There’s no question that getting annual mammograms saves lives. Mammograms can detect potentially cancerous changes in the breast before warning signs or symptoms such as a lump appear so patients can be diagnosed earlier, when the chances of long-term survival are at their highest.
However, mammograms are not flawless. For women with dense breasts, which contain more fibrous and glandular tissue than fat, a cancerous growth or tumor can be hard to see and may be missed altogether. It’s a major area of concern in breast cancer research, and the need to develop screening improvements has become even more urgent in light of recent news that more young women—who are more likely to have dense breasts than older women—are developing the disease.
To mark National Mammography Day, Sadia Zapp, BCRF’s managing director of communications, sat down with Dr. Christopher E. Comstock, a BCRF investigator since 2019, to discuss current tests that help supplement mammography, who needs them, and a promising new screening approach he’s studying now.
Sadia Zapp: Hi, Dr. Comstock, thank you so much for joining us today.
Dr. Christopher Comstock: Thank you. Thank you for having me.
SZ: I'm so excited to speak with you because I think this conversation has the potential to not just be enlightening but also super helpful for a lot of our viewers today. I was sharing earlier that I was diagnosed with breast cancer last year at the age of 36, and I completed active treatment earlier this year. So I'm really excited to talk to you, especially because I was diagnosed at a pretty young age. But first, please introduce yourself and tell us about your work.
CC: I'm a breast imaging radiologist at Memorial Sloan Kettering Cancer Center. I've been just doing breast imaging and breast imaging research for nearly 27 years. And I've been all over—I was at the University of Chicago and Northwestern and then I was section chief at UC San Diego. I then worked my way up to Sloan Kettering, and I've been here for 15 years.
SZ: Will you talk a little bit about your BCRF-supported research?
CC: First, it's so critical for researchers like me to have the support of BCRF. And [BCRF co-founder] Dr. Larry Norton, who I work with…he's seen the power in my particular interest in contrast mammography. He's been instrumental in making this happen and supporting the work.
So we've been doing analog- or film-based mammography for quite a while, and then we moved in the mid-2000s to digital mammography, which is a little bit of an improvement. So mammography has evolved, but it's been little incremental improvements—not vast improvements in screening, detection. And now the latest thing we have is 3D mammography that, again, is a bit of an incremental improvement.
I think the biggest issue with mammographic screening is the underdiagnosis of cancer. We've had eight randomized control trials since the 60s that have shown that patients who mammographic screening, compared to women who don't undergo screening, have anywhere from 20 to 40% reduction in mortality from breast cancer. But surprisingly, the test that's used for screening mammography is probably around a 20% sensitivity.
So it's not a great test, but it's fit the bill over the years because of its availability and low cost. But it’s surprising that such a poor test has managed to reduce mortality that much. What my interest is is something called vascular-based screening. It includes contrast mammography and MRI, which allow you to see the early changes of cancer from the vascularity that occur before you see a tumor that's big enough to detect on mammography and ultrasound. So vascular-based screening can really improve sensitivity.
Despite improvements in therapy over the last 20 years, we still have over 40,000 women dying annually from breast cancer. So, I think the quickest improvement and easiest way to reduce mortality significantly is to find more cancers earlier. And vascular-based screening, such as MRI and contrast mammography, allow us to do that.
The study I’m working on, CMIST [Contrast Enhanced Mammography Imaging Screening Trial] is supported by the BCRF. The trial is comparing our best mammographic screening, which is 3D mammography or digital breast tomosynthesis, with contrast-enhanced MRI to show how many more cancers it detects. Our early, single-institution data looks like it doubles the cancer detection rate.
SZ: That's pretty incredible. When we're talking contrast-enhanced and you say things like “vascular,” what does that mean?
CC: A mammographic screening or 3D mammography screening is just an X-ray of the breast. It's basically a shadow, though 3D mammography adds a bit more information in terms of a 3D view through the breast. But it's not true 3D like an MRI or CAT scan.
A plain X-ray or ultrasound relies on seeing a mass in the breast tissue by its edges or its shape. And some women have more fat-dense breasts, which you can see through easily. But others have more fibroglandular tissue, and you can't see through that as well. So in women with more fibroglandular tissue, it's like looking for a snowman in a snowstorm—the more of a storm there is, the harder it is to see things like a mass. That's the limitation of mammography.
But when you have contrast mammogram, as with a [contrast] MRI, you get an injection of gadolinium [a contrast dye] and the MRI can see that area of contrast, and it's not limited by breast density. The same is true with contrast mammography. It's a mammogram, but you get an injection of iodine contrast beforehand, which is what we use all the time with CAT scans. That allows you to see areas that are hidden in the tissue because it's not limited by breast density.
SZ: For a patient, it's an injection. It's really simple.
CC: Yes, and I think access to mammography is much easier compared to MRI.
SZ: We’re talking about improving technology but without having to roll out hundreds of new machines across mammography centers.
So we've talked about mammography, MRI, and ultrasound. Can we talk a little bit about why and when you choose a particular screening for a particular patient?
CC: How do patients and referring physicians decide? I think that’s an area we need to work on. It becomes a combination of education of the referring physicians of the pros and cons of each test, and also patients, who may have preferences for different tests.
In general, you want a program where you have tailored screening based on the patient's risk, their breast density, and the level of sensitivity they want, since more sensitive tests may result in some false positives.
There are four categories of breast density: A, B, C, and D. For the top two categories, C and D, where breasts are extremely dense, the mammographic sensitivity is going to drop. So those patients should have some supplemental screening, either a contrast mammography or MRI.
So, the decision is based on your breast density, but also on your risk of breast cancer. Let's say you don't have any risk factors and your breasts are not dense. You're probably okay with just mammographic screening or mammography plus ultrasound. If you have dense breasts and you're at intermediate risk, you probably would want to do just contrast mammography. And then if you’re really high risk, we recommend MRI. If you're a gene carrier, which has the highest risk, you may alternate between contrast mammography and MRI.
It's still a work in progress, but the general idea is a combination of your personal history, family history, and genetic models to assess your risk along with your breast density. Also, there is what patient wants. Some patients may not be comfortable in the MRI if they have claustrophobia, and some patients might prefer contrast mammography.
There are other new methods being studied. Dr. Constance Lehmann, who's also a BCRF researcher, has shown that with AI (artificial intelligence) and computer imaging, there's certain patterns in a woman's mammographic pattern that can infer higher risks. I think in the future, we're going to see a whole new type of evaluating risk based on family history but also artificial intelligence, looking at the patterns of your mammogram and combining those to determine what should be the best screening program for you.
SZ: Now we're talking about machine learning to not just detect cancer but also predict future cancer as well. I want to talk a little bit more about the basics of dense breasts, and then we'll get into risk-based screening and what we're hoping to do there as well, and the role that CMIST plays in that.
CC: It's interesting that there's no formalized pathway in breast imaging or in radiology…for adopting new technologies it's kind of hit or miss. So the goal of CMIST is to get some basic phase two data and major publication to really foster adoption. It takes a multicenter trial and a major publication as well as the actual the community of breast imaging. This is a very hot topic. One of my frustrations is that things change very slowly. We know the power of this technology, but it's really working on the research and the science to try to promote adoption. And that way, your referring physicians and other radiologists see that and decide, Hey, this is something we need to offer to our patients.
SZ: One thing I wanted to highlight that you've mentioned but I want to drive home is the success of contrast enhanced mammography as far as reducing false positives or even false negatives. I read that it could be up to 80% improvement, so we're talking about a huge amount.
CC: Yes, it's important information patients and referring physicians should know, whether there are high false positives and unnecessary biopsies. The first time a person has contrast mammography, you're going to see things that might have been there before, but you don't have a comparison. So that's your baseline, like when a patient gets their first mammogram, and there's obviously going to be a higher callback rate. But once you have subsequent screenings for comparison, those further rounds of testing should show fewer false positives and lead to fewer unnecessary biopsies. We already have a good sense that contrast mammography is going to nearly double—80% to 100%—increase in sensitivity. But we want to show also that it's well tolerated and the false positives that are low.
SZ: We’re seeing an increase in incidence of breast cancer in women under the age of 50 and an 8% increase in incidence in women ages 30 to 39. We've long known that incidence has been on the rise, but that was the first study that really quantified that increase in risk in that younger age group. And they’re particularly vulnerable because they're not being screened annually.
One thing I find fascinating is that with so many of these young women, you'll hear how patients brought the diagnosis to their doctor, or said, “I feel this lump” and the doctors will not recommend mammography. And I've been struggling with that because I don't think it's that doctors don't care—I think a lot of it also has to do with the fact that they have less faith in the accuracy of mammography, right? They don't want to put young patients through that screening. They’re more likely to not recommend screening because young women are more likely to have dense breasts, so there’s a higher rate of false positives and unnecessary biopsies.
I always say that when you hear about biopsies, you think, “Oh, they can't be so bad.” But then suddenly you're a patient who needs one, you get the biopsy, and you get that bill and realize this is really a huge financial burden. So one of the reasons I'm so hopeful about CMIST is that it will improve confidence, specifically for people with dense breasts. And that’s the majority of patients under 40.
CC: Right, and in the 40 to 50 range. When to start screening for breast cancer is an area where there's a fair amount of debate. Some organizations and the United States Preventive Task Force kind of left the starting age up to patients talking to their doctors. But the American Cancer Society and the American College of Radiology have, for a long time, recommended screening start at age 40.
Regarding false positives and unnecessary biopsies, I think some of the modeling and the recommendations from the Task Force factored in the anxiety and stress of having an unnecessary biopsy or being called back. I don't think they give women enough credit because I think women would rather have their breast cancer found than feel anxious about getting a biopsy. Yes, there’s short-term anxiety in getting a biopsy and waiting for results, but I think most women would prefer to undergo it rather than find themselves thinking later, I didn't get screening and now I have this palpable lump that with nodes positive, I wish I had done more screening.
So I think recommendations from at least some of the organizations and, like you say, the referring physicians, their hesitancy is because of some of that information from the Task Force, I think they overused the modeling data on anxiety and the harms of mammography. I think you're right that some physicians are hesitant to recommend screening as often as they should. But I think CMIST may help to show that contrast mammography is a much more sensitive test that has lower false positives. So its utilization, particularly in women with dense breasts, will be a big positive.
SZ: I couldn't agree more. There is no comparing the anxiety of screening versus being diagnosed at a later stage. And I think a lot of young patients, specifically when they go in to talk about risk factors, even that can be a longer conversation to be had with doctors—like you were saying, starting in your twenties just talking, just having that conversation with your doctor to outline what your risk factors are and when you should begin screening is really, really important.
Let’s talk about dense breasts. What are they? The FDA just required imaging facilities to let patients know that they have dense breasts. Once you find out you have dense breasts, then what?
CC: There's something called the Mammography Quality Standards Act, which has helped to standardize mammographic screening and the reports, so it's more clear for physicians what they need to do and if they need to do biopsies.
But breast density, as I mentioned, there's four categories. And the breast is like other organs in the body like the pancreas and kidneys—its excretory. It's made to produce milk during pregnancy. It has tubes and structures in it. Imagine it's like the tree branches—the more leaves you have, then the more dense the breasts are. You could have just the branches and not many leaves. That's the non-dense pattern that's mostly fat density. So you can see through the “branches” very easily. You could see if there's a bird sitting on a branch. But some patients have extremely dense breasts where you can't see anything sitting on a branch anywhere.
I think back in the 60s and 70s, when mammography was rolling out during the American Cancer Society’s war on cancer, which is the time when breast imaging became a specialty and when mammography and breast cancer screening really got promoted. And I hate to say it, but I think we oversold mammography. It's not a cholesterol test where you take a sample and you get a number. Mammography is based on the radiologist’s experience and ability to detect these patterns in the image.
But in the last 10 years—partly through grassroots organization and a website called Are You Dense?—the importance of breast density came through because you had women who had a lump, went to their doctor, were biopsied, and told it was a cancer. And they said, Wait a minute, I had a mammogram three, four months ago.
So there was this grassroots movement to create legislation—it's now a federal, but it used to vary by state—that said, Hey, we need to notify patients if they’re in one of the top two density categories. So now with your report from your mammogram, you get a letter that says if you have dense breast tissue and states it’s a risk factor for breast cancer. And that's because the more breast tissue cells you have, the higher chance of one going awry and becoming cancer. So your mammogram may not be as good at detecting cancer and you should talk to your physician about supplemental screening.
For a long time, these letters didn't specify which test you’d need because ultrasound was the default screening test. But again, ultrasound is based on finding the shapes of things. It's not vascular based. MRI does improve sensitivity by maybe 40 to 50% or 30 to 40%, depending on your risk level. But it's also fairly time consuming, expensive, and has false positives. Still, it turned out to be the default test for women who got density letters and looked into supplemental screening.
At Sloan Kettering, in the breast imaging department where I started 12 or 13 years ago, we didn't do many whole-breast screening ultrasounds. But when the laws changed in New York, we did 30 to 40 a day. Now, it's been done for quite a while. But ultrasound is not an efficient mammography placeholder screening. The sensitivity is still not that great. But with MRI and contrast mammography we can double the detection rate.
SZ: How common are dense breasts? They’re more common in younger women, right?
CC: About half of the screening population falls into the dense breast category and the other into the non-dense category. But proportions change over time. Looking at women in their forties and fifties, more than half of them are in the dense category. As women get older and postmenopausal, you see breast density decrease somewhat.
SZ: Dense breasts are a risk factor for being diagnosed with breast cancer. Is that due to the number of breast cells present, as you said earlier?
CC: It's kind of like lottery tickets, right? The more tickets you have, the greater the chance your number could be called. If you buy one ticket, there’s a lower chance of cancer developing. It's related to the sheer number of breast tissue cells.
SZ: You mentioned earlier that for some women, especially those with dense breasts, they would get a combination of screenings. And you also said the screenings could be alternated. What does that mean?
CC: There's not a lot of data, but in general it’s accepted that some patients, instead of getting both their screenings—let's say they're doing mammography and ultrasound or MRI—that if you spread them out, you do one, and another six months later, you're in essence getting a test every six months. And that approach may find something that was not visible at the time of the previous screening. But a few months later it's seen on the second test.
Again, there are no real studies or data, but some people do alternating screening rather than having the screenings at the same time. For those who are gene carriers, BRCA carriers, for example, they're at such a high risk—you might want to do an MRI and then contrast mammography six months later.
SZ: What are some of the most promising areas of research you're most excited about?
CC: As I mentioned, I think the low-hanging fruit is to reduce mortality in a way that we haven't seen since the initiation of mammography. We really need to work on the detection side of things. In the long run, liquid biopsy would be something you could potentially do at home, but it's not going to replace mammographic screening or MRI or contrast mammography, because it's just not sensitive enough right now. But I think it will have a role.
I think for now the goal is contrast mammography and then liquid biopsies further in the future, where you can detect whether a patient has cancer just by doing a quick blood test. The other thing I would bring up is vaccines for breast cancer. So I think we're coming at screening from multiple fronts, and hopefully within the next 20 years we can significantly reduce the number of women who dying of breast cancer.
SZ: That's what every patient wants to hear, and it leaves me quite hopeful for what the future holds. I did want to touch a little bit on CMIST, which is underway and has started to enroll patients. How soon do you think we're bringing that to the bedside? What's the timeframe that you think—in the next five to 10 years—that contrast enhanced will be available for patients?
CC: I think within the next five years the early results should be out, and probably three years from the first round of screening. Once that's published, I think more and more centers will start offering it to their patients. So I don't think it's a long term. I think it's a short term, five years where we see [this become available].
But with this mammographic screening and contrast, mammography is widely available, so it's very easy for a center to offer that. So I can see a much more rapid uptick in offering that to patients who seek it out and say, I'm intermediate risk and I'd like I'd like something more than just ultrasound.
If I can add here—I've been in breast imaging going on 30 years. And I think we're at a time when we're ready for a new paradigm and screening. We've been doing mammography, we have 3D mammography, and we've done ultrasound, but this is breast cancer screening 2.0. Now that we’re moving into vascular-based screening—and I think women are really ready for something new—we've really said, “Here's a major change in how we can screen.”
CC: And I have to say that [CMIST] couldn't have been done without funding from BCRF. Pretty soon, six sites will be open, and we hope to hope to have about 14 sites, despite COVID delays, FDA delays, and then staffing shortages. Larry Norton has been instrumental. He has the the vision and I think, because he uses it on his patients, he sees the power. And we could not have gotten this trial off and running without BCRF and Dr. Norton and Dorraya El-Ashry from BCRF.
SZ: We'll be waiting with bated breath for sure. We’re very excited to see what the results look like so then contract mammography can be adopted across the country and more patients are able to access it and see better results with screening.
CC: You can have me back on when we have the results published.
SZ: I’d love that. Thank you so much for joining us today.
Please remember BCRF in your will planning. Learn More
Breast Cancer Research Foundation28 West 44th Street, Suite 609, New York, NY 10036
General Office: 646-497-2600 | Toll Free: firstname.lastname@example.org | BCRF is a 501 (c)(3) | EIN: 13-3727250