Evidence Explorer / MS-N020

Evidence at a glance

• Population: Breast-cancer survivors in large prospective cohorts.
• Exposure: Tofu, soy milk, edamame and other soy foods—not concentrated isoflavone capsules.
• Finding: Higher postdiagnosis soy-food intake was associated with lower recurrence and mortality, including among tamoxifen users.
• Boundary: No randomized trial shows that starting soy prevents recurrence or treats breast cancer.

What the human evidence found

The Shanghai Breast Cancer Survival Study followed 5,042 women. Comparing the highest with lowest quartile of soy-protein intake, adjusted hazard ratios were 0.71 (95% CI 0.54–0.92) for total mortality and 0.68 (95% CI 0.54–0.87) for recurrence. Four-year mortality was 7.4% versus 10.3%, and recurrence 8.0% versus 11.2%.

A pooled US–China analysis of 9,514 survivors found postdiagnosis intake of at least 10 mg/day of soy isoflavones associated with lower recurrence; breast-cancer-specific mortality was directionally favorable but not statistically definitive. Meta-analyses of cohorts are broadly reassuring.

These data are observational. People eating more soy may differ in activity, body weight, other foods, treatment adherence or access to care. Adjustment cannot remove all confounding.

Limits and what this cannot establish

Food studies cannot validate high-dose isoflavone pills. Laboratory estrogen-receptor effects do not override human outcome data, but neither do favorable cohort associations prove treatment benefit.

This record links the fuller Missing Script review rather than republishing it.

Safety and practical context

Soy is a common allergen. During active weight loss or gastrointestinal symptoms, any diet change should preserve adequate calories and protein. Concentrated extracts deserve medication review during endocrine therapy because their exposure differs from food.

The Shanghai study reported NIH support; one author had received a small research-development fund from the United Soybean Board, which the paper states had no role in study design or interpretation.

Questions for the care team

• Are whole soy foods reasonable within my current nutrition plan?
• Is this product ordinary soy food/protein or a concentrated isoflavone extract?
• Could a major diet change compromise calories, protein or treatment tolerance?

Sources checked

• Related Missing Script review
• Shanghai Breast Cancer Survival Study (PMID 19996398)
• Pooled US–China cohort analysis (PMID 22648714)
• Postdiagnosis soy-food meta-analysis (PMID 23725149)

Appraisal date: October 14, 2026. Last literature search: October 14, 2026.

Educational evidence summary, not medical advice. Research regimens describe what was studied; they are not personal prescribing instructions and should not replace cancer treatment or individualized supportive care.