Critique 144: Critique 144: Alcohol consumption and risk of endometrial cancer — 18 August 2014

Je Y, DeVivo I, Giovannucci E.  Long-term alcohol intake and risk of endometrial cancer in the Nurses’ Health Study, 1980–2010.  British Journal of Cancer 2014;111:186–194; doi: 10.1038/bjc.2014.257.

Authors’ Abstract

Background: Previous epidemiologic studies have shown inconsistent results for the association between alcohol intake and endometrial cancer risk.  Most of the studies, however, assessed alcohol intake after cancer diagnosis, or measured alcohol intake at baseline only.

Methods: We prospectively examined the association between alcohol intake and endometrial cancer risk in the Nurses’ Health Study with 68,067 female participants aged 34–59 years in 1980.  Alcohol intake was measured several times with validated dietary questionnaires.  We calculated cumulative average alcohol intake to represent long-term intakes of individual subjects.  Using Cox proportional hazards models, we estimated incidence rate ratios (RRs) and 95% confidence intervals (CIs) for endometrial cancer risk after controlling for several risk factors simultaneously.

Results:  We identified a total of 794 invasive endometrial adenocarcinoma from 1980 to 2010.  We found an inverse association among alcohol drinkers (multivariable RR=0.81; 95% CI: 0.68–0.96) compared with nondrinkers.  Women with light alcohol intake of <5 g per day (one-half drink per day) had a 22% lower risk of endometrial cancer (multivariable RR=0.78; 95% CI: 0.66–0.94).  Higher intake of alcohol, however, did not provide additional benefits against endometrial cancer: multivariable RRs for 5–14.9 g (~ 1 drink), 15–29.9 g (~ 2 drinks), or ≥ 30 g (≥ 2 drinks) versus 0 g per day were 0.88, 0.83, and 0.78 (95% CI: 0.49–1.25), respectively.  The lower risk among drinkers (~ half drink per day) appeared to be stronger for obese women, but no significant interaction by body mass index was found.

Conclusions: This study provides prospective evidence for an inverse association between light alcohol intake (~ half drink per day) in the long term and endometrial cancer risk, but above that level no significant association was found.

Forum Comments

A recent meta-analysis by Friberg et al, studying the relation of alcohol consumption to the risk of endometrial cancer, demonstrated a J-shaped curve, with a significantly reduced risk of cancer for light drinkers and an increase in risk for heavier drinkers.  Another meta-analysis by Sun et al showed an increase in risk for spirits but no effect on risk of uterine cancer for consumers of beer or wine. 

The present well-done analysis is of importance because it is based on the very large Nurses’ Health Study, with data collected prospectively (rather than after the diagnosis of cancer).  With repeated assessments of data on alcohol consumption, it was possible for the investigators to estimate long-term average alcohol intake.  Further, there were a large number of cases (n=794) of invasive uterine cancer. 

The study showed that no category of drinkers had a relative risk as high as that of abstainers, and there was a statistically significant approximately 20% reduction in risk of uterine cancer for consumers of an average of no more than one-half drink per day (with similar risk ratios for the other categories).  While the estimated risk for the highest drinking category remained below 1.0, there were few heavy drinkers in the cohort, and a possible increase in risk for greater alcohol intake could not be adequately assessed.  The RR for wine drinkers was slightly lower than that of other drinkers, but there were no significant differences according to type of beverage consumed.   

Forum member Van Velden commented: “This a well done study that supports the potential protective effects of moderate alcohol consumption.  Of special importance, is the fact that the researchers identified the association of alcohol and the metabolic syndrome with insulin resistance.  Alcohol intake has been associated with reduced fasting insulin concentrations and improved insulin sensitivity and thus may lower the risk of endometrial cancer by relieving chronic hyperinsulinaemia.  Taking into consideration the inverse association between alcohol consumption and obesity, and the inverse association of alcohol with type 2 diabetes, this finding is of great relevance in view of the epidemic of obesity.  Insulin can stimulate the growth of endometrial cells, increasing the risk of endometrial cancer.  This may explain the mechanism of protection of moderate alcohol consumption and uterine cancer.”  Forum member de Gaetano agreed and added: “The relevance of the data to the obesity epidemic is of particular interest.”

Reviewer Gretkowski agreed that this was an excellent analysis; she  pointed out that “Most of these women were in the overweight BMI range, confounding this study.  Breakdown of dietary constituents in this group (Mediterranean-type diet?) should be considered due to issues related to chronic hyperinsulinemia.”  Reviewer Finkel agreed that this is a well-done study.  He did wonder whether the small differences in effect between wine and spirits drinkers suggested by the data may be of importance.  Could other (non-alcohol) components of wine play a role in this association? 

Forum reviewer Skovenborg stated: “I agree with other reviewers that this is a well-done study with important and biologically plausible results.”  He added, “A question worthy of consideration: The inverse associations were attenuated after multivariable adjustment – and BMI was a primary confounder.  In a previous Nurses’ Health Study, Colditz et al demonstrated a clear inverse relation between moderate alcohol consumption and body mass index; a similar association has been described by Arif et al and by many others.  A causal association between moderate alcohol consumption and reduced BMI in women might be a part of the biological mechanism of effect, and not a real confounder.”

References for Forum review

Arif, Ahmed A., and Rohrer, James E.  Patterns of alcohol drinking and its association with obesity: Data from the Third National Health and Nutrition Survey, 1988-1994. BMC Public Health, 2005;5:126.

Colditz GA, Giovannucci E, Rimm EB, et al.  Alcohol intake in relation to diet and obesity in women and men.  Am J Clin Nutr 1991;54:49-55

Friberg E, Orsini N, Mantzoros CS, Wolk A.  Alcohol intake and endometrial cancer risk: a meta-analysis of prospective studies. Br J Cancer 2010;103:127–131.

Sun Q, Xu L, Bo Zhou, Wang Y, Jing Y, Wang B (2011) Alcohol consumption and the risk of endometrial cancer: a meta-analysis. Asia Pac J Clin Nutr 2011;20:125–133.

Forum Summary

Most epidemiologic studies have shown that moderate alcohol consumption does not increase the risk of uterine cancer, and some have suggested an inverse or J-shaped relation.  In a new analysis from the Nurses’ Health Study, with 68,067 female participants aged 34–59 years in 1980, investigators have related repeatedly assessed long-term alcohol intake, and related the cumulative average intake over time to the risk of invasive uterine cancer.  A total of 794 cases of invasive endometrial adenocarcinoma were identified over a 30 year follow-up period. 

The authors report an inverse association among alcohol drinkers (multivariable RR=0.81; 95% CI: 0.68–0.96) compared with nondrinkers.  Women with an intake of <5 g per day (an average of approximately one-half drink per day) had a 22% lower risk of endometrial cancer (multivariable RR=0.78; 95% CI: 0.66–0.94), with no further decrease evident from larger amounts of alcohol.  In comparison with non-drinkers, the relative risk was similarly reduced for consumers of more than 30 grams of alcohol per day, but the number of subjects in this category was small, and a potential increase in risk from heavy drinking could not be adequately assessed in this study.

Forum members considered this to be a very well-done prospective study, with clear and biologically plausible results.  By having repeated assessments of alcohol, an estimate of long-term average consumption was possible.  The approximately 20% lower risk of uterine cancer among subjects with light alcohol intake remained statistically significant after multivariate adjustments for known potential confounders.  Reviewers agreed with the authors that a potential mechanism for the association could be a reduction in insulin concentrations and improved insulin sensitivity that have been shown to occur with moderate alcohol consumption.  Further, the frequently demonstrated inverse relation for alcohol with obesity may also play a mechanistic role.  Overall, this study adds to accumulating scientific data showing that moderate drinking does not increase the risk of uterine cancer, and probably is associated with a reduction in risk

                                 *                      *                      *

Comments on this paper have been provided by the following members of the International Scientific Forum on Alcohol Research:

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

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

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

Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy

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

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

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