Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer

Sebastian Foersch; Christina Glasner; Ann Christin Woerl; Markus Eckstein; Daniel Christoph Wagner; Stefan Schulz; Franziska Kellers; Aurélie Fernandez; Konstantina Tserea; Michael Kloth; Arndt Hartmann; Achim Heintz; Wilko Weichert; Wilfried Roth; Carol Geppert; Jakob Nikolas Kather; Moritz Jesinghaus

doi:10.1038/s41591-022-02134-1

Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer

Sebastian Foersch, Christina Glasner, Ann Christin Woerl, Markus Eckstein, Daniel Christoph Wagner, Stefan Schulz, Franziska Kellers, Aurélie Fernandez, Konstantina Tserea, Michael Kloth, Arndt Hartmann, Achim Heintz, Wilko Weichert, Wilfried Roth, Carol Geppert, Jakob Nikolas Kather, Moritz Jesinghaus

Chair of Pathology

Research output: Contribution to journal › Article › peer-review

98 Scopus citations

Abstract

Although it has long been known that the immune cell composition has a strong prognostic and predictive value in colorectal cancer (CRC), scoring systems such as the immunoscore (IS) or quantification of intraepithelial lymphocytes are only slowly being adopted into clinical routine use and have their limitations. To address this we established and evaluated a multistain deep learning model (MSDLM) utilizing artificial intelligence (AI) to determine the AImmunoscore (AIS) in more than 1,000 patients with CRC. Our model had high prognostic capabilities and outperformed other clinical, molecular and immune cell-based parameters. It could also be used to predict the response to neoadjuvant therapy in patients with rectal cancer. Using an explainable AI approach, we confirmed that the MSDLM’s decisions were based on established cellular patterns of anti-tumor immunity. Hence, the AIS could provide clinicians with a valuable decision-making tool based on the tumor immune microenvironment.

Original language	English
Pages (from-to)	430-439
Number of pages	10
Journal	Nature Medicine
Volume	29
Issue number	2
DOIs	https://doi.org/10.1038/s41591-022-02134-1
State	Published - Feb 2023

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1038/s41591-022-02134-1

Cite this

Foersch, S., Glasner, C., Woerl, A. C., Eckstein, M., Wagner, D. C., Schulz, S., Kellers, F., Fernandez, A., Tserea, K., Kloth, M., Hartmann, A., Heintz, A., Weichert, W., Roth, W., Geppert, C., Kather, J. N., & Jesinghaus, M. (2023). Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer. Nature Medicine, 29(2), 430-439. https://doi.org/10.1038/s41591-022-02134-1

@article{e721a448f81b499c87646e684e9b7f21,

title = "Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer",

abstract = "Although it has long been known that the immune cell composition has a strong prognostic and predictive value in colorectal cancer (CRC), scoring systems such as the immunoscore (IS) or quantification of intraepithelial lymphocytes are only slowly being adopted into clinical routine use and have their limitations. To address this we established and evaluated a multistain deep learning model (MSDLM) utilizing artificial intelligence (AI) to determine the AImmunoscore (AIS) in more than 1,000 patients with CRC. Our model had high prognostic capabilities and outperformed other clinical, molecular and immune cell-based parameters. It could also be used to predict the response to neoadjuvant therapy in patients with rectal cancer. Using an explainable AI approach, we confirmed that the MSDLM{\textquoteright}s decisions were based on established cellular patterns of anti-tumor immunity. Hence, the AIS could provide clinicians with a valuable decision-making tool based on the tumor immune microenvironment.",

author = "Sebastian Foersch and Christina Glasner and Woerl, {Ann Christin} and Markus Eckstein and Wagner, {Daniel Christoph} and Stefan Schulz and Franziska Kellers and Aur{\'e}lie Fernandez and Konstantina Tserea and Michael Kloth and Arndt Hartmann and Achim Heintz and Wilko Weichert and Wilfried Roth and Carol Geppert and Kather, {Jakob Nikolas} and Moritz Jesinghaus",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive licence to Springer Nature America, Inc.",

year = "2023",

month = feb,

doi = "10.1038/s41591-022-02134-1",

language = "English",

volume = "29",

pages = "430--439",

journal = "Nature Medicine",

issn = "1078-8956",

publisher = "Nature Research",

number = "2",

}

Foersch, S, Glasner, C, Woerl, AC, Eckstein, M, Wagner, DC, Schulz, S, Kellers, F, Fernandez, A, Tserea, K, Kloth, M, Hartmann, A, Heintz, A, Weichert, W, Roth, W, Geppert, C, Kather, JN & Jesinghaus, M 2023, 'Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer', Nature Medicine, vol. 29, no. 2, pp. 430-439. https://doi.org/10.1038/s41591-022-02134-1

TY - JOUR

T1 - Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer

AU - Foersch, Sebastian

AU - Glasner, Christina

AU - Woerl, Ann Christin

AU - Eckstein, Markus

AU - Wagner, Daniel Christoph

AU - Schulz, Stefan

AU - Kellers, Franziska

AU - Fernandez, Aurélie

AU - Tserea, Konstantina

AU - Kloth, Michael

AU - Hartmann, Arndt

AU - Heintz, Achim

AU - Weichert, Wilko

AU - Roth, Wilfried

AU - Geppert, Carol

AU - Kather, Jakob Nikolas

AU - Jesinghaus, Moritz

PY - 2023/2

Y1 - 2023/2

N2 - Although it has long been known that the immune cell composition has a strong prognostic and predictive value in colorectal cancer (CRC), scoring systems such as the immunoscore (IS) or quantification of intraepithelial lymphocytes are only slowly being adopted into clinical routine use and have their limitations. To address this we established and evaluated a multistain deep learning model (MSDLM) utilizing artificial intelligence (AI) to determine the AImmunoscore (AIS) in more than 1,000 patients with CRC. Our model had high prognostic capabilities and outperformed other clinical, molecular and immune cell-based parameters. It could also be used to predict the response to neoadjuvant therapy in patients with rectal cancer. Using an explainable AI approach, we confirmed that the MSDLM’s decisions were based on established cellular patterns of anti-tumor immunity. Hence, the AIS could provide clinicians with a valuable decision-making tool based on the tumor immune microenvironment.

AB - Although it has long been known that the immune cell composition has a strong prognostic and predictive value in colorectal cancer (CRC), scoring systems such as the immunoscore (IS) or quantification of intraepithelial lymphocytes are only slowly being adopted into clinical routine use and have their limitations. To address this we established and evaluated a multistain deep learning model (MSDLM) utilizing artificial intelligence (AI) to determine the AImmunoscore (AIS) in more than 1,000 patients with CRC. Our model had high prognostic capabilities and outperformed other clinical, molecular and immune cell-based parameters. It could also be used to predict the response to neoadjuvant therapy in patients with rectal cancer. Using an explainable AI approach, we confirmed that the MSDLM’s decisions were based on established cellular patterns of anti-tumor immunity. Hence, the AIS could provide clinicians with a valuable decision-making tool based on the tumor immune microenvironment.

UR - http://www.scopus.com/inward/record.url?scp=85145910677&partnerID=8YFLogxK

U2 - 10.1038/s41591-022-02134-1

DO - 10.1038/s41591-022-02134-1

M3 - Article

C2 - 36624314

AN - SCOPUS:85145910677

SN - 1078-8956

VL - 29

SP - 430

EP - 439

JO - Nature Medicine

JF - Nature Medicine

IS - 2

ER -

Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this