乳腺癌
卵巢癌
医学
内科学
肿瘤科
组织学
深度学习
癌症
人工智能
计算机科学
作者
Erik N. Bergstrom,Ammal Abbasi,Marcos Díaz‐Gay,Loïck Galland,Scott M. Lippman,Sylvain Ladoire,Ludmil B. Alexandrov
出处
期刊:Cold Spring Harbor Laboratory - medRxiv
日期:2023-02-26
标识
DOI:10.1101/2023.02.23.23285869
摘要
ABSTRACT Breast and ovarian cancers harboring homologous recombination deficiencies (HRD) can benefit from platinum-based chemotherapies and PARP inhibitors. Standard diagnostic tests for detecting HRD utilize molecular profiling, which is not universally available especially for medically underserved populations. Here, we trained a deep learning approach for predicting genomically derived HRD scores from routinely sampled hematoxylin and eosin (H&E)-stained histopathological slides. For breast cancer, the approach was externally validated on three independent cohorts and allowed predicting patients’ response to platinum treatment. Using transfer learning, we demonstrated the method’s clinical applicability to H&E-images from high-grade ovarian tumors. Importantly, our deep learning approach outperformed existing genomic HRD biomarkers in predicting response to platinum-based therapies across multiple cohorts, providing a complementary approach for detecting HRD in patients across diverse socioeconomic groups. One-Sentence Summary A deep learning approach outperforms molecular tests in predicting platinum response of HRD cancers from histological slides.
科研通智能强力驱动
Strongly Powered by AbleSci AI