Skip to main content
. 2018 Oct 10;13(10):e0204334. doi: 10.1371/journal.pone.0204334

Fig 2. Comparison of the survival prediction of patients with cervical cancer by serum CTC or SCC expression levels alone and the combination of both.

Fig 2

(A) and (B) ROC curve analysis of the optimal cutoff value of serum CTC or SCC level for distinguishing disease-free survival (DFS) from patients with cervical cancer. AUC was 0.694 for serum CTC (P = 0.001) and 0.702 for serum SCC (P = 0.001). (C) ROC analysis compares the survival prediction of patients with cervical cancer by serum CTC or SCC expression levels alone and the combination of both. The result shows that the area under the curve (AUC) of the combined serum CTC and SCC expression is the largest among the three predictors, which demonstrates that predictive accuracy of the combined risk model is better than those of FIGO stage and serum CTC or SCC alone. (D) ROC curve analysis of the predictive accuracy for DFS of patients with cervical cancer by FIGO staging system (AUC = 0.671, P = 0.004).