Abstract CN04-04: Phase II trial of bexarotene in women at high risk of breast cancer: Comparison of protein and RNA biomarkers

Abstract

Abstract CN04-04 Background Breast cancer prevention trials have shown that tamoxifen or raloxifene reduce breast cancer incidence by up to 50%. However, these drugs do not prevent estrogen receptor (ER)-negative breast cancer. Based on previous studies in mice showing RXR-selective retinoids or rexinoids prevent ER-negative breast cancer, we conducted a Phase II trial testing whether the rexinoid bexarotene would suppress breast cell proliferation in women at high risk of breast cancer. Methods In this Phase II double-blind randomized clinical trial, pre- and post-menopausal women at high risk of breast cancer (defined as >10% chance of carrying a BRCA1 or 2 mutation) received 200 mg/m2 bexarotene or placebo for 28 days. Breast core needle biopsies were taken on Days 1 and 29. Pre- and post-treatment samples were assessed for Ki67 expression (primary endpoint) and Cyclin D1 expression (secondary endpoint) by immunohistochemistry (IHC) and Ki67 and cyclin D1 expression (secondary endpoints) by quantitative RT-PCR (qRT-PCR). Results Eighty seven participants enrolled in this study. The median age was 43 years (range: 20-82). Bexarotene was well tolerated, but was associated with more frequent hypertriglyceridemia (57%), subclinical hypothyroidism (49%), and mild skin reactions (34%) compared to placebo (6%, 0%, 6%, respectively). These side effects resolved after completing therapy. 66 women had paired biopsies, and 52 of these pairs were satisfactory for biomarker analysis. For the primary endpoint, there was no change in pre- versus post-treatment Ki67 as measured by IHC staining in either arm, and no difference in the change in Ki67 between the two arms (p=0.88). We also conducted analysis of Ki67 RNA expression as measured by qRT-PCR. Overall, these studies showed a decrease in Ki67 RNA in the bexarotene treated group as compared to an increase in the placebo group, although this difference did not achieve statistical significance (p=0.20). We next investigated the effect of bexarotene on the expression of Cyclin D1 by IHC and by qRT-PCR. These studies showed that there was no difference in the change in Cyclin D1 protein expression overall as measured by IHC (p=0.40). In subgroup analysis of pre- and post-menopausal women, Cyclin D1 protein expression decreased in post-menopausal women in both arms, but this decrease was greater in the bexarotene arm (although not statistically significant, p=0.32). In studies of cyclin D1 RNA, there was an overall decrease in the bexarotene arm that was nearly twice the decrease in the placebo arm, although this difference did not reach statistical significance (p=0.13). In a subgroup analysis of postmenopausal women, the decrease in cyclin D1 was significantly greater in the bexarotene arm (-65%) than in the placebo arm (-24%)(p=0.03). Conclusions This study shows biomarker analysis using IHC and quantitative RT-PCR is feasible using core needle breast biopsies. We did not see any significant reduction in Ki67 protein in the women taking bexarotene as measured by IHC analysis. However, we did observe a reduction in cyclin D1 RNA expression as measured by qRT-PCR, particularly in postmenopausal women taking bexarotene, suggesting a biological effect of bexarotene in women at high risk of breast cancer. These studies support the use of RNA-based biomarkers in Phase II biomarker studies. We are now using these results to plan future cancer prevention trials using rexinoids. Citation Information: Cancer Prev Res 2008;1(7 Suppl):CN04-04.