DNA Polymorphisms and Breast Cancer in a Multi-ethnic Cohort

Institution: University of Southern California
Investigator(s): Brian Henderson, M.D. -
Award Cycle: 2000 (Cycle VI) Grant #: 6IB-0070 Award: $80,949
Award Type: IDEA
Research Priorities
Etiology and Prevention>Etiology: the role of environment and lifestyle



Initial Award Abstract (2000)
The sex steroid hormone estrogen is believed to play a crucial role in the development of breast cancer as it promotes cell division in the breast. It is hypothesized that multiple, sequential cell divisions cause an accumulation of genetic injuries and can result in the production of a cell with a growth advantage over neighboring cells. A breast tumor itself represents an overgrowth of clones arising from such a single progenitor cell.

Genes are the most basic unit of information and can be translated into protein products that carry out the body’s essential functions. Any given gene is defined by its nucleic acid sequence; it is the goal of the Human Genome Project to catalog the sequence and function of all genes in the human genome. In some genes, the nucleic acid sequences may vary at a relatively high frequency among individuals and are termed polymorphic. This proposed project intends to investigate the role of variations in candidate genes involved in the production, activation, or degradation of estrogen. It is hypothesized that polymorphisms in candidate genes will modify a woman’s risk of developing breast cancer over the lifetime.

The first objective of this proposed project is to discover variations in the sequence of nucleic acids in identified candidate genes. To this end, a total of 800 women (400 breast cancer cases and 400 controls) were identified from an ongoing multi-ethnic cohort based in Hawaii and Los Angeles studying diet and cancer. The biologic material required for this effort is already available. Secondly, standardized statistical methods will be employed to quantify the relationship between discovered genetic variations and the risk of breast cancer.

The core of this proposal lies in the hypothesis that even minute variation in candidate gene sequences may vary circulating estrogen levels. Elevated estrogen levels, extended over a woman’s lifetime, has the potential to significantly increase her risk of developing breast cancer. As such, findings that support a role for any or a combination of these candidate genes permit the development of a genetic model of breast cancer. Such a model can provide a basis for targeting "at risk" populations through screening as well as designing appropriate preventive and therapeutic strategies. By investigating these variations in a multiethnic population, results can be widely generalized and may potentially contribute to a better understanding of the etiology of breast cancer.


Final Report (2001)
The sex steroid hormone estrogen is believed to play a crucial role in the development of breast cancer as it promotes cell division in the breast. It is thought that multiple, sequential cell divisions cause an accumulation of genetic injuries and can result in the production of a cell with a growth advantage over neighboring cells. A breast tumor itself represents an overgrowth of clones arising from such a cell. In this current study, we proposed to investigate the role of variations among candidate genes involved in the production, activation, or degradation of estrogen. It has been hypothesized that polymorphisms in candidate genes will modify a woman's risk of developing breast cancer over the lifetime.

Our initial aim was to discover common DNA single nucleotide polymorphisms (SNPs) in the coding regions (exons) and non-coding (introns) regions of such candidate genes and, secondarily, to comprehensively test the association between each SNP and breast cancer. Our final goal was to develop and expand a model of genetic susceptibility to breast cancer.

To this end, we selected a nested case/control sample of female participants from a large, ongoing multiethnic cohort study in Hawaii and Los Angeles, California. This study included a total of 854 women of which 428 incident breast cancer cases and 426. Four ethnic groups were considered in this investigation: African-American (n=234), Latina (n=200), Japanese (n=200) and Caucasian (n=220). Stage of disease was characterized as "localized" (n=206) or "high stage" (n=222)- which included regional or metastatic disease.

We initially focused on the CYP17 gene, a candidate gene that occupies a central role in sex steroid hormone synthesis as its product is a single microsomal enzyme that catalyzes two distinct steroid biosynthetic pathways. A total of 9 SNPs were considered—of these 1 localized to the 5' untranslated region (upstream from the start of the gene, expressed in mRNA but not in the protein itself), 2 were in coding regions (exons), 5 were in non-coding regions of the gene (introns) and the final variant was 3' of the gene (beyond its termination). The prevalence of variants ranged in frequency from 3.9% to 38.1% among controls for all ethnicities combined. It appeared that several of the common polymorphisms were over-represented both among low stage and high stage breast cancer.

Subsequently, we similarly examined the ESRI gene which encodes a similar form of the estrogen receptor - also a promising candidate locus in the pathogenesis of breast cancer. Estrogen receptor not only has a biologic role in normal breast cell development and function but is also a key prognostic indicator for hormonal therapy in breast cancer. We genotyped a total of 32 markers across this gene. We observed that ESRI was structured into discrete haplotypic "blocks" spanning 11 to 101 kb. The number of haplotypes in each block varied greatly by ethnicity, with African-American women displaying the highest degree of haplotypic diversity (4-6 haplotypes) as compared to Caucasians (2-3). Differences in the distribution of haplotypes and individual markers among all breast cancer cases combined and among high stage only were examined.

Although no significant differences have been noted (p>0.05), we posit that genetic markers among candidates involved in the production, activation, or degradation of estrogen may provide us with a greater understanding of variation in breast cancer risk and potentially provide a basis for targeting "at risk" populations.