CHICAGO—Stereoscopic digital mammography, an investigational technique that produces three-dimensional, in-depth views of the breast, may reduce the number of patients having to be recalled due to false positives, compared with standard digital mammography, suggest interim results of an ongoing study.
Results of the first 1,093 women to be enrolled in a single-center trial comparing the two techniques show that stereoscopic digital mammography reduced the number of false-positive detections by nearly half, “a highly significant result,” reported David J. Getty, PhD, Division Scientist at BBN Technologies in Cambridge, MA, which developed the workstation used in the Department of Defense-funded study.
The study was powered to detect even a 15% reduction in mammography recalls with a planned enrollment of 1,500 patients, Dr. Getty said here at the Radiological Society of North America Annual Meeting.
“We are reporting these preliminary data at this time because we have already reached significance,” said Dr. Getty, who has been working on developing the technology for the past 12 years.
False positives are a well-accepted surrogate for mammography recalls, noted primary clinical investigator Carl J. D'Orsi, MD, Professor of Radiology and Director of Breast Imaging at Emory University, where the trial is being conducted.
“With standard mammography, up to 30 percent of women are recalled due to superimposition. By reducing the number of false positives, stereotactic mammography will reduce the number of women having to be recalled,” he said.
Dr. Getty said the new technology is designed to overcome “three subtle problems of standard mammography in which radiologists view the breast in the form of two orthogonal 2-D images.
“In a two-dimensional image of the breast, subtle abnormalities may be masked by underlying or overlying normal tissue, leading to false negatives. Also, normal tissue scattered at different depths can align to mimic a lesion, leading to false-positives. Finally, there is limited volumetric information about 3-D structures,” he explained.
In contrast, stereoscopic digital mammography permits the radiologist to view the internal structure of the breast directly and in depth.
In the approach, a modified digital mammography scanner is used to take vertical and lateral images from two views separated by an angle of six to 10 degrees while the breast remains in a fixed position. A stereo display workstation is used to fuse the images, creating a 3-D view of the breast.
“It's like seeing the breast with a View-Master or in a 3-D movie,” Dr. Getty said.
Dr. D'Orsi added that the image is not only more readable, but also takes less time to read than standard mammography.
In the study, each patient received both a standard digital screening examination and a stereoscopic screening examination. The results were read independently by different radiologists.
If a suspicious finding was found using either method, the patient was called back for diagnostic workup—a film mammogram, as well as biopsies when needed—with the results forming the basis for lesion truth.
“Women served as their own controls, and we worked up everything, so we are confident with these data,” Dr. D'Orsi said.
One or both of the methods detected 259 abnormal lesions, Dr. Getty reported. Of those, 109 were confirmed to be true lesions on diagnostic workup and 150 turned out to be false positives.
Standard mammography produced 103 false positives, while stereo mammography produced 53. This corresponded to a significant, 49%, reduction in false positives, he said.
Of the 109 true lesions, standard digital mammography failed to detect 40, while stereo mammography missed only 24—corresponding to a nearly 40% reduction in false negatives. While this result did not quite reach statistical significance—the p value was 0.06—he said that the “strong trend” suggests that significance will be reached as more patients are studied. The researchers expected to complete enrollment of the planned 1,500 women by the end of the year (just a few weeks after he presented the data at the meeting), he said.
Also, standard mammography missed 41 cases of clustered calcification, while stereo mammography missed only four—translating to a highly significant 80% reduction in the chance of missing a calcified lesion with the newer technique.
This is noteworthy, Dr. D'Orsi said, as “a good deal of DCIS first presents as clustered calcification.”
Entails Only Minor Changes to Standard Digital Mammography Equipment & Software
Joseph H. Tashjian, MD, President of St. Paul (MN) Radiology, MN, said that an advantage of the new technology is that it entails only minor changes to standard digital mammography equipment and software.
“What's also nice is that it allows you to see the breast in 3-D and see in depth,” he explained. “Stereotactic mammography allows you to separate [layers of tissue] out, so you don't have so many false positives where you think something is an abnormality but it turns out just to be overlapping tissue.
“On the other hand, you see fewer false negatives where abnormalities are obscured by overlapping tissue. The stereotactic mammography allows you to visualize the abnormality by separating the overlapping tissue.
But the question we want to ask,” he said, “is whether having a fused stereo image is better than having two images. We still don't have the answer to that.”