Jon Whitney1, German Corredor2, Andrew Janowczyk3, Shridar Ganesan4, Scott Doyle5, John Tomaszewski5, Michael Feldman6, Hannah Gilmore7, Anant Madabhushi3. 1. Department of Biomedical Engineering, Case Western Reserve University, 2071 Martin Luther King Drive, Cleveland, OH, 44106-7207, USA. Jon.whitney@case.edu. 2. Universidad Nacional de Colombia, Bogotá D.C, Colombia. 3. Department of Biomedical Engineering, Case Western Reserve University, 2071 Martin Luther King Drive, Cleveland, OH, 44106-7207, USA. 4. Department of Medicine, Division of Medical Oncology, Rutgers Robert Wood Johnson Medical School, Rutgers Cancer Institute of New Jersey, 195 Little Albany Street, New Brunswick, NJ, 08903, USA. 5. SUNY at the University at Buffalo, 3435 Main Street, Buffalo, NY, USA. 6. Department of Pathology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA. 7. Department of Pathology, University Hospitals, Cleveland Medical Center and Case Western Reserve University, Cleveland, OH, 44106, USA.
Abstract
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.
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 cancerpatients 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.
Authors: Andrew H Beck; Ankur R Sangoi; Samuel Leung; Robert J Marinelli; Torsten O Nielsen; Marc J van de Vijver; Robert B West; Matt van de Rijn; Daphne Koller Journal: Sci Transl Med Date: 2011-11-09 Impact factor: 17.956
Authors: Ben S Wittner; Dennis C Sgroi; Paula D Ryan; Tako J Bruinsma; Annuska M Glas; Anitha Male; Sonika Dahiya; Karleen Habin; Rene Bernards; Daniel A Haber; Laura J Van't Veer; Sridhar Ramaswamy Journal: Clin Cancer Res Date: 2008-05-15 Impact factor: 12.531
Authors: John M S Bartlett; Jane Bayani; Andrea Marshall; Janet A Dunn; Amy Campbell; Carrie Cunningham; Monika S Sobol; Peter S Hall; Christopher J Poole; David A Cameron; Helena M Earl; Daniel W Rea; Iain R Macpherson; Peter Canney; Adele Francis; Christopher McCabe; Sarah E Pinder; Luke Hughes-Davies; Andreas Makris; Robert C Stein Journal: J Natl Cancer Inst Date: 2016-04-29 Impact factor: 13.506
Authors: Molly E Klein; David J Dabbs; Yongli Shuai; Adam M Brufsky; Rachel Jankowitz; Shannon L Puhalla; Rohit Bhargava Journal: Mod Pathol Date: 2013-03-15 Impact factor: 7.842
Authors: Kun-Hsing Yu; Ce Zhang; Gerald J Berry; Russ B Altman; Christopher Ré; Daniel L Rubin; Michael Snyder Journal: Nat Commun Date: 2016-08-16 Impact factor: 14.919
Authors: Mustafa Yousif; Paul J van Diest; Arvydas Laurinavicius; David Rimm; Jeroen van der Laak; Anant Madabhushi; Stuart Schnitt; Liron Pantanowitz Journal: Virchows Arch Date: 2021-11-18 Impact factor: 4.064
Authors: Famke Aeffner; Mark D Zarella; Nathan Buchbinder; Marilyn M Bui; Matthew R Goodman; Douglas J Hartman; Giovanni M Lujan; Mariam A Molani; Anil V Parwani; Kate Lillard; Oliver C Turner; Venkata N P Vemuri; Ana G Yuil-Valdes; Douglas Bowman Journal: J Pathol Inform Date: 2019-03-08
Authors: Haojia Li; Jon Whitney; Kaustav Bera; Hannah Gilmore; Mangesh A Thorat; Sunil Badve; Anant Madabhushi Journal: Breast Cancer Res Date: 2019-10-17 Impact factor: 6.466