New developments in the early detection of Alzheimer’s disease
RetourThe PHARes Translational Population Health Research team at the BPH has developed a promising new algorithm for the early detection of Alzheimer’s disease. This breakthrough comes against a backdrop of active research into the early detection of this neurodegenerative disease.
Development of algorithm and modelling
Lisa Le Scouarnec, epidemiologist and biostatistician, PhD student in the PHARes team led by Carole Dufouil and at the Bordeaux Clinical Investigation Centre (CIC 1401 Inserm/CHU Bordeaux) used data from the Memento cohort, which includes 2,323 patients with mild cognitive impairment.
The team tested six predictive models on a sample of 853 participants, incorporating sociodemographic and cognitive data as well as specific markers of Alzheimer’s disease.
Results and validation
Two models have been particularly effective in the prediction of a pathological threshold level of amyloid deposition in the brain :
- A model including the status of the apolipoprotein E gene
- A model based on blood biomarkers
These models have been validated, confirming their relevance and potential for generalisation, in a second cohort, the Amsterdam Dementia Cohort.
What are the applications?
Although less accurate than traditional tests (imaging or lumbar puncture), this new decision support tool aims to optimise diagnosis by avoiding additional testing in around 30% of people in a population similar to the Memento cohort. It offers a less invasive and more accessible alternative to current screening methods.
What does the future hold ?
The team try to improve the performance of this tool by testing the addition of new blood biomarkers, in particular pTau17. The long-term aim is to provide a screening method that is quick and easy to use and could replace current more invasive and costly techniques.
This advance is part of a wider research effort into early detection of Alzheimer’s disease, with other teams working on similar approaches using artificial intelligence and medical imaging.
The Memento cohort, launched in 2011 and renewed in 2024, included patients who consulted the Memory Resources and Research Centres (CMRR) for mild cognitive impairment or subjective cognitive complaints. They agreed to be monitored for five years, to enable scientists to look for determinants and possible precursors of Alzheimer’s disease in them.
To this end, a standardised collection of socio-demographic, biological, clinical and imaging (MRI) data was carried out for each participant.
Scientific paper :
L. Le Scouarnec et coll. Development and assessment of algorithms for predicting brain amyloid positivity in a population without dementia. Alz Res Therapy, 11 octobre 2024 ; doi:10.1186/s13195-024–01595‑5