New Test May Predict Dementia 9 Years Before Diagnosis
Although scientists still haven’t developed any neuroprotective medications to be used with early onset dementia, researchers have found a way to predict dementia up to 9 years before a diagnosis. Their hope is that once interventions are developed, the test will help researchers select the most appropriate candidates to go into trials.
The test, which uses a new machine learning model, can predict the eventual occurrence of dementia with 82% accuracy.
The Research
A team of researchers from Queen Mary University of London used a machine-learning model analyzed 1,111 fMRI scans of people with and without dementia, which were collected from the UK Biobank. The model was designed to be able to identify changes in 10 key regions within the brain’s default mode network (DMN). The DMN is more active when at rest, during daydreaming, or during introspection.
The control group, which included scans from 1,030 people, did not have dementia. The remaining 81 scans were from people who had not been diagnosed with dementia when their scans were taken but who developed dementia up to 9 years afterward.
Additionally, the team looked for associations between DMN dysconnectivity and various risk factors for dementia. The test found links for genetic risks for the condition, as well as risks from social isolation (which has been linked to the likelihood of developing Alzheimer’s).
The predictive tool isn’t just designed to detect Alzheimer’s, but rather all-cause dementia.
“Some brain areas show reduced activity, but others show increased activity, probably as a compensatory response,” explains the study’s senior author, Professor Charles Marshall, clinical senior lecturer in dementia in the Preventive Neurology Unit at Queen Mary. “We trained a machine learning tool to recognize patterns that were ‘dementia-like.’ In practice, a large majority of dementia is due to either Alzheimer’s disease on its own or mixed Alzheimer’s and vascular dementia. We need to extend the work to show whether or not it is relevant to rarer dementias such as frontotemporal dementia and Lewy body dementia.”
Limitations of the study
Some experts have concerns about the reproducibility of reading the studies and limitations that require replication of results with standardized methods, as well as more diverse populations.
The Alzheimer’s Association’s U.S. Senior Director of Scientific Programs and Outreach, Dr. Claire Sexton, DPhil, who was not involved in the study, expressed three specific concerns regarding the study’s limitations:
- The cohort (drawn from UK Biobank) is predominantly white, healthier than average, and with a higher-than-average socioeconomic status.
- The study reports all-caused dementia based on clinical coding, not diagnostic criteria.
- The definition and examination of DMN dysconnectivity varies substantially across studies.
Still, it’s worth noting that the dysconnectivity predicted both future dementia incidence and time to diagnosis—outperforming models based solely on functional connectivity and brain structure.
The study was published in the June 6, 2024 edition of the journal Nature Mental Health.
MBJ
Wendy Burt-Thomas writes about the brain, mental health and parenting.
Check out the original research:
