Doctors strive, struggle for better dense-breast guidance

Cancer-surveillance researchers published a study in late August that they hoped would clarify the long-term benefits and harms of mammography screening with and without supplemental breast MRI scans — in particular for the roughly 50% of women whose breasts are clinically characterized as dense.

The study was a simulation that compared 60 screening strategies, with variables of mammography type (digital mammography and digital breast tomosynthesis), supplemental MRI, breast density, age of screening initiation (40, 45 o 50), and screening interval (annually or biannually).  

The simulation suggested that women with extremely dense breasts (the highest 10%) benefit most from, and are the best candidates for, additional MRI screening.

One of the coauthors was Dr. Katy Lowry, a radiologist and associate professor at the University of Washington School of Medicine. In a conversation, Lowry explained the paper’s key findings and also acknowledged stakeholders’ struggle to achieve a balance between reducing breast-cancer deaths and causing harms to patients: false positives, callbacks, biopsies and costs that ultimately are found to be unnecessary. 

This conversation has been edited for clarity and length.

Q: What percentage of women have breasts that can be characterized as dense? 

Lowry: About half. Ten percent of women have extremely dense breasts and 40% of women have heterogeneously dense breasts. The roughly 50% of women with dense breasts have a modestly higher risk of breast cancer, and mammograms don’t perform as well as in women in non-dense breasts. 

Q: What is the current mammography guidance for this higher-risk population?

Lowry: The exact wording varies from guideline to guideline, but most professional societies recognize that mammography does not perform as well for women with dense breasts, and that these women are at higher risk of having false-negative mammograms. Because of that, they may want to consider additional screening such as MRI or ultrasound, and should be supported and counseled on the pros and cons. So it’s a little bit of a softer recommendation than, say, the recommendations for MRI in women at very high risk due to their genetics or family history.

Q: What were authors hoping to learn or clarify with this study?

Lowry: We know this is a challenging decision for a lot of women, and we wanted to provide some estimates of the risks and benefits of supplemental MRI screenings for each density group, to hopefully help patients as well as providers trying to counsel patients about screening options.

Q: Could you describe your team’s findings?

Lowry: Not surprisingly, we found that the more intensively you screen a woman, whether it's more frequent mammograms or added MRIs, the more lives you save, but there are also more downsides: More women who don’t have cancer also get called back for additional imaging and biopsies. Those are false positives. And that's always the case with breast-cancer screening. The more we look, the more we find cancer and things that are not cancer.

We did find, though, that the tradeoff seems to be more favorable when you limit MRI screening to that 10% of women with extremely dense breasts. There we tend to see tradeoffs that are not so dissimilar from what we see with routine mammography in the average population.

Q: When you say ‘mammography,’ do you include 3-D breast tomosynthesis?

Lowry: We looked at both digital breast tomosynthesis and digital mammography in this paper. Tomosynthesis is accessible in about 90% of radiology facilities across the country. But quite a few women still get digital mammograms.

With tomosynthesis, the advantage is fewer false positives for the same mortality benefit. But the benefits of MRI for women with dense breasts are similar, whether they are getting screened with digital mammograms or tomo.

Q: It seems like radiologists are frequently sifting data to find some new clarity or an epiphany about which breast screenings make sense at what age and for which densities. Does it ever feel like the information you want is elusive?

Lowry: Yes. It's an inherent challenge because it's often not clear. How much screening is too much? How much is not enough? We don't have a guide that says, ‘We’re willing to do X number of biopsies to save one life.’ That doesn't exist.

That's why a screening policy is often controversial, because you’re trying to achieve a balance of benefits versus harms, and wherever you draw the line, some will think it's too much and others will think it’s not enough. 

But we do these analyses because sometimes patterns do emerge that more easily support one recommendation over another. For example, in this paper, we found that doing annual mammograms and MRIs in women with extremely dense breasts did not increase the mortality benefit that much, but increased false positives quite a bit. So for women with dense breasts but without other risk factors, it might make sense to do MRI every two years instead of every year.

Q: Are women in that 10% of extreme breast density grateful to have access to MRI, or bothered that it is necessary, or both?

Lowry: It's variable. Some women are very motivated to have MRI so they can know they've done everything possible to reduce their own risk of a late stage breast cancer. Some women find the MRI procedure very unpleasant: it requires lying face-down in a loud, narrow tube for about 20 minutes. On the other hand, it doesn’t have the discomfort of breast compression.

Q: What are the downsides of MRI screening for breast cancer? 

Lowry: With mammography, if we have a callback, we can bring the patient in for additional views or do an ultrasound to try to resolve the callback without resorting to a biopsy. With MRI, we're more limited to the information from the scan, so we end up doing more biopsies for abnormal findings compared to screening only with mammography. And the biopsies are often MRI-guided, which are more challenging for patients than ultrasound-guided biopsies and biopsies guided by mammography. 

So having breast MRI is a great capability because it is really good at finding cancer, but it’s not without consequence. The bigger issues, though, are cost and equity. It is substantially more expensive to have an MRI than a mammogram. Depending on your insurance, patients might have a pretty hefty copay.

Q: It seems unfair to financially penalize women who happen to have dense breasts.

Lowry: I think having large out-of-pocket costs for anything that we deem medically necessary is a problem. If we demonstrate that MRI screening is medically appropriate for a patient, then it is unethical to expect them to foot a bill that might impact their ability to pay a mortgage or take their family on vacation. That propagates disparities, because patients with more financial means are more willing to accept those expenses and get the care they need.

Our study wasn’t designed to address factors like access and equity, but these are very real factors that have a big impact on the feasibility and the real-world effectiveness of these guidelines.

For details about UW Medicine, please visit http://uwmedicine.org/about.