Louise A. Brinton, Ph.D., M.P.H.
|Organization:||National Cancer InstituteDivision of Cancer Epidemiology & Genetics, Hormonal and Reproductive Epidemiology Branch|
|Address:||NCI Shady GroveRoom 7E102|
After receiving an M.P.H. in epidemiology from the University of North Carolina at Chapel Hill, Dr. Brinton joined the NCI as a Staff Fellow in 1976. She earned a Ph.D. in epidemiology from The Johns Hopkins School of Hygiene and Public Health in 1979, and subsequently conducted postdoctoral research at Oxford University. Dr. Brinton was appointed Acting Chief of the Environmental Studies Section in 1984, and in 1985 Chief of the Environmental Epidemiology Branch (now called the Hormonal and Reproductive Epidemiology Branch). She served on the Executive Board of the Society for Epidemiologic Research, and was elected President of the organization in 1990. Dr. Brinton has received the PHS Special Recognition Award and the NIH Director's Award for innovative leadership in women's health research. She has also been honored by receipt of the H.A. Tyroler Distinguished Alumni Award from the University of North Carolina and the American College of Epidemiology’s Abraham Lilienfeld Award
Despite extensive epidemiologic study, the causes of a substantial proportion of cancers of the female breast and reproductive system are unknown. Although a number of risk factors have been identified, our understanding of the mechanisms underlying most of these risk factors is incomplete, which may be preventing a fuller understanding of causality. A major emphasis of our current research efforts is to clarify the etiology of these cancers through the assessment of various biomarkers, including genetic markers and endogenous hormones. We have also invested considerable effort in approaching the study of these cancers not as a single disease entity but as a complex of diseases defined by unique etiologic components.
We have recently completed two large case-control studies of breast cancer, one in Poland and the other in three U.S. centers (this latter investigation involved extensive collaboration with a number of extramural investigators). Primary emphasis has been placed on utilizing the biologic samples collected within these studies to assess genetic influences on breast cancer risk. The results from these studies have been important both in terms of their individual findings as well as for contributions to consortial efforts. Samples have also been contributed to genome wide association studies. Given the extensive amount of environmental risk factor data collected within each study, it is anticipated that future efforts will be focused on clarifying interactive influences of genetic and environmental factors. In terms of environmental factors, the Polish study was uniquely positioned to assess effects of occupational exposures given that a large proportion of Polish women worked outside their homes (often in industrialized settings). In this study, we also attempted to carefully assess a wide range of physical activities utilizing both questionnaires as well as accelerometer measurements. The Polish study also involved an extensive pathology component, allowing the construction of tissue microarrays which have enabled disease classifications according to various protein expression markers.
Bone density is recognized as a predictor of breast cancer risk, but the mechanism underlying this relationship is unknown. We are collaborating on a follow-up study of women previously screened for bone density to evaluate their risk of breast and other hormonally-related cancers. Sera collected earlier from these women, together with DNA that will be obtained from buccal cell swabs, offer the opportunity for assessing interrelationships among bone density, endogenous hormones, and genetic polymorphisms.
Although the role of hormones in the etiology of breast and gynecologic cancers is well recognized, it has been difficult to relate endogenous hormones to risk in many previous studies. This may reflect difficulties in measuring such hormones, but recent advances, including development of a liquid chromatography/mass spectography method at a Frederick laboratory, has promise for further clarifying relationships. Using serologic samples donated by over 7,000 women residing in Columbia, MO during the 1970’s and early 1980’s, we plan to measure 15 estrogens and estrogen metabolites for assessment in relation to subsequent breast cancer incidence. Questionnaires administered to these women will allow us to assess the influence of these biomarkers on risk before and after adjustment for other risk predictors and to evaluate how these factors influence levels of these markers.
Male breast cancer, with an incidence <1% that of female breast cancer, has received little prior epidemiologic attention. We used data from the large NIH-AARP cohort study to identify risk factors for this poorly studied tumor. Like female breast cancer, we found that a family history of breast cancer, obesity and physical inactivity appeared to increase risk. Somewhat surprisingly, men who reported a history of bone fractures later in life were at an increased risk of male breast cancer, a relationship opposite to that found in women. Since bone fractures in men relate both to low estrogen and androgen levels, we hypothesized that the variant relationship might relate to alterations in bioavailable estrogens to androgens. We also assessed risk factors for male breast cancer in the large Veteran’s Affairs Medical Care System, where we found that major risk predictors included Klinefelter syndrome, gynecomastia, obesity and orchitis/epididymitis.
Based on these findings, we are interested in further studying male breast cancers, but given the rarity have proposed doing this through consortial efforts. We have approached investigators of the largest available cohort studies to determine their receptivity to contributing data for this pooling effort. Eight large cohorts have agreed to participate in this effort, with NCI contributions including the NIH-AARP and PLCO follow-up studies. We are currently assembling questionnaire datasets and hope to use derived analyses to guide future plans for biologic samples, including those informative to genetic and hormonal relationships.
The Polish breast cancer study also included case groups of both endometrial and ovarian cancer, allowing us to capitalize on comparisons with the large control group assembled for the breast cancer cases. A similar questionnaire was utilized as in the breast cancer study, and various biologic samples (including blood and tissue samples) were collected. Many opportunities exist.
Several years ago, the Gynecologic Oncology Group (GOG) requested our assistance in developing a questionnaire that could be incorporated into the context of some of their ongoing trials. Although GOG trials had collected extensive clinical data, little attention had been focused on other patients characteristics. We therefore developed a detailed risk factor questionnaire, which has now been incorporated into a large endometrial cancer trial. To date, questionnaires have been collected on over 4,000 patients, which will enable a wide variety of hypotheses to be explored, including many that can capitalize on the merging of epidemiologic and clinical data. Plans are currently underway to incorporate our questionnaire into several trials of other gynecologic cancers.
Clues about hormonal mechanisms of carcinogenesis may be derived from studies of cancer risk in populations with known hormonal alterations. A collaborative record-linkage study in Denmark and Sweden found that patients with endometriosis had significant excesses of non-Hodgkin lymphoma and cancers of the ovary and breast. We have also used data from this study to gain insights as to other medical conditions that are related to ovarian and endometrial cancers. The large size of this study allowed us to examine histology specific relationships between endometriosis and ovarian cancer, showing that the association is specific for clear cell and serous cancers.
To address the long-term effects of silicone breast implants, we conducted a retrospective cohort study among women with augmentation mammoplasties, using a comparison group of women with other types of plastic surgery. Despite clinical evidence that implants interfere with the visualization of breast lesions, we found no evidence that implants were associated with altered breast cancer risk. Patients with breast implants, however, presented with somewhat later stages of breast cancer at diagnosis than other patients. Data from the study were also analyzed to evaluate the incidence of other cancers and connective tissue disorders, as well as cause-specific mortality, among implant patients.
Although reports indicate that ovulation-stimulating drugs may predispose to ovarian cancer, the studies are limited by small numbers and imprecise information on causes of infertility and medication use. To clarify this relationship, a large retrospective cohort study abstracted detailed medical data on women who were evaluated and treated for infertility as long ago as the 1960s. Questionnaires were administered to collect additional data on subsequent health events and other risk factors, and patient identifiers were linked to cancer registries and the National Death Index to ascertain information on development of cancers. An initial follow-up of these patients was completed in 2002, and an updated follow-up is soon to be initiated.