The much-anticipated results of the Digital Mammography Imaging Screening Trial (DMIST) were released last week and the findings were largely what was expected: digital mammograms are better than analog x-rays in detecting breast cancer in younger women and those with dense breasts. The study involved 42,760 asymptomatic women who were imaged at 33 sites across the U.S. and Canada since October 2001 and was sponsored by the National Cancer Institute (NCI).
Health Imaging News spoke with the study's senior author, Etta Pisano, MD, of the University of North Carolina at Chapel Hill, about her thoughts regarding the findings and the implications regarding the future of mammography and women's health.
What impact will these results have on the practice of mammography?
I think that more women will be having digital mammography exams, especially the woman for whom the study showed improvements. Unfortunately, they all won't be able to get [digital mammograms] initially, but over time I expect that most women with dense breasts and younger woman and pre-menopausal women will be receiving digital mammography because of our results.
Do you think the results will capture the interest of younger woman and urge them to go for regular screenings?
I hope that most women in their 40s are already getting screened because there is strong data to support the reduction in mortality from breast cancers that screening can provide. Obviously, improved therapies for breast cancer are another part of the reason why mortality is starting to fall. I must say it does worry me a little that if the cost of mammography screening goes up too much that some women will be priced out of it. So the cost-effectiveness analysis is very important. We're going to need to be careful that we don't price-out the uninsured out of mammography completely.
Is screening every two years optimal for women in their 40s?
Well, the recommendations of the American Cancer Society and the American College of Radiology say every other year. There's a lot of debate about this. Mammography is not as effective in terms of mortality for women in their 40s partially because most women in their 40s have dense breasts, but also because breast cancer tends to be more aggressive in younger women. That would argue for doing more screening not less screening for women in their 40s.
Insurance companies probably won't like that, do you think?
If anything, you should cut back on screening as women age because though breast cancer is a more common disease, it moves more slowly and you have more time to find it before it's a serious problem. I'm talking about women 70 and up.
There are still some pending results of DMIST having to do with the cost-effectiveness of digital and film technologies and the effect on participant quality of life due to the expected reduction of false positives. When will these data be available?
The cost-effectiveness analysis is pending and we have reader studies that we ran. We did a lot of other research that hasn't been published yet. We're probably going to be publishing as many as 10 to 20 papers out of the results. This is the primary analysis though, this is not preliminary; this is all we're going to say about the 50,000 woman study. Probably in the next year everything will get published, some sooner than others.
Will any additional research come out of the study?
We're working on getting a little more money because we'd like to do an analysis of the cancers that were missed to try to figure out why they were missed.
CAD played no part in this study, but what role do you see CAD playing with analog and digital mammography? Are you, or other researchers you know of, doing a large study on the effectiveness of CAD? What's your perspective on the prospect of such research?
CAD was not involved in this study. I don't think the literature on CAD and digital mammography is very strong. I don't think there's been much published at all about how effective CAD is for digital mammography. There's a lot of data on CAD for film mammography which is fairly convincing to me about its efficacy especially for non-expert readers. In terms of the future, I'm not personally doing any CAD research and ACRIN [American College of Radiology Imaging Network] is not. As far as I know, there are no large clinical trials of CAD going on that are funded by anybody but industry.
I think there's room for research in the area but I'm just not aware of any. It would be useful