Elizabeth S Burnside1, Karen Drukker2, Hui Li2, Ermelinda Bonaccio3, Margarita Zuley4, Marie Ganott4, Jose M Net5, Elizabeth J Sutton6, Kathleen R Brandt7, Gary J Whitman8, Suzanne D Conzen2, Li Lan2, Yuan Ji9,10, Yitan Zhu10, Carl C Jaffe11, Erich P Huang11, John B Freymann11, Justin S Kirby11, Elizabeth A Morris6, Maryellen L Giger2. 1. Department of Radiology, University of Wisconsin School of Medicine and Public Health, Madison, Wisconsin. 2. Department of Radiology, University of Chicago, Chicago, Illinois. 3. Department of Radiology, Roswell Park Cancer Institute, Buffalo, New York. 4. Department of Radiology, University of Pittsburgh, Pittsburgh, Pennsylvania. 5. University of Miami, Miller School of Medicine, Miami, Florida. 6. Department of Radiology, Memorial Sloan Kettering Cancer Center, New York, New York. 7. Department of Diagnostic Radiology, Mayo Clinic, Rochester, Minnesota. 8. Department of Diagnostic Imaging, The University of Texas MD Anderson Cancer Center, Houston, Texas. 9. Department of Health Studies, University of Chicago, Chicago, Illinois. 10. Program of Computational Genomics and Medicine, NorthShore University HealthSystem, Evanston, Illinois. 11. National Cancer Institute, National Institutes of Health, Rockville, Maryland.
Abstract
BACKGROUND: The objective of this study was to demonstrate that computer-extracted image phenotypes (CEIPs) of biopsy-proven breast cancer on magnetic resonance imaging (MRI) can accurately predict pathologic stage. METHODS: The authors used a data set of deidentified breast MRIs organized by the National Cancer Institute in The Cancer Imaging Archive. In total, 91 biopsy-proven breast cancers were analyzed from patients who had information available on pathologic stage (stage I, n = 22; stage II, n = 58; stage III, n = 11) and surgically verified lymph node status (negative lymph nodes, n = 46; ≥ 1 positive lymph node, n = 44; no lymph nodes examined, n = 1). Tumors were characterized according to 1) radiologist-measured size and 2) CEIP. Then, models were built that combined 2 CEIPs to predict tumor pathologic stage and lymph node involvement, and the models were evaluated in a leave-1-out, cross-validation analysis with the area under the receiver operating characteristic curve (AUC) as the value of interest. RESULTS: Tumor size was the most powerful predictor of pathologic stage, but CEIPs that captured biologic behavior also emerged as predictive (eg, stage I and II vs stage III demonstrated an AUC of 0.83). No size measure was successful in the prediction of positive lymph nodes, but adding a CEIP that described tumor "homogeneity" significantly improved discrimination (AUC = 0.62; P = .003) compared with chance. CONCLUSIONS: The current results indicate that MRI phenotypes have promise for predicting breast cancer pathologic stage and lymph node status. Cancer 2016;122:748-757.
BACKGROUND: The objective of this study was to demonstrate that computer-extracted image phenotypes (CEIPs) of biopsy-proven breast cancer on magnetic resonance imaging (MRI) can accurately predict pathologic stage. METHODS: The authors used a data set of deidentified breast MRIs organized by the National Cancer Institute in The Cancer Imaging Archive. In total, 91 biopsy-proven breast cancers were analyzed from patients who had information available on pathologic stage (stage I, n = 22; stage II, n = 58; stage III, n = 11) and surgically verified lymph node status (negative lymph nodes, n = 46; ≥ 1 positive lymph node, n = 44; no lymph nodes examined, n = 1). Tumors were characterized according to 1) radiologist-measured size and 2) CEIP. Then, models were built that combined 2 CEIPs to predict tumor pathologic stage and lymph node involvement, and the models were evaluated in a leave-1-out, cross-validation analysis with the area under the receiver operating characteristic curve (AUC) as the value of interest. RESULTS:Tumor size was the most powerful predictor of pathologic stage, but CEIPs that captured biologic behavior also emerged as predictive (eg, stage I and II vs stage III demonstrated an AUC of 0.83). No size measure was successful in the prediction of positive lymph nodes, but adding a CEIP that described tumor "homogeneity" significantly improved discrimination (AUC = 0.62; P = .003) compared with chance. CONCLUSIONS: The current results indicate that MRI phenotypes have promise for predicting breast cancer pathologic stage and lymph node status. Cancer 2016;122:748-757.
Authors: C Boetes; R D Mus; R Holland; J O Barentsz; S P Strijk; T Wobbes; J H Hendriks; S H Ruys Journal: Radiology Date: 1995-12 Impact factor: 11.105
Authors: Jia Wu; Yi Cui; Xiaoli Sun; Guohong Cao; Bailiang Li; Debra M Ikeda; Allison W Kurian; Ruijiang Li Journal: Clin Cancer Res Date: 2017-01-10 Impact factor: 12.531
Authors: Yitan Zhu; Abdallah S R Mohamed; Stephen Y Lai; Shengjie Yang; Aasheesh Kanwar; Lin Wei; Mona Kamal; Subhajit Sengupta; Hesham Elhalawani; Heath Skinner; Dennis S Mackin; Jay Shiao; Jay Messer; Andrew Wong; Yao Ding; Lifei Zhang; Laurence Court; Yuan Ji; Clifton D Fuller Journal: JCO Clin Cancer Inform Date: 2019-02
Authors: Jia Wu; Guohong Cao; Xiaoli Sun; Juheon Lee; Daniel L Rubin; Sandy Napel; Allison W Kurian; Bruce L Daniel; Ruijiang Li Journal: Radiology Date: 2018-05-01 Impact factor: 11.105
Authors: Mirella L Altoe; Kevin Kalinsky; Alessandro Marone; Hyun K Kim; Hua Guo; Hanina Hibshoosh; Mariella Tejada; Katherine D Crew; Melissa K Accordino; Meghna S Trivedi; Dawn L Hershman; Andreas H Hielscher Journal: Clin Cancer Res Date: 2021-01-15 Impact factor: 12.531