| Â | AUC (95% CI) | P value | Sensitivity (%) | Specificity (%) | Optimal cutoff |
|---|
PET ATN biomarkers |
A: Global SUVR | 0.93 (0.88–0.97) |  < 0.001*** | 86.7 | 89.4 | 1.25 |
T: MTL SUVR | 0.94 (0.82–0.96) |  < 0.001*** | 82.2 | 95.7 | 1.30 |
T: NEO-T SUVR | 0.95 (0.91–0.99) |  < 0.001*** | 86.7 | 97.9 | 1.23 |
N: metaROI SUVR | 0.83 (0.76–0.90) |  < 0.001*** | 80.9 | 72.2 | 1.37 |
Plasma ATN biomarkers |
A: Aβ42/Aβ40 ratio | 0.63 (0.54–0.73) | 0.011 | 80.9 | 45.6 | 0.04 |
T: P-tau181, pg/ml | 0.93 (0.88–0.98) |  < 0.001*** | 93.3 | 85.1 | 3.23 |
N: T-tau, pg/ml | 0.56 (0.46–0.66) | 0.265 | 17.8 | 97.9 | 5.76 |
N: NfL, pg/ml | 0.65 (0.54–0.75) | 0.005* | 81.1 | 51.1 | 13.63 |
- Each participant was classified as either Aβ-positive (Aβ+) or Aβ-negative (Aβ−) based on 18F-florbetapir PET imaging findings. Each plasma and PET imaging ATN biomarker was examined by receiver operating characteristic (ROC) curve analysis in relation to its ability to predict the Aβ status. The optimal cutoff for each biomarker was selected as the point that maximized the Youden’s index according to the ROC curve analysis; the corresponding sensitivity and specificity were subsequently calculated. Unadjusted P values are presented. Significant P values (P < 0.05) are marked in bold, and those surviving multiple comparisons (Bonferroni’s correction) are marked with asterisks (***, Pc < 0.001; *, Pc < 0.05)
- PET Positron emission tomography, A/T/N Amyloid/Tau/Neurodegeneration, SUVR Standardized uptake value ratio, MTL Medial temporal lobe, NEO-T Temporal neocortex, metaROI meta-analytically derived region of interest, p-tau181 Tau phosphorylated at threonine 181, t-tau Total tau, NfL Neurofilament light chain, AUC Area under the curve, CI Confidence interval
