Deep Learning for Detecting BRCA Mutations in High-Grade Ovarian Cancer Based on an Innovative Tumor Segmentation Method From Whole Slide Images

Bourgade, Raphaël; Rabilloud, Noémie; Perennec, Tanguy; Pécot, Thierry; Garrec, Céline; Guédon, A.; Bézieau, Stéphane; Lespagnol, Alexandra; Tayrac, Marie de; Henno, Sébastien; Sagan, Christine; Toquet, Claire; Mosnier, Jean‐François; Kammerer‐Jacquet, Solène‐Florence; Loussouarn, Delphine

doi:10.1016/j.modpat.2023.100304

Cited by 5 publications

(4 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, digital pathology and associated deep learning-based computer vision algorithms emerged as an alternative diagnostic method to predict HRD status and platinum sensitivity. 7 , 8 , 15 , 36 …”

Section: Discussionmentioning

confidence: 99%

Histopathology and proteomics are synergistic for High-Grade Serous Ovarian Cancer platinum response prediction

Kilim,

Olar,

Biricz

et al. 2024

Preprint

View full text Add to dashboard Cite

Patients with High-Grade Serous Ovarian Cancer (HGSOC) exhibit varied responses to treatment, with 20-30% showingde novoresistance to platinum-based chemotherapy. While hematoxylin-eosin (H&E) pathological slides are used for routine diagnosis of cancer type, they may also contain diagnostically useful information about treatment response. Our study demonstrates that combining H&E-stained Whole Slide Images (WSIs) with proteomic signatures using a multimodal deep learning framework significantly improves the prediction of platinum response in both discovery and validation cohorts. This method outperforms the Homologous Recombination Deficiency (HRD) score in predicting platinum response and overall patient survival. The study sets new performance benchmarks and explores the intersection of histology and proteomics, highlighting phenotypes related to treatment response pathways, including homologous recombination, DNA damage response, nucleotide synthesis, apoptosis, and ER stress. This integrative approach has the potential to improve personalized treatment and provide insights into the therapeutic vulnerabilities of HGSOC.

show abstract

Section: Discussionmentioning

confidence: 99%

Histopathology and proteomics are synergistic for High-Grade Serous Ovarian Cancer platinum response prediction

Kilim,

Olar,

Biricz

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…For example, HRDs for TNBC appear to be recapitulated by a high content of TILs and necrosis, while retraction figures correlated with proficient homologous recombination. Of note, DL has also been applied to WSIs to detect BRCA mutations in high-grade ovarian cancer, based on a tumor segmentation method [ 65 ]. The authors suggest that relevant information for the prediction of BRCA mutations lies more in the tumor context, rather than cell morphology, and that the developed DL tool could be used for prescreening [ 65 ].…”

Section: Prognostic and Predictive Models On Digitalized Hande-staine...mentioning

confidence: 99%

“…Of note, DL has also been applied to WSIs to detect BRCA mutations in high-grade ovarian cancer, based on a tumor segmentation method [ 65 ]. The authors suggest that relevant information for the prediction of BRCA mutations lies more in the tumor context, rather than cell morphology, and that the developed DL tool could be used for prescreening [ 65 ]. An emerging study has led to the development of an interpretable BC molecular subtype classification framework, based on DL utilizing multi-omics datasets (moBRCA-net) [ 66 ].…”

Section: Prognostic and Predictive Models On Digitalized Hande-staine...mentioning

confidence: 99%

Early Breast Cancer Risk Assessment: Integrating Histopathology with Artificial Intelligence

Ivanova,

Pescia,

Trapani

et al. 2024

Cancers

View full text Add to dashboard Cite

Effective risk assessment in early breast cancer is essential for informed clinical decision-making, yet consensus on defining risk categories remains challenging. This paper explores evolving approaches in risk stratification, encompassing histopathological, immunohistochemical, and molecular biomarkers alongside cutting-edge artificial intelligence (AI) techniques. Leveraging machine learning, deep learning, and convolutional neural networks, AI is reshaping predictive algorithms for recurrence risk, thereby revolutionizing diagnostic accuracy and treatment planning. Beyond detection, AI applications extend to histological subtyping, grading, lymph node assessment, and molecular feature identification, fostering personalized therapy decisions. With rising cancer rates, it is crucial to implement AI to accelerate breakthroughs in clinical practice, benefiting both patients and healthcare providers. However, it is important to recognize that while AI offers powerful automation and analysis tools, it lacks the nuanced understanding, clinical context, and ethical considerations inherent to human pathologists in patient care. Hence, the successful integration of AI into clinical practice demands collaborative efforts between medical experts and computational pathologists to optimize patient outcomes.

show abstract

“…Regarding breast cancer, a study showed that a DL model based on the ResNet-101 architecture and trained on 461 whole-slide images can be used to predict mutations in TP53, RB1, CDH1, NF1, and NOTCH2, with AUCs of 0.729 to 0.852 [41], while another study managed to predict germline BRCA mutations from whole-slide images with an AUC of 0.766 using a ResNet architecture [42]. One group reported somatic BRCA mutation prediction in high-grade ovarian cancer with an AUC of 0.681 using a ResNet-based architecture trained on images from 867 patients [43]. In an effort to help differentiate noninvasive follicular thyroid neoplasms with papillary-like nuclear features (NIFTP) from other neoplasms, an Xception-based DL model was trained on 115 whole-slide images to predict the BRAF-RAS score.…”

Section: Ai Tools For Predicting Molecular Alterationsmentioning

confidence: 99%

Breaking Barriers: AI’s Influence on Pathology and Oncology in Resource-Scarce Medical Systems

Vigdorovits,

Köteles,

Olteanu

et al. 2023

Cancers

View full text Add to dashboard Cite

The application of artificial intelligence to improve the access of cancer patients to high-quality medical care is one of the goals of modern medicine. Pathology constitutes the foundation of modern oncologic treatment, and its role has expanded far beyond diagnosis into predicting treatment response and overall survival. However, the funding of pathology is often an afterthought in resource-scarce medical systems. The increased digitalization of pathology has paved the way towards the potential use of artificial intelligence tools for improving pathologist efficiency and extracting more information from tissues. In this review, we provide an overview of the main research directions intersecting with artificial intelligence and pathology in relation to oncology, such as tumor classification, the prediction of molecular alterations, and biomarker quantification. We then discuss examples of tools that have matured into clinical products and gained regulatory approval for clinical use. Finally, we highlight the main hurdles that stand in the way of the digitalization of pathology and the application of artificial intelligence in pathology while also discussing possible solutions.

show abstract

Deep Learning for Detecting BRCA Mutations in High-Grade Ovarian Cancer Based on an Innovative Tumor Segmentation Method From Whole Slide Images

Cited by 5 publications

References 41 publications

Histopathology and proteomics are synergistic for High-Grade Serous Ovarian Cancer platinum response prediction

Histopathology and proteomics are synergistic for High-Grade Serous Ovarian Cancer platinum response prediction

Early Breast Cancer Risk Assessment: Integrating Histopathology with Artificial Intelligence

Breaking Barriers: AI’s Influence on Pathology and Oncology in Resource-Scarce Medical Systems

Contact Info

Product

Resources

About