Diagnostic Blood-Based Biomarkers of Amyloid-β and Tau Pathologies Prior to Alzheimer’s Disease Diagnosis: a Rapid Umbrella Review

Amyloid-β plaques and tau tangles are established hallmarks of Alzheimer’s disease (AD). Early detection of these pathological changes in preclinical and prodromal stages can enable timely intervention and improve outcomes. This umbrella review synthesises evidence from systematic reviews examining diagnostic blood-based biomarkers (BBMs) predictive of amyloid-β and tau pathologies prior to clinical AD diagnosis. We conducted an umbrella review of systematic reviews published between 2018 and 2024, selecting those that synthesised data on BBMs associated with amyloid-β or tau pathologies in adults in preclinical or prodromal AD stages. Searches were performed across Medline, Embase, Cochrane databases, CINAHL, Web of Science, Epistemonikos, and grey literature. A narrative synthesis approach was used. AMSTAR2 was applied for quality appraisal. Eighteen systematic reviews were included. Eight reviews were rated high or moderate quality using AMSTAR 2. Across the 18 reviews, 556 primary studies were represented, and overlap was low (38 studies; 6.8%). Forty‑four blood-based biomarkers (BBMs) were reported as associated with amyloid-β and/or tau pathology, but only three reviews reported diagnostic or prognostic performance metrics (e.g., sensitivity/specificity, PPV/NPV or AUC). Evidence with the clearest translational signal supported use of panels combining amyloid measures (e.g., plasma Aβ42/Aβ40 ratio) with APOE4 + status and/or phosphorylated tau, and plasma GFAP as an aid to distinguish amyloid-positive from amyloid-negative individuals in symptomatic populations. BBMs have the potential to widen access to amyloid and tau pathology assessment earlier in the diagnostic pathway. However, limitations in consistently reported accuracy metrics, heterogeneous populations and assays, and the small number of clinically validated tests mean that clear recommendations for routine clinical implementation cannot yet be made. Future evidence syntheses should prioritise (i) standardised reporting of diagnostic accuracy against reference standards (Aβ-PET/CSF), (ii) head‑to‑head comparisons of leading candidates (p‑tau isoforms, Aβ42/Aβ40, GFAP, NfL) and (iii) evaluation in real‑world diagnostic pathways (primary care, memory clinics).