A recent study published in Nature Medicine has found that a finger-prick dried blood test can accurately track brain changes associated with Alzheimer’s disease, suggesting it could play a role in expanding access to early screening. However, the researchers emphasize that confirmatory testing remains necessary.
The study investigated whether dried capillary blood collected as dried plasma spots (DPS) and dried blood spots (DBS) could reliably measure biomarkers for Alzheimer’s disease compared to traditional venous plasma and cerebrospinal fluid (CSF) tests. The research involved 337 participants from seven European sites, including people with no cognitive impairment, those with mild cognitive impairment or dementia, and individuals with Down syndrome who are at higher genetic risk for Alzheimer’s.
Participants provided fingertip capillary blood samples on cards as DPS and DBS. Of these, 304 also gave paired venous samples. The venous plasma was stored at very low temperatures and shipped on dry ice, while the dried blood cards were transported at room temperature to a central laboratory.
The researchers measured key biomarkers using the single-molecule array (Simoa) platform: p-tau217 from DPS, glial fibrillary acidic protein (GFAP) from DBS, and neurofilament light (NfL) from DPS. These markers are linked to Alzheimer’s pathology and neurodegeneration. Standard cognitive tests such as the Mini-Mental State Examination were used alongside CSF biomarker analysis as reference standards.
Results showed strong correlations between venous plasma p-tau217 and capillary p-tau217, especially at moderate to high concentrations of the analyte. Performance was less robust at lower concentrations where diagnostic discrimination is more challenging. Capillary p-tau217 levels increased with clinical severity of dementia and correlated with age and cognitive performance in ways similar to venous measurements.
Capillary DPS p-tau217 was effective in predicting CSF biomarker positivity but did not perform as well as venous plasma measurements. A two-cut-off strategy allowed for prioritizing sensitivity or specificity depending on threshold selection; most individuals could be classified this way while an intermediate group would need further confirmatory testing. This supports its use primarily as a triage tool rather than a standalone diagnostic method.
Other biomarkers related to astroglial and axonal injury—capillary DBS GFAP and capillary DPS NfL—also showed strong agreement with their venous counterparts. These markers tracked expected patterns based on age and cognition, indicating that card-based sampling methods can capture relevant biological information about neurodegeneration.
In participants with Down syndrome—a population that often faces challenges accessing traditional blood draws—the finger-prick method worked well for detecting differences between dementia states and asymptomatic conditions using both p-tau217 and GFAP measures.
Short-term reproducibility was similar between supervised collection by staff and unsupervised self-collection under controlled conditions, pointing toward possible feasibility of home-based sample collection in future studies or screenings. However, practical issues remain: there were failures in collecting usable samples in 15–25% of cases; dilution effects made direct comparison with standard blood draws difficult; reliable correction factors have yet to be established; measurement of Aβ42 was limited due to low signal strength; although Aβ40 was detectable.
The authors conclude that while finger-pricked dried blood cards provide meaningful measurements of Alzheimer’s-related biomarkers under research conditions—and may support classification across clinical groups including those with Down syndrome—there are still significant limitations preventing immediate clinical adoption. Dilution effects, analytical variability, incomplete validation of diagnostic cut-offs mean this approach is not ready for patient management but may serve as an initial screening tool or for large-scale research studies focused on population health or epidemiology.
Future work will aim to optimize protocols further, validate findings across broader populations, and develop robust pathways for diagnosis using dual thresholds before considering widespread clinical implementation.
