A prospective study of the effect of regular use of non-steroidal anti-inflammatory drugs (NSAIDs) on ovarian cancer risk

B118

Background: Inflammation is hypothesized to be an important contributing mechanism in ovarian carcinogenesis. Ovulation resembles a localized inflammatory response and factors that suppress ovulation (e.g. pregnancy, oral contraceptive use, breastfeeding) or those that prevent inflammation-inducing substances from reaching the upper reproductive tract may decrease ovarian cancer risk (e.g. tubal ligation, hysterectomy) decrease ovarian cancer risk. Given the limited number of modifiable factors in ovarian cancer prevention, identifying whether regular use of commonly used non-steroidal anti-inflammatory agents (NSAIDs) protects against ovarian cancer through reducing of inflammation could have important public health impact. Few prospective studies have considered this association and their findings are inconclusive. We thus sought to examine the effects of regular use of NSAIDs on ovarian cancer risk in two large prospective cohorts of women from whom we collected and updated detailed information on NSAIDs use.
 >Methods: We prospectively assessed the effect of regular use of aspirin and non-aspirin NSAIDs on ovarian cancer risk among 197,835 participants of the Nurses Health Study and Nurses Health Study-II over 24 and 15 years of follow up, respectively. Information on aspirin and non-aspirin NSAIDs use was collected and updated biennially. Information on dose, frequency, and duration of use was obtained in most questionnaires and updated throughout follow-up. We used proportional hazards models to compare incidence rates of ovarian cancer between those who used NSAIDs at least twice a week against participants who did not, taking into account multiple potential confounders.
 >Results: A total of 727 confirmed cases of epithelial ovarian cancer were identified in both cohorts over 3,187,890 person-years. The hazard ratios (HR) and 95% confidence limits (CI) associated with regular use of aspirin and non-aspirin NSAIDs were 1.09 (0.91, 1.32) and 0.89 (0.65,1.23), respectively. There was no evidence of an effect of dose, frequency, or total duration of use of these agents on ovarian cancer risk as the corresponding hazards ratios were comparable to 1. Results were similar across different histologic types of ovarian tumors, though power was limited in these secondary analyses. There was no evidence of modification of the effect of NSAIDs use by body mass index, menopause status, oral contraceptive use, or regularity of menstrual cycle.
 >Conclusion: In this large, prospective, comprehensive study, we found no compelling evidence to support an association between regular use of aspirin or non-aspirin non-steroidal anti-inflammatory agents and ovarian cancer risk.