Fish consumption and prostate cancer risk: a review and meta-analysis.

BACKGROUND Prostate cancer incidence varies 60-fold globally, which suggests the roles of lifestyle and dietary factors in its cause. To our knowledge, a comprehensive assessment of the association between fish consumption and prostate cancer incidence and mortality has not been reported. OBJECTIVE We conducted a meta-analysis of fish intake and prostate cancer by focusing on the incidence of prostate cancer and prostate cancer-specific mortality and included subgroup analyses based on race, fish type, method of fish preparation, and high-grade and high-stage cancer. DESIGN We searched MEDLINE and EMBASE databases (May 2009) for case-control and cohort studies that assessed fish intake and prostate cancer risk. Two authors independently assessed eligibility and extracted data. RESULTS There was no association between fish consumption and a significant reduction in prostate cancer incidence [12 case-control studies (n = 5777 cases and 9805 control subjects), odds ratio (OR): 0.85; 95% CI: 0.72, 1.00; and 12 cohort studies (n = 445,820), relative risk (RR): 1.01; 95% CI: 0.90, 1.14]. It was not possible to perform a meta-analysis for high-grade disease (one case-control study, OR: 1.44; 95% CI: 0.58, 3.03), locally advanced disease (one cohort study, RR: 0.80; 95% CI: 0.61, 1.13), or metastatic disease (one cohort study, RR: 0.56; 95% CI: 0.37, 0.86). There was an association between fish consumption and a significant 63% reduction in prostate cancer-specific mortality [4 cohort studies (n = 49,661), RR: 0.37; 95% CI: 0.18, 0.74]. CONCLUSION Our analyses provide no strong evidence of a protective association of fish consumption with prostate cancer incidence but showed a significant 63% reduction in prostate cancer-specific mortality.