Journal Article
. 2018 Jun;18(1).
doi: 10.1186/s12885-018-4448-9.

Quantitative nuclear histomorphometry predicts oncotype DX risk categories for early stage ER+ breast cancer

Jon Whitney 1 German Corredor 2 Andrew Janowczyk 3 Shridar Ganesan 4 Scott Doyle 5 John Tomaszewski 5 Michael Feldman 6 Hannah Gilmore 7 Anant Madabhushi 3 
  • PMID: 29848291
  •     39 References
  •     10 citations


Background: Gene-expression companion diagnostic tests, such as the Oncotype DX test, assess the risk of early stage Estrogen receptor (ER) positive (+) breast cancers, and guide clinicians in the decision of whether or not to use chemotherapy. However, these tests are typically expensive, time consuming, and tissue-destructive.

Methods: In this paper, we evaluate the ability of computer-extracted nuclear morphology features from routine hematoxylin and eosin (H&E) stained images of 178 early stage ER+ breast cancer patients to predict corresponding risk categories derived using the Oncotype DX test. A total of 216 features corresponding to the nuclear shape and architecture categories from each of the pathologic images were extracted and four feature selection schemes: Ranksum, Principal Component Analysis with Variable Importance on Projection (PCA-VIP), Maximum-Relevance, Minimum Redundancy Mutual Information Difference (MRMR MID), and Maximum-Relevance, Minimum Redundancy - Mutual Information Quotient (MRMR MIQ), were employed to identify the most discriminating features. These features were employed to train 4 machine learning classifiers: Random Forest, Neural Network, Support Vector Machine, and Linear Discriminant Analysis, via 3-fold cross validation.

Results: The four sets of risk categories, and the top Area Under the receiver operating characteristic Curve (AUC) machine classifier performances were: 1) Low ODx and Low mBR grade vs. High ODx and High mBR grade (Low-Low vs. High-High) (AUC = 0.83), 2) Low ODx vs. High ODx (AUC = 0.72), 3) Low ODx vs. Intermediate and High ODx (AUC = 0.58), and 4) Low and Intermediate ODx vs. High ODx (AUC = 0.65). Trained models were tested independent validation set of 53 cases which comprised of Low and High ODx risk, and demonstrated per-patient accuracies ranging from 75 to 86%.

Conclusion: Our results suggest that computerized image analysis of digitized H&E pathology images of early stage ER+ breast cancer might be able predict the corresponding Oncotype DX risk categories.

Systematic analysis of breast cancer morphology uncovers stromal features associated with survival.
Andrew H Beck, Ankur R Sangoi, +6 authors, Daphne Koller.
Sci Transl Med, 2011 Nov 11; 3(108). PMID: 22072638
Highly Cited.
Analytical validation of the Oncotype DX genomic diagnostic test for recurrence prognosis and therapeutic response prediction in node-negative, estrogen receptor-positive breast cancer.
Maureen Cronin, Chithra Sangli, +7 authors, Drew Watson.
Clin Chem, 2007 Apr 28; 53(6). PMID: 17463177
Highly Cited.
Prediction of the Oncotype DX recurrence score: use of pathology-generated equations derived by linear regression analysis.
Molly E Klein, David J Dabbs, +4 authors, Rohit Bhargava.
Mod Pathol, 2013 Mar 19; 26(5). PMID: 23503643    Free PMC article.
Analysis of the MammaPrint breast cancer assay in a predominantly postmenopausal cohort.
Ben S Wittner, Dennis C Sgroi, +9 authors, Sridhar Ramaswamy.
Clin Cancer Res, 2008 May 17; 14(10). PMID: 18483364    Free PMC article.
Interobserver reproducibility of Gleason grading of prostatic carcinoma: general pathologist.
W C Allsbrook, K A Mangold, +3 authors, J I Epstein.
Hum Pathol, 2001 Feb 15; 32(1). PMID: 11172299
Comparing Breast Cancer Multiparameter Tests in the OPTIMA Prelim Trial: No Test Is More Equal Than the Others.
John M S Bartlett, Jane Bayani, +18 authors, OPTIMA TMG.
J Natl Cancer Inst, 2016 May 01; 108(9). PMID: 27130929    Free PMC article.
Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features.
Kun-Hsing Yu, Ce Zhang, +4 authors, Michael Snyder.
Nat Commun, 2016 Aug 17; 7. PMID: 27527408    Free PMC article.
Highly Cited.
Relationship among outcome, stage of disease, and histologic grade for 22,616 cases of breast cancer. The basis for a prognostic index.
D E Henson, L Ries, L S Freedman, M Carriaga.
Cancer, 1991 Nov 15; 68(10). PMID: 1913453
Quantitative relationships of intravascular tumor cells, tumor vessels, and pulmonary metastases following tumor implantation.
L A Liotta, J Kleinerman, G M Saidel.
Cancer Res, 1974 May 01; 34(5). PMID: 4841969
Highly Cited.
Histopathologic variables predict Oncotype DX recurrence score.
Melina B Flanagan, David J Dabbs, +2 authors, Rohit Bhargava.
Mod Pathol, 2008 Mar 25; 21(10). PMID: 18360352
A deep learning based strategy for identifying and associating mitotic activity with gene expression derived risk categories in estrogen receptor positive breast cancers.
David Romo-Bucheli, Andrew Janowczyk, +2 authors, Anant Madabhushi.
Cytometry A, 2017 Feb 14; 91(6). PMID: 28192639    Free PMC article.
Nuclear Shape and Architecture in Benign Fields Predict Biochemical Recurrence in Prostate Cancer Patients Following Radical Prostatectomy: Preliminary Findings.
George Lee, Robert W Veltri, +3 authors, Anant Madabhushi.
Eur Urol Focus, 2017 Jul 30; 3(4-5). PMID: 28753763    Free PMC article.
Predicting response to primary chemotherapy: gene expression profiling of paraffin-embedded core biopsy tissue.
Lida Mina, Sharon E Soule, +8 authors, Kathy D Miller.
Breast Cancer Res Treat, 2006 Oct 14; 103(2). PMID: 17039265
Abnormal nuclear shape in solid tumors reflects mitotic instability.
D Gisselsson, J Björk, +4 authors, N Mandahl.
Am J Pathol, 2001 Jan 06; 158(1). PMID: 11141493    Free PMC article.
Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up.
C W Elston, I O Ellis.
Histopathology, 1991 Nov 01; 19(5). PMID: 1757079
Highly Cited.
Feature selection based on mutual information: criteria of max-dependency, max-relevance, and min-redundancy.
Hanchuan Peng, Fuhui Long, Chris Ding.
IEEE Trans Pattern Anal Mach Intell, 2005 Aug 27; 27(8). PMID: 16119262
Highly Cited.
Development of the 21-gene assay and its application in clinical practice and clinical trials.
Joseph A Sparano, Soonmyung Paik.
J Clin Oncol, 2008 Feb 09; 26(5). PMID: 18258979
Highly Cited. Review.
A comparison of PAM50 intrinsic subtyping with immunohistochemistry and clinical prognostic factors in tamoxifen-treated estrogen receptor-positive breast cancer.
Torsten O Nielsen, Joel S Parker, +12 authors, Matthew J Ellis.
Clin Cancer Res, 2010 Sep 15; 16(21). PMID: 20837693    Free PMC article.
Highly Cited.
Multi-field-of-view framework for distinguishing tumor grade in ER+ breast cancer from entire histopathology slides.
Ajay Basavanhally, Shridar Ganesan, +4 authors, Anant Madabhushi.
IEEE Trans Biomed Eng, 2013 Feb 09; 60(8). PMID: 23392336    Free PMC article.
Deep learning for digital pathology image analysis: A comprehensive tutorial with selected use cases.
Andrew Janowczyk, Anant Madabhushi.
J Pathol Inform, 2016 Aug 27; 7. PMID: 27563488    Free PMC article.
Highly Cited.
Subvisual chromatin changes in cervical epithelium measured by texture image analysis and correlated with HPV.
Martial Guillaud, Karen Adler-Storthz, +7 authors, Calum Macaulay.
Gynecol Oncol, 2005 Sep 29; 99(3 Suppl 1). PMID: 16188299
Multi-field-of-view strategy for image-based outcome prediction of multi-parametric estrogen receptor-positive breast cancer histopathology: Comparison to Oncotype DX.
Ajay Basavanhally, Michael Feldman, +4 authors, Anant Madabhushi.
J Pathol Inform, 2011 Jan 01; 2. PMID: 22811953    Free PMC article.
Cell orientation entropy (COrE): predicting biochemical recurrence from prostate cancer tissue microarrays.
George Lee, Sahirzeeshan Ali, +3 authors, Anant Madabhushi.
Med Image Comput Comput Assist Interv, 2014 Feb 08; 16(Pt 3). PMID: 24505786
Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer.
Soonmyung Paik, Gong Tang, +11 authors, Norman Wolmark.
J Clin Oncol, 2006 May 25; 24(23). PMID: 16720680
Highly Cited.
A fibrotic focus is a prognostic factor and a surrogate marker for hypoxia and (lymph)angiogenesis in breast cancer: review of the literature and proposal on the criteria of evaluation.
G G Van den Eynden, C G Colpaert, +5 authors, T Hasebe.
Histopathology, 2007 Jun 27; 51(4). PMID: 17593207
A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer.
Soonmyung Paik, Steven Shak, +12 authors, Norman Wolmark.
N Engl J Med, 2004 Dec 14; 351(27). PMID: 15591335
Highly Cited.
Use of modified Magee equations and histologic criteria to predict the Oncotype DX recurrence score.
Bradley M Turner, Kristin A Skinner, +6 authors, David G Hicks.
Mod Pathol, 2015 May 02; 28(7). PMID: 25932962
Stromal fibroblasts in cancer initiation and progression.
Neil A Bhowmick, Eric G Neilson, Harold L Moses.
Nature, 2004 Nov 19; 432(7015). PMID: 15549095    Free PMC article.
Highly Cited. Review.
Histological grading and prognosis in breast cancer; a study of 1409 cases of which 359 have been followed for 15 years.
Br J Cancer, 1957 Sep 01; 11(3). PMID: 13499785    Free PMC article.
Highly Cited.
Increased stromelysin 3 gene expression is associated with increased local invasiveness in head and neck squamous cell carcinomas.
D Muller, C Wolf, +7 authors, G Methlin.
Cancer Res, 1993 Jan 01; 53(1). PMID: 7677979
Automated Tubule Nuclei Quantification and Correlation with Oncotype DX risk categories in ER+ Breast Cancer Whole Slide Images.
David Romo-Bucheli, Andrew Janowczyk, +2 authors, Anant Madabhushi.
Sci Rep, 2016 Sep 08; 6. PMID: 27599752    Free PMC article.
Minimum redundancy feature selection from microarray gene expression data.
Chris Ding, Hanchuan Peng.
J Bioinform Comput Biol, 2005 Apr 27; 3(2). PMID: 15852500
Highly Cited.
Novel PCA-VIP scheme for ranking MRI protocols and identifying computer-extracted MRI measurements associated with central gland and peripheral zone prostate tumors.
Shoshana B Ginsburg, Satish E Viswanath, +4 authors, Anant Madabhushi.
J Magn Reson Imaging, 2014 Jun 20; 41(5). PMID: 24943647
Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials.
Early Breast Cancer Trialists' Collaborative Group (EBCTCG).
Lancet, 2005 May 17; 365(9472). PMID: 15894097
Highly Cited.
Dynamic classification using case-specific training cohorts outperforms static gene expression signatures in breast cancer.
Balázs Győrffy, Thomas Karn, +3 authors, Lajos Pusztai.
Int J Cancer, 2014 Oct 03; 136(9). PMID: 25274406    Free PMC article.
An introduction to recursive partitioning: rationale, application, and characteristics of classification and regression trees, bagging, and random forests.
Carolin Strobl, James Malley, Gerhard Tutz.
Psychol Methods, 2009 Dec 09; 14(4). PMID: 19968396    Free PMC article.
Highly Cited.
A quantitative histomorphometric classifier (QuHbIC) identifies aggressive versus indolent p16-positive oropharyngeal squamous cell carcinoma.
James S Lewis, Sahirzeeshan Ali, +2 authors, Anant Madabhushi.
Am J Surg Pathol, 2013 Oct 23; 38(1). PMID: 24145650    Free PMC article.
Cognitive function in breast cancer patients receiving adjuvant chemotherapy.
C B Brezden, K A Phillips, +2 authors, I F Tannock.
J Clin Oncol, 2000 Jul 15; 18(14). PMID: 10894868
Prospective Validation of a 21-Gene Expression Assay in Breast Cancer.
Joseph A Sparano, Robert J Gray, +28 authors, George W Sledge.
N Engl J Med, 2015 Sep 29; 373(21). PMID: 26412349    Free PMC article.
Highly Cited.
Artificial intelligence in digital pathology - new tools for diagnosis and precision oncology.
Kaustav Bera, Kurt A Schalper, +2 authors, Anant Madabhushi.
Nat Rev Clin Oncol, 2019 Aug 11; 16(11). PMID: 31399699    Free PMC article.
Highly Cited. Review.
Quantitative nuclear histomorphometric features are predictive of Oncotype DX risk categories in ductal carcinoma in situ: preliminary findings.
Haojia Li, Jon Whitney, +4 authors, Anant Madabhushi.
Breast Cancer Res, 2019 Oct 19; 21(1). PMID: 31623652    Free PMC article.
A Novel Digital Score for Abundance of Tumour Infiltrating Lymphocytes Predicts Disease Free Survival in Oral Squamous Cell Carcinoma.
Muhammad Shaban, Syed Ali Khurram, +6 authors, Nasir M Rajpoot.
Sci Rep, 2019 Sep 19; 9(1). PMID: 31527658    Free PMC article.
Quantitative Assessment of the Effects of Compression on Deep Learning in Digital Pathology Image Analysis.
Yijiang Chen, Andrew Janowczyk, Anant Madabhushi.
JCO Clin Cancer Inform, 2020 Mar 11; 4. PMID: 32155093    Free PMC article.
Replication Fork Remodeling and Therapy Escape in DNA Damage Response-Deficient Cancers.
Martin Liptay, Joana S Barbosa, Sven Rottenberg.
Front Oncol, 2020 May 21; 10. PMID: 32432041    Free PMC article.
Artificial intelligence in digital breast pathology: Techniques and applications.
Asmaa Ibrahim, Paul Gamble, +4 authors, Emad A Rakha.
Breast, 2020 Jan 15; 49. PMID: 31935669    Free PMC article.
A deep learning model to predict RNA-Seq expression of tumours from whole slide images.
Benoît Schmauch, Alberto Romagnoni, +11 authors, Gilles Wainrib.
Nat Commun, 2020 Aug 05; 11(1). PMID: 32747659    Free PMC article.
Introduction to Digital Image Analysis in Whole-slide Imaging: A White Paper from the Digital Pathology Association.
Famke Aeffner, Mark D Zarella, +11 authors, Douglas Bowman.
J Pathol Inform, 2019 Apr 16; 10. PMID: 30984469    Free PMC article.
Highly Cited. Review.
And They Said It Couldn't Be Done: Predicting Known Driver Mutations From H&E Slides.
Michael C Montalto, Robin Edwards.
J Pathol Inform, 2019 Jun 01; 10. PMID: 31149368    Free PMC article.
Computerized spermatogenesis staging (CSS) of mouse testis sections via quantitative histomorphological analysis.
Jun Xu, Haoda Lu, +6 authors, Eugene Yujun Xu.
Med Image Anal, 2021 Mar 07; 70. PMID: 33676102    Free PMC article.