MRI影像组学可有效预测乳腺癌新辅助治疗疗效OA

Objective To explore the efficacy of neoadjuvant therapy for breast cancer based on MRI radiomics features,clinical parameters,and pathological data.Methods A retrospective analysis was conducted on 123 breast cancer patients who underwent neoadjuvant therapy at the First Affiliated Hospital of Bengbu Medical University from January 2021 to December 2023.Based on the Miller-Payne(MP)grading system,patients were stratified into the major histological response group(MHR,n=71)and the non-major histological response group(NMHR,n=52).Clinical parameters,pathological data and MRI radiomics features were collected and compared between the two groups.Statistical analyses were performed using R software,and the predictive performance of the model was assessed using ROC curves and area under the curve(AUC).Results Analysis of clinical characteristics demonstrated no statistically significant differences between the MHR and NMHR groups regarding age,tumor long-axis diameter,pre-NAC clinical N stage,pre-NAC clinical T stage,and pre-NAC clinical stage(P>0.05).Following the extraction and dimensionality reduction of MRI radiomics features,a support vector machine(SVM)was employed to develop predictive models distinguishing the two groups.The models achieved an AUC of 0.783 in the training sets and 0.727 in the validation sets for both groups.Conclusion This study indicates that conventional clinical factors,including lesion size and lymph node metastasis,have limited value in predicting the response to neoadjuvant therapy,whereas MRI radiomics effectively predicts the efficacy of neoadjuvant therapy for breast cancer and may guide individualized treatment,with the potential to improve patient prognosis.