Saito · Journal of nutritional science 2025 · review of registered clinical trials · n=84 trials

Sulforaphane as a potential therapeutic agent: a comprehensive analysis of clinical trials and mechanistic insights.

Cited 18 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review/registry analysis of registered clinical trials

PubMed 40988712 · doi:10.1017/jns.2025.10033 · record verified 2026-08-26

What was done

Review of clinical trials registered on ClinicalTrials.gov evaluating sulforaphane (SFN) or broccoli-derived extracts across diverse clinical indications. The authors identified 84 registered trials (39 published) to assess clinical efficacy, underlying mechanisms (e.g., Keap1/Nrf2 pathway and histone deacetylase inhibition), and trial publication status.

What was found

Out of 84 identified trials, 39 had published results. In healthy volunteers, SFN enhanced detoxification and reduced inflammation. In oncology, SFN showed promise in early-stage prostate and breast cancer (particularly in GSTM1-positive individuals) but had limited impact in advanced cancer. For neuropsychiatric conditions, symptomatic improvements were observed in autism spectrum disorder and cognitive improvements in schizophrenia, but biomarker integration was absent. For metabolic diseases, glycemic control improved in type 2 diabetes, but no benefit was found for hypertension. SFN had minimal impact on respiratory disease except for supportive effects in allergic rhinitis. Approximately 50% of completed trials remain unpublished. The abstract provides no specific quantitative effect sizes or confidence intervals.

Why it matters

This review maps the current translational evidence for sulforaphane across multiple clinical domains while highlighting that widespread non-publication of completed trials hampers clear clinical recommendations.

Limits

No numerical effect estimates, confidence intervals, or meta-analytic statistics are reported in the abstract. Substantial publication bias exists, with ~50% of completed trials unpublished. Published studies are constrained by small sample sizes and inconsistent outcome measures.

Cited by