Differentiation of Benign and Malignant Sub-1-cm Breast Lesions Using Contrast-Enhanced Sonography

ORIGINAL RESEARCH Differentiation of Benign and Malignant Sub-1-cm Breast Lesions Using Contrast-Enhanced Sonography He Liu, MD, Yuxin Jiang, MD, Qing Dai, MD, Qingli Zhu, MD, Liang Wang, MD, Jing Zhang, MD, Qian Yang, MD Received February 24, 2014, from the Ultrasound Department, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, Beijing, China. Revision requested March 26, 2014. Revised manuscript accepted for publication April 14, 2014. The research was supported by the National Natural Sciences Foundation of China (grant 30900372). Address correspondence to Yuxin Jiang, MD, Ultrasound Department, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences, 1 Shuaifuyuan, Dong Cheng District, 100730 Beijing, China. E-mail: jiangyuxinxh@163.com Abbreviations BI-RADS; Breast Imaging Reporting and Data System; DCIS, ductal carcinoma in situ doi:10.7863/ultra.34.1.117 Objectives The purpose of this study was to prospectively assess the diagnostic efficacy of contrast-enhanced sonography for differential diagnosis of sub-1-cm breast lesions. Methods Contrast-enhanced sonography was performed in 46 women with 46 sub- 1-cm breast lesions scheduled for surgery or biopsy. Histologic results were used as a reference standard. The contrast enhancement pattern, enhancement degree, direction, margin, radial vessels surrounding the lesion, and lesion size discrepancy between contrastenhanced and conventional sonography were evaluated. Results The detection rates for increased size, radial vessels surrounding the lesion, and hyperenhancement on contrast-enhanced sonography were significantly higher in the malignant than the benign group (P <.05). The sensitivity, specificity, and accuracy for increased size were 72.7%, 80.0%, and 78.2%, respectively. The sensitivity, specificity, and accuracy for radial vessels surrounding the lesion were 54.5%, 97.1%, and 87.0%. The sensitivity, specificity, and accuracy for hyperenhancement were 81.8%, 60.0%, and 65.0%. No significant difference was found for the enhancement pattern, direction, and margin between the groups. Conclusions Contrast-enhanced sonographic features of sub-1-cm breast lesions included increased size on contrast-enhanced sonography, radial vessels surrounding the lesion, and hyperenhancement. Identification of these features is useful for differentiation of small breast lesions. Key Words breast cancer; breast ultrasound; contrast-enhanced sonography; differential diagnosis; small lesions S mall (sub-1-cm) breast cancers are diagnosed with increased frequency because of the broader use of screening mammography in the general population. 1 3 Most small cancers are hormone receptor positive, low histologic grade tumors, with a higher proportion of special tumor types associated with an excellent prognosis. 4,5 The relapse-free survival of invasive ductal and lobular lesions less than 1 cm in diameter was reported as 88% at 20 years. 6 The detection and subsequent characterization of small breast lesions are therefore of paramount importance. Unfortunately, as the lesions get smaller, it becomes increasingly difficult for radiologists to distinguish them using morphologic parameters alone. With the use of contrast-enhanced sonography, the morphologic characteristics and hemodynamics of the microvessels can now be appreciated. It is the abnormal vascularity that, if visualized adequately, may hold the key to distinguishing benign from malignant lesions. 2015 by the American Institute of Ultrasound in Medicine J Ultrasound Med 2015; 34:117 123 0278-4297 www.aium.org

The purpose of our study, therefore, was to prospectively evaluate the diagnostic efficacy of contrast-enhanced sonography for differentiation of sub-1-cm breast lesions. Materials and Methods Patients The study design and protocol were approved by the Ethics Committee of our institution, and written informed consent was obtained from all participants. From April 2012 to October 2012, 269 consecutive patients with American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) category 3 to 5 lesions on conventional sonography underwent preoperative contrast- enhanced sonography. The inclusion criterion was the presence of breast lesions on conventional sonography. The exclusion criteria were contraindications to contrast-enhanced sonography, pregnancy or breast-feeding, and treatment with neoadjuvant chemotherapy. When a patient had multiple lesions, only the most suspicious lesion was evaluated. The histopathologic results of surgery or biopsy served as the reference standard for the diagnosis. Of the 269 patients, 46 (age range, 18 74 years; mean age ± SD, 40 ± 11 years; median age, 44 years) with 46 sub-1-cm breast lesions on conventional sonography constituted this study group. Sonographic Examinations An iu22 ultrasound system (Philips Healthcare, Bothell, WA) with a 12 5-MHz linear transducer for conventional sonography and an 8 4-MHz linear transducer for contrastenhanced sonography was used. All sonographic examinations were performed by a single experienced ultrasound physician (H.L., with 8 years of experience in breast sonography and 3 years of experience in breast contrast-enhanced sonography). The examiner was aware of the presence and location of the breast lesion but was blinded to any other previously collected clinical and imaging data. The lesion was located and imaged on conventional sonography. The plane of the maximum lesion diameter was selected for contrast-enhanced sonography. The selected plane included the lesion and its surrounding normal tissue if possible, and it remained unchanged during contrastenhanced sonography. The contrast agent used was SonoVue (Bracco SpA, Milan, Italy). It consists of a lyophilized powder of phospholipid-stabilized microbubbles containing sulfur hexafluoride (SF6) gas with a mean diameter of 2.5 μm. The solution is reconstituted by the addition of 5 ml of sterile saline. A 20-gauge cannula was placed in the antecubital vein of the patients to administer the contrast agent. Contrast-enhanced sonography was performed after a 2.4-mL bolus injection of SonoVue manually via the intravenous cannula, followed by a 5-mL saline solution flush. The selected plane remained unchanged during the examination, and real-time imaging was recorded for up to 3 minutes for later analysis. B-mode pulse inversion harmonic imaging was used for contrast-enhanced sonography, and the machine parameters were adjusted so that the mechanical index was less than 0.1, the imaging depth was 3 or 4 cm, and the single-focus zone was at the bottom. In addition, no pressure was exerted on the transducer to avoid weakening the contrast signals. All static and dynamic images were stored in the iu22 system and then transferred to a computer for subsequent analysis. Image Analysis All sonograms were evaluated by 2 ultrasound physicians (H.L. and Q.Z.), who were blinded to the patients clinical data, mammographic findings, and histories. If disagreement occurred, the readers jointly reviewed the images, discussed, and came to a consensus. Enhancement features were assessed according to our previously published results 7,8 and the literature 9 as follows: (1) enhancement pattern (none, homogeneous, heterogeneous, or peripheral enhancement); (2) radial vessels surrounding the lesion (present or absent); (3) lesion size discrepancy between contrast-enhanced and conventional sonography; (4) enhancement degree compared to surrounding normal breast tissue at the peak time (hyperenhancement, isoenhancement, or hypoenhancement); (5) enhancement direction (centripetal, centrifugal, or diffuse); and (6) enhancement margin (well defined or poorly defined). The maximum lesion diameter on conventional sonography was measured manually. For lesions with circumscribed margins, the hypoechoic core of the tumor was measured. For lesions with microlobulations, angulations, or spiculations, measurements included microlobulations, angulations, or spiculations or corresponded to the distance from the tip of one spicule to the tip of an opposing spicule. The maximum lesion diameter on contrastenhanced sonography was measured manually from maximum-enhancement images. All enhancement components within and adjacent to the region of the tumor on conventional sonography were included; large enhancing blood vessels or other interfering regions (eg, movement artifacts), if present, were excluded on contrast-enhanced sonography. For lesions with no enhancement, the maximum diameter was considered the same as that measured on conventional sonography. On the basis of statistical analysis of random variations by Zeggelink et al, 10 a dif- 118 J Ultrasound Med 2015; 34:117 123

ference in lesion size greater than 2.5 mm was defined as a significant discrepancy between conventional and contrastenhanced sonography. Histopathologic Analysis All pathologic specimens were specifically evaluated by a single pathologist (X. Chang, MD) with more than 10 years of experience in breast pathology. The tumors were histologically classified according to the World Health Organization classification. Lymph node status was recorded as the number of positive lymph nodes of all the lymph nodes collected in patients who received radical mastectomy. Statistical Analysis The SPSS version 13.0 software package (IBM Corporation, Armonk, NY) was used for statistical analysis. Data were expressed as mean ± SD. A χ 2 test was performed to see whether the lesion enhancement features were different between the benign and malignant groups. A paired sample t test was performed to test the difference in lesion size measured between conventional and contrast-enhanced sonography. P <.05 was considered statistically significant. Diagnostic efficacy was quantified by using the standard indices of sensitivity, specificity, and accuracy, and pathologic results were used as a reference standard. Results Histologic Diagnosis Tissue specimens from 46 breast lesions were obtained from surgery (n = 18) or biopsy (n = 28). Histopathologic analysis revealed 11 breast carcinomas and 35 benign breast lesions. Malignant lesions included invasive ductal carcinoma (n = 7), ductal carcinoma in situ (DCIS; n = 3), and invasive lobular carcinoma (n = 1). The mean size was 0.87 ± 0.24 0.5 1.0 cm. On preoperative conventional sonography, 8 were classified as BI-RADS 4, and 3 were classified as BI- RADS 5. Benign lesions included fibroadenoma (n = 16), adenosis (n = 12), intraductal papilloma (n = 5), adenomyoepithelioma (n = 1), and chronic suppurative inflammation (n = 1). The mean size was 0.80 ± 0.15 0.4 1.0 cm. On preoperative conventional sonography, 20 were classified as BI-RADS 3, 14 were classified as BI-RADS 4, and 1 was classified as BI-RADS 5. Enhancement Features Different contrast enhancement features were observed in benign and malignant lesions (Table 1). In the 11 malignant lesions, heterogeneous enhancement was observed in 8 cases (72.7%), including 6 invasive ductal carcinoma, Table 1. Comparison of Contrast Enhancement Features Between Benign and Malignant Sub-1-cm Breast Lesions Feature Malignant (n = 11) Benign (n = 35) P Enhancement pattern No enhancement 1 (9.1) 2 (5.8).575 Homogeneous 2 (18.2) 8 (22.9) Heterogeneous 8 (72.7) 25 (71.4) Peripheral 0 0 Radial or penetrating vessels surrounding the lesion Present 6 (54.5) 1 (2.9) <.001 Absent 5 (45.5) 34 (97.1) Lesion size discrepancy between contrast-enhanced and conventional sonography Increased 8 (72.7) 7 (20.0).001 Unchanged 3 (27.3) 28 (80.0) Enhancement degree a Hyperenhanced 9 (90.0) 14 (42.4).008 Isoenhanced or hypoenhanced 1 (10.0) 19 (57.6) Enhancement direction a Centripetal 7 (70.0) 18 (54.5).312 Centrifugal or diffuse 3 (30.0) 15 (45.5) Enhancement margin a Well defined 4 (40.0) 7 (21.2).214 Poorly defined 6 (60.0) 26 (78.8) Data are presented as number (percent). a One DCIS and 2 adenosis cases were excluded because they showed no contrast enhancement. J Ultrasound Med 2015; 34:117 123 119

1 invasive lobular carcinoma, and 1 DCIS. Homogeneous enhancement was seen in 2 cases (18.2%), including 1 invasive ductal carcinoma and 1 DCIS. No enhancement occurred in 1 DCIS (9.1%). Radial vessels surrounding the lesion were present in 6 cases (54.5%), including 5 invasive ductal carcinoma, and 1 DCIS, and were absent in 5 (45.5%), including 2 invasive ductal carcinoma, 2 DCIS, and 1 invasive lobular carcinoma. There was a significant difference in lesion size between conventional and contrast-enhanced sonography for malignant lesions (P =.004, t test). The size of 8 of 11 breast cancers (72.7%) increased on contrastenhanced sonography; the mean increased size was 0.98 ± 0.44 0.5 1.8 cm. Their histologic types were as follows: 6 invasive ductal carcinoma, 1 invasive lobular carcinoma, and 1 DCIS. The size of 3 of 11 breast cancers (27.3%) remained unchanged on contrast-enhanced sonography; they included 2 DCIS and 1 invasive ductal carcinoma. For enhancement degree, hyperenhancement was indentified in 9 cases (90.0%), including 6 invasive ductal carcinoma, 2 DCIS, and 1 invasive lobular carcinoma. Isoenhancement was shown in 1 invasive ductal carcinoma (10.0%). For enhancement direction, 7 cases (70.0%) were centripetal, and 3 (30.0%) were diffuse. For enhancement margin, 4 cases (40.0%) were well defined, and 6 (60.0%) were poorly defined. In the 35 benign lesions, heterogeneous enhancement was observed in 25 cases (71.4%), including 9 adenosis, 9 fibroadenoma, 5 intraductal papilloma, 1 adenomyoepithelioma, and 1 chronic suppurative inflammation. Homogeneous enhancement was seen in 8 cases (22.9%), including 7 fibroadenoma and 1 adenosis. No enhancement occurred in 2 cases of adenosis (5.8%). Radial vessels surrounding the lesion were present in 1 (2.9%) adenoma and absent in 34 cases (97.1%). There was no significant difference in lesion size between conventional and contrastenhanced sonography for benign lesions (P =.060, t test). The size of 28 of 35 benign lesions (80.0%) remained unchanged, and 7 of 35 (20.0%) increased in size on contrastenhanced sonography. The mean increased size was 0.29 ± 0.12 0.2 0.5 cm. Their histologic types were as follows: 3 fibroadenoma, 2 adenosis, 1 intraductal papilloma, and 1 chronic suppurative inflammation. For enhancement degree, isoenhancement was shown in 19 cases (57.6%), including 8 fibroadenoma, 7 adenosis, 2 intraductal papilloma, 1 adenomyoepithelioma, and 1 chronic suppurative inflammation. Hyperenhancement was identified in 14 cases (42.4%), including 9 fibroadenoma, 3 intraductal papilloma, and 2 adenosis. For enhancement direction, 18 cases (54.5%) were centripetal, and 15 (45.5%) were diffuse. For enhancement margin, 7 cases (21.2%) were well defined, and 26 (78.8%) were poorly defined. The χ 2 test indicated that the detection rates for increased size, radial vessels surrounding the lesion, and hyperenhancement on contrast-enhanced sonography were significantly higher in the malignant than the benign group (P =.001; P <.001; P =.008, respectively). No significant difference was found in the enhancement pattern, direction, and margin between the groups (P =.575; P =.312; P =.214). Malignant lesions were characterized by increased size, radial vessels surrounding the lesion, and hyperenhancement on contrast-enhanced sonography (Figures 1 and 2). The sensitivity, specificity, and accuracy for increased size were 72.7%, 80.0%, and 78.2%, respectively. The sensitivity, specificity, and accuracy for radial vessels surrounding the lesion were 54.5%, 97.1%, and 87.0%. The sensitivity, specificity, and accuracy for hyperenhancement were 81.8%, 60.0%, and 65.0%. Discussion Sub-1-cm breast lesions are most often nonpalpable and are detected by mammography. Patients with such tumors have excellent (>90%) 20-year disease-specific survival. 11,12 An imaging modality that could accurately characterize sub-1-cm breast lesions would be both appropriate and cost-effective by increasing the detection rate for early breast cancers with a good prognosis and by reducing the number of surgical biopsies for benign disease. Contrastenhanced sonography is unique in that it provides vascular distributional and morphologic information, which is likely to be an important clue for discrimination between benign and malignant breast tumors. However, most previous studies 7,13 focused on larger (mean size, 2 cm) breast lesions rather than small ones. To our knowledge, the usefulness of contrast-enhanced sonography in sub-1-cm breast lesions has not been fully evaluated. Our results demonstrated that sub-1-cm breast lesions have their own characteristics on contrast-enhanced sonography. In previous studies, peripheral enhancement with centripetal filling was reported to be specific for malignant lesions, and it correlated with a peripheral zone of high microvessel density and a central zone of relatively lower micro vessel density in malignant lesions. 9 No enhancement was indicative of benign lesions 7,13,14 because the microvessel density in benign lesions was probably not high enough to show detectable enhancement on contrastenhanced sonography. 7,15 However, in our study of sub- 1-cm breast lesions, we did not find any significant differences in peripheral enhancement, no enhancement, or centripetal enhancement between malignant and benign lesions. This finding may have been due to the small 120 J Ultrasound Med 2015; 34:117 123

sample size or, more importantly, the specific histopathologic findings of small breast cancers. As a new tumor is a few millimeters in diameter (stage I), it can grow without inducing a substantial angiogenesis, when the surrounding capillary network is sufficient to nourish its growth. 16 As the tumor grows beyond a few millimeters (stage II), new capillary blood vessels are induced, followed by rapid tumor growth. As the tumor grows larger (stage III), central fibrosis and necrosis may develop if the vascular formation and nutritional supply are insufficient. Small breast cancer Figure 1. In a 45-year-old woman with a palpable mass in the outer upper quadrant of her right breast, pathologic examination proved invasive ductal carcinoma. A, Conventional sonogram showing an irregular hypoechoic mass (calipers) measuring 0.7 0.6 cm. B, On color flow imaging, peripheral blood vessels are present. C, Contrast-enhanced sonogram showing increased size (1.2 0.6 cm; arrows), radial vessels surrounding the lesion (arrowheads), and hyperenhancement. A Figure 2. In a 38-year-old woman with a nonpalpable mass in the outer quadrant of her right breast close to the nipple, pathologic examination proved fibroadenoma. A, Conventional sonogram showing a regular hypoechoic mass (calipers) measuring 0.9 0.6 cm with a clear margin. B, On color flow imaging, no blood vessels are present. C, Contrastenhanced sonogram showing the same size (0.9 0.6 cm; arrows) and isoenhancement. A B B C C J Ultrasound Med 2015; 34:117 123 121

can display no enhancement, probably because its micro - vessel density is not high enough to show detectable enhancement on contrast-enhanced sonography, and peripheral enhancement may be absent if it grows actively without central fibrosis or necrosis in stage II. Analysis of enhancement features revealed a significant difference between malignant and benign lesions for increased size on contrast-enhanced sonography. Our previously published study 8 showed that increased size on contrast-enhanced sonography was observed in about 62.8% of malignant lesions, and the size increased from 0.3 to 1.6 cm. The region corresponding to increased size could be invasive or intraductal carcinoma or adenosis with lobular hyperplasia. In this study, increased size on contrast-enhanced sonography was seen in 72.7% of malignant lesions (8 of 11), and the size increased from 0.5 to 1.8 cm. On retrospective review of the static and dynamic images of these lesions, we found all of them (8 of 8) actually showed a hyperechoic halo or a spicular margin, which was too subtle or inexplicit to be identified on conventional sonography. By comparison of conventional and contrast-enhanced sonograms, these features can be recognized. In our study, a significant discrepancy in lesion size greater than 2.5 mm between conventional and contrast-enhanced sonography was based on statistical analysis of random variations by Zeggelink et al, 10 who provided guidelines to determine which differences between measurements should be attributed to random measurement variations and which to actual underlying differences. In their study, the overall random variations for sonography in the assessment of lesion size was 2.5 mm for lesions no larger than 1.7 cm. In line with the findings by Du et al 13 and Zhang et al, 14 radial vessels surrounding the lesion and hyperenhancement were suggestive of malignancy. Radial vessels surrounding the lesion can be attributed to the fact that vasoactive substances, mainly vascular endothelial growth factor, secreted by breast cancer cells induce a large amount of angiogenesis, especially in the adjacent vessels of breast carcinoma. 17,18 As enhancement is determined by the tissue vascularity, hyperenhancement probably reflects increased angiogenesis in malignant lesions. 19 Inconsistent with previously reported results, 20 the enhancement margin (well defined or poorly defined) revealed no significant difference between malignant and benign lesions. Some authors described a perfusion defect as a specific sign of malignancy, which was observed in 66.7% of malignant lesions. 13 Our study did not include this feature for image analysis because the definition of a perfusion defect was not clear. If it is defined as any areas with no enhancement signals, a perfusion defect will be present with no, heterogeneous, and peripheral enhancement but not homogeneous enhancement. Contrast-enhanced sonography offers an interesting alternative to sonographically guided core biopsy for managing subcentimeter breast masses, possibly allowing shorter-interval follow-ups. The drawbacks to the general use of this technique would be personnel training, the added cost of buying and injecting contrast material, and also insurance reimbursement. Certain limitations existed in our study. First, the inclusion criteria of sub-1-cm lesions limited the number of enrolled patients. Therefore, the sample size was relatively small, and further studies in larger patient populations are needed to confirm our findings. Second, contrast-enhanced sonographic quantitative analysis was not applied. That process should be objective, but in contrast-enhanced sonography, there are some conflicting views on whether contrast medium kinetic analysis aids in the final diagnosis. 20,21 Third, classification of the enhancement patterns was performed visually by 2 radiologists in consensus, so we did not have any information on interobserver variability. In conclusion, sub-1-cm breast lesions have their own characteristics on contrast-enhanced sonography. No significant difference was found in peripheral or absence of enhancement between malignant and benign lesions. Sub-1-cm breast cancers are characterized by increased size, radial vessels surrounding the lesion, and hyper-enhancement on contrast-enhanced sonography. The best discriminator is increased size on contrast-enhanced sonography, with sensitivity, specificity, and accuracy of 72.7%, 80.0%, and 78.2%, respectively. Radial vessels surrounding the lesion had high specificity (97.1%) but poor sensitivity (54.5%). The sensitivity of hyperenhancement was high (81.8%), but the specificity was low (60.0%). Identification of these contrast-enhanced sonographic features is useful for differentiation of small breast lesions. References 1. Vacek PM, Geller BM, Weaver DL, Foster RS Jr. Increased mammography use and its impact on earlier breast cancer detection in Vermont, 1975 1999. Cancer 2002; 94:2160 2168. 2. Luke C, Nguyen AM, Priest K, Roder D. Female breast cancers are getting smaller, but socio-demographic differences remain. Aust NZ J Public Health 2004; 28:312 316. 3. Fracheboud J, Otto SJ, van Dijck JA, Broeders MJ, Verbeek AL, de Koning HJ; National Evaluation Team for Breast Cancer Screening (NETB). Decreased rates of advanced breast cancer due to mammography screening in The Netherlands. Br J Cancer 2004; 91:861 867. 122 J Ultrasound Med 2015; 34:117 123

4. Chen YY, Schnitt SJ. Prognostic factors for patients with breast cancers 1 cm and smaller. Breast Cancer Res Treat 1998; 51:209 225. 5. Cowan WK, Kelly P, Sawan A, et al. The pathological and biological nature of screen-detected breast carcinomas: a morphological and immunohistochemical study. J Pathol 1997; 182:29 35. 6. Gibbs P, Liney GP, Lowry M, Kneeshaw PJ, Turnbull LW. Differentiation of benign and malignant sub-1 cm breast lesions using dynamic contrast enhanced MRI. Breast 2004; 13:115 121. 7. Liu H, Jiang YX, Liu JB, Zhu QL, Sun Q. Evaluation of breast lesions with contrast-enhanced ultrasound using the microvascular imaging technique: initial observations. Breast 2008; 17:532 539. 8. Jiang YX, Liu H, Liu JB, Zhu QL, Sun Q, Chang XY. Breast tumor size assessment: comparison of conventional ultrasound and contrastenhanced ultrasound. Ultrasound Med Biol 2007; 33:1873 1881. 9. Du J, Li FH, Fang H, Xia JG, Zhu CX. Correlation of real-time gray scale contrast-enhanced ultrasonography with microvessel density and vascular endothelial growth factor expression for assessment of angiogenesis in breast lesions. J Ultrasound Med 2008; 27:821 831. 10. Zeggelink WF, Deurloo EE, Bartelink H, Rutgers EJ, Gilhuijs KG. Reproducibility of the assessment of tumor extent in the breast using multiple image modalities. Med Phys 2003; 30:2919 2926. 11. Tabár L, Chen HH, Duffy SW, et al. A novel method for prediction of long-term outcome of women with T1a, T1b, and 10 14 mm invasive breast cancers: a prospective study. Lancet 2000; 355:429 433. 12. Tabár L, Chen HH, Yen AMF, et al. Mammographic tumor features can predict long-term outcomes reliably in women with 1 14-mm invasive carcinoma. Cancer 2004; 101:1745 1759. 13. Du J, Wang L, Wan CF, et al. Differentiating benign from malignant solid breast lesions: combined utility of conventional ultrasound and contrastenhanced ultrasound in comparison with magnetic resonance imaging. Eur J Radiol 2012; 81:3890 3899. 14. Zhang J, Jiang YX, Dai Q, et al. Application of Real-time gray-scale contrast-enhanced ultrasound in the diagnosis of benign and malignant breast lesions [in Chinese]. Zhongguo Yi Xue Ke Xue Yuan Xue Bao2008; 30:49 53. 15. Foster FS, Burns PN, Simpson DH, Wilson SR, Christopher DA, Goertz DE. Ultrasound for the visualization and quantification of tumor microcirculation. Cancer Metastasis Rev 2000; 19:131 138. 16. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis: correlation in invasive breast carcinoma. N Engl J Med 1991; 324:1 8. 17. Du J, Li FH, Fang H, Xia JG, Zhu CX. Microvascular architecture of breast lesions: evaluation with contrast-enhanced ultrasonographic micro flow imaging. J Ultrasound Med 2008; 27:833 842. 18. Drevs J. VEGF and angiogenesis: implications for breast cancer therapy. Eur J Cancer 2008; 6:7 13. 19. Caproni N, Marchisio F, Pecchi A, et al. Contrast-enhanced ultrasound in the characterisation of breast masses: utility of quantitative analysis in comparison with MRI. Eur Radiol 2010; 20:1384 1395. 20. Wan CF, Du J, Fang H, Li F, Wang L. Evaluation of breast lesions by contrast enhanced ultrasound: qualitative and quantitative analysis. Eur J Radiol 2012; 81:444 450. 21. Du WH, Liu LJ, Wang H, Wei Y, Li T. The clinical significance of real-time contrast-enhanced ultrasonography in the differential diagnosis of breast tumor. Cell Biochem Biophys 2012; 63:117 120. J Ultrasound Med 2015; 34:117 123 123