Critique 039: The role that alcohol drinking may play in the risk of cancer – 17 April 2011

Schűtze M, Boeing H, Pischon T, et al, Alcohol attributable burden of incidence of cancer in eight European countries based on results from prospective cohort study.  BMJ 2011; 342:d1584 doi: 10.1136/bmj.d1584 (Published 7 April 2011)

Authors’ Abstract

Objective To compute the burden of cancer attributable to current and former alcohol consumption in eight European countries based on direct relative risk estimates from a cohort study.

Design Combination of prospective cohort study with representative population based data on alcohol exposure.

Setting Eight countries (France, Italy, Spain, United Kingdom, the Netherlands, Greece, Germany, Denmark) participating in the European Prospective Investigation into Cancer and Nutrition (EPIC) study.

Participants 109,118 men and 254,870 women, mainly aged 37-70.

Main outcome measures Hazard rate ratios expressing the relative risk of cancer incidence for former and current alcohol consumption among EPIC participants. Hazard rate ratios combined with representative information on alcohol consumption to calculate alcohol-attributable fractions of causally related cancers by country and sex. Partial alcohol attributable fractions for consumption higher than the recommended upper limit (two drinks a day for men with about 24 g alcohol, one for women with about 12 g alcohol) and the estimated total annual number of cases of alcohol attributable cancer.

Results If we assume causality, among men and women, 10% (95% confidence interval 7 to 13%) and 3% (1 to 5%) of the incidence of total cancer was attributable to former and current alcohol consumption in the selected European countries.  For selected cancers the figures were 44% (31 to 56%) and 25% (5 to 46%) for upper aerodigestive tract, 33% (11 to 54%) and 18% (−3 to 38%) for liver, 17% (10 to 25%) and 4% (−1 to 10%) for colorectal cancer for men and women, respectively, and 5.0% (2 to 8%) for female breast cancer.  A substantial part of the alcohol-attributable fraction in 2008 was associated with alcohol consumption higher than the recommended upper limit: 33,037 of 178,578 alcohol-related cancer cases in men and 17,470 of 397,043 alcohol related cases in women.

Conclusions In western Europe, an important proportion of cases of cancer can be attributable to alcohol consumption, especially consumption higher than the recommended upper limits.  These data support current political efforts to reduce or to abstain from alcohol consumption to reduce the incidence of cancer.

Forum Comments

Overview:  This is a very well-done analysis of data collected from the large EPIC study.  The investigators had good follow-up data over a mean of 8.8 years on more than 300,000 subjects.  There were large numbers of subjects developing the most common types of cancer, but limited numbers of cases of upper aerodigestive cancer and liver cancer.  The authors classified a number of cancers as “alcohol-attributable;” many of these were “alcohol-related,” with the latter meaning that other factors (smoking, diet, obesity, etc.) may be more important “causes” of such diseases.

While this is a valuable dataset, the intentions of the authors in the preparation of this paper seem to focus mainly on indicting alcohol as a major cause of cancer.  Many important observations that would be very helpful to clinicians (such as the role of moderate drinking, rather than intake greater than recommended, as a cause of cancer) are not presented clearly.  For example, the authors give very detailed tables of the proportion of cancer related to alcohol intake, but do not explicitly indicate the numbers of cases that relate just to heavy alcohol intake, even though they state that among men the percentage of cancers related to > 2 drinks/day accounted for 57% to 87% of the total alcohol-attributable fractions; consumption > 1/day accounted for 40% to 98% of the fractions for women.  It would have been preferable if they presented data separately for associations with cancer incidence for moderate drinking and for heavier drinking.

It is also unfortunate that the authors do not present stratified data (separately for non –smokers and smokers) to clearly indicate the role that smoking plays in the development of certain cancers.  For upper aero-digestive cancers, in particular, it would be important to demonstrate how smoking, independently and in an interaction with alcohol, relates to such cancers.  As stated by one Forum reviewer: ¨Smoking is a well-known confounder of the alcohol-cancer association.  However, the authors do not acknowledge that The Million Women Study¹ found no association between cancers of the upper aerodigestive tract and intake of alcohol in non-smoking women.”  Another reviewer comments that the authors state that “In the groups of never smokers the number of cases of cancer was limited in the EPIC study, which led to a limited power to assess the association between the consumption of alcohol and risk of cancer in this subgroup.”  The authors do not comment that this is a serious problem; this raises concerns that the a priori purpose of the study was just to demonstrate an association between any alcohol intake and cancer.

Some results of the study are puzzling.  It is hard to understand that 73% of alcohol-attributable liver cancers – a cancer with liver cirrhosis as an essential precursor – are found in men reporting no more than two drinks a day.  It suggests either that these subjects were former heavy drinkers or that they grossly underestimated their alcohol intake.  Further, the authors have focused on cancers with a causal association with alcohol consumption, and do not mention certain cancers (e.g., renal cancer, thyroid cancer, lymphoma) that are inversely associated with alcohol intake.

A Forum reviewer points out that “the authors assume a linear association between alcohol and risk of cancer among lifetime consumers of alcohol and use regression coefficients to express the risk for cancer incidence per 1g/day increment in alcohol consumption.  However, the comprehensive review of more than 7.000 studies on the association of lifestyle factors and cancer that was done by the World Cancer Research Fund (WCRF) in cooperation with the American Institute for Cancer Research² had found thresholds for colorectal cancers and liver cancers.  For example, the WCRF review states that increased risk of colorectal cancer is only apparent above a threshold of 30 g/day of ethanol for both sexes.² Further, cirrhosis is an essential precursor of liver cancer caused by alcohol.  Morgan et al³ state that of 100 persons drinking more than 60 grams of  alcohol per day, 2 persons on average are going to develop cirrhosis per year.  In the large study by Boffetta et al,⁴ the figure illustrating the association between alcohol consumption and mortality suggests a threshold of about 2 units of alcohol per day for alcohol intake and an increase in cancer mortality,

Net effects of alcohol consumption in the population:  Among the major concerns with this paper are a number of inaccurate statements that relate to the net health effects of alcohol consumption on the population.  The investigators state: “ . . . even though light to moderate alcohol consumption might decrease the risk of cardiovascular disease and mortality, the net effect of alcohol is harmful.”  This statement is clearly in error, and is not consistent with current scientific data: almost all well-done studies have shown reductions in total mortality from alcohol intake.  For example, a recent paper from Finland⁵ on the effects on increases in alcohol consumption due to lower prices showed that, indeed, there was an increase in deaths from alcohol-related diseases.  On the other hand, there were larger decreases in total mortality because the increase in alcohol intake was associated with fewer cardiovascular, pulmonary, and other deaths in the elderly.⁵

It is interesting that the authors acknowledge that “for all cause mortality alcohol consumption is often shown to be associated with a lower risk for up to four drinks a day in men and two drinks a day in women.”  But then they state: “Thus, alcohol consumption should not be recommended to prevent cardiovascular disease or all cause mortality.”  They present no data to support such a statement.

Another reviewer comments that “While the authors have such information in this database, they do not present data on diseases other than cancer or on all-cause mortality, and numerous recent studies clearly show that moderate drinking is associated with decreases in risk of total mortality.  Such comments add further concern about the intentions of the authors in these very complex analyses.”  As one Forum reviewer stated: “This paper reflects tunnel vision, looking only at the focus of this particular study, and, once again, neglecting the net benefits of moderate drinking.  We all agree that excessive drinking is bad for your health (which is certainly not ‘breaking news’!) and the paper adds little of clinical importance by not focusing on differences between moderate and heavy consumption.”

A Forum reviewer states: “The paper presents and interprets their results as if the authors are adopting a ‘guilt-oriented’ approach to alcohol consumption, far from scientific neutrality.  (The paper may be poor for advancing scientific knowledge, but good for a newspaper scoop.)  A physician will never suggest to a patient either moderate drinking or abstention on the basis of such data.”  Another reviewer commented that the presentation of results in this paper tends to be misleading: “It presents data in a way that leads to almost complete obfuscation of the real risk of cancer due to alcohol consumption.”

The statement of the authors of this paper that “for cancer, as shown by many studies including ours, there is no sensible limit below which the risk of cancer is decreased” is nonsensical.  It is not possible to prove a zero-risk limit for any risk factor by epidemiological studies.  However, if most studies find a threshold value, it is reasonable to argue that a threshold is the most probable finding considering the plethora of studies on this topic that have been published.

Giovanni de Gaetano has provided an excellent overview of this paper: “The paper’s conclusion is that only 10% of total cancer in men and 3% in women are attributable to alcohol consumption; however, the proportion of this that is attributable to ‘moderate drinking’ is very much smaller.  This is especially the case for the truly ‘alcohol-attributable’ cancers such as upper aero-digestive cancers and liver cancer, where the risk is appreciably increased only for very heavy drinkers and alcoholics.  The authors’ conclusion that these data support the efforts to reduce alcohol consumption in order to reduce the incidence of cancer is formally correct.  However, as moderate alcohol consumption reduces cardiovascular disease and — what is more relevant — total mortality, their conclusion that one should totally abstain from alcohol is not  justified based on current scientific data.  Further, the authors surprisingly conclude that ‘alcohol consumption should not be recommended to prevent cardiovascular disease or all-cause mortality.’  They should more correctly say that heavy alcohol consumption should be avoided as it is associated not only with cancer incidence increase but also with increased fatal and non-fatal cardiovascular events and total mortality.”

References from Forum review

1.  Allen NE, Beral V, Casabonne D, et al.  Moderate alcohol intake and cancer incidence in women.  J Natl Cancer Inst 2009;101: 296-305.

2.  Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. AICR,      Washington DC, 2007.  http://www.cancerinstitute.org.au/cancer_inst/publications/pdfs/pm-2008-03_alcohol-as-a-cause-of-cancer.pdf).

3.  Morgan TR, Mandayam S, Jamal MM.  Alcohol and hepatocellular carcinoma.  Gastroenterogy  2004;127:S87-96.

4.  Boffetta P, Garfinkel L. Alcohol drinking and mortality among men enrolled in an American Cancer Society prospective study.  Epidemiology 1990;1:342-348.

5.  Herttua K, Mäkelä P, Martikainen P.  An evaluation of the impact of a large reduction in alcohol prices       on alcohol-related and all-cause mortality: time series analysis of a population-based natural experiment.  Int J Epidemiol 2011;40:441-454; doi:10.1093/ije/dyp336.

Forum Summary:  A large group of distinguished scientists have published a very detailed and rather complex paper describing the association between alcohol consumption and cancer.  It is based on data from the EPIC study in Europe, with a mean follow up of 8.8 years for more than 300,000 subjects.  The authors describe an increase in risk of many cancers from alcohol intake, but do not give data permitting the detection of a threshold of intake for an adverse effect on cancer risk.  The investigators conclude that “In western Europe, an important proportion of cases of cancer can be attributable to alcohol consumption, especially consumption higher than the recommended upper limits.”

Members of the Forum were concerned that the authors did not separate moderate consumption from heavy consumption for their main analyses, ignored the demonstrated benefits of moderate drinking on total mortality, and did not point out other environmental influences (such as smoking, diet, obesity, etc.) that often have much larger effects on the risk of many cancers than does alcohol consumption.  The authors make statements such as alcohol has negative effects on total mortality that are not supported by the data presented in their paper, and are contradicted by most large-scale population-based studies.  Overall, while this paper supports the well-known association between heavy drinking and an increased risk of upper aero-digestive and certain other cancers, it adds little information useful for the prevention of most types of cancer.

*                      *                      *

Comments on the present paper were provided by the following members of the International Scientific Forum on Alcohol Research:

Yuqing Zhang, MD, DSc, Epidemiology, Boston University School of Medicine, Boston, MA, USA.

Creina Stockley, clinical pharmacology, Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia.

Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark.

Francesco Orlandi, MD, Dept. of Gastroenterology, Università degli Studi di Ancona. Italy.

Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA.

R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA.

Giovanni de Gaetano, MD, PhD, Research Laboratories, Catholic University, Campobasso, Italy.

Roger Corder, PhD, MRPharmS, William Harvey Research Institute, Queen Mary University of London, UK.