Breast MRI Indications, Interpretation, and Interventions

Screening

Breast MRI is an important adjunct screening tool for women at high risk of developing breast cancer. The American Cancer Society (ACS), American College of Radiology (ACR), and National Comprehensive Cancer Network (NCCN) currently recommend annual breast MRI in addition to annual mammogram in women with ≥20% lifetime risk. Breast cancer risk is determined by personal and family history as well as genetic predisposition ( Box 8.2 ). Although a growing body of evidence supports screening breast MRI of intermediate-risk women, such as those with dense breasts or history of atypia on prior core biopsy, the evidence remains insufficient at this time for ACS to recommend its routine use in this setting. Although MRI has high sensitivity for detecting breast cancer, MRI is not recommended for screening in the general average-risk population due to higher costs and higher rates of benign biopsies.

Diagnostic Imaging in Women With Known Breast Cancer

NCCN guidelines recommend that breast MRI be considered in patients with newly diagnosed breast cancer to establish the extent of ipsilateral disease and screen the contralateral breast. Meta-analyses show that extent of disease MRI detects additional breast cancer approximately 15% of the time in the ipsilateral breast and 4% of the time in the contralateral breast.

MRI’s intrinsically high tissue contrast and ability to assess dynamic enhancement patterns allow for better detection of multifocal (more than one area of malignancy within a quadrant) and multicentric (more than one area of malignancy within multiple quadrants) disease in the ipsilateral breast ( Fig. 8.1 ). MRI assesses direct extension of disease to the overlying skin, nipple areolar complex, pectoralis muscle, and chest wall. Because MRI provides side-by-side imaging that allows for direct comparison of morphology with the contralateral side and includes all three axillae levels and the internal mammary chain, it can be helpful in assessing regional lymph node involvement. Specifically, it can identify suspicious level 1 axillary lymph nodes, for which preoperative ultrasound-guided percutaneous biopsy may be performed, as well as lymph nodes not typically visualized on mammogram and ultrasound, including level 2 axillary, level 3 axillary, and internal mammary. Finally, MRI detects clinically and mammographically occult synchronous cancer in the contralateral breast in about 4% of women with newly diagnosed breast cancer.

Rarely (<1%), breast cancer presents as metastatic axillary adenocarcinoma of unknown primary without a clinically or mammographically evident finding within the breast. In such cases, breast MRI has been shown to identify the otherwise occult breast primary malignancy in 61% of patients, which can significantly impact management approach.

Evaluation of Response to Neoadjuvant Therapy

Breast MRI is used to assess response to neoadjuvant (presurgical) therapy for surgical and adjuvant (postsurgical) therapy planning purposes. Neoadjuvant therapy typically refers to chemotherapy, although hormonal therapy may also be used. Neoadjuvant therapy has several advantages over traditional postsurgical therapy approaches without adverse effect on overall and disease-free survival: (1) It can make breast conservation more feasible in women with locally advanced disease; (2) it can allow for in vivo assessment of the effectiveness of the selected treatment regimen; and (3) it can allow for treatment to begin while the patient carefully considers optimal surgical and reconstruction options. MRI is an excellent imaging tool to determine neoadjuvant therapy effectiveness. Decrease in lesion size and volume and less suspicious enhancement kinetics on follow-up MRIs at various time points during chemotherapy treatment are associated with neoadjuvant treatment outcomes, such as pathologic complete response, allowing for assessment of a primary lesion’s response prior to surgery ( Figs. 8.2 and 8.3 ). For patients with poor or no response to neoadjuvant therapy, a change in treatment regimen may be indicated. Studies show that MRI is a stronger predictor of pathologic response to neoadjuvant therapy than clinical assessment and that it is superior to mammography and ultrasound in assessing for residual disease after completion of neoadjuvant therapy. However, pathologic response to therapy at the time of surgery is still considered the gold standard. Even in cases with no residual suspicious enhancement in the breast after neoadjuvant therapy, a BI-RADS category 6 assessment should still be assigned until surgical excision is completed.

Problem Solving: Evaluation of Suspicious Nipple Discharge

The use of MRI for problem solving of equivocal mammographic or sonographic imaging findings is not supported by research, except for the clinical scenario of suspicious nipple discharge. Suspicious nipple discharge is bloody or clear in color, unilateral, spontaneous, and usually arising from a single duct. For patients over the age of 30, initial imaging workup involves mammogram and ultrasound. When both are negative, either contrast-enhanced MRI or ductography may be considered. Although ductography has historically been the examination of choice, it is invasive and requires nipple discharge to be present on the day of the procedure to allow cannulation of the appropriate duct. Ductography may only show “tip of the iceberg” as the initial obstruction prevents retrograde contrast from passing, thus missing significant upstream disease. Furthermore, ductography does not provide a means for sampling of an abnormality and can only serve as guidance for the surgeon’s excision. MRI is noninvasive and has been shown in multiple studies to have higher sensitivity and specificity compared with ductography. MRI also allows for a means to biopsy the abnormality prior to surgery. This is important, since a malignant diagnosis on core biopsy prior to surgery can allow for a single operation to be performed, particularly in cases of invasive cancer where axillary lymph node sampling is required.

Evaluation of Silicone Implant Integrity

Breast implants can be single- or double-lumen, containing saline, silicone, or a combination of both. Complications, specifically implant rupture, are common. Saline implant rupture is a clinical diagnosis and presents as rapid loss of breast volume with surrounding tissue absorbing the saline solution over several days. Silicone implant rupture is more difficult to detect on clinical examination. Mammography alone has low sensitivity for detecting silicone implant rupture, especially intracapsular implant rupture. MRI is considered the gold standard for diagnosis of intracapsular silicone implant rupture with high sensitivity and specificity. Implant evaluation is discussed in further detail in the “Augmented and Reconstructed Breast” chapter ( Chapter 18 ).

Technical Considerations

Although breast MRI acquisition protocols vary between institutions, there is a set of generally agreed-upon minimum standards and technical considerations. Examinations are performed with patients in the prone position, and breasts are positioned within a dedicated breast surface coil. The use of dedicated breast surface coils helps maximize signal, and the greater number of coil elements allows for higher spatial resolution and reduction of scan time via parallel imaging. Both breasts are imaged, which allows for assessment of symmetry, establishment of bilateral background parenchymal enhancement, and determination of bilateral benign findings. The field of view should include bilateral axillae and chest wall.

Because adequate signal-to-noise ratio (SNR) is required to achieve high imaging quality, and SNR is directly proportional to the strength of the main magnetic field B0, breast MRI is typically performed with a 1.5 T or higher strength magnet. Higher magnetic field strength also allows for better magnetic field homogeneity and more uniform fat suppression.

Although there are emerging noncontrast techniques such as diffusion-weighted imaging (DWI), clinical breast MRI currently requires administration of gadolinium-based contrast for detection of breast cancer, injected intravenously using a power injector at a dose of 0.1 mmol/kg followed by a 20 mL saline flush at a rate of 2 mL/s.

High temporal and spatial resolution is crucial for detection and characterization of lesions on breast MRI. High spatial resolution allows for assessment of small morphologic features such as spiculations and dark internal septations. The ACR Breast Magnetic Resonance Imaging Accreditation Program (BMRAP) requires all T1-weighted images to be acquired with slice thickness of 3 mm or less and a maximum in-plane pixel dimension of 1 mm ³ . High temporal resolution during postcontrast sequences provides information on enhancement kinetics with invasive cancers typically demonstrating early-phase fast enhancement followed by delayed-phase washout. Because achieving high spatial resolution requires longer imaging time, spatial and temporal resolutions must be balanced when designing the imaging protocol. In general, spatial resolution is more important because it allows for improved depiction of lesion morphology, which is the most important feature for determining Breast Imaging Reporting and Data System (BI-RADS) assessment and the probability of malignancy (see interpretation algorithm below).

Fat suppression is essential in order to avoid the intrinsic high signal on T1-weighted images from obscuring enhancing abnormalities. Fat suppression can be achieved with spectral fat suppression during image acquisition or through subtraction techniques during postprocessing. Newer techniques such as the multipoint Dixon technique, which relies on chemical shift phenomena, can provide even more homogeneous fat suppression but are less commonly used due to time penalties and other artifacts (e.g., fat-water swapping).