Answers are grounded in this paper and cite their sources.
Breast cancer screening is once again controversial. It is hard to miss headlines stating that mammography is an imperfect screening test, missing biologically aggressive cancers and picking up indolent cancers that do not need treatment. We, who have devoted our lives to fighting breast cancer, have long been aware of the limitations of mammography, especially in women with extremely dense breast tissue. Mammography, for all its limits, is still the only test proven to decrease mortality in multiple randomized controlled trials and through experience with population-based screening. Detecting small cancers on imaging before they are palpable improves survival, as well as treatment options. Yet, questions persist. Does the test pick up too much and still not enough; is it wrong too often? Perhaps relying on anatomic density, distortions, and secondary byproducts (calcifications) of cancer on mammograms is not enough; we need a better test. As with oncologic treatment, oncologic imaging these days relies more on functionality than anatomy. Screening for breast cancer needs to catch up to this paradigm to better image clinically significant malignant changes. The quest for improvement in sensitivity and specificity of breast cancer screening is precisely why new tests are being developed by the radiology community. Traditional two-dimensional digital mammography is being supplanted by three-dimensional digital breast tomosynthesis (DBT), and contrast-enhanced digital mammography (CEDM) is a test that images vascularity as well as anatomic abnormalities. Screening breast ultrasound is an increasingly requested supplemental test in women of all breast densities. Luckily, we have had the most sensitive test for breast cancer detection at our disposal for decades: breast magnetic resonance imaging (MRI). And it rarely misses invasive breast cancers. Breast MRI can tell us functionally how a lesion is behaving as the images reflect the tumor’s molecular/genetic characteristics. Breast MRI does not use radiation, cannot induce cancers, and is exceedingly safe even though it does require intravenous contrast. Unlike mammography, which generates images based on the density of tissue, MRI creates a “blood flow map,” detecting tumor neovascularity and peritumoral inflammation; this explains its high sensitivity. Therefore, unlike mammography, MRI relies on alterations that correlate with proliferation and possibly metastatic potential. As medical oncology moves into new therapies based on oncogenetics and greater understanding of cell biology, the evolution in imaging is similarly a function of our new understanding of tumor biology. In the article that accompanies this editorial, Kuhl et al show that a rapid (ie, 3-minute) breast MRI is comparable to the accepted standard 21-minute study when screening for cancer. They pared down the imaging protocol to the bare basics (three dimensional maximum intensity projection [MIP] and first postcontrast subtracted T1-weighted image [FAST]), cutting out any nonessential sequences. Although this protocol is likely not appropriate for diagnostic studies, by doing this in the screening setting they achieved a high detection rate without a high false-positive rate, the hallmarks of a quality screening test. They report that 603 screening rounds in 443 asymptomatic women at intermediate to slightly increased risk of breast cancer and with negative mammograms yielded 11 breast cancers, including seven invasive cancers and four ductal carcinoma in situ (DCIS; 18.3 of 1,000). All were Tis or T1, N0, M0, and almost all were path or nuclear grade 2 or 3. Median tumor size was 8.4 mm, and no interval cancers were diagnosed. MIP images allowed quick detection of enhancement that was further evaluated on the FAST images. MIP analysis alone yielded a sensitivity and negative predictive value of 98.9%, which increased to 100% with the FAST images. These results are exciting for they show that an ultrafast MRI scan can detect all those cancers—invasive as well as DCIS—that mammography cannot. The benefits of a fast examination are obvious. Patient tolerance is improved, and there is great potential for cost savings, which is related to shorter scan time. Kuhl et al show us that we can chip away at the cost of MRI and possibly make screening breast MRI competitive with other imaging tests, including mammography and ultrasound. Breast density legislation is now present in many states in the United States. Women and their physicians are requesting supplemental screening above and beyond mammography. It has perplexed many of us in the breast imaging community how screening breast ultrasound appears to be accepted for supplemental screening whereas breast MRI is not. There is not a single study that shows ultrasound detects more cancer than MRI. In fact, studies show that a significant number of cancers are missed on screening ultrasound but are readily found with MRI. Importantly, false positives are much greater with screening ultrasound than with screening MRI. In fact the likelihood of finding a cancer at biopsy performed on the basis of ultrasound screening is less than 10%, whereas it is approximately 30% with MRI. A much higher percentage of biopsies from screening MRI turn out to be cancer.
No takes yet. Share an insight, caveat, or question.
Elizabeth A. Morris (2014) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: