet regularly,

you'll be able to engage in a space

where they feel comfortable

and they'll be able to contribute.

And that kind of works.

Researchers are very keen to come along

and just chat to our clients.

And I don't really understand

how they use the information they get,

but clearly they keep coming

back, so it has value.

And I would say from the people

that are living with a

diagnosis and their carers,

there's a therapeutic value

of just being involved.

It's giving their life meaning.

That gives them a purpose.

And that can make a really big difference.

It's that sense of what

is the point of me.

It helps them retain that

focus of who they are,

for both the carer and the

person with the diagnosis.

It's a way of them contributing.

So they are keen to engage,

but often it's hard for them to do so

because they're being overwhelmed

by the rest of their lives.

- That's really an important point.

Michele, you had something-

- I was saying I've won

the Oxford Discovery Parkinson's Cohort,

which has 1,600

And people find it really reassuring

to be in an observational cohort

because they're seeing

often the same staff.

Perhaps every year or year and a half,

they're being fed back

if there's anything that

needs clinical care.

You know, general health improved

because we're measuring things

like high blood pressure, et cetera,

you know, as part of that assessment.

And we find that people

are quite willing to do procedural tests

like a skin biopsy, blood

tests, and lumbar puncture.

And I just wanna take issue.

It's not painful when we do them, Sanjay.

(group laughing)

It's no more painful than

when you go to the dentist

and have a local anaesthetic which stings.

But after that, and I've just

done one about two hours ago,

so we do them roughly every week

for about three or four individuals

and they are often very surprised

at how straightforward it is.

And certainly, for Parkinson's,

it's becoming actually a

prerequisite that you have

the misfolded protein

alpha-synuclein identified

before you're eligible for a trial entry.

- No, that's really interesting.

- Thank you. You're absolutely correct.

(group laughing)

- Staying with clinical trials perhaps,

you know, I'm interested to hear,

we've obviously had quite

a few clinical trials,

clinical successes in the space

of Alzheimer's disease in particular.

And you know, we're moving to a place now

where we're hoping to have

more therapies in the clinic.

But maybe from a clinician's perspective,

how do you think we can go

about improving trial designs

so that we can more rapidly

get the right therapies

to the right patients at the right time?

I don't know if, Michele,

do you want to answer first?

- So one way to do it

is to use a different,

more adaptive trial design.

And in the UK for Parkinson's,

we now have the Edmund J Safra

Accelerating Clinical Trials

to Parkinson's or the ACT-PD.

And this uses a model called

MAMS, multi-arm, multi-stage.

And it was adapted originally

from trials that were

done in prostate cancer.

What's unique about this is

you, for example, can have three

or four different treatment trial arms

going on simultaneously,

which in the case of ACT-PD

will be three oral medications

all with different evidence

for slowing down Parkinson's progression

and one placebo group.

So fewer people will need the placebo,

and you also have much

quicker go/no-go decisions

at 6 to 12 and 18 months

if you are not seeing any

signal of improvement benefit

within 6 to 12 months.

So I think this is now also being adopted

by Michael J. Fox Foundation for their P2P

or Path to Prevention trial.

And I think it's gonna be the way forward.

We also need better ways of mitigating

against things that can mess up trials.

For example, placebo

effect and expectation bias

and learning effects.

So if patients are really desperate

that they want something to work,

in fact, in the exenatide trial people,

the placebo group just

got better over time

over the two years simply

by taking part in a trial,

having something to aim for,

regular visits with allied

health professionals and doctors,

and just feeling that they

were being better cared for.

- That's really interesting.

I mean, Peter, is that something

that you've come across

in sort of talking with

people with lived experience?

- I would say giving people hope

is a very strong therapeutic

effect you absolutely see

in people.

It is coming back to

the point I made before.

It gives them purpose.

They've got a reason to focus on something

in a world that can

otherwise feel very dark.

It's not dissimilar to the

importance of socialisation.

It's keeping your mind active,

keeping socially active.

This all contributes to that

and that I think that has a big effect.

- That's really interesting.

And I mean, thinking about the patients

that we put onto these trials,

obviously a really important

perspective in drug discovery

is that we have to have the right patients

to make sure that we

see a clinical effect.

So I don't know, Laura,

if you've got any thoughts about that

from the sort of more biomarker discovery

and clinical understanding

and a cohort understanding.

- Well, I was sort of thinking about this,

and I guess, there is, obviously,

there's an amyloid-targeted treatment.

And we can now measure amyloid

and we can point people

to that trial or drug.

And so I guess you are thinking about now

about picking out

cohorts in that same way.

So making them eligible for something.

I'm thinking about other therapies.

Maybe when you're thinking

about designing a target,

you might also think about a biomarker

that goes alongside it.

So if you're thinking about tau

then maybe you want the tau markers.

If you're thinking about

an inflammatory drug,

then maybe you want to be thinking

about inflammatory markers

that go along to check progression

or to think about end

decision points as well.

And so I think there's probably

parallel streams of work

to be done as you think

about the drug targets,

what else are you thinking

about in that particular cohort?

So yeah, there's a lot

with that understanding.

- That's such an important point.

I mean, do you have any thoughts

about how we can create that environment

to make sure we are pairing up people

with that sort of expertise

and understanding what

the biomarkers should be

and how we should go about doing them

and getting the drug discovery scientists

having those conversations early?

- I mean, I think just

having these people.

Panels like this really enable

us to chat to each other.

So I myself as a data

science talking to yourself

as somebody from the

Drug Discovery Institute,

and the clinicians as well.

So it's just having joined

up conversations with people

who are working in this field

and making sure that everybody

is working towards the

same thing, I think.

And so we sort of have the environment.

We just have to make sure we're using it.

- Yeah, it's so important.

And you know, I think that

that is really the only way

we begin to see success, isn't it?

Making sure that our data scientists,

our drug discovery scientists,

and our clinicians are having

that good communication

and obviously making sure

we're involving people

with lived experience

to know exactly what we should be doing

to inform our strategy.

- I think what's really

unique about ACT-PD

is it's had a very active

patient engagement group.

And as a result, we're not doing things

like asking patients to come in

following overnight

withdrawal of medication

'cause that's actually really distressing.

People often get a lot worse,

particularly if they've had

PD for a number of years.

Also, the main outcomes

are actually not the

ones that are in clinic.

And there's a heavier reliance

on other remote assessments

that can be done at home

and digital outcomes,

which I think will be very important,

particularly for phase three trials.

- That sounds really kind

of game-changing actually,

you know, thinking that we can have people

in the comfort of their own homes

still kind of reporting on

the endpoints of studies.

I think that that will be fantastic.

I mean, again, sort of thinking about

some of the clinical trials

that have been ongoing

and that have been in the clinic

and have maybe had mixed success.

And we've had some successes

in the Alzheimer's field

with lecanemab and donanemab,

and obviously that's a little bit mixed,

but I would really like

first to ask maybe Peter

and our clinical colleagues,

you know, what the

perspective of those are

and then think about sort of what's next.

So Peter, you know, what do

the dementia community think

about the current Alzheimer's therapies

that are available to some individuals?

- No, and I think, it's quite a mix

for people actually living with

the condition at the moment.

I mean it's hope on the horizon,

but for most of them, it's

very difficult for them

to engage even with trials.

The criteria are quite strict.

And it can be challenging

with the press raising everybody's hopes

with these big announcements

and then shooting them down

with sensational follow up.

And they don't know what to think

and we have to be very

careful about what we say.

The reality is that it

requires a relatively early,

I'll be corrected, but it

requires an early diagnosis.

You've got to manage the side effects,

which are exclusionary for

a lot of people, et cetera.

So we have to manage expectations

and say that like you

can apply for the trials,

but it might actually not

be good for you to engage.

- Yeah.

And it's so difficult

to manage those expectations, isn't it?

I think, in a clinical setting I mean.

Sanjay, you know, in your experience,

are patients excited about

these sorts of developments

or is it, as Peter mentioned,

a bit of a mixed reception?

- So with most patients who

we diagnose with Alzheimer's,

I would talk to 'em about these therapies,

mainly because they're in the news

and everyone has questions about them.

But in pretty much all the cases,

now, of course, it depends

on what you tell them,

doesn't it?

But in most cases,

when I give an honest

description of the findings,

they all come out thinking,

"No, it's not ready for consumption yet,"

particularly when you

describe, you know, the profile

of the care they would need,

the effects that could happen.

And in particular when you

tell them that the statistic

that it prolongs life by four months-

- But not improving the

quality of their life.

- And this is the thing.

What patients are most afraid

of when they come out clinic

is being a burden.

Can you imagine?

This is actually what a lot

of them say to you is that,

"The other thing I'm most terrified about

is being a burden,"

to their partner, to society.

It must be a terrible

feeling to have actually.

What we need to strive

towards is therapies

that improve the quality of

life that people have, right?

That's what's needed.

And when a therapy is

touted as prolonging life,

yeah, I think they all know

what that means actually.

- I think you've raised a

really important point there

about improving quality of life.

And I wonder, you know, we

have so many different types

of therapies now entering the clinic.

There's been a bit of a

focus on disease modifying

and that being a bit of

the sort of the thing

that everybody's striving for.

But actually, I wonder

if symptomatic treatments

might be something that

patients are, you know,

have more appetite for.

Is that sort of your

perspective there, Sanjay?

- That would be my intuition

in talking to a lot of them.

It's something that can

make them feel better,

even if it's short term.

Exactly of as much

benefit, if not more than,

something that really attracts things on,

which is, you know, that's

how they feel about it.

Of course, in reality,

there will be two months

of better quality life too.

And is the, you know, the amount

of treatment they have worth that?

That's a second question.

That's a nice start question, you know?

That will be addressed

and will improve as well.

- And Michele, what's

your thoughts on that?

Are your patients looking

for symptomatic relief

or disease-modifying therapies?

I mean there is scope for both, but...

- Yeah, well, we're quite

lucky in Parkinson's

'cause you know we have a

lot more symptomatic drugs

that are really quite effective.

We can replace the levodopa.

We can give dopamine agonists.

We can use enzymes to make

it hang around for longer.

Then we also have things

like deep brain stimulation

and we can give pump therapies.

So you know, I think,

less than 5% of all of our

patients with Parkinson's

will be suitable for

these device therapies

or DBS in any case.

And I would posit, probably,

similar or even less

will be suitable for the current

disease-modifying therapies

that we have.

I think there are also lots of other

potentially more effective,

low-cost pragmatic solutions.

And in Parkinson's,

the evidence that regular

high-intensity exercise

three times a week,

and a diet, a change in diet

away from processed foods

to more, you know, less

animal-based proteins,

less fried food,

sort of African heritage diet,

really slows down progression.

But we haven't got

enough data to look at it

beyond 12-month duration.

And that's one of the studies

that I would love to do.

And I don't see why that wouldn't

also be the case actually

in Alzheimer's or other forms of dementia.

- That's really interesting,

and you know, maybe something

for the data scientists

to collaborate with the clinicians on

to try and understand that.

I mean, Laura, is there anything

that you are doing in that field

to understand sort of

prevention versus treatment?

Or do you know of any

research going on around that?

- So I was immediately drawn.

There's quite a lot of work done.

I think there's probably

quite a few diets,

but I know the Mediterranean diet

is one that people have been working with

and there's some data there.

I've been doing a little bit

of work with looking at iron

and looking at whether low or high iron

might be of interest.

But there's lots of different things

that we might be able to do

that might be disease-modifying.

And I think there's a quite

a bit of work to be done

with prevention that might be useful.

I think Sana was sharing stuff earlier

about multi-morbidities

and thinking about

maybe not just dementia,

but thinking about all the other things

that might be going in parallel.

And she was using, well, UK

Biobank, a very big dataset.

So there's a lot of work

to be done in that area,

not just with diets,

but thinking about other things

that might be prevention as well.

- UK Biobank have used

wrist-worn accelerometers

and measured continuous

sort of home living activity

and shown that if you've

got higher levels,

you've got a reduced

risk of future dementia.

- That's great. Thank you.

Peter, maybe just to sort of turn to you

and ask what the dementia

community are looking for

in terms of treatment.

Obviously, this might differ

depending on whether you ask somebody

with a diagnosis of dementia

compared to their carer.

But are patients looking for something

that's going to, you

know, change the course

of their disease,

suppress the symptoms,

or as you know, we've just discussed,

you know, potentially use

a lifestyle intervention

to slow progression?

- I mean I think it's very hard

to give in a generic answer to that

because there's different

types of dementia

and their personality

overlaps with that so much.

They're all different.

And you alluded to earlier, Sanjay,

that you've got to tailor what

you say to the individual.

Some people want to work towards a cure.

That's just what they desire.

Other people are much more pragmatic

and they just might want to,

"I want to lead a high quality

life for as long as I can

and what can help with that?"

It's very varied.

Whether there's a difference

between the person with the

diagnosis and the carer,

I find it very hard to sort

of see the answer to that

because the reality is as

the condition progresses,

the carer becomes the advocate

for the person with the diagnosis.

They're speaking for them

because they even know that person best

and they're struggling to

articulate the views themselves.

So I think separating those two things,

I'm not sure I can give an answer to that.

- Yeah, I can see it can

be very sort of situation

and patient-dependent and it's, yeah.

But I can really see a place

for everything that we've discussed

in, you know, trying to tackle

this really challenging,

these challenging set of

diseases and conditions.

I would maybe like to sort

of go out to the audience

and ask if there are any questions.

So we have a few questions

on Slido already,

but please do keep them coming

if you would like to pose

any questions to our panel.

So the first question is for Sanjay,

and this is, what role

does genetic testing play

in assessing younger dementia patients?

And how does this fit with

established disease paradigms?

- So we would tend only to

test genetically in clinic

on the NHS

if a patient is diagnosed

under the age of 50,

or if they have a first-degree relative,

and usually, first-degree

relative with early onset as well.

Because it's very common to

have a parent with Alzheimer's

and then have a diagnosed with

Alzheimer's just by chance.

In terms of who wants genetic testing,

seems to be about 50/50

if you ask patients whether they want

to be genetically tested,

but half of them don't

'cause of the implications

for their family usually.

And we will do the testing

on clinically necessary grounds

when it will provide

some kind of diagnostic

and treatment implication.

So for, you know, certain types

of frontotemporal dementia,

it's typical we would convince the patient

that they need that test.

We can't convince everyone,

but most of them will

agree to having the test.

But it's not normally done.

I'm talking about these

kind of monogenic genes.

APOE is not tested on

the NHS at the moment.

Again, this ties

into the, you know,

donanemab, lecanemab question.

And at some point, it might be.

But again, the difficulty

is what do you tell someone.

You've got a risk of getting Alzheimer's.

It's increased by a certain proportion.

Yeah, so at the moment,

there's resource limitations

as well on genetic tests.

- Yeah, that's interesting.

And maybe, actually, just a

follow-up question to Laura.

How useful and how informative do you find

that kind of the data from genetic testing

in driving our understanding

of the mechanisms and append disease?

- Well, I think APOE

is extremely important

in driving a lot of

different parts of disease

in Alzheimer's.

It's got an impact.

In many of the research studies I do,

I'm always correcting for APOE

so we consider it wherever we go.

So yes, that test is very

useful to us as data scientists.

But actually having the whole genome

is extremely useful as well.

So to have a look at the other genes,

we can look at genetic

risk scores as well.

So obviously, there are genes

that we already know are associated

to Alzheimer's disease,

but potentially there are others out there

that we haven't found yet.

So I would say that yeah,

genetics is a very important study of.

- Yeah, and I guess having

that sort of APOE biology

on the background of all of

the other sort of changes

that occur in the genome is vital

to aid our understanding.

Great, thank you.

So we have a couple of other questions.

This one is directed to

Michele, but also everybody.

So I'll give Michele the

opportunity to answer first.

What do you think causes the research gap

in finding biomarkers for earlier

and more precise medicine?

And how can we bridge this

gap from basic research?

- Yeah, so that's something

I've thought a lot about

over the 20 years I've been

doing biomarker research.

I think the basic problem

is we're dealing with

a problem in the brain.

And you know, we're not

dealing with a kidney

or a liver or skin problem

where we can easily

access diagnostic tissue,

and we can use that

not just for diagnosis,

for stratification.

And the second main

problem is we have no way

of direct imaging at alpha-synuclein

and its deposition in pathological forms,

both in the brain and in

the extra central regions

where we know it happens.

For example, the

peripheral nervous system,

the gut, the skin, the heart.

So those are the two key issues.

So how do we start to bridge that gap?

And really there's not much.

We're starting to think about this,

I would say, in the last decade.

So we can measure spinal fluid,

and within that, we can look at cells

that have obviously come

from areas of the brain

as well as across the blood-brain

barrier from periphery.

We can then try and link that

to what's happening in

peripheral blood cells, PBMCs.

We can look at proteomic,

epigenomic signatures across both.

In terms of brain imaging, we

can measure dopamine deficit.

That's helpful, but not that helpful.

And we can use perhaps

more advanced sequences,

which we're developing and others are,

to look more at where the pathology is

in particular the substantia

nigra and the locus coeruleus

and using MRI,

try and directly correlate

the signal change

to pathological changes

at post-mortem brain MRI

at high resolution.

So that's another kind

of general approach.

And then we are doing more

peripheral skin biopsies

and tissue biopsies

to sort of start to understand more

about things like the gut-brain axis,

the role of the microbiome,

and how inflammation might be triggered

by something totally peripheral,

but then be weak capitulated in the brain.

- Yeah, no that's really interesting.

And you know, going back

to something you say

about the sort of brain

region-specific changes,

something that I often kind of think about

because you know, pathology

doesn't happen broadly

everywhere, does it?

And there's a lot of heterogeneity

in the way that cells

respond to the pathology.

So I think that's such an inclusive point.

- And lastly, I don't think

we've had animal models

that have particularly

weak recapitulated either.

The clinical phenotype

are the pathology that we see in humans.

And some of the human

pre-formed fibril models

of injecting alpha-synuclein

into the salivary gland

where it goes straight up into the brain

that I've recently seen presented,

I think, are very exciting.

- That sounds fascinating.

- I was just saying that

this is very exciting,

but Alzheimer's disease

is quite a long way behind

in all of this in terms

of peripheral markers.

So we don't have that gut and

all these synuclein markers.

- You've got amyloid and tau PET.

- We do.

So the thing is, amyloid and tau PET

are really difficult to access.

We don't have it in Oxford.

We don't have it in Oxford.

Where do we have it?

We have to send all our

patients to London for this.

And the ligand's really difficult to make.

So yeah, we do not have that signature yet

until we had the blood markers,

which obviously, looking very promising.

But I think other types of marker

are also looking very promising.

And by that, I mean things

like behavioural, cognitive,

and measures that we can do in clinic,

looking at how the brain is functioning.

I always say to the patients

that we can take pictures of your brain,

but the problem with your memory

is not structural necessarily.

It's to do with the way

the brain is working,

the computations and the neurons firing.

And if the problem is at that scale,

and it's certainly early in the disease,

it's at that scale, right?

We see atrophy in the

hippocampus many years

into the disease usually.

So the first biomarker

is probably going to be a

functional-type biomarker

using behavioural and

cognitive probes, I think.

- Oh, that's, yeah,

very, very interesting.

Laura, I don't know if you want to comment

about sort of finding biomarkers

from a more basic research perspective

or data science perspective.

- Well, I guess, I was

gonna conflict a little bit

and say that maybe I'm still searching

for the elusive fluid biomarkers, so.

CSF when we have it.

But also I think the idea

of finding a blood biomarker

that enables a clinician

to do a blood test

at the GP level is really useful.

And we have brilliant

progression with that.

Yeah, with p-tau217.

I think there are trials going on

where this would be moved to clinic.

The question is now actually with this one

is when do we use it?

Because of course, you might

be able to tell a patient

that they have the amyloid,

but what do we do next?

So I think there is, potentially,

there is that diagnostic marker with AD,

but maybe we want to be

thinking about other biomarkers

for tracking maybe progression

or back again to the drug trial.

And thinking about whether we can use that

to tell us more about the disease

and maybe subtypes of the disease as well.

- Sounds about nicely to what

we were talking about earlier

about different branches of research

needing to communicate

well with each other.

- And just in response, if I may.

(group laughing)

Actually, you know, we see

about 10% of our patients

who are amyloid-positive

either in CSA plasma or in PET,

they don't progress to

get Alzheimer's disease.

So there's a,

I don't know if you wanna

call that a protective factor

or some kind of resilience component,

but yes, you can tell someone

they've got the pathology,

but not have the progressive

disease or the dementia.

And we do do that sometimes,

for a few patients who we'll tell that to.

- Thank you.

So I think we've got time

for one more question

before we sort of close out,

but I will maybe swipe

one from the audience.

As Sanjay mentioned,

there's a great variety

of Alzheimer's diseases

and heterogeneity.

So are current efforts in

therapeutic development too broad?

And how should the community

approach this efficiently?

I'm not sure who wants to take that first.

- Sanjay.

(group laughing)

- I do think that people often talk about

a lot of researchers working on prognosis

and what's the value of this?

Yes, we can tell patients how

long they're gonna survive,

but the real benefit of knowing prognosis

is in doing clinical trials.

If you know,

for a particular person tailored

to their genes, biology,

and other pathologies they have,

if you know what the

trajectory of their disease

ought to be,

then that makes all these clinical trials

much more efficient,

much, much more efficient.

I can't really explain how

much variability there is

in this condition.

There are some patients

who've gone for 10 years after diagnosis.

Others for two.

And if your measure is

how long people go on for

when you give them a treatment,

you're gonna need an enormous sample size.

But as soon as we're able to pin them down

to a particular trajectory

based on maybe their premorbid MRI,

their cognitive function,

their particular profile of deficits,

and some biomarkers and so on,

then we can then use this

sort of residualized benefit

in the trial.

And I think that that's

something that is underexplored

and people don't like putting

into trials really very much.

Right? That's my feeling.

- Yeah, I think we don't...

I mean Alzheimer's and Parkinson's,

they're complex

neurodegenerative conditions.

They progress quite slowly

over decades in general.

And there are these extremes

of phenotype and heterogeneity.

But I think we've been

too simplistic thinking

that we just do a drug trial

of one drug versus placebo.

'Cause for cancer, we never do that.

Usually you're needing two or

three different medications

and then over a period of time,

there'll be a repeat staging, scans,

and different therapeutic regimes.

I think that's the first thing.

And I think the individualised trajectory

can be best achieved

through assessing the

patient in their own home

and much more frequent

granularity of measures over time,

which we can achieve at

low cost with digital data.

So I think that,

and maybe having a trial design

where you have a six-month run in

where you then establish

that patient's baseline

before then randomising might

make a really big difference.

And you also want to know

that there is gonna be some

patient that respond very well

and others that don't,

and how are you gonna then look back

to see what best predicted that response.

So target engagement biomarkers I guess.

- Thank you very much.

We are coming up to time now,

and there are still lots of questions,

so thank you so much to everybody

for sending in questions

and hopefully there can be some

more discussion a bit later.

But I would really like to finish up

by a little quick fire

question to everybody.

What do you think will

create the biggest impact

in the way we treat and manage dementia?

So I'd like to go to Peter first.

- So I'm really interested

in what you were saying

about the digital era

and the ability to support

people in their own home.

I think that would

really make a difference.

Certainly in terms of like

as these disease progresses,

you've got the issue with

any change of routine.

Anything that's unusual

is really distressing

and can actually have

a very negative effect.

The idea of having digital

tracking that monitors them,

that gives you the ability

to make some sort of

decision about treatment

without drawing them into hospital,

drawing them into clinic I

think would be very helpful.

- Thank you very much.

Laura, we'll turn to you next.

- It's a really big question.

I was hoping you would

come to me last. (laughs)

So I don't know.

I think 'cause we were

thinking about trials,

and I think in terms of

just my particular field,

I think everyone will have

their own answer on this.

And I was thinking about

data accessibility.

And I think there's a

really important thing now

where we are thinking

we have these trials,

we have things ongoing,

but then the data is

locked away afterwards.

And I think it would be

really useful for onward

if we were able to access

and work with that data in the community.

So that's my one thing, but

I think there are others.

- Thank you very much, Laura.

Sanjay, we'll turn to you.

- So it's a controversial answer.

If you break your leg, you

can fix it, get a new one.

Your brain is you. It's your personality.

And until now, there's been

nothing in the universe

that can do the same thing.

But now, assistive technology

and things that can supplant,

you know, help you with your memory,

on the horizon with AI.

And although you can't replace

a person, the identity,

these devices, these

algorithms are quite good

at emulating particular people

given the amount of

training data you give them.

And there may be ways

in which AI cannot be a brain replacement,

but can kind of facilitate thought.

- Fascinating. Thank you very much.

Michele.

- Just very practical.

I think we may need collaboration,

real collaboration between hospital care

and community care and the patient.

And I think community care

could be better streamlined.

And we would save so many

blocked patient bed days

if we had a way to discharge patients

that needed it sooner into the community,

into the appropriate places with support.

- Wonderful. Thank you so much.

Lots of different perspectives,

and it's been wonderful to hear

your thoughts and comments.

So thank you so much.

And I'd like to thank the audience

for all the questions

that you've submitted,

and thank everybody on

the podcast for listening.

And please join me in

thanking the panel members.

(audience applauding)

(bright music)

That's where we'll leave it.

My thanks to Sanjay,

Michele, Peter, and Laura,

and to everyone who was there on the day.

You can find more

conversations like this one

at Dementia Researcher

wherever you get your podcasts.

Thanks for listening.

- [Announcer] The "Dementia

Researcher Podcast"

was brought to you by

University College London

with generous funding

from the National Institute

for Health and Care Research,

Alzheimer's Research UK,

Alzheimer's Society,

Alzheimer's Association,

and Race Against Dementia.

Dementiaresearcher.nihr.ac.uk

(upbeat music)