A deep learning framework to iDentify prOgnostically releVant cancEr Regions (DOVER) within whole slide histopathology images

Xiangxue Wang; Yufei Zhou; Cristian Barrera; Yijiang Chen; Bolin Song; Cheng Lu; Kailin Yang; Shlomo Koyfman; James Lewis; Kurt Scahlper; Anant Madabhush

doi:10.1016/j.canlet.2025.218134

Back

A deep learning framework to iDentify prOgnostically releVant cancEr Regions (DOVER) within whole slide histopathology images

Journal article

Open access

Peer reviewed

A deep learning framework to iDentify prOgnostically releVant cancEr Regions (DOVER) within whole slide histopathology images

Xiangxue Wang, Yufei Zhou, Cristian Barrera, Yijiang Chen, Bolin Song, Cheng Lu, Kailin Yang, Shlomo Koyfman, James Lewis, Kurt Scahlper, …

Cancer letters, Vol.636, 218134

01/2026

DOI: 10.1016/j.canlet.2025.218134

PMCID: PMC12640559

PMID: 41238099

Files and links (1)

url

https://doi.org/10.1016/j.canlet.2025.218134View

Published (Version of record) Open Access

Abstract

The recent advancements in computational pathology focus on extracting valuable prognostic insights from whole-slide images (WSIs). These methods primarily involve deep learning-based or handcrafted feature representations of the disease’s morphologic patterns associated with outcomes. However, determining the most prognostic regions within tumors remains challenging due to significant morphologic heterogeneity even within manually annotated tumor areas. In other words, the question is not simply what type of representation is appropriate to predict cancer outcomes, but specifically where to mine those representations. To address this issue, a deep learning framework to identify prognostically relevant (PR) cancer regions (DOVER) within WSIs is presented. DOVER leverages patterns mined from the tissue microarray (TMA) spots with the associated long-term clinical outcomes. The prognostic patterns learned from the individual spots of the TMA (morphologically consistent) are then mapped into larger WSIs to locate PR regions for subsequent feature representation and patient outcome prediction. DOVER improves prognostic prediction in terms of c-index over 20% (p<0.05) across 2,041 patients (NSCLC: n=1,141; OPSCC: n=900). Moreover, correlations with quantitative immunofluorescent (QIF) images reveal a diverse CD8+, CD20+, CD4+, and tumor cell distribution in DOVER-selected regions, reflecting a complex interplay between tumor and immune cells. DOVER identifies statistically significant differences between PR regions, both at the molecular and morphological levels. DOVER could help identify specific spatial locations on WSIs that could be used to mine prognostic feature representation for subsequent predictions of clinical outcomes. With additional validation, DOVER could also potentially help to guide AI-informed molecular profiling of tumors. •DOVER objectively identifies prognostically relevant (PR) regions in WSIs.•DOVER maps prognostic patterns from homogeneous TMA spots onto larger WSIs.•Feature representation from high-PR regions improves survival prediction.•Validated DOVER on over 2,000 patients with NSCLC and OPSCC cancer.•Co-registered QIF shows high-PR regions have a mix of immune and cancer cells.

Computational Pathology

Prognostic Prediction

Prognostically Relevance

Details

Title: Subtitle: A deep learning framework to iDentify prOgnostically releVant cancEr Regions (DOVER) within whole slide histopathology images
Creators: Xiangxue Wang - Nanjing University of Information Science and Technology
Yufei Zhou - Case Western Reserve University
Cristian Barrera - The Wallace H. Coulter Department of Biomedical Engineering
Yijiang Chen - Stanford University
Bolin Song - The Wallace H. Coulter Department of Biomedical Engineering
Cheng Lu - Guangdong Provincial People's Hospital
Kailin Yang - Cleveland Clinic
Shlomo Koyfman - Cleveland Clinic
James Lewis - Mayo Clinic Hospital
Kurt Scahlper - Yale University
Anant Madabhush - Georgia Institute of Technology
Resource Type: Journal article
Publication Details: Cancer letters, Vol.636, 218134
DOI: 10.1016/j.canlet.2025.218134
PMID: 41238099
PMCID: PMC12640559
NLM abbreviation: Cancer Lett
ISSN: 0304-3835
eISSN: 1872-7980
Publisher: Elsevier B.V
Grant note: National Natural Science Foundation of China: 62301265 National Cancer Institute: R01CA249992-01A1, R01CA202752-01A1, R01CA208236-01A1, R01CA216579-01A1, R01CA220581-01A1, R01CA257612-01A1, U01CA239055-01, U01CA248226-01, U54CA254566-01 National Heart, Lung and Blood Institute: R01HL151277-01A1, R01HL158071-01A1 National Institute of Biomedical Imaging and Bioengineering: R43EB028736-01 National Center for Research Resources: C06 RR12463-01 United States Department of Veterans Affairs Biomedical Laboratory Research and Development Service: IBX004121A Office of the Assistant Secretary of Defense for Health Affairs, through the Breast Cancer Research Program: W81XWH-19-1-0668 Prostate Cancer Research Program: W81XWH-15-1-0558, W81XWH-20-1-0851 Lung Cancer Research Program: W81XWH-18-1-0440, W81XWH-20-1-0595 Peer Reviewed Cancer Research Program: W81XWH-18-1-0404, W81XWH-21-1-0345 RSNA Research Fellow GrantKidney Precision Medicine Project (KPMP) Glue GrantASTRO-LUNGevity Foundation Radiation Oncology Seed GrantDoD Prostate Cancer Research Program Idea Development Award: W81XWH-18-1-0524 Cancer Tissue and Pathology shared resource of Winship Cancer Institute of Emory UniversityNIH/NCI: UL1TR002548 Clinical and Translational Science Collaborative (CTSC) Cleveland Annual Pilot Award 2020Bristol Myers-SquibbBoehringer-IngelheimEliLillyAstrazenecaDoD Peer Reviewed Cancer Research Program: W81XWH-22-1-0236 Winship Invest Pilot GrantWinship Cancer InstituteMike Slive Foundation for Prostate Cancer Research GrantNCI: U01CA113913 NIH AIM-AHEAD Consortium Development Program
This work is made possible by the National Natural Science Foundation of China under award numbers 62301265, National Cancer Institute under award numbers R01CA249992-01A1, R01CA202752-01A1, R01CA208236-01A1, R01CA216579-01A1, R01CA220581-01A1, R01CA257612-01A1, U01CA239055-01, U01CA248226-01, U54CA254566-01, National Heart, Lung and Blood Institute R01HL151277-01A1, R01HL158071-01A1, National Institute of Biomedical Imaging and Bioengineering R43EB028736-01, National Center for Research Resources under award number C06 RR12463-01, VA Merit Review Award IBX004121A from the United States Department of Veterans Affairs Biomedical Laboratory Research and Development Service, the Office of the Assistant Secretary of Defense for Health Affairs, through the Breast Cancer Research Program (W81XWH-19-1-0668), the Prostate Cancer Research Program (W81XWH-15-1-0558, W81XWH-20-1-0851), the Lung Cancer Research Program (W81XWH-18-1-0440, W81XWH-20-1-0595), the Peer Reviewed Cancer Research Program (W81XWH-18-1-0404, W81XWH-21-1-0345), the Kidney Precision Medicine Project (KPMP) Glue Grant, DoD Prostate Cancer Research Program Idea Development Award W81XWH-18-1-0524, Clinical and Translational Science Collaborative (CTSC) Cleveland Annual Pilot Award 2020 UL1TR002548 and sponsored research agreements from Bristol Myers-Squibb, Boehringer-Ingelheim, EliLilly and Astrazeneca, DoD Peer Reviewed Cancer Research Program (W81XWH-22-1-0236), Winship Invest Pilot Grant, American Cancer Society Institutional Research Grant from the Winship Cancer Institute. Mike Slive Foundation for Prostate Cancer Research Grant, NCI U01CA113913, NIH AIM-AHEAD Consortium Development Program. K. Y. was supported by the RSNA Research Fellow Grant and ASTRO-LUNGevity Foundation Radiation Oncology Seed Grant. Research reported in this publication was supported in part by the Cancer Tissue and Pathology shared resource of Winship Cancer Institute of Emory University and NIH/NCI under award number P30CA138292.
Language: English
Electronic publication date: 11/12/2025
Date published: 01/2026
Academic Unit: Radiation Oncology
Record Identifier: 9985027354202771

Metrics

32 Record Views

See more details