Effect of soymilk consumption on serum estrogen concentrations in premenopausal Japanese women

BACKGROUND: Estrogens have been implicated in the development of breast cancer. Preliminary evidence suggests that consumption of soy products, which contain isoflavones (phytoestrogens), can reduce serum estrogen levels. Our purpose was to determine the effect of soy consumption on serum estrogen levels in premenopausal women by use of a dietary intervention approach. METHODS: Premenopausal Japanese women were randomly assigned to receive either a soymilk-supplemented diet (n = 31) or a normal (control) diet (n = 29). The women in the soymilk-supplemented group were asked to consume about 400 mL of soymilk (containing about 109 mg of isoflavones) daily during a study period that involved three consecutive menstrual cycles. Follicular-phase blood samples were to be obtained in the menstrual cycles preceding (cycle 1) and following (cycle 3) the 2-month dietary intervention. All statistical tests were two-sided. RESULTS: At the end of the study period, estrone and estradiol levels were decreased by 23% and 27%, respectively, in the soymilk-supplemented group and were increased by 0.6% and 4%, respectively, in the control group. The changes for each hormone between the two groups were not statistically significantly different. In the soymilk-supplemented group, menstrual cycle length was increased by nearly 2 days, and, in the control group, it was decreased by approximately 1 day, a difference that was not statistically significant. A subgroup analysis restricted to subjects who provided follicular-phase blood samples on the same day or 1 day apart in menstrual cycles 1 and 3 showed a reduction in serum estrone levels in the soymilk-supplemented group that was of borderline statistical significance (P = .07 for change in serum estrone level in soymilk-supplemented group versus control group). CONCLUSION: Much larger studies will be required to confirm the ability of soy products to reduce serum estrogen levels.