Establishment Of A Comprehensive Prediction Model For Pathologic Complete Response After Neoadjuvant Chemotherapy

Objective: To establish a comprehensive predictive model of pathologic complete response of breast cancer after neoadjuvant chemotherapy,to evaluate the pathological response after neoadjuvant chemotherapy and to guide the choice of clinical operation.Methods: The clinic pathological and imaging data of 98 patients with breast cancer treated with neoadjuvant chemotherapy from January 2015 to October2019,patients were retrospectively analyzed and divided into p CR group and np CR group according to postoperative pathology.Chi-square test was used to analyze the indexes of the two groups,then the indexes of P <0.05 and consider the indicators that may be of clinical significance were included in the multivariate Logistic regression analysis.A comprehensive predictive model of pathological remission of breast cancer after neoadjuvant chemotherapy was constructed by multivariate analysis with statistically significant indexes,and the performance of the model was evaluated by ROC curve.Results: Univariate analysis showed that background parenchymal enhancement,longest diameter,pathological classification could predict pathological complete response of breast cancer,while multivariate analysis showed that background enhancement type,pathological classification were independent predictors of pathological complete response after neoadjuvant chemotherapy(P<0.05)but consider clinical factors,still include longest diameter in the model.The area under the curve,specificity and sensitivity of the comprehensive predictive model of pathological response of breast cancer after neoadjuvant chemotherapy were0.793,85.3% and 65.2% respectively.Conclusion: The comprehensive prediction model of pathological complete response of breast cancer after neoadjuvant chemotherapy has a good ability to predict the state of pathological response and can provide clinical guidance for the selection of appropriate surgical methods.