An accessible digital single-molecule sensing platform for plasma P-tau217 quantification in Alzheimer's disease screening.
BACKGROUND Early detection of Alzheimer's disease (AD) necessitates affordable and accessible blood biomarkers. Plasma phosphorylated tau 217 (P-tau217) is promising, but low-cost platforms for large-scale screening remain limited. METHODS We validated a Digital Single-Molecule Sensing (DiSMS) platform for plasma P-tau217 quantification. A reference interval was established in Cohort I (discovery, n = 325, comprising 143 cognitively unimpaired controls, 80 CE, 34 frontotemporal dementia [FTD], and 68 subcortical ischemic vascular dementia [SIVD] patients). Diagnostic performance was validated in Cohort II (n = 242; 77 CE, 165 controls). Cross-platform concordance was assessed against the Simoa HD-X in 137 paired samples. RESULTS The healthy reference interval was 0.09-0.51 pg/mL. In Cohort II, P-tau217 strongly distinguished AD from controls (accuracy 93.4% [95% CI: 90.3%-96.5%], sensitivity 92.2% [95% CI: 83.2%-96.8%], specificity 93.9% [95% CI: 88.8%-96.9%]). A two-cutoff approach (< 0.40, 0.40-0.50, > 0.50 pg/mL) yielded 94.1% accuracy and 95.9% sensitivity, leaving 8.4% of cases in the intermediate zone. DiSMS exhibited strong concordance with the Simoa (concordance rate = 94.9%; Spearman r = 0.94, P < 0.001). Plasma P-tau217 was significantly higher in AD than FTD (median 0.25 pg/mL) and SIVD (median 0.36 pg/mL) (both P < 0.001). Age-adjusted AUC was moderate for AD versus SIVD (0.777) but limited for AD versus FTD (0.665; sensitivity/specificity below 65%). CONCLUSIONS The DiSMS platform enables accurate, cost-accessible plasma P-tau217 quantification for AD screening, meeting consensus guidelines (≥90%). The limited P-tau217 performance for differentiating AD from FTD (AUC = 0.665) highlights the need for integration with complementary biomarkers or clinical assessments.