Different markers of alcohol consumption, smoking and body mass index in relation to risk of pancreatic cancer. A prospective cohort study within the Malmö Preventive Project.

BACKGROUND/AIM: The association between alcohol consumption and pancreatic cancer is not clear. This study investigates different prediagnostic measurements of alcohol consumption, a laboratory marker (gamma-glutamyltransferase; gamma-GT), and a score measuring alcohol addiction (Mm-MAST), in relation to the risk of pancreatic cancer. Furthermore, the study investigated whether smoking and alcohol consumption interact with each other, or if the risk of pancreatic cancer associated with these factors is modified by obesity or weight gain.
METHODS: A cohort of 33,346 subjects provided prediagnostic information on the above factors. During a mean follow-up of 22.1 years, 183 cases of pancreatic cancer occurred. Cox's analysis yielded relative risks (RR) with 95% confidence intervals (CI).
RESULTS: The highest gamma-GT quartile was associated with a high risk of pancreatic cancer (RR = 2.15, 95% CI = 1.34-3.44), and this association was even stronger in subjects that reported a previous weight gain (RR = 3.61, 95% CI = 1.29-10.09). A high Mm-MAST score was also associated with pancreatic cancer (p = 0.02). Current smoking was associated with pancreatic cancer (RR = 2.34, 95% CI = 1.60-3.43), and obese smokers had an even higher risk (RR = 7.45, 95% CI = 1.65-33.64).
CONCLUSION: High alcohol intake is associated with subsequent risk of pancreatic cancer and this risk may be higher following weight gain. The risk associated with smoking may be even higher in obese subjects. Copyright 2009 S. Karger AG, Basel.