Comparison of Contrast-Enhanced Spectral Mammography and Contrast-Enhanced MRI in Screening Multifocal and Multicentric Lesions in Breast Cancer Patients

Objectives We aimed to determine the difference between contrast-enhanced spectral mammography (CESM) and contrast-enhanced magnetic resonance imaging (CE-MRI) in detecting multifocal and multicentric breast cancer (MMBC). Methods : This study was conducted among breast cancer patients between July 1, 2017, and May 30, 2021. The sensitivity, specificity, and accuracy of CESM and CE-MRI in the diagnosis of MMBC were evaluated with pathological results as the gold standard. Results A total of 188 lesions were detected in 54 patients with MMBC, including 177 breast cancer and 11 benign lesions. Based on CESM and CE-MRI, 4 false-positive cases and 3 false-negative cases and 7 false-positive cases and 1 false-negative case, respectively, were found. The accuracy of CESM was higher than that of MRI (96.3% vs 95.7%), and the specificity was higher than that of MRI (63.6% vs 36.4%). There were no significant differences in the sensitivity, specificity, and accuracy for the detection of MMBC between CESM and CE-MRI (p = 0.500; p = 0.250; p = 0.792). Conclusion CESM is an effective method for the detection of MMBC, which is consistent with the sensitivity and accuracy of CE-MRI.


Introduction
Breast cancer is one of the most common fatal malignancies in women. Preoperative assessment when choosing between total and conservative mastectomy depends on the extent of the cancer and whether the cancer is multifocal or multicentric [1,2]. Compared with breast cancer patients presenting with a single focal lesion, multifocal and multicentric breast cancer (MMBC) showed a higher risk of lymph node metastasis and aggression, a higher degree of malignancy, as well as a poorer prognosis [3,4]. e incidence of MMBC has a wide variation (6%-60%) among different clinical studies, mostly due to the lack of a standardized classification [5]. erefore, a comprehensive preoperative imaging evaluation, such as first-line imaging evaluation by mammography and ultrasonography (US) followed by contrast-enhanced magnetic resonance imaging (CE-MRI), is essential for patients suspected with multiple breast cancers.
Mammography is a conventional modality of breast screening or clinical diagnosis of breast cancers. e cancer detection sensitivity is only about 50% to 60% in dense breasts because of poor contrast between the cancer and the background [6]. us, there is a high possibility of underdetection. US is commonly used for cancer detection, as well as preoperative evaluation of cancer status due to the advantages of being handheld, convenient, and radiation-free, but single US application usually showed poor consistency with pathological diagnosis [7]. A few studies have proposed the limitations of classic ultrasonographic differentiation between benign and malignant masses. Ultrasonographybased texture analysis (USTA) offers a new perspective. It is unclear whether the parameters can reflect the histopathological characteristics of lesions [8]. Recently, preoperative contrast-enhanced ultrasonography only showed good predictive value for the sentinel lymph node detection in patients with breast cancer [8]. CE-MRI is considered to be one of the most accurate imaging methods for the diagnosis of breast cancer [9]. Although it has a high sensitivity for identifying MMBC from 88% to 100%, the specificity and positive predictive value (PPV) of CE-MRI are limited as both benign and malignant lesions presenting enhancement [10]. Additionally, it is more expensive, time-consuming, and not easily accessible compared with mammography.
Recently, an emerging imaging technique of contrastenhanced spectral mammography (CESM) provides a lowenergy mammogram (LE-MG) and a recombined subtracted mammogram (RSM) after intravenously administering iodinated contrast medium images in the same session of examination within a short time. CESM allows both a morphologic evaluation comparable to routine digital mammography and a simultaneous assessment of tumor neovascularity as an indicator of malignancy [11].
Preliminary studies indicated that CESM showed an extremely high sensitivity to breast cancer. To date, a few studies have focused on the comparison of the screening efficiency for malignancies between CESM and CE-MRI [12,13]. e studies reported better accuracy and specificity and a decreased false-positive rate of CESM in breast cancer detection than those of MRI. Recently, a few studies have been available to assess the effectiveness of the radiomics analysis of CESM in discriminating between breast cancers and background parenchymal enhancement (BPE), as well as to discrimination of benign and malignant breast lesions [14,15]. However, to the best of our knowledge, few studies have focused on the diagnostic performance of CESM in the diagnosis of MMBC [1]. Accordingly, this study aimed to compare the diagnostic performance of CESM in the discovery of MMBC compared with CE-MRI to determine whether CESM could result in changes to surgical management.

Inclusion and Exclusion
Criteria. Between July 1, 2017, and May 30, 2021, MMBC patients were included in this study based on the following criteria: (1) the final diagnosis with MMBC was established via core needle biopsy under the guidance of US and/or surgery. Histopathologic diagnosis was utilized as the standard; (2) MRI and CESM were scheduled within 3 days in premenopausal women and within 14 days in postmenopausal women; and (3) we included all types of breast (A-D) based on the American College of Radiology (ACR). Patients were excluded based on the following criteria: (1) the patient with single focal breast cancer; (2) the patient was pregnant or lactating; (3) the patient with claustrophobia; (4) the patient received implanted pacemakers or metal implants previously; and (5) the patient with hyperthyroidism, or a history of allergy to iodine or gadolinium contrast agent, or severe hepatorenal dysfunction with a glomerular filtration rate of <30 ml/min.

CE-MRI Examinations.
e 3.0 T superconducting MR instrument (Magnetom Aera; Siemens, Erlangen, Germany) equipped with the 16-channel breast coil was used for the CE-MRI scan. e patients were in a prone position with bilateral breasts loosing naturally within the breast-dedicated coils. After conventional plain scan, diffusion weighted imaging (DWI) was performed, followed by CE-MRI scan. e contrast medium (Gd-DTPA) was used and injected at a dose of 0.2 mmol/kg and a rate of 2 ml/s. e conventional plain scan sequences were as follows: T1WI sequence (TR 6.0 ms, TE 2.46 ms), fast reversal recovery of T1WI sequence by fat suppression (TR 4,000 ms, TE 54 ms, slice thickness 4.0 mm), and bilateral mammary sagittal T2WI sequence (TR 3,300 ms, TE 70 ms, slice thickness 4.0 mm); and horizontal axis DWI sequence: TR 4,730 ms, TE 47 ms, slice thickness 5.0 mm, b � 0, 400, 800 s/mm 2 . For the CE-MRI, fast 3D dynamic imaging fat suppression T1WI sequence was used: TR 4.66 ms, TE 1.70 ms, slice thickness 1.6 mm. One time phase was scanned before injection of Gd-DTPA, and six time phases were scanned again after injection with an interval of 64 sec. en, the digital subtraction was performed on the transverse and coronal planes. For suspicious masses, a time-signal strength curve was added.

CESM Examinations.
GE digital mammography equipment (Senographe Essential; GE Healthcare, Buc, France) was used for the CESM under the assistance of double-cylinder high-voltage syringe (Ulrich, Germany). Each patient received ioversol injection (1.5 ml/kg) via the radial vein, with a flow rate of 3 ml/s. About within 2 min after ioversol injection, the craniocaudal (CC) position of the affected breast was first photographed, and then the CC images of the healthy breast were obtained. e mediallateral oblique (MLO) position of the affected breast was scanned, and then the MLO images of the healthy breast were obtained. e examination was completed within 6 min  after injection. Briefly, a pair of LE-MG and high-energy MG is obtained, and then the images are subtracted from each other. After CESM, patients were recommended to drink more water, followed by observation for 30 min until presence of no adverse reactions.

Image
Analysis. CESM and CE-MRI images were independently analyzed by three radiologists with 10-year working experience in breast cancer diagnosis who were blinded to patients' conditions. All the images were reviewed based on the criteria from the BI-RADS lexicon [16]. Digital mammography BI-RADS descriptors of lesions can be applied for the morphologic analysis of mass lesions on the low-energy MG imaging of CESM. e RSM of CESM images was evaluated using criteria related to contrast enhancement intensity and morphology according to the MRI part of the BI-RADS lexicon [17]. Subjective judgment of lesion enhancement was performed based on the scale of none, mild, moderate, and strong [18]. e lesions of a BI-RADS category of equal or less than 3 were diagnosed as benign, and those of above BI-RADS 4 (BI-RADS 4 A, B, C and BI-RADS 5) were diagnosed as malignant. For the detection of lesions, multifocal cancer was confirmed in the presence of two or more malignant focus in the same quadrant. Multicentric cancer was confirmed in cases of involvement of different quadrants.

Pathological Evaluation of Specimens.
e specimens were fixed, embedded, and subjected to hematoxylin-eosin (HE) staining and immunohistochemical analysis. Pathological analysis was performed by pathologists with 10-year working experience based on the Pathological Classification and Diagnostic Criteria of Breast Tumors (2012) proposed by the World Health Organization (WHO) [19]. A caliper was utilized to assess the gross specimen for large specimens, while the small specimens were measured under a microscope.

Statistical Analysis.
For statistical analysis, BI-RADS 1-3 and BI-RADS ≥ 4 were defined as suspicious benign and probably malignant, respectively. e chi-square test was used to compare the sensitivity, specificity, and accuracy in detecting MMBC between CESM and CE-MRI. With the pathological results as the criterion standard, the diagnostic accuracy was compared among different images. All statistical analysis was performed with IBM SPSS Statistics for Windows, version 20.0 (IBM Corp; Armonk, NY, USA) software. p < 0.05 was considered to be statistically significant.

Patient Characteristics and Pathological Diagnosis.
e study population consisted of 54 patients. All 54 patients were female (median age: 48.7 years; range: 33-73 years) and were proven to be MMBC (type A: 4; type B: 24; type C:

Discussion
CE-MRI has been acknowledged as one of the most accurate imaging methods for the diagnosis of breast cancer [20]. In a previous study, Rabasco et al. reported a higher sensitivity (100%) and an accuracy rate (98.4%) of MRI in screening MMBC compared with mammography (40%, 92.5%), as well as the combination of mammography and US (57.1%, 92.5%) [21]. Our study demonstrated that CESM possessed a high sensitivity and might serve as a promising candidate for screening MMBC. ere were no significant differences in the sensitivity, specificity, and accuracy between CESM and CE-MRI (p � 0.500; p � 0.250; p � 0.792, respectively).
CESM provided more comprehensive information for the diagnosis of MMBC. In the presence of the combination of LE-MG imaging and RSM imaging, CESM can show the mass and shape of the tumor, as well as the characteristics of malignant calcification and enhancement. In clinical practice, breast cancer patients showed similar enhancement in CESM and CE-MRI, and the morphological descriptors for these two techniques were also similar [22]. CE-MRI contributed to the MMBC diagnosis based on the shape and enhancement curve of the tumor and the signal characteristics of DWI; however, its specificity was comparatively low in a range of 37%-97% [23], as it could not show the characteristic calcification of breast malignancy. is would result in invasive diagnostic tests (e.g., core needle biopsy) that were not necessary. Some scholars concluded that CESM was superior to MRI in detecting noninvasive carcinomas, especially if there were only clustered calcifications in the lesions. Previous studies showed that the specificity of CESM was up to 90% [11]. Our results also confirmed that the specificity of CESM was higher than that of CE-MRI (36.4% vs 63.6%). Among the 177 malignant, 54 lesions (30.51%) showed clustered microcalcifications in the LE-MG imaging.
irty-five lesions (64.8%) were assessed as BI-RADS 5, while for CE-MRI, 22 (40.7%) were diagnosed as BI-RADS 5. erefore, CESM LE-MG imaging can detect more malignant calcifications in the breast than CE-MRI, which increased the confidence in the diagnosis of breast cancer.
CESM showed different enhancement in RSM imaging between benign and malignant lesions. Benign lesions showed weak or no enhancement. Among the 165 malignant lesions detected by CESM, 160 (96.9%) lesions were IDC, in which 154 (96.3%) lesions presented markedly heterogeneous enhancement and the other 6 lesions (3.7%) showed mild enhancement. All the 5 DCIS were manifested as mild enhancement. Hence, IDC was mainly manifested as severe heterogeneous enhancement, while DCIS was mainly manifested as mild enhancement. IDC lesions presented higher enhancement than that of DCIS. Our data also showed that non-mass lesions with mild enhancements or clustered microcalcification lesions with mild enhancements were closely related to DCIS (100%, 5/5), which was concordant with that of the previous literature description [24]. is was speculated to be associated with more aggressive behavior and abundant blood flow of IDC than DCIS. In a previous study, approximately 11% of the high-grade DCIS showed no enhancement [25]. In our study, none of the patients showed such type of breast cancer, which may be related to the fact that the sample size for the carcinoma in situ was relatively small. Interestingly, the smallest DCIS in the CESM images showed a diameter of 3 mm. In our future study, we will focus on finding more characteristics of breast cancer to improve the diagnostic confidence of CESM.
In addition, CESM could provide a more intuitive image, which was beneficial to the preoperative localization of lesions. We found a special case occasionally. In the CESM images of the patients, there was a malignant mass and a nodule (about 3 mm in diameter) with significant enhancement in the subcutaneous tissues of the nipple. e morphology and location were consistent on both CESM and CE-MRI. A small hemangioma was confirmed after surgery ( Figure 4). Obviously, it was easier and more convenient to find out the small nodules based on CESM images. In addition, the body position for CE-MRI was prone position not the supine position. erefore, the localization of small nodules in breast was more difficult in CE-MRI than that of CESM.
ere are some limitations for CESM. For example, it required close cooperation of the patients. During the CESM scan, the breast cancers in the inner quadrant of the breast adjacent to the chest wall were more likely to be neglected. One patient with tumor in the deep internal quadrant close to the chest wall was misdiagnosed as she felt pain after puncture at the wound site ( Figure 5). erefore, CESM examination must be performed prior to targeted puncture. In addition, in order to maximally include all the breast tissues, muscle relax in the thoracic wall was recommended.
Eleven benign lesions were misdiagnosed as MMBC including 5 lesions of fibroadenoma, 3 lesions of fibrocystic lesion with ductal hyperplasia, 2 lesions of breast adenosis with fibroadenoma, and 1 lesion of ductal dilatation of the breast. Seven lesions were misdiagnosed as MMBC by CE-MRI, and four lesions were misdiagnosed as MMBC by CESM. Breast adenosis, fibroadenoma, and ductal dilatation of the breast cases were all presented multifocal and mild enhancement due to which it was difficult to distinguish DCIS. is is associated with the small sample size for carcinoma in situ, together with a lack of experience on the differential diagnosis. is is a retrospective analysis. All the cases included in this study received CESM and MRI within 3-14 days. In our hospital, we can only perform US-guided puncture, rather than X-ray-guided or MRIguided puncture. In cases of few small lesions or multiclustered calcification by X-ray or by MRI, which cannot be detected by US, it is still difficult to remove lesions surgically.
Compared with MR diagnosis, CESM showed the characteristics of less time-consuming, low cost, and sensitivity to calcification; however, it was still a challenge to wrap all of the mass adjacent to the chest wall as it may result in FN. MRI had the advantages of being able to image the entire chest wall and axilla, and it involved no ionizing radiation. Although iodinated contrast was generally considered to be significantly more hazardous than gadolinium contrast, whether the deposition of gadolinium had any actual consequences to human health was unknown.

Conclusions
In conclusion, CESM can display breast lesions and has diagnostic efficacy equivalent to CE-MRI in detecting MMBC. CESM promoted the identification of MMBC, which was beneficial to the surgery selection, generation of negative surgical margins, and avoiding treatment failure.
Data Availability e data that support the findings of this study are available from the corresponding author, upon reasonable request.
Ethical Approval e retrospective study was reviewed and approved by the Ethical Committee of Taian City Central Hospital (No. 2017-04-003). In CESM, the LE-MG and RSM imaging indicated asymmetric density shadow in the lower quadrant, with mild ductal enhancement (wide arrow). e diagnosis was BI-RADS 4A. e pectoralis major muscle was not well displayed, which may not be diagnosed. (e) US indicated a hypoechoic mass with irregular edge (crosses), which was diagnosed as BI-RADS 4C. (f, g) T1WI image of CE-MRI fat compression in axial position showed the first mass with irregular and heterogeneous enhancement, which could be seen in the lower inner quadrant of the right breast (wide arrow). e second lesion showed non-mass and ductal enhancement around the mass (wide arrow). (h) e first mass was confirmed to be IDC after surgery, as shown in HE staining under a magnification of 20×. (i) e non-mass lesion was DCIS, as shown in HE staining under a magnification of 10×.

Consent
Written informed consent was obtained from all patients in this study.