Abstract
BACKGROUND: Mammographic density (MD) is a well-established independent risk factor for breast cancer. However, mammography (MG) exhibits limited sensitivity for detecting cancer in dense breasts. The ultrasound-based glandular tissue component (GTC) classification represents an emerging qualitative assessment approach, yet its predictive value for breast cancer risk in Chinese women remains to be further explored. Automated breast ultrasound (ABUS) provides a reproducible method for acquiring standardized volumetric data to support GTC assessment. METHODS: This retrospective case-control study included 414 women with heterogeneously or extremely dense breasts (203 breast cancer cases and 211 benign controls). Data on demographics, clinical indicators, and the BCSC 5-year risk score were collected. Two physicians independently performed GTC classification on the ABUS images, blinded to the group assignment and each other’s assessments. Univariate and multivariate logistic regression analyses were used to identify risk factors in the overall population and among postmenopausal women. Inter-observer agreement was assessed using the weighted kappa and the intraclass correlation coefficient (ICC). RESULTS: Compared to the benign group, the malignant group had significantly higher values for age, age at menarche, proportion of postmenopausal women, prevalence of positive family history, lesion size, BCSC 5-year risk, and GTC classification (all P < 0.05). Multivariate analysis showed that after adjusting for confounders, GTC classification was an independent risk factor for breast cancer (C: OR = 2.62; D: OR = 3.21, P < 0.001) and was positively associated with breast cancer risk (P < 0.001). This association was more pronounced in postmenopausal women (C: OR = 4.17; D: OR = 7.38). Inter-observer agreement for GTC classification was high (weighted k = 0.810, ICC = 0.888). CONCLUSIONS: Among women with dense breasts, ABUS-based GTC classification is a significant, reproducible, and independent risk factor for breast cancer. The findings of this study provide a theoretical basis for future research to integrate GTC into existing risk models to optimize risk stratification.