Alcohol intake and pancreatic cancer risk: a pooled analysis of fourteen cohort studies.

 

 

 

 

 

 

Reference: 

 

Alcohol Linked to Pancreatic Cancer Risk

A daily drinking habit may modestly increase pancreatic cancer risk, researchers found in a meta-analysis.

Pancreatic cancer risk was 22% higher (95% CI 3% to 45%) for individuals who had two or more drinks a day than for nondrinkers, Stephanie Smith-Warner, Ph.D., of the Harvard School of Public Health, and colleagues reported in the March issue of Cancer Epidemiology Biomarkers and Prevention.

Although risk rose gradually with intake (P=0.05 for trend), fewer than two drinks a day did not affect pancreatic cancer rates significantly in multivariate analysis of primary data from more than 860,000 men and women in 14 prospective studies.
Action Points  

    * Explain to interested patients that the meta-analysis had greater statistical power to find a small association than any individual component study.


    * Caution that the modest pancreatic cancer risk seen in the meta-analysis needs to be taken in the context of both increased and decreased risks of other cancers and diseases linked to alcohol consumption in prior studies.

Two drinks a day may be considered more than moderate intake, noted co-author Jeanine M. Genkinger, Ph.D., of Georgetown University Medical Center in Washington. Guidelines from the American Heart Association and other organizations recommend no more than two drinks a day for men and no more than one a day for women, she said.

"The bottom line is drinking too much is not safe, but what's the safe amount depends on one's risk profile," commented Elizabeth A. Platz, Sc.D., M.P.H., of Johns Hopkins, who was not involved in the study.

She noted that these findings add to a mixed bag of often conflicting results from studies on alcohol's health effects, most notably the reduced risk of cardiovascular disease in contrast to increased risk of a range of cancers.

In the past week alone, a large British survey found that risk of a wide range of cancers rose among women with every 10-g alcoholic drink consumed per week, whereas three other studies found that a glass or two of wine each day reduced risk of esophagitis, Barrett's esophagus, and esophageal adenocarcinoma. (See: Wine Cuts Risk of Esophageal Lesions and Even a Few Drinks a Day Increases Cancer Risk in Women)

Although most of the individual prospective pancreatic cancer studies to look at alcohol intake have had risk ratios above one for the highest intake group versus the lowest, none showed a significant effect on risk.

Because there are so few known risk factors for this disease, however, it's important to understand any potential causes, even if the risks are weak or modest, Dr. Genkinger said.

So to determine whether the association was real and prior studies have simply had too few incident cancers and too small a sample of heavier drinkers to determine an effect, Dr. Smith-Warner's group pooled the findings.

They found that the slight positive association with pancreatic cancer risk for the highest alcohol intake group that drank the equivalent of two or more drinks a day was significant overall (age-adjusted RR 1.36, 95% CI 1.15 to 1.60).

It remained significant in comparison with nondrinkers after full multivariate adjustment for the few consistently-reported risk factors for pancreatic cancer, including age, smoking, diabetes, and height and weight (RR 1.22, 95% CI 1.03 to 1.45).

The association was significant for women (multivariate relative risk 1.41 for 30 versus 0 g per day, 95% CI 1.07 to 1.85) but not men (multivariate RR 1.12, 95% CI 0.89 to 1.39).

The association was also somewhat stronger for adenocarcinomas of the pancreas than for other pancreatic cancer subtypes (multivariate RR 1.45, 95% CI 1.17 to 1.79).

Notably, no single type of alcoholic beverage appeared to carry the brunt of the risk. The associations with pancreatic cancer for the highest versus lowest intake category (five or more drinks a day versus none) were:

    * For wine (multivariate RR 1.01, 95% CI 0.87 to 1.16)
    * For beer (multivariate RR 1.13, 95% CI 0.90 to 1.04)
    * For spirits (multivariate RR 1.05, 95% CI 0.93 to 1.19)

Nor was the link with alcohol modified by smoking status, total folate intake, methionine intake, or multivitamin use.

But body mass index was a factor as alcohol appeared to affect pancreatic cancer risk only among individuals in the normal weight range (P=0.01 for interaction).

The researchers cautioned that their meta-analysis used only recent alcohol intake self-reported on baseline food-frequency questionnaires, which precluded evaluation of the effect of changes in intake over time, lifetime consumption, past drinking habits, and binge drinking patterns.

 

（Source: Med Page Today: 2009 March)