Analysis Of Prognostic Factors Of Tongue Squamous Cell Carcinoma Based On SEER Database And Establishment Of Prediction Model

Objective:The aim of this study is to investigate the factors influencing the prognosis of patients with tongue squamous cell carcinoma(TSCC)through the SEER database,and to establish and verify the nomogram survival prediction model based on these factors,so as to achieve the individualized quantitative assessment of the risk of death in TSCC patients and provide a new direction for clinical work.Methods:Through the Surveillance,Epidemiology,and End Results(SEER)established by the National Cancer Institute(NCI),clinical data of 12595 patients diagnosed with TSCC from 2010 to 2015 were collected,which were divided into Training Set and Validation Set in a ratio of 7:3.A prediction model was established on the basis of the training set,and possible influencing factors were screened by univariate Cox regression analysis,Then the factors with p <0.1 in the univariate regression analysis was included in the multivariate Cox regression analysis,finally,a prediction model was obtained.Based on the multi-factor Cox regression model,the survival prediction model Nomogram plot was established,and the Area Under Curve(AUC)and the concordance index(C-index)of the Receiver Operating Characteristic(ROC)were used to evaluate the differentiation of the prediction model in the training set and the verification set respectively.Calibration curves were used to evaluate the fit of the prediction model.Kaplan-Meier survival curve was used to describe the survival differences among groups,and a bilateral p <0.05 was defined as a significant difference.Results:1.Cox univariate analysis showed that age(p <0.001),race(p <0.001),primary site(p = 0.014),tumor grade(p <0.001),clinical stage(p <0.001),T stage(P <0.001),N stage(p <0.001),M stage(p <0.001),distant metastatic site(p <0.001)and operation method of primary site(p <0.001)were significantly associated with the prognosis of TSCC patients.2.According to the results of Cox multivariate analysis,a total of 8 independent influencing factors of TSCC patients were included in the final nomogram prediction model,including age(p <0.001),race(p <0.001),primary site(p <0.001),tumor grade(p <0.001),T stage(p <0.001),N stage(p <0.001),distant metastatic site(p <0.001),and operation method of primary site(p <0.001).3.The nomogram prediction model had a good fitting degree,and the predicted survival rate had a higher coincidence rate with the actual survival rate of the patient.4.The nomogram prediction model had better discrimination than the TNM stage.The C-index of the training set were 0.725(95% CI: 0.715~0.735)and 0.674(95%CI: 0.664~0.684);the C index of the verification set were respectively 0.726(95% CI: 0.712~0.740)and 0.671(95% CI: 0.647~0.685).5.The accuracy of the nomogram model to predict the 3-year and 5-year overall survival rate of patients was better than that of traditional TNM stage.In the training set,the AUC of the nomogram prediction model predicted the 3-year and 5-year overall survival were 0.746(95% CI: 0.744~0.764)and 0.738(95% CI: 0.729~0.756),respectively,while the AUC of TNM stage predicted 3-year and 5-year overall survival were 0.694(95% CI: 0.686~0.708)and 0.671(95% CI: 0.665~0.693).In the validation set,the AUC of the nomogram prediction model predicted 3-year and 5-year overall survival were 0.745(95% CI: 0.728~0.760)and 0.741(95% CI: 0.721~0.761),respectively,while the AUC of TNM stage predicted 3-year The 5-year overall survival AUC were 0.643(95% CI: 0.623~0.658)and 0.637(95% CI: 0.610~0.656).6.In patients with stage III,IVA,and IVB,T1-2N1-3M0 was all related to the patients' overall survival(p <0.001).Cox multivariate analysis showed that among patients with stage III,IVA,and IVB,T1-2N1-3M0 was independent factor that affected the survival of patients(p <0.001).7.The 5-year survival rate of patients with stages T1-2N1-3M0 was significantly higher than that of patients with TNM stage II(64.4% vs.54.7%,p <0.001).Conclusion:1.For TSCC patient,age,race,primary site,tumor grade,T stage,N stage,and the primary site of metastasis surgical site is an independent prognostic factor.2.The nomogram prediction model established in this study has good discrimination and fitting degree,and can individually predict the 3-year and 5-year survival rates of TSCC patients,to better guide follow-up treatment plans.3.There is a big difference in the prognosis of TSCC patients in the same stage.The TNM stage has a limited ability to predict the prognosis of TSCC patients.