Construct Prognostic Risk Prediction Model Of Lymph Node Metastasis In Early Gastric Cancer

Background and Objective:Gastric cancer is a common malignant tumor of digestive tract,which is highly malignant and easy to metastasize.In patients with gastric cancer,lymph node metastasis is the most common and major route of metastasis,as well as an important cause of death.At present,there are few studies on the prognosis of early gastric cancer patients with lymph node metastasis.Therefore,based on the SEER database,this study established a prognostic risk assessment model with good predictive ability by analyzing the relevant clinicopathological features and independent risk factors of patients with early gastric cancer lymph node metastasis,so as to better assist clinicians in assessing the disease and providing accurate treatment.Methods:A search of The Surveilance Epidemiology,and End Results(SEER)database of the National Cancer Institute,based on inclusion and exclusion criteria,Data from 567 patients diagnosed with early gastric cancer with lymph node metastasis between 2004 and 2018 were included.The data were randomly divided into the training group and the verification group in a ratio of 7:3.First,the difference of clinical baseline characteristics between the two groups was analyzed.Univariate Cox regression analysis was performed for the training group using R Studio 4.2.0 to screen out statistically significant variables.Multi-factor Cox stepwise regression analysis was continued to screen out independent risk factors that could be included in the prediction model,and a column graph was constructed to predict the impact of each independent risk factor on Overall Survival(OS)of target patients.In addition,the 1-year,3-year and 5-year survival rates were predicted according to the total scores of each risk factor item in the graph.Finally,the model ROC curve,calibration chart and decision curve were drawn to evaluate the evaluation effect of the model.Results:Results of univariate analysis in the training group: age,sex,race,primary site,tumor differentiation degree,tumor diameter,T stage,N stage,and whether lymph node dissection was performed were independent factors influencing the prognosis of lymph node metastasis in early gastric cancer.Results of multi-factor analysis in the training group: age,tumor diameter,primary site,degree of tumor differentiation,T stage,N stage and whether lymph node dissection was performed were independent factors influencing the prognosis of lymph node metastasis in early gastric cancer.According to this,the column diagram is drawn,which shows good discrimination and calibration performance in both training set and verification group.The C index of the training group and the verification group were 0.720 and 0.684,respectively,which were both significantly higher than that of the TNM staging system,indicating higher prediction accuracy of the model.Conclusion:The established prediction model has high accuracy in predicting the prognosis and survival risk of patients with early gastric cancer lymph node metastasis,and can provide reference for clinicians to conduct individualized treatment for different patients.