Risk of endometrial cancer in relation to individual nutrients from diet and supplements.

OBJECTIVE Intake of nutrients may influence the risk of endometrial cancer (EC). We aimed to estimate the association of intake of individual nutrients from food and from food plus supplements with EC occurrence. DESIGN A population-based case-control study conducted in Canada (2002-2006). SETTING Nutrient intakes from food and supplements were assessed using an FFQ. Logistic regression was used to estimate EC risk within quartile levels of nutrient intakes. SUBJECTS Incident EC cases (n 506) were identified from the Alberta Cancer Registry, and population controls were frequency- and age-matched to cases (n 981). RESULTS There existed little evidence of an association with EC for the majority of macronutrients and micronutrients examined. We observed a statistically significant increased risk associated with the highest, compared with the lowest, quartile of intake of dietary cholesterol (multivariable-adjusted OR = 1·51, 95 % CI 1·08, 2·11; P for trend = 0·02). Age-adjusted risk at the highest level of intake was significantly reduced for Ca from food sources (OR = 0·73, 95 % CI 0·54, 0·99) but was attenuated in the multivariable model (OR = 0·82, 95 % CI 0·59, 1·13). When intake from supplements was included in Ca intake, risk was significantly reduced by 28 % with higher Ca (multivariable-adjusted OR = 0·72, 95 % CI 0·51, 0·99, P for trend = 0·04). We also observed unexpected increased risks at limited levels of intakes of dietary soluble fibre, vitamin C, thiamin, vitamin B6 and lutein/zeaxanthin, with no evidence for linear trend. CONCLUSIONS The results of our study suggest a positive association between dietary cholesterol and EC risk and an inverse association with Ca intake from food sources and from food plus supplements.