Breast cancer is a heterogeneous disease, manifesting as multiple clinical, histopathologic and molecular subtypes with different therapeutic responses and prognoses. Aberrant DNA methylation contributes to breast cancer through regulation of gene transcription and genome stability, and methylation of cytosine residues in CpG islands of gene promoters can silence tumor suppressor genes critical to breast cancer development and progression. Hypermethylated tumor DNA can be detected in the circulation of breast cancer patients, particularly those with more advanced disease;thus, circulating tumor DNA could be useful as a biomarker for cancer diagnosis, clinical staging, and to monitor therapeutic effects, systemic disease activity and recurrences. Sensitive methods for measuring DNA methylation could permit detection of occult or hidden cancer cells in biological samples such as plasma. In our recent study of array-based DNA methylation profiling of 519 invasive, mainly early-stage breast tumors from cases in the Carolina Breast Cancer Study (CBCS) (phase 1), we identified highly methylated genes significantly correlated with more advanced clinical stage or that predicted disease-specific survival, overall or among specific intrinsic tumor subtypes. The central hypothesis of this study is that DNA methylation patterns are important in distinguishing breast cancers from normal breast tissue and in defining breast tumor phenotypes, and sensitive methylation assays can be used to detect aberrant tumor-associated DNA in the plasma from breast cancer patients. Using 10 hypermethylated prognostic or stage-related CpG markers identified in breast tumors from CBCS (phase 1), we propose to conduct a pilot study to determine whether sufficient quantity and quality of DNA can be recovered from banked plasma samples from CBCS (phase 2) subjects for methylation studies, to optimize sensitive quantitative methylation-specific PCR assays for plasma methylation detection, and to compare these 10 methylation marker profiles in specimens from CBCS breast cancer cases and controls. Primary breast tumor, plasma and peripheral blood leukocytes (PBLs) will be evaluated in 100 breast cancer cases having varying clinical stages and intrinsic subtypes and who had their blood drawn prior to receiving chemotherapy. Plasma and PBLs will be evaluated from age and race-matched controls (n=50). Comparisons of marker methylation in plasma from cases and controls will provide an indication of its specificity for tumor DNA. In addition, we will determine if plasma methylation is correlated with methylation state in the matched tumor, or with characteristics of the cases including stage, node-positivity, age, race or subtype. Finally, analysis of PBL methylation will be examined as a possible source of contamination of plasma samples. We expect the results of this work will facilitate development of improved biomarkers for cancer diagnosis, treatment effects, and prognosis and disease recurrence in breast cancer.
Breast cancer is a heterogeneous disease, manifesting as distinct clinical and molecular subtypes with different therapeutic responses and prognoses, and aberrant DNA methylation contributes to breast cancer development and progression through effects on gene transcription. Methylated tumor DNA can be detected in the circulation of breast cancer patients, and thus could be potentially useful as a biomarker for cancer diagnosis, clinical staging, to monitor therapeutic effects, or detect disease recurrences. Using 10 methylation markers selected from our previous tumor methylation studies in the Carolina Breast Cancer Study (CBCS), we will conduct a pilot study to determine the feasibility of detecting tumor-specific DNA in the plasma of cases from the CBCS. Specifically, we will isolate DNA from banked plasma samples from CBCS phase 2 cases and controls, optimize sensitive quantitative methylation-specific PCR assays to detect circulating tumor DNA in plasma, and then assess methylation of these 10 markers in matched sets of primary tumor, pretreatment plasma samples and PBLs from 100 breast cancer cases as well as 50 plasma and PBLs from age/race matched controls. The primary goals will be to assess the specificity of plasma methylation markers for breast cancer cases and to evaluate their clinical correlates.