| Objective The purpose of this study was to explore the risk factors influencing the prognosis of patients with epithelial ovarian cancer(EOC)and to construct a nomogram model that can individually predict the cancer-specific survival(CSS)of patients with EOC,so as to help clinicians select personalized treatment regiments and analyze the survival prognosis of patients with different stages of EOC.Methods Clinical data of 13,697 patients diagnosed with epithelial ovarian cancer from 1975 to 2018 were collected from the SEER database of the National Cancer Institute of the United States.After screening according to inclusion and exclusion criteria,a total of 8978 eligible patients with epithelial ovarian cancer were included.Clinical characteristics of the patients were statistically analyzed using SPSS 26.0.The patients were randomly divided into the training cohort and validation cohort at 7:3.Univariate Cox regression analysis was performed on the variables,and then variables with P<0.05 were included in multivariate Cox regression analysis to obtain independent prognostic factors affecting CSS.The nomogram model was constructed by using R language software.Internal validation of the line-chart was performed using C-index,AUC,calibration curve,and decision Curve analysis(DCAs).Data of 135 patients with epithelial ovarian cancer in the gynecological Surgery department of Zhejiang Cancer Hospital from December 2006 to December 2019 were collected for external validation.Kaplan-Meier curves were plotted to compare survival outcomes between different risk groups.Results Age,grade,histological type,tumor side,TNM stage,number of resected regional lymph nodes,number of positive lymph nodes,chemotherapy and postoperative residual lesions were independent prognostic factors of CSS.Based on these factors,a nomogram was constructed to predict CSS in patients with epithelial ovarian cancer.The C-index of training cohort,internal validation cohort and external validation cohort were 0.757,0.741 and 0.703,respectively.The calibration curve showed that there was a good consistency between the survival status predicted by the line map and the actual survival status.The results of AUC and DCA showed that the nomogram had good clinical practicability and predictive performance.Kaplan-Meier curve differences among different risk subgroups were statistically significant.Conclusion1.In this study,age,grade,histological type,tumor side,TNM stage,number of resection regional lymph nodes,number of positive lymph nodes,chemotherapy and postoperative residual lesions were independent prognostic factors affecting cancer-specific survival in patients with epithelial ovarian cancer.2.The nomogram model constructed in this study has good correlation and accuracy between the prediction performance and the actual results,and can accurately predict the cancer-specific survival of patients with epithelial ovarian cancer after initial treatment.3.This study provides a convenient,intuitive and practical prediction model for gynecological oncologists,which can help doctors evaluate individual risk of ovarian cancer and select appropriate treatment plan,improve the quality of life of patients to a certain extent,and provide help and support in clinical work and scientific research. |
PDF Full Text Request