Predicting Prostate Cancer Molecular Subtype With Deep Learning on Histopathologic Images
Modern Pathology · 2023
DOI: 10.1016/j.modpat.2023.100247 ↗Kurz gefasst
Fachartikel in der Zeitschrift Modern Pathology, 2023. Verfasst von Eric Erak und 21 weiteren Autor:innen. Beteiligte Einrichtung: Department of Pathology, Johns Hopkins University School of Medicine. Hauptthemen laut MEDLINE: Deep Learning, Prostatic Neoplasms, PTEN Phosphohydrolase. Die Arbeit wurde bisher 35-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS.
Abstract (Originalsprache)
Microscopic examination of prostate cancer has failed to reveal a reproducible association between molecular and morphologic features. However, deep-learning algorithms trained on hematoxylin and eosin (H&E)-stained whole slide images (WSI) may outperform the human eye and help to screen for clinically-relevant genomic alterations. We created deep-learning algorithms to identify prostate tumors with underlying ETS-related gene (ERG) fusions or PTEN deletions using the following 4 stages: (1) automated tumor identification, (2) feature representation learning, (3) classification, and (4) explainability map generation. A novel transformer-based hierarchical architecture was trained on a single representative WSI of the dominant tumor nodule from a radical prostatectomy (RP) cohort with known ERG/PTEN status (n = 224 and n = 205, respectively). Two distinct vision transformer-based networks were used for feature extraction, and a distinct transformer-based model was used for classification. The ERG algorithm performance was validated across 3 RP cohorts, including 64 WSI from the pretraining cohort (AUC, 0.91) and 248 and 375 WSI from 2 independent RP cohorts (AUC, 0.86 and 0.89, respectively). In addition, we tested the ERG algorithm performance in 2 needle biopsy cohorts comprised of 179 and 148 WSI (AUC, 0.78 and 0.80, respectively). Focusing on cases with homogeneous (clonal) PTEN status, PTEN algorithm performance was assessed using 50 WSI reserved from the pretraining cohort (AUC, 0.81), 201 and 337 WSI from 2 independent RP cohorts (AUC, 0.72 and 0.80, respectively), and 151 WSI from a needle biopsy cohort (AUC, 0.75). For explainability, the PTEN algorithm was also applied to 19 WSI with heterogeneous (subclonal) PTEN loss, where the percentage tumor area with predicted PTEN loss correlated with that based on immunohistochemistry (r = 0.58, P = .0097). These deep-learning algorithms to predict ERG/PTEN status prove that H&E images can be used to screen for underlying genomic alterations in prostate cancer.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Pathology, Johns Hopkins University School of Medicine.
- Department of Pathology, Suleyman Demirel University, Turkey.
- Koç University School of Medicine, Turkey.
- Fundacion Santa Fe de Bogota University Hospital, Columbia.
- Hacettepe University, Turkey.
- Department of Pathology, University of Alabama School of Medicine, Alabama.
Förderung
- NCI NIH HHS
Zitieren
APA
Erak, E., Oliveira, L. D., Mendes, A. A., Dairo, O., Ertunc, O., Kulac, I., Baena-Del Valle, J. A., Jones, T., Hicks, J. L., Glavaris, S., Guner, G., Vidal, I. D., Markowski, M., de la Calle, C., Trock, B. J., Meena, A., Joshi, U., Kondragunta, C., Bonthu, S., …, & Lotan, T. L. (2023). Predicting Prostate Cancer Molecular Subtype With Deep Learning on Histopathologic Images. Modern Pathology. https://doi.org/10.1016/j.modpat.2023.100247
Vancouver
Erak E, Oliveira LD, Mendes AA, Dairo O, Ertunc O, Kulac I, et al. Predicting Prostate Cancer Molecular Subtype With Deep Learning on Histopathologic Images. Modern Pathology. 2023. doi:10.1016/j.modpat.2023.100247
BibTeX
@article{Erak2023,
author = {Erak, Eric and Oliveira, Lia DePaula and Mendes, Adrianna A and Dairo, Oluwademilade and Ertunc, Onur and Kulac, Ibrahim and Baena-Del Valle, Javier A and Jones, Tracy and Hicks, Jessica L and Glavaris, Stephanie and Guner, Gunes and Vidal, Igor Damasceno and Markowski, Mark and de la Calle, Claire and Trock, Bruce J and Meena, Avaneesh and Joshi, Uttara and Kondragunta, Chaith and Bonthu, Saikiran and Singhal, Nitin and De Marzo, Angelo M and Lotan, Tamara L},
title = {Predicting Prostate Cancer Molecular Subtype With Deep Learning on Histopathologic Images},
journal = {Modern Pathology},
year = {2023},
doi = {10.1016/j.modpat.2023.100247}
}
EUVIMED