Q and A: Magnetic Resonance Imaging in the Detection and Evaluation of Breast Cancer

Abstract

The experts invited to answer questions raised by the recent ACS guideline and studies offer a range of perspectives. Dr Kramer comments that the ACS guideline does not satisfy some of the most fundamental features of high-quality evidence-based guidelines. In particular, the ACS guideline does not explicitly explain how the evidence considered was assigned to categories of evidence. [Editor's note: Representatives from the ACS were invited to contribute to this article, but we had not received their responses by the time the article went to press]. Dr Kramer notes that none of the recommendations was based on data from a randomized clinical trial which, as many have argued, is the gold standard for demonstrating that a screening test reduces mortality.9 Current study designs also do not permit an estimate of the degree of overdiagnosis associated with breast MRI, because all women received both MRI and mammography. Relatively short randomized clinical trials using interval clinical cancers as the end point could address this issue.9 Both Drs Hortobagyi and Kramer underscore that overdiagnosis is a real possibility with MRI screening in the population addressed by the ACS guideline. Dr Hortobagyi ultimately supports use of MRI in high-risk women with the proviso that these women be made aware of the risk of false- positive findings and the consequences of those findings. In a recent systematic review of the literature, Lord et al10 stressed the need for the development of comprehensive risk prediction models to aid in the identification of women at high risk of breast cancer who would benefit the most from MRI screening. Dr Burstein echoes this call for careful attention to the benefits and risks of MRI screening, including the increased risk of biopsy for benign breast disease. He further encourages patients and clinicians to ensure that the MRI facility has sufficient experience with the technique before permitting a facility to perform this test. In evaluating patients with newly diagnosed breast cancer both Drs Morrow and Newman believe that routine MRI use is inappropriate and leads to unnecessary mastectomies. Although MRI detects cancer that is not evident clinically or by mammogram in 10% to 20% of patients, much of the disease found on MRI is controlled by standard postlumpectomy breast radiation. Even in patients with invasive lobular cancer, a subset often proposed to benefit from MRI, no different local recurrence rate is shown compared to patients with ductal cancer. Although contralateral disease is detected in 2% to 3% of patients undergoing bilateral mastectomy for a unilateral breast cancer, both Drs Morrow and Newman agree that MRI should not be used to determine whether contralateral mastectomy should be performed. It is not clear that newly diagnosed breast cancer patients will derive any survival benefit from treating a contralateral tumor detected by MRI but not evident clinically or mammographically. Dr Newman notes, however, that MRI may assist in management decision-making in selected cases, such as when mastectomy and TRAM flap reconstruction is planned (the TRAM flap can only be harvested once but can be used for bilateral reconstruction if done synchronously), or in a patient at increased hereditary risk for cancer. With respect to the number of adequately trained specialists required to accommodate changes in MRI screening practices that may result from the ACS recommendations, Dr Schall and Ms Hartman admit that there are currently not enough radiologists with expertise in breast MRI to handle the potential increase in MRI procedures. They believe, however, that there is sufficient capacity in the system to accommodate the expected increase in breast MRI procedures and they note that there has been progress in training practitioners.