Phytoestrogens are polyphenolic secondary plant metabolites, which have received attention because of their potentially beneficial health effects. In addition to other polyphenolic compounds, phytoestrogens have structural and functional similarities with 17β-oestradiol and can bind to the oestrogen receptor. Although these compounds have only weak oestrogenic activity (10−2–10−3 compared with 17β-oestradiol), they are present in much higher concentration and compete at the receptor complex. Their anti-oestrogenic activity has raised the possibility that these compounds might be beneficial in the prevention of hormone related diseases, such as breast and prostate cancer, as well as osteoporosis, and for the alleviation of menopausal symptoms. Several observational studies have been conducted to investigate the health effects of phytoestrogens, however, results so far have been inconclusive. Whereas studies conducted in Asian populations with high habitual phytoestrogen intake (e.g. 38 mg/d of soy isoflavones in the Shanghai Women's Health Study(Reference Zhang, Shu and Li1)) have shown a protective effect against breast cancer and a positive effect on bone density, most studies conducted in Western populations with lower habitual phytoestrogen intake failed to show an effect(Reference Ward and Kuhnle2).
Using a newly developed comprehensive food composition database(Reference Kuhnle, Dell'Aquila and Aspinall3–Reference Kuhnle, Dell'Aquila and Aspinall6), including for the first time the phytoestrogen content of foods of animal origin(Reference Kuhnle, Dell'Aquila and Aspinall3), we have investigated associations between dietary isoflavones, lignans, enterolignans and total phytoestrogens, and health in nested case control samples drawn from EPIC Norfolk, a cohort of 25 000 men and women in 1993–2008(Reference Day, Oakes and Luben7). No significant associations between phytoestrogen intake and breast cancer were found, neither with urinary biomarkers nor using dietary data(Reference Ward and Kuhnle2). Similarly, no significant associations have been observed for colorectal cancer, although in men a marginally significant positive association was observed for the lignan secoisiolariciresinol(Reference Day, Oakes and Luben7). Risk for prostate cancer was positively associated with intake of enterolignans, compounds found mainly in dairy products(Reference Kuhnle, Dell'Aquila and Aspinall3). However, this association became non-significant after adjusting for Ca as surrogate marker of dairy intake.
In post-menopausal women, the association between bone density and phytoestrogen intake was marginally significant for the non-soy isoflavone formononetin and enterolignans, however, the latter became non-significant after adjusting for Ca intake. In the lowest quintile of Ca intake (less than 570 mg/d), a marginally significant association between bone density and soy isoflavones was observed(Reference Kuhnle, Ward and Vogiatzoglou8).
These results suggest that phytoestrogens have no significant protective effect against cancer or osteoporosis in the general population with low habitual intake (less than 5 mg/d).
This research was supported by the Food Standards Agency (T05028) and the MRC.
