Development and validation of a circulating tumor DNA-based optimization-prediction model for short-term postoperative recurrence of endometrial cancer

Figure 1 Lasso regression analysis was used for variable selection to identify the most predictive clinical factors. A: The Lasso weight coefficient distribution of the screened variables; B: Penalty coefficient (λ) and binomial deviation diagram; C: Venn diagram showing features selected by univariate Cox regression and Lasso regression.

Figure 2 Prognostic nomogram used to predict the 1-year, 1. 5-year, and 2-year probability of postoperative recurrence-free survival in patients with endometrial cancer. ctDNA: Circulating tumor DNA; CA125: Carcinoembryonic antigen 125.

Figure 3 Calibration curves of the nomogram. A: Training set; B: Validation set. RFS: Recurrence-free survival.

Figure 4 Receiver operating characteristic curves of the nomogram. A: Training set; B: Validation set.

Figure 5 Decision curve analysis of the nomogram. A: 1-year decision curve; B: 1.5-year decision curve; C: 2-year decision curve.

Figure 6 Survival analysis and cumulative hazard analysis of the training and validation sets. A: Kaplan-Meier survival analysis curve for the training set; B: Kaplan-Meier survival analysis curve for the validation set; C: Cumulative hazard curve for the training set; D: Cumulative hazard curve for the validation set.