Critique 109: In women developing breast cancer, moderate alcohol consumption before or after diagnosis does not increase breast cancer mortality and decreases total mortality — 16 April 2013

Newcomb PA, Kampman E, Trentham-Dietz A, Egan KM, Titus LJ, Baron JA, Hampton JM, Passarelli MN, Willett WC.  Alcohol consumption before and after breast cancer diagnosis: Associations with survival from breast cancer, cardiovascular disease, and other causes.  J Clin Oncol 2013; pre-publication. 

Available at   jco.ascopubs.org/cgi/doi/10.1200/JCO.2012.46.5765

Authors Abstract

Purpose.  Alcohol intake is associated with increased risk of breast cancer.  In contrast, the relation between alcohol consumption and breast cancer survival is less clear.

Patients and Methods.  We assessed pre- and postdiagnostic alcohol intake in a cohort of 22,890 women with incident invasive breast cancer who were residents of Wisconsin, Massachusetts, or New Hampshire and diagnosed from 1985 to 2006 at ages 20 to 79 years.  All women reported on prediagnostic intake; a subsample of 4,881 reported on postdiagnostic intake.

Results.  During a median follow-up of 11.3 years from diagnosis, 7,780 deaths occurred, including 3,484 resulting from breast cancer.  Hazard ratios (HR) and 95% CIs were estimated.  Based on a quadratic analysis, moderate alcohol consumption before diagnosis was modestly associated with disease-specific survival (compared with nondrinkers, HR = 0.93 [95% CI, 0.85 to 1.02], 0.85 [95% CI, 0.75 to 0.95], 0.88 [95% CI, 0.75 to 1.02], and 0.89 [95% CI, 0.77 to 1.04] for two or more, three to six, seven to nine, and ≥ 10 drinks/wk, respectively).  Alcohol consumption after diagnosis was not associated with disease-specific survival (compared with nondrinkers, HR = 0.88 [95% CI, 0.61 to 1.27], 0.80 [95% CI, 0.49 to 1.32], 1.01 [95% CI, 0.55 to 1.87], and 0.83 [95% CI, 0.45 to 1.54] for two or more, three to six, seven to nine, and ≥ 10 drinks/wk, respectively).  Results did not vary by beverage type.  Women consuming moderate levels of alcohol, either before or after diagnosis, experienced better cardiovascular and overall survival than nondrinkers.

Conclusion.  Overall alcohol consumption before diagnosis was not associated with disease-specific survival, but we found a suggestion favoring moderate consumption. There was no evidence for an association with postdiagnosis alcohol intake and breast cancer survival.  This study, however, does provide support for a benefit of limited alcohol intake for cardiovascular and overall survival in women with breast cancer.

Forum Comments

There are considerable data from prospective studies showing a slight increase in the risk of developing breast cancer among women who drink alcohol; the usual estimate for women who consume an average of no more than one typical drink per day is generally in the range of a 7 to 10% increase, in comparison with non-drinkers.  Data on the effects of alcohol consumption on survival of women developing breast cancer are less clear, with some studies showing decreased survival and others showing increased survival.  Given that moderate drinking reduces the risk of death from cardiovascular diseases and other diseases of ageing (that are much more common causes of death than breast cancer), it is important that research evaluate both death from breast cancer and death from other conditions, including total mortality.  The present study is based on a large cohort of American women with breast cancer, with almost 8,000 women dying during follow up.    

Specific comments from Forum reviewers:  Overall, Forum reviewers considered this to be an excellent paper, giving important new information on the relation of alcohol to breast cancer.  Reviewer Zhang stated:  “The authors did a good job to differentiate the ‘direct effect’ [i.e., alcohol consumption prior to breast cancer (BrCa) diagnosis] from the ‘total effect’ (that included changes in alcohol consumption after BrCa diagnosis] on cause-specific mortality and total mortality.  Results related to changes in consumption tend to mimic a ‘clinical trial’ of alcohol consumption on mortality among women who have been diagnosed with BrCa because it attempts to answer the question whether a change in drinking habits will affect mortality.”

Zhang continued: “Considering that characteristics of alcohol drinkers may be different from those of non-drinkers, what I am really interested in is how those who were pre-diagnosis drinkers did, or did not, change their consumption after the diagnosis.  (We assume that these ‘drinkers’ continued to have similar characteristics before and after the diagnosis, unlike comparing drinkers with non-drinkers.)  Since post-diagnosis alcohol intake is highly correlated with alcohol intake before the diagnosis of cancer, as the authors commented, it is hard to tell if these two measures of alcohol really test different hypotheses.  However, it is interesting that those women who increased alcohol drinking after diagnosis subsequently had a lower risk of cardiovascular deaths (RR=0.39) than those whose alcohol intake did not change very much (RR=0.64).  Thus, the net effect of alcohol increase vs. no change on cardiovascular mortality among drinkers is 0.39 / 0.64 = 0.61.  On the other hand, those women who decreased alcohol drinking after the diagnosis of BrCa showed a slightly increased risk of death from cardiovascular disease (0.87 / 0.64 = 1.36) than those who did not change their intake, and higher than those who increased their intake (0.87 / 0.39 = 2.23).  You can see a similar pattern for total mortality.”   

Reviewer Finkel stated: “This paper supports others showing that women with breast cancer may drink without increasing their risks of survival, and that they sustain the same cardiovascular benefits as others.  The clean methodology and large number of subjects, as well as the lack of overreaching in conclusions based on their data, makes this paper a ‘keeper.’  The accompanying editorial likewise makes very good sense.”  

Reviewer Skovenborg agreed that “the paper is well-done and adds to our knowledge on how to inform women with breast cancer regarding future alcohol consumption.  However, I have a few comments and questions:

1.         I find it strange that a substantial number of non-drinking women (1 of 5 prediagnostic non-drinkers) would increase their drinking habits after the shock of a breast cancer diagnosis — considering the widely publicized association between alcohol and breast cancer.  The authors do not comment of this behaviour.  I wonder whether the change of drinking habits among prediagnostic non-drinkers is an effect on information bias regarding alcohol consumption.

2.         The stratification of alcohol consumption levels seems to be in great detail regarding a low to moderate intake, while the 10+ drinks/week group might be a very heterogeneous group.  There is no information about the mean alcohol consumption of women in the 10+ drinks/week group.

3.         The consumption of 10+ drinks/week is characterized as heavy alcohol consumption.  For many years a consumption of not more than 14 drinks per week has been regarded as moderate alcohol consumption for women.  However, if you calculate the alcohol content of a drink as defined by the authors you get a clearer picture.  For example, a 5 oz. glass of wine is usually referred to as one ‘typical drink.’  Based on current alcohol levels in many wines, the alcohol content of 150 ml of wine with an alcohol content of, say, 13.5% by volume = 20.25 ml alcohol = 16.2 grams of alcohol; 10 drinks would equal162 grams of alcohol, which amount to 13.5 Danish drinks (12 g alcohol/drink) or 20 British drinks (8 g alcohol/drink).  The reporting of consumption levels in drinks instead of alcohol (in grams) may create confusion in view of the highly different alcohol content of a standard drink in various countries.

4.         The conclusion of the authors: ‘Women consuming moderate levels of alcohol, either before or after diagnosis, experienced better cardiovascular and overall survival than nondrinkers’ is the truth but not the whole truth.  According to data presented, the 10+ drinks women are the only women with a significant reduction of cardiovascular mortality and these women also have a significant reduced total mortality.  The whole truth conclusion might read: ‘Women consuming moderate levels of alcohol before or moderate and heavy levels of alcohol after diagnosis experienced better cardiovascular and overall survival than nondrinkers.’”  Reviewer Waterhouse stated that he could not “figure out what the high level drinkers were consuming, but it appears that ‘high’ is the most protective pattern.”

Forum Summary

Most previous research has shown a slight increase in the risk of breast cancer (BrCa) among women who consume alcohol (versus nondrinkers), even among those whose average is only one drink/day.  The association with death due to breast cancer is less clear, although overall total mortality among moderate drinkers is almost always lower than among non-drinkers; this reduction is generally attributed to a lower risk of cardiovascular disease. 

The present study related alcohol consumption, both prior to the diagnosis of BrCa and after the diagnosis, with survival; deaths from BrCa and from cardiovascular disease, as well as total mortality, were related to alcohol intake.  Specifically, in this study the authors assessed alcohol intake in a cohort of 22,890 women with incident invasive breast cancer (BrCa) who were residents of one of four US states and diagnosed at ages 20 to 79 years.  During a median follow-up period of 11.3 years after diagnosis of BrCa, 7,780 deaths occurred, including 3,484 attributed to breast cancer.  

Based on a quadratic analysis, moderate alcohol consumption before diagnosis showed a tendency towards lower risk of death from BrCa, and a significantly greater reduction in the risk of cardiovascular disease mortality and total mortality.   Alcohol consumption after diagnosis was not associated with disease-specific survival, but was associated with a lower risk of total mortality.  For women consuming alcohol prior to the diagnosis of BrCa, those who decreased their intake showed little effect on mortality, while those who increased their intake showed further lowering of their risk of cardiovascular and total mortality.

Forum reviewers considered this to be an excellent paper providing important new data on the association of alcohol consumption with survival after a diagnosis of invasive breast cancer.  They thought that the statements of the authors accurately reflected their data, as they concluded: “Overall alcohol consumption before diagnosis was not associated with disease-specific survival, but we found a suggestion favoring moderate consumption.  There was no evidence for an association with postdiagnosis alcohol intake and breast cancer survival.  This study, however, does provide support for a benefit of limited alcohol intake for cardiovascular and overall survival in women with breast cancer.” 

An accompanying editorial* agreed with the conclusions of the authors.  The editorial stated, “Based on the best available evidence, including [the present report], it appears that modest alcohol consumption after breast cancer diagnosis, up to approximately one drink per day on average, may be associated with optimal overall survival, without compromising breast cancer-specific survival.”   

*Demark-Wahnefried W.  To your health: How does the latest research on alcohol and breast cancer inform clinical practice?  jco.ascopubs.org/cgi/doi/10.1200/JCO.2013.490466),

                                                                        *                      *                      *

Comments on this critique by the International Scientific Forum on Alcohol Research were provided by the following members:

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

Jean-Marc Orgogozo, MD, Professor of Neurology and Head of the Neurology Divisions, the University Hospital of Bordeaux, Pessac, France

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

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

David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa

Andrew L. Waterhouse, PhD, Marvin Sands Professor, Department of Viticulture and Enology, University of California, Davis; Davis, CA, USA.

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

Dominique Lanzmann-Petithory,MD, PhD, Nutrition/Cardiology, Praticien Hospitalier Hôpital Emile Roux, Paris, France

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