Identification of prognostic risk factors and construction of nomograms for elderly patients with stage III–IV endometrioid endometrial carcinoma: a study based on the SEER database

Background: The incidence of endometrial cancer among elderly women has been increasing year by year, and endometrioid endometrial carcinoma (EC) is the predominant histological subtype of endometrial cancer. Unfortunately, elderly patients with advanced-stage endometrioid EC continue to experience unfavorable prognostic outcomes. To date, reliable instruments for individualized prognosis estimation among elderly patients with stage III–IV endometrioid EC are still lacking. Therefore, this investigation sought to determine prognostic indicators and develop nomograms for the prediction of overall survival (OS) and cancer-specific survival (CSS) in this patient population.

Methods: This study utilized data from the Surveillance, Epidemiology, and End Results (SEER) database, analyzing eligible elderly patients (aged ≥60 years) with stage III–IV endometrioid EC. The Kaplan-Meier approach was employed for survival analysis, while univariate and multivariate Cox regression analyses were used to identify factors independently associated with prognosis. Based on these characteristics, nomograms were constructed and subsequently evaluated via receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA) to assess their discriminative performance, model fit, and potential clinical utility.

Results: Multivariable Cox proportional hazards regression showed that age, tumor grade, N stage, chemotherapy, radiotherapy, surgery, and brain metastasis were independently linked to OS as well as CSS. The nomograms incorporating these variables demonstrated good predictive performance in both the training and validation cohorts. The calculated area under the curve (AUC) for OS ranged from 0.76 to 0.88, while those for CSS ranged from 0.77 to 0.84. Calibration curves demonstrated excellent agreement between the predicted and observed survival probabilities. Moreover, DCA indicated that the nomograms provided favorable net clinical benefits across a broad range of threshold probabilities.

Conclusions: The constructed nomograms demonstrated satisfactory performance in predicting survival outcomes among elderly patients with advanced endometrioid EC, with good discriminatory ability. These models may facilitate individualized risk stratification, although further external validation is warranted before routine clinical application.