A blood test to diagnose Alzheimer’s disease? – 3 questions to Vincent Bouteloup
RetourVincent Bouteloup, a PhD student in Epidemiology in the PHARes team at the BPH, and his colleagues have analysed the predictive value of a biomarker present in the blood, phospho-tau 217 protein (p-tau217), which quantifies the presence of cerebral amyloid plaques causing the lesions characteristic of Alzheimer’s disease.
At present, the quantification of these plaques for clinical management purposes is carried out mainly in specialised centres (such as the Centres Mémoires de Ressources et de Recherche), by analysing cerebrospinal fluid obtained by lumbar puncture.
If measuring p-tau217 in a blood sample correctly predicts the presence of amyloid plaques in the brain, then this rapid, safe and minimally invasive test could one day replace lumbar puncture as the first line of defence.
This work is complementary to that of Lisa Le-Souarnec (PhD student in the PHAres team), who recently published an article on the prediction of amyloid plaques from data obtained during a clinical consultation to help diagnose Alzheimer’s disease at an early stage.
After presenting his work at the International Congress on Alzheimer’s and Parkinson’s Disease (AD/PD) in Vienna on 4 April 2025, we asked Vincent Bouteloup a few questions.
Picture by Mika Baumeister on Unsplash
What is the starting point for this research topic?
The scientific literature on blood markers of Alzheimer’s disease has developed rapidly in recent years. At the same time, two recent international recommendations on the diagnosis of Alzheimer’s disease have been published, contrasting a ‘biological’ view of the disease with a ‘clinico-biological’ approach. The clinico-biological approach considers that Alzheimer’s biomarkers should be interpreted in the light of the patient’s symptoms.
In this context, my thesis supervisors Carole Dufouil and Vincent Planche proposed looking at the importance of clinical aspects in the interpretation of blood markers based on data from the MEMENTO cohort.
The biomarker ptau-217 appeared to be the best candidate for this question.
Over the last 3-4 years, the scientific literature has shown that this marker performs very well compared with other less precise markers targeting other forms of this protein.
We had the opportunity to collaborate with Dr Nicolas Villain at the Institute of Brain and Marrow (ICM), and Prof Kaj Blennow at the University of Gothenburg in Sweden, to carry out the dosage of this marker in the participants included in the MEMENTO study.
What were the main challenges you faced in your study?
The 1st challenge was to find another independent study to replicate the analyses.
Replication of results is a key element in the production of scientific data and is necessary to support the validity of the results. It also makes it possible to aim for publication in more prestigious journals, as is the case here with JAMA Neurology.
For this replication we used data from the French BALTAZAR study, promoted by the AP-HP, whose principal investigator is Pr Olivier Hanon.
The second challenge concerns the possible implementation of this blood marker in clinical practice.
To this day, it provides information about the probability of having or not having the disease, but the interpretation of the results is not 100% accurate. In our work, we estimated two threshold values, one high and one low, as well as an intermediate band for which the results are considered inconclusive.
Our results show that ptau-217 is most effective in the presence of specific symptoms, typical of those found in Alzheimer’s patients. In the absence of these symptoms, our study shows that the risk of wrongly concluding that amyloid plaques are present (false positives) is very high.
In practice, you don’t want to treat a patient without making sure that he or she is really ill. Initiating treatment is not a trivial decision, and can have major consequences for patients and their families. For the 1st time, the EMA recently approved the commercialisation of an anti-amyloid treatment, lecanemab, subject to certain conditions.
Can you explain how long the research has taken and what the next steps will be before this test is available to everyone?
The dosage of Alzheimer’s markers in the blood has existed for a number of years, but it is only recently that the performance of dosage kits has reached a sufficient level of precision for them to be used on a large scale. The process of developing these assay techniques has taken so long, firstly because of the very low concentration of these biomarkers – a few picograms per ml of blood – and secondly because, unlike cerebrospinal fluid taken by lumbar puncture, which is relatively homogenous, blood carries a lot of elements that can interfere with the measurements obtained. To measure only the markers of interest, it was necessary to develop highly specific assay methods adapted to very small quantities.
In MEMENTO, we were able to measure certain blood markers a few years ago, but this is a costly and time-consuming operation for the research teams.
Most of the studies published in the field of Alzheimer’s now have these markers, and there is starting to be some scientific literature on repeated measures in relation to the progression of the disease.
Nevertheless, in my opinion, the question of clinical usefulness has not been sufficiently studied. Several publications propose theoretical schemes for implementing these markers. Scientific studies need to be carried out to specifically address the question of the value of these markers in clinical practice, as well as their economic viability in terms of, for example, coverage by the French health insurance system, because at present all this is still hypothetical.
For my part, I’m submitting my thesis at the end of the year. We’re going to continue our collaboration with the ICM on other complementary markers, and study the value of repeated measurements, again using MEMENTO. Wherever possible, we will continue to collaborate with other research teams, as was the case with the BALTAZAR study.