Chronic inflammation has been implicated in the etiology of gastric cancer. Prostaglandin E(2) (PGE(2)) is one of the major end-products of the cyclooxygenase-2 pathway, an enzyme that is an important mediator of inflammation. Using a novel method of quantifying the primary urinary metabolite of PGE(2) (PGE-M; 11alpha-hydroxy-9,15-dioxo-2,3,4,5-tetranorprostane-1,20-dioic acid), we evaluated urinary PGE-M concentrations in association with subsequent risk of development of gastric cancer in the Shanghai Women's Health Study, a large population-based prospective cohort, using a nested case-control study design. Controls were matched (1:1) to 153 gastric cancer cases by menopausal status; age, time, and date of sample collection; time interval since last meal; and availability of urine sample. Odds ratios (95% confidence intervals) were calculated using conditional logistic regression adjusted for potential confounders. Baseline urinary PGE-M levels were slightly higher among gastric cancer cases with a median of 6.4 ng/mg creatinine (interquartile range, 3.4-11.2) compared with 5.4 ng/mg creatinine among controls (interquartile range, 2.8-9.0), but this difference was not statistically significant (P = 0.34, Wilcoxon). With increasing quartiles of urinary PGE-M levels, the odds ratios (95% confidence intervals) for risk of gastric cancer increased in quartiles 2 to 4: 1.00 (0.48-2.08), 1.40 (0.67-2.91), and 1.98 (0.95-4.13), with a statistically significant test for trend (P = 0.04). The association persisted after additional adjustment for Helicobacter pylori status and was slightly strengthened among non-nonsteroidal anti-inflammatory drug users, subjects with positive H. pylori status, and for cases diagnosed within 46 months after study enrollment. Our findings suggest that higher levels of urinary PGE-M, a marker of inflammation, may be associated with gastric cancer risk.