Cargando…

Predicting EGFR mutational status from pathology images using a real-world dataset

Treatment of non-small cell lung cancer is increasingly biomarker driven with multiple genomic alterations, including those in the epidermal growth factor receptor (EGFR) gene, that benefit from targeted therapies. We developed a set of algorithms to assess EGFR status and morphology using a real-wo...

Descripción completa

Detalles Bibliográficos
Autores principales: Pao, James J., Biggs, Mikayla, Duncan, Daniel, Lin, Douglas I., Davis, Richard, Huang, Richard S. P., Ferguson, Donna, Janovitz, Tyler, Hiemenz, Matthew C., Eddy, Nathanial R., Lehnert, Erik, Cabili, Moran N., Frampton, Garrett M., Hegde, Priti S., Albacker, Lee A.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Nature Publishing Group UK 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10020556/
https://www.ncbi.nlm.nih.gov/pubmed/36927889
http://dx.doi.org/10.1038/s41598-023-31284-6
_version_ 1784908285322723328
author Pao, James J.
Biggs, Mikayla
Duncan, Daniel
Lin, Douglas I.
Davis, Richard
Huang, Richard S. P.
Ferguson, Donna
Janovitz, Tyler
Hiemenz, Matthew C.
Eddy, Nathanial R.
Lehnert, Erik
Cabili, Moran N.
Frampton, Garrett M.
Hegde, Priti S.
Albacker, Lee A.
author_facet Pao, James J.
Biggs, Mikayla
Duncan, Daniel
Lin, Douglas I.
Davis, Richard
Huang, Richard S. P.
Ferguson, Donna
Janovitz, Tyler
Hiemenz, Matthew C.
Eddy, Nathanial R.
Lehnert, Erik
Cabili, Moran N.
Frampton, Garrett M.
Hegde, Priti S.
Albacker, Lee A.
author_sort Pao, James J.
collection PubMed
description Treatment of non-small cell lung cancer is increasingly biomarker driven with multiple genomic alterations, including those in the epidermal growth factor receptor (EGFR) gene, that benefit from targeted therapies. We developed a set of algorithms to assess EGFR status and morphology using a real-world advanced lung adenocarcinoma cohort of 2099 patients with hematoxylin and eosin (H&E) images exhibiting high morphological diversity and low tumor content relative to public datasets. The best performing EGFR algorithm was attention-based and achieved an area under the curve (AUC) of 0.870, a negative predictive value (NPV) of 0.954 and a positive predictive value (PPV) of 0.410 in a validation cohort reflecting the 15% prevalence of EGFR mutations in lung adenocarcinoma. The attention model outperformed a heuristic-based model focused exclusively on tumor regions, and we show that although the attention model also extracts signal primarily from tumor morphology, it extracts additional signal from non-tumor tissue regions. Further analysis of high-attention regions by pathologists showed associations of predicted EGFR negativity with solid growth patterns and higher peritumoral immune presence. This algorithm highlights the potential of deep learning tools to provide instantaneous rule-out screening for biomarker alterations and may help prioritize the use of scarce tissue for biomarker testing.
format Online
Article
Text
id pubmed-10020556
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Nature Publishing Group UK
record_format MEDLINE/PubMed
spelling pubmed-100205562023-03-18 Predicting EGFR mutational status from pathology images using a real-world dataset Pao, James J. Biggs, Mikayla Duncan, Daniel Lin, Douglas I. Davis, Richard Huang, Richard S. P. Ferguson, Donna Janovitz, Tyler Hiemenz, Matthew C. Eddy, Nathanial R. Lehnert, Erik Cabili, Moran N. Frampton, Garrett M. Hegde, Priti S. Albacker, Lee A. Sci Rep Article Treatment of non-small cell lung cancer is increasingly biomarker driven with multiple genomic alterations, including those in the epidermal growth factor receptor (EGFR) gene, that benefit from targeted therapies. We developed a set of algorithms to assess EGFR status and morphology using a real-world advanced lung adenocarcinoma cohort of 2099 patients with hematoxylin and eosin (H&E) images exhibiting high morphological diversity and low tumor content relative to public datasets. The best performing EGFR algorithm was attention-based and achieved an area under the curve (AUC) of 0.870, a negative predictive value (NPV) of 0.954 and a positive predictive value (PPV) of 0.410 in a validation cohort reflecting the 15% prevalence of EGFR mutations in lung adenocarcinoma. The attention model outperformed a heuristic-based model focused exclusively on tumor regions, and we show that although the attention model also extracts signal primarily from tumor morphology, it extracts additional signal from non-tumor tissue regions. Further analysis of high-attention regions by pathologists showed associations of predicted EGFR negativity with solid growth patterns and higher peritumoral immune presence. This algorithm highlights the potential of deep learning tools to provide instantaneous rule-out screening for biomarker alterations and may help prioritize the use of scarce tissue for biomarker testing. Nature Publishing Group UK 2023-03-16 /pmc/articles/PMC10020556/ /pubmed/36927889 http://dx.doi.org/10.1038/s41598-023-31284-6 Text en © The Author(s) 2023 https://creativecommons.org/licenses/by/4.0/Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) .
spellingShingle Article
Pao, James J.
Biggs, Mikayla
Duncan, Daniel
Lin, Douglas I.
Davis, Richard
Huang, Richard S. P.
Ferguson, Donna
Janovitz, Tyler
Hiemenz, Matthew C.
Eddy, Nathanial R.
Lehnert, Erik
Cabili, Moran N.
Frampton, Garrett M.
Hegde, Priti S.
Albacker, Lee A.
Predicting EGFR mutational status from pathology images using a real-world dataset
title Predicting EGFR mutational status from pathology images using a real-world dataset
title_full Predicting EGFR mutational status from pathology images using a real-world dataset
title_fullStr Predicting EGFR mutational status from pathology images using a real-world dataset
title_full_unstemmed Predicting EGFR mutational status from pathology images using a real-world dataset
title_short Predicting EGFR mutational status from pathology images using a real-world dataset
title_sort predicting egfr mutational status from pathology images using a real-world dataset
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10020556/
https://www.ncbi.nlm.nih.gov/pubmed/36927889
http://dx.doi.org/10.1038/s41598-023-31284-6
work_keys_str_mv AT paojamesj predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT biggsmikayla predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT duncandaniel predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT lindouglasi predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT davisrichard predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT huangrichardsp predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT fergusondonna predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT janovitztyler predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT hiemenzmatthewc predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT eddynathanialr predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT lehnerterik predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT cabilimorann predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT framptongarrettm predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT hegdepritis predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset
AT albackerleea predictingegfrmutationalstatusfrompathologyimagesusingarealworlddataset