Abstract PO5-25-10: Clinical and Epidemiologic Significance of HER2-Low Expression in Breast Cancer in the Pathways Study

Abstract Background: A subset of breast cancers (BC) typically classified as HER2-negative do express low levels of the protein and can be responsive to targeted therapy trastuzumab-deruxtecan. It is unclear if HER2-low tumors have a biology distinct from that in HER2 negative BC. The goal of this study was to evaluate the clinical and epidemiologic features of tumors with HER2-low expression compared with those classified as HER2-negative in a large population study. Material and methods: The Pathways Study is a prospective multi-ethnic cohort study of women with BC enrolled between 2006-2013 within Kaiser Permanente Northern California (KPNC). Breast biomarker results were extracted from pathology reports and underwent centralized pathology review. HER2-low was defined as 1+ or 2+ (negative by in situ hybridization); HER2-negative was defined as 0+. Other data were collected by self-report or extraction from electronic health records at the KPNC cancer registry, including BC risk factors, tumor characteristics AJCC stage, ER/PR positive or negative status, treatment modality (chemotherapy, radiation therapy, hormonal therapy, and type of surgery), and survival outcomes recurrence free survival (RFS), breast cancer specific mortality (BCSM), and overall survival (OS). The clinical and epidemiologic variables were tested in association with HER2 status (low/negative) by t-tests. Associations of HER2-low with survival outcomes were calculated by proportional hazards regression. Women were included in this analysis if they had a documented HER2 value that was not classified as HER2-positive. Results: Of the 2,200 eligible cases, 1,295 (57.2%) had tumors that were HER2-low. Hormone receptors (HR) were positive in 1,956 (88.9%) cases. HER2-low expression was observed more often in HR-positive cases than in HR-negative, 1,144 (58.4%) vs. 115 (47.1%), respectively (p=.0005). Patients with HER2-low tumors were less likely than those with HER2-negative BC to have family history: 232 of 1259 (18.6%) vs. 214 of 941 (22.8%), respectively (p=.015). This difference was true for patients with HR-positive tumors but not for HR-negative. HER2-low expression appeared to be more common in Asian patients in the HR-negative group than in patients from any other race or ethnicity group. In the HR-negative group, patients with HER2-low had better OS, BCSM, and RFS (p=.019, .014, and .034, respectively) than those with HER2-negative disease. Multivariable analyses were performed for the HR-negative group and adjusted for age at diagnosis, tumor AJCC stage, and type of treatment. Compared with patients with HER2-negative status, patients with HER2-low tumors had lower risk for OS (hazard ratio=0.48, 95% confidence interval: 0.28 to 0.81, p= .006); for RFS (hazard ratio=0.49, 95% confidence interval: 0.27 to 0.90, p= .021); and for BCSM (hazard ratio=0.36, 95% confidence interval: 0.17 to 0.76, p= .007). In analyses stratified by race and ethnicity and controlled for age at diagnosis, AJCC stage and treatment type, Black patients with HER2-low tumors had lower risk for RFS (hazard ratio=0.44, 95% confidence interval: 0.21 to 0.94, p= .035) compared with patients with HER2-negative tumors. Conclusions: In this large prospective study with well annotated epidemiologic, clinical, and outcome data, we observed some clinical and epidemiologic differences between HER2-low and HER2-negative patients, raising the possibility that HER2-low might be a unique biologic entity. Citation Format: Thaer Khoury, Rochelle Payne Ondracek, Song Yao, Lucas Mendocino, Warren Davis, Angela Omilian, Marilyn Kwan, Janise Roh, Lia D'Addario, Emily Valice, Daniel Fernandez, Isaac Ergas, Christine Ambrosone, Lawrence Kushi. Clinical and Epidemiologic Significance of HER2-Low Expression in Breast Cancer in the Pathways Study abstract. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO5-25-10.