Dr. Michael Berry
Physicians are increasingly relying on preoperative breast MRI to assess patients’ qualifications for accelerated partial breast irradiation (ABPI), a trend that has generated a great deal of controversy among healthcare professionals. Supporters of breast MRI say it reduces re-excision rates and enables women to make more informed treatment decisions. Opponents argue it increases costs and leads to unnecessary procedures without increasing survival rates.
In our last issue, Dr. Julie Barone of Oncology Associates of San Diego shared her views on breast MRI, including why she would never recommend APBI without one. This month, breast surgeon Michael Berry, M.D., F.A.C.S. of The Breast Clinic of Memphis explains why he thinks MRI is a useful but over-utilized tool.
Q: In your opinion, what role should breast MRI play in the patient selection process for APBI?
Breast MRI is beneficial in certain cases, but it should be used selectively. I understand the desire to identify multicentric and multifocal disease, and it should absolutely be available to patients – but it’s not appropriate for every single breast cancer patient.
MRI is the least reliable breast imaging modality we have, with a false positive rate of approximately 40%. The technology is so sensitive that it frequently overestimates the extent of disease, which can lead to unnecessary mastectomies or needlessly disqualify patients for APBI.
When breast MRI is over-utilized, its impact on the healthcare system is enormous. It increases the cost of medicine, delays treatment and leads to additional biopsies that are rarely positive for additional cancer. And yet we’re not seeing an increase in survival rates. In my opinion, MRI doesn’t change the game enough to justify the strain it causes.
A meta-analysis published last year in the journal CA examined many of the trials utilizing MRI in newly diagnosed breast cancer. The results indicate that MRI does not reduce re-excision rates and causes false positives in terms of detection and unnecessary surgery. The researchers concluded that there is little evidence to support the routine use of breast MRI in newly diagnosed breast cancers.
Breast MRI has a lot of supporters, but when you look at the data, there are some real issues with the technology. Unfortunately, I think the acceptance of this tool has outpaced its actual performance.
Q: In which cases do you recommend a preoperative MRI for your patients?
There are certainly cases where MRI is not only warranted, but extremely helpful as a diagnostic tool. In my experience, approximately 30% of patients of truly require breast MRI to help me further evaluate the breast tissue. MRI is best utilized in cases where mammography or ultrasound is compromised, such as women with dense or complex breasts, occult cancers or diseases that appear amorphous on mammograms.
A patient’s relative breast cancer risk is also a factor in my decision to recommend breast MRI. I will certainly recommend MRI for a patient with a known genetic mutation, which carries a 60% risk of contralateral cancer. I’m also much more likely to recommend it for younger patients. However, it’s important to remember that these are simply factors, and each patient must be evaluated independently to assess her individual need for an MRI.
If you can visualize definite dimensions via mammography and ultrasound, MRI is most likely not going to be useful. Approximately 40% of patients have mammographic, sonographic and physical examinations that are more than adequate to evaluate the size and extent of the disease without additional imaging.
The remaining 30% of patients are on the cusp – you feel as if you can adequately manage the disease without MRI, but you recommend it as an option. Unfortunately, I think many physicians would agree that a majority of MRIs in these cases are suggested from a defensive medicine standpoint rather than the best interest of the patient.
Q: Research indicates up to 14% of patients initially deemed eligible for APBI were later found to be ineligible based on MRI findings. Have you observed a similar proportion?
I won’t use breast MRI to disqualify anyone from APBI, unless we’ve actually found multicentric or multifocal disease that would exclude them from breast conservation therapy. To determine if a patient is eligible for APBI, I wait for the pathology results to show the actual size of the disease, whether we have adequate margins and if the breast is cosmetically acceptable. But in terms of using MRI to determine whether the tumor size meets APBI criteria, it too frequently overestimates the extent of disease to rely on it for patient selection.
But many physicians do rely on MRI for this purpose. These tend to be the physicians that have the least experience with breast MRI. They are less aware of its disadvantages and tend to trust the technology more than they trust themselves. That was my experience early on, until I began to notice how frequently MRI findings didn’t correlate with other imaging modalities.
Q: Do you think preoperative MRI causes more women to choose mastectomy or creates significant delays in treatment, as some studies suggest?
Research shows that an increase in mastectomy rates tends to mirror the incidence of MRI proliferation. But the problem is that the clinical evidence doesn’t show a corresponding increase in survival rates, or that a majority of these cancers were actually more extensive than originally suspected. We’re either discovering inconsequential disease or relying on MRI-findings that overestimate the extent of the disease. We’re not saving more lives, but we are losing more breasts.
Delay is also clearly an issue, though I think the impact is more psychological than clinical. When I talk to patients about MRI, I make sure they understand it will take a few days to schedule the MRI, and the exam will be interpreted a few days after that. If there’s an abnormality, it’s going to take a few days to schedule an ultrasound and biopsy and another couple days to get the pathology results. So by definition, we are delaying surgery by one or two weeks for the sake of additional imaging. While this doesn’t make a significant clinical difference, it can have a great deal of psychological impact. If a woman is anxious about surgery and she falls into the group of patients who most likely won’t benefit from breast MRI, it simply doesn’t make sense to create an unnecessary delay in her treatment.
Q: Is there a particular imaging modality or set of criteria you find most useful for APBI patient selection?
Digital mammography is beneficial for identifying the extent of disease process for non-invasive diseases. For invasive diseases, breast ultrasound is the best for establishing the actual extent and size of the invasive component. For non-invasive diseases like DCIS, ultrasound is fairly ineffective, but it’s the most correlative imaging modality for evaluating invasive disease.
We have institutional guidelines based on the original ASBS inclusion criteria. In my conversations with radiation oncologists, we understand the ASTRO treatment recommendations but we think they’re being overly restrictive despite the positive data we have on partial breast radiation.
Dr. Berry is a Breast Surgical Oncologist and is board certified by the American Board of Surgery. He received his fellowship training in Breast Surgical Oncology through the Susan G. Komen Breast Cancer Foundation fellowship of Breast Oncology at the University of Texas Southwestern Medical Center at Dallas. He is the director of the Breast Clinic of Memphis which he started in 2002. Visit http://www.breastclinicmemphis.com for more information.