| Objective:To investigate the efficacy of three lymph node staging systems for survival prediction of intrahepatic cholangiocarcinoma(ICC),including log odds of positive lymph nodes(LODDS),lymph nodes ratio(LNR),and presence of regional lymph node metastasis(N staging),and to construct a nomogram model for survival prediction of intrahepatic cholangiocarcinoma(ICC).Methods:Data of 10112 patients with intrahepatic cholangiocarcinoma from 2004 to 2015 were collected from the SEER database,and clinical and pathological data of 802 patients with intrahepatic cholangiocarcinoma were included by screening criteria.Then,they were randomly divided into modeling group(n=402)and validation group(n=400).The predictive value of LODDS,LNR,and N stages in the prognosis of intrahepatic cholangiocarcinoma(ICC)was evaluated by calculating the C index,AIC information criteria,and AUC area.The independent risk factors related to overall survival(OS)were screened by univariate and multivariate Cox proportional risk regression analysis,and a prediction model was built based on the results.The Bootstrap method and validation group were used for validation.Calibration curves and receiver operating characteristic(ROC)curves were drawn to evaluate the prediction performance of the nomogram model.Results:Univariate and multivariate Cox proportional risk regression analysis showed that age,sex,tumor size,tumor grade,T stage,M stage,and LODDS,LNR,and N stages were independent risk factors for OS in patients with intrahepatic cholangioca-rcinoma.By calculating the C index(LODDS:0.701;LNR: 0.684;N staging: 0.677),AIC Information Criteria(LODDS:2169;LNR:2175;N staging: 2171),AUC area(LODS: 0.720 1 year,0.790 3 years;LNR: 0.702 for 1 year,0.762 for 3 years;N staging: 0.692 over 1 year,0.754 over 3 years)found that LODDS had the best predictive performance.Based on multivariate Cox regression analysis,the LODDS was incorporated into the model to establish a nomogram for survival prediction.Its C index was calculated to be 0.703(95%CI 0.666 ~ 0.740).The calculated calibrati-on curve showed a good agreement between the predicted and actual values.The data of the modeling group and the validation group were substituted into the nomogram model to draw the ROC curves of 1,3 and 5 years survival prediction for patients with intrahepatic cholangiocarcinoma,and the results of the area under the curve(1 year for the modeling group :0.720;3 years: 0.790;5 years: 0.811;1 year in the validation group: 0.659;0.760 3 years;5 years 0.740)showed that it had good predictive value.Conclusion:LODDS staging is the best predictor of survival for intrahepatic cholangiocarc-inoma.The LODDS-based nomogram model is a good model for intuitively predicting survival in patients with intrahepatic cholangiocarcinoma. |