Although higher circulating levels of oestrogen are related to postmenopausal breast cancer risk, limited information is available regarding effects of diet on endogenous oestrogen. Thus, we examined associations between macronutrient intakes and serum oestrogen with consideration of polymorphisms in oestrogen-metabolising genes. In this cross-sectional study, 784 naturally menopaused Japanese women aged 47–69 years were selected from participants of the Japan Multi-Institutional Collaborative Cohort Study. We documented dietary intakes, measured serum concentrations of oestrone (E1) and oestradiol (E2) and genotyped polymorphisms in oestrogen-metabolising CYP19A1 (rs4441215 and rs936306) and HSD17B1 (rs605059) genes. Trends and interactions were examined using linear regression models. In addition, we calculated the ratios of the oestrogen concentrations of the second to the highest quartiles (Q2–Q4) of dietary intake to those of the lowest quartiles (Q1). After adjustment for potential confounders, E2 was significantly associated with intake of carbohydrate and noodles; ratios of Q4 v. Q1 were 1·15 (95 % CI 1·04, 1·28) and 1·15 (95 % CI 1·04, 1·26), respectively. In contrast, E2 levels were inversely associated with intake of total energy, SFA and n-3 highly unsaturated fatty acids (n-3 HUFA); ratios of Q4 v. Q1 were 0·90 (95 % CI 0·82, 0·99), 0·89 (95 % CI 0·81, 0·98) and 0·91 (95 % CI 0·83, 1·00), respectively. In stratified analysis by polymorphisms, the rs605059 genotype of HSD17B1 significantly modified associations of E2 with intake of n-3 HUFA and fish; the associations were limited to those with the CC genotype. Macronutrient intakes were associated with serum E2 level, and these associations may be modified by HSD17B1 polymorphism in postmenopausal women.

Hydroxysteroid (17-beta)dehydrogenase 1(HSD17B1) plays a central role in sex steroid hormone metabolism. HSD17B1 polymorphic variants may contribute to cancer susceptibility. Numerous investigations have been conducted to assess the association between HSD17B1 Ser312Gly polymorphism and cancer risk in multiple ethnicities, yet these have produced inconsistent results. We therefore performed this comprehensive meta-analysis to attempt to provide a quality assessment of the association of interest. Odds ratios (ORs) with 95% confidence intervals (CIs) were used to evaluate the strength of associations. After a systematic literature search of several major public databases, 20 studies involving 29,460 cases and 36,687 controls were included in this meta-analysis. No significant association was found between HSD17B1 Ser312Gly polymorphism and cancer risk. However, Ser312Gly polymorphism showed a significantly decreased risk for Caucasians (there were 44,284 Caucasians for analysis, comprising 19,889 cases and 24,395 controls) in the subgroup analysis by ethnicity (dominant: OR = 0.958, 95% CI = 0.919–0.998; and allele comparing: OR = 0.973, 95% CI = 0.947–0.999). And there was the same trend towards risk in the population-based (PB) controls (homozygous: OR = 0.951, 95% CI = 0.908–0.997 and allele comparing: OR = 0.976, 95% CI = 0.954–0.999), but not among Asians or hospital-based (HB) controls. In addition, no association was observed in the stratified analysis for breast cancer studies by source of control, ethnicity and quality score. These findings suggested that the HSD17B1 Ser312Gly polymorphism might confer genetic cancer susceptibility in an ethnic-dependent manner, especially among Caucasians. Well-designed, large-scale studies are warranted to validate these findings.