An article released today by the British medical journal The Lancet suggests that MRI is more effective than mammograms in detecting early, non-invasive breast cancer lesions. (I could not find a link to this article when I posted this blog.)

The implications of the article and an accompanying editorial are that we could do a much better job of finding breast cancer earlier in its course, and save more lives from breast cancer than is currently possible by relying on screening mammography alone.

This is clearly going to be a controversial issue, given our reliance on mammography as the best available breast cancer screening tool for women at average risk.

For women at a high lifetime risk of being diagnosed with breast cancer, the American Cancer Society in March released a guideline suggesting that these women should have breast cancer screening with MRI in addition to regular mammograms.   However, the Society did not find MRI to be useful in women at average risk and recommended against its use in that circumstance.

The situation is further complicated by the fact that the number of centers nationwide who have the equipment and expertise to accurately read and follow-up on screening MRIs is very limited.   This is a point on which both the Lancet researchers and the authors of the Society’s guidelines agree.

What did the Lancet study report?

The theory behind the German researchers’ study was that we are applying the wrong criteria to MRI screening for the purposes of diagnosing early, non-invasive breast cancer (called DCIS, for “ductal carcinoma in situ”).

Traditional film or digital-based mammograms rely on patterns of calcium deposits in the breast to indicate that a suspicious lesion may be present.

As the authors point out in their study, these calcifications are the result of a cancer growing, then cells dying, and the “debris” from that process becoming calcified in the breast.

They go on to say that most studies of MRI in the screening of breast cancer rely on finding the same calcifications on the MRI that might be seen on the mammogram.

The key, however, to finding early breast cancers on MRI should rely not on detecting calcium deposits, but on finding patterns suggestive of increased vascularity, or blood vessel growth and blood flow, say the researchers.

When looked at this way, the authors concluded, the MRI has a much better chance of finding abnormal (possibly cancerous) lesions in the breast.

When they applied these criteria in a study where women had both an MRI and a mammogram—and where the radiologists who were reading the films were not aware of the results of either study—the doctors were able to find many more early DCIS lesions with the MRI than they were with the mammogram. 

This held true for all groups of women who participated, including women who were at high risk of breast cancer, women who had previously been diagnosed and treated for breast cancer, and women who simply were anxious about the risk of breast cancer but had no predisposing family history or other risk factor which suggested they had a greater chance of developing the disease.

In addition, according to the study, the MRI was much more successful than the mammograms in detecting DCIS lesions that were considered “high grade,” which were those more likely to progress to more typical invasive cancer at some time in the future.

According to the authors, these are the lesions that we should be looking for.  If we found these lesions in greater numbers, they suggest, we would substantially improve the outlook for women who have breast cancer. 

In other words, find the high-risk non-invasive cancer before it becomes invasive.  The treatments would be simpler, and the outlook much better.

Interestingly in this study, and perhaps unexpectedly, routine mammograms were not as successful as MRI in finding these particular high-risk DCIS lesions. The difference in the ability of MRI to find these early cancers was substantially greater compared to mammography.

The implications of the study are clear: if we are really going to make an impact on reducing the incidence and burden of invasive breast cancer, we must do a better job of finding these high grade DCIS lesions by using MRI routinely in the screening of all women for breast cancer.

But the authors were careful to point out that there were some significant cautions that had to be considered before recommending such a significant policy change.

First, they noted that their study was done in an institution by highly trained breast imaging radiology specialists with years of experience who had access to the best equipment.  This is not representative of the circumstances in most parts of the United States or the world.

In their words, “For the time being, few radiologists can offer a level of expertise for MRI that comes close to that required for diagnostic mammography…  Additionally, there are currently no standards that would define appropriate technical requirements for breast MRI.  Therefore our results are unlikely to be reproducible in a community breast imaging service at present.”  (Emphasis mine)

The authors go on to recommend that a large clinical trial is necessary to further investigate whether or not MRI can indeed find more high-grade DCIS lesions compared to regular mammography, and whether that translates into a better outlook for women with breast cancer.

In the editorial published with this paper, another breast-imaging researcher reaffirmed the importance of the results of this trial.

“It is widely believed that mammography is more sensitive in detecting DCIS than is MRI.  However, Kuhl found that the sensitivity of MRI for DCIS is much higher than that of mammography, especially for high-grade lesions, which are thought to be more prone to progress to invasive carcinomas,” wrote the editorialist.

The authors also noted that the ability of MRI to predict whether a suspicious lesion seen on a study was cancer (called the positive predictive value) was very similar to that of mammography, while the ability of MRI to pick up the high-grade DCIS lesions was much greater.

In other words, in contrast to current thinking which suggests that MRI leads to many more biopsies that turn out to be benign compared to mammography, by changing the criteria for biopsy the predictive value of both studies is almost the same.

The editorial concluded, “MRI should thus no longer be regarded as an adjunct to mammography but as a distinct method to detect breast cancer in its earliest stage.  A large multicentre breast-screening trial with MRI in the general population is essential.”

Before you go to your doctor and demand a screening MRI (which neither the article nor the editorial would support for the reasons noted), you should know that other experts have some qualms about this study.

Debbie Saslow, PhD, who is the American Cancer Society’s director of breast and gynecologic cancer, has a number of thoughts that are worth noting.

Among them is the fact that we already know that MRI is better at detecting DCIS in screening high risk women and follow-up of women with abnormal mammograms or abnormal clinical findings.

The study only looked at 29 women who were at average risk, according to Dr. Saslow.  Those are the women who represent the overwhelming number of those who have screening mammograms.  Therefore, it is difficult to extrapolate these findings to the large number of women who receive screening mammograms.

One other consideration discussed by Dr. Saslow was the reality that we have to consider not only limited availability of qualified centers and radiologists who are able to perform and interpret MRI breast screening studies, but we also have to be concerned about costs. 

The reality is that an MRI cost is about 10 times that of a mammogram.  Multiply that by millions of women, and the costs become prohibitive.

Dr. Saslow concludes that for an average risk woman, the harms of MRI outweigh the risks, and there have been no studies which have assessed MRI screening of women who were not at high risk.

As I mentioned, there is certain to be discussion and controversy over this study.

As is frequently the case, other experts will weigh in on the issue.  It is even possible that perhaps a larger scale study will be done to look more closely at the question of whether MRI breast cancer screening is more effective in finding early cancers, and can reduce treatment costs and side effects as well as increase survival from breast cancer. 

For now, this remains simply a report of an interesting clinical trial. 

There is much yet to be learned about the use of MRI in routine breast cancer screening.  This is not something that is ready for implementation today.   We need larger trials and experience with this approach. 

Much more needs to be done, including improved certification of centers and radiologists who offer this service.

On those points, all of the experts agree.