Low-fat diet lowers dietary glycation compounds for years in postmenopausal women

By Hugo Francisco de SouzaReviewed by Susha Cheriyedath, M.Sc.Jul 29 2026

Analysis of the long-running WHI trial offers new clues about how everyday eating patterns may reduce estimated intake of compounds linked to inflammation and metabolic dysfunction.

Study: Low-fat dietary pattern and dietary advanced glycation end-products intake: a secondary analysis of the Women’s Health Initiative randomized trial. Image Credit: Elena Eryomenko / Shutterstock

In a recent study published in the British Journal of Cancer, researchers evaluated whether a long-term low-fat dietary intervention could reduce estimated intake of dietary advanced glycation end-products (dAGEs) in postmenopausal women. The analysis included 40,209 participants with dietary data from the Women’s Health Initiative (WHI) Dietary Modification randomized trial, which ran for approximately 7 years.

Study findings revealed that assignment to a low-fat, plant-forward dietary pattern was associated with substantial reductions in estimated dAGE intake by Year 1 that persisted throughout follow-up, compared with a usual-diet control group. These findings suggest that broad dietary-pattern changes may reduce estimated dAGE intake without requiring specific cooking instructions.

Background

Over decades of clinical and epidemiological research, advanced glycation end-products (AGEs) have been associated with systemic metabolic dysfunction, chronic inflammation, weight gain, and several cancers, including breast cancer. AGEs are reactive metabolites formed endogenously or consumed in the diet, particularly through high-fat and processed foods and foods prepared by dry-heat methods.

In long-term follow-up of the Women’s Health Initiative (WHI) Dietary Modification (DM) randomized trial, postmenopausal women assigned to a low-fat dietary intervention had a statistically significant reduction in deaths from breast cancer over 20 years of follow-up (hazard ratio [HR] 0.79, 95% confidence interval [CI] 0.64–0.97, P = 0.02).

Subsequent exploratory analyses examined whether baseline metabolic syndrome burden modified this mortality finding. Among intervention participants, those with three or four metabolic syndrome components at baseline had 69% lower breast cancer mortality than those with no components (HR 0.31, 95% CI 0.14–0.69; interaction P = 0.01). However, this finding did not establish that improvements in metabolic health mediated the dietary intervention’s effect.

Notably, because animal fats, red and processed meats, baked or fried foods, and high-fat spreads are major dietary contributors to dAGE intake, the authors of the present study hypothesized that the WHI intervention's emphasis on lowering total fat while increasing fruits, vegetables, and grains may have naturally reduced participants’ dAGE intake.

About the study

The present study aimed to examine whether a broad-scale low-fat dietary pattern was associated with sustained reductions in dAGE intake among postmenopausal women under free-living conditions by assessing the longitudinal effects of the WHI low-fat dietary pattern intervention on dAGE intake over 7 years.

The secondary analysis was conducted on data obtained from 40,209 postmenopausal women (ages 50–79 years). These participants were originally randomized in a 40:60 ratio to either the low-fat dietary pattern intervention (n = 16,198) or a usual-diet comparison group (n = 24,011).

The dietary intervention targeted a reduction in total fat intake to 20% of total daily energy alongside increased daily consumption of fruits, vegetables, and grains, delivered through 18 dietitian- and nutritionist-led group sessions during the first year and quarterly maintenance sessions thereafter. Dietary intake was assessed using self-reported Food Frequency Questionnaires (FFQs) at baseline, Year 1, and in rotating annual subgroups through Year 7.

Individual dAGE scores were estimated using a commonly referenced Nε-carboxymethyl-lysine (CML)-AGE database containing values for 540 foods, energy-adjusted using nutrient density methods, and expressed as kilo Units per 1,000 kilocalories (kU/1000 kcal).

Statistical evaluations utilized generalized linear models (GLMs) with repeated measures to assess differences between the trial arms, changes across study visits, and whether changes over time differed between the intervention and comparison groups.

Study findings

Baseline mean dAGE scores were nearly identical between trial arms prior to intervention (7,542 kU/1000 kcal in the intervention group and 7,514 kU/1000 kcal in the comparison group; P = 0.15).

However, by the Year 1 follow-up, participants in the low-fat intervention group had a sharp decline in model-estimated mean dAGE score (5,362 kU/1000 kcal), whereas the usual-diet comparison group had a higher model-estimated mean score (7,159 kU/1000 kcal; β = -1,796.96 kU/1000 kcal, P < 0.0001).

This substantial reduction persisted, although the difference between groups narrowed over time. During the middle pooled follow-up period, model-estimated mean dAGE scores remained lower in the intervention group (5,818 kU/1000 kcal) than in controls (7,458 kU/1000 kcal; β = -1,639.87 kU/1000 kcal, P < 0.0001).

Similarly, during the final pooled follow-up period, the model-estimated mean dAGE score for intervention participants was 6,236 kU/1000 kcal, compared with 7,669 kU/1000 kcal in controls (β = -1,432.46 kU/1000 kcal, P < 0.0001). The authors noted that earlier WHI dietary analyses had documented marked decreases in total fat, added fats, meat, and desserts, reflecting reduced consumption of red and processed meats and baked goods, alongside increased consumption of plant-forward foods. These changes were consistent with, but did not establish the mechanism underlying, the lower dAGE estimates.

Conclusions

The present study provides evidence from a secondary analysis of randomized trial data suggesting that a generalized low-fat, plant-forward eating pattern was associated with meaningful, lasting reductions in estimated dietary AGE intake among participating postmenopausal women. Importantly, these outcomes suggest that changes to overall eating patterns may lower dAGE scores under free-living conditions without requiring specific cooking or food-preparation guidance.

While the study was limited by its secondary, hypothesis-generating design, reliance on self-reported FFQs and CML-based food database estimates, and inability to directly link dAGE estimates to specific foods, cooking practices, or biological AGE exposure, it supports future research examining whether reduced dAGE intake contributes to favorable breast cancer outcomes. The present analysis did not evaluate cancer incidence, mortality, or causal mediation.