Environmental and social determinants of mammographic features

Percent mammographic density (PMD) is one of the strongest risk factors for, and is considered an intermediate marker of, breast cancer. Recent wide spread uptake of digital mammography, and advances in image analyses have identified additional mammographic features including texture variation (gray-scale and spatial variation)

in breast density and deep learning-based risk scores that predict breast cancer risk independent of PMD. Because of the strong associations with breast cancer risk there is a need to identify potential modifiable risk factors of mammographic features. Differences in breast cancer risk by socioeconomic status (SES) have long

been established, as well as observations of spatial variation in the US. It has become clear that social determinants of health (SDOH), have an impact on health conditions including cancer. Healthy People 2030 has identified five key domains of SDOH that impact health: economic stability, education, health care, neighborhood

environment, and social/community context. Additionally, there is a small, but growing body of evidence suggesting a role of environmental exposures (e.g., air pollution) with breast cancer risk and mammographic features. However, most studies have examined the health effects associated with one environmental exposure

or SDOH at a time, ignoring that individuals experience multiple exposures simultaneously. Also, few studies have been able to assess changes in mammographic features over the menopausal transition. Using the resources of the nationwide prospective Nurses’ Health Study 3 (NHS3), we are uniquely positioned to study the complex associations of multiple environmental exposures and SDOH on mammographic

features predictive of breast cancer. NHS3 is an open cohort currently consisting of 33,780 active US female nurses and nursing students (mean age at enrollment 34.0). To date 12,891 participants (38.2%) reported having at least one mammogram. We propose to collect and analyze digital mammographic images from 4,000

participants selected to maximize diversity in age, race and ethnicity, and region of residence. This proposal takes a comprehensive approach to understanding the impact of demographic factors, SDOH and a wide range of environmental factors including air pollution, UV, radon, on mammographic features.

The specific aims are: (1) determine how multiple environmental exposures, demographic factors, and SDOH are associated with mammographic features, and (2) with change in mammographic features from pre- to postmenopause; and (3) determine the joint associations of environmental and SDOH characteristics with

baseline PMD and imaging features and change of the same. We expect our findings to provide valuable information on the role of modifiable exposures on mammographic features and thus, subsequent breast cancer risk. These findings will provide valuable information for individual- and population-level prevention, risk assessments, and policy decisions, with the long-

term goal of reducing the burden of breast cancer.