Screening Of Influencing Factors Of Adjuvant Therapy For Rectal Cancer And Establishment Of Prediction Model

Background: Rectal cancer is a common malignant tumor of digestive tract in my country,which has the characteristics of high incidence,poor prognosis,and difficulty in early diagnosis.Although there has been great progress in the treatment of rectal cancer,some patients still experience recurrence and metastasis after surgery.Currently,there is no clinical method to predict the recurrence or metastasis of tumor after surgery.Objective: Based on the clinical data of the patients,this study constructed a model that can predict the effect of adjuvant therapy in patients with rectal cancer.The model predicts the sensitivity of patients to adjuvant therapy,and provides a reference for the formulation and adjustment of clinical treatment plans.Methods: According to the inclusion and exclusion criteria,the clinical data of patients in the modeling group and the validation group were collected,and the patients in the modeling group were divided into recurrence and metastasis group and non recurrence and metastasis group.Through univariate analysis,the differences of23 variables such as age and gender between the two groups were compared to screen the variables that may affect the recurrence and metastasis of patients.Then,using multivariate logistic regression analysis,the variables with predictive value are assigned according to the regression coefficient,and the joint predictors are constructed.Finally,the effectiveness of joint predictors is evaluated,and it is internally verified by validation group data.Results: A total of 264 patients in the modeling group and 149 in the validation group were included in this study.The modeling group was divided into a recurrence and metastasis group and a non-recurrence and metastasis group according to whether there was tumor recurrence and metastasis within 5 years after surgery.The results of univariate analysis showed that postoperative recurrence and metastasis were associated with gender,age,BMI,hypertension,diabetes mellitus,colon polyps,neoadjuvant therapy,tumor location,mucous adenocarcinoma in tissue,pulse The presence or absence of tumor thrombus,postoperative complications,preoperative CA199,hemoglobin,albumin,triglyceride,and maximum tumor diameter were not related to the degree of differentiation,the presence or absence of cancer nodules,nerve invasion,cholesterol levels,and preoperative CEA It is related to the difference in the number of dissection and metastatic lymph nodes;the regression coefficients of the above variables were obtained by multivariate Logistic regression analysis,and the calculation formula of the combined predictor L was constructed: L=3.779-3.037*differentiation degree(moderately differentiated)-2.856*differentiation degree(Well differentiated)+0.019*preoperative CEA-0.447*cholesterol-0.087*removed lymph nodes+0.159*metastatic lymph nodes+0.911*nerve invasion+0.494* cancer nodules.The ROC curve analysis was performed and the area under the curve AUC was calculated.After that,the Youden index was calculated to be 0.4533,the critical value was 0.24468,the sensitivity was 0.5510,and the specificity was 0.9023.The prediction results of the validation group were consistent with the modeling group.Conclusion: The prediction model of adjuvant therapy for rectal cancer based on patient clinical data is of great value in predicting postoperative recurrence and metastasis of rectal cancer patients,and has a certain reference significance for the evaluation of the effect of adjuvant therapy.