Leveraging immunoprecipitation mass spectrometry and neuropathology to identify the clinical value of different blood tau species in Alzheimer's disease

J Neurol. 2026 Oct 31;273(11):695. doi: 10.1007/s00415-026-14232-y.

ABSTRACT

INTRODUCTION: Advances in plasma-based assays have allowed for the study of several tau and p-tau epitopes. P-tau217 has emerged as one of the most robust markers of Alzheimer's disease (AD) pathology. Despite their strong diagnostic performance, plasma tau biomarkers require further validation to clarify their clinical utility.

OBJECTIVE: To investigate the associations between a panel of phosphorylated and non-phosphorylated tau species measured in antemortem plasma and clinical features of autopsy-confirmed AD.

METHODS: Cross-sectional data from 123 participants from the BU ADRC who had antemortem blood draws and brain donation following death. Participants completed neuropsychological testing during life as part of their National Alzheimer's Coordinating Center-Uniform Data Set (NACC UDS) evaluation. AD neuropathology was diagnosed using NIA-Reagan criteria. Time interval between testing at last visit and autopsy ranged from 0 to 12 years, mean = 5.54 years. Raw neuropsychological test scores were used for analysis. Sample size ranged from 65 to 91 across the tests. Plasma samples were analyzed using immunoprecipitation mass spectrometry (IP-MS) for phosphorylated tau species (p-tau-181, 199, 202, 205, 217, 231) and non-phosphorylated tau (195-209, 212-221) peptides. Two ratios were analyzed, p-tau217/tau212-221 and p-tau205/tau195-209.

RESULTS: 123 participants took part in this study, 55 (44.7%) female, and 6 (4.9%) Black/African American. The mean (SD) age at blood draw was 78.17 (8.55). The strongest association between plasma biomarkers with neuropsychological test outcomes was observed with the ratio p-tau217/tau212-221 (b = - 0.564-0.404, p_low-high = 0.00-0.18), followed by p-tau217 (b = - 0.478-0.377, p_low-high = 0.00-0.28), both significantly correlating with six of the eight neuropsychological measures (p < 0.05). Followed by the ratio p-tau205/tau195-209 (b = - 0.513-0.356, p_low-high = 0.00-0.55), p-tau205 (b = - 0.365-0.226, p_low-high = 0.00-0.44), then p-tau231 (b = - 0.319-0.237, p_low-high = 0.02-0.62). Fewer significant relationships and weaker effect sizes were observed among the other biomarkers. The effect sizes of the ratio p-tau217/tau212-221, followed closely by p-tau217 on neuropsychological test outcomes, were observed to be the most similar to the effect sizes of autopsy-confirmed AD (b = - 0.564-0.393).

CONCLUSIONS: Plasma p-tau217 and p-tau ratios were comparable to autopsy-confirmed AD in predicting clinical outcomes of AD. This study highlights the promise and potential utility of plasma biomarkers for symptom monitoring.

PMID:42965704 | DOI:10.1007/s00415-026-14232-y