GIÁ TRỊ CỦA ĐẶC ĐIỂM SIÊU ÂM TRONG DỰ ĐOÁN TÌNH TRẠNG HER2 Ở CÁC KHỐI UNG THƯ VÚ NGUYÊN PHÁT

Abstract

Objective: To evaluate the correlation and predictive value of ultrasound features for HER2 status in primary breast cancer lesions. Subjects and Methods: A retrospective study analyzing 145 patients with 152 primary breast cancer lesions, all of which had confirmed HER2 status by immunohistochemistry. Ultrasound features were analyzed using univariate and multivariate logistic regression to identify independent predictive factors. The performance of the final prediction model was assessed using receiver operating characteristic (ROC) curve analysis. Results: Among 152 breast cancer lesions, 39 cases (25.7%) were HER2-positive (+) and 113 cases (74.3%) were HER2-negative (−). Multivariate logistic regression analysis showed that indistinct margins (OR = 10.72; p = 0.049) and microcalcifications (OR = 3.89; p = 0.002) were statistically significant independent predictors of HER2 positivity. The predictive model combining these two features demonstrated good diagnostic performance, with an area under the ROC curve (AUC) of 0.775 (95% CI: 0.69–0.86; p < 0.001). Conclusion: A predictive model combining margin characteristics and the presence of microcalcifications has value in early prediction of HER2 status, providing effective support for clinical decision-making.