Development of a novel web-based calculator for predicting overall survival in early-onset cervical cancer patients with positive lymph node metastasis

Background: Cervical cancer (CC) is a malignant tumor affecting the female genital system and ranks as second most common cancer among younger women. This study aimed to identify key clinicopathological and lymph nodal characteristics associated with the overall survival (OS) of CC patients aged <45 years and develop an interactive webbased calculator to assess patient prognosis. Methods: The Surveillance, Epidemiology and End Results (SEER) database was searched for cases diagnosed with CC from 2004 to 2015, which were then randomly divided into a training (n = 3720, 70%) and a validation (n = 1661, 30%) set. Least absolute shrinkage and selection operator (LASSO) regression was used to identify relevant predictors and construct a nomogram incorporating the most significant variables. In addition, its performance was assessed using C-index values, area under curve (AUC) values, calibration plots and KaplanMeier curves, and an online prediction tool was constructed. Results: In the training cohort, the C-index for the proposed nomogram was 0.809 (95% Confidence Interval (CI): 0.802-0.816), and in the validation set, it was 0.811 (95% CI: 0.801-0.821). The AUC values for 1-, 3- and 5-year OS were 0.880, 0.856 and 0.842 in the training set and 0.911, 0.843 and 0.829 in the validation set, respectively. The calibration curves demonstrated the reliable predictive performance of the nomogram, with the nomogram demonstrating good calibration and discrimination abilities in the validation set. Conclusions: The developed nomogram and online tool for CC patients aged <45 years demonstrated promising utility in potentially assisting clinicians to predict patient prognosis and develop more informed treatment strategies for these patients.