Cargando…

Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study

Deep learning (DL) can predict microsatellite instability (MSI) from routine histopathology slides of colorectal cancer (CRC). However, it is unclear whether DL can also predict other biomarkers with high performance and whether DL predictions generalize to external patient populations. Here, we acq...

Descripción completa

Detalles Bibliográficos
Autores principales: Niehues, Jan Moritz, Quirke, Philip, West, Nicholas P., Grabsch, Heike I., van Treeck, Marko, Schirris, Yoni, Veldhuizen, Gregory P., Hutchins, Gordon G.A., Richman, Susan D., Foersch, Sebastian, Brinker, Titus J., Fukuoka, Junya, Bychkov, Andrey, Uegami, Wataru, Truhn, Daniel, Brenner, Hermann, Brobeil, Alexander, Hoffmeister, Michael, Kather, Jakob Nikolas
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10140458/
https://www.ncbi.nlm.nih.gov/pubmed/36958327
http://dx.doi.org/10.1016/j.xcrm.2023.100980
_version_ 1785033165703741440
author Niehues, Jan Moritz
Quirke, Philip
West, Nicholas P.
Grabsch, Heike I.
van Treeck, Marko
Schirris, Yoni
Veldhuizen, Gregory P.
Hutchins, Gordon G.A.
Richman, Susan D.
Foersch, Sebastian
Brinker, Titus J.
Fukuoka, Junya
Bychkov, Andrey
Uegami, Wataru
Truhn, Daniel
Brenner, Hermann
Brobeil, Alexander
Hoffmeister, Michael
Kather, Jakob Nikolas
author_facet Niehues, Jan Moritz
Quirke, Philip
West, Nicholas P.
Grabsch, Heike I.
van Treeck, Marko
Schirris, Yoni
Veldhuizen, Gregory P.
Hutchins, Gordon G.A.
Richman, Susan D.
Foersch, Sebastian
Brinker, Titus J.
Fukuoka, Junya
Bychkov, Andrey
Uegami, Wataru
Truhn, Daniel
Brenner, Hermann
Brobeil, Alexander
Hoffmeister, Michael
Kather, Jakob Nikolas
author_sort Niehues, Jan Moritz
collection PubMed
description Deep learning (DL) can predict microsatellite instability (MSI) from routine histopathology slides of colorectal cancer (CRC). However, it is unclear whether DL can also predict other biomarkers with high performance and whether DL predictions generalize to external patient populations. Here, we acquire CRC tissue samples from two large multi-centric studies. We systematically compare six different state-of-the-art DL architectures to predict biomarkers from pathology slides, including MSI and mutations in BRAF, KRAS, NRAS, and PIK3CA. Using a large external validation cohort to provide a realistic evaluation setting, we show that models using self-supervised, attention-based multiple-instance learning consistently outperform previous approaches while offering explainable visualizations of the indicative regions and morphologies. While the prediction of MSI and BRAF mutations reaches a clinical-grade performance, mutation prediction of PIK3CA, KRAS, and NRAS was clinically insufficient.
format Online
Article
Text
id pubmed-10140458
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Elsevier
record_format MEDLINE/PubMed
spelling pubmed-101404582023-04-29 Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study Niehues, Jan Moritz Quirke, Philip West, Nicholas P. Grabsch, Heike I. van Treeck, Marko Schirris, Yoni Veldhuizen, Gregory P. Hutchins, Gordon G.A. Richman, Susan D. Foersch, Sebastian Brinker, Titus J. Fukuoka, Junya Bychkov, Andrey Uegami, Wataru Truhn, Daniel Brenner, Hermann Brobeil, Alexander Hoffmeister, Michael Kather, Jakob Nikolas Cell Rep Med Article Deep learning (DL) can predict microsatellite instability (MSI) from routine histopathology slides of colorectal cancer (CRC). However, it is unclear whether DL can also predict other biomarkers with high performance and whether DL predictions generalize to external patient populations. Here, we acquire CRC tissue samples from two large multi-centric studies. We systematically compare six different state-of-the-art DL architectures to predict biomarkers from pathology slides, including MSI and mutations in BRAF, KRAS, NRAS, and PIK3CA. Using a large external validation cohort to provide a realistic evaluation setting, we show that models using self-supervised, attention-based multiple-instance learning consistently outperform previous approaches while offering explainable visualizations of the indicative regions and morphologies. While the prediction of MSI and BRAF mutations reaches a clinical-grade performance, mutation prediction of PIK3CA, KRAS, and NRAS was clinically insufficient. Elsevier 2023-03-22 /pmc/articles/PMC10140458/ /pubmed/36958327 http://dx.doi.org/10.1016/j.xcrm.2023.100980 Text en © 2023 The Author(s) https://creativecommons.org/licenses/by/4.0/This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Niehues, Jan Moritz
Quirke, Philip
West, Nicholas P.
Grabsch, Heike I.
van Treeck, Marko
Schirris, Yoni
Veldhuizen, Gregory P.
Hutchins, Gordon G.A.
Richman, Susan D.
Foersch, Sebastian
Brinker, Titus J.
Fukuoka, Junya
Bychkov, Andrey
Uegami, Wataru
Truhn, Daniel
Brenner, Hermann
Brobeil, Alexander
Hoffmeister, Michael
Kather, Jakob Nikolas
Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study
title Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study
title_full Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study
title_fullStr Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study
title_full_unstemmed Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study
title_short Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study
title_sort generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: a retrospective multi-centric study
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10140458/
https://www.ncbi.nlm.nih.gov/pubmed/36958327
http://dx.doi.org/10.1016/j.xcrm.2023.100980
work_keys_str_mv AT niehuesjanmoritz generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT quirkephilip generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT westnicholasp generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT grabschheikei generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT vantreeckmarko generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT schirrisyoni generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT veldhuizengregoryp generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT hutchinsgordonga generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT richmansusand generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT foerschsebastian generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT brinkertitusj generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT fukuokajunya generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT bychkovandrey generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT uegamiwataru generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT truhndaniel generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT brennerhermann generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT brobeilalexander generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT hoffmeistermichael generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy
AT katherjakobnikolas generalizablebiomarkerpredictionfromcancerpathologyslideswithselfsuperviseddeeplearningaretrospectivemulticentricstudy