Volume and Morphology of Fibroglandular Tissue for Breast Cancer Risk Prediction

描述（由申请人提供）：纤维腺组织的体积和形态学用于乳腺癌风险预测乳房密度作为乳腺癌发展的一个强有力的风险预测因子的作用已经被许多研究所证实。乳腺癌预防协议组建议将定量乳腺密度纳入癌症风险预测模型，但如何可靠地测量定量密度参数仍是一个活跃的研究领域。除了密度的多少外，致密组织的形态分布模式也可能对风险预测起作用，而这只能在三维图像上进行分析。在本R21应用中，我们将评估基于mri的密度参数的作用，包括纤维腺组织的体积和形态，并使用病例对照研究设计建立风险预测模型。提出了三个目标。Aim-1将开发一个全自动分割软件来分割乳房和纤维腺组织。这个软件将提供给共享，它将

DESCRIPTION (provided by applicant): Volume and Morphology of Fibroglandular Tissue for Breast Cancer Risk Prediction The role of breast density as a strong risk predictor for development of breast cancer has been established by many studies. The Breast Cancer Prevention Collaborative Group has recommended that quantitative breast density should be incorporated into the cancer risk prediction model, but how to reliably measure quantitative density parameters is still an active research area. In addition to the amount of density, the morphological distribution pattern of dense tissue may also play a role in risk prediction, which can only be analyzed on 3-dimensional images. In this R21 application we will evaluate the role of MRI-based density parameters, including the volume and the morphology of the fibroglandular tissue, and build a risk prediction model using a case-control study design. Three aims are proposed. Aim-1 will develop a fully automated segmentation software to segment the breast and the fibroglandular tissue. This software will be made available for sharing, and it will provide a very useful tool for researchers in the breast densitometry research field to analyze large datasets. Aim-2 will develop a risk prediction model based on the MRI-analyzed fibroglandular tissue volume and the morphological distribution pattern, in combination with six basic risk factors (age, hormonal use, family history, prior benign disease, weight, number of live birth) to differentiate between patients who were found to have cancer in screening MRI (cases) vs. matching controls. We have access to a large screening MRI database for retrospective analysis. It is estimated that 220 cancer cases will be available, and by using a 1:5 ratio we will select 1,100 matching controls for analysis. Then Aim-3 will evaluate how the fibroglandular tissue volume and morphological index may be used to improve the risk prediction accuracy, by comparing to the risks estimated by using existing standard models. The history sheet that each subject filled out will be used to calculate the risk scores by using Gail, Claus, BRCAPRO and Tyrer-Cuzick models. The ability of these existing models in differentiating between the cancer cases and controls will be compared to that analyzed using the MRI-density model developed in Aim-2, and the results will allow us to evaluate the added value of breast density in risk prediction. The success of this R21 will build a great foundation for a subsequent longitudinal study, using a prospective screening database that is being collected now within the Athena Breast Health Network formed by five University of California campuses.