Effect of Preventive Hormonal Therapy on Breast Density: A Systematic Qualitative Review

Table 2

Raloxifene (RLX): characteristics of the reviewed studies (𝑛=9).

Authors/year	Endpoint	Design	High risk definition	Population/number of cases	Age	Method of breast density measurement/time	Mammograms readers	Reproducibility	BD baseline
Jackson et al. 2003 [42]	Effect of raloxifene and HRT on breast density	Randomized, open-label, parallel designed trial	NA	356 postmenopausal women with osteopenia or osteoporosis/280 assigned randomly/193 included in the final analysis: RLX: 109; ccHT: 84^1	Mean 66.7 y; average y postmenopause: 17.9; RLX (mean ± SD): 66.9 y ± 5.3; ccHT: 66.4 y ± 4.5	Visually according to BI-RADs classification at baseline and after 12 months of therapy	Three radiologists blinded to treatment assignment	Weighted Kappa for interrater reliability at baseline: 0.57 to 0.70; after 12 months: 0.51 to 0.66	BI-RADs I: 16.6–22.8%; II: 65.3–75.6%; III: 7.8–10.9%; IV: 0-1%
Eilertsen et al. 2008 [43]	Effect of raloxifene and different regimens of HT on breast density	Open-label, randomized, comparative clinical trial	NA	358 postmenopausal women/202 assigned randomly/178 included in the final analysis: low-dose combined HT: 𝑛=44, conventional-dose combined HT: 𝑛=45, tibolone: 𝑛=45, and RLX: 𝑛=44^2	45–65 yrs: low-dose combined HT (mean ± SD): 54.8 y ± 5.0; conventional-dose combined HT: 56.1 y ± 3.6; tibolone: 54.9 y ± 4.7; RLX: 56.3 y ± 4.9	Volumetric breast density by a fully automated technique in full-field digital mammograms at baseline and after 12 weeks	NA	NA	Low-dose HT: 8.6 (5.1–17.1); conventional-dose HT: 8.3 (5.5–11.4); tibolone: 7.5 (5.3–12.3); RLX: 7.7 (5.7–11.1)^3
Eng-Wong et al. 2008 [44]	Effect of raloxifene on breast density	Retrospective analysis of a phase II prospective trial	Gail 5-year risk ≥ 1.7%, or a family history of BC, or previous LCIS, ADH, DCIS	37 high risk premenopausal	43 y (35–47)	On digitized mammograms using a semiquantitative technique and using MRI T1-weighted images to determine breast MRI volume using a semiautomatic method at baseline, after 1 and 2 years, and 1 year posttreatment	Two radiologists	Previously tested in another validation study for MRIV (94.95% agreement). For mammograms: interradiologist correlation: 1 y: 0.63; 2 y: 0.62; 3 y (one year off  RLX): 0.39	Mean BD (range) 39% (7–78)
Freedman et al. 2001 [45]	Effect of raloxifene (at one of two doses) and HRT on breast density	Subgroup analysis of a prospective double-blind, randomized, placebo-controlled trial	NA	619 postmenopausal women with previous hysterectomy/168 included in the final analysis: placebo (𝑛=45); RLX 60 mg/day (𝑛=45); 150 mg/day (𝑛=42); estrogen 0.625 mg/day (𝑛=36)	Mean 53 y (45–60); mean y after menopause: 6.; Placebo (mean ± SD): 52.2 y ± 5.1; RLX 60 mg/day: 54.1 y ± 4.2; 150 mg/day: 53.3 y ± 4.8; estrogen: 52.1 y ± 4.6	Digital scanning and computer-assisted segmentation of mammograms at baseline and 2 years	One radiologist blinded to treatment	Previously tested in another validation study	Placebo: 9.8%±9.6%; RLX 60 mg/day: 9.3%±9.1%; 150 mg/day: 8.1%±6.6%; estrogen: 13.5%±11.5%
Harvey et al. 2009 [46]	Effect of bazedoxifene compared with raloxifene or placebo on breast density	Retrospective analysis of a phase III randomized placebo- and active-controlled trial	NA	7 609 postmenopausal women with osteoporosis/1 243 eligible for participation; 622 eligible for digitization and 442 included in the final analysis: bazedoxifene 20 mg (𝑛=92); bazedoxifene 40 mg (𝑛=106); RLX 60 mg (𝑛=119); placebo (𝑛=125)	≤62 y; mean age: 58.7 y. Mean ± SD: bazedoxifene 20 mg/day: 58.3 y ± 2.5; 40 mg/day: 58.8 y ± 2.4; RLX 60 mg/day: 58.8 y ± 2.6; placebo: 58.8 y ± 2.5	On digitized mammograms using an interactive thresholding software at baseline and at 2 years	One radiologist blinded to time sequence and treatment arm	NA	Bazedoxifene 20 mg: 26.4% (SD 18.7%); bazedoxifene 40 mg: 25.8% (SD 19.1%); RLX 60 mg: 27.6% (SD 19.3%); placebo: 27.2% (SD 18.1%)
Nielsen et al. 2009 [47]	Effect of transdermal estradiol compared with raloxifene on breast density or heterogeneity	Post hoc analysis of a prospective randomized study	NA	500 women at least 5 y postmenopausal/270 included in the final analysis: RLX 60 mg/day (𝑛=135); low-dose estradiol  (𝑛=135)^4	66 years (55–80); mean ± SD: RLX: 66.7 y ± 0.5; estradiol: 66.3 y ± 0.5	Visually according to BI-RADs classification, area percentage and computer-based (E2-specific) heterogeneity examination of digitized mammograms at baseline and at 2 years	One radiologist blinded to treatment	Previously tested in another validation study: intraobserver variability between baseline and 2-year assessments = 0.79 (range 0.70–0.86)	RLX 60 mg: 16% (5–31); low-dose estradiol: 16% (7–24)
Lasco et al. 2006 [48]	Effects of long-term raloxifene on breast density	Prospective case control study	NA	70 postmenopausal women with normal body weight/cases: women with osteoporosis receiving RLX 60 mg/day (𝑛=50); controls: women without osteoporosis (𝑛=20)	Cases: 52.4 ± 4.1 y, menopausal age: 42.1 ± 3.9 y; controls: 53.6 ± 3.5 y, menopausal age: 43.1 ± 3.6 y	Digital scanning and computer-assisted segmentation of mammograms at baseline and 2 years	NA	NA	Image mean index (IMI)^5: ~3.35 for both groups
Christodoulakos et al. 2002 [49]	Effect of raloxifene compared with tibolone on breast density	Randomized, comparative clinical trial	NA	131 postmenopausal women/tibolone 2.5 mg/day (𝑛=56), RLX 60 mg/day (𝑛=48) and controls: no risk factors of osteoporosis or denied treatment: (𝑛=27)	Mean age ± SD: tibolone: 52.6 y ± 4.8, RLX: 53.9 y ± 3.9, controls: 51.4 y ± 7.6, mean months since menopause: range 69.8–88.3	Visually according to Wolfe ^6classification at baseline and 12 months	Two radiologists blinded to treatment arm	In case of discrepancies (9.2–13%), films were reevaluated by the 2 radiologists together for consensus	N1: tibolone: 19 (36.5%), RLX: 23 (47.9%), controls: 6 (22.2%); P1: tibolone: 16 (30.8%), RLX: 10 (20.8%), controls: 7 (25.9%); P2: tibolone: 15 (28.8%), RLX: 12 (25%), controls: 11 (40.7%); DY: tibolone: 2 (3.8%), RLX: 3 (6.3%), controls: 3 (11.1%)^7
Cirpan et al. 2006 [50]	Effect of raloxifene on breast density	Retrospective study	NA	55 postmenopausal women with osteopenia or osteoporosis	Mean age ± SD: 50.4 y ± 3.6 (43–58)	Visually according to BI-RADs classification at baseline and after 12 to 16 months of therapy	NA	NA	BI-RADs category I: 𝑛=8 (14.5%); II: 𝑛=28 (50.9%); III: 𝑛=17 (30.9%); IV: 𝑛=2 (3.6%)

1ccHT: continuous-combined HT: conjugated equine estrogen 0.625 mg/day + medroxyprogesterone acetate 2.5 mg/day. ²Conventional-dose HT = 2 mg 17 𝛽-estradiol and 1 mg norethisterone acetate, low-dose HT = 1 mg 17 𝛽-estradiol and 0.5 mg norethisterone acetate. NA: not available or not applicable. ³Values are represented as median (25th–75th percentile of interquartile range). LCIS: lobular carcinoma in situ, ADH: atypical ductal hyperplasia, and DCIS: ductal carcinoma in situ. ⁴Weekly patch delivering 0.014 mg estradiol/day. ⁵Image Mean Index (IMI): computer-assisted algorithm calculation including identification and delimitation of the same region of interest for each image and the differentiation of gray levels into 10 classes. ⁶Wolfe classification: N1: parenchymal pattern composed almost entirely of fat with trabeculae but no visible ducts, P1: pattern composed mainly of fat with fibroglandular tissue that constitutes 25% of the breast, P2: pattern composed of fibroglandular tissue appearing as a heterogeneously dense breast that occupies more than 25% of the breast, and DY: extremely dense tissue. ⁷Tibolone versus controls: 𝑃=0.32 or RLX versus controls: 𝑃=0.18.