Effect of the third generation selective estrogen receptor modulator arzoxifene on mammographic breast density
562 Background: Arzoxifene (ARZ) is currently being studied for treatment of breast cancer patients in a Phase II trial because of tamoxifen-like efficacy but lack of uterine agonist effect. We conducted a Phase II chemoprevention trial in women at high risk for development of breast cancer on the basis of personal or family history. Methods: Potential subjects had multiple biomarkers assessed, including random periareolar fine needle aspiration (RPFNA) with breast epithelial cells processed for cytomorphology and immunocytochemistry. Women who exhibited cytologic hyperplasia ± atypia were eligible for enrollment. Subjects were stratified on the basis of atypia, estrogen receptor expression, menopause status, germline BrCa1/2 mutation status, and accrual site. Subjects were randomized (double-blind) between placebo and ARZ (LY353381.HCI, 20 mg daily) for 6 mo, with an option to continue on study for another 6 mo while receiving open-label ARZ. Assessments conducted at baseline, 6 mo, and 12 mo included mammographic breast density. Mammograms were digitized to image files which were cropped to remove labels and dates, and then identified by a study subject ID number and a random coding for baseline, 6 or 12 mo. This allowed the reader (GU) to view the three files for a subject, but to remain blinded as to the sequence of the films or the study agent. The files were assessed for mammographic density using the Madena computer-assisted system. Results: Of 199 subjects enrolled on the study, 52% were pre-menopausal; with 101 women randomized to placebo and 98 to ARZ. At baseline, mean values were comparable for placebo and ARZ groups for breast area (∼244 cm2), total dense area (∼100 cm2), and the percent of the breast at increased density (41.3% vs 46.2%). After 6 mo, there were minimal changes in total breast area (P=0.13); but statistically significant decreases (P<0.001) for the comparison of placebo vs ARZ (2-sided T-test) for change in both dense area (+3.8 vs −12.9 cm2) and percent breast density (+0.8% vs −4.6%). Conclusions: The 3rd generation SERM arzoxifene administered for 6 mo produces statistically significant decreases in mammographic breast density relative to placebo in women at high risk for development of breast cancer. No significant financial relationships to disclose.