Today, I'm going to discuss emerging dementia therapies, and this is an area

that's really exploded in the last 12 months in Australia.

It's been brewing a little bit longer than that in international regions,

but it's now suddenly become extremely topical, and I know a number of your

patients will be asking about this, I'm sure.

So today, I want to give you some background as to what the emerging dementia therapies are.

Try and give you some sense of which patients we should be considering for these

therapies, so who's going to be eligible for treatment.

I want to try and address this concern that people still have,

you know, do these treatments work?

And I want to give you some sense of what the risk of these treatments are,

because obviously that's really critical as well.

And then finally, I want to give you some sense of what's involved for patients,

should they be eligible and undergo treatment.

These are my disclosures.

So, at a very basic level, these emerging dementia therapies that we're talking

about target underlying Alzheimer's disease pathology, and more specifically,

underlying amyloid pathology.

So, we've known for a number of decades that amyloid pathology is both characteristic

of Alzheimer's disease, but it's also one of the earliest pathological changes.

The amyloid deposition that precedes Alzheimer's disease probably begins several

decades before there are any outward signs of trouble.

And over the last 10 to 20 years, we've developed more and more sophisticated

ways of picking up some of these very,

very early changes and trying to get to the root of what it is that triggers

off the pathological cascade that ultimately leads to Alzheimer's disease dementia.

And so you can see here on this slide, this is our basic concept of how Alzheimer's

disease pathology begins.

There are early changes that can be detected either through fluid or imaging biomarkers.

And over time, the pathology is just quietly accumulating, even though there

are absolutely no clinically detectable problems in the patient at all.

And then at some point, you know, the pathology reaches a threshold at which

suddenly patients begin to exhibit symptoms, which may be very mild at first,

but over time become more advanced.

And most of the therapies that we're interested in really are trying to target

the earliest changes to avoid, you know, severe or delay, prevent,

you know, severe dementia from developing.

So as I mentioned, amyloid pathology is the most characteristic and earliest

feature of Alzheimer's pathology,

but the amyloid deposition then facilitates deposition of another protein called

tau and it's probably the tau protein deposition that causes the damage.

This leads to neuronal loss which we can detect as atrophy on an MRI scan and

that neuronal loss also underlies the cognitive impairment that we see in our patients.

So, very basically, these new therapies target amyloid plaques in Alzheimer's disease.

They aim to clear amyloid, which hopefully then prevents tau deposition and

therefore reduces or prevents neuronal loss.

Now, there are three anti-amyloid therapies that have reached the stage of being

approved for clinical use, at least in the United States.

There's one medication called adacanumab, which was the first medication to be approved by the FDA.

Now, this therapy, the approval process was a little bit controversial.

The clinical data probably wasn't quite there to approve the medication.

And for that reason, it never really took off in the American marketplace.

And it was actually withdrawn from the market.

But two other medications followed a similar path and had more robust clinical

data to support their use and

these therapies have now been available overseas for two and three years.

In Australia, we're just getting to the stage where these therapies are being licensed for use.

So the first therapy licensed for use in Australia is a medication called Denanimab.

Now this medication actually was the second licensed for use in the United States

and most international markets.

So the order in which these were licensed is a little bit out of whack with the rest of the world.

Lacanamab is the other medication. Now, when I initially wrote this talk,

Lacanamab had not been approved for use in Australia.

In fact, it had been rejected twice, but just in the last couple of weeks, it's been approved.

So it's, as you can see, a very dynamic field that's changing very,

very quickly, almost on a weekly basis at the moment.

Now, most of this talk is going to be about denanomab. Now, that's not because

one drug is clearly better than

the other or because I have a clear bias towards one company or another.

The reality is this is the only medication that had been licensed for approval

when I wrote this talk, and that's why we're going to concentrate on denanomab.

Many of the concepts are similar, and I think it's the broader message about

anti-amyloid therapies that I want to get across rather than a discussion about

which drug is better for who and why.

Now, as we've mentioned, amyloid deposition

is the earliest and characteristic feature of Alzheimer's disease And denanomib

aims to target this by targeting an insoluble modified N-terminal truncated

form of amyloid Which is present only in brain amyloid plaques And we know from

the phase 2 studies that,

first of all, if we can look at the,

Hopefully you can see my pointer.

Hopefully you can see that, but this bottom panel here on the right-hand side of the slide,

the medication is very effective at clearing amyloid as detected by an amyloid PET scan.

So we know that it works very well at clearing the amyloid.

And that's been, I think, fairly well established for a number of these drugs for a number of years.

The piece of the puzzle that's been a bit harder to prove has been the clinical

benefit of that amyloid clearance.

So we got an early signal here with a Phase II trial, and that led on to the

Phase III randomized double-blind placebo-controlled trial, which is called

the TRAIL-BLAZER study.

Now, you'll see that there's a number of kind of offshoot studies,

as there often is, looking at different subgroups and different ways of giving

the medication, but the TRAIL-BLAZER is the kind of overarching name that they've used for these studies.

Now, importantly, the Trailblazer study looked at patients between the ages

of 60 and 85 years, and they were really targeting patients who had either mild

cognitive impairment or mild dementia.

Now, that's a little bit tricky as to how you define mild as opposed to,

say, moderate dementia, but certainly mild cognitive impairment we would accept

as someone who has impaired performance on cognitive screening tools.

So, for instance, a mini mental score of less than 26, or my preferred measure,

a MOCA score of less than 26, but has no obvious functional impairment.

That would be someone with mild cognitive impairment.

Someone with mild dementia does have some functional impairment,

but it's relatively mild.

And there are complicated ways of trying to decide if it's mild or more significant than that.

But that's the group that we're targeting here with this particular study.

Now, patients had to have confirmed Alzheimer's disease pathology.

Now, in this particular study, they confirmed it using an amyloid PET scan.

But, you know, the amyloid PET scan is generally felt to be synonymous or gives

you the same information as a lumbar puncture looking for underlying amyloid pathology.

So those two are considered interchangeable in terms of confirming Alzheimer's pathology.

They also did what's called a tau PET. Now, this is something that's much more

a kind of clinical research tool.

It's not widely available at a clinical level, but it's important for reasons

that I'll come to shortly.

Now, one of the important kind of inclusion criteria was an imaging-based one

where patients had to have less than four what we call cerebral microhemorrhages.

So these are little areas of previous bleeding. And again, I'll come to why

that's important in a moment.

Less than one area of superficial siderosis.

No evidence of what we call ARIA-E. Again, I'll explain what that is in a moment,

but it's essentially an area of inflammation related to amyloid pathology.

And patients had to have no severe white matter disease. Okay.

So I think boiling that right down, patients needed to have either mild cognitive

impairment, mild dementia.

Amyloid pathology had to be confirmed, you know, in this case by amyloid PET,

but in practice by amyloid CSF.

And they had to have no evidence of previous brain swelling and no significant

evidence of previous brain bleeding.

They included APOE4 status but that

wasn't part of the inclusion or exclusion criteria for this

particular study but again I'll come to why that's important in a

moment now in terms of the tau pet and we're very interested to see you know

if we've got this concept that the first thing is amyloid deposition that leads

to tau deposition and then it's the tau deposition that does the damage that

we're very interested to see well what happens if you treat people before they've

got a lot of tau deposition.

And so they did this sort of sub-study, if you like, comparing patients who

had low to medium levels of tau versus patients who had high levels of tau.

And so this is, I think, where we're eventually going to get to.

If we're very efficient in finding patients of mild symptoms and biomarker evidence

of disease, it's probably that group of patients that we want to treat.

So this is really a critical kind of clinical question.

Now what they found, so obviously with both groups, high tau and low to medium

tau, the amyloid was cleared very effectively.

I think that's on the next slide actually, but you can see that after 76 weeks,

about 70 to 80% of patients will have amyloid clearing. So that's about 18 months.

Most people will have amyloid cleared on the amyloid PET scan.

But it's the clinical benefit that's important there. So we can see here in

the left-hand panel that the low to medium tau population had a dramatic slowing

in the rate of decline on this measure,

which is a sort of clinical measure of dementia progression, if you like, the IADRS.

Now, importantly, even the patients who had higher levels of tau,

they still benefited. It's just that the benefit was greater for those patients

that had low to medium levels of tau.

So I think the implication of this and some of the further follow-up studies

that have come subsequently is that we're going to be trying to identify people

at a very early age, confirm that they have Alzheimer's pathology,

and then time our intervention.

So whether we wait for some measure of tau deposition to become positive,

or whether we treat people based on other characteristics, that part's not quite clear.

But I think we're going to be targeting people with early pathological disease

when they have very subtle, maybe even minimal symptoms to try and prevent deterioration.

And I'm not showing the data today, but some studies, follow-up studies have

shown that if you treat people very early, you can actually reduce their decline

to the point where it's almost non-existent.

I mean, it's an amazing sort of set of results that we're talking about here.

All well and good, but what are the risks? So I've mentioned a little bit about

ARIA-E and ARIA-H, which is the reaction related to hemorrhage.

But before we get to that, you know, like any other infusion-based therapy,

patients can have infusion reactions.

These are generally pretty minor and easily managed.

These therapies can be delivered in a private practice setting.

They don't have to be in a hospital, unlike some other infusion-based therapies.

So that's pretty straightforward.

But the real concern is what we call amyloid-related imaging abnormalities or ARIA.

And this is sort of subclassified into ARIA-E.

E because it's an American designation, but they mean edema,

which is areas of swelling within the brain.

Or ARIA-H, which is related to hemorrhage.

Now, most of these ARIA-E and ARIA-H episodes are completely asymptomatic.

But because there's a lot of concern about what could happen with these therapies,

very regular monitoring with MRI is actually required for these therapies.

So two things about that. First of all, patients have to be able to have MRIs.

So if there's some reason why they can't have an MRI, they can't have these therapies.

They need to have regular MRIs. These adverse effects cannot be detected any other way.

So patients need to be able to have regular MRIs.

And the other thing is to say is that the licensing of these medications by

the TGA has actually mandated that.

So it's not sort of a nice to have, you should have an MRI scan,

it's no big deal if you don't. It's literally a condition of licensing.

So patients really need to have the MRI scans.

As I mentioned, now people worry about the top level here.

We're talking about episodes of ARIA, either ARIA-E or ARIA-H.

With denanomab treatment over an 18-month period, 36.8% of patients had a detectable

episode of ARIA-E or ARIA-H.

That looks huge, right? No doubt.

But first of all, patients who are on the placebo treatment had 15%.

So it's still a doubling of that risk.

But some of these changes that we're talking about are a feature of underlying

Alzheimer's disease pathology anyway.

So it's important to recognise that. And the other point to make here is that

the rate of symptomatic ARIA-E was actually only 6%.

So when you consider that Alzheimer's disease is a very important cause of death,

it's a huge cause of disability in our community.

We've got a therapy that causes symptomatic side effects in only 6% to people. Actually.

When symptomatic, these episodes of ARIA-E were usually relatively mild,

a bit of headache and confusion.

Most of the time, therapy would need to be suspended for a period of time until

things settle down, and then you could potentially resume treatment.

Severe or life-threatening adverse reactions to these medications is thankfully very, very low.

Probably the biggest concern we have is if somebody's accumulating areas of

ARIA-H, so little areas of cerebral microbleeds that we see on an MRI scan,

and then they come to an emergency department that isn't treating them for their dementia,

and people think that they may have had a stroke, and then they give them a thrombolytic therapy,

that's really the concern we have more with REH at this stage,

rather than symptomatic episodes related to medication only.

Now, I should just mention as well, these rates of ARIA-E relate to the initial

treatment protocol that was used, okay?

But what they realised, the pharma study people realised fairly early on is

that if you change the way that you give the first couple of months of therapy,

you can actually probably halve the rate of ARIA-E.

So just by shuffling, I think it's just one vial of medication from the first

infusion time to the third infusion time, you can halve the rate of ARIA, ARIA-E.

ARIA-H just seems to be a little bit more, you know, independent,

I guess, of the medication. So that didn't really change.

But it's ARIA-E that changes with the way that you give the medication.

There's another important point here which we'll come to Which relates to APOE4 status So again,

APOE4 homozygotes In other words,

patients who have two copies of the APOE4 gene Have the greatest risk of developing

ARIA-E And so you'll see that the licensing actually excludes those patients

from these treatments Therefore,

the real world rate of ARIA-E is likely to be much, much lower than the data

that we're showing you here.

So, now hopefully it makes sense as to why we need to exclude patients who've

got areas of cerebral microbleeds or underlying brain hemorrhage prior to any treatment.

Now, in practice, that means greater than four microhemorrhages on an iron-sensitive MRI sequence.

Now, the studies were done using what's called gradient echo MRI.

Now, this is actually an old-fashioned technique. In Australia,

most radiology practices use a more sensitive technique, which is called SWI

So we're trying to work out, well, do we go back to the past and do gradient

echoes Because that's what the study said,

Or do we use the more sensitive, more up-to-date method I guess the good thing

about using a more up-to-date method is that it will be more conservative And

we'll hopefully have less problems,

The bad thing about using SWI is that we'll be excluding a few more patients,

but that's an issue we have to resolve.

And as mentioned, patients who have two copies of the APOE4 gene are excluded from treatment.

So in the last six months, I've gone full 180 from never really testing APOE4

status to now testing it all the time. So that's a dramatic change in my practice.

So we've mentioned that this is a therapy for mild cognitive impairment and

mild dementia, not moderate or severe.

Now, that's tricky because often the patients who have moderate or severe dementia,

it's their family members that are most motivated to get these treatments.

But, you know, often they'll have disease that's too severe.

And as I mentioned, they need to have evidence of underlying Alzheimer's disease pathology.

I think generally speaking, that's going to be via CSF or lumbar puncture.

For most patients, it's more affordable from a system-wide level.

And, you know, we do have this testing available in Australia.

It's not covered by Medicare, so that is an issue, but it's relatively cheap.

So the processing of the sample alone is about $250.

Amyloid PET scans, we do have them available here at Macquarie.

That technology is always going to be more expensive than a lumbar puncture,

but obviously much less invasive.

So patients are happy to pay a higher fee but don't want to have the more invasive

test that's certainly an option for them and we need to do testing for the APOE4 gene.

As I mentioned, patients have to have MRIs regularly throughout their treatment course.

And the treatment protocol I'm establishing, I think we're going to do a baseline

scan on the scanner that we intend to use throughout the course of treatment.

So I think, strictly speaking, if you've had an MRI in the last three months, that'll probably do.

But I think best practice will be, let's do a scan on the scanner that we're

going to rely on to detect any changes over the course of treatment.

So this is the treatment schedule. You can see it's quite busy.

By my count, we're talking six MRI scans in the first 12 months and a monthly

infusion for denanumab.

In passing, lacanumab is actually a fortnightly treatment.

So there's even more infusion dates, about a similar rate of MRIs and other

things. I've highlighted there amyloid PET.

That's an optional thing. in the denanimab study they actually stopped treatment

if amyloid had been cleared.

So about 20% of patients cleared amyloid after six months and about 50% cleared at 12 months.

So if you're a patient who had your amyloid cleared and confirmed an amyloid

PET you could potentially save yourself six to 12 months of treatment.

So these, I hope, are the key takeaways for you.

I think it's clear that the early diagnosis of Alzheimer's disease may allow

us to institute some of these disease-modifying therapies, and these are truly

disease-modifying therapies.

The first ones that we've had for Alzheimer's disease are, in fact,

any neurodegenerative disease.

These treatments are approved for patients with mild cognitive impairment and

mild dementia, but we need to make sure that the safety criteria are met.

The treatment involves monthly infusions for 6 to 18 months.

There's a lot of monitoring for safety but it is at this stage an 18-month course

of treatment and then finished.

Now, I sort of didn't put this in because I wanted to see what happened when

I wrote the talk but these therapies are not covered by the PBS at this point.

Okay, I think it's a matter of time before they are approved But at the moment,

patients will have to self-fund.

So they'll have to pay for the medication.

If they're privately insured, some of the infusion-based costs and things may

be covered or partially covered.

But the therapy alone is a costly exercise.

So that's something that people need to be aware of as well.

And, you know, I think we're just starting to get some of the long-term follow-up data.

But it shows so far that the curves between the treated and untreated people are still separating.

So I think, you know, as with many kind of new treatments, you know,

you start with a particular group of patients with that disease and then you

start to work your way backwards.

I think we'll get to a point where we're detecting disease in people with almost

zero symptoms and we're preventing them from developing dementia.

I think that's the direction that we're heading in.