Hormone levels during dietary changes in premenopausal African-American women

In the United States, 5-year survival rates of 69% and 84%, respectively, have recently been reported for African-American and Caucasian women diagnosed with breast cancer. Differences in the levels of endogenous sex hormones in these populations could explain some of the variation in survival rates, since estrogen is recognized as a risk factor for this type of cancer.

Dietary factors are known to affect endogenous hormone levels; therefore, our study was designed to determine the serum hormone levels of African-American women consuming a typical North American diet, to determine the effect of a low-fat and high-fiber diet on their serum hormone levels, and to compare the base-line serum hormone levels in the African-American women with hormone data from our study of Caucasian women (n = 68) consuming the same control diet.

Twenty-one healthy, premenopausal, African-American women who agreed to eat only food prepared in a clinical study unit were recruited into the study. The control diet was similar to their usual diet, being high in fat (40% of calories from fat) and low in fiber (12 g/day), and was consumed on average for 3 weeks. The concentrations of estrone (E1), estrone sulfate (E1SO4), estradiol (E2), free E2, androstenedione, and sex hormone-binding globulin (SHBG) in serum samples obtained from the participants during the last week of the control diet and during the follicular phase of their menstrual cycle were determined. The women were then switched to a diet low in fat (20% of calories as fat) and high in fiber (40 g/day); they consumed this diet for two menstrual cycles before blood samples were collected for determination of serum hormone levels. Repeated-measures regression modeling was used to investigate the relationship between diet and hormone levels in African-American and Caucasian women. All P values resulted from two-sided statistical tests.

Analysis of serum hormone levels in the African-American women indicated that the change in diet caused a significant decrease in E2 (-8.5%; 95% confidence interval [CI] = -16.1% to -0.3%; P < or = .03) and E1SO4 (-16.2%; 95% CI = -22.1% to -9.8%; P < .0001) and a significant increase in androstenedione levels (+18.3%; 95% CI = +10.3% to +26.8%; P < .0001). SHBG levels of the African-American women were 5.6% (95% CI = -14.0% to +3.7%) lower for those on the experimental diet compared with those on the control diet, but the difference was not statistically significant. Comparison of control serum hormone values in the African-American women in this study with those in Caucasian women previously studied indicated that the Caucasian women had statistically significant lower levels of E1 (-37%; 95% CI = -61.2% to -16.4%; P < or = .0002), E2 (-54.5%; 95% CI = -90.9% to -25.1%; P < or = .0001), free E2 (-30.2%; 95% CI = -65.7% to -2.3%; P < .03), and androstenedione (-48.3%; 95% CI = -83.7% to -19.7%; P < or = .0004).

African-American women appear to have higher levels of serum hormones than Caucasian women, and dietary modification can result in a lowering of serum estrogens.