- Historically, imaging of the patient with newly diagnosed breast cancer has played 3 major roles: Mapping the extent of disease for surgical planning, screening the contralateral breast, and assessing response to treatment. Suspicious axillary nodes can also be identified and biopsied prior to treatment. As treatment has evolved, the impact of imaging on each of these roles has been more thoroughly examined and questioned. Initial core biopsy pathology now routinely includes provisional tumor grade, estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor 2 (HER2) receptor status; tumor genetic testing is often performed for T1-3, N0-1 ER(+) disease to help guide treatment decisions. Such receptor information and grade are now included in tumor staging as treatment and prognosis vary by tumor subtype.
- Surgical treatment has been the mainstay of breast cancer therapy for decades. The decision to offer breast-conserving surgery (BCS) is based on the premise that overall survival is equivalent to patients who receive more invasive surgery (mastectomy). Expectations of BCS include adequate resection of the known malignancy with clear margins and a good cosmetic result. Standard breast-conserving therapy (BCT) includes adjuvant radiation therapy (XRT) following BCS. XRT is increasingly used with benefit after mastectomy &/or for regional nodal irradiation (RNI) in appropriate patients. Positive margins increase the risk of recurrence. Increasingly, neoadjuvant (primary) chemotherapy (NAC) is being used to shrink the tumor prior to surgery, particularly with HER2(+) or triple-negative breast cancer (TNBC). NAC can produce a complete pathologic response: Trials are now underway to forego surgery if there is no evidence of residual tumor on imaging and biopsy post NAC.
Locoregional Therapy Planning
- Locoregional therapy selection is determined and measured by outcomes that include reexcision rates, conversion to mastectomy, local recurrence, and contralateral breast cancer. Mammography with magnification views of suspicious Ca⁺⁺ is routinely performed to map the extent of malignancy on mammography. HER2(+) tumors are particularly likely to have an extensive intraductal component with Ca⁺⁺. An average of 48% of breasts thought to have unifocal tumor on mammography and clinically will have additional disease at detailed pathology. US and MR will depict some of this, but the benefit to survival outcomes for patients is unclear. For invasive tumors ≤ 3 cm, US is very accurate at depicting tumor size. For larger tumors, MR is more accurate.
- While early studies of pretreatment MR showed equivocal results, there is increasing evidence of reduced reoperation rates and reduced recurrence in women who had preoperative MR. Early studies show that digital breast tomosynthesis (DBT) improves detection of multifocal or multicentric disease compared to 2D full-field digital mammography (FFDM), but MR still improves disease depiction even after DBT. Contrast-enhanced mammography (CEM) is also being evaluated for this purpose and depicts invasive cancers with similar sensitivity to MR and probably greater specificity (fewer false positives), though DCIS is less well seen.
- In the AJCC 8th edition, staging includes "clinically suspicious" nodes and that includes those suspicious on imaging as well as those that are palpable. US should include evaluation of the ipsilateral axillary nodal basins, at least to include documentation of the number of suspicious nodes. US should start with the inferior level I nodes, which are likely to be the sentinel nodes, and proceed through mid and upper level I, then medially to level II and level III as long as there continue to be suspicious nodes.
- Additional considerations in local surgical planning include the suspicion of pectoralis invasion when MR shows abnormal enhancement of the pectoralis muscle contiguous with current breast cancer. This abnormal enhancement may demonstrate an infiltrative pattern without change in muscle contour or focal mass enhancement within the muscle. Suspicion of pectoralis muscle invasion should be reported for surgical planning but does not constitute chest wall invasion. The diagnosis of chest wall invasion requires direct involvement of the serratus anterior or intercostal muscles by breast cancer and may also involve ribs. Treatment of chest wall involvement comprises NAC, XRT, ± chest wall resection if grossly involved. Modified radical mastectomy is performed if there is full-thickness pectoralis muscle involvement.
- Second opinion review of outside imaging prior to treatment has been validated not only for breast but for other cancers, musculoskeletal imaging, neuroimaging, and body imaging. Such review can identify additional suspected disease or discount findings originally considered suspicious. On average, such review changes management in nearly 20% of cases with an average 40% discrepant rate for breast imaging. Pathology review is also recommended for any diagnosis of invasive cancer, DCIS, or atypia with resulting reduced under- and overtreatment; discrepant rates average 41-52% for atypia, 9-18% for DCIS, and 4-5% for invasive carcinoma.
Multifocal and Multicentric Breast Cancer, EIC
- Multifocal breast cancer is defined as ≥ 2 separate foci in the same breast, separated by ≥ 0.5 cm and ≤ 4 cm, in the same duct system. Multicentric breast cancer is defined as ≥ 2 separate foci in the same breast where the greatest distance between 2 foci is variably > 4-5 cm, and typically involve different breast quadrants, usually in different duct systems. Multiple suspicious masses ± Ca⁺⁺ are a feature on MMG. The MR sensitivity for additional malignant foci ranges from 53-60%. An average of 16% of patients planning BCS will show additional malignant foci in the ipsilateral breast on MR with an average of 30% of women with ILC having additional foci on MR. Luminal-B and HER2(+) subtypes are 2-3x more likely to have multifocal/multicentric disease than other subtypes. The greatest yield is in the same quadrant with 66% PPV for suspicious findings on MR. Multifocal/multicentric breast cancer confers a 2.4x ↑ risk of nodal metastases vs. unifocal tumor after correcting for size and grade. 10-year disease-specific survival is 2x lower in multifocal &/or multicentric or diffuse tumor vs. unifocal tumor. The finding of multifocal &/or multicentric breast cancer is a contraindication to partial breast irradiation.
- An extensive intraductal component (EIC) is defined as invasive ductal carcinoma (IDC) with associated DCIS that represents at least 25% of the main tumor mass with satellite foci of DCIS beyond the main tumor mass. Invasive cancers with an EIC are more likely to demonstrate Ca⁺⁺ associated with a mass and clumped linear or segmental nonmass enhancement in the DCIS component on MR. Large (> 2.5-cm²) areas of malignant-appearing Ca⁺⁺ on mammography require magnification views to depict extent. Preoperative bracketing with needle localization may be necessary to facilitate negative surgical margins, and mammograms may be performed to assure complete removal at BCS prior to XRT if Ca⁺⁺ are extensive. US targeted to any associated mass or asymmetry is helpful to guide biopsy and confirm an invasive component. These EIC cancers demonstrate increased frequency of positive margins at lumpectomy and high rates of residual disease at reexcision (38-74% vs. 11-42%) when compared to tumors without EIC. Invasive cancers with EIC are a contraindication to partial breast irradiation. HER2(+) tumors are 2-3x more likely to have an EIC than other subtypes.
- Skin invasion is most commonly related to direct tumor extension involving the overlying skin; with this alone, T stage is still defined by tumor size. It is crucial to distinguish this from skin ulceration due to tumor, satellite malignant skin nodules, or overlying edema ± peau d'orange = T4/stage IIIB disease. Skin invasion is best depicted on MR (abnormal enhancement extends to include skin, often directly over tumor) or US (mass directly extends to involve the skin/subcutaneous interface). Skin punch biopsy confirms the diagnosis if no parenchymal mass is evident. Whole-body staging (e.g., PET/CT) to exclude distant metastasis is necessary. Treatment includes mastectomy ± chemotherapy (neoadjuvant or adjuvant) or BCT with good response to NAC to include resection of the involved skin.
- Axillary nodal metastatic disease remains the strongest prognostic factor. Breast cancer-specific survival (BCSS) and recurrence rates are directly related to the extent of axillary involvement.
- Sentinel lymph node (SLN) biopsy is performed for prognostic information, and axillary lymph node dissection (ALND) is performed for local control and complete staging. The American College of Surgeons Oncology Group (ACOSOG) Z0011 trial showed that among breast cancer patients with limited (1-2 positive) metastatic SLN treated with BCT and systemic therapy, the use of SLN biopsy alone had equivalent locoregional recurrence (LRR) and 5-year disease-free survival (DFS) to ALND. SLN biopsy is less accurate after NAC with a false-negative rate of over 12%. In patients having NAC, US-guided FNA or core biopsy is first performed of a suspicious axillary node, and a clip should be placed. ACOSOG Z1071 trial evaluated SLN biopsy in patients with clinically node-positive (cN1) disease receiving NAC and showed that patients with 3 or fewer suspected metastatic nodes on initial evaluation could be adequately treated with targeted axillary dissection post NAC to include the clipped, known metastatic node. RNI is beneficial with even one metastatic axillary node.
- Internal mammary adenopathy (IMN) is best diagnosed on US as round, hypoechoic mass(es) in the intercostal space adjacent to the internal mammary vessels, on CT/MR as small, round, enhancing mass(es) adjacent to internal mammary vessels, or on PET/CT as soft tissue masses in the internal mammary chain with increased FDG uptake. The demonstration of IMN on lymphoscintigraphy ranges from < 2% to 38% depending on the technique and radioisotope used. IM SLN drainage has been demonstrated in younger patients, is less likely with UOQ tumors, more likely with smaller/medial tumors, and associated with significantly worse distant disease-free survival (DDFS) (HR 1.6; 95% CI, 1.03 to 2.6; P = .04) but not LRR or overall survival. Most centers do not perform IMN biopsies due to concerns about morbidity and lack of established survival benefit; IMN suspicious on imaging (US, MR, CT, PET/CT) are generally considered positive. Ipsilateral IMN (+) without axillary lymph node (AXLN) metastasis is stage N2b, and metastasis to both ipsilateral IMN and AXLN is stage N3b disease.
Contralateral Breast Cancer
- Contralateral breast cancer (CBC) can be synchronous [(SCBC), occurs within 1 year of incident tumor diagnosis] or metachronous [(MCBC), 2nd primary cancer diagnosed after incident tumor treatment)]. The incidence of SCBC after mammography and clinical breast examination (CBE) is 2%; incidence by screening US after negative mammography and CBE is 3% and by MR is 5.5%. ER(+) tumors have the lowest risk for CBC with 1.6x risk if the tumor is ER(-), 2.0x risk if HER2(+), and 1.4x risk if triple-negative. CBC in women on tamoxifen therapy for ≥ 5 years are 4.4x more likely ER(-). SCBC are often more advanced at presentation, more frequently invasive lobular carcinoma (ILC), and more frequently demonstrate multicentric disease. The cumulative risk for MCBC is 0.5-1.0% per year. Risk differs based on age at the time of diagnosis.
- In a meta-analysis of 3,252 women with unilateral breast cancer, 131 contralateral malignancies were detected by MR alone, 35% of which were DCIS.
Inflammatory Breast Cancer
- Inflammatory breast cancer (IBC) is locally advanced breast cancer that presents with rapid onset (within 6 months) of inflammatory skin change and breast swelling. Clinical exam reveals diffuse erythema, warmth, pain, induration, and peau d'orange skin changes that involve ≥ 1/3 of breast. Due to the propensity for early systemic metastases, induction chemotherapy (anthracycline, taxanes) is the treatment of choice. Mastectomy and axillary dissection followed by chest wall XRT after mastectomy are standard. MR is useful for the assessment of disease extent prior to, during, and after NAC.
- Imaging is important for staging at diagnosis. Characteristic mammography findings are skin and trabecular thickening, global or focal asymmetry, irregular mass(es), and axillary adenopathy in up to 50% of patients at diagnosis. Skin thickening and edema associated with diffuse hyperechogenicity and dilated lymphatics are typical US findings. US identifies a mammographically occult mass in up to 90% of patients; masses are typically irregular, hypoechoic, and demonstrate posterior shadowing. The diagnosis can be established by skin punch biopsy or US-guided core biopsy of the breast, which also allows for biomarker evaluation. IBC is usually poorly differentiated IDC, stage T4d/IIIB, with dermal lymphatic tumor emboli and lymphovascular invasion. Staging MR reveals streaky hyperintensities on T2W/STIR MR (diffuse edema), skin enhancement, and irregular contiguous or coalescent masses, occasionally with a reticular dendritic pattern. Systemic staging includes CT of chest/abdomen and bone scan; PET/CT is optional.
- The goal of NAC is to provide the option of BCS to patients who would have required mastectomy if surgery had been performed at initial diagnosis. MR is relatively accurate at predicting pathologic complete response (pCR), though pCR is not necessary for the success of BCT. MR has the advantage of determining whether tumor shrinkage is concentric or has left scattered nests of tumor cells and may be helpful in determining the suitability of patients for BCS. Multivariate analysis has shown that receptor subtypes are the only significant predictors of pCR. Triple-receptor (+) and HER2(+) breast cancers respond best to NAC and are the subtypes best assessed by MR for response to NAC. A meta-analysis of 25 studies, including a total of 1,212 patients, reported a specificity of 91% (95%CI 91 to 92) and sensitivity of 63% (95%CI 56 to 70) for MR prediction of pCR. The ACRIN 6657 trial comparing MR (tumor size by longest diameter, volumetric measurement, peak signal enhancement ratio) and clinical size showed advantage in early MR volumetric assessment of tumor size change for prediction of pCR. The study also showed that MR longest diameter post treatment correlated best with residual tumor size at surgery.
- Residual cancer burden (RCB) at pathology post NAC includes tumor bed area (2 diameters, usually based on pretreatment MR/imaging size), overall cancer cellularity, % that is in-situ disease, number of metastatic nodes and largest nodal deposit. RCB0 = no residual disease, pCR; RCB-I = minimal residual disease; RCB-II = moderate residual disease; and RCB-III = extensive residual disease. Lower RCB predicts improved distant relapse free survival.
- Chang Sen LQ et al: Impact of second-opinion interpretation of breast imaging studies in patients not currently diagnosed with breast cancer. J Am Coll Radiol. 15(7):980-87.e1, 2018
- Coffey K et al: The impact of patient-initiated subspecialty review on patient care. J Am Coll Radiol. 15(8):1109-15, 2018
- Falkner NM et al: Added value of second biopsy target in screen-detected widespread suspicious breast calcifications. J Med Imaging Radiat Oncol. 62(3):299-306, 2018
- González-Huebra I et al: Is it worth to perform preoperative MRI for breast cancer after mammography, tomosynthesis, and ultrasound? Magn Reson Imaging. 57:317-22, 2018
- Raghavendra A et al: Clinical findings and outcomes of MRI staging of breast cancer in a diverse population. Breast Cancer Res Treat. ePub, 2018
- Rosenkrantz AB et al: Discrepancy rates and clinical impact of imaging secondary interpretations: a systematic review and meta-analysis. J Am Coll Radiol. 15(9):1222-1231, 2018
- Scheel JR et al: MRI, clinical examination, and mammography for preoperative assessment of residual disease and pathologic complete response after neoadjuvant chemotherapy for breast cancer: ACRIN 6657 Trial. AJR Am J Roentgenol. 210(6):1376-85, 2018
- Tosteson ANA et al: Second opinion strategies in breast pathology: a decision analysis addressing over-treatment, under-treatment, and care costs. Breast Cancer Res Treat. 167(1):195-203, 2018
- Krammer J et al: Value of additional digital breast tomosynthesis for preoperative staging of breast cancer in dense breasts. Anticancer Res. 37(9):5255-61, 2017
- Kuhl CK et al: Impact of preoperative breast MR imaging and MR-guided surgery on diagnosis and surgical outcome of women with invasive breast cancer with and without DCIS component. Radiology. 161449, 2017
- Grimm LJ et al: Can breast cancer molecular subtype help to select patients for preoperative MR imaging? Radiology. 274(2):352-8, 2015
- Houssami N et al: Meta-analysis of the association of breast cancer subtype and pathologic complete response to neoadjuvant chemotherapy. Eur J Cancer. 48(18):3342-54, 2012
- Hylton NM et al: Locally advanced breast cancer: MR imaging for prediction of response to neoadjuvant chemotherapy--results from ACRIN 6657/I-SPY TRIAL. Radiology. 263(3):663-72, 2012
- Morrow M et al: MRI for breast cancer screening, diagnosis, and treatment. Lancet. 378(9805):1804-11, 2011
- Tot T et al: Breast cancer multifocality, disease extent, and survival. Hum Pathol. 42(11):1761-9, 2011
Related Differential Diagnoses