A prospective study of plasma C-peptide and colorectal cancer risk in men.

Background: Colorectal cancer and type 2 diabetes share many risk factors, and hyperinsulinemia appears to be associated with an increased risk of colorectal cancer. We used the concentration of plasma C-peptide (an indicator of insulin production) to determine whether insulin and insulin resistance are associated with the risk of developing colorectal cancer.

Methods: We conducted a nested case-control study in the Physicians' Health Study. Plasma samples were collected from 14 916 cancer-free men from August 1982 through December 1984. Plasma C-peptide concentration, measured with an enzyme-linked immunosorbent assay, was available for 176 case patients who developed incident colorectal cancer through December 31, 1995, and 294 age- and smoking status-matched control subjects. Information on four other insulin resistance-related factors at baseline was obtained. We used conditional logistic regression models to investigate associations. All statistical tests were two-sided.

Results: Plasma C-peptide concentration was positively associated with age, body mass index (BMI), and number of insulin resistance-related factors, and it was inversely associated with fasting time before the blood draw, alcohol consumption, and vigorous exercise. An increased concentration of plasma C-peptide was statistically significantly associated with an increased risk of colorectal cancer (relative risk [RR] for the highest versus lowest quintile of plasma C-peptide = 2.7, 95% confidence interval [CI] = 1.2 to 6.2; P(trend) =.047), after adjusting for age, smoking status, fasting, BMI, alcohol consumption, vigorous exercise, and aspirin assignment in the Physicians' Health Study. The association became even stronger when the analysis was further adjusted for factors related to insulin resistance other than insulin levels (RR for the highest versus lowest quintile = 3.4, 95% CI = 1.4 to 8.3; P(trend) =.02) or when data from case patients who were diagnosed during the first 5 years of follow-up were excluded (RR for the highest versus the lowest quintile = 3.4, 95% CI = 1.3 to 8.8; P(trend) =.03). Adjusting for plasma levels of insulin-like growth factor I (IGF-I) and its binding protein 3 (IGFBP-3) did not materially change the results. There was no apparent modification of risk by BMI, insulin resistance-related factors, or vigorous exercise.

Conclusion: Elevated insulin production, as reflected by elevated concentrations of plasma C-peptide, may predict the risk of developing colorectal cancer, independently of BMI, factors related to insulin resistance, or levels of IGF-I and IGFBP-3.