%0 Journal Article
%A Kather, Jakob Nikolas
%A Heij, Lara R
%A Grabsch, Heike I
%A Loeffler, Chiara
%A Echle, Amelie
%A Muti, Hannah Sophie
%A Krause, Jeremias
%A Niehues, Jan M
%A Sommer, Kai A J
%A Bankhead, Peter
%A Kooreman, Loes F S
%A Schulte, Jefree J
%A Cipriani, Nicole A
%A Buelow, Roman D
%A Boor, Peter
%A Ortiz-Brüchle, Nadi-Na
%A Hanby, Andrew M
%A Speirs, Valerie
%A Kochanny, Sara
%A Patnaik, Akash
%A Srisuwananukorn, Andrew
%A Brenner, Hermann
%A Hoffmeister, Michael
%A van den Brandt, Piet A
%A Jäger, Dirk
%A Trautwein, Christian
%A Pearson, Alexander T
%A Luedde, Tom
%T Pan-cancer image-based detection of clinically actionable genetic alterations.
%J Nature cancer
%V 1
%N 8
%@ 2662-1347
%C London
%I Nature Research
%M DKFZ-2021-00719
%P 789 - 799
%D 2020
%X Molecular alterations in cancer can cause phenotypic changes in tumor cells and their micro-environment. Routine histopathology tissue slides - which are ubiquitously available - can reflect such morphological changes. Here, we show that deep learning can consistently infer a wide range of genetic mutations, molecular tumor subtypes, gene expression signatures and standard pathology biomarkers directly from routine histology. We developed, optimized, validated and publicly released a one-stop-shop workflow and applied it to tissue slides of more than 5000 patients across multiple solid tumors. Our findings show that a single deep learning algorithm can be trained to predict a wide range of molecular alterations from routine, paraffin-embedded histology slides stained with hematoxylin and eosin. These predictions generalize to other populations and are spatially resolved. Our method can be implemented on mobile hardware, potentially enabling point-of-care diagnostics for personalized cancer treatment. More generally, this approach could elucidate and quantify genotype-phenotype links in cancer.
%F PUB:(DE-HGF)16
%9 Journal Article
%$ pmid:33763651
%2 pmc:PMC7610412
%R 10.1038/s43018-020-0087-6
%U https://inrepo02.dkfz.de/record/168179