MRI background enhancement: Its relationship with breast density and breast cancer risk

Abstract

Background: Increased breast density increases breast cancer risk. MRI breast is increasingly used for imaging as unlike mammography, the sensitivity of MRI breast is not affected by breast density but rather, the amount of background enhancement. It has been suggested that increased background enhancement may be related to the amount of proliferative breast tissue present; hence this may also be related to increased breast cancer risk. To date there have been no study published assessing background enhancement and its relation to breast cancer risk. Method: A retrospective review 45 mammography films and breast MRI scans of 36 patients (9 bilateral) with normal breast imaging was performed and scored independently by 3 radiologists for breast density using BI-RADS criteria. MRI breast density was scored on the non-contrast enhanced T1-weighted series whereas background enhancement was quantified on the first contrast-enhanced T1-weighted fat-suppressed series. Background enhancement was scored as none-minimal, mild, moderate, marked. Weighted kappa was used to assess the inter-modality agreement. A partial approximation to the Gail risk score was calculated from the available risk factors (age, menarche, number of biopsies, births and relatives with breast cancer), using coefficients from a Caucasian population; the resulting values were converted to ranks and was correlated with background enhancement. Result: There was substantial within-reader agreement between mammographic and MRI density (kappa=0.68; p=0.001). There were no correlation between background enhancement and breast density. All 6 patients with background-enhancement were pre-menopausal. Although not statistically significant (p<0.37), it is suggestive that 4 of the 6 (67%) had a history of breast cancer while only 6 of the 13 premenopausal non-enhancers (46%) had cancer. Moreover, the mean of the rank risk for enhancing premenopausal women was 29.5 while the non-enhancing pre-menopausal women was 16.6 (p=0.048 by one-tailed t-test). Conclusion: We found good correlation between mammographic and MRI breast density. There was no correlation between breast density and MRI background enhancement. There is a suggestion that MRI background enhancement may be associated with a higher risk of breast cancer.