Nomograms for predicting overall survival and cancer-specific survival in elderly patients with epithelial ovarian cancer

Abstract

Abstract Background Epithelial ovarian cancer (EOC) is one of the most fatal gynecological malignancies among elderly patients. We aim to construct two nomograms to predict the overall survival (OS) and cancer-specific survival (CSS) in elderly EOC patients. Methods Elderly patients with EOC between 2000 and 2019 were selected from the Surveillance, Epidemiology, and End Results (SEER) database. Enrolled patients were randomly divided into the training and validation set at a ratio of 7:3. The OS and CSS were recognized as endpoint times. The independent prognostic factors from the multivariate analysis were used to establish nomograms for predicting the 3-, 5- and 10-year OS and CSS of elderly EOC patients. The improvement of predictive ability and clinical benefits were evaluated by consistency index (C-index), receiver operating characteristic (ROC), calibration curve, decision curve (DCA), net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Finally, the treatment efficacy of surgery and chemotherapy in low-, medium-, and high-risk groups were displayed by Kaplan-Meier curves. Results A total of 5,588 elderly EOC patients were obtained and randomly assigned to the training set (n = 3724) and validation set (n = 1864). The independent prognostic factors were utilized to construct nomograms for OS and CSS. The C-index of the OS nomogram and CSS nomogram were 0.755 and 0.700 in the training cohort. In the validation cohort, the C-index of the OS nomogram and CSS nomogram were 0.746 and 0.696. The calibration curve demonstrated good concordance between the predicted survival rates and actual observations. Moreover, the NRI, IDI, and DCA curves determined the outperformance of the nomogram compared with the AJCC stage system. Besides, surgery had no benefit on the prognosis in the high-risk group. Chemotherapy had a better prognosis in the medium-, and high-risk groups, but not for the low-risk group. Conclusions We developed and validated nomograms for predicting OS and CSS in elderly EOC patients to help gynecologists to develop an appropriate individualized therapeutic schedule.