Development And Validation Of A Predictive Model Associated With Lymph Nodes Metastasis Of Gastric Signet Ring Carcinoma

Objective: Gastric signet ring cell carcinoma(GSRC)is a special type of gastric cancer(GC),its risk factors and prognostic factors are still controversial,lymph node metastasis(LNM)is an important prognostic factor affecting the survival of GC patients.It is critical to accurately predict lymph node metastasis(LNM)status for GSRC.The aim of this study is to establish and verify a nomogram to predict LNM status in GSRC patients.Methods: A total of 2789 GSRC patients were enrolled from Surveillance,Epidemiology and End Results(SEER)database and Yijishan Hospital of Wannan Medical College(YJS).A predictive model was established using logistic regression based on SEER cohort.The Concordance index(C-index),receiver operating characteristic(ROC)curve,plots of calibration curves and decision curve analysis(DCA)were used to evaluate the performance of the model.The YJS cohort was used to validate the robustness of the model.Results: Four independent predictors of LNM in SEER cohort selected through logistic regression analysis were identified including: primary sites,depth of infiltration,tumor size and number of lymph node examined(NLNE).Then,a nomogram was constructed by incorporating these predictors.The C-index of model were 0.800(95%CI = 0.781-0.819)and 0.837(95%CI = 0.784-0.890)in the SEER and YJS cohorts,respectively,the AUC of model were 0.798 and 0.826 in the SEER and YJS cohorts,respectively.These results show that the model has good predictive performance.In both SEER and YJS cohorts,the calibration plots testified forecast of nomogram have good accordance with actual LNM,the outcomes of DCA supported good clinical benefits.All of them showed excellent performance and clinical useful of the model in our study.Conclusion: Primary site of the tumor,depth of invasion,tumor size,and number of lymph nodes examined during surgery were independent predictors of the occurrence of LNM in GSRC,we proposed a predictive model for LNM in GSRC patients to assist physicians in making treatment decisions.The nomogram is advantageous for treatment planning in GSRC patients.