The EPIC-Norfolk Study
Dietary phytoestrogens and breast cancer risk.
The American Journal of Clinical Nutrition 2004 ; 79: 282-8.
Keinan-Boker L, van der Schouw YT, Grobbee DE, and Peeters PH
DOI : 10.1093/ajcn/79.2.282
PubMed ID : 14749235
PMCID :
URL : https://academic.oup.com/ajcn/article/79/2/282/4690093
Abstract
A high intake of phytoestrogens, particularly isoflavones, has been suggested to decrease breast cancer risk. Results from human studies are inconclusive.
We investigated the association between phytoestrogen intake and breast cancer risk in a large prospective study in a Dutch population with a habitually low phytoestrogen intake.
The study population consisted of 15 555 women aged 49-70 y who constituted a Dutch cohort of the European Prospective Investigation into Cancer and Nutriton (EPIC; 1993-1997). Data concerning habitual dietary intake in the preceding year were obtained by using a validated food-frequency questionnaire. The content of isoflavones and lignans in relevant food items was estimated through a literature search, the use of food-composition tables, and contact with experts. Newly diagnosed breast cancer cases up to 1 January 2001 were identified through linkage with the Comprehensive Cancer Center Middle Netherlands. Hazard ratios for the disease were estimated by Cox proportional hazard analysis for quartiles of isoflavone and lignan intake. Associations were adjusted for known breast cancer risk factors and daily energy intake.
A total of 280 women were newly diagnosed with breast cancer during follow-up. The median daily intakes of isoflavones and lignans were 0.4 (interquartile range: 0.3-0.5) and 0.7 (0.5-0.8) mg/d, respectively. Relative to the respective lowest intake quartiles, the hazard ratios for the highest intake quartiles for isoflavones and lignans were 1.0 (95% CI: 0.7, 1.5) and 0.7 (0.5, 1.1), respectively. Tests for trend were nonsignificant.
In Western populations, a high intake of isoflavones or mammalian lignans is not significantly related to breast cancer risk.