Women with dense breasts are increasingly being screened with magnetic resonance imaging (MRI), which is clearly the best way to detect small cancers in this population, according to Elizabeth Morris, MD, FACR, FSBI, FISMRM, Professor and Chair of the Department of Radiology at the University of California, Davis. Dr. Morris is former Chief of Breast Imaging Service at Memorial Sloan Kettering Cancer Center, New York. Dr. Morris outlined the advantages of MRI in the subset of women with dense breasts and described other disrupters of traditional mammography at the virtual edition of PER’s Annual Miami Breast Cancer Conference.1
Advantages of MRI
MRI uses contrast, which detects tumors with angiogenesis. Thus, it is a preferred screening tool for certain populations, such as women with extremely dense breasts. Density not only increases the risk of cancer but can mask tumors, and this is where contrast is helpful, Dr. Morris said.
“MRI uses contrast, and this is what allows us to detect disease at a smaller size and earlier stage than if we waited for secondary signs of malignancy on mammography.”— Elizabeth Morris, MD, FACR, FSBI, FISMRM
Tweet this quote
As a contrast-based method, MRI detects more biologically relevant disease than traditional screening methods, even three-dimensional (3D) mammography. “It’s not so much that MRI picks up earlier breast cancer—it’s that it uses contrast, and this is what allows us to detect disease at a smaller size and earlier stage than if we waited for secondary signs of malignancy on mammography,” she explained.
Two less expensive alternatives to MRI are abbreviated MRI and contrast mammography. Abbreviated MRI takes about 3 minutes to perform and produces excellent images. With traditional MRI, radiologists focus on the “end of the curve,” which reflects gadolinium uptake. With abbreviated MRI, they focus on the earlier part of the curve, which has become possible with the advent of rapid scanning. Using precontrast and postcontrast imaging, radiologists obtain the information they need to read the image, explained Dr. Morris.
Quality is not compromised with abbreviated MRI, she emphasized. According to Dr. Morris, it is a faster, less expensive test with equal cancer detection and false-positive rates. In addition, many patients may tolerate it better than other methods, she added.
Recent Publications Support Benefit
Two publications in 2020 supported the benefit of using MRI to screen women with dense breasts.
First, the DENSE trial in the Netherlands evaluated the benefit of MRI (traditional, not abbreviated) in women with extremely dense breasts as a supplement to a negative screening mammogram.2 Of 40,373 women who underwent mammography, 8,061 were invited to undergo MRI, and 32,312 were assigned to have mammography alone.
Supplemental MRI was found to detect more cancers—16.5 per 1,000 screenings as compared with 6.8 per 1,000 screenings with mammography—and these tumors were largely high grade and therefore potentially deleterious cancers. The consequence of this greater detection was a notable decrease in the rate of interval cancer—from 5.06 per 1,000 screenings to 0.83 per 1,000 screenings—a difference of 4.2 per 1,000 examinations (P < .001).
“Is abbreviated MRI ready for the clinic? Yes, it is, and it is being offered at many centers.”— Elizabeth Morris, MD, FACR, FSBI, FISMRM
Tweet this quote
“The findings led to coverage of MRI screening for women with extremely dense breasts in the Dutch National Screening Program. In the United States, we would like to do MRIs in heterogeneously dense breasts, not just extremely dense breasts; however, we need data before insurers will cover this. We are very encouraged by these results,” commented Dr. Morris.
The second study showed that abbreviated MRI detected more cancers than traditional screening with 3D tomosynthesis.3 Invasive cancers were twice as likely to be detected by MRI, and no invasive cancers were found with tomosynthesis alone.
“Is abbreviated MRI ready for the clinic? Yes, it is, and it is being offered at many centers,” Dr. Morris said. “These days, any MRI [hardware] can perform abbreviated MRI if it’s accredited to do so. It’s not a matter of hardware. It’s the implementation and the radiologist’s comfort with it.”
“The issue is billing,” noted Dr. Morris. “There’s no CPT [current procedural terminology] code. Patients often have to pay cash or charge through their insurance. That’s the stumbling block more than the technology or being able to use it.”
Advances in imaging beyond abbreviated MRI are on the horizon. Coming soon is “ultrafast” MRI, which produces even better images within seconds of the gadolinium injection and offers the opportunity for dynamic information during routine screening, Dr. Morris shared. Ultrafast MRI can easily differentiate tumors from the surrounding parenchyma and can pick up very small cancers, such as a 4-mm triple-negative tumor in a high-risk patient, she added.
Concerns about gadolinium deposition (particularly in the brain) have led researchers to ask whether contrast is necessary at all in MRI. There are currently some MRI techniques where contrast is not needed (ie, diffusion-weighted imaging, diffusion-tensor imaging), but although they can reveal cancer, such approaches are not yet sensitive enough for clinical use. Most centers are now using safer agents, with less risk of deposition and other potentially harmful effects, commented Dr. Morris.
“Trials of screening without contrast are under way, and they will be interesting to keep an eye on,” she added.
Of great interest in radiology is the use of artificial intelligence (AI) and machine learning. “AI machine learning has disrupted radiology. We cannot only analyze images and detect cancer with these algorithms, but we can also assess individual risk,” Dr. Morris explained.
A recent study evaluated a mammography-based deep learning model for risk prediction, finding it to be more accurate than established clinical risk models.4 The retrospective study included 88,994 consecutive screening mammograms for 38,571 women; cancer outcomes were obtained from tumor registries, and risk factor data were extracted from questionnaires and electronic medical records. Deep learning models outperformed the clinical risk model (which included breast density, Tyrer-Cuzick version 8) in predicting individual risk.
“As breast imagers, we know that two breasts are never alike. Every woman’s mammogram is unique to her, like a thumbprint, and contains information that does, in fact, predict her future risk of cancer. The pattern of the parenchyma gives important information that could be telling about risk. With MRI, we have a breast parenchyma pattern and enhancement associated with it. This is an interesting space that will be exciting to follow,” concluded Dr. Morris.
DISCLOSURE: Dr. Morris has received institutional research funding from GRAIL.
2. Bakker MF, de Lange SV, Pijnappel RM, et al: Supplemental MRI screening for women with extremely dense breast tissue. N Engl J Med 381:2091-2102, 2019.
3. Comstock CE, Gatsonis C, Newstead GM, et al: Comparison of abbreviated breast MRI vs digital breast tomosynthesis for breast cancer detection among women with dense breasts undergoing screening. JAMA 323:746-756, 2020.
4. Yala A, Lehman C, Schuster T, et al: A deep learning mammography-based model for improved breast cancer risk prediction. Radiology 292:60-66, 2019.