Abstract A5: A pilot study of circulating and intrabreast tissue estrogen levels among postmenopausal BRCA1/2 mutation-positive women

Abstract

Abstract Background: Increasing evidence suggests that BRCA1-related breast cancers may be a consequence of genotoxic variants of hormonal carcinogenesis rather than the physiological properties of ovarian hormones. Catechol estrogens, 2-OHE1, 2-OHE2, and 4-OHE1 and the 16 -hydroxlyated estrogens are strong promoters of breast carcinogenesis. In BRCA1 mutation carriers, estrogens may elevate risk by increasing genomic instability within the background of impaired DNA repair function (Y Hu et al., Oncogene, 2005). Currently, the influence of BRCA2 on transcriptional activity of ER is unknown. The relationship between circulating estrogen metabolites (EM) and EM in breast tissue (e.g., in nipple aspirate fluid (NAF) and ductal lavage supernatant (DLS) is not well-established in either BRCA1/2 mutation carriers or in non-carriers. Methods: We conducted a cross-sectional study of EM in serum, NAF, and DLS among postmenopausal (i.e., spontaneous amenorrhea for ≥12 months or bilateral oophorectomy) BRCA1/2 carriers ages 31 to 57 enrolled in the Clinical Genetics Branch's Breast Imaging Study. Serum, NAF and DLS were collected (using standard methods) at the same visit, but both NAF and DLS could not always be collected from each participant. Total EM (conjugated and free forms of estrone (E1), estradiol (E2), estriol (E3) and up to 12 EM were measured in paired NAF and serum samples (n=19 women) and paired DLS and serum samples (n=25 women) using a novel quantitative liquid chromatography-tandem mass spectrometry method previously validated for serum (X Xu et al., Anal Chem, 2007). Spearman's correlation coefficients were calculated to examine the strength of the relationship between serum and tissue-specific EM. Results: The intraclass correlation coefficients (ICC) for EM assessed in replicate specimens demonstrated excellent reproducibility for all specimen types (ICCs ranged from 0.98–0.99 for EM measured in individuals with replicate NAF, DLS, and serum specimens). Total (conjugated plus free) levels of circulating serum EM were strongly and positively correlated with tissue-specific EM measured in NAF (E1, r=0.99, p<0.0001; E2, r=0.73, p=0.0004; E3, 0.88, p<0.0001; 2-MeOE1, r=0.57, p=0.01; 2-MeOE2, r=0.44, p=0.06; 2-OHE1, r=0.73, p=0.0004; 2-OHE2, r=0.58, p=0.009) and DLS (E1, r=0.97, p<0.0001; E2, r=0.72, p<0.0001; E3, 0.79, p<0.0001; 2-MeOE2, r=0.42, p=0.04. Conclusions: Our data demonstrate that EM can be detected and reliably measured in novel breast tissue-derived biospecimens (i.e., NAF and DLS). Further, our data suggest that circulating EM are strongly and positively correlated with tissue-specific EM measured in NAF and DLS among postmenopausal BRCA1/2 mutation carriers. Correlations between EM and mammographic density will be presented (analyses ongoing). Our pilot data will guide future larger studies designed to determine whether circulating levels of EM may be used to predict intra-breast levels, using existing samples from a more diverse group of women with varying degrees of breast cancer risk and measured mammographic density, a potent breast cancer risk factor. Ultimately, we hope to identify key EM correlates of mammographic density for use as surrogate biomarkers of breast cancer risk and risk modulation measures related to chemoprevention and lifestyle modification. Citation Information: Cancer Prev Res 2010;3(1 Suppl):A5.