Session 3: Epidemiology & Molecular Epidemiology of Breast Cancer
Facilitator: Leslie Bernstein (University of Southern California)
- Building on what we know from past breast cancer epidemiology, how can we incorporate new information and tools to more fully understand individual susceptibility to environmental carcinogens?
- How can we incorporate emerging information on environmental response genes and breast cancer genetics into new epidemiology studies?
- What new methods/tools do we need to incorporate into future research to improve the assessment of exposures across the lifespan?
The third workshop session focused on molecular epidemiology and exposure assessment in breast cancer research. Leslie Bernstein and Richard Santen provided some background for this discussion. Bernstein classified risk factors as modifiable or non-modifiable, although she acknowledged that this classification is in some ways arbitrary and subjective. Some of the risk factors are gender, age, race/ethnicity, breast feeding history, family cancer history, age at menarche, age at menopause, pregnancy history, obesity, radiation exposure, exogenous hormones (i.e., oral contraceptives and replacement therapy), breast density, diet and alcohol consumption. Other external environmental exposures may also play a significant role including estrogenic compounds (pesticides, organochlorines, PAHs, plastic additives) smoking, electromagnetic fields, NSAIDs, antidepressants, sunlight and vitamin D. Breast cancer is complex, at both the biochemical and the biological/developmental levels.
Bernstein outlined an approach for studying breast cancer epidemiology involving the following steps: search for risk factor, search for biological relationship between risk factor and breast cancer, search for consistency in relationships, develop hypothesis and identify related risk factors, carry out intervention trials to test hypothesis. Accurate exposure assessment is important in epidemiological studies of breast cancer. Study designs should consider genetic factors, age and ethnicity.
Breast cancer epidemiological studies would benefit greatly from identification of useful biomarkers of susceptibility, sensitivity, exposure or effect. Suitable parameters might be a circulating protein or growth factor, breast mammographic density or breast volume. Several genes have been identified as potential biomarkers of susceptibility to breast cancer (i.e., NAT1, NAT2, GSTM, XRCC1, RAD51, CYP genes, ER, PR and others). Gene-gene and gene-environment interaction studies will be needed to understand the significance of polymorphic variants in these genes. Bernstein indicated that large numbers would be needed in epidemiological studies analyzing multigene and complex gene-environment interactions that play a role in breast cancer. Bernstein acknowledged that to date, breast cancer data lacks consistency, and she attributed the inconsistency to insufficient basic understanding of breast cancer biology. This lack of knowledge limits the researcher's ability to generate valid testable hypotheses. Bernstein also pointed to the lack of a systematic approach in breast cancer research.
Key Discussion Points:
- Breast cancer clusters
Concerns were raised about whether research on breast cancer clusters is particularly informative. Individuals have high mobility, and they are exposed to many different environments at different life stages. Therefore, cumulative exposure in different environments can influence their health status. In addition, the individuals in a cluster are often highly heterogeneous. There is more variation in breast cancer incidence internationally than regionally or locally in the U.S. Some research on breast cancer clusters is recommended in the context of a broad research program on breast cancer etiology.
- Estrogen-negative breast cancer
Estrogen-negative breast cancer epidemiology is less well understood than estrogen-positive breast cancer epidemiology. Estrogen-negative breast cancer is understudied, and more research on this type of disease is warranted.
- Long-term, multigenerational, large scale prospective studies of breast cancer
It will be important to follow breast cancer risk in long-term studies. It may be useful to attempt to enlist the daughters of participants in clinical trials or research studies to assess periods of susceptibility. Funding for long term multigenerational studies could be an issue.
- Occupational exposure in the etiology of breast cancer
Women in the workplace are exposed to many chemicals. Occupational exposure should continue to be explored in breast cancer research.
- Partnership with NIEHS Children's Centers
The resources and expertise of the NIEHS Children's Centers could be useful. Partnerships and joint programs between NIEHS Children's Centers and Breast Cancer Research Centers should be considered to allow for follow up of children through puberty. Studies assessing biomarkers and hormone levels may be assessed for a picture of exposure to chemicals during specific periods of time.
- Identification of gene/polymorphisms for assessment in epidemiological studies
Researchers stated that, initially, studies should only look at genes and polymorphisms that are associated with a known functional status.
- Risk Estimation
Current risk estimation methods could help by providing updated/better models that indicate the proportion of risk that environment contributes to BCA. Estimates vary widely, and better information on this question would be helpful in many ways.
- Focus points
Researchers noted that four points that should be focused on for epidemiological studies are:
- Accurate exposure assessment
- Evaluation of individual exposures
- Persistence of chemicals and body burdens
- Identification and evaluation in intermediate markers