Plasma p‐tau217 predicts PET‐based pathological staging for precision Alzheimer disease assessment

Abstract INTRODUCTION While positron emission tomography (PET) is the standard for pathological staging, its limited availability necessitates accessible alternatives. We evaluated plasma biomarkers for detecting PET‐based stages using single‐axis (Thal/Braak) and integrated A/T composite models. METHODS We enrolled 237 AD spectrum participants undergoing multimodal assessments including amyloid/tau PET and plasma biomarker analysis (phosphorylated tau p‐tau 217, %p‐tau217, and amyloid beta Aβ 42/40 ratio). Detecting and discriminative performance was assessed using receiver operating characteristics (ROC) analysis and probability‐based stage prediction. RESULTS Plasma p‐tau217‐based biomarkers showed excellent detecting performance for early amyloid (Thal I–II; area under the curve values > 0.96) and intermediate tau (Braak III–IV; area under the curve values > 0.92). Probability‐based prediction identified therapeutic window thresholds of 1.895–5.077 pg/mL. Notably, integrated A/T composite staging yielded highly consistent thresholds (< 3% variance). DISCUSSION Plasma p‐tau217‐based biomarkers accurately reflect PET‐based staging across frameworks. The convergent therapeutic window thresholds demonstrate robust biological transitions, enabling accessible identification of optimal candidates for disease‐modifying therapies.