Etiology
In developing its priorities for research funding, the Breast Cancer Research Council (BCRC) identified etiology as a critical topic. A major goal of the BCRP is to reduce the incidence and morbidity of breast cancer among women in California. Understanding the etiology (causes) and pathogenesis (development) of breast cancer was deemed an essential first step in developing methods to prevent breast cancer.
It is well known that breast cancer affects some groups in the population at higher rates than others. A better understanding of who gets breast cancer and who doesn't might lead to clues about risks and protective factors for breast cancer occurrence. Under the broad topic of etiology, the BCRC envisioned a variety of possible studies, ranging from laboratory research looking at potential cancer-causing agents and genetic abnormalities in breast cancer cells or tissue, to large scale population studies that might examine dietary, hormonal, and environmental factors associated with breast cancer. The BCRC thought that these would be important first steps in enhancing our ability to recommend interventions for larger scale clinical testing. The BCRC deemed this a high priority area because of its potential to shed light on the causes of breast cancer and the development of preventive intervention strategies in the future.
The funded studies focus on examining various potential causes of breast cancer, including: an agent found in cooked meat, industrial pollutants, a virus found in cows (potentially passed through ingestion of beef and milk) and hormones. The examination of these possible causes of breast cancer offer the potential for a variety of prevention measures.
Research Program Awards
Linking a Dietary Carcinogen to Breast Cancer Susceptibility
James Felton, Ph.D.
Lawrence Livermore National Laboratory
Lifestyle has been shown to be involved in causing breast cancer, but the specific causes are not well understood. Correctly identifying the risk factors could play a large role in reducing breast cancer incidence and deaths. Many studies suggest carcinogenic (cancer-causing) chemicals in the diet play an important role in causing human breast cancer. A carcinogen (a chemical implicated in causing cancer) that damages DNA, called PhlP, causes mammary carcinomas in female rats, and is found in cooked meats associated with dietary fat. Like most carcinogens, it is metabolized (chemically processed in the liver and other tissues) to an active carcinogenic form in the body.
The susceptibility of women to breast carcinogens is caused by several complex factors, but seems to be related to at least 2 important things: 1) some individuals get a higher exposure from diet or lifestyle; and 2) individuals differ in their response to these exposures due to differences in their body's metabolism and hormonal responses. We propose to develop methods to measure the metabolism of PhlP in individual women. This will require analysis of urine samples for very small amounts of this carcinogen and its metabolites. To link human breast tumor causation to PhlP and its metabolites in urine, rats will be used to make the connection between the rats and humans. Experimental work will involve the use of modern analytical chemistry methods to isolate and purify PhlP and its metabolites. Determining the relative amounts of PhlP and the ratios of the major metabolites will make it possible to determine how each individual woman metabolizes PhlP and provide information on the dietary dose of PhlP.
Completion of the studies proposed here will help to 'close the loop" between the studies of breast cancer causation in the rat and studies of human breast cancer. It is clear that PhlP exposure is widespread, and PhlP has been found in home-cooked and restaurant foods at levels up to hundreds of parts-per-billion. Interestingly, there are no laws to regulate or limit the exposure to PhlP, since this mutagen and animal carcinogen is formed naturally upon cooking meats. For this reason, PhlP is one of the few carcinogens that can be studied in humans and offers a unique opportunity to test the extrapolation from animal breast cancer causation to studies of breast cancer risk in humans. Upon completion of these experiments, we should be able to better determine who is susceptible and who might benefit from dietary changes that can be made to lower their risk for breast cancer. This work will lead to meaningful information on a possible breast cancer-causing agent, and to better methods to link animal studies on cancer causation to humans.
Xenoestrogens and Breast Cancer in African American Women
Peggy Reynolds, Ph.D.
California Public Health Foundation
The purpose of our study is to identify possible environmental risk factors for breast cancer in African American women. Most of the known risk factors for breast cancer are reproductive factors that are not amenable to change. However, it has been estimated that 50-70% of women with breast cancer do not have any of the established risk factors. The incidence of breast cancer in African American women under 40 has risen dramatically and now exceeds the incidence in white women under 40. Breast cancer is more common in urban, industrialized areas suggesting that some aspect of the urban environment may play a role in the development of breast cancer.
Since the late 1940's a number of chemical compounds with hormonal properties have contaminated our environment and the food chain. Some of these compounds are organochlorine pesticides such as DDT, and industrial compounds known as polychlorinated biphenyls (PCBs) and dioxins. These compounds have disruptive effects on the reproductive, hormonal and immune systems of a variety of wildlife species and laboratory animals. Communities of color may be disproportionately exposed to industrial pollutants in inner-city environments. Many inner city residents have migrated from the rural south where they may have accumulated higher levels of DDT and related compounds.
A few recent studies have considered the role these chemicals may play in causing breast cancer through their hormonal effects on breast tissue. Some studies have had conflicting results and left many unanswered questions. Most were only able to test for total levels of DDT and PCBs. As these compounds are complex mixtures, it is likely that some of the individual substances within the mixtures may act differently. Only one study has looked at these substances in minority women. That study found much higher levels of DDT and PCBs in African American women compared to white women. It also revealed different relationships between the compounds and breast cancer in the different ethnic groups.
We have developed the capability of testing the individual components of DDT, other pesticides, PCBs and dioxins that are known to have hormonal or cancer causing effects in animals. Our proposed study will compare the levels of this expanded list of compounds_specifically including dioxins_in African American women with breast cancer to those undergoing biopsies for noncancerous conditions. Women will also be given a dietary questionnaire and interview to determine other risk factors for breast cancer. The proposed study builds on the strengths of a comparable study we are conducting in a population that is predominantly composed of white women.
We expect the results of these two studies to make a significant contribution to the current knowledge about possible environmental causes of breast cancer and may help explain the higher rates of breast cancer among younger African American women.
Innovative, Developmental and Exploratory (IDEA) Awards
Bovine Leukemia Virus and Human Breast Cancer Risk
Gertrude Buehring, Ph.D.
University of California, Berkeley
The purpose of the research proposed here is to determine whether: 1) humans can become infected with bovine leukemia virus; 2) such an infection occurs through consumption of foodstuffs from infected cows; 3) infection with BLV could lead to developing breast cancer. It is well-established that breast cancer in the mouse is caused by a virus, the mouse mammary tumor virus, which is passed from mother to nursing babies via the milk. Attempts to find an analogous virus in human milk passed from mother to baby have failed. Since humans drink more cow's milk than they do human milk, perhaps cows would be a more promising target in the search for a milk-transmitted agent of human breast cancer.
Bovine leukemia virus (BLV) is a cancer-causing virus of cattle which can be passed from mother cow to calf via the milk. BLV commonly infects dairy and beef cattle and is found in the marketed milk and meat of these animals. Most infected cattle do not actually develop leukemia, but remain healthy and are not removed from the herd. Consumption of unpasteurized dairy products or undercooked beef could possibly allow transmission of infectious virus to humans. BLV infections are not limited to cattle. The virus can infect sheep and nonhuman primates experimentally and cause cancer in the sheep. In the laboratory it can infect the cells of many species including humans and other primates. We recently discovered that BLV may infect the breast cells of cows and cause these cells to behave more like cancer cells. This indicates that the tissue preference of this "leukemia" virus is not limited to blood cells.
We propose here a pilot study to examine the first aspect of the overall proposal, whether humans can become infected with BLV. Human breast tissues removed during surgery, breast tissue sections received from a pathologist, cells from milk and colostrum, and blood cells will be searched for evidence of different components of BLV using cellular and molecular techniques (immunocytochemistry, PCR, and in situ reverse transcriptase PCR). Human blood will be tested for antibodies to BLV. Our preliminary results suggest that the breast cells of some women are infected with BLV and that some humans have antibodies to BLV. The potential impact of the project is through prevention of these cancers by elimination of BLV in cattle, interception of its transmission from cattle to humans, or vaccination of humans.
New Investigators Awards
Breast Cancer Risk Factors and Hormone Receptors
Ann Geiger, Ph.D.
Kaiser Foundation Research Institute
This epidemiologic study was designed to improve our understanding of how breast cancer develops in a way which may help us prevent breast cancer. The study specifically aims to determine whether established risk factors for breast cancer are related to the presence of estrogen and progesterone receptors in breast tumors.
Hormonal changes are a common explanation of how various factors increase risk for breast cancer. For example, beginning menstruation at a younger age or menopause at an older age are both risk factors for breast cancer. Both factors cause women to experience an increased number of menstrual cycles over their lifetime. The estrogen and progesterone present in these cycles stimulate the growth of breast tissue through estrogen and progesterone receptors present on breast cells. Greater amounts of estrogen and progesterone promote more growth than usual. This growth may increase the risk of breast cancer by increasing the chances of a genetic error (as cells reproduce themselves) or allowing genetic errors to be expressed more frequently.
The idea (hypothesis) to be tested in this study is that tumors with both estrogen and progesterone receptors will be more common in women with hormonally-related risk factors such as early age of first menstruation, late age at menopause, and late age at first birth. The accompanying hypothesis is that tumors without estrogen and progesterone receptors will be more common in women with risk factors which are not related to hormones. Such risk factors include having a mother or sister diagnosed with breast cancer. Finally, tumors with one receptor but not the other may occur in women with or without hormonally-related risk factors.
Female members of the Kaiser Permanente Southern California Medical Care Plan diagnosed with breast cancer from November 1996 to November 1998 will be included in this study. These patients will be asked to complete a short questionnaire about established breast cancer risk factors. Information about the presence of estrogen and progesterone receptors in these patients' tumors will be collected from pathologists' reports. Then the presence of breast cancer risk factors will be examined in four groups of patients with similar receptor status: both receptors positive; both receptors negative; estrogen positive with progesterone negative; and estrogen negative with progesterone positive.
This study was designed in response to two related BCRP priority issues: etiology and prevention. Improving our understanding of how breast cancer develops may suggest ways in which breast cancer can be prevented, thus having a dramatic impact on the health and well-being of the women in California and elsewhere.
Architecture of the ErbB2 Molecule Leading to Breast Cancer
Cara Marks, Ph.D.
University of California, San Francisco
Breast cancer occurs when breast cells escape normal cell growth controls and grow without restriction. Proteins known as receptors play a crucial role in this process. In healthy tissue, receptors transmit a signal to the cell to grow when a smaller protein, called a growth factor or ligand, binds to the receptor. Ligand binding induces a change in the shape of the receptor, which causes receptors next to each other to connect. This change in shape and resulting connection between two receptors on the outside of a cell acts like a domino falling over and starts a chain reaction throughout the cell that causes the cell to grow. Thus, all that is needed to signal a cell to grow is a connection between two receptors.
Such a connection between receptors can occur spontaneously (without the growth factor binding to a receptor) if there are a lot of receptors on the surface of the cell, such that they frequently bump into each other. Thus, cells that have abnormally high levels of receptors can grow without receiving signals from growth factors. Indeed, the most common abnormality in human cancers involves receptor overproduction. Such overproduction has been reported in breast, prostate, ovarian, bladder and lung cancers. One particular family of receptors called the epidermal growth factor receptors (EGF/ErbB1-4) has been implicated more than any other. In approximately 30% of breast cancer patients, the ErbB2 receptor is overproduced, resulting in aggressive and uncontrolled growth of tumor cells.
In this project, we propose to examine the architectural features of the contact between ErbB2 receptors which leads to the abnormal growth characteristic of cancer. This will not only increase our understanding of the mechanism by which ErbB2 overproduction results in unregulated breast cell growth, but may also provide insights into the mechanism of action of other growth factor receptors. The long term goal of this proposal is to develop small molecules to arrest breast cancer growth by blocking interactions between ErbB2 biomolecules. To accomplish this goal the architectural features of ErbB2 must be determined. The approach described uses molecular biology, molecular modeling and x-ray crystallography to investigate the architectural details of ErbB2. The combined use of these techniques will allow us to overcome the hurdles frequently encountered with structure determination.
Using this knowledge, drugs can be designed which specifically block this pathway and stop breast cancer tumor growth.
Postdoctoral Fellowship Awards
Breast Cancer and Genetic Differences in Estrogen Formation
Nicole Probst-Hensch, Ph.D.
University of Southern California
The goal of our research is to better understand the role of estrogens and of the enzyme aromatase in breast cancer. There is a good deal of evidence that women with high concentrations of estrogens in the blood are at an increased risk of developing breast cancer. The female body produces estrogens at different sites; before menopause, most of the estrogen production takes place in the ovary, while after menopause, the majority of estrogens are formed in adipose (fatty) tissue. A single enzyme (enzymes are proteins that act to increase the rate at which various chemical reactions within the body take place) aromatase, is responsible for the production of estrogen in postmenopausal women. It is currently not known if individuals are different in how efficiently the enzyme aromatase works to make estrogen. If there are such differences, we would expect persons whose aromatase works more efficiently to have higher estrogen levels, and to therefore have a higher breast cancer risk. It is possible that differences in aromatase activity could be due to different forms of the gene (called polymorphisms) that produce this enzyme.
It has not been systematically investigated whether such genetic differences frequently exist. So far, we only know about the existence of two polymorphisms, but it is currently unknown how they effect aromatase activity. We propose a pilot study to determine whether aromatase activity differs between individuals and whether such differences are related to breast cancer risk. If we actually find that aromatase activity differs between people, there would be many important consequences of such a result. First, we would need to identify the exact genetic changes that control aromatase activity. Within our own study population, we can investigate whether aromatase activity affects breast cancer risk more strongly in a certain subgroup of women, for example in women who take the pill, or in women who eat a high-fat diet. Second, it would be important to study whether differences in aromatase activity exists between racial groups, and whether these differences are in part responsible for the racial differences in breast cancer risk. Third, medications exist that stop aromatase function. They are currently used for treating existing breast cancer. If we find that a high aromatase activity increases the risk of breast cancer, the question arises whether these medications, called aromatase inhibitors, could be given to women at high risk before they actually develop breast cancer.
Genetic Susceptibility to Breast Cancer
Heather Spencer-Feigelson, Ph.D.
University of Southern California
The purpose of this grant is to investigate genes involved in the production of estrogen and its movement both in the blood throughout the body and in and out of cells, and to determine their role in breast cancer causation. A substantial amount of experimental, clinical and epidemiologic evidence indicates that hormones (most importantly, estrogen and progesterone) play a major role in the causation of breast cancer. The known risk factors for breast cancer can be understood as measures of the total lifetime exposure of the breast to estrogen and, perhaps, progesterone. Exposure to these hormones cause breast cells to grow and divide. Estrogen and other hormones are made in the body from cholesterol through a complex series of steps. These steps are determined by genes. We will test the idea (hypothesis) that certain genes involved in the making of estrogen, such as those that change estrogen from less active to more active forms, also play important roles in breast cancer. We also hypothesize that changes in the estrogen receptor (ER) gene, which works in concert with estrogen and other factors to signal breast cells to divide, are associated with breast cancer risk.
We will conduct a study to examine the possible associations between three genes and breast cancer in a total of 1,600 women from an ethnically diverse cohort (group of women) in Los Angeles County and Hawaii established in 1993. "Cases" will be those women with a first time occurrence of breast cancer. They will be compared to healthy women of similar age and ethnicity who have no history of cancer ("controls"). Approximately equal numbers of women will be included from the following four racial/ethnic groups: African-American, Hispanic, Japanese, and White.
We will explore possible associations between breast cancer and these genes within and across different ethnic groups, and also seek to identify environmental factors, e.g., dietary components, such as fat, or hormone use, that may interact with these genes and modify risk. We have
