Deep Learning-Based Pathology Image Analysis Enhances Magee Feature Correlation With Oncotype DX Breast Recurrence Score.

ER+ breast cancer Magee equation Oncotype DX score deep learning-based algorithm digital pathology

Journal

Frontiers in medicine
ISSN: 2296-858X
Titre abrégé: Front Med (Lausanne)
Pays: Switzerland
ID NLM: 101648047

Informations de publication

Date de publication:
2022
Historique:
received: 28 02 2022
accepted: 18 05 2022
entrez: 1 7 2022
pubmed: 2 7 2022
medline: 2 7 2022
Statut: epublish

Résumé

Oncotype DX Recurrence Score (RS) has been widely used to predict chemotherapy benefits in patients with estrogen receptor-positive breast cancer. Studies showed that the features used in Magee equations correlate with RS. We aimed to examine whether deep learning (DL)-based histology image analyses can enhance such correlations. We retrieved 382 cases with RS diagnosed between 2011 and 2015 from the Emory University and the Ohio State University. All patients received surgery. DL models were developed to detect nuclei of tumor cells and tumor-infiltrating lymphocytes (TILs) and segment tumor cell nuclei in hematoxylin and eosin (H&E) stained histopathology whole slide images (WSIs). Based on the DL-based analysis, we derived image features from WSIs, such as tumor cell number, TIL number variance, and nuclear grades. The entire patient cohorts were divided into one training set (125 cases) and two validation sets (82 and 175 cases) based on the data sources and WSI resolutions. The training set was used to train the linear regression models to predict RS. For prediction performance comparison, we used independent variables from Magee features alone or the combination of WSI-derived image and Magee features. The Pearson's correlation coefficients between the actual RS and predicted RS by DL-based analysis were 0.7058 ( Our results suggest that DL-based digital pathological features can enhance Magee feature correlation with RS.

Sections du résumé

Background UNASSIGNED
Oncotype DX Recurrence Score (RS) has been widely used to predict chemotherapy benefits in patients with estrogen receptor-positive breast cancer. Studies showed that the features used in Magee equations correlate with RS. We aimed to examine whether deep learning (DL)-based histology image analyses can enhance such correlations.
Methods UNASSIGNED
We retrieved 382 cases with RS diagnosed between 2011 and 2015 from the Emory University and the Ohio State University. All patients received surgery. DL models were developed to detect nuclei of tumor cells and tumor-infiltrating lymphocytes (TILs) and segment tumor cell nuclei in hematoxylin and eosin (H&E) stained histopathology whole slide images (WSIs). Based on the DL-based analysis, we derived image features from WSIs, such as tumor cell number, TIL number variance, and nuclear grades. The entire patient cohorts were divided into one training set (125 cases) and two validation sets (82 and 175 cases) based on the data sources and WSI resolutions. The training set was used to train the linear regression models to predict RS. For prediction performance comparison, we used independent variables from Magee features alone or the combination of WSI-derived image and Magee features.
Results UNASSIGNED
The Pearson's correlation coefficients between the actual RS and predicted RS by DL-based analysis were 0.7058 (
Conclusion UNASSIGNED
Our results suggest that DL-based digital pathological features can enhance Magee feature correlation with RS.

Identifiants

pubmed: 35775006
doi: 10.3389/fmed.2022.886763
pmc: PMC9239530
doi:

Types de publication

Journal Article

Langues

eng

Pagination

886763

Subventions

Organisme : NCI NIH HHS
ID : U01 CA242936
Pays : United States

Informations de copyright

Copyright © 2022 Li, Wang, Li, Dababneh, Wang, Zhao, Smith, Teodoro, Li, Kong and Li.

Déclaration de conflit d'intérêts

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Références

Semin Cancer Biol. 2018 Oct;52(Pt 2):151-157
pubmed: 29990622
Int J Breast Cancer. 2017;2017:1257078
pubmed: 28168058
Clin Breast Cancer. 2020 Feb;20(1):19-24
pubmed: 31806448
J Clin Oncol. 2016 Apr 1;34(10):1134-50
pubmed: 26858339
Breast Cancer Res Treat. 2017 May;163(1):51-61
pubmed: 28243897
Breast Cancer Res. 2020 May 14;22(1):47
pubmed: 32408905
Breast Cancer Res. 2016 Aug 11;18(1):84
pubmed: 27515302
Hum Pathol. 2017 Jun;64:7-12
pubmed: 28153508
Ann Surg Oncol. 2016 Jul;23(7):2242-8
pubmed: 26965699
J Clin Oncol. 2003 Sep 1;21(17):3357-65
pubmed: 12847142
J Natl Cancer Inst. 2003 Nov 19;95(22):1692-702
pubmed: 14625260
BMC Cancer. 2018 Mar 24;18(1):320
pubmed: 29573739
IEEE Trans Pattern Anal Mach Intell. 2017 Jun;39(6):1137-1149
pubmed: 27295650
Mol Cancer. 2019 Mar 30;18(1):67
pubmed: 30927930
J Pathol. 2020 Apr;250(5):667-684
pubmed: 32129476
J Clin Oncol. 2016 Dec 20;34(36):4390-4397
pubmed: 27998227
Clin Breast Cancer. 2021 Dec;21(6):e738-e747
pubmed: 33727019
Breast. 2013 Oct;22(5):879-86
pubmed: 23643806
Breast Cancer Res Treat. 2019 Jan;173(1):55-64
pubmed: 30302588
Breast Cancer Res. 2020 Jan 28;22(1):12
pubmed: 31992350
Cancer Res Treat. 2019 Jul;51(3):1073-1085
pubmed: 30384581
Oncoimmunology. 2016 Dec 14;6(1):e1257452
pubmed: 28197375
J Cancer Res Clin Oncol. 2018 May;144(5):799-807
pubmed: 29427210
Cancer Invest. 2010 Nov;28(9):978-82
pubmed: 20690804
Cancer Commun (Lond). 2020 Apr;40(4):154-166
pubmed: 32277744
J Clin Oncol. 2020 Apr 20;38(12):1346-1366
pubmed: 31928404
Breast Cancer Res Treat. 2017 Jul;164(1):245-246
pubmed: 28393310
Breast Cancer Res Treat. 2018 Aug;171(1):1-9
pubmed: 29774470
Clin Cancer Res. 2020 May 15;26(10):2411-2421
pubmed: 31992588
Mod Pathol. 2021 Jan;34(1):77-84
pubmed: 32661297
J Clin Oncol. 2006 Aug 10;24(23):3726-34
pubmed: 16720680
Mod Pathol. 2008 Oct;21(10):1255-61
pubmed: 18360352
Annu Rev Biomed Eng. 2017 Jun 21;19:221-248
pubmed: 28301734
Clin Cancer Res. 2013 Aug 1;19(15):4196-205
pubmed: 23757354
Breast Cancer Res Treat. 2020 Oct;183(3):599-606
pubmed: 32715443
Clin Breast Cancer. 2017 Aug;17(5):356-366
pubmed: 28576631
J Clin Oncol. 2011 Nov 10;29(32):4273-8
pubmed: 21990413
J Clin Oncol. 2018 Jul 10;36(20):2105-2122
pubmed: 29846122
Breast. 2019 Aug;46:116-125
pubmed: 31146185
Biomark Med. 2019 Aug;13(12):1005-1012
pubmed: 31234641
J Natl Cancer Inst. 2001 Jul 4;93(13):979-89
pubmed: 11438563
Am J Clin Pathol. 2016 Oct;146(4):496-502
pubmed: 27686176
Cancers (Basel). 2019 Jul 17;11(7):
pubmed: 31319547
Am J Clin Pathol. 2016 Jun;145(6):871-8
pubmed: 27298399
IEEE Trans Med Imaging. 2016 Aug;35(8):1962-71
pubmed: 27164577
Arch Pathol Lab Med. 2017 Apr;141(4):490-498
pubmed: 28353377
N Engl J Med. 2015 Nov 19;373(21):2005-14
pubmed: 26412349
Lancet Digit Health. 2020 Aug;2(8):e407-e416
pubmed: 33328045
Lab Invest. 2020 Oct;100(10):1367-1383
pubmed: 32661341
Breast Cancer. 2020 Sep;27(5):1007-1016
pubmed: 32385567
Lancet Oncol. 2013 Oct;14(11):1067-1076
pubmed: 24035531
Clin Breast Cancer. 2019 Aug;19(4):236-245
pubmed: 30846407
Semin Cancer Biol. 2020 Feb;60:202-213
pubmed: 31377307
Breast Cancer Res Treat. 2018 Sep;171(2):303-313
pubmed: 29948405
Breast Cancer Res Treat. 2014 Nov;148(1):117-23
pubmed: 25262341
Breast Cancer Res Treat. 2017 Feb;161(3):435-441
pubmed: 27928699
Breast Cancer Res Treat. 2012 Jan;131(2):413-24
pubmed: 21369717
Am J Pathol. 2020 Jul;190(7):1491-1504
pubmed: 32277893
Hematol Oncol Stem Cell Ther. 2019 Jun;12(2):89-96
pubmed: 30796885
Pathol Res Pract. 2020 Nov;216(11):153155
pubmed: 32871536
Arch Pathol Lab Med. 2019 Jul;143(7):859-868
pubmed: 30295070
Med Image Anal. 2017 Dec;42:60-88
pubmed: 28778026
J Clin Oncol. 2021 Jan 10;39(2):126-135
pubmed: 33108242
N Engl J Med. 2019 Jun 20;380(25):2395-2405
pubmed: 31157962
Breast Cancer Res. 2020 Nov 4;22(1):121
pubmed: 33148288
Mod Pathol. 2013 May;26(5):658-64
pubmed: 23503643
Asian Pac J Cancer Prev. 2020 Dec 01;21(12):3587-3593
pubmed: 33369456
Am J Clin Pathol. 2017 Aug 01;148(2):167-172
pubmed: 28898988
J Magn Reson Imaging. 2019 Feb;49(2):518-524
pubmed: 30129697
Med Image Anal. 2019 May;54:280-296
pubmed: 30959445
Am J Clin Pathol. 2018 Oct 24;150(6):491-498
pubmed: 30084931
J Clin Oncol. 2008 Feb 10;26(5):721-8
pubmed: 18258979
Cancer Invest. 2010 Nov;28(9):969-77
pubmed: 20873988
Mod Pathol. 2020 Aug;33(8):1563-1570
pubmed: 32203092

Auteurs

Hongxiao Li (H)

Department of Mathematics and Statistics, Georgia State University, Atlanta, GA, United States.
Institute of Biomedical Engineering, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, China.

Jigang Wang (J)

Department of Pathology, The Affiliated Hospital of Qingdao University, Qingdao, China.
Department of Pathology and Laboratory Medicine, Emory University, Atlanta, GA, United States.

Zaibo Li (Z)

Department of Pathology, The Ohio State University, Columbus, OH, United States.

Melad Dababneh (M)

Department of Pathology and Laboratory Medicine, Emory University, Atlanta, GA, United States.

Fusheng Wang (F)

Department of Computer Science, Stony Brook University, Stony Brook, NY, United States.

Peng Zhao (P)

Department of Pathology, The Affiliated Hospital of Qingdao University, Qingdao, China.

Geoffrey H Smith (GH)

Department of Pathology and Laboratory Medicine, Emory University, Atlanta, GA, United States.

George Teodoro (G)

Department of Computer Science, Federal University of Minas Gerais, Belo Horizonte, Brazil.

Meijie Li (M)

Department of Mathematics and Statistics, Georgia State University, Atlanta, GA, United States.

Jun Kong (J)

Department of Mathematics and Statistics, Georgia State University, Atlanta, GA, United States.
Department of Computer Science, Georgia State University, Atlanta, GA, United States.
Department of Computer Science, Emory University, Atlanta, GA, United States.

Xiaoxian Li (X)

Department of Pathology and Laboratory Medicine, Emory University, Atlanta, GA, United States.

Classifications MeSH