Table 3.
Diagnostic efficacy of individual plasma biomarkers assessed by ROC curve analysis.
| Diagnostic comparison setting | Plasma biomarker | AUC | 95% CI | Cutoff (pg/mL) | Sensitivity (%) | Specificity (%) |
|---|---|---|---|---|---|---|
| Distinguishing Alzheimer’s disease from HC | p-Tau217 | 0.894 | 0.803–0.951 | 2.17 | 0.700 | 100.0 |
| p-Tau181 | 0.862 | 0.764–0.930 | 1.82 | 0.775 | 96.7 | |
| NfL | 0.832 | 0.728–0.909 | 83.21 | 0.800 | 77.5 | |
| GFAP | 0.769 | 0.655–0.861 | 49.53 | 67.5 | 85.0 | |
| Aβ42/40 | 0.751 | 0.635–0.846 | 0.075 | 95.00 | 25.0 | |
| Aβ1-40 | 0.699 | 0.579–0.803 | 130.2 | 60.0 | 85.00 | |
| Aβ1-42 | 0.584 | 0.457–0.704 | 10.80 | 25.0 | 90.00 | |
| Distinguishing CSVD from HC | GFAP | 0.881 | 0.786–0.943 | 36.39 | 95.0 | 70.0 |
| NfL | 0.886 | 0.793–0.947 | 84.35 | 72.5 | 77.5 | |
| p-Tau181 | 0.762 | 0.647–0.855 | 1.82 | 77.5 | 67.5 | |
| p-Tau217 | 0.718 | 0.600–0.820 | 1.45 | 75.0 | 67.5 | |
| Aβ1-40 | 0.526 | 0.399–0.650 | 130.8 | 32.5 | 85.0 | |
| Aβ42/40 | 0.414 | 0.293–0.544 | 0.0718 | 22.5 | 92.5 | |
| Aβ1-42 | 0.376 | 0.231–0.457 | 4.00 | 95.0 | 71.5 | |
| Distinguishing Alzheimer’s disease from CSVD | p-Tau217 | 0.877 | 0.781–0.941 | 3.07 | 52.5 | 85.0 |
| p-Tau181 | 0.782 | 0.671–0.871 | 3.24 | 52.5 | 77.5 | |
| Aβ1-40 | 0.733 | 0.617–0.832 | 123.7 | 70.0 | 62.5 | |
| NfL | 0.619 | 0.493–0.735 | 174.5 | 45.0 | 87.5 | |
| GFAP | 0.614 | 0.451–0.698 | 163.81 | 64.0 | 70.0 |
AUC, area under the receiver operating characteristic curve; CI, confidence interval; Cutoff, optimal cut-off value (maximized Youden’s index); Sen, sensitivity; Spe, specificity; Aβ, amyloid-β; p-Tau: phosphorylated tau protein; NfL, neurofilament light chain; GFAP, glial fibrillary acidic protein.