Choi Y-J, Myung S-K, Lee J-H. Light Alcohol Drinking and Risk of Cancer: A Meta-analysis of Cohort Studies. Cancer Research and Treatment 2017; pre-publication release.
Purpose To determine whether light alcohol drinking increases the risk of cancer by using a meta-analysis of cohort studies because the newly revised 2015 European Code against Cancer 4th edition on alcohol and cancer was based on critical flaws in the interpretation and citation of the previous meta-analyses.
Materials and Methods PubMed and EMBASE were searched in April, 2016. Two authors independently reviewed and selected cohort studies on the association between very light (≤0.5 drink/day), light (≤1 drink/day), or moderate drinking (1-2 drinks/day) and the risk of cancer incidence and mortality. A pooled relative risk with its 95% confidence interval was calculated by a random-effects meta-analysis. Main outcome measures were cancer incidence and mortality.
Results A total of 60 cohort studies from 135 articles were included in the final analysis. Very light drinking or light drinking was not associated with the incidence of most cancers except for female breast cancer in women and male colorectal cancer. Conversely, light drinking was associated with a decreased incidence of both female and male lung cancer significantly and both female and male thyroid cancer marginally significantly. Moderate drinking significantly increased the incidence of male colorectal cancer and female breast cancer, whereas it decreased the incidence of both female and male hematologic malignancy.
Conclusion We found that very light or light alcohol drinking was not associated with the risk of most of the common cancers except for the mild increase in the incidence of breast cancer in women and colorectal cancer in men.
Epidemiologic studies for many decades have shown a slight increase in the risk of breast cancer associated with alcohol for women, even those reporting only light drinking. For other types of cancer, there has usually been no appreciable increase in risk except for consumers of greater amounts of alcohol, i.e, a threshold for effect. Nevertheless, many scientific publications and recommended guidelines have touted the association of alcohol intake with cancer risk without describing differences according to the amount consumed. Frequently, such reports are based on publications that include data from case-control studies, which are unreliable at evaluating cause, or other deficiencies, as discussed below. This has led to public confusion about the actual risk of cancer related to alcohol.
In 2015, Scoccianti et al published an article based on recent recommendations from the International Alcohol Research Council (IARC), which concluded: “If you drink alcohol of any type, limit your drinking. Not drinking is better for cancer prevention.” The authors of the present paper dispute that assertion that no alcohol is optimum, especially for overall health outcomes. They have conducted an extensive new meta-analysis to try to develop a more accurate overview of the relationship by limiting their study to cohort studies of high quality. Their analyses are based on data from 60 cohort studies, described in 135 articles. They conclude that, other than breast cancer and possibly colon cancer in men, risks are not increased for most types of cancer from very light drinking.
Potential reasons why even light drinking may be related to increased cancer risk in some epidemiologic studies: Forum member Ellison suggested that there are a number of reasons why an increase in risk for cancer might be considered present for even light drinkers, even if the actual increase in risk may be primarily (or exclusively) from greater intake. “First, some studies showing this relation have been based on lower quality studies, including case-control studies, where recall of earlier alcohol consumption is more likely to be biased. Another reason that the estimates of the exposure (alcohol) may be imprecise is that most epidemiologic studies have not included adequate data on the pattern of drinking (binge versus regular moderate, and with or without food), that may strongly affect the relation of alcohol to cancer.
“Another important factor leading to a false implication of light drinking may be the failure to adjust for important confounders, especially associated lifestyle factors such as smoking (much more common among heavy drinkers), or for other factors such as socio-economic status (SES) of subjects (based on income, education, job level, etc.). Even when reporting similar amounts of alcohol, subjects at lower SES levels have been shown to have many more adverse effects of alcohol, including alcohol-attributable cancers, than those with higher SES (as recently described by Katikireddi et al). Also, the occasional report that shows an increase in known alcohol-attributable cancers and other conditions from light drinking may well be the result of under-reporting of alcohol intake.” This is discussed in the section below.
Forum member Finkel noted: “I find no fault with this paper. But we should not place much weight on small differences, say in cancer incidence related to drinking-quantity categories, nor is fine distinctions between drinking categories a good idea. Cutting them too fine will ensure error. I say this even should the statistical reliability calculate within given margins. We should demand clear-cut, readily seen differences. (Now I suppose I’ve incurred the wrath of all the statisticians.) And once again we lack potentially important and relevant data: nutrition, especially regarding folate; beverage choices; drinking pattern; eating while drinking; and so on.” And Finkel added: “Once again, as so often happens, in discussing the significance of their results the authors ignore that cardiovascular and other health benefits of light-moderate drinking vastly outweigh any liabilities that may be incurred in increasing cancer risk, even among women, even with regard to breast cancer.”
Reviewer Skovenborg noted: “I agree with other Forum members that it would have been nice to have more information about important factors like drinking pattern, drinking with/without meals, beverage choice and better alcohol exposure data. It is interesting that very light drinking is associated with a decreased mortality of breast cancer. The association of alcohol consumption with malignant melanoma makes me wonder once again – how to explain this observation? That said, I think that the paper and its conclusions are sound and well done.”
Reviewer Van Velden stated: “I agree with these comments, especially the fact that most reports do not take lifestyle factors into consideration. This includes SES, smoking, healthy diet, and pattern of alcohol consumption. Alcohol can never be seen in isolation, but as a lifestyle pattern. It is a well-balanced paper.”
Underreporting of alcohol consumption as a factor in estimating causation in epidemiologic studies: Klatsky et al have demonstrated in their very large Kaiser-Permanente studies that associations between reported light drinking and cancer are often seen only in subjects who appear to be underreporting their alcohol intake, but not in those deemed to not be underreporters (Klatsky et al, 2014). These authors have carefully reviewed the extensive medical records collected in the Kaiser-Permanente study, often over decades, for evidence of an alcohol use disorder at any time; these included reports of alcoholism, social problems related to alcohol, alcoholic psychosis, excess blood alcohol level, alcohol poisoning, alcoholic cardiomyopathy, alcoholic fatty liver or cirrhosis, etc. Then, when these subjects stated that they consumed “1 to 2 drinks/day” in a subsequent study of alcohol and cancer, they were classified as “likely underreporters.” For other subjects who had similar amounts of data previously collected that did not list any such conditions, they were classified as “unlikely underreporters.”
In an analysis of alcohol and cancer in their cohort, Klatsky and his colleagues report that, overall: “Persons reporting light–moderate drinking had increased cancer risk in this cohort. For example, the hazard ratios (95 % confidence intervals) for risk of any cancer were 1.10 (1.04–1.17) at 1 drink per day and 1.15 (1.08–1.23) at 1–2 drinks per day. Increased risk of cancer was concentrated in the stratum suspected of underreporting. For example, among persons reporting 1–2 drinks per day risk of any cancer was 1.33 (1.21–1.45) among those suspected of underreporting but 0.98 (0.87–1.09) among those not suspected of such underreporting. These disparities were similar for the alcohol-related composite and for breast cancer.” Further, the same disparities had earlier been reported for hypertension (Klatsky et al, 2006), with an increased risk of hypertension among persons reporting 1-2 drinks per day being seen only among those who had been identified as probable underreporters of their alcohol intake.
There findings suggest that some of the increases in cancer risk reported in previous epidemiologic studies to occur from light drinking may well be due to a poor assessment of alcohol intake, in that subjects who actually drink in excess may claim to be light or moderate drinkers. Obviously, better means of estimating not only the amount of alcohol but the pattern of drinking are needed to more precisely judge the relation of light drinking to cancer and other diseases.
The overall relation of alcohol intake to selected cancers and mortality: Forum member Pajak had a number of observations: “The final number of participants included in the analysis is impressing but the numbers of studies on cancer of particular sites are much less. Sometimes inclusion of one study only puts a question to the results of the meta-analysis. In reference to the authors’ final statement to change the recommendations, my problem would be that they have paid little attention to potential confounders. Then, protective effect for lung cancer (alcohol consumption is correlated with smoking) and protective effect on esophagus cancer sounds a little tricky. There is no debate on social determinants as confounders. For breast cancer and colorectal for which the number of the studies seems to be sufficient the relation does not support the “no effect” hypothesis but the relations seems to be weak (OR=1.09 and 1.04 respectively) and despite statistical significance, is not necessarily related to causality. I think that the amount of work done for this paper is impressing, but still the authors should present more skepticism towards their own results.”
Forum member De Gaetano also had some cogent remarks on the results of this study: “This is an important meta-analysis that limits the impact a previous meta-analysis by Bagnardi et al (2015) had on the discussion related to moderate alcohol consumption and cancer, in particular breast cancer. The distinction between case-control and cohort studies is crucial. Overall, data show the following: (1) Very light drinking increases breast cancer incidence by 4% but reduces breast cancer mortality by 21%; (2) Light drinking increases breast cancer incidence by 9% , but has no association with mortality; (3) Moderate consumption increases breast cancer incidence by 13% and mortality by 4%.
“Breast cancer incidence is important, but more important is specific mortality, that appears very marginally or not increased by very low to moderate alcohol consumption. Even more important is considering all-cause mortality, information not included in this meta-analysis. A number of large meta-analyses indicate that total mortality is significantly reduced by very light/moderate drinking.
“As the significant beneficial effect of light to moderate alcohol drinking on cardiovascular (CV) risk is without doubt, one may conclude that people at high CV risk but low alcohol-related cancer risk may safely drink in a moderate and regular way. In contrast, people with increased alcohol-related cancer risk (such as young women with a familial history of breast cancer) should preferably abstain from alcohol. However, when the same women arrive at menopause, their CV risk would greatly increase, while cancer risk would remain rather stable: at that moment, a large CV and total mortality risk prevention by moderate alcohol consumption would become more prevalent than a relatively small increase of breast cancer risk.”
Forum member Stockley reviewed previous data on alcohol consumption and certain types of cancer. “Although not seen consistently in studies, a meta-analysis of 27 cohort and 34 case-control studies by Fedirko et al also provided strong evidence for an association between alcohol drinking of >1 drink/day (>12.5 g/day ethanol) and colorectal cancer risk. The relationship was stronger for men than for women possibly reflecting the limited number of studies reporting data on high alcohol intake among women as well as lower average alcohol consumption in women as compared with men.
“From a brief scan of the literature, however, conclusions drawn re the relationship of alcohol consumption and risk of lung cancer are more inconsistent. For example, Benedetti et al simply saw a positive relationship, Chao et al suggested that there was a dose-dependent relationship with a threshold of above three standards drinks/day, the older studies of De Stefani et al and Potter et al suggested that there was no relationship, while seven studies suggested that the relationship was inverse. Indeed, the relationship between alcohol consumption and lung cancer risk may actually be J-shaped, such that there may be a protective effect for light alcohol consumption, and a weaker protective effect for moderate drinking, although other meta-analyses suggest that alcohol consumption does not play an independent role in lung cancer aetiology (Bagnardi et al, 2011), and that residual confounding by smoking may explain part of observed relationships with heavier alcohol consumption (Korte et al, Freudenheim et al).
Forum member Mattivi noted: “This study brings the focus on the role between light alcohol consumption and breast cancer in women and male colorectal cancer (CRC). The meta-analysis would bring additional progress in understanding alcohol’s enhancing effect on these cancers if other specific factors could have explored. It has been suggested that factors, such as obesity, folate deficiency, and genetic susceptibility may contribute additional CRC risk for those consuming alcohol (Klarich et al). It has also been reported that increased dietary folate intake reduces breast cancer risk for women with higher alcohol intake level. However, this was not the case for those with lower alcohol intake (Chen et al). Since there is currently only moderate evidence of the role between changes in folate metabolism and colon and breast cancer, it would have been particularly interesting to explore the specific case of light drinkers. To provide effective nutritional advice to the population, a better mechanistic understanding of the interactions between alcohol and other risk factors and the complex biological mechanisms involved for breast and colorectal cancer are needed.”
References from Forum critique
Bagnardi V, Rota M, Botteri E, Scotti L, Jenab M, Bellocco R, et al. Alcohol consumption and lung cancer risk in never smokers: a meta-analysis. Ann Oncol 2011;22:2631-2639.
Bagnardi V, Rota M, Botteri E, Tramacere I, Islami F, et al. Alcohol consumption and site-specific cancer risk: a comprehensive dose–response meta-analysis. Br J Cancer 2015;112:580–593. doi: 10.1038/bjc.2014.579.
Benedetti A, Parent ME, Siemiatycki J. Lifetime consumption of alcoholic beverages and risk of 13 types of cancer in men: results from a case-control study in Montreal. Cancer Detect Prev 2009;32:352-362.
Chao C, Li Q, Zhang F, White E. Alcohol consumption and risk of lung cancer in the VITamins And Lifestyle Study. Nutr Cancer 2011;63:880-888.
Chen P, et al. Higher dietary folate intake reduces the breast cancer risk: a systematic review and meta-analysis. Br J Cancer 2014;110:2327-2338, DOI: 10.1038/bjc.2014.155.
De Stefani E, Correa P, Fierro L, Fontham ET, Chen V, Zavala D. The effect of alcohol on the risk of lung cancer in Uruguay. Cancer Epidemiol Biomarkers Prev 1993;2: 21-26.
Fedirko V, Tramacere I, Bagnardi V, Rota M. Scotti L, Islami F, et al. Alcohol drinking and colorectal cancer risk: an overall and dose-response meta-analysis of published studies. Ann Oncol 2011;22:1958-1972.
Freudenheim JL, Ritz J, Smith-Warner S-A, Albanes D, Bandera EV, et al. Alcohol consumption and risk of lung cancer: a pooled analysis of cohort studies. Am J Clin Nutr 2005;82:657-667.
Katikireddi SV, Whitley E, Lewsey J, Gray L, Leyland AH. Socioeconomic status as an effect modifier of alcohol consumption and harm: analysis of linked cohort data. Lancet Public Health 2017. Online publication May 10, 2017. http://dx.doi.org/10.1016/S2468-2667(17)30078-6
Klarich DS et al., Moderate Alcohol Consumption and Colorectal Cancer Risk, Alcoholism-Clin Exp Res 2015;39:1280-1291. DOI: 10.1111/acer.12778.
Klatsky AL, Gunderson E, D G, Kipp H, Udaltsova N, Friedman GD. Higher prevalence of systemic HTN among moderate alcohol drinkers: exploring the role of under-reporting. J Stud Alcohol 2006;67:421–428.
Klatsky AL, Udaltsova N, Li Y, Baer D, Nicole Tran H, Friedman GD. Moderate alcohol intake and cancer: the role of underreporting. Cancer Causes Control 2014;25:693-699. doi: 10.1007/s10552-014-0372-8.
Korte JE, Brennan P, Henley SJ, Boffetta P. Dose-specific meta-analysis and sensitivity analysis of the relation between alcohol consumption and lung cancer risk. Am J Epidemiol 2002;155:496-506.
Potter JD, Sellers TA, Folsom AR, McGovern PG. Alcohol, beer, and lung cancer in postmenopausal women. The Iowa Women’s Health Study. Ann Epidemiol 1992;2:587-595.
Scoccianti C, Cecchini M, Anderson AS, Berrino F, Boutron-Ruault MC, Espina C, et al. European Code against Cancer 4th Edition: Alcohol drinking and cancer. Cancer Epidemiol. 2015;39 Suppl 1:S67-74.
The association between the consumption of alcohol and the risk of cancer has been of great interest for many decades. There are a number of types of cancer, especially those of the upper aero-digestive tract (such as mouth, tongue, pharynx, etc.) that are clearly increased among heavy drinkers, especially among subjects who are also heavy smokers. Cancer of the liver can be a result of alcoholic liver cirrhosis, related to long-term heavy drinking. Further, an increased risk for many other cancers have been shown for heavy drinkers, but generally not for light or moderate drinkers.
The risk of breast cancer in women, however, is usually found to be slightly higher in even light drinkers than it is among non-drinkers. This has been a common finding, although some studies suggest that the pattern of drinking, the estimated level of underreporting of alcohol intake, use of hormone replacement therapy, and level of folate intake may all affect this association. For colorectal cancer, some studies suggest an increase in risk even among light-to-moderate drinkers. This has led some to proclaim that no alcohol consumption is preferable for the prevention of cancer. Others suggest that it is important to consider not only cancer, but other diseases; for example, coronary heart disease, ischemic stroke, diabetes, and dementia (all of which are important causes of disability and death) occur less frequently among moderate drinkers than among non-drinkers. Further, the risk of total mortality is almost always found to be lower among light to moderate drinkers than among abstainers.
The present analysis was designed to help determine the association of light or moderate drinking with the risk of cancer. It consisted of a meta-analysis based on a total of 60 studies from 135 articles and included only cohort studies of high quality. The analysis focused on the association between very light (≤0.5 drink/day), light (≤1 drink/day), or moderate drinking (1-2 drinks/day) and the risk of cancer incidence and mortality.
The authors found that very-light drinking was not associated with the incidence of most cancers except for female breast cancer and male colorectal cancer, and was associated with a decreased incidence of both female and male lung cancer and both female and male thyroid cancer. Moderate drinking significantly increased the incidence of male colorectal cancer and female breast cancer, whereas it decreased the incidence of both female and male hematologic malignancy.
Forum members considered this to be a well-done meta-analysis, based on a large number of studies. The main weaknesses of the study related to the lack of ability of the authors to evaluate a number of known confounders: underreporting of alcohol intake, the pattern of drinking or the type of beverage consumed, socio-economic status, and folate intake or other dietary factors. Further, the authors did not describe the effects of alcohol consumption on total mortality.
The results of this study, obviously, support most previous epidemiologic evidence on the association between alcohol consumption and cancer. Larger amounts of alcohol are associated with increased risk of a number of cancers, and even light consumption is associated with the risk of breast cancer, and possibly colorectal cancer. However, given the inability to adjust for many confounders in this study, and lack of data on the net effects in terms of total mortality, it remains difficult for scientists to provide scientifically sound and balanced guidelines regarding the health consequences of alcohol consumption that are applicable to everyone in the population.
* * * * * *
The following members provided comments for this critique by the International Scientific Forum on Alcohol Research:
Elizabeth Barrett-Connor, MD, Distinguished Professor, Division of Epidemiology, Department of Family Medicine and Public Health and Department of Medicine, University of California, San Diego, La Jolla, CA USA
Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy
R. Curtis Ellison, MD, Professor of Medicine & Public Health, Boston University School of Medicine, Boston, MA, USA
Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA
Fulvio Mattivi, MSc, Head of the Department of Food Quality and Nutrition, Research and Innovation Centre, Fondazione Edmund Mach, in San Michele all’Adige, Italy
Professor Andrzej Pająk, Epidemiology and Population Studies, Jagiellonian University Medical College, Kraków, Poland
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia, Australia
Pierre-Louis Teissedre, PhD, Faculty of Oenology–ISVV, University Victor Segalen Bordeaux 2, Bordeaux, France
Fulvio Ursini, MD, Dept. of Biological Chemistry, University of Padova, Padova, Italy
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa
Andrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA