Critique 094: The complex association between moderate alcohol consumption and breast cancer ——– 23 October 2012
Brooks PJ, Zakhari S. Moderate alcohol consumption and breast cancer in women: From epidemiology to mechanisms and interventions. Alcohol Clin Exp Res 2012; pre-publication: DOI: 10.1111/j.1530-0277.2012.01888.x.
Epidemiologic studies indicate that moderate alcohol consumption increases breast cancer risk in women. Understanding the mechanistic basis of this relationship has important implications for women’s health and breast cancer prevention. In this commentary, we focus on some recent epidemiologic studies linking moderate alcohol consumption to breast cancer risk and place the results of those studies within the framework of our current understanding of the temporal and mechanistic basis of human carcinogenesis.
This analysis supports the hypothesis that alcohol acts as a weak cumulative breast carcinogen and may also be a tumor promoter. We discuss the implications of these mechanisms for the prevention and treatment of alcohol-related breast cancer and present some considerations for future studies. Moderate alcohol consumption has been shown to benefit cardiovascular health and recently been associated with healthy aging. Therefore, a better understanding of how moderate alcohol consumption impacts breast cancer risk will allow women to make better informed decisions about the risks and benefits of alcohol consumption in the context of their overall health and at different stages of their life. Such mechanistic information is also important for the development of rational clinical interventions to reduce ethanol-related breast cancer mortality.
Background: Most epidemiologic studies have shown a slight increase in the occurrence of breast cancer among women who report alcohol consumption, even among those consuming an average of only seven drinks per week. The present study utilizes data from recent well-done epidemiologic studies and mechanistic studies to evaluate two separate hypotheses that may explain the alcohol-breast cancer association: (1) alcohol acts as a promoter of breast cancer progression (it may increase growth of a pre-existing cancer so that it becomes clinically evident); and (2) long-term alcohol use serves as a weak cumulative breast carcinogen.
Key points of paper: The authors make several important points at the onset: alcohol intake is usually under-reported by subjects (which could exaggerate the harm associated with light drinking, or even lead to spurious associations at lower levels of consumption). Further, most epidemiologic studies are not able to adjust for the pattern of drinking, and usually include binge drinkers and regular moderate drinkers who have similar total weekly alcohol consumption within the same category. Given that the blood alcohol level may be the most important mechanism for effects on cancer risk, the pattern by which a woman consumes a given amount of alcohol is especially important in interpreting associations.
Alcohol as a promoter of progression of a previously existing breast cancer: Another point of the authors is that the development of breast cancer requires many years, and is much longer than the observational periods of almost all epidemiologic studies. Hence, any breast cancer that may be initially diagnosed during the follow-up of such studies was presumably present long before the study began, and if an increase is found for consumers of alcohol it must be due to promotion of growth of an already existing tumor. Estrogen has been proposed as a mechanism by which alcohol consumption could increase the risk of detectible breast cancer, but the authors emphasize that research data indicate that moderate drinking does not increase estrogen levels in post-menopausal women. The authors conclude: “Therefore, the observed relationship between daily consumption of 10 g of alcohol and breast cancer risk in postmenopausal women cannot be explained by a mechanism involving increased serum estrogen levels.” The authors state that there is the possibility that alcohol could stimulate epithelial-mesenchymal transition (EMT) or global DNA methylation, which could increase risk, but current data are insufficient to evaluate these mechanisms.
Alcohol as a weak cumulative breast cancer carcinogen: Current epidemiologic data are insufficient to test the validity of the second hypothesis: that chronic alcohol use over the lifetime may act as a weak cumulative breast carcinogen. Most studies have limited data on the long-term alcohol exposure of subjects; most use “baseline” data, usually when the women are in middle-age or older, to estimate their alcohol exposure. Given that cancer develops over many decades, the baseline exposure may be important only to the degree that it reflects previous intake. The authors describe recent data on the effects that varying levels of alcohol dehydrogenase (ADH) could have on blood levels of aldehyde, a known carcinogen. Further, alcohol–inducible CYP2E1 could be another mechanism by which alcohol affects risk; higher levels of blood alcohol concentration (as after binge drinking) would be expected to have the most effect. If either of these mechanisms turns out to be key, this raises the possibility of novel approaches for the prevention of breast cancer using specific medications to block such effects.
The net effect on health of alcohol consumption: The authors also point out that scientific data strongly indicate that regular moderate drinking in post-menopausal women is associated with a much lower risk of diseases much more common than breast cancer, especially cardiovascular disease and diabetes. Further, moderate alcohol consumption is associated with a decrease in disability in old age and in total mortality. Thus, advice to women will vary according to age, to genetic factors (currently poorly understood), and to many environmental factors. Physicians are somewhat limited at present in being able to provide the best advice to individual patients, but the results of further research to help better define the hypothetical mechanisms described in this paper could be critical.
Specific comments by Forum members: Forum members considered that this paper provided a very balanced and thorough review of the topic. Said reviewer Skovenborg: “I find the paper sound with plausible hypotheses and an intelligent discussion of the two hypotheses. The only thing I am missing in the discussion of drinking pattern is the influence of drinking with food on the blood alcohol level, which is considerable and too often forgotten.”
Forum reviewer Finkel also liked the paper and thought that the authors exhibited both command and competence in their discussions of this complex subject. “The authors consider many essentials, all of which complicate their job, but which make it all the more valid: cytokinetics, hormonal interactions, drinking patterns, the possible role of binge drinking in the midst of what might appear to be moderation, the health benefits of drinking, the need to refine drinking pattern understanding, possible practical applications, and the value and limitations of epidemiology. True, they pose more questions than answers, but isn’t that the beginning of understanding, particularly of such a confused and conflicted subject?” Finkel adds “Not touched on are the differential effects, if any, of various beverages, the role of folic acid, or genetic susceptibilities, among other factors, but it’s too much to ask as yet. This is a most complex set of relationships, no doubt encompassing a number of subsets of people afflicted with breast cancer.”
Forum reviewer Gretkowski believed that this article offers a concise review of the contemporary opinion on the ethanol/breast carcinogenesis issue. “Drinking amounts, patterns that could effect alcohol elimination or cumulative toxicity of free radicals, are all treated as co-models within a model of estrogen elevation/tumor initiator or promoter model.”
She continues: “Arguably, estrogen levels might need to be followed over a prolonged period of time with very sensitive assay to determine no absolute estrogen level increase actually occurs. On the other hand, the idea of measuring serum estrogen levels is quite simplistic and not surprising that it is not through pure measurement of levels that we could discern a dose-dependent effect. Differences in bound vs unbound hormone, as well as dihydroepiandrostenedione conversion levels peripherally, and subsequent estrogen levels or receptor occupation levels may hold more information. This could all shed light on the higher incidence of recurrence in ER+ patients who consume alcohol in moderation, with alcohol clearly fitting a different lock than tamoxifen or aromatase inhibitors.”
Reviewer Ellison emphasizes the statements of the authors regarding future directions of research. “It may be especially important to judge the effects of the pattern of drinking on cancer risk, with the assumption that the effects of regular consumption of small amounts of alcohol will be quite different from those of binge drinking leading to high blood alcohol levels. Further animal studies of the effects of the long-term administration of small amounts of alcohol in terms of breast cancer risk would also be important.”
Reviewer Van Velden adds a further element that needs consideration: “This is a well written and balanced paper. What we really miss is the nutrigenetic aspects of alcohol and breast cancer, as well as the protective effects of folic acid. Alcohol may switch on certain carcinogenic genetic polymorphisms; diet does play a very important role in the promotion of breast cancer.”
Forum reviewer Ursini thought that the authors may have been over zealous in their speculations and conclusions. “While attempts to find mechanisms for the carcinogenic effect of alcohol are important, we know too little about the overall underlying causes of cancer. This issue is not so easy to achieve, even for compounds that are by far much more efficient than alcohol in increasing an epidemiologically detectible risk. In this respect, and in light of the minimal increase in risk associated with alcohol intake, searching for or discussing the mechanism on the basis of the limited available biological data may not be scientifically sound.
“Also, the suggestion that pre- or post-menopausal drinkers will accept or reject a potentially harmful habit does not appear well founded, particularly in view of the demonstrated effects of abstinence on increasing cardiovascular disease and total mortality. By no means can the speculations such as those reported in this paper be used to support women abstaining from alcohol to prevent breast cancer.”
An excellent review article from two scientists at the National Institue on Alcohol Abuse and Alcoholism describes the epidemiologic and basic scientific evidence linking alcohol consumption to the risk of breast cancer. The authors point out deficiencies in the epidemiologic data, especially that the pattern of drinking (regular moderate versus binge drinking) has generally not been taken into consideration (and the latter pattern can be associated with much higher blood alcohol concentrations). Further, epidemiologic studies usually provide data for only a short period of time, while the development of cancer may relate to exposures over many decades. The authors also comment upon the effect that under-reporting of alcohol by study participants could exaggerate effects on cancer risk from light drinking. They discuss two hypotheses that could relate alcohol to breast cancer risk: alcohol as a breast tumor promoter, and alcohol as a weak cumulative breast carcinogen, and present evidence from epidemiology and basic science that would relate to each hypothesis.
Overall, Forum reviewers were enthusiastic about this review paper, considering that it clearly outlined some of the difficulties scientists have in determining the causes of cancer. They agreed with the authors’ statements regarding the necessity to consider the overall net effects of moderate drinking, including reductions in the risk of cardiovascular disease and total mortality. They also agreed that future epidemiologic studies should focus on the pattern of drinking (and not just the average weekly amount of alcohol), and with their suggestions for future animal studies. One Forum reviewer cautioned that our understanding of the causes of breast cancer is still very incomplete, limiting our ability to provide well-founded recommendations to the public regarding moderate drinking as it relates to breast cancer risk.
* * *
Coments on this paper were provided by the following members of the International Scientific Forum on Alcohol Research:
Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark
Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA
Lynn Gretkowski, MD, Obstetrics/Gynecology, Mountainview, CA, Stanford University, Stanford, CA, USA
R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA
David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa
Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway
Fulvio Ursini, MD, Dept. of Biological Chemistry, University of Padova, Padova, Italy