Deep learning-based image analysis predicts PD-L1 status from H&E-stained histopathology images in breast cancer

• Published in: Nature communications 2022-11, Vol.13 (1), p.6753-6753, Article 6753
• Main Authors: Shamai, Gil, Livne, Amir, Polónia, António, Sabo, Edmond, Cretu, Alexandra, Bar-Sela, Gil, Kimmel, Ron
• Format: Article
• Language: English
• Subjects: 13/105; 13/51; 631/114/1305; 631/114/1564; 631/67/1347; 692/308/53/2423; 692/308/575; Annotations; B7-H1 Antigen - metabolism; Biomarkers; Biomarkers, Tumor - metabolism; Breast cancer; Breast Neoplasms - genetics; Datasets; Decision support systems; Deep Learning; Female; Hematoxylin; Histopathology; Humanities and Social Sciences; Humans; Image analysis; Image processing; Immunohistochemistry; Immunotherapy; Ligands; Lung Neoplasms - pathology; Medical imaging; multidisciplinary; PD-L1 protein; Quality assurance; Science; Science (multidisciplinary); Staining and Labeling
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-022-34275-9

Programmed death ligand-1 (PD-L1) has been recently adopted for breast cancer as a predictive biomarker for immunotherapies. The cost, time, and variability of PD-L1 quantification by immunohistochemistry (IHC) are a challenge. In contrast, hematoxylin and eosin (H&E) is a robust staining used routinely for cancer diagnosis. Here, we show that PD-L1 expression can be predicted from H&E-stained images by employing state-of-the-art deep learning techniques. With the help of two expert pathologists and a designed annotation software, we construct a dataset to assess the feasibility of PD-L1 prediction from H&E in breast cancer. In a cohort of 3,376 patients, our system predicts the PD-L1 status in a high area under the curve (AUC) of 0.91 – 0.93. Our system is validated on two external datasets, including an independent clinical trial cohort, showing consistent prediction performance. Furthermore, the proposed system predicts which cases are prone to pathologists miss-interpretation, showing it can serve as a decision support and quality assurance system in clinical practice. Programmed death ligand-1 (PD-L1) has been recently adopted for breast cancer as a predictive biomarker for immunotherapies. Here, the authors show that PD-L1 expression can be predicted from H&E-stained images using deep learning.