Researchers at the Smilow Cancer Center of Yale-New Haven Hospital may have identified a more accurate way to diagnose breast cancer. A study conducted by Yale School of Medicine Assistant Professor of Diagnostic Radiology Jaime Geisel shows that 3D imaging, also called tomosynthesis, can reveal breast abnormalities that 2D mammograms alone overlook. Geisel said she found an 11 percent greater cancer detection rate in patients who were screened through both 2D and 3D methods than in those who only received a 2D mammogram. Geisel presented her study to the ARRS Medical Imaging Society on April 19 in Washington, D.C. She met with the News Monday to talk about her experimental methods, the implications of her findings and her plans to pursue this research further.
Q: What prompted you to run this study?
A: Tomosynthesis is a relatively new development in mammography — Smilow received a 3D imaging unit in August 2011, and we were the first in Connecticut to have it. During our first few months of experience with it, we realized we were finding cancers that we couldn’t see with 2D images, and I wanted to do this study to add validity that we were finding more cancers with the new unit. I know there is new stuff coming out all the time, and not everything new is better, but in this case I think tomosynthesis is better. I just sort of wanted to create more awareness that this new, better technology is out there and can be used to leave fewer cases of breast cancer undiagnosed.
Q: How did you go about analyzing whether tomosynthesis is more accurate than a typical 2D mammogram?
A: We essentially looked at data from the first year of having tomosynthesis, and divided every patient who had a screening mammogram into two groups — one for people who just had 2D imaging and another for people who had 2D plus 3D imaging. We found more cancers in the group which had 3D imaging. Actually, we saw 11 percent increased detection, which isn’t too large but it’s still significant. We also noticed that the cancers found through 3D screening tended to more often be in patients with dense breasts.
All of this shows that there are cancers we can’t see on a 2D mammogram that we can only see on a 3D mammogram. With 2D imaging, it’s as if we’re looking at the front page of a book, whereas the 3D method takes multiple images and gives us more information. It’s still not perfect, but it still picks up on more cancers, and it’s better especially for patients with dense breasts.
Q: Do you plan to continue investigating tomosynthesis?
A: Definitely — my study is ongoing. As of now we’ve only analyzed one year’s worth of data, so we need to follow up and see what happens next year to determine if there were any cancers we missed. We want higher numbers, so we’ll be looking at data from other years of tomosynthesis as well. Really, my goal is to look at 3D mammography from every angle, so I also hope to analyze recall rates to see how often we are asking women to come back after a mammogram and to see how often we require six month follow-ups for abnormalities that are probably benign. Both of these rates seem to be reduced by tomosynthesis.
Q: Is tomosynthesis new to the field of breast cancer research, and how do you hope it will change breast cancer diagnosis?
A: Tomosynthesis is a very new concept, and is probably the most exciting thing that’s happened in breast imaging in the past couple years. The FDA only approved it in February 2011, but I know other institutions are also showing similar findings to what we are seeing. We’ve been so impressed with 3D mammography that soon we’ll have five units at Smilow, and we’re hoping to eventually replace all our units with 3D units.
A lot of other places haven’t bought 3D units yet because there is no reimbursement, meaning there is no extra charge to the insurance company when a patient is screened through tomosynthesis. But I’m hoping that in the future, there will be more widespread availability for all women having mammograms so that more patients have access to accurate diagnosis.