Development and validation of nomograms for epithelial ovarian cancer: a SEER population-based, real-world study

Aim: To develop and internally validate nomograms to predict the overall survival (OS) and the cancer-specific survival (CSS) of patients with epithelial ovarian cancer (EOC). Methods: A total of 9001 EOC patients diagnosed between 2010 and 2013 were randomly divided into the training (n = 6301) and validation (n = 2700) cohorts. Nomogram and bootstrap validation were used to assess the predictive values of the models, including discrimination, calibration and clinical benefit. Results: In the validation cohort, the concordance statistic values were 0.733 for OS and 0.747 for CSS. Calibration plots and decision curve analyses demonstrated moderate accuracy and clinical applicability. Conclusion: Nomograms were user-friendly tools for guiding clinical treatment and estimating prognosis. Lay abstract We developed and validated nomograms, based on discrimination, calibration and decision curve analysis, that better predicted 3- and 5-year survival for patients with ovarian cancer. These nomograms were user-friendly and accurate for prognostic prediction, and they may be promising tools for predicting the prognosis of ovarian cancer patients.