Abstract P3-01-03: Breast density profile in a prospective large breast cancer screening program in South Brazil.

Abstract Introduction: Women involved in breast cancer (BC) screening of NMPOA cohort, in South Brazil, are undergoing their 8th year of follow up. As mammography breast density is a well-established risk factor for breast cancer, since the beginning of this study, all mammograms were characterized according to breast density Birads pattern. Recently, a study published in the JNCI showed an association of this pattern of breast density with tumors that presented more aggressive characteristics. Objective: To characterize the mammography density patterns of women involved in a prospective NMPOA cohort, and describe the mammography density appearance of those diagnosed with BC, correlating it with genomic classification and stages. Methods: First screening mammogram of all women in the cohort was reviewed. All mammograms were classified according to BIRADS® patterns of density, which take into account the percentage of fibroglandular tissue (A: less than 25%; B: from 25 to 50%; C: from 50% to 75%, and D: more the 75%). Also, we compared breast density among the 17 incidental breast cancers diagnosed during follow up. Genomic classification and staging of the diagnosed cancers was described. Results: A total of 2943 first mammograms were done until the end of women accrued inclusion in the cohort (April of 2006). Women's age ranged from 40 to 69 years. From these, 820 were A density, 893 B, 970 C and 260 D (graphic 1). Until November 2011, 17 cancer cases were found in the screening population. From these, 04 were A, 06 were B, 06 were C and 1 were D (graphic 2). Cancer staging and genomic classification according to each mammogram density are summarized on Tables 1 and 2. No differences were observed in BC staging when we compared the A and B density with C and D (p = 1.0). The same occurred with the genomic classification (p = 0.48). Conclusion: Our screening population has a predominance of low mammogram densities, probably due to the high body mass index observed in previous studies of this cohort. With respect to cancer cases, the absence of statistical significance results suggests that longer follow up will be necessary for definitive conclusions. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-01-03.