New and emerging drug therapies for Alzheimer disease

[Music] Welcome to the Australian Prescriber Podcast. An independent, no-nonsense podcast for busy health professionals.

Until recently, Alzheimer's disease hasn't been typically thought about as a therapeutic area subject to breakthroughs and therapies. It was a disease where at best, we treated to relieve the clinical features without any hope of modifying the underlying disease. Things change though, and there's been a lot of talk internationally very recently about very expensive therapies in Alzheimer's disease, which might have the chance to change the game. So where do we really stand now and how do we move forward in the future?

Today, I'm speaking to Associate Professor Louise Waite, the Head of Geriatric Medicine at Concord in Sydney and a dementia specialist who has written an article in Australian Prescriber entitled ‘New and emerging drug therapies for Alzheimer disease’. Louise, welcome to the podcast.

Thanks so much, David.

So when you wrote your last Australian Prescriber review back in 2015, ‘Treatment for Alzheimer's disease: has anything changed?’, it seemed like it was a bit of a static landscape, perhaps maybe we can talk a bit about that first, about what we've traditionally done in Alzheimer's disease.

That's right, David. So things were pretty static back in 2015. At that time, we had the 3 cholinesterase inhibitors. There were monoclonal antibodies in trial at that stage, but none of them had actually come up with any positive results. And even at that time, there was talk of other pathways that contributed to Alzheimer's disease.

So tell me about this process of discovery that we've had in Alzheimer's disease because it feels like the kind of thing where we've taken smart clinical observations, done the right studies, and we've gradually got ourselves towards a potential therapeutic target.

So driving all of this is the fact that we have some genetic causes of Alzheimer's disease and we know all those genetic causes work by affecting the amyloid precursor protein, which causes abnormal processing of amyloid and then that aggregation of abnormal amyloid. So that's very much been the amyloid cascade hypothesis that's driven Alzheimer's disease and has really been pivotal in the research over the last 20 to 30 years. In development has been anti-amyloid monoclonal antibodies, and the way they work is that they go in and they bind to the amyloid protein and remove the amyloid protein from the brain.

Some of the initial trials were not positive, and the thought was that those trials were not positive because we were intervening too late. So more recent trials have focused on mild Alzheimer's disease and even that phase before we actually hit a dementia diagnosis where people have got positive biomarkers and the biomarkers that we look at are those proteins that we know are abnormal in Alzheimer's disease, amyloid and tau in particular. So these most recent therapies are focused on that amyloid cascade hypothesis and effective removal of amyloid from the brain.

How confident can we be that amyloid and tau really are causative rather than just being standards to a pathological process?

That's the million-dollar question and still under debate. So there's the amyloid amyloidists and the amyloid naysayers. I think until there are definitively trials that show that removing amyloid doesn't modify the disease in the long term, the amyloid cascade hypothesis is going to continue to dominate. It is now recognised that amyloid is not the only thing that matters in Alzheimer's disease. So there's a lot of talk now about precision medicine in dementia, working out what aspects in that individual have contributed to the dementia, and that might be things like inflammation, metabolic syndromes. That's probably where we're headed in the future with dementia and it's recognised that it's unlikely that the amyloid agents are going to be the sole agent that we use moving forward in the future with dementia and specifically with Alzheimer's disease.

With Alzheimer's in terms of therapies, it's been hard to be able to try and find the exact population that is right for intervention, partially because you've said about the complex nature of dementia, but then also partially as well to try and intervene at the right time. Tell us about this process about going earlier and earlier in dementia. We might want to try and use the most expensive and complicated therapies on the people with the worst dementia, but that's probably not necessarily the best approach, is it?

Probably not. Certainly, the way things are looked at now, and most recently, one of the trials showed that earlier intervention was more effective. So one of the monoclonal antibody trials confirmed that early intervention was more positive. There's a couple of aspects to your question, David. The first thing is if you've got pure Alzheimer's disease, the indication is that earlier intervention is likely to be better. But I'm going to put my geriatrician's hat on next and say that in Australia, 43% of people are over the age of 85 with dementia, and there are a lot of beautiful pathological series looking at the pathology of people within that age group. And Alzheimer's occurs rarely on its own. Alzheimer's is usually one of multiple pathologies causing dementia in that age group. So with my geriatrician's hat on, I look and think for the majority of people with dementia, it's going to be a mixed dementia and the monoclonal antibodies will only treat one aspect of their dementia.

So I think if we are looking at a cure for dementia, I don't think we're going to have a cure. We might have something that might modify certain aspects of that dementia and will be more effective when we've got nice, clean cases of dementia. By way of another example, there was a study in Sweden recently and they looked at people over the age of 70 and looked at who in their study would actually fit in the monoclonal antibody trials, and the first thing they found was that 10% of people would be eligible. And then when they looked at more specific details, only about 6% of people would be eligible for the monoclonal antibodies. There's a long list of exclusion criteria and just even practical things with the monoclonal antibodies, there's a lot of adverse effects associated with them. So there's what we call ARIA. So ARIA is A-R-I-A, and that stands for amyloid-related imaging abnormalities, and they can occur in around a third of people who get the monoclonal antibodies.

There's 2 types of ARIA. There's ARIA-H, which means hemosiderin or microhemorrhages, and there's the ARIA-E, which is the American version of saying oedema. So these people can get either cerebral oedema or microhaemorrhages and hemosiderin deposits on the brain associated with the monoclonal antibodies. When we have that, the only way we can identify it is with MRI scans. So where I started with that is saying that some people just can't have the monoclonal antibodies. So if you can't have an MRI scan, you can't have a monoclonal antibody because you can't have the actual monitoring that's required for it. So there are some really core aspects in working out who is and who isn't going to be eligible for the monoclonal antibodies.

So it sounds like we've got to try and pick these people who might benefit early enough. How hard is that to be able to predict the people who are going to benefit in the future to treat them early enough in their disease course?

It's going to be quite a complex and involved process to identify that small subgroup of people who are going to be eligible. It's going to be very involved with biomarkers. So these people are all going to need to have an MRI scan first to make sure they don't have an excess number of ARIA-E or ARIA-H, which can occur just in Alzheimer's disease without monoclonal antibodies. If they've got them, they're not going to be eligible because of the risk associated with monoclonal antibodies. Also, going to need to do biomarker studies to show that they've actually got amyloid in the brain. And there are cutoff criteria for people having enough amyloid in the brain to be eligible for these therapies. So the first thing is going to be identifying (1) do they fit the diagnostic criteria and then (2) do they have any exclusion criteria that increases their risk in having the therapies?

Beyond the MRIs, there's also the added complexity of apoE epsilon 4, homozygosity and heterozygosity. We know the strongest risk factor for Alzheimer's disease is having an ε4 allele, and if you are homozygous, you're worse. If you've got one copy of ε4, you're at high risk. And if you've got 2 copies of ε4, you're at even higher risk. So having homozygosity for ε4 is the strongest risk factor for Alzheimer's disease. But we also know having homozygosity for ε4 increases your risk of ARIA as well and reduces your response to the monoclonal antibodies.

So it's really going to need to be quite a nuanced discussion in terms of, yes, you've got these 2 alleles, you're not going to respond as well, you're going to have a higher risk of ARIA. And then the implications of that when you start doing that sort of genetic testing that it has (1) on the individual, but then also on that individual's family and the counselling that's going to be required around that.

The other thing, again, I'm going to put my geriatrician's hat on, there's going to be 90% plus of people who are not appropriate for monoclonal antibodies. They're still going to have a progressive dementia. We need to be able to provide care for these people. So I still have strong concerns that yes, these monoclonal antibodies will assist a few, but the vast majority it's not going to, and we're still going to need to be able to provide good dementia care for these people and manage their illness as we would usually.

So with all that in mind, let's talk a little bit about the clinical trials which have caused so much controversy. We've got 3 monoclonal antibodies at the moment that are in active discussion in terms of registration. Tell me a little bit about how the clinical trials have played out.

Aducanumab was the first one approved by FDA. The producers of aducanumab have actually withdrawn it from the market for commercial reasons. Lecanemab has been approved by the FDA in the US and we're still waiting to hear what's happening with donanemab. And both lecanemab and donanemab will be assessed by the TGA here in Australia. So at this point, none of the monoclonal antibodies are approved for use in Australia.

So these clinical trials obviously were quite high profile. Do you think that these have been slam dunk winners as far as the clinical trials are concerned? It certainly seems to be the case that there are some questions that have been raised about how effective these therapies are based on their clinical trials.

Absolutely. There's quite a lot of controversy around the medications. From my point of view, there's absolutely no doubt these drugs remove amyloid from the brain. The people who are very strong supporters of these medications argue that, well, we haven't had them long enough to see what their effect is going to be. These are disease-modifying drugs. They're not like the cholinesterase inhibitors, which are just symptomatic drugs. When I put my simple clinician's hat on, I stop and think, well, if they're modifying the disease, I would like to see measurable benefit other than just removal of amyloid. If we think that it's really beneficial, I should see some cognitive benefits. When again, I put my simple clinician's hat on and I compare it to the cholinesterase inhibitor trials accepting and recognising that cholinesterase inhibitors are symptomatic rather than disease-modifying therapies, the cholinesterase inhibitors actually come out ahead on scores like the ADAS-Cog and CDR-Sum of Boxes.

So if we're going to give these to people and they're going to expect that their cognition is going to improve on them, that's not going to happen. For me, I would like to see longer trials to be convinced of the fact that these drugs are really going to modify the disease course and give people extra quality of life. There's alternative analyses done of the drug trials that suggests that you might get an extra 5.3 months of better cognition over the 18-month duration of the trial. But those analyses are not done in the typical way where we would analyse trials. They're not done in terms of measuring scores at the same time point to see what the outcome is. So there's a bit of controversy in the way those trials are analysed and the definitions of what's clinically meaningful to a patient versus what we believe is clinically meaningful. They're 2 quite distinct options.

So if we look at what the trials would say, they would not reach criteria for being clinically meaningful, but I've heard very powerful advocates with Alzheimer's disease saying that that's meaningful to them. So there's controversy around what benefit is and how we define it from different perspectives.

What about the safety of these medicines? Have there been any concerns about that?

There's a couple of safety concerns with these medications. We know about a quarter of people get infusion reactions, so most of those are mild, and that's typically flu like symptoms with the infusion of the monoclonal antibody. Obviously, sometimes those can be quite severe and require inotropic support if you get one of the really severe infusion reactions. The other adverse effect I've talked about is the amyloid-related imaging abnormalities, and they can occur in about a third of people and the 2 subtypes of amyloid-related imaging abnormalities in terms of microhaemorrhages, hemosiderin and oedema.

The drug companies tell us that they've done a set of cognitive scores after the onset of ARIA-E and ARIA-H, and that there's no marked difference. There is one trial that suggests that ARIA-E and ARIA-H does adversely affect cognition. And we know that there’s shrinkage of the brain [in people who receive anti-amyloid monoclonal antibody therapy] and that shrinkage is faster in people who get ARIA-E and ARIA-H as one of the side effects. And one of the arguments about that shrinkage of the brain that's called pseudo-atrophy is ‘well isn't that good, we've got all the amyloid out of the brain, so now the brain is smaller’, whereas others are a bit less convinced that that's actually the case that pseudo-atrophy might actually have some [adverse] cognitive impact as well. There are still a lot of unanswered questions around the monoclonal antibodies.

And I guess to go to the point that you brought up at the very beginning that surprise, surprise, Alzheimer's disease is not a simple straightforward disease and there might be more than one process at play. This seems to be the kind of thing where we need to think about all the things that could be going on in the actual disease process itself.

Absolutely. And if you go back to good medical practice and look at prevention, there are a number of metabolic risk factors associated with Alzheimer's disease, and the Lancet Commission has come up with a number of lifestyle factors that might impact the community prevalence of dementia. So we need also to focus on things like education, physical activity, smoking, all those sort of classic things that we know are good for long-term health. We need to focus on those to reduce the risk of developing dementia, so we can't take our eyes off those aspects either.

Before we finish, can we go a little bit blue sky here and think about what ideally would happen, to be able to get these kinds of therapies to the patients who might benefit from them most in the most effective way possible? It sounds like there are a few different things would have to change about the way we practise medicine to be able to get this to happen.

I think none of us have really worked out how we can manage that, because we know the number of people in Australia already with dementia is huge, and the number of people who are going to be interested in these drugs are likely to be huge, but we know that less than 10% of people are going to be appropriate for them. So there's going to need to be some very clear work done on how we actually work out who's appropriate. And then once we identify it, just even at a most basic level who's appropriate, how we then triage those people to make sure those people get all the biomarker studies that they need done to make sure they're eligible, how we make sure those people have the MRI scans, and then how we make sure those people who we actually start on the drug can have the screening. So we need screening MRIs to watch for ARIA once we start someone on a monoclonal antibody.

So there is a huge amount of work that we're going to need to do to work out how we can safely administer this. It's clearly going to be a multidisciplinary effort. We can't do it without our monoclonal antibody infusion nurses. We can't do it without our nuclear medicine physicians reading the PET scans. And given the specialised aspect of it, there's going to need to be specific radiologists who can read the MRI scans, who are familiar with ARIA and what the criteria are for mild, moderate, severe ARIA, and then people who can manage ARIA when it does happen. So at its worst, ARIA is like a cerebral vasculitis. Basically, it gets treated like a cerebral vasculitis with pulse steroids and cyclophosphamide. You can really only have that sort of care delivered in a multidisciplinary fashion. I think it's going to need to be tertiary facilities who've got the capacity to manage things like that moving forward.

For a country like Australia, it obviously has major impacts because we're going to have a rural–urban divide. If these medications are paid for privately, at the moment they're about US$26,500, it's going to restrict them if it's not on Pharmaceutical Benefits [Scheme], to the rich urban dwellers pretty much, which that's obviously going to create additional challenges moving forward.

It sounds like we've got a lot to work through before this dream becomes a reality.

Oh, without doubt, yeah.

Well, thank you so much for joining us today, Louise.

Thanks, David.

[Music]

Louise Waite has no disclosures, and I'm a member of the Drug Utilisation Sub Committee of the PBAC. Thanks once again for joining us on the Australian Prescriber Podcast.