Skip to main content
. 2017 Sep 6;12(9):e0184338. doi: 10.1371/journal.pone.0184338

Table 4. Statistical values of all metrics for differentiating between lung cancers and benign lesions and NSCLC and SCLC.

Metrics AUC P Value Sensitivity Specificity
Lung cancers VS Benign lesions
    CEA 0.769 <0.001a 76.92% 65.63%
    CYFRA21-1 0.744 <0.001a 40.83% 100.00%
    SCC-Ag 0.641 0.012a 40.83% 84.38%
    NSE 0.756 <0.001a 76.33% 68.75%
    SUVmax 0.767 <0.001a 88.76% 59.38%
    Combined tumor markers 0.871 <0.001a 66.86% 93.75%
    SUVmax+Tumor markers 0.902 <0.001a 84.02% 87.50%
NSCLC VS SCLC
    CEA 0.603 0.037a 50.00% 70.25%
    CYFRA21-1 0.623 0.013a 72.92% 51.24%
    SCC-Ag 0.600 0.043a 56.25% 62.81%
    NSE 0.805 <0.001a 60.42% 90.08%
    SUVmax 0.559 0.229 75.00% 44.63%
    Combined tumor markers 0.911 <0.001a 75.00% 92.56%
    SUVmax+Tumor markers 0.916 <0.001a 87.50% 84.30%

a P<0.05.

AUC: area under the curve; CEA: carcino embryonie antigen, CYFRA21-1: cytokeratin 19 fragments, SCC-Ag: squmaous cell carcinoma antigen, NSE: neuron specific enolase, SUVmax: maximal standard uptake value, SCLC: small-cell lung carcinoma, NSCLC: non-small-cell lung carcinoma.