Skip to main content
. 2021 Jun 10;12:656632. doi: 10.3389/fimmu.2021.656632

Table 3.

Prediction performance of the signature in the validation set by the fixed cut-off point.

Distinction AUC (95% CI) Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI) NPV (95% CI) Accuracy (95% CI) Disease prevalence (95% CI) P value
AR vs STA (n = 179) 0.84 (0.778 - 0.899) 70.18 (56.60% - 81.57%) 80.33 (72.16% - 86.97%) 62.5 (52.85% - 71.25%) 85.22 (79.31% – 89.66%) 77 (70.24% - 83.03%) 31.84 (25.09% - 39.21%) <0.0001
AR vs OGIs (n=142) 0.703 (0.617 - 0.789) 70.18 (95% CI: 56.60 - 81.57) 60.00 (95% CI: 48.80 - 70.48) 54.05 (95% CI: 46.31 - 61.61) 75.00 (95% CI: 66.02 - 82.24) 64 (95% CI: 55.61 - 71.96) 40.14 (32.01 to 48.69) <0.0001
AR vs STA + OGIs (n = 264) 0.783* (0.721 - 0.845) 70.18 (56.60% - 81.57%) 71.98 (65.33% - 77.98%) 40.82 (34.35% - 47.62%) 89.76 (85.37% - 92.94%) 72 (65.74% - 76.95%) 21.59 (16.78% - 27.05%) <0.0001

*AUC was calculated based on AR predicted probability of the signature for patients including AR (n=57), STA (n=122), OGIs (n=85), and no-AR (STA and OGIs, n=207) in QC-passed samples. PPV, positive predictive value; NPV, negative predictive value.