New Alzheimer's test aims to catch disease much earlier

Alzheimer’s disease is a progressive brain condition that slowly erodes memory and thinking. It often creeps up on people unnoticed for up to twenty years. But by the time symptoms do appear, treatments can usually only slow the decline by a modest amount. But researchers in Bath and Bristol believe they may have found a way to detect Alzheimer’s much earlier. Their three-minute brainwave test can spot memory problems in people with mild cognitive impairment, a condition that often precedes Alzheimer’s. The study has been led by cognitive neuroscientist George Stothart from the University of Bath…

George - Currently, we have to wait for people to turn up at their doctors and say to the doctor that they think something's up. Now, unfortunately, that can take a very long time to reach that tipping point for people. So often it's taken a crisis to occur in that person's life, that they perhaps they found that they can't do their job anymore, or they have got lost when they've left the house. The problem with that, from a medical point of view, is by the time that's happening, the disease has already done an awful lot of damage to that person's brain. And it's really too late to start doing anything about that disease. And so we want to be able to pick up that disease much, much earlier. So we've developed a new test, which is called Fastball. And it's a very quick three minute test that's entirely passive. So what that means is the participant doesn't have to do anything. All we ask them to do is watch a screen. And on that screen, we play some images very rapidly, three images per second, it's quite quick. And what we do is we repeat some of those images from time to time. Now, whilst they're watching these images, they wear a special cap, which measures their brain activity. It's called an EEG cap. And we can measure their brainwaves with that cap. Now what happens when one of those images repeats is the brain has a little recognition signal that it elicits, and we can detect that with the EEG cap. So we use the EEG cap to measure people's recognition memory. And by doing that, what we've been able to show in our experiments is that people with Alzheimer's disease show really reduced recognition memory. And in our paper that's just come out recently, we've shown that this memory is also reduced in people at very high risk of Alzheimer's disease. So people who don't yet have a diagnosis, but are quite likely to get one in the next few years.

Chris - How specific is it for Alzheimer's disease? So if I, for instance, had some other reason why I had a memory deficit or some other chronic condition that was eroding memory, how specific is your test to say, yes, Chris has got Alzheimer's versus Chris has some other memory deficit?

George - That's a really good question. And it's one we're in the middle of answering. Now we don't have the answer to it yet, but we are in the middle of two very large scale studies. We're recruiting patients from neurology clinics with all sorts of different neurological problems. And that will give us exactly those accuracy numbers so that we can tell whether it's sensitive and specific to Alzheimer's disease, or whether it's just a general marker of memory impairment. Now, my prediction, my hope is that it is specific to Alzheimer's disease. And the reason I think it will be is because we built this test from the bottom up. The starting point was to identify which area of the brain is impacted very early on in Alzheimer's disease. And it's a little part of the brain buried deep in the middle of the brain called the perirhinal cortex. And so we built the task to rely very heavily on that brain area.

Chris - And how sensitive is it? So in other words, if I came to you with some undetectable clinical features to me, I don't think there's anything wrong with me. How good are you with this test at telling me I'm going to get Alzheimer's disease? And how much lead time can you bring with this?

George - We don't know how accurate it would be in the scenario you've described, because what you're describing there is a screening tool, picking up impairment in people without any symptoms at all. It takes a long time to run studies like that, because you have to test people and then you have to wait 10 or 20 years for them to develop the disease. And you have to test enough people in the first place that enough people go on to develop that disease. So we don't have the answer for that yet, but our two large studies will tell us how accurate and sensitive we are at detecting established Alzheimer's disease, but also in detecting Alzheimer's disease in those at risk. So those with what we call mild cognitive impairment, or there's another group called subjective cognitive impairment. And these are people who say to their doctor they're having problems with their memory, but when they come and do the traditional pen and paper tests, they score fine. So they're an important group to look at as well.

Chris - And when you test these people, how good is your test at hitting the nail on the head? So in other words, your test agrees with a neurologist or an elderly care expert, how often?

George - We had a paper in 2021, which had Alzheimer's classification accuracy. And that showed that we could get up to about 90, 92% accuracy in telling whether someone had Alzheimer's disease versus a healthy older adult. So that's good accuracy and sensitivity, but clinically that's not much use because those patients had already been diagnosed. So we want to be more accurate than that earlier on. To do that, we need to test, as I mentioned before, large numbers of people, a proportion of whom then go on to develop Alzheimer's disease. So we will learn that accuracy number from our current study in a few years time as we follow them up.

Chris - Is the longer term ambition here that if you can get this giving you a really long lead time, with the drugs that we foresee coming, we've only really tested those on people with already manifest Alzheimer's problems and got limited benefits from them, slowing down the condition, haven't we? People are saying, well, if we could intervene earlier, maybe the gains would be bigger.

George - Absolutely. Earlier diagnosis brings so many benefits. I mean, it brings benefits, potential benefits for the patient, given greater certainty, given them the opportunity to make lifestyle interventions, modifications, which has been shown to slow disease progress. But also, as you say, the drug development and the ability to use disease modifying therapies earlier. And also, who knows how that might improve and help the actual drug development itself. So not only will the drugs that we have work better, but could it be that we actually improve the development of those drugs in the first place by doing better stratification of patients into clinical trials and better characterization of patients in these clinical trials.