SAN ANTONIO – MRI screening of women who were at average risk of breast cancer and had a negative screening mammogram resulted in the diagnosis of 11 cases of cancer per 1,000 women screened, Dr. Simone Schrading said at the San Antonio Breast Cancer Symposium.
"The additional cancer detection rate is high, even in heavily prescreened women. In experienced hands, the positive predictive value of MRI screening in this average-risk cohort was comparable to that seen in mammographic screening programs or in MRI high-risk screening cohorts," said Dr. Schrading, of the University of Aachen, Germany.
While MRI is well established as a screening method for women at high familial risk of breast cancer, there have been no data to support its use in average-risk women.
Courtesy ©SABCS/Todd Buchanan2012
Dr. Simone Schrading
Dr. Schrading reported on a prospective, single-center study evaluating the additional cancer yield and accuracy of breast MRI screening in 1,725 women at average risk for breast cancer. All had normal clinical breast examinations and digital screening mammograms. In addition, 89% of the women had undergone high-frequency breast ultrasound screening, which was normal in all cases. None of the subjects had a personal or family history of breast or ovarian cancer, and none had been diagnosed with breast proliferative changes or atypia. Their mean age was 55 years (range, 42-71 years).
MRI screening detected breast cancers in 18 of these patients, for a detection rate of 11 per 1,000 screened. Seven malignancies were ductal carcinoma in situ, and the other 11 were invasive breast cancer. The mean size of the invasive cancers was 11 mm. Five of the 7 ductal carcinomas in situ and 6 of 11 invasive cancers were high-grade cancers. On the other hand, the stage distribution of the invasive cancers was favorable: All were staged as pN0, M0.
Nearly 91% of screening MRIs were negative as defined by a BIRADS 1 or 2 rating. Another 5.9% were rated BIRADS 3; follow-up MRIs in this group of patients showed no breast cancers. Suspicious lesions – that is, BIRADS 4 or 5 – were noted in 54 patients, or 3.2% of the study group, and they were evaluated by MRI-guided biopsy. The lesions proved benign in 28 of 54 cases, high risk in 8, and malignant in 18.
Thus, the positive predictive value of breast MRI screening in this average-risk, extensively prescreened population was 33% if only BIRADS 4 and 5 cancers were defined as true positives, and 48% if BIRADS 4 and 5 high-risk lesions were also counted.
The age distribution of the patients with MRI-detected breast cancers was similar to that of the total study population. Mammographic breast density did not predict the likelihood of identifying breast cancer by MRI.
Dr. Laura J. Esserman rose from the audience to question the cost-benefit ratio of introducing MRI screening for the vast population of average-risk women.
The aggregate cost of breast cancer screening in the United States is already at least $8 billion annually. Utilizing MRI to screen average-risk women would push that figure up by an order of magnitude, observed Dr. Esserman, professor of surgery and of radiology and director of the Carol Franc Buck Breast Care Center at the University of California, San Francisco.
"MRI screening is still very expensive, it’s true," Dr. Schrading replied. "The main reason for the high cost of MRI is the long acquisition time and the long reading time. The long acquisition time is because the MRI protocols we have today are designed for diagnostic purposes, not for the purpose of screening. Our goal is to use a screening protocol for MRI, where the acquisition time is only 3 minutes and the reading time is less than 30 seconds. This might be a way to reduce the cost of MRI," she said.
The study was conducted free of commercial support. Dr. Schrading declared having no financial conflicts of interest.