Abstract:
Objective To investigate the value of digital breast tomosynthesis (DBT) combined with special position radiography for predicting the efficiency of neoadjuvant chemotherapy (NAC) in breast cancer.
Methods A retrospective cohort study was conducted. The clinical and imaging data of 125 female patients with breast masses (with an average age of (43.4±8.2) years) admitted to the Fourth Hospital of Hebei Medical University from June 2024 to March 2025 were collected. All enrolled patients underwent DBT combined with special position radiography, ultrasonography (US), and MRI. The longest diameter of the breast masses was recorded, the least separation distance (LSD) from the mass edge to the adjacent skin was evaluated, and the breast masses were graded using the breast imaging reporting and data system. Intergroup comparisons of measurement data were performed using independent-sample t-test; intergroup comparisons of counting data were performed using paired chi-square test. Pearson's correlation analysis was used to evaluate the correlation between the longest diameter of breast masses measured by different imaging examination methods and histopathological examination results. Kappa consistency test was used to analyze the agreement between the benign and malignant assessment of breast masses by different imaging examination methods and histopathological examination results. Paired chi-square test was used to compare the sensitivity and negative predictive value of different imaging examination methods in predicting malignant breast lesions.
Results The longest diameters of breast masses measured by DBT combined with special position radiography, US, and MRI were moderately correlated with histopathological examination results (r=0.59, 0.50, and 0.51, respectively; all P<0.05). The accuracy of DBT combined with special position radiography in predicting the longest diameter of breast masses was 62.40%, which was higher than those of US (44.80%) and MRI (46.40%), and the differences were statistically significant (χ2=7.79 and 6.45, respectively; both P<0.05). The LSD measured by US was significantly higher (t=5.90, P<0.05) than the histopathological examination results, followed by MRI (t=2.50, P=0.02). The agreement between DBT combined with special position radiography and histopathological examination results in evaluating the benign and malignant nature of breast masses was good, with a Kappa value of 0.76, which was higher than those of US (0.70) and MRI (0.72). The sensitivity and negative predictive value of DBT combined with special position radiography for predicting malignant lesions were 92.22% and 82.50%, respectively, which were significantly higher than 76.67% and 60.38% for US and 80.00% and 64.00% for MRI (χ2=2.06–2.93, all P<0.05).
Conclusion In the comprehensive imaging assessment of patients with breast cancer before NAC, DBT combined with special position radiography showed certain clinical practical advantages in predicting the longest diameter, nature of breast masses and the LSD from the breast mass edge to the adjacent skin, thus showing important clinical application prospects.