Molecular characterization of breast tumors in relation to risk factors and survival
Breast cancer is not a single entity and contains several subtypes that have different etiologies and prognoses, ranging from highly fatal to fully curable cancers. The overall goal of this proposal is to understand the determinants of risk and prognosis of breast cancer subtypes. We plan to achieve this goal through two interrelated projects:
1. Development and validation of subtype-specific risk and prognostication models for breast cancer subtypes. This work will be done in collaboration with the Breast Cancer Stratification (B-CAST) Project, a recently funded EU Horizon 2020 program grant embedded within the infrastructure and collaborations of the Breast Cancer Association Consortium (BCAC). B-CAST will evaluate relationships between risk factors, tumor subtypes and their clinical prognosis through the molecular characterization of ~20,000 tumors from large-scale epidemiological studies. Molecular characterization of breast tumors will involve staining and scoring of 15 immunohistochemical (IHC) markers on tissue microarrays (TMAs) and targeted sequencing of paired somatic and germline DNA. Risk factor information will include data on environmental factors (broadly defined as reproductive and hormonal history, lifestyle) and OncoArray genotyping data. In addition to providing biological insights, this project will translate this knowledge to expand and improve risk prediction and prognostication tools for breast cancer. B-CAST also aims to disseminate research findings and tools to the relevant stakeholders for implementation of evidence-based risk-stratified strategies for prevention, early detection, and clinical management of breast cancer. PLCO is already part of the OncoArray genotyping project for which we selected 3,136 cases and 3,136 controls using incidence density sampling. We are currently using data from this nested study to calibrate new risk models integrating data on environmental and genetic risk factors (manuscript in preparation). The addition of tumor data will allow us to contribute to the consortial efforts in B-CAST. Given the prospective design of PLCO, a key contribution of PLCO will be on the calibration of subtype-specific models.
2. Relationship of circulating estrogen metabolites and molecular characteristics of breast tumors. In a recently completed meta-analysis of four cohort studies (including PLCO) relating endogenous estrogen metabolites to breast cancer risk, we found that among postmenopausal women with similar overall estrogen levels, an increase in 2-hydroxylation pathway estrogen metabolites was significantly associated with reduced breast cancer risk (manuscript in preparation). These data support the hypothesis that much of the carcinogenicity of estrogens relates to their mitogenic effects, thus acting as relatively late-stage promoters. There have been spirited debates within the laboratory community about which mechanism (mitogenic or mutagenic) might be more important in estrogen-mediated carcinogenesis; some believe that both are likely to be involved. We aim to determine whether pre-diagnostic circulating estrogen metabolites are related to breast cancer subtypes and mutational signatures in the tumors that later develop.
Specific Aims
1. To stain and score a 15 immunohistochemical (IHC) marker panel on breast cancer tissue microarrays (TMAs). The panel will include established markers ER, PR, HER2, CK56, EGFR, Ki67, BCL-2, as well as additional markers being determined as part of the B-CAST project. Marker selection is based on to identify ???
2. To perform targeted sequencing of paired somatic and germline DNA of ~555kb B-CAST panel including 80 genes and copy number variation (CNV) probes. B-CAST panel genes are selected by examining whole exome sequencing (WES) data on 11 published studies (1,405 cases) based on mutation frequency, as well as Genome MuSiC and MutSigCV to identify drivers.
3. To determine circulating estrogen metabolites in pre-diagnostic serum samples from cases with breast tumor tissue. Liquid chromatography-tandem mass spectroscopy assays will be performed at the Laboratory of Proteomics and Analytical Technologies, Cancer Research Technology Program, Leidos Biomedical Research, Inc. (Frederick, MD) as described previously in a PLCO report (Fuhrman, B. J., C. Schairer, M. H. Gail, J. Boyd-Morin, X. Xu, L. Y. Sue, S. S. Buys et al. . 2012. Estrogen metabolism and risk of breast cancer in postmenopausal women. J Natl Cancer Inst 104 (4):326-339. doi:10.1093/jnci/djr531).
4. To evaluate relationships between tumor markers and risk factors, including reproductive history, lifestyle and germline genetic variation, and prognosis (overall and breast cancer specific mortality).
5. To evaluate relationships between tumor markers and circulating estrogen metabolites.
6. To develop and validate risk and prognostication models integrating information on risk factors, prognosis and tumor markers.
Gretchen Gierach (National Cancer Institute)
Bob Hoover (National Cancer Institute)
Montserrat Garcia-Closas (National Cancer Institute)
Rose Yang (National Cancer Institute)
- Use available tumor RNA samples from PLCO breast cancer cases to run a NanoString targeted gene panel assay
- Application of digital pathology to characterize breast tumor microenvironment and its relationship with risk and survival