Van Patten, C.L. (2002). Effect of soy phytoestrogens on hot flashes in postmenopausal women with breast cancer: A randomized, controlled clinical trial. Journal of Clinical Oncology, 20, 1449–1455.

The study examined the effect of soy phytoestrogens on hot flashes in postmenopausal women with breast cancer:

The study randomized 157 participants with a mean age of 55 years and who had been previously treated for breast cancer from August 1998 to February 2000. Nine women (6%) became ineligible after randomization, and 25 (16%) dropped out because of time commitment (n = 9), intolerance of the study beverage (n = 10, 7 in the soy group, 3 in the placebo group), or other reasons (n = 6). The remaining 123 women completed the study by June 2000.

•  Inclusion criteria:
  • Women had completed treatment for breast cancer more than four months prior to enrollment (tamoxifen use was allowed), were menopausal (12 or more months of amenorrhea), and had not used HRT for 4 or more months, were experiencing troubling hot flashes, defined as a score (frequency × intensity) of 10 or more per week.
  • Women using complementary therapies and prescription medications, including tamoxifen, were eligible if no change in therapy for four months or longer.
  • All participants were instructed to avoid soy-based foods and soy supplements during the study.
• Exclusion criteria:
  • Based primarily on factors that modify estrogen or phytoestrogen metabolism or that had the potential to require medical intervention: smokers, using antibiotics, inflammatory bowel disease, liver impairment (gamma-glutamyl transferase and alkaline phosphatase of ≥ 1.5 times normal), or recurrent breast cancer.
  • Also, soy allergy or regular consumption of soy foods.

In this randomized, placebo-controlled, double-blind clinical trial, participants were stratified for tamoxifen use and randomized to a soy beverage (n = 59) containing 90 mg of isoflavones or to a placebo rice beverage (n= 64).

Women recorded the number/severity of hot flashes with a daily diary for 4 weeks at baseline, then for 12 weeks while consuming 500 ml of a soy or placebo beverage daily. The primary outcome variable was the mean 24-hour hot flash score, created by summing the hot flash score (frequency × intensity) during the day and night. The main analysis, with Student’s t test, was a comparison between groups in the change in the mean 24-hour hot flash score during the 4 weeks of baseline compared with the last 4 weeks of treatment. This analysis was also conducted for the hot flash number and score during the day and night and the hot flash number per 24 hours. Secondary analyses included a comparison between groups of:

1. Consumption and acceptability ratings for each beverage
2. Frequency of side effects
3. Responses to the study exit questionnaire
4. Serum isoflavone concentrations

The average serum isoflavone concentration of the soy beverage was also calculated. All statistical tests were two-tailed and used a significance level of alpha = 0.05.

This trial does not support the use of a soy beverage containing phytoestrogens as a treatment for hot flashes in breast cancer survivors. The soy beverage did not alleviate hot flashes in women with breast cancer any more than did a placebo. Mild gastrointestinal side effects were experienced by both groups but occurred with greater frequency and severity with soy.