Analysis of the Identification value of multiparametric magnetic resonance imaging in HER-2 low expression versus HER-2-positive breast cancer

Abstract

Abstract OBJECTIVE: To investigate the differential diagnostic value of magnetic resonance imaging (MRI) between HER-2 low expression and HER-2-positive breast cancer. Methods:We retrospectively analyzed 52 cases of HER-2 low expression breast cancer and 26 cases of HER-2 positive breast cancer treated in our hospital from 2014-01-01 to 2022-04-12. The patients with breast cancer were first examined by MRI and later confirmed by pathological biopsy. The basic clinical case profiles and the characteristics of lesion-related MRI signs were statistically analyzed between the HER-2 low-expressing breast cancer group and the HER-2-positive breast cancer group. RESULTS: The two groups showed statistically significant differences (P<0.05) in the maximum diameter of the mass (P=0.02), internal enhancement features (P=0.048), ADC values (P=0.001), and histological grading (P=0.001). The remaining clinicopathological and magnetic resonance features such as, age, mass distribution, mass morphology, mass margin and TIC curve type were not statistically different (P > 0.05). logistic multivariate regression model showed that: maximum mass diameter, ADC value and histological grade were independent predictors to distinguish between the two types of breast cancer, and mass diameter (≤2 cm) group (OR=0.306, P = 0.027), lower ADC values (OR=331.254, P=0.001), and lower histological grade (OR=5.001, P=0.001) were more likely to be HER-2 low expressing breast cancers. The ROC prediction model incorporating ADC values had good efficacy in discriminating the HER-2 low expression breast cancer group from the HER-2 positive breast cancer group with an area under the curve (AUC) of 0.691. CONCLUSION: There are some differences between the tumor biological characteristics of patients in the HER-2 low expression breast cancer group and the HER-2 positive breast cancer group, and the use of histologic grading, ADC values, maximum diameter of the mass, and other clinical The use of clinical pathological and magnetic resonance features such as histological grading, ADC value, maximum diameter of the mass combined with logistic regression analysis and ROC curve to construct a prediction model can provide some help in the differentiation of the two.