There is a long-standing interest in finding whether allergic disorders cause or prevent cancer. We meta-analyzed associations of allergies with prostate, breast, and colorectal cancers. We searched MEDLINE, EMBASE, and Web of Science between 1966 and 2008 and eligible reference lists. A total of 16 observational studies were included. Meta-analyses revealed no evidence that asthma, hay fever, or 'any allergy' are associated with cancers of the breast (relative risk, RR = 0.93, 95% CI: 0.73-1.19 for asthma; 1.04, 0.94-1.16 for hay fever; 1.01, 0.94-1.08 for any allergy); prostate (RR = 0.93, 95% CI: 0.76-1.15 for asthma; 0.96, 0.87-1.05 for hay fever; 1.01, 0.87-1.17 for any allergy); or colorectum (RR = 0.95, 95% CI: 0.77-1.16 for asthma; 0.95, 0.86-1.05 for hay fever; 0.94, 0.85-1.04 for any allergy). There was a positive association of atopy (assessed by allergen-specific IgE or skin prick testing) with prostate cancer (RR = 1.43, 95% CI: 1.08-1.91), but not breast (1.18, 0.90-1.55) or colorectal (1.32, 0.69-2.53) cancers. There is little epidemiological support for the immune surveillance theory or antigenic stimulation theory in breast or colorectal carcinogenesis. The findings for prostate cancer warrant further investigation.