Genetic and Environmental Modifiers of Breast Cancer Risk

Institution: University of California, Irvine
Investigator(s): Argyrios  Ziogas , Ph.D. -
Award Cycle: 2000 (Cycle VI) Grant #: 6PB-0061 Award: $287,979
Award Type: Request for Applications
Research Priorities
Prevention & Risk Reduction>Active lifestyle modification: what women can do



Initial Award Abstract (2000)
A major risk factor for breast cancer is family history of the disease. There has been significant progress toward the identification of genes responsible for inherited predisposition to breast cancer. In the last few years the BRCA1 and BRCA2 mutations were have been identified and BRCA1 appears to be responsible for the majority of families where there is early age of onset of breast and ovarian cancer. Original estimates suggested those breast cancer families with BRCA1 and BRCA2 mutations account for about 50% of the hereditary component of breast cancer. However, more recent studies suggest that BRCA1 mutations account for only 10-20% at most of inherited breast cancers, and that BRCA2 mutations account for half this fraction of families.

The purpose of this proposal is to examine the associations of genes that interact with environmental carcinogens to cause breast cancer. In particular, we utilize an existing family-based breast cancer registry consisting of 1176 families to conduct the analysis of gene-environment interaction. The specific aims are: 1) Examine whether the polymorphic variants of genes involved in the metabolism of estrogen are associated with breast cancer; 2) Examine whether these polymorphic variants of genes that are involved in the metabolism of environmental carcinogens, are associated with breast cancer; 3) Explore whether the association between an estrogen- or carcinogen-metabolism gene and breast cancer varies with respect to lifestyle (reproductive, hormone-related and environmental risk factors) as well as genetic (BRCA1/BRCA2 or other metabolism gene) factors; and 4) Utilize the existence of a well-characterized population-based breast cancer family registry to examine associations and interactions between breast cancer risk and genetic or environmental variants. In particular, we will use family based designs that are appropriate for genetic studies.

Results will enable us to improve individualized risk prediction and assist in the timing and targeting of preventive strategies, and will add to the understanudying of the biological basis of BRCA1/2 caused carcinogenesis and interactions between the estrogen- and carcinogen-metabolism genes and their interactions with the reproductive, hormonal and environmental risk factors as well as the major susceptibility genes (BRCA1 and BRCA2).


Final Report (2005)
A major risk factor for breast cancer is family history of breast cancer among first-degree relatives. While 20% of all breast cancer cases have positive family history of breast cancer, a very small percent 5-10% of the familial breast cancer is accounted by the two major susceptibility genes BRCA1 and BRCA2. Thus there is a large percent of familial breast cancer that is not explained by BRCA1 and BRCA2 genes. Since human breast cancer is both a genetic and environmental disease, other genes that they distribute more frequent in the population and their interactions with lifestyle exposures may be involved in the development of breast cancer.

In this study, we examined the mechanism by which genes that are common in the population and are involved in the metabolism of estrogen or the metabolism of environmental carcinogens may be associated with breast cancer.

Most of the studies that have looked at the above questions have been done either on a small numbers of cases or on a convenience sample. We utilized the existence of a well-characterized population-based breast cancer family registry to examine associations and interactions between breast cancer risk and genetic or environmental variants. We ascertained all female invasive breast cancer cases from Orange County, California, diagnosed between March 1, 1994 to February 28, 1995. There were 1479 such invasive breast cases, of which 1008 breast cancer patients agree to participate in our study and completed a family history interview. Participants and non-participants were statistically significant different with respect to the age at diagnosis (p<0.0001), with younger probands (a proband is the affected individual through whom a family with a genetic disorder is ascertained) having higher participation rates and stage of the disease (p<0.0001). Those with localized disease had higher participation rates compared to regional or metastatic disease. No statistically significant differences were observed with respect to race/ethnicity.

In our analysis, we used a number of different study designs and we examined the relative efficiency of those designs. These designs included: 1) studies where we compared breast cancer cases to their older unaffected sisters or cousins with respect to genes and lifestyle-factors; 2) studies where we compared breast cancer cases to unaffected women with respect to genes and lifestyle-factors; and, 3) studies that incorporate the history of cancer experience among the relatives of breast cancer cases (first-degree and second-degree). We observed statistical significant interactions with respect to breast cancer risk between progesterone use and COMT (p=0.05) and progesterone use and CYP1A1 (p=0.05). There was a statistical association between the COMT variant and progesterone use (the percent of COMT L-L among users was 18.1% compared to non users where the percent of COMT L-L was 26.9%). Similarly for CYP1A1 the percent of WW among users was 86.3% compared to 78.7% among non-users. (This case-case analysis does not estimate the risk (odds ratios) but tests the association of the variant and the risk factor conditional that they are independent on the source population.)

In addition, we observed significant interactions between BMI at 18 years old and CYP1A1 (p=0.0141) and GSTM and oral contraceptive use (p=0.0008). These results indicate that women with the CYP1A1 variant {WM or MM) were more likely to have smaller BMI at 18 years old (22.6% with BMI <19, 23.0% with BMI 19-21, and 11.7% with BMI at 18 >21). Similarly, there was smaller number of women with the null GSTM among oral contraceptive non-users 54.4% compared to users 41.0%.

No significant interactions between smoking and GSTM or GSTT were observed

Using a case control study design where cases were unaffected women were randomly selected from the population, we performed multivariate analysis where we adjusted for age (age at diagnosis for cases and age at interview for controls) and race/ethnicity. We observed a statistically significant effect with the GSTT gene (p=0.049). There are two forms of this gene the genotype (MM) among controls was 16.6% compared to 22.9% among cases. A woman will be at increase risk for breast cancer if she was a carrier of the variant on the GSTT gene (MM).

Using relatives of cases as controls we observed no statistical significant associations. This was expected since relatives of cases (sisters or cousins) are more likely to share common genetic variants with the proband (case that were compared) compared to population controls since relatives share the same ancestry.

We also conducted a comparison of the family-based and case-control study designs with respect to relative efficiency. In general, both designs showed similar estimates for the odds ratios however, the family-based design was more efficient (as measured by the ratio of the variances of the estimates). The efficiency was ranging from 16%-155% with the family design being more efficient when the allele frequency was smaller. Since our samples were population based we estimated that the frequency of BRCA1 and BRCA2 mutations among breast cancer patients is 1.3% and 1.9% respectively.

Results of this study will enable us to improve individualized risk prediction and assist in the timing and targeting of preventive strategies.


Symposium Abstract (2003)
A major risk factor for breast cancer is family history of breast cancer among first-degree relatives. While 20% of all breast cancer cases have positive family history of breast cancer, a very small percent (5-10%) of familial breast cancer is accounted for by the two major susceptibility genes BRCA1 and BRCA2. Thus there is a large percent of familial breast cancer not explained by BRCA1 and BRCA2 genes. Since human breast cancer is both a genetic and environmental disease, there are likely to be other genes that are distributed more frequently in the population and are involved in the development of breast cancer when they interact with environmental or lifestyle factors.

In this study, we set out to examine the mechanism by which genes that are common in the population and are involved in the metabolism of estrogen or the metabolism of environmental carcinogens are associated with breast cancer. In addition, we will explore whether the association between an estrogen- or carcinogen-metabolism gene and breast cancer varies with respect to reproductive, hormone-related, and environmental risk factors.

Most of the studies that have looked at the above questions were done either on small numbers of cases or on a convenience sample. We will utilize the existence of a well-characterized population-based breast cancer family registry to examine associations and interactions between breast cancer risk and genetic or environmental variants. In particular, we will use family-based designs that are appropriate for genetic studies. These designs include: (1) studies where we will compare breast cancer cases to their older unaffected sisters or cousins with respect to genes and lifestyle factors; (2) studies where we will compare breast cancer cases to unaffected women with respect to genes and lifestyle factors; and (3) studies that incorporate the history of cancer experience among the relatives of breast cancer cases (first-degree and second-degree).

Results of the study will enable us to improve individualized risk prediction and assist in the timing and targeting of preventive strategies. We will also add to the understanding of the biological basis between breast cancer and genes involved in estrogen- and carcinogen-metabolism and their interactions with reproductive, hormonal, and lifestyle risk factors.