Breast Cancer Susceptibility Genes in Very High Risk Women

Institution: University of Southern California
Investigator(s): Ann Hamilton, Ph.D. -
Award Cycle: 1999 (Cycle V) Grant #: 5IB-0116 Award: $129,247
Award Type: IDEA
Research Priorities
Etiology and Prevention>Etiology: the role of environment and lifestyle

Initial Award Abstract (1999)
A woman with relatives who have had breast cancer has a higher risk of developing the disease than a woman who does not have relatives with the disease. However, within identical twin pairs, the twin of a breast cancer case has an even higher risk of developing the disease than other women with a family history of breast cancer. We have developed a large twin registry and have identified nearly 1200 identical twin pairs in which one twin has had breast cancer and 276 pairs in which both twins have had breast cancer. Twins in the registry completed a questionnaire when they first volunteered to be part of the registry. We have studied the responses within pairs where both had breast cancer and we found that some well known risk factors for breast cancer, especially those related to earlier development and onset of menstruation, were strongly related to the twin who developed breast cancer first in the pair. Since these women are at a high genetic risk of disease, we want to test our hypothesis that genetic factors related to estrogen metabolism may occur more often in these pairs than in others where breast cancer is not as common.

To test our hypothesis we need to obtain DNA from tissue blocks that were preserved when the breast cancer was first diagnosed and treated. In an earlier study we have begun to obtain these blocks and will complete the process with this study. In total, we will obtain tissue from 200 pairs of identical twins where both have breast cancer, 200 pairs where one has breast cancer, and 100 pairs in which neither have had breast cancer. We will compare the frequency of one genetic polymorphism (called the A2 allele on the CYP 17 gene) among these groups. This genetic polymorphism has been shown to be related to higher levels of estrogen in some women. We want to determine if this genetic factor occurs more often in the pairs where both have breast cancer than in the other pairs.

This study is very important because not much is known about specific genetic factors related to breast cancer, other than the widely publicized BRCA1/2 genes. While those genes carry a high risk of developing breast cancer, they are not very common in the population and do not account for all of the cases who have reported having a family history of the disease. We know that young women with breast cancer are more likely to have a genetic factor related to the development of their disease; however, the actual number of cases caused by genetic factors is much greater among those diagnosed after menopause, because breast cancer is so much more common then. The genes responsible for this late disease are not known, but are more difficult to identify because only a small proportion of those with the gene get the cancer, presumably because other factors are required. The twins in this study represent a high genetic risk group and the estrogen related risk factors that we have seen are especially strong in the older twins. Thus the study can contribute important knowledge to identifying the additional genetic factors related to breast cancer.

Final Report (2002)
A woman with a family history of breast cancer has a higher risk of developing breast cancer than a woman without a family history. Among identical twin pairs, those in which breast cancer has occurred in both twins (concordant pairs) have greater genetic predisposition than other women with a family history. The genetic factors previously identified (i.e., BRCA1/2) do not account for all cases with a genetic risk. Thus, concordant twin pairs are an important group in which other genetic factors can be studied. This study utilizes breast cancer twins from a large registry of twins with cancer to test whether suspected genetic factors occur more frequently in concordant pairs than in pairs with only one member affected (discordant pairs) or in control women (without breast cancer).

In particular, this study has focused on CYP17 which is a gene in the estrogen metabolism. The CYP17 gene codes for an enzyme (called cytochrome P450cl7a) that helps regulate estrogen biosynthesis and metabolism. The gene, located on chromosome 10, has two copies or alleles. A polymorphism or variation of the gene has been shown to increase the amount of estrogen produced. One allele carrying this polymorphism has been called the A2 allele and the version of the gene without it is called the Al allele. There are three combinations of these alleles that a person can have: A1/A1, A1/A2, and A2/A2. Women who have at least one copy of the A2 allele have been shown to have higher levels of estrogen in some studies, while other studies have found increased levels only among those with two copies of the A2 allele. Other studies which have looked at the association of having the A2 allele with risk of breast cancer have shown mixed results. If the A2 allele occurred more frequently in the breast cancer concordant twin pairs, then this would provide evidence supporting a genetic risk for breast cancer associated with CYP 17.

DNA has been obtained from tissue blocks that were preserved when the breast cancer was first diagnosed, or from buccal smears from control women. The goal was to obtain tissue or buccal smears from 200 concordant pairs of identical twins, 200 discordant pairs, and 200 control women. A previous (non-BCRP study) obtained tissue blocks from some of the pairs and additional tissue blocks have been obtained during the current study. We have received signed informed consents and tissue blocks or buccal smears from 136 discordant pairs, 124 concordant pairs, and 105 control women. Additional follow-up of twins, completion of consent forms, and recruitment of control women is ongoing under a second (non-BCRP) grant obtained to continue this work. The CYP17 laboratory work has required additional testing to assure accuracy of results. We previously tested samples from one member of 167 pairs (97 discordant and 70 concordant) using conventional PCR methods and found some inconsistencies when retesting a sample. We found that this was due to some difficulties in using the archived tissue samples. As a result, to assure accuracy, we are retesting all archived tissue samples 4-5 times using a linear PCR method. At present, samples from 29 discordant and 29 concordant pairs have been retested with this more rigorous procedure. In addition, samples from 83 control women have been tested. Results from these samples has shown that the proportion of women with at least one A2 allele was higher in the affected pairs compared to our unaffected women (76% in discordant pairs, 69% in concordant pairs, and 53% in control women, p=.058); however the proportion in the affected pairs was similar to control women from other studies. We will also compare the prevalence of the CYP17 genotype with other tumor characteristics (ER and tumor grade) and breast cancer risk factors (age at diagnosis, family history, age at menarche, parity, OC use).

We are unable to make final conclusions at this time regarding the CYP17 polymorphism and breast cancer risk. Our final results are pending the completion of the laboratory analyses on all samples. This resource of DNA from these twins will be used to test other candidate genes in the future.