Researchers at the National Cancer Institute developed "poly-metabolite scores" that distinguish between high- and low-ultra-processed food diets in both observational and experimental settings.
"Poly-metabolite scores, developed in [IDATA participants with varying diets, are predictive of UPF [ultra-processed food] intake and could advance epidemiological research on UPF and health," wrote lead author Leila Abar, PhD, colleagues.
Data was analyzed from 718 adults aged 50 to 74 years who provided biospecimens and dietary information over 12 months. Using rigorous statistical approaches, including LASSO regression, they identified 191 serum and 293 urinary metabolites significantly correlated with UPF intake. Four key metabolites emerged as robust predictors across both blood and urine samples: (S)C(S)S-S-Methylcysteine sulfoxide; N2, N5-diacetylornithine; pentoic acid (all negatively correlated with UPF intake); and N6-carboxymethyllysine (positively correlated).
N6-carboxymethyllysine, which increased with higher UPF consumption, has been previously linked to diabetes and cardiometabolic diseases. Conversely, (S)C(S)S-S-Methylcysteine sulfoxide, a biomarker of cruciferous vegetable intake, decreased with higher UPF consumption.
The researchers validated their findings in a post-hoc analysis of a randomized, controlled, crossover-feeding trial with 20 participants who consumed either an 80% or 0% UPF energy diet for 2 weeks each. The poly-metabolite scores successfully differentiated between diet phases (P < .001), confirming their utility as objective measures. In a subset of four individuals who also consumed diets with 30% energy from UPF, the researchers observed a stepwise increase in scores with increasing UPF levels.
These findings address the need for better assessment of UPF, which constitute more than 50% of calories in the American diet but have been difficult to measure accurately due to limitations in self-reported dietary data.
The study authors noted that their findings will need to be replicated in more diverse populations, as participants were primarily older U.S. adults whose diets may differ from other groups.
Disclosures can be found in the published study.
Source: PLOS Medicine