Abstract 575: Correlation of serum vitamin D levels and changes in breast density

Proceedings: AACR 103rd Annual Meeting 2012‐‐ Mar 31‐Apr 4, 2012; Chicago, IL Background: Mammographic density is an independent risk factor for breast cancer associated with a large relative and attributable risk for the disease. Given the prevalence of women with increased breast density, the attributable risk may be as high as 33%. The largest factor influencing density is heredity, but various lines of evidence suggest that modification of breast density is associated with a decrease in breast cancer risk. Studies have found vitamin D levels to be inversely related to breast density. Here we present a cross-sectional analysis of vitamin D and breast density in women enrolled in the Vermont Cancer Centers High Risk Breast Program. Longitudinal associations and correlations of vitamin D and breast density were evaluated. Methods: The High Risk Breast Program database, for women with strong family history, BRCA mutations, high risk pathology or prior chemoradiotherapy for Hodgkins Disease, was initiated in 2003 and contains over 500 women. A subset of 51 women ranging in age from 31-69 with a mean lifetime Gail risk of 21% is the subject of this analysis. Data collected included mammograms, a medical and dietary history, BMI, and serum Vitamin D levels. Breast density was interpreted (at baseline, 2 years and 4 years) from mammograms using a visual analogue scale and reliability was evaluated. Univariate and multivariate analyses were performed using SAS statistical software. Results: The intra-observer variability for density determination using a visual analogue scale was comparable to published studies (R=0.82). Baseline BMI showed an inverse univariate trend with percent of dense tissue at both time points (p=0.001). No correlation was found between any determinant and change in breast density measured at the 2 year time point. A subset of patients (n=37) was followed for 2 additional years. A model for change in percent dense tissue was obtained for this subset of patients using multivariate analysis that included baseline BMI and change in serum Vitamin D level (R2=0.23). The other factors with a univariate relationship were not found to contribute to the multivariate model. Conclusions: The HRBP database has become an invaluable resource, significantly reducing the financial resources necessary to complete this study and aiding the expansion of the study when needed. Information has been gained with high potential impact on prevention and early detection of breast cancer; first is the reproducibility of the visual analogue scale, a method of quantifying density which is readily available to clinicians and research staff with little training and minimal resources. Second, this study illuminates the importance of a longer follow up period for the detection of changes in mammographic breast density. Third, this study validates the association between serum vitamin D levels and percent breast density, and more importantly the inverse relationship between change in vitamin D levels and change in breast density. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 575. doi:1538-7445.AM2012-575