THE VALUE OF ULTRASOUND IN PREDICTION BREAST CANCER MOLECULAR SUBTYPES
Tran Thi Hue1,2, Dang Thi Hanh 2,3, Le Thi Mai Huong2, Nguyen Van Tuan4,
Nguyen Thu Huong1, Nguyen Duy Hung2,5, Nguyen Minh Duc6
1 – Breast center, Vinmec Times City International Hospital. Ha Noi, Viet Nam.
2 – Department of Radiology, Hanoi Medical University. Ha Noi, Viet Nam.
3 – Department of Radiology, Ha Dong General Hospital. Ha Noi, Viet Nam.
4 – Radiology Center, Hanoi Medical University Hospital. Hanoi, Vietnam.
5 – Department of Radiology, Viet Duc Hospital. Ha Noi, Viet Nam.
|
|
6 – Department of Radiology, Pham Ngoc Thach University of Medicine.Ho Chi Minh City, Viet Nam.
Purpose. To evaluate the association between ultrasound imaging features and molecular subtypes of breast cancer.
Materials and methods. A retrospective study was conducted involving 133 patients with a total of 139 invasive breast carcinoma lesions. All lesions underwent immunohistochemical (IHC) analysis to classify molecular subtypes. Breast ultrasound was performed on all patients prior to biopsy or surgical intervention. The ultrasound images were retrospectively reviewed by two radiologists, each with over five years of experience, who were blinded to the IHC results to ensure statistical independence. Univariate and multivariate logistic regression models were utilized to assess the association between ultrasound characteristics and breast cancer molecular subtypes.
Results and disscussion. The study analyzed 133 patients diagnosed with breast cancer, among whom 6 presented with two separate tumors, resulting in a total of 139 lesions. These lesions were classified into four molecular subtypes: Luminal A (LA) accounted for 39.6% (n=55), Luminal B (LB) for 36% (n=50), HER2-enriched for 14.4% (n=20), and triple-negative breast cancer (TNBC) represented the smallest proportion at 10% (n=14). Ultrasound imaging features demonstrated significant associations with certain molecular subtypes. Lesions classified as LA were characterized by posterior acoustic shadowing (odds ratio [OR] = 3.0), the presence of echogenic rim (OR = 6.2), and the absence of lymph node metastasis (OR = 4.8). In contrast, HER2-enriched lesions were significantly associated with lymph node metastasis (OR = 4.8), indistinct tumor margins (OR = 15.6), and the presence of microcalcifications (OR = 4.8). TNBC lesions were predominantly associated with posterior acoustic enhancement (OR = 11.4), the absence of microcalcifications (OR = 7.6), and the absence of an echogenic rim (OR = 7.1). No statistically significant correlation was observed between ultrasound features and the LB subtype.
Conclusion. The characteristics of margin, posterior acoustic features, microcalcifications, boundary and lymph node status help differentiate between the LA, HER2-enriched and TNBC subtypes.
Keywords: Breast cancer, molecular subtypes, ultrasound, immunohistochemistry.
Corresponding author: Nguyen Duy Hung, e-mail: Этот e-mail адрес защищен от спам-ботов, для его просмотра у Вас должен быть включен Javascript
For citation: Tran Thi Hue, Dang Thi Hanh, Le Thi Mai Huong, Nguyen Van Tuan, Nguyen Thu Huong, Nguyen Duy Hung, Nguyen Minh Duc. The value of ultrasound in prediction breast cancer molecular subtypes. REJR 2025; 15(1):95-106. DOI: 10.21569/2222-7415-2025-15-1-95-106.
Received: 13.02.25 Accepted: 24.02.25