Prognostic nomogram for early-stage cervical cancer in the elderly: A SEER database analysis

Background: To identify key clinical factors affecting the survival of elderly patients with early-stage cervical cancer and to construct a nomogram for predicting their prognosis. Methods: Patients (aged >= 65 years old) diagnosed with cervical cancer between 2004 and 2015 at clinical stages IA to IIA were included in this study. Diagnosis was confirmed via pathological examination, and the cases were randomly divided into a training or a validation group in a 7:3 ratio. Univariate and multivariable Cox regression analyses were performed to identify independent factors affecting the prognosis of elderly early-stage cervical cancer patients, based on which a nomogram was constructed to predict their 12-, 24- and 36-month overall survival (OS). The nomogram's performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves and decision curve analysis (DCA) curves. Results: A total of 686 patients were identified as eligible and assessed. Multivariable Cox proportional hazard regression analysis revealed that age, tumor diameter, marital status and surgical intervention were independent prognostic factors for elderly individuals with early-stage cervical cancer, which were then used to construct the nomogram. The calibration curves showed a strong correlation between predicted and observed survival rates, and Kaplan-Meier survival curves for different risk subgroups demonstrated significant survival differences (P < 0.001). DCA confirmed the nomogram's clinical utility in predicting the prognosis of elderly patients with early-stage cervical cancer. Conclusion: The prognostic model developed in this study can accurately predict the OS of elderly patients with early-stage cervical cancer, showing high concordance with actual clinical outcomes.