| Objectives:The aim of this study is to investigate the pertinent markers of patients’ peripheral blood samples taken one week prior to surgery and the clinicopathological characteristics of postoperative patients with esophageal squamous cell carcinoma(ESCC),and to develop a clinical prediction model to assess the predictive value of the combined indicators for the long-term prognosis of ESCC patients,as well as the model’s predictive ability.Methods:A descriptive and predictive study was conducted.Clinical data of patients who underwent radical resection of esophageal squamous cell carcinoma at Department of Thoracic Surgery,Northern Jiangsu People’s Hospital between January 2015 and July 2017 were collected.The Receiver Operating Characteristic Curve(ROC curve)was drawn in this study.The optimal cut-off values of PLR,NLR,LMR,SII,SIRI,MPV/PCT,AFR,PNI,CA19-9,CA72-4 and CEA were determined,that is,Youden index(sensitivity+specificity-1)was calculated,and the value corresponding to the maximum value of Youden index was defined as the optimal cut-off value.Peripheral blood indicators were grouped and assigned according to this standard.In addition,the area under the ROC curve(AUC)was used to further compare the predictive ability of multiple blood indicators and tumor markers for lymph node metastasis,invasion depth,differentiation degree and prognosis of patients with esophageal cancer,and to evaluate the sensitivity and specificity of the model.Kaplan-Meier method was used to determine the 1-,3-,and 5-year survival rates of patients with esophageal squamous cell carcinoma.Survival curves were drawn to compare the two groups of patients.The Log rank test was used to assess the significance of survival between groups.Univariate analysis was used to determine the variables with significant effect on prognosis,and multivariate COX regressionanalysis was used to establish the prediction model of independent prognostic factors for esophageal cancer.The survival package and rms package in R software were used to draw the Nomogram for predicting the 3-year and 5-year survival rates of patients with esophageal squamous cell carcinoma.The Harrell’s concordance index(C-index)was used to evaluate the performance of the model,with higher C-index indicating better performance.To further evaluate the predictive accuracy of the model,a graphical calibration method was used to evaluate the agreement between the predicted survival rate and the actual survival rate.Excel 2021 and SPSS 26.0 were used for data sorting and summary and statistical analysis,respectively.Medcalc software was used to calculate the cut-off value,and R 4.0 software was used to calculate the survival rate by Kaplan-Meier method,draw the Nomogram and analyze the model validation.If the continuous variables obey the normal distribution,the data distribution was described by the mean ± standard deviation,and the group t test was used for analysis.If the normal distribution was not followed,the median[P50(P25-P75)]was used to describe the data distribution,and the Wilcoxon rank sum test was used for analysis.Categorical data were expressed as frequency and percentage(%),and the chi-square test or Fisher’s exact test was used for comparison of categorical data between groups.Logistic regression analysis was used to analyze the related risk factors of tumor differentiation,invasion depth and lymph node metastasis.Survival analysis was performed by Kaplan-Meier method and survival curve(Log-rank test)analysis to compare the survival rate between groups.Multivariate COX regression model was used to analyze the factors affecting the survival of ESCC.P<0.05 was considered statistically significant.Results:1.There were significant differences in tumor location,tumor size,depth of invasion,vascular invasion,lymph node metastasis,degree of differentiation and CEA between the survival group and the death group,and there were borderline differences in age and AFR between the survival group and the death group.2.The optimal cut-off values for various variables were established using the ROC curve,and the variations in survival rates under various cut-off values were examined.It was discovered that the survival rates of individuals with esophageal cancer under various haematological characteristics varied significantly.The survival rate of patients was substantially decreased by elevated CEA and CA72-4,as well as by decreased LMR,SIRI,and AFR.3.In ESCC patients,being male,having a history of alcohol use,and having increased CEA are all risk factors for lymph node metastasis.In ESCC patients,both reduced LMR and elevated SIRI are risk factors for tumor vascular invasion.In ESCC patients,advanced age is a risk factor for tumor neuroinvasion.4.The findings of the correlation study demonstrated a statistically significant relationship between SIRI and gender,smoking and drinking experience,tumor size,depth of invasion,5-year survival rate,and survival time.The correlation between SII and tumor size,invasion depth,and survival time was significantly significant.PNI was numerically significantly correlated with gender,smoking history,and tumor location.The depth of invasion was correlated with PLR,and the difference was statistically significant.NLR was numerically significantly correlated with gender,smoking history,and tumour size 4.5 cm.LMR was statistically significantly correlated with gender,smoking and alcohol history,tumor site,invasion depth,and 5-year survival rate.CA19-9 was correlated with survival time,and the difference was statistically significant.The difference in CA72-4 and the variables smoking,drinking,tumor location,5-year survival rate,and survival duration was statistically significant.The 5-year survival rate,survival time,smoking and alcohol were all correlated with CEA,and the difference was statistically significant.5.According to the findings of univariate COX regression analysis,the prognosis of patients with esophageal cancer was influenced by LMR,SIRI,AFR,CEA,CA724,tumour location,vascular infiltration,degree of differentiation,and depth of penetration.Increased SIRI,decreased AFR,increased CA72-4,and increased CEA were identified as independent risk variables for the prognosis of esophageal cancer patients by multivariate COX regression analysis.6.The combined model(SIRI+AFR+CEA+CA72-4)had the highest C-index,at 0.728(95%CI:0.687-0.728),indicating that it was the most accurate model.The comprehensive model including those four indicators and clinicopathological features was superior to those of four indicators alone,according to the statistical test of C-index.Conclusions:1.Tumor location,tumor size,depth of invasion,vascular invasion,lymph node metastasis,degree of differentiation,CEA,age,AFR were associated with prognosis.Gender,CEA and alcohol consumption are risk factors for lymph node metastasis in patients with esophageal cancer.LMR and SIRI are risk factors for vascular invasion in patients with esophageal cancer.Age is a risk factor for perineural invasion in patients with esophageal cancer.2.In patients with esophageal cancer,being male,having higher CEA,and drinking alcohol are risk factors for lymph node metastases,whereas lower LMR and higher SIRI are risk factors for vascular invasion.Perineural invasion is more likely to occur in esophageal cancer patients who are older.3.LMR,AFR,CEA,CA72-4,tumor location,vascular invasion,degree of differentiation and depth of invasion are associated with poor prognosis of patients with esophageal cancer.SIRI,AFR,CA72-4,and CEA are among them and are independent risk variables for how well esophageal cancer patients may fare.Combining clinicopathological characteristics with the comprehensive model created by SIRI,AFR,CA72-4,and CEA can considerably increase the ability to predict the prognosis of ESCC.4.The Nomogram is a practical tool for assessing the prognosis of ESCC since it exhibits good prediction consistency and reliability. |
PDF Full Text Request