Dr. Conant is a pioneer in the development of digital mammography, and a leader in research on the use and benefits of early mammography screening and on the role of MRI and PET scanning. She is also the recipient of grants from the National Institutes of Health to compare standard surgical biopsy with digital mammography and stereotactic core breast biopsy.
Emily Conant, MD
Breast cancer is the most common malignancy in women, accounting for 27% of all female cancers. And although we have made significant progress in the detection and treatment, breast cancer remains the number two cause of cancer mortality among women, accounting for more than 40,000 deaths this year alone. Multiple studies have evaluated risk factors for breast cancer over the past several decades, and several risk prediction models exist. However, 75% of women with this cancer have no risk factors; therefore, proper breast cancer screening remains the single most important tool in the detection and treatment of breast cancer.
In accordance with Breast Cancer Awareness Month, ONCOLOGY spoke with Emily Conant, MD, professor of radiology and chief of imaging at the University of Pennsylvania Medical Center in Philadelphia. Dr. Conant is a pioneer in the development of digital mammography, and a leader in research on the use and benefits of early mammography screening and on the role of MRI and PET scanning. She is also the recipient of grants from the National Institutes of Health to compare standard surgical biopsy with digital mammography and stereotactic core breast biopsy.
-Interviewed by Ron Piana
ONCOLOGY: No screening test has been more carefully studied than mammography; however, the true value of this test remains hotly debated. Should annual screening mammography still be the gold standard in breast cancer detection or is it time to rethink our screening recommendations?
DR. CONANT: I’m still a strong believer in the benefits of yearly screening mammography after age 40 in the general population. It’s very important to carefully analyze the data and weigh the pros and cons and the risk to benefits that exist within any study. Naturally, because mammography is such a widely used screening tool, it gets quite a bit of scrutiny. But it is all about the data, how it’s sliced and diced. Unfortunately many studies are not ideally run.
For example a recent study out of Norway used a very different style of screening. The investigators were looking at women over the age of 50 with invasive cancers, not early-stage DCIS, which in the US we pride ourselves in finding on mammography. Moreover, they screened every 2 years, which is different from how we screen, and the data were from pre-digital screening, which also affects the specificity of the outcomes.
But what we do know is that annual screening mammography decreases breast cancer mortality rates and our challenge is to come up with ways that maximize the benefits and reduce the potential for false positives and over-diagnosis. However, the more you look the more you find; uncertain reading can lead to unnecessary biopsies, which is very stressful for patients.
It is also important about how one defines a false-positive within the clinical context of breast health. For instance, most of the studies being debated included into one mammographic data bucket: false-positives that were biopsied and ended up not being cancer along with women who were found to have benign cysts. In my mind those are two very different situations. We want to avoid the unnecessary biopsies, but a having a breast cyst detected on mammogram is in fact a very educational process for a woman. Studies have shown that those types of false-positives, although initially worrisome, engage women to have more personal control over their healthcare decisions.
In short, to cut back on the use and availability of a life-saving tool like mammography without continuing to refine and improve its efficacy would be like going back in time, and that would be harmful for women’s breast health.
ONCOLOGY: You mentioned tomosynthesis in relation to false negatives/positives. Does this technology help distinguish the fine line between specificity and sensitivity?
DR. CONANT: False readings are a problem in breast screening. They cause unnecessary procedures, unnecessary anxiety, and they waste precious healthcare dollars. So moving forward we need to develop technologies that have better specificity. Tomosynthesis is an exciting new technology that builds from the digital platform, meaning that it’s from a digital mammogram platform. Early studies have shown that tomosynthesis can actually decrease false-positives by about 30%, which is significant.
Instead of looking at the breast in what we call a projection, meaning all the images are super-imposed on top of each other, tomosynthesis peels back the layers of the breast in a three-dimensional format so the radiologist can actually scroll through the image. Like the sun being hidden by clouds, this technology peels away the cloudy tissues of the breast, making it easier to detect tumor.
So this technology not only helps decrease false-positives, but it also gives us the ability to detect smaller cancers hidden behind fluffy breast tissue that we don’t see with standard mammography. So, using tomosynthesis along with the digital mammography platform gives us the ability to advance a relatively inexpensive, widely accessible screening tool, one that we can continue to improve upon.
ONCOLOGY: MRI has been shown to be a successful screening tool in certain women. Where is MRI in the current screening scenario?
DR. CONANT: First off, MRI is currently too expensive for the general screening population; insurance companies simply won’t pay for it. But MRI is valuable in certain high-risk groups, for instance women who have BRCA1 and 2 mutations, for women with extremely strong family breast cancer history, or for some women with newly diagnosed cancer and are trying to map out how extensive the disease is.
MRI is a powerful technology but, like digital mammography, it is also susceptible to many false-positives. Moreover, even if a woman’s insurance covers MRI, that does not mean she should forego mammogram. The two tools are complementary, but they look at the breast in different ways. Mammograms show calcifications, the early sign of tumor development, whereas MRI doesn’t show calcium, it is based more on the blood, the vascularity of the tumor. Studies have shown that some cancers are found only on mammography and some only on MRI. So, moving forward our goal is to combine the best qualities from existing technologies with which to build screening tools that are highly effective and selective in detecting cancer.
ONCOLOGY: Have we reached a point in which genetic testing is an integral part of breast cancer screening?
DR. CONANT: At U Penn we’re actually doing some research looking at how we can best personalize our screening for each woman, not only the women with BRCA mutations that we know are high-risk, but also intermediate risk women and women who, because of their complex breast physiology might also be at risk. So we are aggregating all the clinical information such as the imaging, demographics, history, and genetics into one package, thus giving us the ability to tailor the screening technology and follow-up to her personal needs.
ONCOLOGY: Working within today’s financially challenged healthcare environment, please describe the best way women can maintain breast health.
DR. CONANT: That is an important question, one that women need to ask so that they can become empowered to seek out the best care possible. It’s sort of a three-prong approach. A comprehensive breast health program requires substantive discussions between doctor and patient, maintaining a yearly mammogram schedule, and enhanced breast awareness, which includes regular self-breast exam. The mission for women is to find a central clinical venue that provides all of the above; breast health education and proper screening is the foundation of comprehensive detection and treatment.