Long-Term Antibiotic Use As A Breast Cancer Risk Factor

Grant #: 1R03CA092765-01
RFA/PA: PAR98-023
PI Name: Taplin, Stephen H
Title: Long-Term Antibiotic Use As A Breast Cancer Risk Factor
Institute: Center For Health Studies
Funding Started: 2002

Oral antibiotic use disrupts normal intestinal microflora and can interfere with a number of metabolic processes potentially relevant to breast cancer development, such as steroid hormone metabolism and formation of several bioactive phytochemicals. The only reported epidemiologic study of antibiotic use and breast cancer risk was a Finnish cohort study that found a modest association in pre-menopausal women. The purpose of this 2-year study is to examine the association between long-term oral antibiotic use and primary, invasive breast cancer. The design is a case-control study among women aged 20 to 85, enrolled at least 5 years in a non-profit health maintenance organization (HMO), Group Health Cooperative of Puget Sound (GHC). Cases will be all female enrollees diagnosed with incident, invasive breast cancer between 1993 and 1999 (approximately 1,500 women). Cases will be identified through the Western Washington Cancer Surveillance System, which is part of the Surveillance, Epidemiology and End Results (SEER) Program of the National Cancer Institute. Controls will be approximately 4,500 women without breast cancer, frequency matched to cases on length of enrollment, calendar years of enrollment, and age. Oral antibiotic use will be identified for each woman from her period of enrollment (or beginning at 1977 if enrollment was prior to that time) and ending 2 years prior to breast cancer diagnosis for cases or comparable index period for controls. The GHC Pharmacy Database, a unique and comprehensive database of all prescriptions filled at GHC since 1977, will be used to identify antibiotic exposure. Women with long-term antibiotic use will be defined as those with any period of antibiotic use lasting 6 or more months continuously. Unconditional logistic regression will be used to estimate the odds ratios and 95% confidence intervals for long-term oral antibiotic use, relative to short-term use and non-use. Frequency matching variables and other potential confounders, such as oral contraceptive use, will be accessed using the GHC data systems and will be evaluated with logistic regression. The study will have sufficient power to detect odds ratios between 1.2 and 1.3, depending on the final sample size and proportion of women taking long-term antibiotics. Ultimately, this research has the potential to offer insight into the means by which breast cancer occurs. GHC is one of the few settings in the United States well suited to conduct this study, because it has a comprehensive listing of antibiotic prescriptions recorded prior to breast cancer diagnosis, and permits a population-based study with excellent case ascertainment.