198P Deep learning prediction of response to antibody-drug conjugates in patients with metastatic breast cancer using digital pathology

While antibody-drug conjugates (ADCs) have altered the therapeutic landscape of metastatic breast cancer (mBC), robust predictive biomarkers are lacking. Our aim was to employ digital pathology and machine learning (ML) to assess tumor heterogeneity and develop a predictive model for patients with mBC treated with ADCs.