Abstract The results of previous studies of recreational physical activity and risk of epithelial ovarian cancer are mixed, with some studies supporting an inverse association and others, including several prospective studies, observing no association or a suggestive positive association with vigorous activity. Physical activity may decrease ovarian cancer risk by reducing circulating levels of estrogens, decreasing body fat and estrogen synthesis, or reducing the frequency of ovulation. However, plausible mechanisms also exist for a positive association, including possible ovulatory changes with regular vigorous activity, increased pituitary production of gonadotropins, or changes in progesterone or androgen concentrations. In an earlier analysis of the Nurses’ Health Study (NHS) with follow-up through 1996, there was no clear association between physical activity level and ovarian cancer incidence, but there was a suggestion of an increase in risk with frequent vigorous activity. We reanalyzed this association using 10 additional years of follow-up in the NHS and adding data from the younger Nurses’ Health Study II (NHSII) cohort. We calculated current and cumulative average physical activity in hours per week and metabolic equivalent task-hours (MET-hours) per week using data collected every 2-6 years beginning in 1986 (NHS) or 1989 (NHSII). We used Cox proportional hazards regression to model the age- and multivariable-adjusted incidence rate ratios of ovarian cancer for each category of physical activity, compared to the lowest category. Our analysis included 179,443 women with detailed physical activity data and 652 incident, confirmed cases diagnosed through 2006 (NHS) or 2007 (NHSII). There was no clear association between hours per week of current or cumulative average activity and ovarian cancer risk. However, there was evidence of a positive association between cumulative average MET-hours of total activity per week and risk of all epithelial cancers and the serous invasive subtype. The multivariable-adjusted incidence rate ratios and 95% confidence intervals for each MET-hour category, compared to <2.5 MET-hours, were 1.27 (0.90, 1.80) for 2.5-<5 MET-hours, 1.06 (0.77, 1.48) for 5-<10 MET-hours, 1.17 (0.86, 1.60) for 10-<20 MET-hours, 1.56 (1.13, 2.17) for 20-<30 MET-hours, and 1.34 (0.96, 1.87) for ≥30 MET-hours. The corresponding estimates for serous invasive cancers were 1.42 (0.86, 2.35), 1.14 (0.71, 1.84), 1.49 (0.95, 2.32), 1.82 (1.14, 2.92), and 1.48 (0.92, 2.40). Our results do not support an inverse association between physical activity and ovarian cancer risk, and suggest a possible increase in risk with high levels of activity, particularly for the serous invasive subtype. Lagged analyses and analyses of specific types and intensities of activity are ongoing. Citation Format: {Authors}. {Abstract title} abstract. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 1830.