Our findings suggest that alcohol consumption is weakly associated with endometrial cancer risk. Dose–response analyses showed a non-linear association between risk and number of drinks of alcohol per day: consumption of up to 13
g of alcohol per day (one drink) seemed to be weakly protective, whereas exposure to more than two drinks (>26
g of alcohol) per day may increase risk.
Our analysis must be interpreted in the context of the limited available data. Some degree of non-differential misclassification of alcohol exposure is probable, but this would be expected to attenuate the true relationship. As in all meta-analyses, the possibility of publication bias is of concern, but was not suggested by a formal statistical test. Our meta-analysis has several strengths. First, it was based on prospective studies, which are less susceptible to bias (e.g., recall and selection bias). Second, the dose–response analyses included a wide range of alcohol intake.
A relationship between alcohol and risk of endometrial cancer is biologically plausible. Alcohol increases oestrogen levels (Gavaler and Van Thiel, 1992
; Hankinson et al, 1995
; Onland-Moret et al, 2005
; Rinaldi et al, 2006
), which in turn have been shown to increase risk by stimulating the proliferation of endometrial cells (Graham and Clarke, 1997
). The EPIC (European Prospective Investigation into Cancer and Nutrition), which is the largest study on alcohol consumption and sex-steroid concentrations, observed a statistically significant increase in blood oestrone levels among women consuming on an average approximately two drinks per day or more compared with non-drinkers (Rinaldi et al, 2006
). Furthermore, an intake of 30
g of alcohol per day has been shown to improve insulin sensitivity and reduce fasting insulin concentrations (Davies et al, 2002
), thereby potentially decreasing endometrial cancer risk, although higher intakes do not seem to have these effects (Carlsson et al, 2005
). Insulin has been shown to stimulate the growth of endometrial stromal cells by binding to insulin receptors in the endometrium (Nagamani and Stuart, 1998
). Hyperinsulinaemia may also increase levels of free oestrogen through decreasing concentrations of circulating sex hormone-binding globulin (Nestler et al, 1991
; Kazer, 1995
). Finally, hyperinsulinaemia, through decreasing levels of IGFBP-1, increases circulating free IGF-1, which, by binding and activating IGF-1 receptors in the endometrium, stimulates cell proliferation (Corocleanu, 1993
; Irwin et al, 1993
; Ordener et al, 1993
; Murphy, 1994
; Thiet et al, 1994
; Weiderpass et al, 2003
). However, we cannot rule out the possibility that part of the lowered RR observed among women drinking up to one drink per day is because the reference category includes former drinkers and women with health problems.
Our results have important public health implications, given the large number of women consuming alcohol, and the increasing incidence of endometrial cancer in Western societies. The results from this meta-analysis suggest a J-shaped association between alcohol intake and endometrial cancer risk. Moderate alcohol consumption might protect against endometrial cancer, whereas high alcohol consumption may increase risk.