Influence Of Chemotherapeutic Drug-induced Liver Injury To Early Breast Cancer

Objective:To explore the early breast cancer during chemotherapy in patients with liver damage effect on survival.Methods:A retrospective analysis on January 1,2014 to December 31,2018 in the second affiliated hospital of Nanchang university of 559 patients with AJCC stageⅠ,Ⅱbreast cancer chemotherapy patients clinical data,patients are women;Age 26 ～ 75,the median age of 50.According to whether occur during chemotherapy in drug-induced liver damage,liver damage can be divided into the group and the liver damage,the Kaplan-Meier method of single factor analysis to evaluate the clinical,pathological features,liver damage,whether chemotherapy in patients with early breast cancer disease-free survival(DFS),the influence of the Kaplan-Meier survival curve drawing,there will be a significant variable analysis of the multivariate COX proportional hazards model.Liver damage to liver damage and the two groups of patients with pathology and clinical factors analysis of baseline,match the variable tendency of the baseline not neat after Kaplan Meier-method of single factor and COX multiple factors analysis in patients with early breast cancer DFS and draw Kaplan Meier survival curve,it is concluded that liver damage were independent prognostic factors for early breast cancer,and in different molecular classification of subgroup analysis of liver damage in which classification effect is larger.According to each variable and DFS logistic binary regression,there will be a statistically significant variables in the combined forecast DFS multiple linear regression equation,and the ROC curve test performance of the regression equation.Results:Single factor analysis results show that in all patients with early breast cancer molecular classification,tumor size,regional lymph node status,liver function,PR,tumor histological grade in breast cancer patients was significantly associated with DFS.Those are selected in the COX multiple factors analysis,tumor size(P = 0.004),regional lymph node status(P = 0.002),liver(P = 0.003),tumor histological grade(P= 0.031)is still independent prognostic factors for DFS influence breast cancer patients.Two groups were divided according to whether liver damage had occurred.After matching the baseline variables,single factor analysis showed that molecular classification,tumor size,regional lymph node status,liver function,PR and breast cancer patients was significantly associated with DFS.Then we selected those factors into the COX proportional hazards regression model,after controlling other variables,tumor size(P = 0.002),regional lymph node status(P < 0.001),liver(P = 0.042)is still affect the independent prognostic factors for breast cancer patients DFS,Kaplan and Meier survival analysis showed that no liver damage is the liver damage group has better disease-free survival(DFS)(P < 0.001).Different molecular classification of subgroup analysis found that liver damage of LUMINAL B breast cancer with DFS effect is remarkable.With LUMINAL B breast cancer patients undergoing chemotherapy,targeted and endocrine therapy.Establish binary logistic regression,liver damage(P = 0.032),tumor size,lymph node status(P = 0.002)(P =0.001)with early breast cancer patients was significantly associated with DFS.These factors have significant impact on DFS are selected into multiple factors binomial regression modeling and the ROC curve.And establish the early breast cancer recurrence risk model,regression equation when the diagnosis is:logit(P)=-1.461+1.760X1+1.141X2+(3.601X31+2.290X32)。X1 is a lymph node staging(N0X1=0,N1 则 X1=1),X2 refers to the presence of liver function,(without liver damage is X1 = 0,with liver damage,the X1 = 1),the X3 refers to tumor size(T1 X31=0,X32=0,T2 则 X31=1,X32=0,T3 则 X31=1,X32=0).Joint diagnosis area under the ROC curve was 0.798(95%CI:0.691-0.905),P<0.001。Conclusions:The occurrence of liver damage during chemotherapy is a poor prognostic factor in early breast cancer patients and can be used as an independent prognostic indicator for predicting disease-free survival of early breast cancer patients,particularly prominent in type LUMINAL B breast cancer.