Development and Internal Validation of a Nomogram for Predicting Recurrence in Endometrial Cancer Based on...

Area of Science:

• Gynecologic Oncology
• Cancer Biomarkers
• Clinical Prediction Models

Background:

• Endometrial cancer (EC) is a prevalent gynecologic malignancy.
• Accurate identification of patients at high risk for EC recurrence is crucial for personalized management.
• Developing robust prediction models aids in tailoring follow-up and treatment strategies.

Purpose of the Study:

• To develop and validate a nomogram for predicting the risk of endometrial cancer recurrence post-treatment.
• To evaluate the model's predictive accuracy and clinical utility.

Main Methods:

• Retrospective analysis of 486 endometrial cancer patients.
• Development of a nomogram using identified predictors in a training group (n=389).
• Validation of the nomogram using various statistical metrics including ROC, calibration plots, NRI, IDI, and DCA.

Main Results:

• A nomogram incorporating six predictors was constructed and validated.
• The model demonstrated high discrimination and calibration (AUC=0.890).
• Significant improvements in prediction accuracy (NRI, IDI) were observed compared to existing patterns.

Conclusions:

• A validated nomogram based on pathological and immunohistochemical factors accurately predicts EC recurrence.
• The developed nomogram offers superior predictive performance compared to the ESGO-ESTRO-ESP pattern.
• This tool can be utilized for clinical risk stratification and personalized patient management.