The Prognostic Factors In Patients With Stage Ⅱ Or Ⅲ Gastric Cancer And Construction Of Prediction Model

Objective: Based on the clinicopathological characteristics of stage Ⅱor Ⅲ gastric cancer patients,an individualized model was established to predict the net survival benefit of adjuvant chemotherapy in patients with stage Ⅱor Ⅲ gastric cancer.Methods: In this retrospective study analysis,selected from the affiliated hospital of Qingdao university research into standard and in January 1,2013 to December 31,2019,during the surgery Ⅱor Ⅲ stage,354 cases of patients with gastric cancer,according to the and proportion were randomly divided into training set and validation set,the training set of 248 cases used to construct the prediction model,Patients’ clinicopathological characteristics were used as covariates and their association with overall survival(OS)and disease-free survival(DFS)with or without adjuvant chemotherapy was assessed.The differentiation and calibration of the model were internally verified using self-lifting sampling method.The validation set data of the remaining 106 cases were used for external validation of the model.Results: Of the 354 patients analyzed,242(68.6%)were male and 112(31.4%)were female,with a median(interquartile difference)age of 57 years(49-66 years).Age,lesion site,degree of differentiation,carcinoembryonic antigen,CA19-9,depth of invasion,lymph node metastasis and adjuvant chemotherapy were significantly correlated with overall survival and disease-free survival(P<0.05).The survival prediction model has good differentiation and calibration degree,and has relatively high bootstrap-correction consistency index in training set and validation set.In the validation set,the consistency index for overall survival was 0.693(95%CI,0.671-0.715)and for disease-free survival was 0.704(95%CI,0.681-0.728).Two histograms were built on the basis of specific input variables to estimate an individual’s net survival benefit from adjuvant chemotherapy.Conclusion: The survival prediction model can be used for individualized prediction of the expected survival benefit of adjuvant chemotherapy in patients with stage Ⅱor Ⅲgastric cancer.