Conference Coverage

The better mammogram: Experts explore sensitivity of new modalities


 

EXPERT ANALYSIS FROM NAMS 2017

 

– Is it time to think about “the better mammogram” as the new standard of care? Can nuclear medicine provide a cost-effective workaround for imaging of women with dense breasts? According to two leading breast imaging researchers, it may be time to revisit current screening practices to take best advantage of today’s imaging technology.

“Digital breast tomosynthesis is the new kid on the block for screening,” said Emily F. Conant, MD, professor of radiology and chief of breast imaging at the University of Pennsylvania, Philadelphia. “It’s becoming the new standard of care in mammography,” she said, speaking during a plenary session at the annual meeting of the North American Menopause Society.

Digital breast tomosynthesis (DBT) can involve simultaneous acquisition of a conventional 2D mammogram along with a series of images to create a 3D image. Another protocol, which delivers a lower radiation dose, produces a “synthetic” 2D reconstruction of 3D mammography.*

Mammogram versus molecular breast imaging Courtesy Dr. Deborah J. Rhodes
The mammogram was read as negative while the molecular breast imaging shows an area of uptake corresponding to an 11-cm invasive lobular cancer.
In either case, tomosynthesis, said Dr. Conant, is “a digital reformatting of data” that allows the radiologist to “open the book” of the 2D image to flip through the pages, seeing the 1-mm slices that are the final tomosynthesis product.

In addition to making visible tumors that otherwise might be obscured by the overlay of dense breast tissue, DBT can help reduce the recall rate, with the 3D images providing immediate clarification at the initial appointment. Studies show that the recall rate can go down by up to 31%, Dr. Conant said.

DBT has been shown to increase detection of invasive cancers, but it does not pick up more ductal carcinoma in situ, Dr. Conant said. This fact helps address the problem of overdiagnosis of small tumors that might regress. Overall, cancer detection is reported to increase by up to 53% with DBT, Dr. Conant said.

When primarily retrospective American studies are taken together with smaller prospective European studies, “the improvement in outcomes achieved with DBT directly addresses the major concerns regarding screening for breast cancer with mammography,” she said.

However, so far the studies have not offered DBT routinely to all comers. Since 2011, DBT has been offered to every woman screened at the University of Pennsylvania, at no additional cost. This created “a sort of natural experiment – there was no bias as to who got it.” Three consecutive years’ worth of outcomes have now been analyzed, Dr. Conant said.

Patient-level data from the University of Pennsylvania experience show statistically significant reductions in recall rate from diagnostic mammography alone. Also, researchers saw a steady increase in the rate of cancers detected per 1,000 patients, from 4.6 with digital mammography alone, to 6.1 by year three of DBT (JAMA Oncol. 2016 Jun 1;2[6]:737-43). This reflected the institutional learning curve with DBT, Dr. Conant said.

She said that the data also showed “a promising trend down in false negatives,” with an early reduction in cancers that were missed by DBT. Time is needed for mature cancer registry data to bear out these early trends, she added.

Other recent data show that DBT has promise to improve detection rates in a population of great interest – younger women, where there are often too many false positives and not enough cancers found, Dr. Conant said. If the risk-benefit ratio for DBT continues to play out as the data pile up, “I would strongly suggest that we should be doing screening in the 40s,” she said.

An important caveat, noted Dr. Conant, is that whether tomosynthesis is used or not, mammography captures anatomy, not physiology, and very dense breast tissue may still obscure a tumor, even when the tomographic slices are peeled back.

Though “DBT is ‘the better mammogram,’ additional outcome data are needed,” she said, including studies that compare modalities, include subgroup analyses, and better delineate the effect of cancer biology.
 

Molecular breast imaging

Another imaging modality uses nuclear medicine to capture the physiologic changes that accompany cancer. Molecular breast imaging (MBI), or scintimammography, can help “unveil the reservoir of hidden cancers in dense breasts,” said Deborah J. Rhodes, MD, professor of medicine at the Mayo Clinic, Rochester, Minn.

Dr. Rhodes – along with Michael O’Connor, PhD, Connie Hruska, PhD, Katie Hunt, MD, and Amy Conners, MD, her collaborators at the Mayo Clinic – uses a specialized array of gamma cameras to detect uptake of an injected radionuclide that’s preferentially avid for tumor tissue. This technique can unmask smaller tumors not seen on mammogram because it’s not impeded by having to “see” through dense breast tissue.

The radiation dose for an MBI study is a bit more than for DBT, but less than a coronary calcium score scan. The cost is about one-tenth that of breast magnetic resonance imaging (MRI), and interpretation is relatively straightforward, said Dr. Rhodes, who also presented data at the North American Menopause Society plenary.

“The traditional measure of mammography’s performance inflates its effectiveness,” especially in dense breast tissue, said Dr. Rhodes. “What is the sensitivity of mammography in the dense breast? It depends on what you measure it against.”

When cancers detected by MRI or MBI are added, the sensitivity of mammography drops from the 86.9% reported by the Breast Cancer Surveillance Consortium to 21%-31%, according to several published studies.

In one study, Dr. Rhodes and her Mayo colleagues found that the diagnostic yield per 1,000 patients with dense breasts by mammogram alone was 1.9 cancers. When MBI was added, that figure jumped to 8.8 cancers per 1,000 patients, an incremental gain of 363%.

“Tumor size matters profoundly,” she added. “If a tumor is detected above 2 cm, long-term survival drops below 50%.”

That contrasts with the better-than-80% long-term survival rate seen for those with sub-centimeter tumors, even in node-positive disease. “Only a third of tumors are detected when they are less than 1 cm” with regular screening mammography, Dr. Rhodes said.

However, in 2016 the U.S. Preventive Services Task Force concluded that the current evidence was insufficient to assess whether adjunctive screening for breast cancer using breast ultrasonography, MRI, DBT, or other methods should be used in women with dense breasts. The USPSTF noted that there weren’t studies that addressed the effect of supplemental screening on breast cancer morbidity or mortality.

The problem is that it can take 20 years or more to demonstrate mortality reduction, meaning that “no other imaging modality can compete” with mammography when this yardstick is used, Dr. Rhodes said. “This insistence on a mortality endpoint before we change practice” is impeding progress in screening, she said.

The American College of Obstetricians and Gynecologists “does not recommend adjunctive tests to screening mammography in women with dense breasts who are asymptomatic and have no additional risk factors.” However, the organization “strongly supports additional research to identify more effective screening methods” that will improve outcomes and minimize false positives in women with dense breasts.

Though DBT is becoming more widely available, MBI is still primarily used in research centers. Both Dr. Conant and Dr. Rhodes acknowledged that since these techniques are not required to be covered by insurance, payment – and patient access – may vary. Both physicians said their home institutions have worked hard to keep costs down for their studies.

Dr. Conant is consultant or advisory board member for Hologic. Dr. Rhodes reported having no conflicts of interest.

*Correction, 11/15/2017: An earlier version of this story misstated the synthetic mammography protocol.

   Comments ()

Recommended for You

News & Commentary

Quizzes from MD-IQ

Research Summaries from ClinicalEdge

Next Article: