Deep Learning Predicts Survival Across Squamous Tumor Entities From Routine Pathology: Insights From Head and Neck, Esophagus, Lung, and Cervical Cancer
Modern Pathology · 2025
DOI: 10.1016/j.modpat.2025.100845 ↗Kurz gefasst
Fachartikel in der Zeitschrift Modern Pathology, 2025. Verfasst von Verena Bitto und 4 weiteren Autor:innen. Beteiligte Einrichtung: German Cancer Research Center (DKFZ), Division of Radiooncology/Radiobiology, Heidelberg, Germany. Electronic address: verena.bitto@dkfz-heidelberg.d…. Hauptthemen laut MEDLINE: Deep Learning, Esophageal Neoplasms, Lung Neoplasms, Head and Neck Neoplasms. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Computational pathology-based models are becoming increasingly popular for extracting biomarkers from images of cancer tissue. However, their validity is often only demonstrated on a single unseen validation cohort, limiting insights into their generalizability and posing challenges for explainability. In this study, we developed models to predict overall survival using hematoxylin and eosin slides from formalin-fixed paraffin-embedded samples in head and neck squamous cell carcinoma. By validating our models across diverse squamous tumor entities, including head and neck (hazard ratio [HR], 1.58; 95% CI, 1.17-2.12; P = .003), esophageal (nonsignificant), lung (HR, 1.31; 95% CI, 1.13-1.52; P < .001), and cervical (HR, 1.39; 95% CI, 1.10-1.75; P = .005) squamous cell carcinomas, we showed that the predicted risk score captures relevant information for survival beyond head and neck squamous cell carcinoma. Correlation analysis indicated that the predicted risk score is strongly associated with various clinical factors, including human papillomavirus status, tumor volume, and smoking history, although the specific factors vary across cohorts. These results emphasize the relevance of comprehensive validation and in-depth assessment of computational pathology-based models to better characterize the underlying patterns they learn during training.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- German Cancer Research Center (DKFZ), Division of Radiooncology/Radiobiology, Heidelberg, Germany. Electronic address: verena.bitto@dkfz-heidelberg.de.
- Else Kroener Fresenius Center for Digital Health, Technical University Dresden, Dresden, Germany; Department of Thoracic Surgery, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan…
- German Cancer Research Center (DKFZ), Division of Radiooncology/Radiobiology, Heidelberg, Germany; German Cancer Consortium (DKTK), Core Center Heidelberg, Heidelberg, Germany.
- Else Kroener Fresenius Center for Digital Health, Technical University Dresden, Dresden, Germany; Medical Department I, University Hospital and Faculty of Medicine Carl Gustav Carus, Technische Universität Dresden, Dres…
Zitieren
APA
Bitto, V., Jiang, X., Baumann, M., Kather, J. N., & Kurth, I. (2025). Deep Learning Predicts Survival Across Squamous Tumor Entities From Routine Pathology: Insights From Head and Neck, Esophagus, Lung, and Cervical Cancer. Modern Pathology. https://doi.org/10.1016/j.modpat.2025.100845
Vancouver
Bitto V, Jiang X, Baumann M, Kather JN, Kurth I. Deep Learning Predicts Survival Across Squamous Tumor Entities From Routine Pathology: Insights From Head and Neck, Esophagus, Lung, and Cervical Cancer. Modern Pathology. 2025. doi:10.1016/j.modpat.2025.100845
BibTeX
@article{Bitto2025,
author = {Bitto, Verena and Jiang, Xiaofeng and Baumann, Michael and Kather, Jakob Nikolas and Kurth, Ina},
title = {Deep Learning Predicts Survival Across Squamous Tumor Entities From Routine Pathology: Insights From Head and Neck, Esophagus, Lung, and Cervical Cancer},
journal = {Modern Pathology},
year = {2025},
doi = {10.1016/j.modpat.2025.100845}
}
EUVIMED