Predicting continuous amyloid PET levels with CSF and plasma brain-derived p-tau217.
Level 3 - non-randomized controlled study
Cross-sectional biomarker validation study within an observational cohort
PubMed 42557955 · doi:10.1002/alz.71723
What was done
Researchers evaluated conventional and brain-derived (BD) CSF and plasma p-tau217 quantified with the Alamar NULISAseq panel in 924 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI). They analyzed the relationship between BD p-tau217 and continuous amyloid-beta (Aβ) burden measured by amyloid PET Centiloids (CL), as well as classification accuracy across different CL thresholds.
What was found
Plasma BD p-tau217 demonstrated the strongest continuous correlation with amyloid PET CL (R² = 0.63) and lowest prediction error across the spectrum (root mean square error = 24.5). Prediction was precise at low CL levels in cognitively unimpaired participants and reasonably accurate across intermediate-to-high CL levels in cognitively impaired participants. Brain-derived enrichment improved performance in plasma, but not in CSF.
Why it matters
Plasma BD p-tau217 enables continuous estimation of brain amyloid load from a routine blood draw rather than just a binary positive/negative result, potentially simplifying disease staging and treatment monitoring.
Limits
The abstract does not report exact classification performance metrics (such as AUC, sensitivity, or specificity) across binary Centiloid cutoffs, nor confidence intervals for R² or RMSE. Findings are derived from the ADNI cohort, which may limit generalizability to community populations, and longitudinal predictive utility was not evaluated.
Cited by
- contradicts Higher blood levels of phosphorylated tau 217 (p-tau217) serve as a biomarker indicating the loss of synapses in the brain.