Expression Of VWCE,DPT,SCUBE3 And Prediction Of Clinical Prognosis In Breast Cancer

Background: Breast cancer is a disease caused by a variety of pathogenic factors in vivo and in vitro.At present,ER,PR and HER-2 are not only important prognostic factors for breast cancer,but also targets for commonly used endocrine therapy and targeted therapies.However,not all patients can benefit from it,and patients may develop resistance.Therefore,in addition to developing more new drugs,looking for drug resistance mechanisms and treatment methods,it is also very important to screen new candidate targets that can have a prognostic effect on the prognosis of breast cancer,and at the same time,synthesize new targets.Genes establish a predictive model for breast cancer prognosis and provide a reference for clinical treatment.Objective: 1.Through the differential screening of breast cancer tissue second-generation high-throughput gene sequencing and screening of a series of genes associated with invasion and metastasis of breast cancer,a clinical sample was validated,combined with common clinical and pathological indicators,and multivariate analysis to identify breast cancer Factors related to the prognosis of cancer,to obtain new targets that can predict the prognosis of breast cancer.2.Establish and validate breast cancer prognosis prediction models to guide clinical treatment.Methods: 1.Immunohistochemical method was used to detect the expression of VWCE,DPT,SCUBE3 genes and common clinical prognostic genes(ER,PR,HER-2)protein expression in 140 breast cancer patients with an average of more than 5 years of follow-up.2.Using the Kruskal Wallis rank sum test and Fisher's exact calculation probability grouped population differences,multiple COX proportional hazards models were used to perform multiple regression analysis of prognostic factors to screen for factors that independently influenced prognosis.Survival curves were plotted using the Kaplan-Meier method.3.A combination of independent predictors was used to establish a predictive model of prognosis and prognosis of breast cancer.Conventional regression models and new machine learning and prediction models were used to compare the sensitivity and the area under the special OC curve to assess the pros and cons.Results: 1.The up-regulation of VWCE,DPT,and SCUBE3 in breast cancer can increase the risk of death,increase the expression of VWCE,and increase the risk of breast cancer death by 21% in patients with lower VWCE expression.Patients with elevated DPT expression had a 35% higher risk of death from breast cancer than patients with low DPT expression.The increase in SCUBE3 expression increased the risk of death by 139%.In patients with elevated VWCE expression,the risk of breast cancer recurrence was lower than that of VWCE patients by 32%.Three genes were risk factors for the prognosis of breast cancer.2.Age,ER,PR,HER2,the number of lymph node positive,clinical stage,VWCE,DPT,and SCUBE3 were used as traditional prognostic models and machine learning prediction models.Among the predictive factors with the best model prediction effect were age,ER,PR,HER2,lymph node positive,clinical stage,SCUBE3,model scoring index:-7.45061+0.10702×AGE-0.48150×(TNM=2)-0.05943×(TNM=3)-1.57532×(ER=1)-0.64376×(PR=1)-0.05522×(HER2=1)+1.07075×(SCUBE3=1)+0.29063×Number of regional lymph nodes.AUC is 0.87.The machine prediction and learning model prediction model AUC was 0.92.3.The machine prediction and learning model is more effective in predicting the risk of death from breast cancer.Conclusion: VWCE,DPT,and SCUBE3 are independent prognostic factors for breast cancer.The breast cancer prognosis and machine learning and prediction model established with clinical stage,lymph node positive number,ER,and PR can be used to determine the risk of death of patients and have a higher prognostic value.Provide guidance for individualized treatment and provide potential targets for clinical practice.