Background: Several studies of male colon cancer have found positive associations with body size and composition. It is uncertain whether this relationship is due to non-adipose mass, adipose mass, distribution of adipose mass such as central adiposity, or all three.
Methods: In a prospective cohort study of men aged 27-75 at recruitment in 1990-1994, body measurements were taken by interviewers. Fat mass and fat-free mass (FFM) were estimated from bioelectrical impedance analysis. Waist circumference and waist-to-hips ratio (WHR) estimated central adiposity. Incident colon cancers were ascertained via the population cancer registry. Altogether, 16,556 men contributed 145,433 person-years and 153 colon cancers.
Results: Rate ratios (RRs) comparing men in the fourth quartile with those in the first quartile were as follows: FFM 2.3 [95% confidence interval (CI) 1.4-3.7]; height 1.9 (95% CI 1.1-3.1); waist circumference 2.1 (95% CI 1.3-3.5); WHR 2.1 (95% CI 1.3-3.4); fat mass 1.8 (95% CI 1.1-3.0); and body mass index 1.7 (95% CI 1.1-2.8). When continuous measures of FFM and WHR were modeled together, the RR for FFM per 10 kg was 1.37 (95% CI 1.04-1.80) and the RR for WHR per 0.1 unit was 1.65 (95% CI 1.28-2.13). After adjustment for FFM and WHR, the RRs for fat mass and body mass index were no longer statistically significant.
Conclusion: Male colon cancer appears to be related to body size and composition by two different pathways, via central adiposity and via non-adipose mass.