1.6 million women undergo breast biopsy for benign breast disease (BBD) annually in the United States. 90% of these show no evidence of cellular atypia, and are at only modestly increased risk for breast cancer, but this group represents the great majority who develop cancer among women with BBD, since women with identifiable risk factors, such as atypia or strongly positive family history are a small minority. BBD lesions in women who later progress to cancer very likely display molecular changes which predate overt cancer by years. An assay detecting such changes will provide a critical improvement in our ability accurately identify high risk women for preventive interventions (particularly if it distinguishes risk of estrogen receptor (ER) positive vs. negative breast cancer), and allow better targeting advanced surveillance strategies to women who will benefit the most. Aberrant promoter methylation has been documented very early in cancer progression. We have successfully adapted genome-wide array-based tools interrogating the methylome to formalin-fixed tissue samples, creating an opportunity to study archival samples with known long-term outcomes. For the past 15 years, we have built an integrated clinical-pathological relational database of all electronic data of over 75,000 breast patients at this institution since 1985. We identified women with non-atypical BBD who subsequently developed breast cancer (cases), and matched them to women with BBD who remained cancer- free (controls) with documented follow-up of a minimum of 10 years. In a pilot study, we demonstrated that molecular signatures in archival BBD tissue matched signatures present in subsequent invasive breast cancer (IBC); furthermore, these signatures were distinct from those in control BBD patients with no evidence of subsequent IBC, and differed in ER+ vs ER- cases. We now propose to combine our scientific and clinical resources to enable the discovery and validation of risk markers derived from BBD. We have identified over 500 cases with BBD predating IBC and paired IBC samples with known ER status with available archival tissue blocks. We will perform a case-control study of the methylomes of archival non-atypical non-familial BBD from 185 cases as well as their subsequent IBC, and 75 control samples, matched by follow-up, age and presence of hyperplasia. We will also capture known risk factors for IBC, mammographic density and body mass index, assess involution of terminal duct lobular units (TDLU), and characterize the status of key hormone receptors by semiquantitative immunohistochemistry. Our hypothesis is that molecular changes are present in BBD prior to the development of IBC, and are distinct in women who subsequently develop ER + versus ER- IBC. Our overarching goal is the identification of molecular changes in breast tissue that will allow the accurate risk stratification of all women undergoing breast biopsy showing no atypia. In the future, such biomarkers may be applicable to minimal samples of breast tissue such as those obtained with random fine needle aspiration or in nipple fluid.
The optimal breast cancer screening strategy for women is controversial, since regular mammography lacks accuracy, leading to cost and morbidity in women at low risk, while missing cases of potentially lethal breast cancer. Abnormalities in the methylation state of DNA occur very early in cancer progression, and are now measurable in benign breast disease (BBD) in women who subsequently developed cancer, and can be compared with the methylation state of those who did not. This study will lead to the development of clinically useful risk assessment tools, identifying molecular signatures of high risk women with more accuracy than present models; furthermore, these molecular signatures may differ for hormone positive and hormone negative breast cancer, which could substantially inform the choice of preventive strategies.
