BACKGROUND: Epidemiological studies of selenium and vitamin E, two antioxidants hypothesized to reduce prostate cancer risk, have shown no discernible benefit. It has been proposed, however, that tobacco smoking may modify the effect of these nutrients. MATERIALS AND METHODS: We performed a meta-analysis of studies evaluating the relation of vitamin E and selenium exposure to prostate cancer risk in never smokers vs. ever smokers and, when feasible, former and current smokers. Overall and stratum-specific meta-risk ratios (meta-RRs) and 95% confidence intervals (CIs) were calculated using random-effects models. RESULTS: A total of 21 studies have met the inclusion criteria. Meta-RR (95% CI) estimates of prostate cancer associated with vitamin E use were 1.03 (0.95-1.11) in never smokers and 0.98 (0.90-1.07) in ever-smokers. For selenium, meta-RRs were 1.09 (0.78-1.52 and 0.76 (0.60-0.96) for never and ever-smokers, respectively; however, results for current smokers were weaker than those for former smokers. Sub-analyses according to different exposure assessment methods and outcome definitions produced similar results across strata. CONCLUSION: The association between vitamin E and prostate cancer is not modified by smoking. Selenium exposure is associated with lower prostate cancer risk among ever-smokers; however, the lack of an association for current smokers indicates that this finding needs to be interpreted with caution.