Critique 042: An increase in risk of certain types of gastric cancer from heavy drinking, but no increase from moderate alcohol consumption – 17 May 2011

Tramacere I, Negri E, Pelucchi C, Bagnardi V, Rota M, Scotti L, Islami F, Corrao G, La Vecchia C, Boffetta P.  A meta-analysis on alcohol drinking and gastric cancer risk.  Ann Oncology 2011;doi:10.1093/annonc/mdr135

Authors’ Abstract

Background:  Whether an association between alcohol drinking and gastric cancer risk exists is an open question.  In order to provide a definite quantification of the association between alcohol drinking and gastric cancer risk, we conducted a meta-analysis of available data.

Patients and methods:  We carried out a PubMed search of articles published up to June 2010 and identified 44 case-control and 15 cohort studies, including a total of 34,557 gastric cancer cases.  We derived meta-analytic estimates using random-effects models, taking into account correlation between estimates.  We carried out a dose-risk analysis using nonlinear random-effects meta-regression models.

Results:  Compared with nondrinkers, the pooled relative risk (RR) was 1.07 [95% confidence interval (CI) 1.01–1.13] for alcohol drinkers and 1.20 (95% CI 1.01–1.44) for heavy alcohol drinkers (≥ 4 drinks per day).  The pooled estimates were apparently higher for gastric noncardia (RR for heavy drinkers = 1.17, 95% CI 0.78–1.75) than for gastric cardia (RR = 0.99, 95% CI 0.67–1.47) adenocarcinoma.  The dose-risk model estimated a RR of 0.95 (95% CI 0.91–0.99) for 10 g/day and 1.14 (95% CI 1.08–1.21) for 50 g/day.

Conclusions: This meta-analysis provides definite evidence of a lack of association between moderate alcohol drinking and gastric cancer risk.  There was, however, a positive association with heavy alcohol drinking.

Forum Comments

There have been conflicting reports on whether or not alcohol consumption, especially moderate drinking, relates to the risk of gastric cancer.  Forum reviewers of this paper were unanimous in stating that this paper presents a thorough, well-done meta-analysis that helps clarify the relation between alcohol consumption and gastric cancer.  The data suggest no relation with moderate drinking, but a slight increase in risk with heavier consumption.  No data are presented in this paper on the association with cancer for different types of alcoholic beverages, and data were not available on the pattern of drinking.

One Forum reviewer continued to worry about residual confounding from smoking for alcohol’s relation to stomach cancer, especially that of the gastric cardia.  In some studies, there is little effect of alcohol among non-smokers for many upper aero-digestive cancers.  The authors of the present study state: “Among nonsmokers, the summary RR for alcohol drinking was 1.06 (95% CI 0.96–1.17; I-squared = 0.0%, P = 0.520), based on four case–control and three cohort studies (data not shown).”

Potential mechanisms for an effect of alcohol on gastric cancer:  For cardiovascular disease, the J-shaped pattern seen for the association with alcohol consumption is usually explained by a large number of biologic and physiologic mechanisms (increases in HDL-cholesterol, decreased clotting, increased fibrinolysis, improved endothelial function, etc., especially from moderate drinking).  However, fewer potential mechanisms have been identified to explain a J-shaped effect for cancer, as was shown in this study.

Forum member Fulvio Ursini believes that the J-shaped curve demonstrated here may be further evidence of a hormetic effect of alcohol.  That is, that a substance (such as alcohol, poisons, or other stressors) may have a generally favorable effect in small doses but a toxic effect at higher doses.  “The hormetic effect is well known for several ‘toxic’ compounds (including carcinogens) as well as for ionizing radiation.  The remarkable evidence emerging from this study is that the hormetic effect of ethanol, already observed for cardiovascular disease and degenerative diseases in general, is true for cancer as well.”

Professor Ursini adds: “The reasonable hypothesis of an involvement of the Keap1- Nrf2 system (activated by electrophiles produced by ethanol metabolism and shifting the cellular redox status towards a more reduced environment), suggested by in vitro studies, needs now an epidemiological validation.”  Another Forum member, David Vauzour, concurs: “This association is likely to be the results of ethanol on signalling pathways.  One of the best explanations may indeed be the Keap1-nrf-2 pathway, although other potential mechanisms aiming at modulating the redox homeostasis cannot be discarded.  Literature is scarce regarding the impact of alcohol on cellular signalling and this topic needs to be further investigated.” 

The rather striking lack of an effect of alcohol on gastric cancer risk among Asians may relate to their generally lower levels of certain ALDH2 enzymes that help clear aldehyde from the blood.  In general, Asians tend to have the inactive ALDH2 *2 allele more commonly than do westerners, and such subjects tend to consume less alcohol.  In a recent study from Korea on this topic, published after the present analysis was completed, Shin et al1 concluded: “A dose-response association between alcohol intake and the risk for gastric cancer was not observed.  However, ALDH2 polymorphisms were found to modify the susceptibility to the development of gastric cancer associated with alcohol intake,” (which showed higher risk for persons with the ALDH2 *1/*2 genotype).  The International Agency for Research on Cancer (IARC) has recently concluded that “the heterozygous (ALDH2 *1/*2) genotype contributes substantially to oesophageal carcinogenesis via acetaldehydemia derived from the metabolism of ethanol in alcoholic beverages.”2  Overall, findings suggest that there may be an alcohol-ALDH2 genotype interaction in several types of cancer.

References from Forum review

1.  Shin CM, Kim N, Cho SI, Kim JS, Jung HC, Song IS.  Association between alcohol intake and risk for gastric cancer with regard to ALDH2 genotype in the Korean population.  Int J Epidemiol 2011 Apr 19. [Epub ahead of print].

2.  Baan R, Straif K, Grosse Y, et al.  Carcinogenicity of alcoholic beverages.  Lancet Oncol 2007;8:292–293.

Forum Summary:  This well-done meta-analysis supports other data generated on the risk of alcohol consumption and gastric cancer – that is – that the risk may be real for heavy alcohol consumption but not for moderate intake.  The type of gastric cancer relating to heavier alcohol intake in this study tended to be tumors involving the noncardia, but differences between the association with tumors of the cardia were not significant. 

 There was no increased risk of gastric cancer from alcohol among Asians; this may be due to lower alcohol intake; there is a greater prevalence among Asians of the inactive ALDH2 genotype that is associated with flushing and other adverse effects of alcohol, and such subjects tend to drink less alcohol.  However, a number of studies have shown higher risk for upper aero-digestive cancers for subjects with this ALDH2 genotype, so the overall reason for the lower risk among Asians in this study remains unclear.  The main outcome of the study is that there is no increase in the risk of gastric cancer associated with the moderate intake of alcohol.

                                                                         *                      *                      *

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

Fulvio Ursini, MD, Dept. of Biological Chemistry, University of Padova, Padova, Italy.

Luc Djoussé, MD, DSc, Dept. of Medicine, Division of Aging, Brigham & Women’s Hospital and Harvard Medical School, Boston, MA, USA.

David Vauzour, PhD, Senior Research Associate, Department of Nutrition, Norwich Medical School, University of East Anglia, Norwich, UK.

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

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

Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway.

Gordon Troup, MSc, DSc, School of Physics, Monash University, Victoria, Australia.

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