Combining ultrasound and immunohistochemistry to differentiate HER-2-low from HER-2-null breast cancer: A diagnostic value analysis

This study assesses conventional ultrasound integrated with immunohistochemistry (IHC) for differentiating HER-2-low from HER-2-null breast carcinomas. The cohort comprised 185 surgically treated patients with histologically confirmed invasive breast cancer. All patients underwent surgical treatment. All breast cancer cases were analyzed by ultrasonography preoperatively, including size, number of lesions, aspect ratio, echotexture, cystic change, margin, shape, hyperechoic halo, posterior acoustic, calcification, and blood vessel abundance. Five immunohistochemical factors were assessed: ER, PR, Ki-67, EGFR, and AR. After univariate analysis, the multivariate logistics regression included factors with statistical significance. Diagnostic accuracy of three strategies—IHC alone, ultrasonic features alone, and their combination—was evaluated through sensitivity, specificity, and AUC metrics. The optimal model was rendered as a nomogram. Comparative metrics including AUC, sensitivity, and specificity were calculated for three models, immunohistochemistry, ultrasound features, and the combination of ultrasound features and immunohistochemistry, were 0.654 (95%CI: 0.577–0.741), 61.4 %, 64.8 %; 0.689 (95%CI: 0.609–0.768), 69.3 %, 62.0 %; and 0.720 (95%CI: 0.644–0.796), 71.9 %, 62.0 %, respectively. The final combined model identified four independent predictors: PR status and Ki-67 from IHC, along with echotexture, and shape from ultrasound. Although ultrasound or immunohistochemistry alone has moderate predictive value for HER-2-low breast cancer, their combination significantly enhances diagnostic efficacy with synergistic benefits.