System Maintenance Notice

Please note that maintenance work is scheduled between 4 AM and 5 AM - CET. During this time, you may experience reduced performance. Thanks for your patience.

Publications

Journal Article Epublish

PubMed: 41188261 DOI: 10.1038/s41598-025-20804-1

Prognostic prediction in soft-tissue sarcomas using deep learning and digital pathology of tumor and margin areas.

Sci Rep 15(1):38534

Authors: Audrey Michot, Van-Linh Le, Jean-Michel Coindre, Valérie Velasco, Malika Soussi, Nouria Mesli, Antoine Italiano, Maud Toulmonde, Axel Le Cesne, Sylvie Bonvalot, Lucile Vanhersecke, Charles Honoré, Carine Ngo, François Le Loarer, Olivier Saut, Amandine Crombé

Affiliations

  • Department of Surgical Oncology, Institut Bergonie, Bordeaux, France.
  • Bordeaux Institute of Oncology, BRIC U1312, INSERM, Institut Bergonié, Université de Bordeaux, 33000, Bordeaux, France.
  • Department of Data sciences, Institut Bergonie, Bordeaux, France.
  • Department of Pathology, Institut Bergonie, Bordeaux, France.
  • Department of Pathology, Gustave Roussy, Villejuif, France.
  • Department of Oncology, Institut Bergonie, Bordeaux, France.
  • Department of Oncology, Gustave Roussy, Villejuif, France.
  • Department of Surgical Oncology, Institut Curie, Paris, France.
  • Department of Surgical Oncology, Gustave Roussy, Villejuif, France.
  • INRIA, CNRS, Bordeaux, France.
  • Bordeaux Institute of Oncology, BRIC U1312, INSERM, Institut Bergonié, Université de Bordeaux, 33000, Bordeaux, France.
  • Department of Radiology, Bordeaux University Hospital, Bordeaux, France.
  • Department of Radiology, Institut Bergonie, Bordeaux, France.

Abstract

The histological FNCLCC grade is the primary prognostic factor in soft-tissue sarcoma (STS) but fails to fully capture high risk patients. This study aimed to develop and validate a deep learning (DL) model to predict metastatic relapse-free survival (MFS) using digital hematoxylin and eosin-stained whole-slide images. A retrospective analysis was conducted on 308 STS patients from two cancer centers, divided into a training cohort (149 patients) and two independent validation cohorts (64 and 95 patients). Supervised multi-instance learning convolutional neural network models were trained on distinct tumor regions-center (C), periphery (P), and margins (R)-to optimize predictive performance. Univariable analysis showed DL models using tumor center (DL-C), periphery (DL-P), and their combination (DL-CP) were consistently associated with MFS across cohorts, while models incorporating margins (DL-R and DL-CPR) demonstrated less reliable associations. Multivariable Cox regression confirmed that high risk scores from DL models were independent predictors of MFS. The DL-CP model outperformed FNCLCC grading in prognostic accuracy, with c-indices ≥ 0.74 in validation cohorts. Adding tumor margin information did not improve predictions.DL models focusing on tumor center and periphery provide superior prognostic value in STS, offering a streamlined, effective approach for digital pathology-based risk stratification.

MeSH

Telepathology, Neoplasms, Pathology, Sarcoma, Deep Learning, Natural Science Disciplines, Learning, Tissues