Background: Although blood donations may reduce body iron stores, to date, prospective data on frequent blood donation and colorectal cancer risk are limited.
Methodology/principal findings: We tested whether frequent blood donation is associated with a lower risk of colorectal cancer in the Health Professionals Follow-up Study. We prospectively followed 35,121 men who provide the information on lifetime number of blood donations in 1992 through 2008. Serum ferritin levels were measured in a random sample of 305 men. Cox proportional hazard regression models were used to calculate the multivariable relative risks (RRs, 95%CIs) after adjusting for age and other established colorectal cancer risk factors. We documented 684 incident colorectal cancer cases and 224 deaths from colorectal cancer. The mean serum ferritin levels varied from 178 µg/L for men who did not donate blood to 98 µg/L for men who had at least 30 donations. Age-adjusted results for both incidence and mortality were essentially the same as the multivariable-adjusted results. Comparing with non-donors, the multivariable RRs (95%CIs) for colorectal cancer incidence were 0.92 (0.77, 1.11) for 1-5 donation, 0.85 (0.64, 1.11) for 6-9 donations, 0.96 (0.73, 1.26) for 10-19 donations, 0.91 (0.63, 1.32) for 20-29 donations, and 0.97 (0.68, 1.38) for at least 30 donations (P(trend) = 0.92). The multivariable RRs for colorectal cancer mortality were 0.99 (0.72, 1.36) for 1-5 donation, 0.93 (0.57, 1.51) for 6-9 donations, 0.85 (0.50, 1.42) for 10-19 donations, and 1.14 (0.72, 1.83) for at least 20 donations (P(trend) = 0.82). The results did not vary by cancer sub-sites, intake levels of total iron, heme iron, or family history of colorectal cancer.
Conclusions/significance: Frequent blood donations were not associated with colorectal cancer incidence and mortality in men. Our results do not support an important role of body iron stores in colorectal carcinogenesis.