Abstract
BACKGROUND:
Type
2 diabetes mellitus (T2DM) is associated with an increased risk of
colorectal cancer (CRC); however, studies differentiating between
subsites of CRC are limited. We investigated how diabetes mellitus (DM)
was associated with subsite-specific CRC risk in men and women.
METHODS:
The
Netherlands Cohort Study on diet and cancer is a prospective study
among 120 852 men and women aged 55-69 years old at baseline in 1986.
Information on DM, anthropometric, dietary and lifestyle factors was
self-reported at baseline. T2DM was defined as the diagnosis of DM after
30 years of age. Incident CRC cases were identified by record linkage
with the Netherlands cancer registry and the Dutch pathology registry.
After 17.3 years of follow-up, 1735 incident male CRC cases and 1321
female CRC cases were available for analyses. Subsite-specific hazard
ratios (HRs) for CRC were estimated in case-cohort analyses using Cox
regression.
RESULTS:
At
baseline, 3.1% of subcohort members reported T2DM, of whom 80% were
diagnosed after 50 years of age. Multivariable-adjusted models showed
that the risk of proximal colon cancer was significantly increased in
women with T2DM versus women without T2DM (HR=1.80, 95% confidence
interval: 1.10-2.94). There was no association between T2DM and the risk
of overall CRC, distal colon cancer and rectal cancer in women. In men,
T2DM was not associated with overall CRC (HR=0.98, 95% confidence
interval: 0.64-1.50), or with risk at any subsite.
CONCLUSIONS:
This prospective study showed an increased risk of proximal colon cancer in women with T2DM compared with non-T2DM women.
