EUVIMEDUpptäck EUVIMED
Open access · CC BY 4.0

Deep learning‐based H&E‐derived risk scores in colorectal cancer: associations with tumour morphology, biology, and predicted drug response

Nic G Reitsam iD, Xiaofeng Jiang, Junhao Liang, Bianca Grosser, Veselin Grozdanov iD, Chiara Ml Loeffler, Marco Gustav iD, Tim Lenz iD, Hannah S Muti, Zunamys I Carrero, Nicholas P West iD, Philip Quirke iD and 12 more

The Journal of Pathology · 2026

DOI: 10.1002/path.70039 ↗

At a glance

Journal article in The Journal of Pathology, 2026. Written by Nic G Reitsam and 23 co-authors. Institution: Pathology, Faculty of Medicine, University of Augsburg, Augsburg, Germany. Major MEDLINE topics: Colorectal Neoplasms, Deep Learning, Antineoplastic Agents, Image Interpretation, Computer-Assisted. Cited 4 times so far (Crossref); 54 references. The full text is freely available (open access, CC BY 4.0). Funded by: German Federal Ministry of Health, Max-Eder-Programme of the German Cancer Aid, German Federal Ministry of Education and Research.

AbstractSubjectsBibliografiska uppgifterCiteRelated studies

Abstract (original language)

Over recent years, several deep learning (DL) models have been presented to predict colorectal cancer (CRC) patient survival directly from haematoxylin and eosin (H&E)‐stained routine whole‐slide images (WSIs). Unlike traditional studies that rely on manually defined histopathological features, weakly supervised DL allows training directly on clinical endpoints without prior specification of the model's focus. This offers a unique opportunity to study the tissue morphology underlying these predictions, improving our understanding of disease biology. Here, we present a comprehensive analysis of the clinicopathological features, tumour morphology and biology, as well as gene expression‐based predicted drug response of over 4,000 CRC patients derived from four different international cohorts with available H&E‐inferred DL‐based risk scores (low‐ versus high‐risk as well as absolute risk scores). The results from our study suggest that conventional clinicopathological risk factors, such as grade of differentiation, presence of lymph node metastasis, tumour budding, and percentage of tumour necrosis, are positively associated with DL‐based risk scores. Moreover, CRCs with direct tumour–adipocyte interactions are enriched in the DL‐based high‐risk group. Through detailed morphologic review, we provide comprehensive evidence that direct tumour–adipocyte interaction, a high degree of tumour budding, and poorly differentiated morphology are linked to high DL‐based risk scores. Transcriptomic and genetic subgroups show only limited association with H&E‐derived DL‐based risk scores. Moreover, we present data suggesting that DL‐based low‐ versus high‐risk CRCs may be characterised by differential drug sensitivity. Our study highlights that DL‐based risk scores derived from H&E WSIs not only align with established clinicopathological features but also highlight morphological features, such as tumour–adipocyte interaction, that are not routinely captured by established clinicopathological scoring systems. Moreover, DL‐based risk groups may be associated with a differential treatment response, underlining their potential to guide patient stratification in routine clinical practice. © 2026 The Author(s). The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Subjects

MeSH terms (MEDLINE)

Colorectal NeoplasmsDeep LearningAntineoplastic AgentsImage Interpretation, Computer-AssistedHumansRisk AssessmentRisk FactorsBiomarkers, TumorPredictive Value of TestsPredictive Learning ModelsGene Expression Profiling

Author keywords

biomarkercolorectal cancercomputational pathologydeep learningdrug responsegene expressionhistopathologypathologywhole‐slide image

Substances

Antineoplastic AgentsBiomarkers, Tumor

Authors and institutions

  • Pathology, Faculty of Medicine, University of Augsburg, Augsburg, Germany.
  • Bavarian Center for Cancer Research, BZKF, Augsburg, Germany.
  • Else Kroener Fresenius Center for Digital Health, Faculty of Medicine, TUD Dresden University of Technology, Dresden, Germany.
  • Department of Thoracic Surgery, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, University of Electronic Science and Technology of PR China (UESTC), Chengdu, PR C…
  • State Key Laboratory of Precision Measurement Technology and Instruments, Department of Precision Instrument, Tsinghua University, Beijing, PR China.
  • Institute of Pathology, Technical University of Munich, Munich, Germany.

ORCID profiles: Nic G Reitsam, Veselin Grozdanov, Marco Gustav, Tim Lenz, Nicholas P West, Philip Quirke, Heike I Grabsch

Funding

  • German Federal Ministry of Health
  • Max-Eder-Programme of the German Cancer Aid
  • German Federal Ministry of Education and Research
  • German Academic Exchange Service
  • German Federal Joint Committee
  • European Union's Horizon Europe and Innovation Programme
  • National Institute for Health and Care Research
  • Leeds Biomedical Research Centre

Cite

APA

Reitsam, N. G., Jiang, X., Liang, J., Grosser, B., Grozdanov, V., Loeffler, C. M., Gustav, M., Lenz, T., Muti, H. S., Carrero, Z. I., West, N. P., Quirke, P., Foersch, S., Jesinghaus, M., Müller, W., Yuan, T., Hoffmeister, M., Brenner, H., Jonnagaddala, J., …, & Kather, J. N. (2026). Deep learning‐based H&E‐derived risk scores in colorectal cancer: associations with tumour morphology, biology, and predicted drug response. The Journal of Pathology. https://doi.org/10.1002/path.70039

Vancouver

Reitsam NG, Jiang X, Liang J, Grosser B, Grozdanov V, Loeffler CM, et al. Deep learning‐based H&E‐derived risk scores in colorectal cancer: associations with tumour morphology, biology, and predicted drug response. The Journal of Pathology. 2026. doi:10.1002/path.70039

BibTeX
@article{Reitsam2026,
  author = {Reitsam, Nic G and Jiang, Xiaofeng and Liang, Junhao and Grosser, Bianca and Grozdanov, Veselin and Loeffler, Chiara Ml and Gustav, Marco and Lenz, Tim and Muti, Hannah S and Carrero, Zunamys I and West, Nicholas P and Quirke, Philip and Foersch, Sebastian and Jesinghaus, Moritz and Müller, Wolfram and Yuan, Tanwei and Hoffmeister, Michael and Brenner, Hermann and Jonnagaddala, Jitendra and Hawkins, Nicholas J and Ward, Robyn L and Grabsch, Heike I and Märkl, Bruno and Kather, Jakob N},
  title = {Deep learning‐based H&E‐derived risk scores in colorectal cancer: associations with tumour morphology, biology, and predicted drug response},
  journal = {The Journal of Pathology},
  year = {2026},
  doi = {10.1002/path.70039}
}
Start for free →