Etiology: Finding the Causes
Over the past 50 years, scientists have identified a number of factors that increase cancer risk. They have used an epidemiologic approach, comparing a group of people with the disease to a cancer-free group for differences in environmental exposures, diet and other lifestyle factors. However, using this approach, they have been able to identify only a portion of the factors affecting breast cancer risk, and haven't been able to explain the biological mechanisms that trigger the disease. Within the last decade, epidemiologic methods have been combined with a new understanding of events at the cellular level. Scientists are investigating the likely role of genes in determining how a cell responds to an environmental exposure, and whether this response begins the cell's journey toward cancer or not. Increasingly, BCRP funds research that uses this combined approach, to apply the knowledge being gained in human genetics toward uncovering the precise causes of breast cancer.
Research Conclusions Research in Progress Research Initiated in 2000 |
Research Conclusions
Environment and Gene/Environment Interactions: Nature vs. Nurture
Bovine Leukemia Virus and Human Breast Cancer Risk
Tumor viruses such as Mouse Mammary Tumor Virus can cause breast cancer in mice. Gertrude Buehring, Ph.D., of the University of California, Berkeley investigated whether a similar phenomenon occurs in humans. Bovine leukemia virus (BLV) is found in cow's milk. Humans who drink cow's milk are exposed to the virus, but will only be affected if BLV can infect human cells. Dr. Buehring investigated whether human cells can be infected with BLV, by looking for the BLV genome in human cells and for an immune reaction to BLV. She developed a test to identify whether a person has been exposed to BLV and which cells are infected by it. The results indicate that BLV can indeed infect human cells. In future studies, she will determine whether there is a correlation between BLV infection and breast cancer development.
Hormones and Nutrition: Understanding the Modern Woman's Lifestyle
Estrogen Formation (Biosynthesis) and Breast Cancer Risk
A higher breast cancer risk is associated with a higher lifetime exposure to estrogen. An important class of genes produces enzymes and proteins that control the formation and action of estrogen. The more active the enzyme or protein, the more estrogen action, and the higher the risk. Gerhard A. Coetzee, Ph.D., of the University of Southern California Norris Cancer Center, Los Angeles focused on a gene called CYP17, which produces the enzyme that controls the ultimate formation of all the body's estrogen. Strong epidemiological evidence exists that this gene is involved in breast cancer risk. Dr. Coetzee identified several variants of this gene and tested them to better understand how these genetic variations could cause differences in the level of the enzyme or its activity. Although he failed to detect a molecular mechanism that underlies the epidemiologic association, he was able to exclude some obvious candidate mechanisms. He concluded that the differences must be too subtle to detect with today's technologies. Further studies are therefore needed to improve methods that would allow for a better understanding of how enzymes produced by the CYP17 gene relate to breast cancer risk.
Physical Activity, Diet and Menarche in a Multi-Ethnic Cohort
Women who have their first menstrual period at an early age appear to have an increased chance of developing breast cancer. So do women who have a larger lifetime number of menstrual cycles in which they ovulate (release an egg). Carol Koprowski, Ph.D., R.D., of the University of Southern California, Los Angeles studied the relationship between physical activity, diet, when a woman has her first menstrual period (menarche), and whether or not ovulation occurs during early menstrual cycles. The results indicated that girls who spent 13 or more hours per week in physical activity were more likely to have their first menstrual periods at later ages, compared to girls who spent less than five hours per week. The amount of calories the girl consumed did not appear to be an important factor in when a girl had her first menstrual period. However, girls with higher caloric intake were more likely to ovulate during their menstrual cycles. Girls who spent more hours in physical activity were less likely to ovulate. The results suggest that diet and physical activity during childhood and adolescence can affect breast cancer risk.
Role of Estrogen in the Origin of Breast Cancer
The estrogen receptor (ER) is a protein in breast cells that binds with the hormone estrogen. This binding triggers changes in the genes and other processes within the cell. The estrogen receptor can turn on genes through two different regulating sequences on the genes' structure, called ERE and AP1. Peter Kushner, Ph.D., of the University of California, San Francisco designed an estrogen receptor that turned on only the ERE or AP1 and examined how turning genes on at each location affected the development of the mammary gland and tumors in mice. His team made mammary gland cells with genes carrying extra AP1, and these cells divided at an abnormally fast rate in response to estrogen. After this study began, a second estrogen receptor beta (ERb) was discovered independently by other investigators. In an unplanned bonus to this study, Dr. Kushner found that ERb often activates the AP1 sites 10 times more efficiently than ER (now called ERa) in the presence of anti-estrogens. This may explain at least in part why cells develop resistance to the anti-estrogen chemotherapy drug tamoxifen. This function of ER b will be an important consideration when designing anti-estrogen drugs in the future. Dr. Kushner also found that a cancer gene, src, can strengthen the effects of estrogen on breast cancer cells. Breast cancer cells often have higher levels of src activity, indicating that src could provide a new target for anti-estrogen therapy. More on this work can be found in Science 277:1508-1510 (1997) and Clinical Cancer Research 5:251-256 (1999).
Radiation, Reproductive & Menstrual Factors & Breast Cancer
Women diagnosed with breast cancer at an age younger than 40 may be different from those diagnosed at an older age. Deirdre Hill, Ph.D., at the University of Southern California, Los Angeles is exploring reasons that may help explain cancer that develops before 40. These include having a close family member with breast cancer, having cysts or lumps in the breast (benign breast disease) and possible exposure to cancer-causing events or substances in childhood. Women who get X-rays at today's levels probably do not have a higher risk of breast cancer. However, Dr. Hill wanted to find out whether women who got X-rays in childhood (before age 20) in the past (when dose levels were higher), might have an increased risk, especially if they also have benign breast disease or a close family member with breast cancer. Overall, breast cancer risk was only slightly increased among women who received childhood radiation. However, among women with benign breast disease, breast cancer risk was 2.4 times higher in those who received radiation during childhood than those who had not. Among women without benign breast disease, childhood radiation exposure was not related to breast cancer risk. Women with breast cancer in a family member did not have an increased breast cancer risk following childhood radiation, unless they also had benign breast disease. Among women with both benign breast disease and a family member with breast cancer, breast cancer risk was 3.4 times higher. Radiation exposure at age 20 or older was not related to breast cancer risk among women with benign breast disease, but it was possibly related to increased risk among women with breast cancer in a family member. The radiation doses that women received in this study are much higher than those received today, so the increases in risk do not apply to current exposures. These results may contribute to further understanding of the causes of breast cancer, and to prevention for women at high risk.
Other Searches for the Causes
Predictors of Recurrent Breast Tumors In Women With DCIS
Ductal carcinoma in situ (DCIS), is a pre-malignant breast lesion, usually found through a mammogram. Some cases of DCIS turn into invasive breast cancer, and some will recur as DCIS after surgery. Others do neither. Currently, there's no way to predict what will happen for an individual woman, so there's no consensus about treatment. Some women have a mastectomy, others a lumpectomy, and others lumpectomy with radiation. Karla Kerlikowske, M.D., of the University of California, San Francisco investigated better ways to define which DCIS will recur after lumpectomy.
Working from data from medical records, Dr. Kerlikowske measured epidemiologic, clinical, tumor, and tumor function characteristics of 1060 women who had DCIS. A total of 208 (19.6%) women had developed recurrent disease in the same breast. Of these, 56% were DCIS and 44% were invasive cancer, 49 women (4.6%) developed recurrent disease in their other breast. The median time for recurrence was 73 months. Although BCRP funding for the study has ended, Dr. Kerlikowske will complete analysis next year on the epidemiology, pathology and genes or proteins that provide markers for cancer. This information will be essential for predicting the risk of recurrent breast tumors after lumpectomy. It will allow women at high risk of recurrence to consider radiation therapy and possibly mastectomy, and those at low risk to avoid receiving unnecessary treatment.
Research in Progress
Environment and Gene/Environment Interactions: Nature vs. Nurture
Mammographic Density, Cancer, Inheritance and Acquired Risk in Twins. Women whose breast tissue appears denser than average on a mammogram have a higher risk of breast cancer. Thomas M. Mack, M.D., of the University of Southern California, Los Angeles is trying to confirm whether this is true even within pairs of identical twins. He is also investigating the extent to which breast density is inherited. In addition, he is investigating whether certain adult exposures and experiences, known to be related to breast cancer risk, such as diet and taking hormone replacement therapy, modify breast density. He is obtaining risk factor information by interviewing both members of up to 2500 pairs of California twins, some of whom have had breast cancer. He is also using their mammograms to measure their breast density. He will then compare sets of identical and fraternal twins for breast density, past experiences, and the subsequent appearance of breast cancer. He has identified 3202 potentially eligible individuals from 1601 pairs, contacted 2133 of them, and obtained the cooperation of 1600 or 75% of those contacted.
Oral Contraceptives, Hormonal Risk Factors and BRCA1. A large number of breast cancers diagnosed at an early age may be due to mutations in the breast cancer susceptibility gene, BRCA1. Recent evidence suggests, however, that not all women with one of these mutations will develop breast cancer. Giske Ursin, M.D., Ph.D., at the University of Southern California, Los Angeles is conducting a large study to determine to what extent hormones, use of oral contraceptives, reproductive history, physical exercise and other non-genetic factors increase the risk of breast cancer in women with a BRCA1 mutation. She is using tissue samples from the women to sequence their BRCA1 genes, using the new ABI 3700 sequencer. She is investigating whether a certain type of mutation on the gene (non-truncating mutations) is important in causing breast cancer, and what role non-genetic factors may play in cancer development in women with this type of BRCA1 gene mutation.
Breast Cancer Susceptibility Genes In Very High Risk Women. A woman with relatives who have had breast cancer has a higher risk of developing the disease than a woman whose relatives haven't had the disease. Moreover, a woman whose identical twin has breast cancer is at an even higher risk than other women with a family history of breast cancer. Ann S. Hamilton, Ph.D., at the University of Southern California, Los Angeles is testing the hypothesis that a structure on the CYP 17 gene may occur more often in twins where one or both of the women have had breast cancer than in twins who haven't had breast cancer. She is currently recruiting study participants.
The Insulin-Like Growth Factor (IGF) System and Breast Cancer. The insulin-like growth factor (IGF) system consists of several proteins produced by many cells in the body and circulating in the blood. They can work together to increase or decrease cell division. The IGF system may contribute to breast cancer by causing breast cells to divide, either directly or by working along with estrogens. A recent study found a strong association between high blood levels of one protein in the IGF system, IGF-1, and increased breast cancer risk in pre-menopausal white women. Brian E. Henderson, M.D., also at University of Southern California, Los Angeles is investigating whether levels of IGF in the blood differ among women of varying ethnic groups. He is also investigating whether a repeat of a sequence [CA] in the structure of the IGF-1 gene leads to higher levels of IGF-1 in the blood. If so, he will investigate whether the sequence repeat is associated with postmenopausal breast cancer. In addition, IGFBP-3 is another protein in the system that binds with IGF-1. Dr. Henderson is investigating whether a single change in the structure of the gene that produces the IGFBP-3 protein is associated with postmenopausal breast cancer. Preliminary analyses found no differences in circulating IGF-1 levels in the blood samples from a multi-ethnic population (African American, Japanese, Latina) after controlling for age and weight. Preliminary analysis of blood from 460 healthy women did indicate racial and ethnic differences in mean blood levels of IGF-1 and IGFBP-3, with the highest values found among Japanese and African- American women and the lowest values among Latina women. Identifying the gene structures that determine IGF system levels may provide greater understanding of individual variations in breast cancer risk, and potentially serve as indicators for women who would benefit most from screening or specific types of therapy.
Unique Genes Expressed in Cancer Cells. Craig V. Byus, Ph.D., of the University of California, Riverside is trying to isolate the genes that are turned off or on by the enzyme ornithine decarboxylase (ODC) in cells. It appears that high levels of ODC turns on or off genes that allow a cell to become more like cancer. Dr. Byus engineered a breast cancer cell line that had high and stable levels of ODC activity, and identified a number of gene fragments that behaved differently in this high-ODC cell line, compared to a cell line that does not have high levels of ODC. However, he encountered difficulties in this approach. The procedure he used did not allow him to isolate entire genes, and the gene fragments appear to be due to the techniques he used in the experiment. He has engineered a new series of cell lines that do not have this problem, where the ODC can be artificially manipulated to any level. In the upcoming year, he will continue to analyze the genes turned on and off in response to high levels of ODC to provide a better understanding of the process by which normal breast tissue becomes cancerous.
Hormones and Nutrition: Understanding the Modern Woman's Lifestyle
Gene-Diet/Tobacco Interactions in Breast Cancer in Asians. Certain substances (heterocyclic aromatic amines or HAs; polycyclic aromatic hydrocarbons or PAHs; and arylamines such as 4-aminobiphenyls, or 4-ABP) are known to cause genetic changes and cancer in rodents. Humans are exposed to these substances through cooking, smoking, and working in some occupations. If these substances are involved in breast cancer, then variations in the structure of genes that produce the enzymes that metabolize these substances are likely to influence the risk for breast cancer. Anna H. Wu, Ph.D., of the University of Southern California, Los Angeles is investigating the roles of a series of these metabolism genes (NAT1, NAT2, CYP1A1, GSTM1, GSTT1, GSTP1) in a group of Chinese, Japanese, and Filipino-American women. Some of the women have had breast cancer and the control group has not. Dr. Wu is using a molecular biology based technique, polymerase chain reaction or PCR. Another year will be needed to complete collection of blood specimens, identification of the genes, and data analysis. This study will help to identify factors responsible for the increase in breast cancer in Asian-Americans and will also help to clarify the role of HAs, PAHs and arylamines in breast cancer.
Other Searches for the Causes
Breast Cancer Risk Factors: Lesbian and Heterosexual Women. Only a small amount of information is known about lesbians and breast cancer. However, scientists have proposed that lesbians' risk of getting the disease may be two to three times higher than that of heterosexual women. Suzanne L. Dibble, D.N.Sc., at the University of California, San Francisco and Stephanie Roberts. M.D., at Lyon Martin Women's Health Services, San Francisco clinic are trying to determine whether lesbians indeed have a higher risk. They are distributing surveys to lesbians age 40 and older throughout the state of California. They are also asking each lesbian participant to have a heterosexual female friend who lives in California fill out an identical survey, and if they have a sister, for the sister closest in age to fill one out, too. From questionnaires distributed, they have had a 33% response rate from lesbians (637 completed surveys), a 22% response rate from friends (434 surveys), and a 30% response rate from sisters (275 complete surveys and 301 with no sister). The typical lesbian responding to the survey is 50.4 years old, white (85.9%), educated (mean 17.1 years), employed (77%), and urban (53%). In the coming year, they will focus their efforts on reaching more lesbians of color; more elderly, less educated, and rural lesbians; and more lesbians with living sisters.
USC/Norris Breast Cancer Research Training Program. An important goal of the BCRP is to encourage and support the training in breast cancer for new research scientists. Ronald K. Ross M.D., and Michael Press, M.D., maintain the multi-faceted Breast Cancer Research Training Program at the University of Southern California, Los Angeles. The program is now in its fifth year. Faculty include epidemiologists, prevention scientists, behavioral scientists, tumor biologists and molecular geneticists, along with radiation, surgical and medical oncologists. The program encourages trainees to fully utilize the numerous patient and data resources available at the Norris Comprehensive Cancer Center. These include the SEER cancer registry; a large registry of breast cancer in twins; dietary and lifestyle data from large, racially and ethnically diverse groups of women; a large and diverse clinical population at affiliated hospitals; and several large tumor and other biological sample banks for tumor biology and molecular genetics studies. The trainees also participate in a variety of specialized activities related to breast cancer. This past year, five trainees were supported after being matched with an appropriate faculty mentor. Trainees are interdisciplinary in their interests, which include pathology, molecular and cell biology, and molecular epidemiology. Ongoing trainee research projects include: Migratory and Metastatic Potential of HER-2/neu; Improving Retroviral Vectors for Gene Therapy of Breast Cancer; and Anti-angiogenic Gene Therapy for Breast Cancer, among others.
Research Initiated in 2000
Environment and Gene/Environment Interactions: Nature vs. Nurture
Influence of Localized DDT Exposure on Breast Cancer. DDT is not eliminated from the body after exposure. It remains stored in the fat surrounding the breast tissue where cancer originates, and can act like hormones. So the question of whether DDT might increase breast cancer risk is of concern. Vicki L. Davis, Ph.D., at the Center for Women's Health, Cedars-Sinai Medical Center, Los Angeles will try, by clever experiments in animal models, to isolate the effects of the two main compounds formed when DDT is metabolized in the body. One of these compounds may stimulate cell growth. The other may interfere with the hormone androgen. Androgen is normally present in the tissue surrounding the breast cells; it inhibits the growth of cells prone to becoming cancerous. The results could shed light on whether DDT accumulation in breast tissue can influence the cell division processes that lead to breast cancer.
DNA Polymorphisms and Breast Cancer in a Multi-Ethnic Cohort. Naturally occurring variations in human genetic makeup not only account for the visible differences between people, but also influence risk for certain diseases. Such variations include those in the BRCA1 and BCRA2 genes, which are responsible for a small portion of inherited breast cancer risk. However, variations in many other genes may interact with each other and the environment to affect the risk of breast cancer. Brian Henderson, M.D., of the University of Southern California, Los Angeles will use a new, accurate, rapid, reliable and inexpensive method (single base extension--SBE--combined with fluorescence resonance energy transfer, FRET) to categorize variations in 25 genes that affect key hormone levels in the body. He is testing genes from 800 women: 200 African-American, 200 Asian-American, 200 Latina and 200 white. Half of each group have had breast cancer and half have not. Next, he will use statistical methods to quantify the relationship between genetic variations and the risk of breast cancer.
Hormones and Nutrition: Understanding the Modern Woman's Lifestyle
Mammography Density and Sex Steroid Genes. Sue Ingles, Ph.D., of the University of Southern California, Los Angeles will investigate whether some genetic variations could provide a measure of future risk of breast cancer. Women whose breasts appear to be more dense on their mammograms have a greater breast cancer risk. Since hormonal levels are known to influence breast density, variations in the genes that produce female sex hormones may be associated with breast density. If this is the case, it would help us better understand the role of these hormones in breast cancer. Dr. Ingles will also examine whether women with specific variations in their genes who also are on hormone replacement therapy have very dense breasts. This could provide preliminary evidence that might be used to identify women who are at higher breast cancer risk if they use hormone replacement therapy.
Physical Activity and Diet in Adolescents with Disabilities. Carol Koprowski, Ph.D., and K. Sarah Hall, Ph.D., at California State University, Northridge will develop assessment tools to measure physical activity and diet in adolescent girls with disabilities. The tools will be used in a future planned study of the relationship between physical activity, diet and serum hormone levels in this population. The information gathered would be useful for creating lifestyle intervention programs for this underserved group during a susceptible time in their physical development.
Genes Determining Estrogen Susceptibility in Breast Cancer. Many breast tumors are initially estrogen dependent and can be treated with anti-estrogen drugs. However, some tumors eventually progress to an estrogen-independent form and become drug resistant. Wensheng Wei, Ph.D., at Stanford University, Palo Alto will investigate possible mechanisms that might explain why. He will employ a recently-developed procedure to identify, isolate and map the structure of estrogen susceptibility genes (ESGs) that are required for breast cells to respond to estrogen stimulation. Dr. Wei hopes to ultimately identify the changes in the ESGs that lead to breast tumors becoming estrogen independent.
Tamoxifen-Induced Endometrial Cell Transformation. Zhimin Lu, Ph.D., of the Salk Institute for Biological Studies, La Jolla will try to understand the mechanism of the chemotherapy drug tamoxifen's severe side effects. He will also seek approaches for developing more effective breast cancer therapies and prevention without these side effects. Tamoxifen blocks estrogen. It also inhibits a group of enzymes called Protein Kinase C. One of these enzymes, PKC d, may suppress the development and growth of tumors, so inhibiting it may promote tumor formation. Dr. Lu will investigate whether tamoxifen, by inhibiting PKC d, does have a tumor-promoting effect on rat or mouse cells. He will also examine how tamoxifen in human uterine cells causes PKC d to fail to inhibit the cancer process. Finally, he will investigate which cancer-related genes are turned on and which tumor suppressor genes are turned off after tamoxifen treatment.
Other Searches for the Causes
Breast Cancer in California Teachers--Regional Variations. Large differences in breast cancer rates from one geographic area to another have long puzzled the concerned public and have generated many hypotheses about what might cause breast cancer. Most studies investigating geographic breast cancer patterns have used population-level data, with no individual data on personal risk factors. Peggy Reynolds, Ph.D., of the Public Health Institute, Berkeley will use the personal information available on 133,000 active and retired female California school employees participating in the California Teachers Study (CTS). She will determine the degree to which established and suspected risk factors explain the geographic patterns of breast cancer incidence observed in this group of women.
