Off-campus WSU users: To download campus access dissertations, please use the following link to log into our proxy server with your WSU access ID and password, then click the "Off-campus Download" button below.

Non-WSU users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Access Type

WSU Access

Date of Award

January 2018

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Cancer Biology

First Advisor

Michele L. Cote

Abstract

African American women (AAW) suffer a higher breast cancer mortality burden than women of other ethnicities in the US. More likely to be diagnosed with aggressive subtypes resistant to therapy and with rapidly fatal course than European American women (EAW), AAW may benefit greatly from earlier detection of breast cancers. However, it remains difficult to predict with a high degree of accuracy which women will develop breast cancer. Current risk assessment is especially poor for AAW, where models consistently underestimate risk in the subset of women with a prior biopsy. Risk assessment can be improved with the inclusion of new risk factors and, for AAW, race-specific estimates of risk factors. Here we characterized current and novel radiologic and pathologic tissue-based risk factors to improve risk assessment in an understudied population.

We utilized the Detroit BBD cohort to examine several risk factors. We first assessed subsequent breast cancer risk associated with fibroadenomas, a previously-described risk factor. In a nested case/control study, we assessed whether previously-described BI-RADS density scores and Tabár patterns were associated with breast cancer. We also examined whether a complexity indicator, summarizing features routinely described on mammogram but not yet examined as a risk factor, was associated with breast cancer. Finally, in a subset of the nested case/control study additionally age-matched to population-level controls, we examined whether crown-like structures of the breast (CLS-B) were associated with breast cancer risk. We used several uni- and multivariable logistic, ordinal logistic, and conditional logistic models to estimate associations between risk factors and breast cancer.

In Aim 1, fibroadenomas on biopsy were not associated with a breast cancer risk increase over population level risk, unlike prior studies in EA women. In Aim 2, nodular patterns on mammogram, assessed by Tabár classification or our complexity indicator, was more strongly associated with breast cancer than density. These findings suggest that AAW with BBD may benefit from additionally assessing parenchymal patterns on mammography. In Aim 3, we found that CLS-B was associated with breast cancer independent from BMI and BBD and may serve as a histologic marker of risk. These findings suggests differences in risk by race, though we cannot rule out secular differences between our contemporary cohort and other cohorts. These dissertation results, once replicated in other studies, can inform risk assessment tools to better identify women at increased breast cancer risk who may benefit from increased surveillance or chemoprevention.

Off-campus Download

Share

COinS