Background parenchymal uptake on molecular breast imaging as a breast cancer risk factor: a case-control study

Abstract

Background: Molecular breast imaging (MBI) is a functional test used for supplemental screening of women with mammographically dense breasts. Additionally, MBI depicts variable levels of background parenchymal uptake (BPU) within nonmalignant, dense fibroglandular tissue. We investigated whether BPU is a risk factor for breast cancer.

Methods: We conducted a retrospective case-control study of 3027 eligible women who had undergone MBI between February 2004 and February 2014. Sixty-two incident breast cancer cases were identified. A total of 179 controls were matched on age, menopausal status, and MBI year. Two radiologists blinded to case status independently assessed BPU as one of four categories: photopenic, minimal to mild, moderate, or marked. Conditional logistic regression analysis was performed to estimate the associations (OR) of BPU categories (moderate or marked vs. minimal to mild or photopenic) and breast cancer risk, adjusted for other risk factors.

Results: The median age was 60.2 years (range 38-86 years) for cases vs. 60.2 years (range 38-88 years) for controls (p = 0.88). Women with moderate or marked BPU had a 3.4-fold (95 % CI 1.6-7.3) and 4.8-fold (95 % CI 2.1-10.8) increased risk of breast cancer, respectively, compared with women with photopenic or minimal to mild BPU, for two radiologists. The results were similar after adjustment for BI-RADS density (OR 3.3 [95 % CI 1.6-7.2] and OR 4.6 [95 % CI 2.1-10.5]) or postmenopausal hormone use (OR 3.6 [95 % CI 1.7-7.7] and OR 5.0 [95 % CI 2.2-11.4]). The association of BPU with breast cancer remained in analyses limited to postmenopausal women only (OR 3.8 [95 % CI 1.5-9.3] and OR 4.1 [95 % CI 1.6-10.2]) and invasive breast cancer cases only (OR 3.6 [95 % CI 1.5-8.8] and OR 4.4 [95 % CI 1.7-11.1]). Variable BPU was observed among women with similar mammographic density; the distribution of BPU categories differed across density categories (p < 0.0001).

Conclusions: This study provides the first evidence for BPU as a risk factor for breast cancer. Among women with dense breasts, who comprise >40 % of the screening population, BPU may serve as a functional imaging biomarker to identify the subset at greatest risk.

Keywords: Breast cancer risk; Breast density; Mammography; Molecular breast imaging; Tc-99m sestamibi.

Fig. 1

Examples of background parenchymal uptake categories. Molecular breast imaging (MBI) examinations and corresponding full-field digital mammograms from four different women are shown. All images were acquired in the mediolateral oblique projection. MBI with photopenic background parenchymal uptake (BPU) (a) and corresponding mammogram (b). MBI with minimal to mild BPU (c) and corresponding mammogram (d). MBI with moderate BPU (e) and corresponding mammogram (f). MBI with marked BPU (g) and corresponding mammogram (h)

Fig. 2

Example of incident breast cancer case. Right and left mediolateral oblique views are shown. In this 41-year-old premenopausal woman, screening full-field digital mammography (a) was negative with Breast Imaging-Reporting and Data System density category of extremely dense. Supplemental screening molecular breast imaging (MBI) (b) performed at the time of the screening mammogram showed marked background parenchymal uptake. The next 2 years of screening mammography were negative. At 2.6 years after the initial MBI, the patient presented for diagnostic workup of nipple retraction. (c) MBI performed at the time showed a lesion in the left breast (arrow) that was diagnosed as a 9-mm, grade I invasive ductal carcinoma

Fig. 3

Mosaic plots show distribution of background parenchymal uptake (BPU) categories on molecular breast imaging (MBI) as a function of Breast Imaging-Reporting and Data System (BI-RADS) breast density categories for cases (a) and controls (b) interpreted by reader 1 and cases (c) and controls (d) interpreted by reader 2. BI-RADS 1 = almost entirely fat; BI-RADS 2 = scattered fibroglandular densities; BI-RADS 3 = heterogeneously dense; BI-RADS 4 = extremely dense