Gladys G. Lo, MD

Breast MRI has been available for over a decade. It is only now, however, that the examination is becoming recognized as an indispensable adjunct to mammography and ultrasound. Several key factors contribute to this acceptance of clinical breast MRI.

The breast MRI protocol is approaching standardization, high-resolution images are acquired routinely with 1.5T and 3T scanners, and MRI breast biopsy devices are commercially available.

Most breast MRI examinations can be completed within 30 to 40 minutes. Centers typically acquire precontrast T1- and T2-weighted images, perform a dynamic contrast examination, and then acquire delayed enhancement T1-weighted images. Some centers still image one breast per examination, though newer hardware and software allow both breasts to be examined at the same session.

Our hospital began 3T breast MRI in November 2004. This immediately allowed us to image both breasts simultaneously during the dynamic sequence. The 1.5T scanner we had previously been using could perform dynamic imaging on only one breast at a time, though this constraint was later removed following an equipment upgrade. More important, the 3T system provided much higher spatial resolution than our 1.5T system, owing to the inherent increase in signal-to-noise ratio and the increased imaging matrix.

One well-known disadvantage associated with 3T systems is the increased deposition of radiofrequency energy or higher specific absorption rate (SAR).1 Other disadvantages include the prolonged T1 relaxation time and higher susceptibility artifacts. The use of parallel imaging techniques can counter these shortcomings. Because parallel imaging shortens the acquisition time, RF deposition and SAR effects are reduced. Combining higher bandwidth with parallel imaging also effectively lowers the TE, resulting in decreased susceptibility. The synergistic effect of parallel imaging with 3T then produces images with high spatial and temporal resolution.

Our protocol for 3T breast MRI takes approximately 30 minutes:

dynamic 3D axial, fat-saturated T1-weighted (x 5)

delayed enhancement sagittal, fat-saturated T1-weighted

delayed enhancement axial, fat-saturated T1-weighted

T1-weighted images are used to demonstrate anatomy, while T2-weighted images help characterize lesions. High signal observed on T2-weighted MRI can indicate a cyst, a myxoid fibroadenoma, or a mucinous tumor. Carcinoma may not be bright and may even be of lower signal than surrounding breast tissue.2

Diffusion-weighted MRI is feasible at 3T and takes only 90 seconds without contrast. Diffusion-weighted imaging (DWI) probes the Brownian movement of water molecules. Recent studies have suggested that DWI may be useful in differentiating malignant from benign breast tumors.3 We have added DWI to our protocol and are in the process of collecting the relevant data.

Both breasts are imaged five times, for one minute each, in the dynamic sequence. We prefer axial imaging over coronal because it is easier to compare the results with those from mammography. Another consideration is that breast ducts are usually oriented axially and will appear truncated on the coronal view. It is not possible at present to do simultaneous bilateral sagittal scans during the dynamic sequence.

The advent of high-resolution MR scanners means that lesions' morphology (shape and margin) and enhancement characteristics (internal enhancement, pattern of enhancement, and kinetic curves) can be analyzed carefully.4 Precontrast scans can show some of the above characteristics, but the dynamic contrast examination is crucial for lesion interpretation. Breast tumors show up well on this examination owing to angiogenesis.

The American College of Radiology's Breast Imaging Reporting and Database system (BI-RADS) provides a valuable standard for terminology used to report MRI findings (see table).5-7 For example, an irregularly shaped mass with spiculation and

heterogeneous or rim enhancement is considered highly suspicious for malignancy. Similarly, a nonmass enhancement that is asymmetrical with a segmental or regional pattern is a strong indicator of ductal carcinoma in situ.8 Features such as smooth borders or nonenhancing septa, which can be seen in a certain percentage of fibroadenomas, favor benignancy (Figure 1). Enhancing foci, defined as lesions measuring < 5 mm, are usually of no clinical significance.9

Signal intensity/time graphs are obtained for each enhancing lesion at the site of maximal enhancement. Three types of curves can be distinguished (Figure 2).10 Type I curves show continuous enhancement and are generally associated with benign lesions. Type II curves, which exhibit a rapid uptake of contrast then plateau, can indicate benign or malignant lesions. Type III curves demonstrate a rapid uptake of contrast with wash-out and are most often related to malignant lesions.

MRI is now considered to be the most sensitive method of evaluating the extent of breast cancer and to be superior to both mammography and ultrasound.11 Breast MRI has very high sensitivity: greater than or equal to 90% for breast cancer and near 100% for invasive breast carcinoma.12,13 Studies of breast MRI in high-risk patients, such as women with BRCA1 and BRCA2 genes, show a specificity of 93% to 99%.14 One group has reported the sensitivity of MRI for DCIS to be 89%.15 False positives do occur, though, due to rapidly enhancing fibroadenomas, fibrocystic changes, or papillomas. Rare false negatives can also be caused by nonenhancing DCIS or infiltrating lobular carcinoma.16 Ultrasound can be performed as a second look, but this modality misses up to 77% of MR-detected lesions.17

MRI-guided biopsy is consequently an essential part of any program offering a comprehensive breast MRI service. User-friendly, MR-compatible biopsy devices are now commercially available. Practitioners can choose from fine-needle aspiration, core biopsy, and vacuum-assisted core biopsy devices. MR-guided needle localization is also possible.

Indications for breast MRI fall into the following categories: screening, diagnosis, staging of biopsy-proven carcinoma, and treatment monitoring.16 Breast MRI has been used in the screening setting to evaluate high-risk patients. These include individuals with the BRCA1 or BRCA2 gene, women with a family history of breast cancer, women with very dense breast tissue, or those with silicone implants that may have obscured pathology.

Breast MRI should be used judiciously in cases where the results of mammography, ultrasound, and/or physical examination are equivocal; e.g., an abnormality that is seen on only one view of the mammogram, a patient with bloody nipple discharge and a negative ductogram, a questionable palpable abnormality with no correlation on mammography or ultrasound, and a patient with axillary adenopathy. Figure 3 shows a 59-year-old woman who had a lumpectomy for breast carcinoma seven years ago. There is a questionable mass at the scar. The mammogram showed architectural distortion indistinguishable from the scar. Ultrasound showed shadowing, but MRI clearly demonstrated a recurrent enhancing tumor adjacent to the scar.

Locoregional staging of tumor extent is vital, given that surgery is generally the treatment of choice for breast cancer. This indication accounts for more than 50% of our referrals. The accuracy of this staging will determine the success of breast conservation treatment. Any remaining tumor at the time of surgery will increase the chance of recurrence.11 Multifocal tumors (more than one tumor in each quadrant) and multicentric tumors (tumors in more than one quadrant) occur in 6% to 34% of breast cancer cases.18 Contralateral tumors occur 3.8% to 5.4% of the time.16 Invasive lobular carcinoma is difficult to detect with conventional imaging. Breast MRI is reportedly the most effective way of determining the extent of this cancer. Associated findings of lymphadenopathy, chest wall involvement, and skin retraction can also be shown on MRI (Figure 4).

Axillary nodes are divided into three levels according to their position in relation to the pectoralis minor muscle. Those below the pectoralis minor are termed level one nodes. Those behind it are level two nodes, and those above it level three nodes. Level one nodes are included in the dynamic scan field-of-view. Normal nodes are uniform in shape with a fatty hilum. Abnormal nodes are rounder and enlarged, with tumor replacing the fatty hilum and possibly irregular borders. There is considerable overlap, however, between benign and malignant lymph nodes based on their morphologic features.

Dextran-coated ultrasmall superparamagnetic iron oxide (USPIO) particles have been used to stage axillary nodes.19 Researchers showed the sensitivity of this technique to be 100%, specificity 98%, and accuracy 98%. In normal lymph nodes, these USPIO particles are ingested by macrophages after intravenous injection. They are not ingested by tumor-laden nodes. Consequently, lymph nodes that do not take up USPIO particles harbor metastatic disease.

MRI cannot yet replace sentinel-node biopsy, which searches for macrometastases and micrometastases in the first node into which the tumor drains. Internal mammary adenopathy may be suspected if this node is larger than 5 mm.20 This node is found adjacent to the internal mammary artery and vein. Its involvement indicates a worse prognosis. Neither dissection nor irradiation of these nodes improves survival.

Another strength of MRI is its ability to show any tumor involvement in the chest wall, which would indicate involvement in the underlying intercostal muscles. Involvement of the pectoralis major muscle is not considered to be chest wall involvement. Skin involvement is shown by thickening and retraction of the skin.

Breast MRI can additionally be used to monitor treatment response. Many patients now undergo neoadjuvant therapy when tumors are too large or too extensive to survey.16 Staging is used prior, during, and after neoadjuvant therapy to look for response. Because tumor enhancement at MRI is due to angiogenesis and many drugs have antiangiogenic properties, a responding tumor should decrease in size and enhance less. Practitioners should expect to see a smaller tumor on images and a change in the signal intensity/time graph from a type II/III curve to a type I curve (Figure 5).

In conclusion, 3T breast MRI is now a clinical reality. Very high spatial and temporal resolution can be obtained by partnering 3T with parallel imaging techniques. The high spatial resolution enhances the morphologic features of lesions and increases diagnostic confidence. High temporal resolution means the entire breast can be imaged in a minute, the enhancing lesion is visualized well above background parenchyma, and reliable kinetic graphs can be obtained.1 The availability of user-friendly biopsy devices will help clarify the status of lesions that are seen only on MRI. Further improvements in sensitivity and specificity can be expected as radiologists gain more experience working with 3T.

DR. LO is radiologist-in-charge in the department of diagnostic and interventional radiology at the Hong Kong Sanatorium and Hospital.

Kuhl CK, Jost P, Morakkabati N, et al. Contrast-enhanced MR imaging of the breast at 3.0 and 1.5 T in the same patients: initial experience. Radiology 2006;239(3):666-676.

Malich A, Fischer DR, Wurdinger S, et al. Potential MRI interpretation model: differentiation of benign from malignant breast masses. AJR 2005;185(4):964-970.

Guo Y, Cai YQ, Cai ZL, et al. Differentiation of clinically benign and malignant breast lesions using diffusion-weighted imaging. J Magn Reson Imaging 2002;16:172-178.

Kuhl CK. Dynamic breast magnetic resonance imaging. In: Morris E, Liberman L, eds. Breast MRI: diagnosis and intervention. New York: Springer, 2005:79-139.

Technical report of the international working group on breast MRI. J Magn Reson Imaging 1999;10(6):978-1015.

American College of Radiology. Breast imaging reporting and data system (BI-RADS), 3rd ed. Reston, VA: American College of Radiology, 1998.

Morris EA. Breast magnetic resonance imaging lexicon. In: Morris E, Liberman L, eds. Breast MRI: diagnosis and intervention. New York: Springer, 2005, p. 51-78.

Nunes LW. Architectural-based interpretations of breast MR imaging. Magn Reson Imaging Clin North Am 2001;9(2):303-320.

Liberman L, Mason G, Morris EA, et al, Does size matter? Positive predictive value of MRI-detected breast lesions as a function of lesion size, AJR 2006;186(2):426-430.

Kuhl CK, Mielcareck P, Klaschik S, et al. Are signal time course data useful for differential diagnosis of enhancing lesions in dynamic breast MR imaging? Radiology 1999;211(1):101-110.

Dershaw D. Magnetic resonance imaging as a clinical tool. In: Morris E, Liberman L, eds. Breast MRI: diagnosis and intervention. New York: Springer, 2005:256-265.

Orel SG. MR imaging of the breast. Magn Reson Imaging Clin North Am 2001;9(2):273-288.

Lee CH. Problem solving MR imaging of the breast. Radiologic Clin North Am 2004;42(5):919-934.

Warner E, Plewes DB, Hill KA, et al. Surveillance of BRCA1 and BRAC2 mutation carriers with magnetic resonance imaging, ultrasound, mammography and clinical breast examination. JAMA 2004;292(11):1317-1325.

Menell JH, Morris EA, Dershaw DD, et al. Determination of presence and extent pure ductal carcinoma in situ by mammography and MR imaging [abstract]. AJR 2003;180(suppl):52-53.

Daniel B, Ikeda DM. Magnetic resonance imaging of breast cancer and MRI guided breast biopsy. In: Ikeda DM, ed. Breast imaging: the requisites. Philadelphia: Elsevier Mosby, 2004:189-224.

LaTrenta LR, Menell JH, Morris EA, et al. Breast lesions detected with MR imaging: utility and histopathologic importance of identification with US. Radiology 2003;227(3):856-861.

Liberman L, Morris EA, Dershaw DD, et al. MR imaging of the ipsilateral breast in women with percutaneously proven breast cancer. AJR 2003;180(4):901-910.

Memarsadeghi M, Riedl CC, et al. Axillary lymph node metastases in patients with breast carcinomas: assessment with nonenhanced versus uspio-enhanced MR imaging. Radiology 2006;241(2):367-377.

Kinoshita T, Odagiri, K, Kazuo A, et al. Evaluation of small internal mammary lymph node metastases in breast cancer by MRI. Radiat Med 1999;17(3):189-193.

Gladys G. Lo, MD

Articles

Three-T breast MR imaging moves into clinical arena