Brain-derived phosphorylated tau 217 (p-tau 217), measured in blood samples from 213 Alzheimer's cases and 213 matched controls drawn at a mean age of 62.8 years, predicted clinically diagnosed Alzheimer's disease over a median 19.4-year follow-up with an odds ratio of 1.90 (95% CI: 1.51–2.40). The biomarker achieved an area under the curve of 0.80 for age- and sex-adjusted models, rising to 0.82 when APOE-ε4 status was added. Critically, adding 129 additional neuro- and inflammation-related proteins provided no meaningful incremental discrimination beyond p-tau 217 alone.
This finding matters enormously for Alzheimer's prevention architecture. Current prevention trials struggle to enroll participants during the decades-long preclinical window when interventions are most likely to succeed. An accessible blood test achieving 80% discriminative accuracy at roughly age 63 — nearly two decades pre-diagnosis — could transform trial recruitment and eventually clinical screening. Plasma p-tau 217 has been gaining traction in recent years as a confirmatory diagnostic tool, but extending its predictive horizon to 20 years is a substantially more ambitious and clinically consequential claim. The EPIC-Oxford cohort is population-based and well-characterized, strengthening generalizability, though the nested case-control design (213 matched pairs) limits statistical power and cannot establish causation. Hospital record–based ascertainment likely undercounts dementia cases. As a preprint not yet peer-reviewed, these results require independent replication and methodological scrutiny before influencing clinical practice. If validated, this would represent a paradigm shift in preclinical Alzheimer's identification.