The study, published in Critical Reviews in Food Science and Nutrition, harnessed targeted metabolomics to scan the blood of 15,200 participants from the European Prospective Investigation into Cancer and Nutrition, or EPIC, a long-running observational powerhouse that has been tracking diet and disease across ten countries for decades. Plasma samples were interrogated using mass spectrometry platforms capable of quantifying hundreds of small molecules, from amino acids to complex lipids. By combining these measurements with detailed food-frequency questionnaires classified under the Nova system, the team could isolate metabolic shifts uniquely tied to ultra-processed foods, independent of age, smoking, physical activity, and total caloric intake.

What emerged was a constellation of 22 circulating metabolites strongly associated with UPF intake. Many of the perturbed molecules belong to the lipid family, but not in a random fashion. Higher UPF consumption was accompanied by elevations in specific acylcarnitines and other fatty acid derivatives that serve as established biomarkers of impaired mitochondrial fatty acid oxidation. These compounds accumulate when cells struggle to shuttle long-chain fats into mitochondria efficiently, a bottleneck that signals metabolic inflexibility and has been linked to insulin resistance, type 2 diabetes, and cardiovascular stress. Simultaneously, the blood of high UPF consumers exhibited a depletion of several essential lipids—including certain phospholipid species that maintain cell membrane fluidity, ion permeability, and receptor signaling. Such a dual pattern of toxic accumulation and protective loss suggests that ultra-processed diets do not merely add empty calories but actively disrupt the lipid infrastructure that cells rely on to communicate and generate energy.

Adding a further layer of detail, the researchers identified a distinctive profile of eight circulating fatty acids. One of the most telling signals was an elevation of stearic acid, a long-chain saturated fat that, at high levels, often indicates either excessive saturated fat intake or a metabolic state in which the body is synthesizing its own lipids far too aggressively. Alongside stearic acid, the team noted increased concentrations of very-long-chain polyunsaturated fatty acids. This paradoxical coupling—industrial fats alongside molecules typically associated with healthier oil sources—implies something remarkable: the metabolic insult of ultra-processed foods extends well beyond the mere fat content printed on packaging. It appears to ignite endogenous lipogenesis, the liver’s de novo assembly of fatty acids from surplus carbohydrates, particularly from the refined sugars and rapidly digestible starches that pervade UPFs. Even individuals consuming relatively modest amounts of industrial trans fats showed persistent traces of these compounds lingering in their plasma, a testament to how slowly the body can clear such molecular intruders.

Mechanistically, these signatures paint a picture of mitochondrial and peroxisomal stress. The simultaneous rise in odd-chain acylcarnitines and decline in plasmalogens—ether-linked phospholipids critical for protecting membranes against oxidative damage—hint at a system under siege from reactive oxygen species. This oxidative burden may be compounded by the cocktail of emulsifiers, preservatives, and neoformed contaminants that characterize ultra-processed formulations but are absent in whole foods. While observational in nature, the study’s statistical architecture was carefully reinforced. Dr. Jessica Blanco-Lopez and colleagues tested multiple regression models, sensitivity analyses, and adjustment strategies; the metabolic signature proved robust across all of them, lending confidence that the signals are not artifacts of confounding.

The implications for chronic disease are sobering. The observed metabolite disturbances converge on pathways already implicated in the onset of obesity, colorectal cancer, and atherogenic dyslipidemia. When protective structural lipids erode and fatty acid oxidation stalls, tissues begin to accumulate ectopic fat in the liver, muscle, and pancreas, fueling a cascade of inflammation and hormonal dysregulation. The findings offer one of the first molecular-level bridges between the epidemiological data—which have long linked UPF intake to higher mortality—and the elusive biological “how.”

Nevertheless, the cross-sectional design means causality cannot be established yet. The blood samples capture a single metabolic snapshot, so it remains possible that people with pre-existing metabolic vulnerabilities preferentially consume ultra-processed products, rather than the foods directly causing every observed change. Longitudinal studies that follow individuals over time, combined with repeated metabolomic profiling, will be essential to cement the directionality of these relationships. The authors are already advocating for future work that layers on untargeted metabolomics, which could capture thousands of additional unknown compounds and uncover nutritional fingerprints that current technology cannot yet annotate.

The study arrives at a moment when global regulatory bodies are wrestling with how to classify and communicate UPF risks. A metabolic signature of this resolution could one day serve as a composite biomarker in nutrition trials, allowing scientists to measure the biological efficacy of whole-food interventions without waiting decades for hard clinical endpoints. It also provides a stark reminder that food processing leaves an imprint deep within human biochemistry, one that may persist long after the last packaged snack is consumed.

Subject of Research: Cells
Article Title: Circulating metabolites and fatty acids associated with ultra-processed food consumption: results from the EPIC study
News Publication Date: Not provided
Web References: 10.1080/10408398.2026.2629025
References: Blanco-Lopez, J. et al. (2026). Circulating metabolites and fatty acids associated with ultra-processed food consumption: results from the EPIC study. Critical Reviews in Food Science and Nutrition. DOI: 10.1080/10408398.2026.2629025
Image Credits: Not provided
Keywords: Ultra-processed foods, metabolomics, fatty acids, metabolic stress, mitochondrial dysfunction, EPIC study, Nova classification, nutrition, lipidomics, public health