Breast and Colon Cancer Family Registries

For Families: About the Research

The Breast and Colon Cancer Family Registries offer unique opportunities for interdisciplinary and collaborative studies. Current research projects utilizing the registries resources span the fields of:

Gene-Environment Interactions

Breast Cancer Risk Modifiers in BRCA Mutation Carriers
Principal Investigator: Alice Whittemore, Ph.D.

Lifetime breast cancer risks among carriers of mutations of the BRCA1 and BRCA2 genes have been estimated at 40-80%. Thus, some 20-60% of carriers live to advanced ages without developing the disease, which suggests that other genes or personal attributes may modify carriers' risks. At present however, little is known about such personal characteristics. There is an urgent need to determine which, if any, modifiable lifestyle characteristics may alter a carrier's risk of developing breast cancer, to assist her in making rational, informed choices about such preventive options as prophylactic mastectomy. We will use uniformly collected data from young (aged < 50 years) case (N=425) and control (N=352) carriers of deleterious BRCA1 or BRCA2 mutations to evaluate associations between breast cancer risk and five modifiable characteristics:

• history of oral contraceptive use
• history of diagnostic or therapeutic chest irradiation prior to diagnosis
• alcohol consumption
• cigarette smoking
• physical activity patterns during puberty, young adulthood, and middle age

We will focus on carriers under age 50 years at diagnosis (cases) or interview (controls) who have participated in the B-CFR and in clinical studies in New York, Ontario and Australia. We will use unconditional logistic regression to estimate odds-ratios relating these attributes to breast cancer risk while controlling for potential confounders, and use robust variance estimators to account for any correlation present in attributes of related carriers. These data on modifiable characteristics from a large group of young carriers of BRCA1 or BRCA2 mutations represent a unique resource for advancing our knowledge about breast cancer prevention in premenopausal women at high risk for the disease. The proposed analysis will provide new information on alternatives to mastectomy as a preventive strategy for these women.

Gene Discovery

Genetic Linkage in Colorectal Cancer Families
Principal Investigator: John Potter, Ph.D.

Colorectal cancer (CRC) is a common, serious disease that clusters in families, such that individuals with an affected sibling are at almost 3-fold increased risk compared to those in the general population. Little, if any, of the observed familial clustering has yet been explained by shared environmental exposures. Known genetic syndromes such as hereditary nonpolyposis colorectal cancer (HNPCC) and familial adenomatous polyposis (FAP) are thought to account for less than 2% of cases. Unidentified susceptibility loci are probably important in much of the remaining non-syndromic familial colorectal cancer.

We aim to identify novel CRC susceptibility loci using families collected via the C-CFR. Recruited CRC families, who are not shown to carry HNPCC- or FAP-predisposing mutations, will be included in a two-stage gene-mapping strategy. First, we will perform a genome-wide linkage analysis using 528 affected relative pairs in 323 families. Approximately 400 microsatellite markers will be genotyped and assessed for evidence of linkage to CRC using parametric and non-parametric methods; chromosomal regions will be identified for follow-up. Second, individuals from 323 families will be genotyped using more densely-spaced microsatellite markers in genomic regions suggested by the initial scan. Parametric and non-parametric linkage methods will assess evidence for CRC linkage.

Strengths of this effort include use of an existing CRC family collection, a wealth of preliminary data (including screening for known mutations), and a successful collaborative history among investigators. A key future aim is to combine these families with those of other CRC linkage studies in the US and UK to amass an extensive resource with further increased power to understand CRC genetic susceptibility.

Clinical

Prognosis in BRCA1 Associated Breast Cancer
Principal Investigator: Pamela Goodwin, M.D., M.Sc.

An international population-based, prospective cohort study, conducted in Northern California, Australia and Ontario (Canada), will build upon activities of the B-CFR to study women with BRCA1 associated or sporadic breast cancer to address the following objectives:

• To examine the impact of germline mutations in BRCA1 on risk of distant recurrence and death in women with newly diagnosed locoregional breast cancer (examination of the impact of mutations on distant recurrence is our primary objective).
• To explore the frequency and prognostic importance of traditional pathologic factors and selected molecular markers (p53, HER2/neu) on distant recurrence and death in BRCA1 mutation carriers.
• To describe the associations of treatment (systemic therapy, surgery, radiation) with outcomes (local breast events, contralateral breast events, distant recurrence, death) for BRCA1 mutation carriers and sporadic cases in order to generate hypotheses regarding treatment effects that can be tested in future randomized trials.

Specifically, we will review medical records to collect detailed clinical, treatment, and outcome information on women enrolled onto the prognostic study and we will perform a series of immunohistochemical assays for hormone receptors, p53 and HER2/neu. Additional information will be drawn from previously funded B-CFR activities including detailed pathology review, results of mutation analysis, family history and demographic data. Data from these two sources will be incorporated into the statistical analysis which will address our three study objectives.

Our proposed research improves upon existing research because it is population-based, prospective, involves rigorous control of clinical and pathologic prognostic factors, and includes sufficient subjects with BRCA1 associated hereditary breast cancer (100-140 women) that objectives can be addressed with adequate power (>80% power to detect hazard ratios under 2.0). It will also lead to the establishment of a well characterized cohort of over 3,000 women with standard clinical and pathologic data that can be used as a resource for future research into prognostic effects of mutations in BRCA2 (once testing is complete) and in other yet to be discovered breast cancer predisposition genes.

Genetic Characterization

ATM and CHK2 Mutations in Population-Based Families from the Breast CFR
Principal Investigator: Jonine Bernstein, Ph.D.

It was recently reported that two specific mutations of the ATM gene (7271T>G and IVS10-6T>G) are associated with substantially increased breast cancer risk among members of multiple-case families. Given this high estimate of breast cancer risk, it is important to verify these findings in other population-based families. In addition, the CHK2 gene, which encodes a substrate for the ATM kinase, has also been reported to have a specific mutant allele, 1100delC, that is associated with breast cancer in families negative for BRCA1 and BRCA2 mutations. We are screening the case probands of three B-CFR sites (the Australian Breast Cancer Family Registry, the Northern California Family Registry, and the Ontario Registry for Studies of Familial Breast Cancer) for these three mutations in two genes whose products are related by a common biochemical pathway involved in the cellular response to ionizing radiation. We also are screening the family members of those case probands found to carry a mutation for the same mutation. We will estimate the prevalence among cases as a function of family history of breast cancer and the penetrance of these mutations, and examine the possibility of a modifying effect of medical radiation exposure on the penetrance of these mutations.

Behavioral and Prevention:

Promoting Colon Cancer Screening Among Genetically Defined High-Risk Populations within the Colon Cancer Family Registry
Principal Investigator: Dennis J. Ahnen, M.D.

The primary purpose of this project is to test the hypothesis that a tailored, telephone-based educational and barriers counseling intervention will improve the adherence to colonoscopic screening in members of high-risk families. The secondary goals of the proposal are to generate additional natural history information about the relationship between screening intervals and colonic pathologic findings in these high-risk families to make more informed recommendations about colonoscopy screening intervals.

The study population include at risk members of both very high-risk (carriers of a known predisposing HNPCC gene mutation and families that meet the Amsterdam-I or the Amsterdam-II criteria) and high-risk (families that do not meet the Amsterdam-I or -II criteria but who have multiple FDRs with colorectal cancer or have one FDR with colorectal cancer at a young age (<50 years) families.

The study design is a controlled trial of telephone interviews (control group) vs. interviews plus a telephone-based educational and barriers identification and counseling intervention with a reinforcement mail-out tailored to that individual’s self-reported barriers (intervention group). The primary endpoints of the study are self-reported and documented colonoscopic screening behavior. The secondary endpoints are the incidence of colonic neoplasia (adenomas and carcinomas) in the two groups and the relationship between screening intervals and the incidence and stage of colonic neoplasia.

Minorities and Special Populations

Epidemiologic Research on Ethnic/Racial Minorities in the Colon CFR
Principal Investigator: Loic Le Marchand, M.D., Ph.D.

The frequency of colorectal cancer (CRC) varies four-fold among US ethnic/racial minorities, with Japanese Americans and African Americans having a marked excess risk for the disease. Moreover, in contrast to Japanese American patients who experience a better survival than whites, African Americans with CRC present at a later stage and have a shorter, stage-adjusted survival than whites. The poor outcome of African American patients is not completely explained by their reduced access to medical care. We will use data and samples collected from a total of 1,514 African Americans, 745 Japanese Americans and 750 white population-based families in the C-CFR to test the following hypotheses:

• DNA mismatch repair deficiency is a more common pathogenic pathway to CRC in African Americans, compared to whites, as reflected by a greater prevalence of tumor microsatellite instability (MSI).
• Functional gene variants and environmental exposures related to chemical carcinogenesis and methyl group metabolism are associated with an increased risk of MSI-high CRC, particularly in African Americans and Japanese Americans.
• Functional variants in cell proliferation- and angiogenesis-related genes are associated with risk of advanced-stage CRC and poor overall survival, particularly in African Americans.
• Functional variants in genes related to response to 5-fluorouracil/oxaliplatin chemotherapy are associated with shorter disease-free and overall survival, especially in African Americans.
• Other gene variants that are rare in whites but relatively common in African Americans and Japanese Americans are associated with CRC and may significantly contribute to the etiology of the disease in these groups.

The insight into biological pathways gained by this project may lead to new means of prevention and treatment that may help not only to alleviate health disparities in the US but also better control CRC among all ethnic/racial groups.