Purpose:To investigate whether qualitative magnetic resonance (MR) imaging assessments of background parenchymal enhancement (BPE), amount of fibroglandular tissue (FGT), and mammographic density are associated with risk of developing breast cancer in women who are at high risk. Materials and Methods:In this institutional review board-approved HIPAA-compliant retrospective study, all screening breast MR images obtained from January 2006 to December 2011 in women aged 18 years or older and at high risk for but without a history of breast cancer were identified. Women in whom breast cancer was diagnosed after index MR imaging comprised the cancer cohort, and one-to-one matching (age and BRCA status) of each woman with breast cancer to a control subject was performed by using MR images obtained in women who did not develop breast cancer with follow-up time maximized. Amount of BPE, BPE pattern (peripheral vs central), amount of FGT at MR imaging, and mammographic density were assessed on index images. Imaging features were compared between cancer and control cohorts by using conditional logistic regression. Results:Twenty-three women at high risk (mean age, 47 years 6 10 [standard deviation]; six women had BRCA mutations) with no history of breast cancer underwent screening breast MR imaging; in these women, a diagnosis of breast cancer (invasive, n = 12; in situ, n = 11) was made during the follow-up interval. Women with mild, moderate, or marked BPE were nine times more likely to receive a diagnosis of breast cancer during the follow-up interval than were those with minimal BPE (P = .007; odds ratio = 9.0; 95% confidence interval: 1.1, 71.0). BPE pattern, MR imaging amount of FGT, and mammographic density were not significantly different between the cohorts (P = .5, P = .5, and P = .4, respectively). Conclusion:Greater BPE was associated with a higher probability of developing breast cancer in women at high risk for cancer and warrants further study.q RSNA, 2015

ammographic breast density can mask cancers at mammography and is an independent risk factor for breast cancer (1-3). Legislation mandating patients be notified of mammographic breast density has passed in more than 30 states, and a federal bill is under consideration. Details of state legislation vary, but most states require direct reporting to the patient that breast density can mask cancers at mammography and that the patient may benefit from additional testing. Qualitative assessment of mammographic breast density is subjective and varies widely between radiologists (4-10). In a study of 83 radiologists who assessed breast density, Sprague et al (4) found extreme variation in qualitative density assessment per the Breast Imaging Reporting and Data System (BI-RADS), with 6%-85% of mammograms assessed as either heterogeneously or extremely dense depending on radiologist interpretation. In a study of 34 radiologists, the intraradiologist agreement of density assessments among women who underwent two examinations varied from 62% to 87% (6). Commercially available methods for automated assessment of breast density do exist; however, they yield mixed results in agreement with expert qualitative density assessments, with k scores of 0.32-0.61 (11,12). These methods tend to result in over-or underreporting of breast density when compared with qualitative assessment by radiologists (11,13). A recent study found significant differences in density assessments in the same 4170 women with two software programs (Volpara, Volpara Solutions, Wellington, New Zealand; Quantra, Hologic, Bedford, Mass), with the software programs showing 37% and 51%, respectively, of women had dense breast tissue. In the same set of mammograms, radiologists determined 43% of the women had dense breast tissue (13). Deep learning (DL) has been gaining traction in radiology (12,14-17). Specifically, there has been preliminary work with DL methods to assess breast density (12,18); however, none of these techniques have been implemented in clinical practice, raising questions about clinical acceptance by practicing radiologists and the effect on patient care. In contrast, our purpose was to develop a DL algorithm we could use to reliably assess breast density and to measure the acceptance of its predictions in real-time clinical practice. We hypothesize that DL models can be applied to assess breast density at the same level as experienced breast imagers and that they can be accepted into routine clinical practice.

No abstract