By david Liu, PHD

Tuesday July 24, 2012 (foodconsumer.org) -- Eating too much foods high in fructose like high fructose corn syrup may increase risk of pancreatic cancer, according to a new study in Annals of Oncology.

The study led by D. Aune of School of Public Health, Imperial College in London and colleagues found eating each additional 25 grams of fructose per day was associated with a 22 percent increased risk of developing pancreatic cancer.

The study authors say it has been hypothesized that dietary carbohydrates, glycemic load and glycemic index affect the risk of pancreatic cancer, but epidemiological studies have been inconsistent. So they set out to review previous studies and conducted a meta-analysis of data from previous studies.

The meta-analysis conducted based on ten cohort studies indicated that an increase of 10 glycemic index units, 50 glycemic load units, 100 grams of total carbohydrates and 25 grams of sucrose per day were not associated with pancreatic cancer.

However, eating 25 grams of fructose per day was found associated with 22 percent increase in the risk.

Pancreatic cancer is the most deadly cancer.  Only 4.4 percent of pancreatic cancer patients are able to live over five years after diagnosis.

Another meta-analysis led by Jeanine M. Genkinger of Mailman School of Public Health in New York and colleagues also found sugar-sweetened carbonated soft drink intake was associated with increased risk of pancreatic cancer.

The analysis of data from 14 cohort studies involving 853,894 men and women with 2185 diagnosed with incident pancreatic cancer showed that those who had more than 8 oz. per day were almost 20 percent more likely than those who did not use sugar sweetened soft beverages to be diagnosed with the disease.

The most common ingredient in sugar sweetened soft beverages is high fructose corn syrup, which consists of almost half fructose and half glucose.  Prior research has found animals eating fructose gained more weight than those eating glucose.

The study was published in Cancer Epidemiol Biomarkers Prev; 21(2); 305–18