A new groundbreaking blood test, p-tau217, holds potential as an Alzheimer’s disease indicator. When incorporated into a two-step process, it demonstrates exceptional accuracy in detecting or ruling out brain amyloidosis, a crucial early sign
This innovative approach has been unveiled by researchers from the University of Gothenburg, working alongside colleagues from the University of Lund and Montreal, Canada.
Blood biomarkers for Alzheimer’s
Recent years have seen significant research efforts dedicated to creating blood biomarkers for Alzheimer’s. Tau protein has been an area of focus, specifically its phosphorylated form (p-tau), which plays a crucial role in AD pathology.
Blood-based p-tau biomarkers, particularly p-tau217, hold significant potential as practical tools for screening individuals experiencing memory issues or early cognitive symptoms associated with Alzheimer’s disease.
However, even if promising, a concern is that categorising early-stage patients as either having AD or not may still yield a relatively high number of incorrect positive and negative results (individuals with a negative test result who prove to have AD based on other examinations such as amyloid PET scans).
Considering not only ethical and psychological concerns induced by possible misdiagnosis along with the high costs and potential medical risks associated with starting treatments in individuals who don’t have the target disease, but researchers at the University of Gothenburg and their collaborators have also devised a new approach for using blood biomarkers in clinical practice.
Two-step risk stratification model
The two-step model comprises a primary step that employs a diagnostic model, which considers plasma p-tau217, age, and APOE e4 status to categorise patients with mild cognitive impairment (MCI) based on their risk of having amyloid PET positivity.
The second is based on confirmatory testing with CSF Ab42/40 ratio (or amyloid EPT), but it’s only performed when the first step’s results are unclear.
The workflow was tested on 348 individuals with mild cognitive impairment from the Swedish BioFINDER studies at Lund University. It was then confirmed in a separate TRIAD cohort at McGill University in Montreal, Canada, using a different method to analyse plasma p-tau217.
High-risk identification
The model was evaluated using three threshold strategies to categorise participants into low, intermediate, high-risk, and high-risk groups for having Alzheimer’s disease-type pathology.
At the stringent lower probability thresholds, prioritising a 97.5% sensitivity to avoid missing patients with Aβ positive results, only 6.6% of false negatives were identified. On the other hand, the stringent 97.5% specificity threshold, aimed at minimising the misclassification of Aβ negative patients as ‘high risk,’ yielded only 2.3% false positives.
When applying these stringent sensitivity/specificity thresholds, 41% of patients fell into the intermediate risk category, compared to 29% of patients using the 95% threshold. Further assessments of this intermediate group with CSF Aβ42/40 showed a strong agreement of 86% with amyloid PET results.
Conclusion and future implications
This study uses a two-step model with a blood test (p-tau217) to categorise MCI patients into high, low, and intermediate risk for brain amyloidosis and early Alzheimer’s disease. The test’s first step accurately identifies high-risk patients for potential diagnosis and treatment initiation or specialist referral for future treatment.
The low-risk group rules out Alzheimer’s, and the intermediate group, comprising about one-third of patients, reduces the need for costly confirmatory tests, saving on healthcare expenses.
