Meta-analysis: Alcohol Intake Linked to Prostate Cancer Risk

Nick Mulcahy

November 18, 2016

There is new evidence of an association between alcohol consumption and the risk for prostate cancer, according to the authors of the latest meta-analysis on the subject.

A Canadian and Australian team reports that they found, for the first time among meta-analyses, a significant dose-response relationship between level of alcohol intake and risk for prostate cancer, starting with low-volume consumption.

When compared to lifetime abstainers, low-volume drinkers (up to two drinks per day) had an 8% greater risk for prostate cancer (relative risk [RR] = 1.08, P < .001); medium-volume drinkers (up to four drinks per day) had a 7% greater risk (RR = 1.07, P < .01); high-volume drinkers (up to 6 drinks per day), a 14% greater risk (RR = 1.14, P < .001); and higher-volume drinkers (6 drinks or more per day), an 18% greater risk (RR = 1.18, P < .001).

The total number of grams of alcohol for each volume category was wide. Two drinks per day ranged from 1.30 to 24 g; up to 4 drinks, 25 to 44 g; up to 6 drinks, 25 to 64 g; and more than six drinks, > 65 g.

The risk estimates were adjusted for a number of variables, including smoking.

Overall, there was a significant dose-response relationship for current drinkers (P trend < .01), report Tim Stockwell, PhD, director of the Center for Addictions Research of British Columbia at the University of Victoria, and his colleagues.

The new analysis, which included 27 case-control and cohort studies, was published online November 15 in BMC Cancer.

So, should prostate cancer be added to the list of cancers for which drinking alcohol is a risk factor?

Possibly, say the authors.

"Prostate cancer may need to be incorporated in future estimates of the burden of disease alongside other cancers (eg, breast, esophagus, colon, liver)," they write, referring to some of the malignancies for which alcohol is a well-established risk factor.

Dr Stockwell told Medscape Medical News that evidence of an association between alcohol intake and prostate cancer "has been identified by many studies and is considered borderline for the relationship being labelled as definitively causal."

He added: "Research on this topic is growing."

The new study helps make the case for alcohol as a risk factor, said Jürgen Rehm, PhD, professor and chair, addiction policy, Dalla Lana School of Public Health, University of Toronto, Canada.

Dr Rehm, who was not involved with study, told Medscape Medical News that "any final decision will have to be made by the International Agency for Research on Cancer [IARC], which is the most respected agency in this area."

The IARC first declared that alcohol is a carcinogen in 1988.

Dr Rehm said the IARC's decisions about individual cancers are based on epidemiology and basic research, "with special emphasis on biological pathways."

Currently, there is no biological explanation or evidence as to how alcohol could cause prostate cancer, the study authors acknowledge.

But they list some hypothetical mechanisms, including a genotoxic effect of acetaldehyde, increased estrogen concentration, and changes in folate metabolism. Dr Stockwell said that alcohol will not be accepted as a risk factor for prostate cancer until there is evidence of a "definitive causal mechanism."

Findings of previous meta-analyses and literature reviews have been mixed concerning the association of prostate cancer with alcohol consumption, say the authors.

But these efforts have all been flawed, they contend. "None have adequately considered the effects of confounding and bias, including potential biases caused by misclassification of former and occasional drinkers in the abstainer reference groups," they write.

The authors explain that the practice of lumping together former drinkers and occasional drinkers with true abstainers is important because former drinkers, some of whom quit drinking alcohol because of serious health problems, may especially inflate the prostate cancer risk among abstainers.

Dr Rehm confirmed that this was a problem: "The misclassification is indeed a methodological problem which is plaguing alcohol research."

Of 27 studies included in the new meta-analysis, 16 had former drinker bias, and 15 had occasional drinker bias. Only six were free from both types of bias.

Among these latter six studies that had no misclassification errors and thus no related bias, the risk estimates were much higher. For example, among these six studies, low-volume drinkers had a 23% greater risk for prostate cancer compared to abstainers (RR = 1.23, P = 0.0143). By way of comparison, in the analysis of the 27 studies, 8% had a greater risk for prostate cancer. However, the authors acknowledge that the total number of men in this smaller sample weakened the authority of these findings.

In their meta-analysis, Dr Stockwell and colleagues only included studies that reported a man's total alcohol consumption. They excluded studies that were based on drinking only specific beverages, such as wine, whiskey, vodka, sake, or other hard liquors.

All of the studies assessed both quantity and frequency of total alcohol consumption for at least 1 week.

The study was funded by the US National Institutes of Health. Dr Stockwell has received funds from Systembolaget, the Swedish retail alcohol monopoly. Dr Rehm has disclosed no relevant financial relationships.

BMC Cancer. Published online November 15, 2016. Abstract

Follow Medscape senior journalist Nick Mulcahy on Twitter: @MulcahyNick

For more from Medscape Oncology, follow us on Twitter: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.