Digital mammography vendors start looking to next step

August 8, 2001

Full-field digital mammography is receiving a full-court press by imaging vendors, several of which are in various stages of device testing or FDA review. But regulatory approval is only the first hurdle. More serious is the challenge to show a clearcut

Full-field digital mammography is receiving a full-court press by imaging vendors, several of which are in various stages of device testing or FDA review. But regulatory approval is only the first hurdle. More serious is the challenge to show a clearcut clinical edge over its conventional counterpart.

“If you look at the beginnings of any digital modality, first-generation products didn’t realize their ultimate potential,” said Don Blomstrom, national sales manager for mammography at Instrumentarium. “Developing the digital mammography machine is the easy part. It’s the presentation and management of that information once it is acquired that’s hard.”

Help may come from a new trial conducted under the auspices of the American College of Radiology Imaging Network (ACRIN). The three-year study will employ digital devices from several competing vendors and compare the clinical performance of digital and conventional screen-film mammography.

“We hope the trial will answer the question of whether digital mammography has any clinical benefit over screen film. I don’t think anybody’s really answered that,” said Dr. Laurie Fajardo, director of breast imaging at Johns Hopkins University and principal investigator to Hopkins in the $25 million National Cancer Institute-funded study.

The ACRIN trial will involve 49,500 asymptomatic women at 20 participating centers in the U.S. and Canada. All women will undergo both digital and screen-film mammography and will be followed for up to 15 months.

The trial will measure clinical efficacy and evaluate two related issues. The first is the relative cost-effectiveness of film and digital technologies. The second is the impact on patient quality of life that results from an expected reduction of false-positive mammograms with digital mammography.

“Many experts believe digital mammography is better than conventional mammography, and this study should provide the definitive answer,” Fajardo said.

Vendors are not waiting for those results. GE’s Senographe 2000D is already FDA-approved and in use. Others are not far behind.

In late July, Fischer Imaging received an approvable letter from the FDA for its SenoScan full-field digital mammography system (see cover story). In May, Hologic submitted a premarket approval application to the FDA for its Lorad full-field digital device. Fuji is testing computed radiography for mammography. And Instrumentarium is in the early clinical testing phase of its Diamond digital mammography unit.

Digital may hold numerous advantages over screen-film mammography: quick image access, efficient organization of double-reads, the option for telemammography and CAD, and no film costs, according to Dr. Ingvar Andersson, chair of diagnostic radiology at Sweden’s Malmö University.

But there are lingering technical limitations. Digital mammography suffers in comparison to its screen-film counterpart in terms of sufficient spatial and gray-scale resolution. And many digital mammography workstations are too slow for a high-volume screening program.

“The devices that are now or will soon be available incorporate any number of trade-offs and compromises,” said Roman Janer, vice president of Lorad’s digital imaging program. “But the next generation of equipment, which relies on amorphous selenium receptors, will offer much better performance due to direct-conversion technology.”

Such improved performance will have a direct impact on the ability to use digital machines for diagnostic purposes-such as digital subtraction angiography studies of breast vasculature that are less invasive that the current method, Janer said.

The Lorad system undergoing PMA review by the FDA uses charge-coupled device technology, he said. But the company is already working on next-generation equipment that will incorporate selenium plates. It’s a direction in which other companies, including Instrumentarium, are also going.

“Selenium (receptors) will be a second-generation enhancement,” Blomstrom said. “Kodak is using it, Siemens has announced that it will be using it, and Lorad is using it. It’s safe to say that selenium is going to be the de facto standard in a couple of years.”

Because of these advances on the horizon, participation in trials like ACRIN is a double-edged sword, Janer said. The trial will use currently available technology. With the study due for completion in 2004, the results will necessarily reflect the performance of outmoded devices.

“The images acquired using the equipment we have today are excellent,” he said. “But they will be even better when acquired by the next-generation device.”

In addition to the challenge of demonstrating advantages of digital technology versus screen-film, vendors must also position their products to differentiate them from competitors.

Until now, for example, GE has been able to boast that it offers the only FDA-approved device on the market. But with final approval for Fischer’s system possibly just weeks away, that edge is fading. Now GE is gearing up for FDA approval of its combined digital mammography and computer-aided detection system.

The pairing is expected to give both technologies a market boost. CAD will be a standard feature of next-generation workstations, according to Andersson. In clinical trials, CAD has been shown to increase breast cancer detection rates without decreasing specificity. CAD and digital mammography go hand-in-hand, Blomstrom said, particularly when it comes to using some of the advanced image processing that digital technology affords, such as 3-D rendering. Instrumentarium has incorporated a 3-D feature in its digital system, and he believes the system will be more than what radiologists expect to see.

“Radiologists won’t have to look at two views of the breast and in their minds construct the 3-D relationship,” Blomstrom said. “They’ll be able to do it electronically, by separating layers of interest and margins.”

Vendors will need to work on both sides of the technology fence when they pair CAD with mammography, Janer said. Because film will be around for many years to come, vendors need to accommodate customers who have both digital and conventional mammography devices.

“The approach we are following at Lorad is that CAD is an independent black box,” he said. “You need to have a CAD engine for your direct digital images. Our product will provide the connectivity that facilitates transferring the CAD data to the soft-copy review workstation. It’s the sort of open architecture concept that customers who buy CAD systems today will need in the future. The device they buy is not going to become obsolete when they purchase a digital mammography system from us.”

Whether it’s 3-D rendering, combined CAD capability, or selenium versus CCD, it will take time for the market to sort out exactly what customers want from digital mammography. The only sure way to find out is to get systems out and in use, Blomstrom said.

“Somebody had to break ground from a regulatory standpoint, and now that that has happened, we can all move,” he said.