Multivariate Analysis Of Collecting Duct Carcinoma And Establishment Of Visualization Model

Objective: Investigating the factors related to the prognosis of patients with collecting duct carcinoma and seting up the Nomogram model,providing a method for clinicians to evaluate the prognosis of patients with collecting duct carcinoma.Methods: From 2004 to 2016,the clinical data of all patients with collecting duct carcinoma diagnosed by pathological diagnosis were downloaded from Epidemiology,and End Results Database(SEER)in the United States.According to the set inclusion and exclusion criteria,we get the final number of included cases.The continuous data of included patients were converted into classified data and assigned values.Descriptive statistical analysis was performed on enrolled patient variables,Kaplan-Merier univariate survival analysis and log-rank method was used to test differences between groups.The relevant variables affecting the prognosis of patients with collecting duct carcinoma were screened out by setting the best level p=0.05,and the statistically significant variables were add to cox proportional-hazards model and the independent risk factors affecting the prognosis of collecting duct carcinoma were obtained.Then the independent risk factors affect the prognosis of patients with collecting duct carcinoma were add to the Nomogram model.Using C-index and area under ROC curve to test the conformity of the nomogram model.The compliance of the Nomogram model is verified by drawing a standard curve.Results: A total of 235 patients with collecting duct carcinoma were included based on SEER databases.Kaplan-merier univariate survival analysis shows that patients’ age,Grade,T stage,N stage,M stage and whether surgical treatment are associated with prognosis of patients with collecting duct carcinoma(P<0.05).Cox multivariate proportional risk regression model suggests that T stage,M stage and surgical treatment are independent risk factors for the prognosis of patients with collecting duct carcinoma.Based on Cox multivariate proportional risk regression model,a line chart model was constructed.After internal correcting,the consistency index is 0.737 and the area under ROC curve is 0.80.The standard curve fits the straight line with the slope of 1 better,and the constructed Nomogram modell is close to the actual risk,and the model conforms well.Conclusions: 1.Kaplan-merier univariate survival analysis shows that the factors influencing the prognosis of patients with renal collecting duct carcinoma are age,Grade,T stage,N stage,M stage and whether surgical treatment.Cox multivariate survival analysis shows that the independent factors affecting the prognosis of patients with collecting duct carcinoma are T stage,M stage and whether surgical treatment.2.Using discrimination and compliance to test the Nomogram model shows good.This model can be used as a method for clinicians to evaluate the survival and prognosis of patients with collecting duct carcinoma.