Establishment And Validation Of The Prognostic Nomogram For Early-stage Biliary Tract Cancer Based On SEER Database

Background:Biliary tract cancer refers to malignant tumors originating from the biliary system,mainly including bile duct cancer and gallbladder cancer,among which bile duct cancer can be classified into intrahepatic bile duct cancer,hilar cholangiocarcinoma and distal cholangiocarcinoma according to their specific anatomical locations.Currently,AJCC staging is commonly used to assess and predict the prognosis of biliary tract cancer,but for early stage biliary tract cancer,AJCC staging alone cannot meet the clinical needs.Therefore,this study aimed to establish and validate a new Nomogram for predicting the prognosis of cancer-specific survival(CSS)in early biliary tract cancer.Methods:In this study,cases were screened separately from the National Cancer Institute Surveillance,Epidemiology,and End Results Program database(SEER),and a total of 943 patients with early-stage intrahepatic cholangiocarcinoma,496 patients with early-stage hilar cholangiocarcinoma,615 patients with early-stage distal cholangiocarcinoma,and 867 patients with early-stage gallbladder cancer were screened according to different screening criteria.After that,the groups were randomly divided into training and validation groups according to the ratio of 7:3.In each study,Univariate and multivariate COX analyses were used to screen independent risk factors in the training group,and a prognostic staging model based on independent risk factors was established.The accuracy of Nomogram was verified by C-index,ROC curve,and calibration curve.DCA curves were used to evaluate the clinical utility of Nomogram and compared with AJCC staging.Finally,the risk stratification of each group of patients was performed according to the total score calculated by Nomogram,and the stratification effect of Nomogram was evaluated.Results:Multivariate COX analyses of patients with early-stage intrahepatic cholangiocarcinoma revealed 7 variables of age,sex,marital status,tumor size,histological grade,surgery and radiotherapy as independent risk factors for their CSS.Multivariate COX analyses of patients with early-stage hilar cholangiocarcinoma showed that 6 variables,including age,tumor size,histologic grade,T-stage,surgery and chemotherapy,were independent risk factors for CSS.Multivariate COX analyses of patients with early distal cholangiocarcinoma showed that 3 variables,including histological grade,AJCC stage and surgery,were independent risk factors for CSS.Multivariate COX analyses of patients with early-stage gallbladder cancer showed that the 5 variables of age,tumor size,histologic,AJCC stage and surgery were independent risk factors for CSS.The independent risk factors identified from the above analysis were modeled separately for their respective prognosis,and the Nomogram was subsequently drawn.The C-index of the Nomogram for patients with early-stage intrahepatic cholangiocarcinoma was 0.724 in the training group and 0.676 in the validation group;the C-index of the Nomogram for patients with early-stage hilar cholangiocarcinoma was 0.738 in the training group and 0.676 in the validation group.index was 0.738 in the training group and 0.749 in the validation group;Nomogram for patients with early-stage distal cholangiocarcinoma had a C-index of 0.690 in the training group and 0.727 in the validation group;The prognostic Nomogram for patients with early-stage gallbladder cancer had a C-index of 0.663 in the training group and a model C-index of 0.632 in the validation group.The ROC curves of the 4 Nomogram at 1,3 and 5 years showed that these models had good predictive accuracy,while the calibration curves also validated their good predictive performance.In addition,the DCA curves all showed that the 4 Nomogram outperformed the AJCC staging system.Conclusion:In patients with early-stage intrahepatic cholangiocarcinoma,early-stage hilar cholangiocarcinoma,early-stage distal cholangiocarcinoma,and early-stage gallbladder cancer,the Nomogram,which was constructed by combining general information and treatment modalities,has good predictive power for patients’ prognosis,and has better predictive power than the traditional AJCC staging.