Zi-jian Zhang ^* Muhammad S. Azhar ZhongTao Liu Yun-Peng Huang ^* Wang Yongxiang Hui Zhou Li Xiong ^* Yu Wen ^* Heng Zou ^*

• Second Xiangya Hospital, Central South University, Changsha, China

Background: Clinically, the diagnosis and treatment of cholangiocarcinoma are generally different according to the location of occurrence, and the studies rarely consider the differences between different pathological types. Cholangiocarcinomas in large-and middle-sized intrahepatic bile ducts are mostly mucinous, while in small sized bile duct are not; mucinous extrahepatic cholangiocarcinomas are also more common than mucinous intrahepatic cholangiocarcinoma. However, it is unclear whether these pathological type differences are related to the prognosis. Methods: Data of total 22509 patients was analyzed from Surveillance, Epidemiology, and End Results program database out of which 22299 patients were diagnosed with common adeno cholangiocarcinoma while 210 were diagnosed with mucinous cholangiocarcinoma. Based on the propensity score matching (PSM) analysis, between these two groups' clinical, demographic, and therapeutic features were contrasted. The data were analyzed using Cox and LASSO regression analysis and Kaplan-Meier survival curves. Ultimately, overall survival (OS) and cancer specific survival (CSS) related prognostic models were established and validated in test and external datasets and nomograms were created to forecast these patients' prognosis. Results: There was no difference in prognosis between mucinous cholangiocarcinoma and adeno cholangiocarcinoma. Therefore, we constructed prognostic model and nomogram that can be used for mucinous and adeno cholangiocarcinoma at the same time. By comparing the 9 independent key characteristics i.e Age, tumor size, the number of primary tumors, AJCC stage, Grade, lymph node status, metastasis, surgery and chemotherapy, risk scores were calculated for each individual. By integrating these two pathological types in OS and CSS prognostic models, effective prognosis prediction results could be achieved in multiple datasets (OS: AUC 0.70-0.87; CSS: AUC 0.74-0.89). Conclusion: Age, tumor size, the number of primary tumors, AJCC stage, Grade, lymph node status, metastasis, surgery and chemotherapy are the independent prognostic factors in OS or CSS of the patients with mucinous and ordinary cholangiocarcinoma. Nomogram that can be used for mucinous and adeno cholangiocarcinoma at the same time is of significance in clinical practice and management of cholangiocarcinoma.