Nonsteroidal anti-inflammatory drugs and breast cancer risk: The multiethnic cohort.

B212

Introduction: Previous studies on Nonsteroidal Anti-Inflammatory Drugs (NSAIDs) and breast cancer have produced mixed results. Although the relationship between cyclooxygenase-2 expression and breast cancer has been shown to be biologically plausible, the use of NSAIDs to reduce the risk of breast cancer is not well established. Methods: Incident breast cancer cases (N = 4010) from the Multiethnic Cohort (African-Americans, Caucasians, Japanese, Native Hawaiians, and Latinos from Hawaii and Los Angeles County) were identified from 1993 through 2002. Exposure data were obtained from a self-administered questionnaire administered at baseline. Multivariable Cox Proportional Hazards regression analyses provide estimates of hazard rate ratios (HRs) and 95% confidence intervals (95% CIs). Results: We observed no association between breast cancer risk and duration of aspirin use (HR = 1.00, 95% CI: 0.89 - 1.12 for six or more years of use), other NSAID use (HR = 0.85, 95% CI: 0.70 - 1.02 for six or more years of use), or total NSAID use (HR = 0.97, 95% CI: 0.86 - 1.09 for eleven or more years of use). No consistent associations between medication use and breast cancer risk were observed across strata of ethnicity, body mass index, tumor stage, or age. However, other NSAID use for 6 or more years was protective for African-American and Caucasian women (HR = 0.46, 95% CI: 0.27 - 0.79 and HR = 0.69, 95% CI: 0.48 - 1.00, respectively). Conclusions: Our study found aspirin, other NSAIDs and total NSAID use to be unrelated to breast cancer risk, except for African-American and Caucasian long-term users. More studies on newer NSAIDs need to be done to resolve their role on breast cancer risk.