Mark Palacio is Executive Editor of Applied Radiology.
With any screening modality, the primary point of controversy lies not with the screening but with the cost of follow-up, e.g. how accurate is the screening at detecting truly malignant pathology. In breast cancer, declining mortality rates have bolstered the use of mammography as a screening tool. However, recent revision to the guidelines for screening mammography have created a firestorm of debate over the cost associated with following-up suspicious lesions (for more see the Guest Editorial in this issue on page 4). In the quest to determine who should be biopsied, several modalities have emerged and this article addresses recent literature in an attempt to provide a glimpse of the options that exist to better characterize suspicious breast lesions.
The primary methods of follow-up include magnetic resonance imaging (MRI) and ultrasound. Ultrasound has a relatively low cost compared with any other follow-up modality. MRI has been shown to be a beneficial diagnostic adjunct for many women with certain high-risk factors for breast cancer, e.g., possessing the BRCA1 or BRCA2 genetic mutations, those who have dense breasts and those with a family history of breast cancer. Coupled with these two mainstays are the relatively new entries of positron emission mammography (PEM) and breast specific gamma imaging (BSGI). Both of these modalities show comparable results with MRI.
Biopsy systems are now available for all four alternative imaging methods, allowing clinicians to immediately characterize lesions. Generally, results indicate that the tools work better alongside than instead of one another.
As the clinical gold-standard for follow-up, MRI generally sets the benchmark against which the sensitivity and specificity of other imaging tools are measured. MRI provides a look at both breasts and in some studies has shown significant benefit in detecting mammographically occult lesions in the contralateral breast. However, MRI is not for everyone, especially those who are claustrophobic or who have renal insufficiency. Additionally, MRI carries a hefty price tag and so the modality is used judiciously.
A recent study found that contralateral breast screening with MRI should be considered in postmenopausal women with newly diagnosed breast cancer, even those aged ≥70 years at diagnosis.1 The study evaluated the prevalence of synchronous, occult, contralateral breast cancer detected by MRI but not by mammography or clinical breast examination. A total of 425 women were evaluated, of whom 129 were aged ≥70 years. A contralateral biopsy was performed solely on the basis of MRI in 72 of the 425 women. Fully 16 of the 72 women had pathologically confirmed carcinoma, including 7 in the older subgroup. The prevalence of clinically and mammographically occult contralateral carcinoma, detected by MRI, was 3.8% overall and 5.4% in the group of older women. When potential risk factors for contralateral breast cancer were evaluated, postmenopausal status was the only significant predictor of contralateral cancer detected by MRI.
Another study showed that a second-look ultrasound, directed by MRI results, was useful for decision making as part of a diagnostic workup.2 Malignant lesions detected by MRI were likely to have an ultrasound correlate, especially when they presented as masses on MRI. However, the sonographic findings of these lesions were often subtle, and a careful scanning technique was needed for successful MRI-ultrasound correlation. The retrospective review of 158 patient records, of 202 breast abnormalities initially detected on MRI, showed that ultrasound correlation was made in 115 (57%) of lesions. This included 33 malignant lesions and 82 benign lesions. The remaining 87 lesions were not sonographically correlated and included 11 malignant lesions and 76 nonmalignant lesions. Mass lesions, as identified on MRI, were more likely to have a sonographic correlate than non–mass-like lesions (65% vs. 12%, respectively), according to the study. Malignant mass lesions were more likely to show an ultrasound correlation (85%). Malignant lesions with successful sonographic correlation tended to present with subtle sonographic findings.
Because of its relatively low cost, ultrasound has entered the clinical armamentarium as the second look for suspicious mammographic findings. In the past 5 years, advancements have been made in elastography, which is proving itself to be an effective, convenient technique that can be added to routine breast ultrasound. Elastography may help radiologists better distinguish cancerous breast lesions from benign lesions. It improves ultrasound’s specificity by measuring the compressibility and mechanical properties of a lesion. Since cancerous tumors tend to be stiffer than surrounding benign tissue, a more compressible lesion on elastography is less likely to be malignant.
Ultrasound alone has a relatively low specificity. Numerous studies show that approximately 80% of biopsied breast lesions are benign, according to the American Cancer Society. Elastography is generally considered as a promising method of improving ultra-sound’s ability to characterize benign from malignant tissue. A study at the 95th Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA) showed that elastography properly identified 98% of lesions that were malignant on biopsy and 82% of lesions that were benign on biopsy.3 Elastography was also more accurate than ultrasound in gauging the size of the lesions, according to the study.
“There is a lot of room to improve specificity with ultrasound, and elastography can help us do that,” said the study’s lead author, Stamatia V. Destounis, MD, in an RSNA press statement. Dr. Destounis is a diagnostic radiologist at Elizabeth Wende Breast Care, Rochester, NY. “It is an easy way to eliminate needle biopsy for something that’s probably benign.”
Another group of researchers presented an RSNA review of 7-year data using elastography as an adjunct to routine breast ultrasound. They showed elastography safely decreased the number of biopsies of benign lesions, while offering the potential to more precisely map tumors.4 Researchers found that correlating elastography results with B-mode, color Doppler, and subsequent histology and follow-up was crucial. Elasticity combined with B-mode and color Doppler had 99% sensitivity (for invasive cancers) and 47% specificity. They found that 90% of cancers were stiff and appeared larger on strain imaging with better demonstration of tumor extent. Subsequently 40% of benign lesions appeared significantly smaller. There was no stiffness exhibited in normal breast parenchyma, fat islands and gynaecomastia.
While more novel to the market, dedicated nuclear medicine techniques are showing promising results. For instance, one large, multiyear study of positron emission mammography (PEM, Naviscan, San Diego, CA) showed that the scanners significantly outperformed MRI when differentiating between benign and cancerous lesions. The prospective study demonstrated that combining PEM and breast MRI dramatically increased a radiologist’s ability to detect potentially cancerous lesions compared with relying on MRI alone. PEM scanners are high-resolution breast PET systems that can show the location as well as the metabolic phase of a lesion.
The study examined 388 women with newly-diagnosed breast cancers, and it focused on detecting secondary, mammographically occult tumors. Researchers found that PEM scans had 80% specificity, accurately distinguishing 151 of 189 benign additional lesions. MRI had 66% specificity for the same group of benign lesions. The study also found that the most reliable way to identify lesions was to combine the technologies. PEM and MRI scanning together identified an additional 31 (out of 116 total) lesions.
“We looked very carefully for additional cancers and were surprised that barely half of these were shown on MRI,” said Wendie Berg, MD, PhD, of Owings Mills, MD, and principal investigator for the trial, in a prepared statement. “Clearly there is need for better surgical planning information. PEM significantly improved detection of additional disease over MRI alone.”
Breast specific gamma imaging (BSGI), from Dilon Diagnostics, Newport News, VA, is another molecular breast imaging technique that can visualize lesions independent of tissue density. BSGI is a high-resolution, small field-of-view gamma camera that has recently received U.S. Food and Drug Administration (FDA) 510(k) marketing clearance for a lesion localization system.
At the 2009 RSNA Annual Meeting, researchers presented several papers discussing the results of BSGI. Most promisingly, the technology was shown to be highly sensitive for the detection of ductal carcinoma in situ (DCIS).5 In a retrospective study of 55 women with 57 biopsy-proven DCIS lesions, BSGI was shown to have 89.5% sensitivity. Of the 38 cases of biopsy-proven DCIS in 34 women, 89.5% were detected with BSGI. Findings indicated that the pathologic tumor size of the DCIS ranged from 0.1 cm to 3.1 cm in 33 cases. BSGI had 90.5% sensitivity for 1 cm or smaller DCIS, and it detected DCIS as small as 1 mm.
“The sensitivity of BSGI detection of DCIS is comparable to that reported for MRI detection of DCIS (87.9% and 92%). However, BSGI is more cost-effective than MRI, can be performed in all patients regardless of claustrophobia, renal insufficiency, metal/cardiac implants or patient weight,” said Rachel Brem, MD, in a prepared statement. Dr. Brem is Director of Breast Imaging and Intervention at George Washington University Medical Center in Washington, D.C., and Vice Chair of the Department of Radiology.
While the methods described present some of the more tried and true alternative imaging, newer entries including computer-aided detection (CAD) applications, breast tomosynthesis, dedicated cone-beam breast computed tomography and contrast-enhanced ultrasound are also making headway in the United States and overseas. In the near future, radiologists will still rely on a combination of these tools in order to properly characterize breast lesions, ensuring that fewer negative biopsies are performed.
- Bernard JR Jr., Vallow LA, Deperi ER, et al. In newly diagnosed breast cancer, screening MRI of the contralateral breast detects mammographically occult cancer, even in elderly women: The Mayo Clinic in Florida experience. Breast J. 2010;Epub ahead of print.
- Abe H, Schmidt RA, Shah RN, et al. MR-directed (“second-look”) ultrasound examination for breast lesions detected initially on MRI: MR and sonographic findings. AJR Am J Roentgenol. 2010;194:370-377.
- Destounis S, Arieno A, Skolny M, et al. The role of elasticity imaging (EI) in a large community-based breast imaging center. Presented at the 95th Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA). LL-BR4059-B04: Nov. 30, 2009.
- Putturaya S, Stewart V, Zaman N, et al. Analysis of the clinical utility of real time elasticity imaging as a part of routine symptomatic breast ultrasound. Presented at the 95thScientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA). SSK02-02: Dec. 2, 2009.
- Torrente J, Rapelyea J and Brem R. Sensitivity of breast specific gamma imaging (BSGI) for the detection of ductal carcinoma in situ. Presented at the 95th Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA). VM21-11: Nov. 30, 2009.