Two reports—one in the New England Journal of Medicine and the other in the American Cancer Society’s CA A Journal for Clinicians—have been published that will go a long way towards helping patients and doctors make reasoned recommendations regarding the appropriate use of MRI as a screening tool for breast cancer. (Links for these articles were not available at the time this blog was posted.)
We have known for some time that mammography has its limitations, especially in the evaluation of certain groups of women. We have also known that MRI, which is an expensive and occasionally distressing test, can sometimes pick up early breast lesions that might otherwise be missed by conventional mammography, even when done by very competent radiologists.
As good as MRI might be, it also has significant limitations. Chief among those limitations was the fact that MRI picks up many lesions in the breast that turn out not to be cancerous, requiring additional biopsies that may otherwise have been avoidable.
Although the increased risk of what we call a false positive lesion might be acceptable in some women at high risk of breast cancer, it would not be acceptable if applied to the population at large.
As experience with MRI has progressed, and as the techniques became more refined and additional tools such as MRI-guided breast biopsy have become available in some parts of the country, it became appropriate to revisit some of the questions about MRI and its appropriate use as a breast cancer screening tool.
The article and an editorial in the New England Journal discuss recent research in evaluating the effectiveness of MRI in women who themselves have already been diagnosed with breast cancer.
As the authors of the paper note, a woman with a diagnosis of breast cancer in one breast is known to be at increased risk of developing breast cancer in the opposite breast.
If doctors could find that breast cancer at the same time as the original diagnosis in the primary breast (instead of at a later date which is all too frequently the case), then it could mean a single episode of treatment for both cancers, instead of having to come back and retreat a woman at some time in the future when the second breast cancer appeared.
In addition, if the MRI could provide some level of reassurance that no cancer is present in the opposite breast, it may influence some women and their doctors not to do a prophylactic—or preventive—mastectomy on the breast opposite from where the original cancer was diagnosed.
The researchers represented a panel of 25 institutions and private practices that specialized in MRI of the breast.
They examined the results of doing a breast MRI in the breast opposite (or contralateral) to where the primary breast cancer was diagnosed in 969 women who had recently been diagnosed with breast cancer.
None of these women had a breast cancer detected by mammogram or clinical breast exam before the MRI was done.
The researchers found a total of 33 breast cancers in the contralateral breast with MRI within 1 year after entry into the study. 30 of these were the result of a positive breast MRI examination. Three were not found on MRI: one in a woman who had a biopsy of a lesion that was reported to be probably benign on the MRI, and two in women who had a prophylactic (or preventive) mastectomy of the contralateral breast where the MRI was read as negative for cancer.
All three of these cancers were ductal carcinomas in situ, which means they did not invade the breast tissue and were very early cancers. They measured 1, 3 and 4 mm in diameter (There are 254 mm in an inch, so they were very, very tiny cancers.)
Looking at the data another way, the doctors recommended a biopsy based on a positive MRI scan in 135 women (or 13.9%), and 121 had the biopsy performed as recommended. Of these, 30 (24.8%) of the biopsies showed cancer. 91 of the lesions biopsied were benign.
For the 30 cancers diagnosed by MRI, of the 27 were information was available, none had lymph node involvement. The average diameter was 1.09 cm (again, an inch is 2.54 cm). 96.7% of the cancers were stage 0 or stage 1, which have an excellent prognosis, and one cancer was larger and was stage 2.
The authors concluded that their study showed that MRI can improve the detection of breast cancer in the contralateral breast in a woman who has just been diagnosed with breast cancer, even when a mammogram and clinical examination is negative.
The editorial which accompanied the article in the New England Journal (written by Dr. Robert Smith, who is my colleague at the American Cancer Society) noted the progress that has been made in the early detection and treatment of breast cancer over the past 14 years from 1989 through 2003.
Dr. Smith also emphasized that to a woman with recently diagnosed breast cancer, the knowledge that there is a low likelihood of breast cancer in the opposite breast is also a matter of great importance.
The editorial extends the discussion, however, to another topic that is also crucial when it comes to the use of MRI in the early diagnosis of breast cancer, and that is making certain that this valuable test is used only for women for whom it may have the most benefit.
In short, this is not a test to be applied routinely to all women at this time in place of a mammogram.
Perhaps you have seen some of the ads advertising breast MRI imaging centers as being superior to mammography. This is probably more common in large cities like
There are several criteria that should be applied to an MRI center to determine if it is a quality operation.
For example, if a center says it can perform a breast MRI, it should also have the capability to do an MRI guided breast biopsy if a suspicious lesion is found. If the center doesn’t have that capability, it is probably in your interest to go elsewhere.
Dr. Smith, in his editorial, emphasizes the need for standardized quality programs to assure women that the center performing their MRI meets certain recognized standards, much as was the case a decade ago in setting standards for mammography to assure at least a baseline level of quality for that procedure as well.
The New England Journal article focuses on one very important high risk group of women who should consider an MRI for the early detection of breast cancer.
There are many other women at high risk who also could benefit from adding an MRI to their breast cancer screening program—in addition to a mammogram and clinical breast examination once annually.
Until now, however, there have not been any clear-cut guidelines as to which women at high risk are likely to benefit from the addition of a breast MRI once a year.
Simultaneously with the release of the New England Journal article, my colleagues and volunteers from the American Cancer Society published another article which outlined specific recommendations for women and their physicians as to when an MRI should be performed for screening, in addition to an annual screening mammogram.
This report, in the current issue of CA: A Journal for Clinicians, updates our prior recommendations from 2003, when the evidence about MRI screening in breast cancer was not sufficient to make specific recommendations.
The expert panel who wrote the article now makes very specific recommendations for MRI screening for breast cancer.
(The recommendations are too detailed to provide in this blog, but you can find them by going to the actual article.)
Among the groups included where screening is recommended are:
–Women who are BRCA positive;
–The first degree relatives (that is, mother, sister, daughter) of someone who is known to be BRCA positive, but the woman declines to get the test, and
–Women whose lifetime risk of developing breast cancer is 25% or greater, as measured by one of the available breast cancer risk models. These models include the Gail, Claus and Tyrer-Cusick models, as described in the article.
The expert panel also recommended that women who had radiation therapy to the chest as part of cancer treatment (in particular, Hodgkin’s disease patients) when they were between ages 10 and 30 should have an MRI as part of their breast cancer screening program.
There are some situations where there is not enough evidence to recommend for or against the use of MRI as part of a screening program.
These include women where the models mentioned above show a 15-20% lifetime risk of getting breast cancer, where the woman has lobular carcinoma in situ or atypical lobular hyperplasia, or if she has a history of atypical ductal hyperplasia.
Also of note is that the panel was “indeterminate” in making a recommendation regarding women who have dense breasts on mammography, or in women who themselves had a personal history of breast cancer including DCIS (in contrast to the prior article, where MRI was studied ONLY at the time of primary diagnosis).
And, perhaps of considerable importance, is the recommendation that women who have a lifetime risk of breast cancer that is less than 15% should not have MRI included in their breast screening program.
There is obviously much more to this report than can be listed in this blog. For example, there is a discussion whether the MRI and mammogram should be done at the same time, or staggered every six months (mammogram on day 1, MRI in 6 months, mammogram in one year, and so on).
In addition, there is a very detailed discussion about MRI quality and the fact that if the MRI is not done right by someone who has considerable experience doing breast MRI, it should not be done at all.
This discussion has very practical implications.
A couple of weeks ago I had a conversation with a friend whose young wife had just been diagnosed with breast cancer. One of the topics we reviewed was whether or not she should have an MRI of the breast not involved with the cancer to see if another tumor was present in that breast as well.
This past week, during a casual conversation with another friend, the topic of breast cancer risk in his wife, who was a Hodgkin’s disease survivor, came up. Again, the question was raised whether or not she should have an MRI as part of her routine surveillance for the early detection of breast cancer.
Both of these discussions highlighted what we know and what we do not know about the use of MRI in the early detection and screening for breast cancer.
Until now, we simply have not had good evidence upon which we could make reasonable recommendations to women with these types of histories, and many other women who had similar questions about the use of MRI as a screening tool for breast cancer.
Now, with the evidence at hand, I can tell my friend that his wife with recently diagnosed breast cancer that she should have an MRI of the opposite breast before she undergoes her breast cancer surgery (she did, and it was negative).
And my other friend’s wife—who had Hodgkin’s Disease and is now in her 50’s—should also consider adding an MRI to her annual screening.
We still don’t have all of the answers, and both articles point out the issues that remain to be clarified.
But we are much further along in developing the evidence that will guide us in making the best recommendations we can to women and their doctors on how to use an important imaging technique for the right women at the right time, when it counts the most in reducing the potential burden and suffering from breast cancer.