Dartmouth Medicine HomeCurrent IssueAbout UsContact UsSearchPodcasts

Vital Signs:
Study of innovative breast-imaging methods gathers benchmarks

Three out of four women who undergo a breast biopsy turn out not to have breast cancer. But one out of four asymptomatic breast cancers is not detected by mammography. Although startling, these are not new findings. They are, however, the impetus behind growing national and international research that's aimed at finding more accurate methods to detect and diagnose breast cancer.

"Unfortunately, what often gets quoted to the public is that mammography is 90% sensitive and we can find a cancer that is the size of a head of a pin, which is true in certain people," says radiologist Steven Poplack, M.D., who is codirector of breast imaging/ mammography at DHMC. But, he adds, "that's not the whole story. . . . [Mammography] does have downsides, and it has room for improvement."

New types: For the past five years, he and Keith Paulsen, Ph.D., a professor of engineering, have been examining three new types of spectroscopic breast imaging—near-infrared (NIR), electrical impedance (EI), and microwave imaging (MI). Their most recent paper, of which Paulsen was the principal author, appeared in the May 2004 issue of Radiology.

Poplack and Paulsen's team collected data for each imaging technique from 23 women who had no clinical or mammographic findings of breast cancer. The first of its kind, the study now gives the investigators essential data on tissue characteristics in non-cancerous breasts of varying age and composition, thereby providing baseline information for future studies of these techniques.

Photo by Joseph Mehling
Radiologist Steve Poplack, left, and engineering professor Keith Paulsen, right, are collaborating to find more effective alternatives to mammography.

The researchers hope that these new techniques will "improve characterization of breast disease," says Poplack. "If someone has a lump or a mammographic abnormality, [the hope is that] we could say that based on these alternative exams there's no chance there would be breast cancer here and therefore it does not need to have a biopsy." And beyond the prospects for better, more accurate detection of cancer is the opportunity to learn more in general about breast disease, breast structure, and breast function.

Unlike mammography, NIR, EI, and MI do not involve ionizing radiation—which in theory may harm the breast, though no studies have shown this conclusively. (A fourth method, magnetic resonance elastography, is also being explored by the research team but was not included in this study.)

"We don't really know if radiation at a very low level . . . truly causes cancer in the breast," explains Poplack. "We think that it might, and I should add that the risk-benefit of mammography is far in favor of doing mammography. But, having said that," he adds, "it would be nice to have an option where you didn't have to expose the breast to ionizing radiation."

By operating in other regions of the electromagnetic spectrum, these modalities detect differences in a tissue's electrical properties— its permittivity (its capacity to store an electrical charge) and conductivity—and in its light-scattering and -absorption properties. Certain tissue characteristics can indicate if cancer is present or not. For example, NIR spectroscopy, which measures light scattering and absorption, allows researchers to map the total hemoglobin concentration and saturation in a section of tissue. This is of "particular clinical interest," states the Radiology article, since high hemoglobin concentration is associated with angiogenesis in a tumor—the process by which it develops blood vessels to feed its voracious appetite.

Clinical trials: The initial clinical trials of NIR, EI, and MI appear promising. The research team has already encountered some subtle breast cancers that are "very conspicuous" when using one or more of the new modalities. But many larger trials will be needed to determine whether the techniques have significant value in breast cancer detection and diagnosis.

Paulsen notes that one of the challenges facing the group "is that the mammogram is done in a completely different geometry." Mammography requires compression of the breast, whereas the new procedures do not (and that's a plus, since the compression is very uncomfortable for many women). But that means researchers must calculate how regions of the breast correspond among the various procedures in order to judge the sensitivity of the modalities. "We can estimate that reasonably well," says Paulsen. "But in this next round, we want to do more accurate socalled registration between the alternative images and . . . conventional imaging." This will allow for refinement of the equipment before the next tests.

Poplack and Paulsen are also nearing completion on another study of 150 women. Half of the women have breast abnormalities and are slated for biopsies, while the other half, of corresponding ages and breast densities, are a control group. These results should indicate whether there is a difference in the way the techniques work in diseased and in healthy breasts.

The team hopes, says Poplack, in five years to be able to make "some pretty definite recommendations about which of these modalities, either alone or in combination, should be tried in a multi-institutional trial as the next big step."

For the time being, their most recent published study remains a "small step but an important step," adds Paulsen.

Jennifer Durgin

If you would like to offer any feedback about this article, we would welcome getting your comments at DartMed@Dartmouth.edu.

Back to Table of Contents

Dartmouth Medical SchoolDartmouth-Hitchcock Medical CenterWhite River Junction VAMCNorris Cotton Cancer CenterDartmouth College