Breast Screening Continues to Advance at the Radiological Society of North American (RSNA) Meeting
Debra Monticciolo, MD: Hello. I'm Dr. Debra Monticciolo, President of the Society of Breast Imaging and Chief of Breast Imaging and Vice Chair for Research in the Department of Radiology at Scott and White Healthcare and Professor of Radiology at Texas A&M Health Science Center. I have with me Dr. Phil Evans, the Immediate Past President of the Society of Breast Imaging and a Clinical Professor of Radiology at the University of Texas Southwestern Medical Center. He also holds the George and Carol Poston Professorship in Breast Cancer Research and was recently installed as the President of the American Cancer Society. We're here at the 2011 meeting of the RSNA to discuss the most recent developments in breast imaging. So Phil, what have you seen at this year's meeting that you find really important?
W. Phil Evans, MD: Well, there have been lots of things, Debbie, that I think have been quite important. This year there has been a big emphasis on nuclear medicine techniques, on tomosynthesis, on contrast-enhanced spectral mammography, and even on genomics and imaging. There are going to be some new advances in MRI. We'll talk about some ultrasound that was discussed as well.
A Federal Law for Breast Ultrasonography?
Dr. Monticciolo: So let's talk first about screening with ultrasound. There have been several screening sessions, and there are research papers on ultrasound. Do you have any comments on that?
Dr. Evans: Screening breast ultrasound has really gotten a lot of play since the ACRIN 6666 trial came out and was completed. Of course, that trial found that you could, in conjunction with mammography, find more breast cancer using screening ultrasound and mammography together than you could just with mammography alone. So several states now have laws concerning breast density. Texas is one, and Connecticut was the first. Letters to patients tell them that if they have dense breasts they may need to have supplemental screening. I think it has been kind of assumed that breast ultrasound would be one of the ways that that would come out in terms of screening. There were discussions about the new laws that have come into effect. There is also a possibility of a federal law coming into effect. It's going to be very interesting to see how that all plays out because even though you can find more breast cancer with breast ultrasound, we all know, I think, that if you really want to find the most breast cancer you can find it with MRI in addition to mammography.
Dr. Monticciolo: Yes. My concern is we all like ultrasound because there's no radiation to the patient. It's very accessible. We all have it and we're all using it, and we're all looking for more uses of this technique. The problem is that we're finding a lot of things that aren't cancer, and so we do have this supplemental yield but the positive predictive value for biopsy is very low. Even in high-risk women, we know that 90% of the biopsies are going to be benign. We're generating a lot of biopsies, and the question is, will that ultimately be good for our patients? I know there is a lot of push to use this technique. It's just a question of the risk-benefit ratio for the patients. No one has proven that it has really made a difference in life. I think we're going to have trouble proving mortality benefit, but you're correct. If you really want to find the most cancers most accurately, MRI is the way to go.
Dr. Evans: I think it's also interesting that there has been so much evidence over the years about screening mammography and how successful screening mammography is in reducing mortality from breast cancer, and yet there's a lot of discussion in the world these days about whether or not mammography is useful. I think there was a study here at the RSNA done in Rochester, New York. In one of the practices there -- a very large mammography practice -- the investigator looked at women between the ages of 40 and 49 and compared those with a family history of breast cancer, a mother or a sister, and those without a family history of breast cancer. This was over about a 9-year period, and they found almost exactly the same number of invasive breast cancers in the women with family history as those without family history. So it just goes to show that even the age 40-49 continues to be very important because we know also that in terms of years of life saved, women who have breast cancer in their 40s benefit the most. They comprise 40% of the years of life saved, which is a very important number.
Dr. Monticciolo: Yes, and it's interesting that mammography is probably one of the most tested modalities anywhere in medicine. It has got a proven mortality benefit and still we seem to catch a lot of fire. The controversy just never seems to go away. I think most of the issues now are really related to cost, but no one wants to admit that. Most of us, I think, would feel more comfortable having patients know how well mammography works and tell them, "Well, we just can't afford it," which is what the government should be saying, rather than that it doesn't work. So it's a real problem for us. I don't know where we'll end up going with the supplemental screening because that's an added cost as well. Mammography is the only modality that has proven efficacy and proven mortality benefit. So tell me, have you seen some of the papers on automated full breast ultrasound?
Automated Ultrasonography and Molecular Breast Imaging
Dr. Evans: I have seen some of the papers, and I think that there are questions about whether one can use automated breast ultrasound. I still think that that remains unclear. All the studies have been done either using technologists with hand-held devices or physicians using hand-held devices. In some of the newer studies with automated breast ultrasound, I think you have seen recall rates that are pretty high.
Dr. Monticciolo: Yes, it's really concerning. This is one effort to try to make ultrasound easier to use in practice because it is automated, but there was a study showing that 40% of the recalls from automated ultrasound -- almost half-- disappeared when hand-held was used. So these were not real abnormalities. It certainly is going to need a lot more study before we can go forward.
Dr. Evans: I think one of the more interesting things at this meeting had to do with molecular breast imaging.
Dr. Monticciolo: Oh yes.
Dr. Evans: This has been mainly used at the Mayo Clinic, and it has been used in conjunction with mammography in women with dense breasts. One particular paper that was presented showed a reduction in the dose. They had been giving 20 millicuries, I believe, of technetium-99m sestamibi, and now they're using 8 millicuries. They're even using a format where they can look and see what the images would look like if they had used 4 millicuries. Apparently when you get down to 4 millicuries that's pretty close to what you would get with a screening mammogram. They were finding many more cancers with molecular imaging than with screening mammographies. That is something to look forward to in the future to see how it actually pans out.
Dr. Monticciolo: Yes, and we all want to find more cancers, so we're always interested in technologies that suggest we can do that. My concern with the nuclear techniques is that they are trying to get the breast dose equivalent or at least close to mammography, but the breast is not the critical organ for either PEM [positron emission mammography] or BSGI [breast-specific gamma imaging]. The whole body is getting radiated, and that's an issue that's not being addressed very directly in the reports, and the patients need to know about it. The dose to the colon or the bladder, depending on which pharmaceutical you use, is quite high. It's certainly going to be better if it's less than 20 millicuries, but it's pretty high at 20 millicuries so they really need to cut that down. I'd be interested to see what that does with the whole-body dose and the critical organ dose because in my mind that's what limits the routine use of these techniques. You really can't use them year after year.
Dr. Evans: Yes, I think that's an excellent point, and I think that's something that really needs to be studied in more detail.
Dr. Monticciolo: We had some interesting papers on the use of MR for preoperative evaluation, and it has come under a lot of fire. I found that interesting. There certainly needs to be more literature because right now surgeons are giving us a little bit of a hard time using MR, saying that we don't have enough data showing a decrease in either recurrence or reexcision rates. There were several papers in a session earlier this week showing that MR did decrease recurrence rates, both ipsilateral and contralateral breast, as well as decreased positive margins and reexcision rates, particularly in DCIS [ductal carcinoma in situ] but in invasive cancer as well. I'm looking forward to seeing the published papers so we can look at the methodology thoroughly. Right now, the surgeons are saying, "Boy, we don't have the data," or "We have data that shows it's not good." I've reviewed that literature. I don't think it's very strong. It's not robust. We only have 1 randomized, control trial, so these papers really do need to find their way into print so that we can use this information.
Before we wrap up, I wanted to ask you a little bit about the concept of overdiagnosis because that's discussed quite a bit and it's gaining traction in the media. Give us a few words on that if you would.
Dr. Evans: Well, "overdiagnosis" is an interesting word because what it means is that, say you biopsy a cancer, you take the tissue, and you look at it under the microscope. It looks like cancer, but if you had followed that along for many, many years it may actually have done nothing to harm the patient in any way. That is considered overdiagnosis because you biopsied and treated something that wouldn't hurt the patient. At this particular stage, we don't have any way of knowing, when we have histologically detected cancer, whether 'it would harm the patient or not. We assume that because it has shown up on the mammogram and appears to be growing because it's there and it wasn't there before, that it' is going to harm the patient. This has to do with the genomic aspects of breast imaging.
So I think we know just from basic experience that anybody in breast imaging who has followed cancers or seen them over the years knows that cancers that won't do anything to harm the patient are very, very few in number.
Dr. Monticciolo: Have you ever seen one?
Dr. Evans: I have seen cancers that have been stable for a long time but then all of a sudden they developed and did something to really, really change dramatically.
Dr. Monticciolo: Have you ever seen one regress?
Dr. Evans: I've never seen one get smaller, but I've seen them get larger over time. I'm sure they grow at different rates. I'm sure that many of them, almost all the ones we see, can be devastating to people after a period of time. There may be some cases of low-grade DCIS, which are curative with surgery, that may never progress, but we certainly know that high-grade DCIS can be a devastating disease and can kill the patient because it will progress to invasive disease. I would say that probably 1% -- less than 1% -- of breast cancers fall into the overdiagnosis category. It seems to be much different from prostate cancer, where there may be a much higher percentage that never kills a patient. So this is something that I think is going to bear further research, but I think breast cancer and prostate cancer are very different. I think overdiagnosis with breast cancer is very limited if at all.
Dr. Monticciolo: I agree with you. We had a discussion, a group of leaders in breast imaging through the Society of Breast Imaging, and collectively there wasn't anybody who had been in practice less than 20 years. Several of our leaders have been in practice 30 or 40 years. We talked about the fact that none of us have seen these phenomena en masse. We've seen slow-growing tumors, and maybe a slower-growing tumor in an older patient would not progress to kill that patient. None of us had seen a tumor just spontaneously go away, which is what's seen in the literature that goes along with these claims of overdiagnosis. I'm concerned that it has gotten traction in the media. I think we have to drive home the message that when we see a cancer we need to treat it because we don't know which are going to be the worst players and which may grow so slowly they won't kill the patient. We do know some prognostic indicators, and we know when we see aggressive disease that we have to be extra aggressive. I don't think we can go to the step of, "Well, I'm just not going to treat it and just let it go."
Dr. Evans: No, we certainly can't. There is really such a small percentage of breast cancers that would fall into that category that we have to treat all breast cancers at this stage of the game until we get some definite genomic information that may help us out.
Dr. Monticciolo: True. So would you like to make any other summary comments before we close?
Dr. Evans: Well, I'd just like to say that once again the RSNA has been a fruitful time for us. We've learned a lot. We've seen new things, and we've found that some of the things that we've done for a long time are still good.
Dr. Monticciolo: Yes.
Dr. Evans: But there are always new things that will be coming along, and as each day goes by we'll have new modalities.
Dr. Monticciolo: New opportunities.
Dr. Evans: And new opportunities to find more breast cancers.
Dr. Monticciolo: Yes, and I'll just add that I would like to encourage all women age 40 and older to get an annual mammogram. It's still the best test for finding breast cancer and it's a proven lifesaver. Thank you for joining us. This is Debra Monticciolo and Phil Evans for Medscape Radiology.