Table 3 Performance of the models and SEPT9 for advanced neoplasia detection in the model validation cohort
Advanced Neoplasia classification | AA detection % | CRC detection % | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
AUC (95% CI) | Specificity % (95% CI) | Sensitivity % (95% CI) | NPV % (95% CI) | PPV % (95% CI) | D-AA | P-AA | All | Stage I | Stage II | Stage III | Stage IV | All | |
GALNT9, UPF3A | 0.896 (0.835–0.958) | 100 (91–100) | 78.8 (67–88) | 73.6 (60–85) | 100 (93–100) | 87 | 68.4 | 78.6 | 87.5 | 100 | 100 | 33.3 | 79.2 |
GALNT9, UPF3A, WARS, LDB2 | 0.864 (0.798–0.931) | 97.4 (87–100) | 62.1 (49–74) | 60.3 (47–72) | 97.6 (87–100) | 78.3 | 52.6 | 66.7 | 37.5 | 75 | 100 | 33.3 | 54.2 |
20 methylation biomarkers, sex | 0.553 (0.407–0.639) | 17.9 (8–34) | 92.4 (83–97) | 58.3 (28–85) | 65.6 (55–75) | 91.3 | 94.7 | 92.9 | 87.5 | 100 | 100 | 83.3 | 91.7 |
SEPT9 | 0.504 (0.389–0.618) | 100 (91–100) | 15.1 (8–26) | 41.1 (31–52) | 100 (69–100) | 4.3 | 10.5 | 7.1 | 0 | 37.5 | 50 | 50 | 29.2 |