1
|
Nissan N, Massasa EEM, Bauer E, Halshtok-Neiman O, Shalmon A, Gotlieb M, Faermann R, Samoocha D, Yagil Y, Ziv-Baran T, Anaby D, Sklair-Levy M. MRI can accurately diagnose breast cancer during lactation. Eur Radiol 2023; 33:2935-2944. [PMID: 36348090 DOI: 10.1007/s00330-022-09234-z] [Citation(s) in RCA: 6] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/04/2022] [Revised: 08/27/2022] [Accepted: 10/10/2022] [Indexed: 11/09/2022]
Abstract
OBJECTIVE To test the diagnostic performance of breast dynamic contrast-enhanced (DCE) MRI during lactation. MATERIALS AND METHODS Datasets of 198 lactating patients, including 66 pregnancy-associated breast cancer (PABC) patients and 132 controls, who were scanned by DCE on 1.5-T MRI, were retrospectively evaluated. Six blinded, expert radiologists independently read a single DCE maximal intensity projection (MIP) image for each case and were asked to determine whether malignancy was suspected and the background-parenchymal-enhancement (BPE) grade. Likewise, computer-aided diagnosis CAD MIP images were independently read by the readers. Contrast-to-noise ratio (CNR) analysis was measured and compared among four consecutive acquisitions of DCE subtraction images. RESULTS For MIP-DCE images, the readers achieved the following means: sensitivity 93.3%, specificity 80.3%, positive-predictive-value 70.4, negative-predictive-value 96.2, and diagnostic accuracy of 84.6%, with a substantial inter-rater agreement (Kappa = 0.673, p value < 0.001). Most false-positive interpretations were attributed to either the MIP presentation, an underlying benign lesion, or an asymmetric appearance due to prior treatments. CAD's derived diagnostic accuracy was similar (p = 0.41). BPE grades were significantly increased in the healthy controls compared to the PABC cohort (p < 0.001). CNR significantly decreased by 11-13% in each of the four post-contrast images (p < 0.001). CONCLUSION Breast DCE MRI maintains its high efficiency among the lactating population, probably due to a vascular-steal phenomenon, which causes a significant reduction of BPE in cancer cases. Upon validation by prospective, multicenter trials, this study could open up the opportunity for breast MRI to be indicated in the screening and diagnosis of lactating patients, with the aim of facilitating an earlier diagnosis of PABC. KEY POINTS • A single DCE MIP image was sufficient to reach a mean sensitivity of 93.3% and NPV of 96.2%, to stress the high efficiency of breast MRI during lactation. • Reduction in BPE among PABC patients compared to the lactating controls suggests that several factors, including a possible vascular steal phenomenon, may affect cancer patients. • Reduction in CNR along four consecutive post-contrast acquisitions highlights the differences in breast carcinoma and BPE kinetics and explains the sufficient conspicuity on the first subtracted image.
Collapse
Affiliation(s)
- Noam Nissan
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel.
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel.
| | - Efi Efraim Moss Massasa
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
| | - Ethan Bauer
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Osnat Halshtok-Neiman
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Anat Shalmon
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Michael Gotlieb
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Renata Faermann
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - David Samoocha
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Yael Yagil
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Tomer Ziv-Baran
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Debbie Anaby
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Miri Sklair-Levy
- Department of Radiology, Sheba Medical Center, Emek Ha-Ella 1 st. Tel Hashomer, 5265601, Ramat Gan, Israel
- Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
| |
Collapse
|
2
|
Feng B, Liu Z, Liu Y, Chen Y, Zhou H, Cui E, Li X, Chen X, Li R, Yu T, Zhang L, Long W. Predicting lymphovascular invasion in clinically node-negative breast cancer detected by abbreviated magnetic resonance imaging: Transfer learning vs. radiomics. Front Oncol 2022; 12:890659. [PMID: 36185309 PMCID: PMC9520481 DOI: 10.3389/fonc.2022.890659] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/06/2022] [Accepted: 08/25/2022] [Indexed: 11/13/2022] Open
Abstract
Objective To compare the performance of abbreviated breast magnetic resonance imaging (AB-MRI)-based transfer learning (TL) algorithm and radionics analysis for lymphovascular invasion (LVI) prediction in patients with clinically node-negative invasive breast cancer (IBC). Methods Between November 2017 and October 2020, 233 clinically node-negative IBCs detected by AB-MRI were retrospectively enrolled. One hundred thirty IBCs from center 1 (37 LVI-positive and 93 LVI-negative) were assigned as the training cohort and 103 from center 2 (25 LVI-positive and 78 LVI-negative) as the validation cohort. Based on AB-MRI, a TL signature (TLS) and a radiomics signature (RS) were built with the least absolute shrinkage and selection operator (LASSO) logistic regression. Their diagnostic performances were validated and compared using areas under the receiver operating curve (AUCs), net reclassification improvement (NRI), integrated discrimination improvement (IDI), decision curve analysis (DCA), and stratification analysis. A convolutional filter visualization technique was used to map the response areas of LVI on the AB-MRI. Results In the validation cohort, compared with RS, the TLS showed better capability in discriminating LVI-positive from LVI-negative lesions (AUC: 0.852 vs. 0.726, p < 0.001; IDI = 0.092, p < 0.001; NRI = 0.554, p < 0.001). The diagnostic performance of TLS was not affected by the menstrual state, molecular subtype, or contrast agent type (all p > 0.05). Moreover, DCA showed that the TLS added more net benefit than RS for clinical utility. Conclusions An AB-MRI-based TLS was superior to RS for preoperative LVI prediction in patients with clinically node-negative IBC.
Collapse
Affiliation(s)
- Bao Feng
- Department of Radiology, Jiangmen Central Hospital, Jiangmen, Guangdong, China
- School of Electronic Information and Automation, Guilin University of Aerospace Technology, Guilin, China
| | - Zhuangsheng Liu
- Department of Radiology, Jiangmen Central Hospital, Jiangmen, Guangdong, China
| | - Yu Liu
- School of Electronic Information and Automation, Guilin University of Aerospace Technology, Guilin, China
| | - Yehang Chen
- School of Electronic Information and Automation, Guilin University of Aerospace Technology, Guilin, China
| | - Haoyang Zhou
- School of Electronic Information and Automation, Guilin University of Aerospace Technology, Guilin, China
| | - Enming Cui
- Department of Radiology, Jiangmen Central Hospital, Jiangmen, Guangdong, China
| | - Xiaoping Li
- Department of Breast, Jiangmen Central Hospital, Jiangmen, Guangdong, China
| | - Xiangmeng Chen
- Department of Radiology, Jiangmen Central Hospital, Jiangmen, Guangdong, China
| | - Ronggang Li
- Department of Pathology, Jiangmen Central Hospital, Jiangmen, Guangdong, China
| | - Tianyou Yu
- School of Automation Science and Engineering, South China University of Technology, Guangzhou, China
| | - Ling Zhang
- Department of Radiology, Nanfang Hospital, Southern Medical University, Guangzhou, China
- *Correspondence: Ling Zhang, ; Wansheng Long,
| | - Wansheng Long
- Department of Radiology, Jiangmen Central Hospital, Jiangmen, Guangdong, China
- *Correspondence: Ling Zhang, ; Wansheng Long,
| |
Collapse
|