A pooled analysis of mammographic density and breast cancer risk

描述（由申请人提供）：乳腺X线摄影图像中评估的乳腺密度是乳腺癌风险的强预测因子，但相关性的强度可能因体重而异。当乳腺密度被用作干预研究的中间终点和个体化乳腺癌风险的预测指标时，了解这些异质性的来源是很重要的。拟议的项目利用了以前关于乳房X线摄影密度和乳腺癌风险的病例对照研究的现有数据。拟议项目的具体目标是：检查不同体重指数和乳房大小的乳房X线摄影密度与乳腺癌风险的关系;阐明比较从不同地点和使用不同技术获得的乳房X线摄影图像的方法学问题;并启动一个工作组，就乳房X线摄影密度的研究进行合作。该项目将结合联合收割机的数据，从四个病例对照研究乳腺摄影密度。受试者总数将为4，229例，其中1，747例病例和2，482例对照，种族分布为45%白人、38%亚洲人、8%非裔美国人和9%其他，BMI和乳房大小差异很大。数据库将基于所有相关变量的通用格式，包括乳腺癌状态和分期、乳腺密度百分比和绝对密度、人体测量和生殖变量。每例受试者的乳房X线照片将由一名阅片人进行评估，以确保使用定量评估方法进行的研究之间密度读数的可比性。我们将应用一般线性模型来比较以前通过不同方法获得的密度读数与新的标准化读数。将使用SUDAAN逻辑回归程序并调整混杂因素来估计与不同BMI水平和乳房大小的密度百分比相关的乳腺癌风险。研究人员将计划合作研究，并考虑邀请更多研究人员为未来乳腺摄影密度的创新研究贡献他们的乳腺摄影数据的可能性。该项目将提供关于体重如何影响与乳腺密度相关的乳腺癌风险的新信息。这些发现将有助于乳腺癌风险评估和乳腺癌预防。

DESCRIPTION (provided by applicant): Breast density as assessed in mammographic images is a strong predictor of breast cancer risk, but the strength of the association may differ with body weight. It is important to understand these sources of heterogeneity when breast density is used as an intermediate endpoint in intervention studies and as a predictor of individualized breast cancer risk. The proposed project makes use of existing data from previous case-control studies on mammographic density and breast cancer risk. The specific aims of the proposed project are to examine breast cancer risk associated with mammographic density across levels of body mass index (BMI) and breast size; to elucidate methodologic issues in comparing mammographic images that were obtained from different locations and using different technologies; and to initiate a working group to collaborate on studies of mammographic densities. This project will combine data from four case-control studies on mammographic densities. The total number of subjects will be 4,229 with 1,747 cases and 2,482 controls with an ethnic distribution of 45% Caucasian, 38% Asian, 8% African- American, and 9% Others and a wide variation in BMI and breast size. The data base will be based on a common format for all variables of interest, breast cancer status and stage, breast density as percent and as absolute density, anthropometric, and reproductive variables. One mammogram for each subject will be assessed by one reader to assure the comparability of density readings across studies using a quantitative assessment method. We will apply general linear models to compare the previous density readings obtained through different methods with the new standardized readings. Breast cancer risk associated with percent density for different levels of BMI and breast size will be estimated using the SUDAAN logistic regression procedure and adjusting for confounders. The investigators will plan collaborative studies and consider the possibility of inviting additional researchers to contribute their mammographic data for future innovative investigations of mammographic density. This project will provide new information on how body weight affects breast cancer risk related to breast density. These findings will be useful for breast cancer risk estimation and breast cancer prevention.