%0 Journal Article
%A Echle, Amelie
%A Grabsch, Heike Irmgard
%A Quirke, Philip
%A van den Brandt, Piet A
%A West, Nicholas P
%A Hutchins, Gordon G A
%A Heij, Lara R
%A Tan, Xiuxiang
%A Richman, Susan D
%A Krause, Jeremias
%A Alwers, Elizabeth
%A Jenniskens, Josien
%A Offermans, Kelly
%A Gray, Richard
%A Brenner, Hermann
%A Chang-Claude, Jenny
%A Trautwein, Christian
%A Pearson, Alexander T
%A Boor, Peter
%A Luedde, Tom
%A Gaisa, Nadine Therese
%A Hoffmeister, Michael
%A Nikolas Kather, Jakob
%T Clinical-grade Detection of Microsatellite Instability in Colorectal Tumors by Deep Learning.
%J Gastroenterology
%V 159
%N 4
%@ 0016-5085
%C Philadelphia, Pa. [u.a.]
%I Saunders
%M DKFZ-2020-01144
%P 1406-1416.e11
%D 2020
%Z 2020 Oct;159(4):1406-1416.e11
%X Microsatellite instability (MSI) and mismatch-repair deficiency (dMMR) in colorectal tumors are used to select treatment for patients. Deep learning can detect MSI and dMMR in tumor samples on routine histology slides faster and cheaper than molecular assays. But clinical application of this technology requires high performance and multisite validation, which have not yet been performed.We collected hematoxylin and eosin-stained slides, and findings from molecular analyses for MSI and dMMR, from 8836 colorectal tumors (of all stages) included in the MSIDETECT consortium study, from Germany, the Netherlands, the United Kingdom, and the United States. Specimens with dMMR were identified by immunohistochemistry analyses of tissue microarrays for loss of MLH1, MSH2, MSH6, and/or PMS2. Specimens with MSI were identified by genetic analyses. We trained a deep-learning detector to identify samples with MSI from these slides; performance was assessed by cross-validation (n=6406 specimens) and validated in an external cohort (n=771 specimens). Prespecified endpoints were area under the receiver operating characteristic (AUROC) curve and area under the precision-recall curve (AUPRC).The deep-learning detector identified specimens with dMMR or MSI with a mean AUROC curve of 0.92 (lower bound 0.91, upper bound 0.93) and an AUPRC of 0.63 (range, 0.59-0.65), or 67
%F PUB:(DE-HGF)16
%9 Journal Article
%$ pmid:32562722
%R 10.1053/j.gastro.2020.06.021
%U https://inrepo02.dkfz.de/record/156827