- Alzheimer’s disease, the most common form of dementia, affects almost 7 million people in the United States alone.
- Following a diagnosis, it is difficult to predict how the disease will progress in each person.
- Now, a research team in Amsterdam has designed a model that can predict cognitive decline in people with mild cognitive impairment or mild dementia due to Alzheimer’s disease.
- The scientists aim to use the model to develop an app that clinicians can use to help them personalize treatments and forecasts for patients.
A diagnosis of Alzheimer’s disease is becoming increasingly common. The Alzheimer’s Association reports that one in three seniors in the United States will die with Alzheimer’s or another dementia.
Some memory decline is a normal part of aging, but problems with memory and thinking that start to impact daily functioning are often a sign of
In some people, MCI can develop into dementia, such as Alzheimer’s disease, but it is hard to predict whether a person with MCI will go on to develop dementia.
Katherine Gray, Head of Research at Alzheimer’s Society, told Medical News Today that: “Among the almost a million people living with dementia in the [United Kingdom], we know that no two people have the same journey. Symptoms often progress at different rates and the availability and standard of dementia care across the UK can vary extensively.”
Now, a team from Amsterdam University Medical Center has developed a model that can predict cognitive decline in people with MCI or mild dementia due to Alzheimer’s disease. The study appears in the journal Neurology.
Wiesje van der Flier, PhD, full professor, scientific director of Amsterdam UMC’s Alzheimer Centre, and senior author on the study, told MNT that the model might one day be used to tailor Alzheimer’s care for individual patients.
“I think in the future it can. When patients get a diagnosis, their next question is: what can I expect? Or — what is my prognosis? This model provides a first attempt to an answer. It provides a prognosis for cognitive decline on an individual basis,” she told us.
Gray also commented that:
“This is encouraging research as it suggests that researchers can produce a model which is able to predict how the symptoms experienced by people with mild cognitive impairment or early Alzheimer’s disease might change. Predicting how these symptoms may change cognition over time is vitally important for those living with dementia and their carers to prepare for the future, ultimately leading to better care.”
The researchers recruited participants from the Amsterdam Dementia Cohort for their longitudinal study. In total, 961 people were included, of whom 310 had MCI and 651 had mild dementia due to Alzheimer’s disease. Their mean age was 65 years, and 49% of them were women.
All were amyloid-positive, meaning that the researchers detected amyloid biomarkers in their
The researchers used the
In this 5-minute test, a person can achieve a maximum score of 30. Clinicians roughly interpret scores as follows:
- 25–30: no impairment
- 20–24: mild dementia
- 15–20: moderate dementia
- 14 and below: severe dementia.
Over the course of the study, MMSE scores for all participants decreased, indicating a decline in cognitive abilities.
For those with MCI, MMSE scores declined from a mean of 26.4 at baseline, to 21 after 5 years. People with mild dementia showed a greater decline, from a starting mean of 22.4, the mean score reduced to 7.8 after 5 years.
For both groups, cognitive decline accelerated over time.
Using these scores, together with MRI scan results and biomarkers, the researchers modelled MMSE scores over time for both MCI and mild dementia.
“This is a really interesting study and provides the foundation for tools which could be highly beneficial to patients with Alzheimer’s disease and their families,” Scott Kaiser, MD, a board-certified geriatrician at Providence Saint John’s Health Center in Santa Monica, CA, told MNT.
“The study subjects were selected from a large cohort, the Amsterdam Dementia Cohort, with a wide range of robust clinical data available and the quality of methodology engaged in the predictive modeling appears to be a real strength of this work,” he added.
“[Our model] shows how difficult such a prognosis is, and that there remains uncertainty. From research, we know that patients and their families value this information, even if there is uncertainty. This model helps to facilitate doctor-patient communication about this uncertainty,” said van der Flier.
The researchers constructed models predicting MMSE, which they then used to estimate time to reach an MMSE of 20 (mild dementia) for those with MCI, and 15 (moderate dementia) for those with mild dementia, using different baseline CSF amyloid and MMSE measurements.
They also predicted how long it would take to reach these threshold MMSE scores, if a treatment intervention reduced decline by 30%.
Van der Flier explained why they did this, saying that:
“The predictions can also be used by a clinician to discuss with a patient the potential effect of a treatment. Recently emerging treatments have been shown to reduce the rate of decline by around 30%. Applying this to the model provides a starting point for communication between clinician and patient on the potential benefit.”
However, Karen Miller, PhD, a neuropsychologist and geropsychologist, and senior director of the Brain Wellness and Lifestyle Programs at Pacific Neuroscience Institute in Santa Monica, CA, who was not involved in this research, emphasized the uncertainty in the model.
“While the study is rigorous from a research perspective and comprehensive in terms of variable included (cognitive measures, genetics, brain imaging), the variability for any given patient still leaves the provider with a range of years in terms of any given trajectory for the unique patient,” she told MNT.
“We’re excited to see how research like this will progress in the future and deliver lasting change to people living with dementia,” said Gray.
Together with developing the predictive model, the researchers have also designed a prototype app for clinicians, as van der Flier explained.
“In the tool (adappt.health) we are currently developing, there is also a communication sheet which is intended for clinicians to share with patients and carers, to explain what the prediction entails. Also, there is patient-facing information about the disease, diagnosis and prognosis,” she told us.
Although this is an early model for predicting cognitive decline in Alzheimer’s, Kaiser welcomed the research as a step forward in giving Alzheimer’s patients and their carers more information about what to expect following diagnosis:
“This research lays the foundation for the types of prognostic tools that can not only give us a sense of what might be expected along the road ahead, but also, how we might alter that course by addressing a variety of modifiable risk factors to change the disease trajectory and improve our chances of maintaining higher levels of cognitive health and function over a longer horizon.”
