Plasma p‑tau217 'Clock' Predicts Timing of Symptomatic Alzheimer’s Disease

Investigators described building plasma biomarker “clock” models based on plasma p‑tau217 with the stated goal of estimating the age at which an individual is likely to develop symptomatic Alzheimer disease. The authors said the models predicted symptom onset within a margin of about 3–4 years. The work was presented as an effort to translate longitudinal biomarker trajectories into an estimated timeline for symptomatic onset.

To define the interval between elevated plasma p‑tau217 and symptom onset, the analysis was reported to draw on two independent research initiatives: the Knight Alzheimer Disease Research Center cohort and the Alzheimer’s Disease Neuroimaging Initiative cohort. The combined dataset was described as 603 community-dwelling older adults. The report also described an age-related pattern, stating that older individuals tended to have a shorter interval between elevated p‑tau217 and the start of symptoms than younger participants. In this framing, the “clock” estimates were derived from observed trajectories across the two cohorts.

Across the cohorts, plasma p‑tau217 measurement was described as spanning multiple assay platforms. In the WashU Knight ADRC cohort, p‑tau217 was measured with PrecivityAD2, characterized in the report as a clinically available diagnostic blood test; in ADNI, p‑tau217 was reported to be measured using assays from other companies, including one test described as cleared by the U.S. Food and Drug Administration. The authors were quoted as saying the approach worked with p‑tau217-based diagnostic tests beyond the specific platform used in the WashU cohort, which they presented as evidence of robustness across assays. Overall, the evaluation was presented as considering more than one p‑tau217 assay platform when assessing model behavior.

The report stated that the authors shared the full code used to develop the models so other researchers could further refine them, and it also described a web-based application for exploring the clock models in more detail. In author-attributed comments, potential research uses were framed around making clinical trials more efficient by identifying individuals likely to develop symptoms within a specified period of time. The same comments described future refinement as potentially incorporating additional blood biomarkers said to be associated with cognitive symptoms in Alzheimer disease, with the intent of improving estimates. In this account, the shared code and web tool were presented as enabling external exploration and iterative development aligned with the authors’ stated research goals.

Key Takeaways:

• The report described plasma p‑tau217 “clock” models intended to estimate the timing/age of symptomatic Alzheimer disease onset, with performance characterized in the report as predicting onset within a small multi-year window.
• Model development and evaluation were described as using longitudinal data from the Knight ADRC and ADNI cohorts, with p‑tau217 measured via PrecivityAD2 in one cohort and other commercial assays (including an FDA-cleared test) in the other, alongside a cross-assay generalizability claim.
• The authors were reported to have shared model code and a web-based app, and they described potential trial-focused uses and possible refinement by adding other blood biomarkers as part of continued research development.