Lv · Science bulletin 2026 · Multicenter cross-sectional diagnostic accuracy study · n=431

Head-to-head comparison of plasma biomarker assays across platforms for amyloid pathology in a multicenter cohort.

Cited 0 times in the scientific literature.

Level 3 - non-randomized controlled study

Multicenter cross-sectional diagnostic validation study against an amyloid-PET reference standard

PubMed 42633973 · doi:10.1016/j.scib.2026.08.018 · record verified 2026-08-26

What was done

This multicenter diagnostic validation study evaluated 9 plasma phosphorylated tau 217 (p-tau217) assays (6 including Aβ42 measurements) for identifying amyloid positron emission tomography (PET) positivity. Blinded, batch-matched plasma samples from 431 participants across 10 memory clinics in China (median age 68.0 years; 61.7% women; 64.0% amyloid PET-positive; 30 cognitively unimpaired, 230 mild cognitive impairment, 171 dementia) were analyzed in a central laboratory across chemiluminescence, single-molecule, and multiplex bead-based flow cytometric platforms.

What was found

Seven of the 9 assays demonstrated strong performance with areas under the curve (AUC) from 0.899 to 0.930, while 2 assays had significantly lower performance (AUC <0.800; P < 0.001). Using a two-cutoff approach calibrated to 90% sensitivity and 90% specificity, the 7 high-performing assays produced an intermediate zone of 2.8% to 13.7%. Performance remained robust within mild cognitive impairment and dementia subgroups. Adding Aβ42 produced assay- and population-dependent effects. Previously published and manufacturer-recommended cutoffs exhibited variable performance in this cohort.

Why it matters

This head-to-head comparison shows that multiple commercially available and emerging plasma p-tau217 platforms reliably detect Alzheimer amyloid pathology in memory-clinic settings. However, marked variation in cutoff transferability indicates that thresholds must be validated and standardized before clinical implementation.

Limits

The study is cross-sectional without longitudinal clinical outcomes. The sample was restricted to memory clinics in China with a high prevalence of amyloid positivity (64.0%) and included only 30 cognitively unimpaired individuals, limiting generalizability to primary care or preclinical screening populations.

Cited by