BACKGROUND

Clear cell adenocarcinoma of the cervix (CCAC) is a rare malignancy without a well-established prognostic model. Our study aimed to develop and validate a nomogram to estimate overall survival in CCAC patients.

METHODS

We collected data on 630 CCAC patients from the Surveillance, Epidemiology, and End Results (SEER) database (2000-2021). Missing clinicopathological data were imputed using the missForest package. The imputed dataset served as the training cohort, while the dataset with missing values removed acted as the validation cohort. The nomogram's performance was assessed through discriminative ability, calibration, C-index, AUC, and calibration plots. Clinical benefits were compared against the International Federation of Gynecology and Obstetrics (FIGO) staging using decision curve analysis (DCA), net reclassification index (NRI), and integrated discrimination improvement (IDI).

RESULTS

The nomogram, based on nine variables, demonstrated strong discriminative power, with C-index values of 0.82 for the training cohort and 0.81 for the validation cohort, and AUCs exceeding 0.80 in both sets. Calibration plots showed a strong agreement between the nomogram's predictions and actual outcomes in both cohorts. The NRI values for the training set were 0.21 for 3-year, 0.20 for 5-year, and 0.30 for 10-year overall survival (OS) predictions, and for the validation set were 0.34 for 3-year, 0.25 for 5-year, and 0.31 for 10-year OS predictions. The IDI results for the training set were 0.17 across 3-, 5-, and 10-year OS predictions, and for the validation set were 0.21 for 3-year, 0.17 for 5-year, and 0.15 for 10-year OS predictions. The nomogram significantly outperformed the FIGO criteria ( < 0.01), and DCA highlighted its superior clinical utility in identifying high-risk patients.

CONCLUSION

The nomogram, which integrates treatment data, was successfully developed and validated to assist clinicians in assessing the prognosis of CCAC patients. It demonstrated superior performance to FIGO criteria in predicting overall survival.