Comparative risk assessment of carcinogens in alcoholic beverages using the margin of exposure approach Dirk W. Lachenmeier 1,2 , Maria C. Przybylski 2 and Ju ¨rgen Rehm 1,3,4 1 Epidemiological Research Unit, Institute for Clinical Psychology and Psychotherapy, Technische Universit€ at Dresden, Dresden, Germany 2 Chemisches und Veterin€ aruntersuchungsamt (CVUA) Karlsruhe, Karlsruhe, Germany 3 Social and Epidemiological Research Department, Centre for Addiction and Mental Health (CAMH), Toronto, Canada 4 Dalla Lana School of Public Health, University of Toronto, Toronto, Canada Alcoholic beverages have been classified as carcinogenic to humans. As alcoholic beverages are multicomponent mixtures containing several carcinogenic compounds, a quantitative approach is necessary to compare the risks. Fifteen known and suspected human carcinogens (acetaldehyde, acrylamide, aflatoxins, arsenic, benzene, cadmium, ethanol, ethyl carbamate, formaldehyde, furan, lead, 4-methylimidazole, N-nitrosodimethylamine, ochratoxin A and safrole) occurring in alcoholic beverages were identified based on monograph reviews by the International Agency for Research on Cancer. The margin of exposure (MOE) approach was used for comparative risk assessment. MOE compares a toxicological threshold with the exposure. MOEs above 10,000 are judged as low priority for risk management action. MOEs were calculated for different drinking scenarios (low risk and heavy drinking) and different levels of contamination for four beverage groups (beer, wine, spirits and unrecorded alcohol). The lowest MOEs were found for ethanol (3.1 for low risk and 0.8 for heavy drinking). Inorganic lead and arsenic have average MOEs between 10 and 300, followed by acetaldehyde, cadmium and ethyl carbamate between 1,000 and 10,000. All other compounds had average MOEs above 10,000 independent of beverage type. Ethanol was identified as the most important carcinogen in alcoholic beverages, with clear dose response. Some other compounds (lead, arsenic, ethyl carbamate, acetaldehyde) may pose risks below thresholds normally tolerated for food contaminants, but from a cost-effectiveness point of view, the focus should be on reducing alcohol consumption in general rather than on mitigative measures for some contaminants that contribute only to a limited extent (if at all) to the total health risk. Since the first observation in France in the beginning of the last century that the consumption of absinthe was related to esophageal cancer, 1 epidemiology has established a causal relationship between alcohol consumption in general (i.e., in- dependent of beverage type) and the occurrence of cancer. In 1988, the IARC classified alcoholic beverages into group 1 as carcinogenic to humans. 1 At this time, a causal relationship between alcohol consumption and the occurrence of malig- nant tumors of the oral cavity, pharynx, larynx, esophagus and liver was established. In the following IARC evaluations, colorectal cancer and female breast cancer were added to the list of cancer sites with causal relationship, while only limited evidence points to stomach and pancreas as further sites. 2,3 While the epidemiological evidence on the carcinogenicity of alcoholic beverages had been sufficiently established for several decades, the principal mechanism underlying this relationship has been a matter of debate. For a long time, it was assumed that ethanol itself was not a direct carcinogen. The 1988 IARC monograph, for example, stated that there is inadequate evidence for the carcinogenicity of ethanol in ex- perimental animals. 1 However, this statement was based on lack of well-controlled and well-designed experimental studies rather than on a clear absence of effect. Since then, two adequately designed long-term animal studies have clearly demonstrated that ethanol causes a dose-related increase in cancer in mice and rats at sites similar to those observed in humans (liver and oral cavity). 4,5 As a result of this new evi- dence, the 2007 IARC evaluation concluded that there is suf- ficient evidence in experimental animals for the Key words: alcoholic beverages, risk assessment, dose–response relationship, margin of exposure, epidemiology Abbreviations: BMD: Benchmark dose; BMDL: Confidence limit of the BMD; CPDB: Carcinogenic potency database; EFSA: European food safety authority; EPA: Environmental protection agency; IARC: International agency for research on cancer; IPCS: International programme on chemical safety; JECFA: Joint FAO/WHO expert committee on food additives; MOE: Margin of exposure; NDMA: N- nitrosodimethylamine; NTP: National toxicology program Additional Supporting Information may be found in the online version of this article. Grant sponsor: European Community 7th Framework Programme, Grant Agreement No 266813 - Addiction and Lifestyles in Contemporary Europe-Reframing Addictions Project (ALICE RAP) DOI: 10.1002/ijc.27553 History: Received 3 Jan 2012; Accepted 12 Mar 2012; Online 24 Mar 2012 Correspondence to: Dirk W. Lachenmeier, Chemisches und Veterin€ aruntersuchungsamt (CVUA) Karlsruhe, Weissenburger Strasse 3, D-76187 Karlsruhe, Germany, Fax: þ49-721-926-3549, E-mail: Lachenmeier@web.de Epidemiology Int. J. Cancer: 131, E995–E1003 (2012) V C 2012 UICC International Journal of Cancer IJC