Weakly-supervised deep learning models enable HER2-low prediction from H &E stained slides

• DOI: 10.1186/s13058-024-01863-0
• Publication Date: 2024-08-19T10:02:57Z
• Authors: Renan Valieris; Luan Martins; Alexandre Defelic...; Adriana Passos Bueno; Cynthia Aparecida...; Dirce Carraro; Emmanuel Dias-Neto; Rafael A. Rosales; Jose Marcio Barro...; Israel Tojal da S...
• Keywords: Surgical oncology

Human epidermal growth factor receptor 2 (HER2)-low breast cancer has emerged as a new subtype of tumor, for which novel antibody-drug conjugates have shown beneficial effects. Assessment HER2 requires several immunohistochemistry tests with an additional in situ hybridization test if case is classified 2+. Therefore, cost-effective methods to speed up the assessment are highly desirable.