Related Experiment Video
Updated: Aug 6, 2025
08:16
Comparative Lesions Analysis Through a Targeted Sequencing Approach
Published on: November 5, 2019
6.8K
Direct prediction of Homologous Recombination Deficiency from routine histology in ten different tumor types with
Chiara Maria Lavinia Loeffler1,2,3, Omar S M El Nahhas2, Hannah Sophie Muti2,4
1Department of Medicine III, University Hospital RWTH Aachen, Aachen, Germany.
Medrxiv : the Preprint Server for Health Sciences
|March 22, 2023
Summary
Deep learning accurately predicts Homologous Recombination Deficiency (HRD) from standard histology images, simplifying biomarker testing for PARP inhibitor therapy across multiple cancer types.
Area of Science:
- Computational pathology
- Oncology
- Biomarker discovery
Background:
- Homologous Recombination Deficiency (HRD) is a key biomarker for predicting response to PARP inhibitors (PARPi).
- Current HRD testing methods are complex and challenging.
- There is a need for simpler, more accessible HRD detection methods.
Conclusions:
- Deep learning models can accurately predict HRD status directly from H&E histology slides.
- This approach is effective both within and across multiple tumor types.
- DL-based HRD prediction offers a simplified alternative to complex genomic testing, potentially expanding patient eligibility for PARPi therapy.
