A nomogram for predicting recurrence in endometrial cancer patients: a population-based analysis.

Objective: Endometrial cancer recurrence is one of the main factors leading to increased mortality, and there is a lack of predictive models. Our study aimed to establish a nomogram predictive model to predict recurrence in endometrial cancer patients. Method: Screen 517 endometrial cancer patients who came to Nanjing Drum Tower Hospital from 2008 to 2018. All these data are listed as the training group, and then 70% and 60% are randomly divided into verification groups 1 and 2. Univariate, Multivariate logistic regression, stepwise regression were used to select variables for nomogram. Nomogram identification and calibration were evaluated by concordance index (c-index), area under receiver operating characteristic curve (AUC) over time and calibration plot Function. By decision curve analysis (DCA), net reclassification index (NRI), integrated discrimination improvement (IDI), we compared and quantified the net benefit of nomogram and ESMO-ESGO-ESTRO model-based prediction of tumor recurrence. Results: A nomogram predictive model of endometrial cancer recurrence was established with the eight variables screened. The c-index (for the training cohort and for the validation cohort) and the time-dependent AUC showed good discriminative power of the nomogram. Calibration plots showed good agreement between nomogram predictions and actual observations in both the training and validation sets. Conclusions: We developed and validated a predictive model of endometrial cancer recurrence to assist clinicians in assessing recurrence in endometrial cancer patients.