The potential role of coffee consumption in the development of various types of cancer has been extensively investigated in epidemiologic studies. How coffee consumption may modulate risk of gastric cancer, however, remains a subject open for investigation. To better quantify this relation, we quantitatively summarized evidence from prospective studies. Eligible studies were identified on PubMed and Embase databases. The summary risk estimates were obtained using the random-effects model. Subgroup, sensitivity and dose-response analyses were conducted. The present meta-analysis included 12 prospective cohort studies. A pooled analysis of these studies suggested that coffee consumption (highest vs. lowest consumption) was not associated with risk of gastric cancer (RR = 1.12, 95% CI = 0.93-1.36). In the subgroup analysis, significant increased risk was detected in the U.S. studies (RR = 1.36, 95% CI = 1.06-1.74) and in the studies with <10 years of follow-up (RR = 1.24, 95% CI = 1.00-1.54), and the greatest increase in risk was observed in those studies without adjustment for smoking (RR = 1.48, 95% CI = 1.13-1.93). There was some evidence of publication bias (P for Egger's test = 0.03). Cumulative evidence from prospective studies suggests that coffee consumption is not associated with risk of gastric cancer. The observed positive results may be confounded by smoking and need further investigation.