Soy & Tamoxifen for Breast Cancer Prevention in High-Risk Premenopausal Women

Grant #: Research Grant
PI Name: Tice J
Title: Soy and Tamoxifen for Breast Cancer Prevention in High-Risk Premenopausal Women
Institute: Department of Defense

Background: The use of soy products in women at high risk for breast cancer is controversial.

Purpose: To conduct a feasibility study to assess the efficacy and safety of dietary soy for breast cancer prevention in premenopausal women at elevated risk of breast cancer. Mammographic breast density, a potential surrogate marker for breast cancer risk, was used as the primary entry criterion and the primary outcome.

Methods: The PREVENT is a randomized, placebo-controlled study of 40 pre-menopausal women with breast density ≥ 50% on mammography. Women were randomized to either 25 mg/d of soy protein containing 50 mg total isoflavones or 25 mg/day of milk protein containing 0 mg of total isoflavones for 6 months. The primary outcome measure is the change in percent breast density on mammography timed to the menstrual cycle. Density is being measured by using a computer-based threshold method to identify the percentage of total breast area that is dense on screening mammography. Many of the women also had novel measures of breast density including volumetric density, parenchymal complexity, and MRI. Additional outcome measures include analyses of nipple aspirate fluid for cytology, estrogen metabolites, and serum IGF-1 and IGFBP-3. Feasibility was assessed by evaluating patient enrollment, compliance, and satisfaction.

Results: Initial recruitment for the study was hampered by the inclusion of a tamoxifen arm as a positive control. After eliminating the tamoxifen arm, the investigators randomized 49 women. The final close-out visit was scheduled for January 2005. At randomization, the average 5-year Gail risk was 2.0% and the average breast density was 74% (range 59% - 90%). Patients reported some difficulties consuming the study protein. Among the 7 women who dropped out after randomization, 2 found the protein powder intolerable and one was concerned about weight gain. However, compliance by packet count was 88% among women completing the 6 month visit. The only side effects reported by more than 2 women were stomach upset (8%) and constipation (14%). There were no serious adverse events. Among the 25 women with 6 month follow-up data at the time of this analysis, there was no net change in weight (-0.4 kg) and the average change in breast density was -1.1% (range -13% to +14%). Change scores measured using volumetric density and MRI had less variability.

Conclusions: Recruitment for a prevention study that included a tamoxifen arm was difficult but targeting recruitment efforts to women with a history of dense mammograms was successful. The results of this study lay the foundation for a larger, definitive trial of soy by identifying impediments to recruitment and retention and providing estimates for sample size calculations based on change in breast density.