Clinical trial recommends digital over film mammography for high-risk women

A clinical imaging study published in The New England Journal of Medicine on Sept. 16 could spark substantial new demand for full-field digital mammography.

The Digital Mammographic Imaging Screening Trial (DMIST) found that digital mammography was significantly more accurate than film mammography, when screening for breast cancer in three groups of women: those who are under age 50, those who have radiographically dense breasts, and those who are pre- or perimenopausal. Consequently, the authors of the study recommended the use of digital mammography for women in these groups.

Findings from the study will likely boost sales of digital mammography units, according to industry pundits, as the results suggest clinical value from the use of these systems in a substantial number of patients. Short-term sales may be limited, depending on the institution and its stage of the annual business cycle. The long-term effect, however, will run through this cycle and may be amplified if study results successfully fuel efforts to increase reimbursement for the use of digital mammography equipment.

"There's already a pretty broadly used premium in insurance reimbursement for digital studies, which has been under a particularly withering attack from Blue Cross Blue Shield groups across the county," said Bill Healy, vice president and general manager of the Breast Cancer Detection business line for Hologic. "In some cases, insurers have been able to successfully eliminate it and bring it down to the analog level. Hopefully, this (study) will arrest this and let the modest premium stay in place."

The potential market for digital mammography is huge. Women the study found likely to benefit the most from digital mammography make up 40% of the women over age 50 and 60% of women under age 50, said Georgia Hitzke, vice president of clinical and product management for Hologic. A study conducted by the medical device analysis company Millennium Research Group and released prior to DMIST predicted revenues from digital breast imaging would rise to $480 million by 2009, up from less than $100 million today.

The digital mammography market has been growing at a double-digit annual rate since the systems were introduced in 2000. The total number of full-field digital systems in the U.S. at the end of August, however, was only 985, according to the FDA. With approximately 13,000 FDA-approved mammography centers in the country, digital mammography systems have barely scratched the surface in terms of overall market penetration.

This study could be what the digital mammography industry has been waiting for. Previous clinical trials that compared screen-film and digital mammography showed no difference in the accuracy of diagnosis, concluding that the two technologies were comparable in their ability to find breast cancers. These trials were limited, however, because they included only one type of digital detector or lacked the statistical power to reveal small differences in diagnostic accuracy. DMIST was designed to find such small differences.

DMIST was also unequaled in size and scope. It is the largest screening study of its kind, said Vince Polkus, a mammography applications product manager for GE Healthcare. The study evaluated imaging data obtained from 42,760 women who were screened at 33 sites in the U.S. and Canada, ranging from large academic institutions to freestanding diagnostic imaging centers.

The study also found differences in diagnostic accuracy in a population of women that is notoriously difficult to screen for breast malignancy with standard x-ray mammography. Because screen-film mammography is a 2D technique, normal healthy tissue, particularly in dense glandular breasts, often overlies and obscures nearby cancers. The difference in contrast between a lesion and fibrous glandular background tissue is often so slight it cannot easily be identified.

Digital mammography not only has detectors with a wide dynamic range, which enhance the ability to differentiate between subtle changes in contrast, but it allows fine-tuning of images on soft-copy readers to adjust the brightness and contrast of images and apply a variety of image processing tools.

For the DMIST study population as a whole, as well as for other subgroups of patients, there was no difference in diagnostic accuracy between screen-film and digital mammography. Digital mammography nevertheless was easier to use. The authors reported that digital mammography provided easier access to images and computer-aided diagnosis as well as more efficient transmission, retrieval, and storage of images. Digital mammograms also could be obtained with a lower average dose of radiation.

To Bob Britain, vice president of medical products for the National Electrical Manufacturers Association, DMIST will help bring mammography into the 21st century, because it recognizes the advantages of digital technology.

"There are very good things about digital imaging. It's efficient when reading images. It's efficient when transmitting images to different experts to get a second opinion. And it's very efficient when storing, archiving, and comparing images with each subsequent visit," he said. "With this kind of (study results), physicians will be more likely to invest in the higher cost of digital equipment because of the efficiencies it brings."

Making the switch to digital mammography will be challenging, however. Adding an incremental digital mammography system to an existing infrastructure that can transfer digital images is fairly straightforward. But imaging centers that do not already have digital capability will need to tie digital mammography into an overarching IT system. There is also a learning curve for radiologists who have never read images on soft-copy workstations.

Mammography centers, which must contend with low rates of reimbursement, have been reluctant to invest $400,000 to $500,000 for digital technology, an amount four times more than they would pay for analog equipment.

Early adopters of the technology were willing to bite the economic bullet because they were comfortable that digital mammography was clinically comparable to screen-film mammography and looked forward to savings that would result down the road from reducing the cost of chemicals, film, and archiving and from increasing throughput.

Other centers have waited for a clear answer to the question of whether digital mammography is better.

"We're hoping the DMIST trial will convince more customers that full-field digital mammography is the real deal and they will choose to adopt it," Polkus said.