Plasma p ‐Tau217 assay performance in Japan: from preclinical/prodromal cohort to DMT registry
Year: 2025
Type: article
Study design: Observational Study
Abstract: BACKGROUND: There is an urgent need to determine the utility of plasma biomarkers, particularly tau phosphorylated at threonine 217 (p-tau217), in predicting amyloid PET positivity and monitoring the efficacy of disease-modifying therapies (DMT) in the early stages of Alzheimer's disease (AD). METHOD: Participants in the Japanese trial-ready cohort for preclinical/prodromal AD (J-TRC), which included 474 non-demented individuals (CDR 0: 331, CDR 0.5: 143), were assessed by amyloid PET (florbetapir or flutemetamol) and plasma biomarker assays. Plasma Aβ42 and Aβ40 were measured by IP mass spectrometry (Shimadzu), and p-tau217 was measured by immunoassay on the Meso Scale Discovery platform (Eli Lilly). The study was conducted by collaborative groups from the J-TRC and the AD DMT Registry. RESULT: Aβ-PET was abnormal in 81 participants. High accuracy in detecting an abnormal Aβ-PET scan was achieved by plasma p-tau217 (AUC = 0.913) or Aβ42/40 (AUC = 0.856), which was improved by combining plasma biomarkers and by including clinical information in the models. The highest AUCs were observed for the models combining p-tau217/Aβ42 ratio, APOE, age, sex in the whole cohort (AUC = 0.936), p-tau217, Aβ42/Aβ40 ratio, APOE, age, sex in the CDR 0 group (AUC = 0.948), and p-tau217/Aβ42 ratio, APOE, age, sex in the CDR 0.5 group (AUC = 0.955), respectively. CONCLUSION: These results indicate that plasma p-tau217 alone, or combinations of plasma Aβ biomarkers and p-tau217, are effective in predicting Aβ-PET positivity in preclinical and prodromal AD. The ability of the p-tau217 plasma assays to prescreen eligible patients and monitor the efficacy of anti-Aβ antibodies and other DMTs is also being tested in a real-world setting in the Japanese early symptomatic AD patient registry.
Source: Alzheimer s & Dementia
Author: Takeshi Iwatsubo
Institution: The University of Tokyo
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
Estimated APC paid: $4,200
OpenAlex ID: W7117111144
FWCI:
Cites:
Cited by:
Related to:
