Flavonoid-Rich Diet Tied to Lower Breast Cancer Risk

Women with the highest intake of flavonoid-rich foods had a 15% lower risk of developing breast cancer than those with the lowest intake, according to a prospective cohort study of more than 93,000 women published in npj Breast Cancer.¹ The association was strongest among women with high genetic susceptibility to the disease, a finding the authors say supports incorporating dietary counseling into risk-reduction strategies regardless of a patient's inherited risk profile.

The study, led by Rui Yang, MD, and colleagues at China Medical University in Shenyang, China, analyzed data from 93,271 women in the UK Biobank who completed at least one 24-hour dietary assessment (Oxford WebQ) and had genotyping data available. Over a median follow-up of 11.8 years, 3110 incident breast cancer cases were recorded.

Investigators quantified habitual intake of flavonoid-rich foods using the Flavodiet Score (FDS), a composite measure derived from reported daily servings of tea, red wine, apples, berries, grapes, oranges, grapefruit, sweet peppers, onions, and dark chocolate. Participants were also stratified by a polygenic risk score (PRS) built from 168 breast cancer-associated single nucleotide polymorphisms identified in prior genome-wide association studies.

Key Findings

In adjusted models, women in the highest FDS quintile had a HR of 0.85 (95% CI, 0.75-0.95) for breast cancer compared with those in the lowest quintile, with a statistically significant linear trend (P =.031). Among individual flavonoid subclasses, higher intake of proanthocyanidins was associated with the most robust inverse association (HR, 0.81; 95% CI, 0.72-0.92; P for trend =.002), while flavanones showed a similar but less consistent pattern (HR, 0.87; 95% CI, 0.78-0.98). No significant associations were observed for total flavonoid intake or several other subclasses after full covariate adjustment.

Among individual foods, high apple intake (HR, 0.90; 95% CI, 0.82-0.99) and high orange intake (HR, 0.83; 95% CI, 0.75-0.92) were each associated with reduced risk, whereas tea and onion intake showed no significant association.

Genetic Risk Modification

As expected, PRS was strongly associated with breast cancer risk overall, with women in the highest PRS quintile facing more than double the risk of those in the lowest quintile (HR, 2.71; 95% CI, 2.40-3.06). When stratified by genetic risk category, the protective association with FDS reached statistical significance only among women with high genetic risk, where those in the top FDS quintile had a 29% lower risk than those in the bottom quintile (HR, 0.71; 95% CI, 0.58-0.87; P for trend =.005). No significant multiplicative interaction between FDS and PRS was detected (P >.05), though additive interaction measures suggested some synergy in select strata.

Limitations and Future Directions

The authors noted several limitations, including the observational design, which precludes causal inference; potential recall and reporting bias inherent to 24-hour dietary questionnaires; and a study population restricted to participants of white British ancestry, which may limit generalizability. Sensitivity analyses excluding early cases (to address reverse causality), removing red wine from the score (to address alcohol-reporting bias), and using multiple imputation for missing covariates produced consistent results.

The authors proposed several biological mechanisms that could underlie the association, including flavonoid-mediated reduction of oxidative stress and inflammatory signaling, and interaction of certain flavonoids with estrogen receptor-alpha pathways. They concluded that adherence to a flavonoid-rich dietary pattern may offer a modifiable avenue for breast cancer risk reduction, particularly worth exploring in future interventional studies, including among women with elevated genetic risk.

REFERENCE

1. Yang R, Xu YL, Wang Y, et al. Flavonoid-rich foods, genetic risk, and female breast cancer risk: a prospective cohort study. npj Breast Cancer. Published online 2026. doi:10.1038/s41523-026-00989-y