Hao Xu1, Jieke Liu1, Zhe Chen1, Chunhua Wang1, Yuanyuan Liu1, Min Wang1, Peng Zhou1, Hongbing Luo2, Jing Ren3. 1. Department of Radiology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China. 2. Department of Radiology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China. rohbin@163.com. 3. Department of Radiology, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China. 13880611648@163.com.
Abstract
OBJECTIVES: To develop and validate radiomic models for preoperative prediction of intraductal component in invasive breast cancer (IBC-IC) using the intratumoral and peritumoral features derived from dynamic contrast-enhanced MRI (DCE-MRI). METHODS: The prediction models were developed in a primary cohort of 183 consecutive patients from September 2017 to December 2018, consisting of 45 IBC-IC and 138 invasive breast cancers (IBC). The validation cohort of 111 patients (27 IBC-IC and 84 IBC) from February 2019 to January 2020 was enrolled to test the prediction models. A total of 208 radiomic features were extracted from the intratumoral and peritumoral regions of MRI-visible tumors. Then the radiomic features were selected and combined with clinical characteristics to construct predicting models using the least absolute shrinkage and selection operator. The area under the curve (AUC) of receiver operating characteristic, sensitivity, and specificity were used to evaluate the performance of radiomic models. RESULTS: Four radiomic models for prediction of IBC-IC were built including intratumoral radiomic signature, peritumoral radiomic signature, peritumoral radiomic nomogram, and combined intratumoral and peritumoral radiomic signature. The combined intratumoral and peritumoral radiomic signature had the optimal diagnostic performance, with the AUC, sensitivity, and specificity of 0.821 (0.758-0.874), 0.822 (0.680-0.920), and 0.739 (0.658-0.810) in the primary cohort and 0.815 (0.730-0.882), 0.778 (0.577-0.914), and 0.738 (0.631-0.828) in the validation cohort. CONCLUSIONS: The radiomic model based on the combined intratumoral and peritumoral features from DCE-MRI showed a good ability to preoperatively predict IBC-IC, which might facilitate the individualized surgical planning for patients with breast cancer before breast-conserving surgery. KEY POINTS: •·Preoperative prediction of intraductal component in invasive breast cancer is crucial for breast-conserving surgery planning. • Peritumoral radiomic features of invasive breast cancer contain useful information to predict intraductal components. •·Radiomics is a promising non-invasive method to facilitate individualized surgical planning for patients with breast cancer before breast-conserving surgery.
OBJECTIVES: To develop and validate radiomic models for preoperative prediction of intraductal component in invasive breast cancer (IBC-IC) using the intratumoral and peritumoral features derived from dynamic contrast-enhanced MRI (DCE-MRI). METHODS: The prediction models were developed in a primary cohort of 183 consecutive patients from September 2017 to December 2018, consisting of 45 IBC-IC and 138 invasive breast cancers (IBC). The validation cohort of 111 patients (27 IBC-IC and 84 IBC) from February 2019 to January 2020 was enrolled to test the prediction models. A total of 208 radiomic features were extracted from the intratumoral and peritumoral regions of MRI-visible tumors. Then the radiomic features were selected and combined with clinical characteristics to construct predicting models using the least absolute shrinkage and selection operator. The area under the curve (AUC) of receiver operating characteristic, sensitivity, and specificity were used to evaluate the performance of radiomic models. RESULTS: Four radiomic models for prediction of IBC-IC were built including intratumoral radiomic signature, peritumoral radiomic signature, peritumoral radiomic nomogram, and combined intratumoral and peritumoral radiomic signature. The combined intratumoral and peritumoral radiomic signature had the optimal diagnostic performance, with the AUC, sensitivity, and specificity of 0.821 (0.758-0.874), 0.822 (0.680-0.920), and 0.739 (0.658-0.810) in the primary cohort and 0.815 (0.730-0.882), 0.778 (0.577-0.914), and 0.738 (0.631-0.828) in the validation cohort. CONCLUSIONS: The radiomic model based on the combined intratumoral and peritumoral features from DCE-MRI showed a good ability to preoperatively predict IBC-IC, which might facilitate the individualized surgical planning for patients with breast cancer before breast-conserving surgery. KEY POINTS: •·Preoperative prediction of intraductal component in invasive breast cancer is crucial for breast-conserving surgery planning. • Peritumoral radiomic features of invasive breast cancer contain useful information to predict intraductal components. •·Radiomics is a promising non-invasive method to facilitate individualized surgical planning for patients with breast cancer before breast-conserving surgery.
Authors: A C Voogd; M Nielsen; J L Peterse; M Blichert-Toft; H Bartelink; M Overgaard; G van Tienhoven; K W Andersen; R J Sylvester; J A van Dongen Journal: J Clin Oncol Date: 2001-03-15 Impact factor: 44.544
Authors: M Luke Marinovich; Lamiae Azizi; Petra Macaskill; Les Irwig; Monica Morrow; Lawrence J Solin; Nehmat Houssami Journal: Ann Surg Oncol Date: 2016-08-15 Impact factor: 5.344
Authors: Marissa C van Maaren; Linda de Munck; Geertruida H de Bock; Jan J Jobsen; Thijs van Dalen; Sabine C Linn; Philip Poortmans; Luc J A Strobbe; Sabine Siesling Journal: Lancet Oncol Date: 2016-06-22 Impact factor: 41.316
Authors: Martin Sjöström; S Laura Chang; Nick Fishbane; Elai Davicioni; Shuang G Zhao; Linda Hartman; Erik Holmberg; Felix Y Feng; Corey W Speers; Lori J Pierce; Per Malmström; Mårten Fernö; Per Karlsson Journal: J Clin Oncol Date: 2019-10-16 Impact factor: 44.544