Abstract
Background
The purpose of this study was to Construct and validate a novel nomogram to predict cancer-specific survival (CSS) in patients with ependymoma.
Methods
Clinical data from the Surveillance, Epidemiology, and End Results database spanning 2000 to 2018 were used for the analysis. Data were randomly categorized into development and validation groups in a 7:3 ratio. Univariate and multivariate Cox regression analyses and LASSO regression were conducted to identify independent risk factors for CSS. Predictive models were evaluated using calibration plots, concordance index (C-index), and area under the receiver operating characteristic curve (AUC). Additionally, decision curve analysis and clinical impact curves were conducted.
Results
The final sample comprised 2,340 patients. Multivariate analysis identified race, age, histological type, surgery, and tumor site as independent predictors of CSS. A nomogram incorporating these risk factors was developed to predict CSS in patients with ependymomas. The calibration plot shows strong agreement between predicted and actual values. The C-index for the training cohort was 0.746 (95% CI: 0.715–0.777), and for the validation cohort, it was 0.743 (95% CI: 0.698–0.788). Additionally, both AUC and decision curve analysis analyses indicated robust performance and clinical significance benefits. Kaplan–Meier curves further demonstrated the nomogram’s strong ability to predict patient outcomes.
Conclusions
Our nomogram may be of great value in predicting the prognosis of patients with ependymoma. This predictive model will assist doctors and patients in devising effective clinical strategies.