br H Zahreddine K L
 H. Zahreddine, K.L.B. Borden, Mechanism and insights into drug resistance in cancer, Front. Pharmacol. 4 (28) (2013) 1–8.  A.F. Castillo, U.D. Orlando, P. Lopez, A.R. Solano, P.M. Maloberti, E.J. Podesta, Gene expression profile and signaling pathways in MCF-7 breast cancer cells mediated by Acyl-Coa synthetase 4 overexpression, Transcriptomics 3 (2)
E. Bieberich, Guggulsterone and bexarotene induce secretion of exosome-associated breast cancer resistance protein and reduce doxorubicin resistance in MDA-MB-231 cells, Int. J. Cancer 137 (7) (2015) 1610–1620.  B. Askari, J.E. Kanter, A.M. Sherrid, D.L. Golej, A.T. Bender, J. Liu, W.A. Hsueh, J.A. Beavo, R.A. Coleman, K.E. Bornfeldt, Rosiglitazone inhibits acyl-CoA synthe-tase activity and fatty Cyclosporin H partitioning to diacylglycerol and triacylglycerol via a peroxisome proliferator-activated receptor-gamma-independent mechanism in human arterial smooth muscle cells and macrophages, Diabetes 56 (4) (2007) 1143–1152.
 B. Fischer, C. Frei, U. Moura, R. Stahel, E. Felley-Bosco, Inhibition of phosphoino-sitide-3 kinase pathway down regulates ABCG2 function and sensitizes malignant pleural mesothelioma to chemotherapy, Lung Cancer (Amsterdam, Netherlands) 78
F. Penault-Llorca, M. Bamdad, BCRP and P-gp relay overexpression in triple
negative basal-like breast cancer cell line: a prospective role in resistance to Olaparib, Sci. Rep. 5 (2015) 12670.
 P. Palasuberniam, X. Yang, D. Kraus, P. Jones, K.A. Myers, B. Chen, ABCG2 trans-porter inhibitor restores the sensitivity of triple negative breast cancer cells to aminolevulinic acid-mediated photodynamic therapy, Sci. Rep. 5 (2015) 13298.  E. Isik, H. Demirbilek, J.A.L. Houghton, S. Ellard, S.E. Flanagan, K. Hussain, Congenital hyperinsulinism and evolution to sulfonylurea-responsive diabetes later in life due to a novel homozygous p.L171F ABCC8 mutation, J. Clin. Res. Pediatr. Endocrinol. (2018).
L.S. Steelman, J.A. McCubrey, Inhibition of GSK-3beta activity can result in drug and hormonal resistance and alter sensitivity to targeted therapy in MCF-7 breast cancer cells, Cell Cycle (Georgetown, Tex.) 13 (5) (2014) 820–833.
63 Original Investigation
Additive Benefit of Radiomics Over Size Alone in the Distinction Between Benign Lesions and Luminal A Cancers on a Large Clinical Breast MRI Dataset
HeatherD1X M. Whitney, PhD,D2X NathanD3X S. Taylor,D4KarenD5X Drukker, PhD,D6X AlexandraD7X V. Edwards, MA,D8X JohnD9X Papaioannou, MS,D10X David1X Schacht, MD,D12X MPH, MaryellenD13X L. Giger, PhDD14X
Rationale and Objectives: The objective of this study was to demonstrate improvement in distinguishing between benign lesions and luminal A breast cancers in a large clinical breast magnetic resonance imaging database by using quantitative radiomics over maximum linear size alone.
Materials and Methods: In this retrospective study, 264 benign lesions and 390 luminal A breast cancers were automatically segmented from dynamic contrast-enhanced breast magnetic resonance images. Thirty-eight radiomic features were extracted. Tenfold cross valida-tion was performed to assess the ability to distinguish between lesions and cancers using maximum linear size alone and lesion signatures obtained with stepwise feature selection and a linear discriminant analysis classifier including and excluding size features. Area under the receiver operating characteristic curve (AUC) was used as the figure of merit.
Results: For maximum linear size alone, AUC and 95% confidence interval was 0.684 (0.642, 0.724) compared to 0.728 (0.687, 0.766) (P = 0.005) and 0.729 (0.689, 0.767) (P = 0.005) for lesion signature feature selection protocols including and excluding size features, respectively. The features of irregularity and entropy were chosen in all folds when size features were included and excluded. AUC for the radiomic signature using feature selection from all features was statistically equivalent to using feature selection from all features excluding size features, within an equivalence margin of 2%.
Conclusions: Inclusion of multiple radiomic features, automatically extracted from magnetic resonance images, in a lesion signature sig-nificantly improved the ability to distinguish between benign lesions and luminal A breast cancers, compared to using maximum linear size alone. The radiomic features of irregularity and entropy appear to play an important but not a solitary role within the context of feature selection and computer-aided diagnosis.
Key Words: Breast cancer; MRI; luminal A; radiomics.
© 2018 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.