Correlative Factors And Prediction Model Construction Of Pathological Complete Response To Neoadjuvant Chemotherapy In Breast Cancer

Objective According to the clinical and pathological information of patients,the factors of pathological complete remission of neoadjuvant chemotherapy for breast cancer were studied,and the prediction model of neoadjuvant chemotherapy efficacy was established to provide scientific basis for the selection of clinical treatment decision.Methods A retrospective analysis of 107 breast cancer patients who received neoadjuvant chemotherapy with anthracycline+taxane chemotherapy combined with or without targeted therapy in Zhejiang Provincial People’s Hospital from January 2019 to December 2021 was performed.Univariate analysis was used to study The correlation between different clinical and pathological data and pathological complete remission of breast cancer was analyzed.Significant influencing factors were incorporated into binary logistic regression analysis to construct a predictive model for statistical analysis,and the ROC curve was used to evaluate the performance of this model.In addition,patients were divided into EC-T group and EC-TH(P)group whether targeted therapy was used in neoadjuvant chemotherapy.Multivariate logistic regression analysis and ROC curve were used to construct pCR prediction model and titer verification of different treatment regimens respectively.Results 58(54.2%)of 107 patients achieved pCR after completing NACT.Univariate analysis found:chemotherapy regimen,ER status,HER-2 status,KI-67 level,tumor molecular type,WHO stage,D0 tumor length,D0 tumor size,D2 tumor shrinkage rate,D4 tumor shrinkage The results of tumor shrinkage rate and D6 tumor shrinkage rate were significantly correlated with the pCR results after NACT(P<0.05).Multivariate logistic regression analysis,tumor stage(OR:3.842,P=0.018),HER-2 status(OR:4.552,P=0.012),KI-67 level(OR:1.06,P=0.004),D2 tumor shrinkage Changes in indicators such as tumor shrinkage rate(OR:1.046,P=0.008),D4 tumor shrinkage rate(OR:1.047,P=0.001),and D6 tumor shrinkage rate(OR:1.032,P=0.013)are the predictors of whether NACT can be achieved after NACT.Independent influencing factors of pCR.The area under the curve of the ROC curve analysis model was 0.910,the prediction cutoff value was 0.522,the sensitivity was 0.857,and the specificity was 0.828.After grouping according to different treatment regimens,in the EC-T group,the D2 tumor shrinkage rate(OR:1.049,P=0.035),the D4 tumor shrinkage rate(OR:1.036,P=0.025),the level of KI-67(OR:1.036,P=0.025)1.048,P=0.031)can be used as an independent factor for predicting pCR,the model cutoff value is 0.300,the sensitivity is 0.920,and the specificity is 0.727.In the EC-TH(P)group,D4 tumor shrinkage rate(OR:1.070,P=0.025),D6 tumor shrinkage rate(OR:1.068,P=0.047),KI-67 level(OR:1.107,P=0.009)can be used as an independent factor for predicting pCR,with a cut-off value of 0.747,a sensitivity of 0.833,and a specificity of 0.929.Conclusion Tumor stage,HER-2 status,and KI-67 level can be used as effective predictors of pCR achieved in neoadjuvant chemotherapy for breast cancer.It is possible to obtain pCR.At the same time,the tumor shrinkage rate in the 2nd cycle,the tumor shrinkage rate in the 4th cycle,and the level of KI-67 can predict the pCR of patients with EC-T regimen neoadjuvant chemotherapy,while the tumor shrinkage rate in the 4th cycle,the tumor shrinkage rate in the 6th cycle,the KI-67 level The level of 67 was an independent factor predicting pCR in patients with EC-T combined with targeted therapy.