Dr. Daniel Kopans

Dr. Daniel Kopans - the father of breast tomosynthesis

November 30, 2014
by Lauren Dubinsky, Senior Reporter
Dr. Daniel Kopans, a professor in the department of radiology at Harvard Medical School and the director of breast imaging at Massachusetts General Hospital and widely considered to be the “father of breast tomosynthesis,” spoke with HCBN.

HCBN: What role do you see tomosynthesis playing in the future? Will it be the “go-to” technology for breast imaging or will it be a specialty?
I invented Digital Breast Tomosynthesis (DBT), and my group at the Massachusetts General Hospital led by Richard Moore, Dr. Loren Niklason and Dr. Tao Wu developed it because one of the major problems for radiologists is the superimposition of normal breast tissues on 2-Dimensional mammograms. Breast cancers can be missed because they are hidden behind normal breast tissues, and normal breast tissues can superimpose on one another forming “summation shadows” that look like cancer but are not.

Clearly, failing to find small cancers because they are hidden is an important problem. Since summation shadows are responsible for 25 percent of the women recalled from screening for additional evaluation, eliminating these reduces the recall rate (called false positives by some). DBT increases the detection of small invasive breast cancer (the ones we want to find) by approximately 30 to 40 percent while reducing the need to recall women from screening for additional evaluation by approximately 20 to 40 percent.

Usually, the effort to find additional cancers involves lowering the threshold of concern (increasing the recall rate and lowered “specificity”). The benefit of DBT is that it allows for the detection of more small cancers (increased sensitivity) while also decreasing the recall rate (increased specificity) — a true “win win” situation. DBT will replace conventional 2-D mammography simply because it is better than 2-D.
Once the need to obtain full exposure 2-D images while obtaining DBT images (double dose) is eliminated by the adoption of synthetic 2-D images, there will be no reason to not adopt a test that finds more cancers with a lower recall rate. Having its own billing code will eliminate that barrier to adoption. Missing a cancer because DBT was not used for screening will become a legal liability once it becomes the standard of care.

HCBN: In your opinion, how long will it take for tomosynthesis to reach its maximum market saturation?
Barring some unforeseen impediment, now that GE Healthcare has FDA approval, I suspect that adoption will move quickly. Radiologists were slow to adopt Digital Mammography because we recognized that it was really not much better than 2-D mammography at finding cancers. The Digital Mammographic Imaging Screening Trial was not very convincing. I changed us to all digital at the Massachusetts General Hospital because we could (the funds were available); the entire Radiology Department was becoming Digital; and the logistics of Digital made sense (no more film files to manage; simultaneous viewing of the images; etc,).

HCBN: Other than reimbursement, are there any other challenges standing in the way of wide-spread adoption?
I do not see any major impediments. Most of the opposition to DBT that I have seen is from those who have never used it. I have not met anyone who would return to 2-D mammography once they have experienced its advantages.

It takes getting used to the fact that it takes longer to review a screening DBT study, but it took awhile to get used to CT instead of plain X-rays of the chest and abdomen, yet we recognized that CT was superior and we now, routinely, review multiple slices instead of two “plain films.” In addition, since DBT leads to fewer recalls and fewer “diagnostic” evaluations, the considerable time saved in not having to do diagnostic imaging can be cost accounted to screening DBT reducing much of the extra time involved.

HCBN: Are there any improvements for breast tomosynthesis on your wish list?
Just as 2-D mammography screening is not the ultimate answer to breast cancer — it does not find all cancers and does not find all cancers early enough to result in a cure — DBT is not going to end all deaths from breast cancer.

DBT is a major advance, but there will still be cancers that are hidden even on DBT and others that are metastatic before they can be detected. The death rate from breast cancer has been significantly reduced since 1990 by the introduction of screening and earlier detection.

DBT will help to further drive down the death rate. There will still be some cancers that are not visible on DBT, but are detectable using ultrasound and certainly, using MRI. Our immediate goal at the MGH, actually one that we have hoped to do for many years, is to try to incorporate ultrasound into the DBT system so that ultrasound and DBT can be acquired together. Having registered images should allow for maximizing the detection of cancer while further improving specificity.