•
•
•
•
In a multidisciplinary effort, Dr. Regina Ziegler, Division of Cancer Epidemiology and Genetics, National Cancer Institute; Dr. Larry Keefer, Center for Cancer Research, National Cancer Institute; and Drs. Xia Xu and Tim Veenstra, Mass Spectroscopy Facility, NCI-Frederick, have developed a sensitive, accurate, precise liquid chromatography/mass spectrometry (LC/MS) method to measure simultaneously the ~15 estrogens and estrogen metabolites in human urine. The procedure is simple and rapid and can easily measure in < 0.5 ml of urine the low levels of estrogen in postmenopausal women. In the past, it has not been possible to measure accurately absolute levels of estrogen exposure and patterns of estrogen metabolism, especially in postmenopausal women. At present, we are using the method to explore the role of estrogen metabolism in the etiology of breast cancer; but the technique will benefit many areas of research. The technique should be applicable not only in etiologic research on hormone-related cancers but also in identifying high-risk populations and monitoring treatment. It should be relevant to studies of cancer, heart disease, osteoporosis, endocrine disorders, infertility, and other diseases/conditions influenced by levels of endogenous steroid hormones.
Our research includes both laboratory and epidemiologic investigations. A postdoctoral fellow could focus on either approach, or become involved in both types of studies. In the laboratory, we are currently extending our high performance liquid chromatography-electospray ionization-tandem mass spectrometry technique to plasma, serum, and tissue. We have already discovered that we can detect in blood the same broad range of estrogen metabolites we are measuring in urine. We are also modifying the single chemical derivatization we use so that we can measure androgens and androgen metabolites concurrently with the estrogens. Since so little in known about estrogen metabolism in women, we are evaluating interindividual and intraindividual variability in absolute estrogen levels and estrogen metabolism, as well as the influence of age, menopausal status, and stage of menstrual cycle. Preliminary results indicate the interindividual variability is very high, with urinary estrone, estradiol, and estriol, adjusted for creatinine, varying 10-fold in postmenopausal women, and several other estrogen metabolites varying 100-fold. We are also optimizing our techniques for collecting and stabilizing urine and blood samples in epidemiologic studies.
Possibly most exciting are our two epidemiologic studies of breast cancer. While it is generally acknowledged that estrogen increases the risk of breast cancer in postmenopausal women, almost nothing is known about the importance of the different estrogens and patterns of estrogen metabolism. We are applying our LC/MS technique to urine samples collected in studies of breast cancer in Asian-American women and U.S. women. In the population-based Asian-American breast cancer study, a six-fold gradient in breast cancer risk by migration patterns was observed, comparable to the international differences in breast cancer incidence rates. The U.S. study utilizes a large cohort of nurses, with extensive information on accepted and putative breast cancer risk factors and biomarkers. Once our technique is modified for blood samples, we will apply it in a study of incident breast cancer nested in the Prostate, Lung, Colon, and Ovarian Cancer Screening Trial cohort, which is part of the Cohort Consortium study of breast cancer, endogenous hormones, and related genetic variability.
