• Preventive medicine | Understanding risks before they manifest
• Predictive medicine | From patterns to forecasting
• Precision medicine | Individuality in context
• Population-based medicine | Power in numbers
• Participatory medicine | Empowered by omics
• The dynamic duo of the post-genomic era

When we speak about 5P medicine – preventive, predictive, precision, population-based, and participatory – the conversation often gravitates toward molecular measures of health. Yet, one essential influence on human biology that deserves a seat at the 5P table is the exposome.

Defined at the Banbury conference as “the integrated compilation of all physical, chemical, biological, and psychosocial influences that impact biology”, the exposome is becoming a necessary part of the omics and medical toolkits, and a particularly promising one when combined with metabolomics.

Metabolomists know that metabolic readouts integrate influences from both our genome and our environment. Exposomics allows us to map the upstream exposures that metabolomics reflects downstream, but it also contributes to the design of impactful metabolomic studies.

Exposomics is defined as “the field that studies the comprehensive and cumulative effects of the exposome on biological systems by integrating data from a variety of interdisciplinary methodologies and data streams” (Miller et al. 2025). These methodologies include mass spectrometry and NMR, as for metabolomics, but also dietary information, health monitoring records, medical questionnaires, geospatial data, meteorological data, and much more.

Because the effects of exogenous factors are known functions of time and intensity of exposure, exposomics is the only omic that emphasizes these parameters in the definition of its scope. There is much here to be learned for metabolomics enthusiasts.

I never tire of explaining how the flexibility and sensitivity of metabolomics is a strength rather than a weakness. But these are characteristics of exposomics too. For this reason, when combined, exposomics and metabolomics form a dynamic duo that leverages the strength of sensitive health measures in all its might.

I got confirmation of this once again recently, while recording an episode of The Metabolomist podcast where Léa Maitre from the Barcelona Institute of Global Health explains the unique strength of metabolomics in a multiomic study of early life exposures: “Metabolomics was the better omic to measure cross associations. [It was the strongest] when we measured the exposure and the omics at the same time in childhood.” You can listen to the full episode here.

This is just one example of the synergies that we unlock when we combine metabolomics and exposomics. In this blog, I will focus on the end applications of these technologies and how our dynamic duo ties to each of the 5Ps. Whether your focus is exclusively on precision medicine or you are looking for a truly holistic view of health, I hope these examples will encourage you to start integrating these two powerful omics in your research.

Preventive medicine | Understanding risks before they manifest

Preventive medicine aims to avoid disease altogether. Thus, prevention is only as strong as our ability to identify risks. Exposomics brings clarity by capturing environmental and behavioral factors such as air pollution, diet, stress, and chemical exposures that influence long-term health trajectories. Environmental and behavioral exposures strongly shape health, including drug response and chronic disease risk. Exposomics thus provides a critical foundation for anticipating and reducing exposure-derived health risks.

Metabolomics contributes here by identifying metabolic signatures linked to exposure-induced biological changes. For example, in a study of the composition of breast milk from mothers with apparently healthy infants versus stunted infants, even a small targeted metabolomic panel could identify signatures pointing to different nutrition levels (Hampel et al 2022). In the study I discuss with Léa Maitre on the podcast, metabolomics helped identify patterns linked to exposures in early childhood (Maitre, Bustamante et al. 2022) that can be followed in longitudinal studies or serve as a basis for mining the catalogue of exposome-related cohorts put together in the IHEN project.

Exposomics combined with metabolomics moves prevention from generic advice to evidence based, exposure and phenotype-specific interventions.

Predictive medicine | From patterns to forecasting

Predictive medicine hinges on data that can forecast health outcomes years before symptoms appear. Exposomics offers exactly that: the ability to quantify the cumulative external pressures shaping one’s biological trajectory. A review by Wan et al. (2025) highlights how exposomics supports diagnosis, disease prediction, early detection, and treatment prediction.

Metabolomics is also well-positioned to reflect the progressive drift of the metabolome from health towards disease outcomes. But one of its best known use is as a source of biomarkers predictive of patient drug response in pharmacometabolomics.

In non small cell lung cancer, quantitative metabolomics has shown that a patient’s baseline metabolic phenotype—shaped not just by genetics but also by diet, microbiome, inflammation and prior exposures—can predict response to immunotherapy, illustrating how the metabolome translates the cumulative exposome into actionable insight for predictive and personalized treatment (Lee et al. 2024).

In other words, exposomics tells us what happened, and metabolomics tells us how the phenotype changed; a powerful predictive duo when we want to leverage the impact of the environment on health.

Precision medicine | Individuality in context

The promise of precision medicine is the ability to tailor treatments to the individual. Genomics contributes the blueprint, but exposomics adds the context; the influences that shape how that blueprint is expressed. Metabolomics, in turn, contributes the resulting phenotype and some of the effectors of this impact on genome expression.

A type of exposure not always recognized by the public but highly relevant in medicine is the intentional exposure to chemicals such as pharmaceutical drugs. Not only do drugs influence our metabolome, but the levels of their downstream metabolic products when they pass through our organs are a powerful way to stratify patients. This is another powerful combination of exposomics and metabolomics.

In the ADNI cohort, metabolomics enabled stratification of individuals not only by disease stage, but also by medication exposure, revealing how drugs act as a critical and often overlooked dimension of the exposome (St John-Williams et al. 2017). By accounting for polypharmacy and treatment effects, this approach demonstrated how metabolomics can support more precise interpretation of molecular phenotypes and more informed patient stratification in clinical research.

In the field of nutrition research, stratification based on metabolomic profile, or “metabotyping” has become a popular tool, as it works well together with variables related to diet, another lesser-known source of deliberate exposures. In a 2023 randomized controlled trial, metabotypes were used to stratify individuals and deliver personalized dietary advice, demonstrating that people with different metabolic phenotypes respond differently to the same nutritional guidance. Leveraging metabolomics for stratification, this study demonstrated how to enable precision nutrition by translating dietary exposures into actionable, metabotype specific interventions rather than population level recommendations (Hillesheim & Brennan 2023). And in this case, the end result most likely will entail the modulation of the very exposures investigated (the diet), turning this knowledge into quickly actionable insights.

Population-based medicine | Power in numbers

The first population-based cohorts were built with genomics in mind, searching for the genetic determinants of disease. This approach opened the door for a new wave of knowledge, but it couldn’t answer all questions. Today, at the population level, exposomics reveals patterns that inform on non-genetic influencers of health especially relevant in the study of complex chronic disease.

Exposures vary dramatically between regions, occupations, socioeconomic backgrounds, and lifestyles, and the study of exposomics quickly takes us to investigate health disparities, environmental injustice, and geographically clustered risks, which are all likely to translate to metabolic differences too.

The HELIX cohort has been a pioneer in the integration of exposomics with other omics, notably combining over 200 measures of exposures with blood and urine metabolomics (Maitre et al. 2022). A follow up study investigated the links between the metabolome, health outcomes and chemical classes with known effects on health, namely endocrine disruptors. The study shows that childhood exposure to endocrine disrupting chemicals, including persistent pollutants, was associated with alterations in the metabolome, including differences in tryptophan derivatives. This work highlights the role of combined exposomics and metabolomics approaches in capturing early life biological responses to chronic environmental exposures at the population level (Fabbri et al. 2023).

Participatory medicine | Empowered by omics

When individuals engage in their own health decisions, this is one of the most direct applications of research that can be. The tenets of participatory medicine are easy-to-use sample collection, ideally performed at home to be extra accessible and reduce discriminations in access to health, and quantitative, robust measures of health that can be compared to reference values from the healthy population.

Today, measures of both exposures and health are already found in many homes, from wearables, to sensors, but also local environmental measures that lead to actionable big data. Tools that combine these measures of the exposome with reliable (metabol)omics measures will provide the solutions that will enable the application of omics-based knowledge in the home, at a scale of n=1.

Today, these offerings largely sit with private companies offering personalized fitness monitoring and advice. Tomorrow, the communities built around exposomics and metabolomics will be the cornerstone of the strategies implemented by healthcare systems providing regular checkups based on samples collected at home and sent in the mail, online questionnaires and exposure data collected by relevant home/health appliances and local exposome mapping.

The dynamic duo of the post-genomic era

To fully realize the goals of 5P medicine, we must integrate data from all layers of the biological and environmental ecosystem. Metabolomics provides the clearest snapshot of a phenotype influenced by both genetics and environment. Exposomics contributes the context in which drivers such as drugs, environmental pollutants, diet and socioeconomic factors influence this phenotype.

The intersection of these two rich omic layers hosts not only a sensitive measure of health outcomes but a wealth of information about determinants of health.
Increasingly used in population-based medicine, driving tailored approaches in preventive, predictive and precision medicine, and soon to enter the realm of participatory medicine, the combination of exposomics and metabolomics is about to revolutionize how we understand and modulate health.

• History & Evolution
• Biosynthesis & dietary uptake
• EPA and the microbiome
• EPA and inflammation
• EPA and neurology
• EPA and oncology
• EPA and 5P medicine
• References

History & Evolution

1957: first isolated (Klenk, E. and Eberhagen, D., 1957) | 1964: EPA linked to prostaglandin production (Bergström, S. et al., 1964) | 1978: discovery of cardiovascular protective effects

Eicosapentaenoic acid (EPA) is a long-chain omega-3 polyunsaturated fatty acid (PUFA) and a precursor to many lipid signaling molecules involved in inflammation and cardiometabolic regulation. The name reflects its structure: “eicosa” refers to its 20 carbon atoms, while “pentaenoic” refers to five double bonds (Castro, L. et al., 2016).

EPA was first isolated from cod liver oil in 1957 by Klenk and Eberhagen (Klenk, E. and Eberhagen, D., 1957). At the time, lipid research focused on omega-6 fatty acids, particularly arachidonic acid (AA) and linoleic acid, as their roles in cholesterol metabolism and prostaglandin synthesis made them more compelling targets (Spector, A. and Kim, H., 2019). Although Bergström later showed that EPA could also be enzymatically converted into prostaglandin E3 (Bergström, S. et al., 1964), this pathway attracted little attention and interesting early findings on omega-3 fatty acids were largely overlooked.

This changed in 1978, when Dyerberg and Bang observed notably low rates of myocardial infarction among Greenland Eskimos, who typically followed a diet of marine lipids rich in EPA and docosahexaenoic acid (DHA) (Dyerberg, J. and Bang, O., 1978).

By the 1980s, epidemiological and clinical studies linked diets rich in marine omega-3 fatty acids to lower rates of cardiovascular disease, putting EPA at the heart of vascular biology. EPA has since been shown to have protective effects against wide-ranging diseases, including obesity, diabetes, chronic obstructive pulmonary disease and certain cancers (Watabe, S. et al., 2024).

EPA is found primarily in marine microalgae, fish and other marine animals, where it accumulates through the food chain.

Biosynthesis vs. dietary uptake

In humans, EPA comes mainly from the diet, particularly oily fish, seafood and fish-based supplements. However, this is not a reliable source. Fish do not produce EPA themselves but obtain it from algae, so their EPA content varies depending on their diet and environment (Watabe, S. et al., 2024). Concerns about sustainability and contamination of fish and fish-derived supplements have increased interest in alternative sources such as algal oils (Sousa, S. et al., 2024).

EPA can also be produced endogenously, but only in small amounts. It is synthesized from the essential fatty acid alpha-linolenic acid (ALA), found in foods such as flaxseed, walnuts and vegetable oils (Watabe, S. et al., 2024). This conversion takes place mainly in the liver through a series of enzymatic reactions. The process is quite inefficient: only 1-15% of ALA is converted to EPA, and this can be further reduced by alcohol consumption, smoking, inactivity and dietary factors (Sousa, S. et al., 2024).

As a result, endogenous synthesis does not make a significant contribution to circulating EPA and, for most people, dietary intake is the main source. Recommended daily intake is around 250-500 mg per day (together with downstream metabolite DHA), rising to 700-1000 mg for pregnant and lactating women (Sousa, S. et al., 2024).

Once in circulation, EPA is incorporated into cell membranes where it can be converted into lipid mediators, including prostaglandins, leukotrienes and resolvins (Igarashi, M. et al., 2013).

EPA and the microbiome

Research shows that EPA interacts with the gut microbiome, directly affecting bacterial growth, diversity and survival. As with other PUFAs, EPA has been shown in experimental settings to have an antibacterial effect against Gram-positive species such as Bacillus cereus and Staphylococcus aureus, which are both causes of infection (Le, PNT. and Desbois, A., 2017).

Studies report a decrease in the Firmicutes/Bacteroidetes ratio and increases in genera such as Bifidobacterium, Lachnospira, Roseburia and Lactobacillus (Amedei, A. et al., 2025). EPA has also been found to modulate levels of pro-inflammatory molecules and affect short-chain fatty acid (SCFA) concentrations (Fu, Y. et al., 2021). These effects are associated with improved cardiovascular risk profiles.

In turn, beneficial bacteria including Bifidobacterium, Lactobacillus and Akkermansia have been shown to influence EPA absorption and metabolism, indicating a two-way interaction between EPA and the gut microbiome (Amedei, A. et al., 2025).

EPA and inflammation

Inflammation is one of the main pathways through which EPA affects human health.
As a precursor to lipid signaling molecules, EPA influences how inflammatory mediators are produced. In cell membranes, it partly replaces omega-6 fatty acids such as arachidonic acid, changing the pool of fatty acids available for enzymatic conversion and leading to the production of eicosanoids that are generally less pro-inflammatory (Crupi, R. and Cuzzocrea, S., 2022). EPA is also converted into specialized pro-resolving mediators, such as resolvins, which help limit and resolve inflammation. More broadly, these changes in membrane composition influence cell signaling, gene expression and the production of cytokines, adhesion molecules and other inflammatory mediators (Calder, P., 2010).

These pathways are also influenced by the exposome. Environmental factors such as diet, pollutants and oxidative stress can alter fatty acid composition and downstream inflammatory signaling (Hinman, J. et al., 2025). Studies show that higher omega-3 status, including EPA levels, is associated with increased pro-resolving and reduced pro-inflammatory lipid mediator responses following exposures such as air pollution, highlighting EPA’s role in shaping inflammatory responses to environmental stressors (Chen, H. et al., 2025).

Clinically, this translates into measurable effects on disease-related inflammation. EPA has been shown to reduce pro-inflammatory cytokines and oxidative stress, improve endothelial function and support stabilization of atherosclerotic plaques, all of which are relevant to cardiovascular risk (Crupi, R. and Cuzzocrea, S., 2022). In inflammatory conditions such as rheumatoid arthritis, higher intakes of EPA (typically with DHA) are associated with reduced inflammatory activity, although relatively high and sustained doses are often required (Calder, P., 2010). A 2024 literature review found that EPA is linked to improved metabolic and inflammatory profiles in non-communicable diet-related conditions such as obesity, diabetes and cardiovascular disease (Banaszak, M. et al., 2024).

A 2021 study using targeted lipidomics linked EPA’s anti-inflammatory effects to improvements in depression (Borsini, A. et al., 2021). Using a “depression in a dish” model of human hippocampal cells, researchers showed that EPA-derived lipid mediators protected neurons from inflammation-related damage and supported neurogenesis. In patients with major depressive disorder, higher levels of these metabolites were associated with reduced symptom severity, suggesting a mechanistic link between EPA, inflammation and depression.

EPA and neurology

EPA has been studied in neurology because of its effect on synaptic function, neuroinflammation and neuronal survival. In a preclinical study, EPA was taken up by both neurons and glial cells (Kawashima, A. et al., 2010). This increased long-term potentiation in the hippocampus and activated intracellular signaling involving phosphoinositide 3-kinase (PI3K) and protein kinase B (Akt), while reducing markers of cell death. These findings suggest a potential role in protecting against neurodegeneration and support interest in EPA in conditions such as Alzheimer’s disease, Huntington’s disease and schizophrenia.

More recent work has looked at EPA in neurodevelopmental disorders and epilepsy, where inflammation, oxidative stress and synaptic dysfunction are recurring factors (Li, M. et al., 2025). Omega-3 fatty acids, including EPA, may help in attention deficit hyperactivity disorder (ADHD), autism spectrum disorder and Tourette syndrome, but the evidence is mixed and clinical effects are often modest. For example, in ADHD, some studies report improvement in attention or behavior from higher EPA doses, while others find no clear benefit. In epilepsy, EPA and other omega-3 fatty acids have been explored mainly as adjuncts rather than stand-alone treatments, with some studies reporting reduced inflammatory markers or seizure frequency.

Omics studies suggest that lipid dysregulation, including EPA, may contribute to neurological disease progression. For example, in a 2025 longitudinal metabolomic-lipidomic study of 767 people with multiple sclerosis, decreased serum EPA was associated with slower walking speed and reduced manual dexterity (Noroozi, R. et al., 2026).

A 2026 study found that after repetitive mild traumatic brain injury, EPA accumulates in the brain under baseline conditions but is then depleted during injury-related metabolic stress (Karakaya, E. et al., 2026). This was associated with impaired endothelial repair, reduced angiogenic signaling and vascular and cognitive decline.

These findings make EPA a metabolite of interest as a marker of and potential therapeutic target in neuroinflammatory and neurovascular processes.

EPA and oncology

The relationship between EPA and cancer is unclear. The metabolite has been shown to suppress proliferation in colon, pancreatic, breast, esophageal and other cancer cell lines (Mizoguchi, K. et al., 2014), and EPA-derived resolvins seem to have an anti-tumor effect (Kiyasu, Y. et al., 2024).

However, clinical studies show mixed results. In animal models, for example, fish oil has been shown to both suppress colorectal carcinogenesis and to promote colitis-associated cancer (Kiyasu, Y. et al., 2024).

Evidence in patients is also limited. Small studies in patients undergoing chemotherapy report changes in inflammatory markers and immune function with EPA (often combined with DHA) (Munhoz, J. et al., 2025). However, findings are inconsistent, and there is currently insufficient evidence to support routine use in cancer care.

EPA and 5P medicine

EPA’s links to diet, genetics and inflammation make it an interesting metabolite for the study of chronic, non-communicable diseases, which are a top priority in population health. In epidemiological research, circulating omega-3 fatty acids, including EPA, are used as objective biomarkers of dietary intake and have been associated with cardiovascular outcomes and mortality risk across large cohorts and meta-analyses (Jiang, H. et al., 2022).

Studies also show that EPA’s effects depend on both exposure (diet) and host biology (genetics), with implications for precision medicine. For example, in the seAFOod polyp prevention trial, EPA supplementation was found to reduce adenoma risk by around 50% in individuals with a specific FADS variant, while those without the allele saw no benefit (Sun, G. et al., 2024).

EPA also seems to play a role in aging, making it increasingly relevant as populations age and the burden of chronic disease grows (Hinman, J. et al., 2025). Studies show changes in PUFA profiles and inflammatory and oxidative stress markers across the lifespan (Aiello, A. et al., 2024). In long-lived women, a lower AA to EPA ratio was associated with higher antioxidant capacity (Hinman, J. et al., 2025).

References

Aiello, A. et al. (2024). Polyunsaturated fatty acid status and markers of oxidative stress and inflammation across the lifespan: A cross-sectional study in a cohort with long-lived individuals. Experimental Gerontology, 195, 112531. DOI: https://doi.org/10.1016/j.exger.2024.112531.

Amedei, A. et al. (2025). Potential and Future Therapeutic Applications of Eicosapentaenoic/Docosahexaenoic Acid and Probiotics in Chronic Low-Grade Inflammation. Biomedicines, 13(10), 2428. DOI: https://doi.org/10.3390/biomedicines13102428.

Banaszak, M. et al. (2024). Role of Omega-3 fatty acids eicosapentaenoic (EPA) and docosahexaenoic (DHA) as modulatory and anti-inflammatory agents in noncommunicable diet-related diseases – Reports from the last 10 years. Clinical Nutrition ESPEN, 63, 240-258. DOI: https://doi.org/10.1016/j.clnesp.2024.06.053.

Bergström, S. et al. (1964). The Enzymatic Conversion of Essential Fatty Acids into Prostaglandins: PROSTAGLANDINS AND RELATED FACTORS 34. Journal of Biological Chemistry, 239, PC4006-PC4008. DOI: https://doi.org/10.1016/S0021-9258(18)91234-2.

Borsini, A. et al. (2021). Omega-3 polyunsaturated fatty acids protect against inflammation through production of LOX and CYP450 lipid mediators: relevance for major depression and for human hippocampal neurogenesis. Molecular Psychiatry, 26, 6773–6788. DOI: https://doi.org/10.1038/s41380-021-01160-8.

Calder, P. (2010). Omega-3 Fatty Acids and Inflammatory Processes. Nutrients, 2(3), 355–374. DOI: https://doi.org/10.3390/nu2030355.

Castro, L. et al. (2016). Long-chain polyunsaturated fatty acid biosynthesis in chordates: Insights into the evolution of Fads and Elovl gene repertoire. Progress in Lipid Research, 62, 25-40. DOI: https://doi.org/10.1016/j.plipres.2016.01.001.

Chen, H. et al. (2025). Omega-3 Fatty Acid Intake and Oxylipin Production in Response to Short-Term Ambient Air Pollution Exposure in Healthy Adults. Toxics, 13(12), 1063. DOI: https://doi.org/10.3390/toxics13121063.

Crupi, R. and Cuzzocrea, S. (2022). Role of EPA in Inflammation: Mechanisms, Effects, and Clinical Relevance. Biomolecules, 12(2), 242. DOI: https://doi.org/10.3390/biom12020242.

Dyerberg, J. and Bang, O. (1978). Dietary fat and thrombosis. The Lancet, 311(8056), 152. DOI: https://doi.org/10.1016/S0140-6736(78)90448-8.

Fu, Y. et al. (2021). Associations among Dietary Omega-3 Polyunsaturated Fatty Acids, the Gut Microbiota, and Intestinal Immunity. Mediators Inflamm, 8879227. DOI: https://doi.org/10.1155/2021/8879227.

Hinman, J. et al. (2025). Effects of molecular interactions between the exposome and oxylipin metabolism on healthspan. Front. Physiol., 16, DOI: https://doi.org/10.3389/fphys.2025.1584195.

Igarashi, M. et al. (2013). Kinetics of eicosapentaenoic acid in brain, heart and liver of conscious rats fed a high n-3 PUFA containing diet. Prostaglandins Leukot Essent Fatty Acids, 89(6), 403–412. DOI: https://doi.org/10.1016/j.plefa.2013.09.004.

Jiang, H. et al. (2022). Omega-3 polyunsaturated fatty acid biomarkers and risk of type 2 diabetes, cardiovascular disease, cancer, and mortality. Clin Nutr, 41(8), 798-1807. DOI: https://doi.org/10.1016/j.clnu.2022.06.034.

Karakaya, E. et al. (2026). Eicosapentaenoic acid reprograms cerebrovascular metabolism and impairs repair after brain injury, with relevance to chronic traumatic encephalopathy. Cell Reports, 117135. DOI: https://doi.org/10.1016/j.celrep.2026.117135.

Kawashima, A. et al. (2010). Effects of eicosapentaenoic acid on synaptic plasticity, fatty acid profile and phosphoinositide 3-kinase signaling in rat hippocampus and differentiated PC12 cells. J Nutr Biochem, 21(4), 268-77. DOI: https://doi.org/10.1016/j.jnutbio.2008.12.015. Epub 2009.

Kiyasu, Y. et al. (2024). EPA, DHA, and resolvin effects on cancer risk: the underexplored mechanisms. Prostaglandins Other Lipid Mediat, 174, 106854. DOI: https://doi.org/10.1016/j.prostaglandins.2024.106854.

Klenk, E. and Eberhagen, D. (1957). [Occurrence of delta5, 8, 11, 14, 17-n-eicosapentanoic acid in cod liver oil and its isolation] (Article in German). Hoppe Seylers Z Physiol Chem, 307(1), 42-8.

Le, PNT. and Desbois, A. (2017). Antibacterial Effect of Eicosapentaenoic Acid against Bacillus cereus and Staphylococcus aureus: Killing Kinetics, Selection for Resistance, and Potential Cellular Target. Mar Drugs, 15(11), 334. DOI: https://doi.org/10.3390/md15110334.

Li, M. et al. (2025). Omega-3 fatty acids: multi-target mechanisms and therapeutic applications in neurodevelopmental disorders and epilepsy. Front. Nutr. , 12, 1598588. DOI: https://doi.org/10.3389/fnut.2025.1598588.

Mizoguchi, K. et al. (2014). Induction of Apoptosis by Eicosapentaenoic Acid in Esophageal Squamous Cell Carcinoma. Anticancer Research December, 34(12), 7145-7149. PMID: 25503142.

Munhoz, J. et al. (2025). Perspective: Implications of Docosahexaenoic Acid and Eicosapentaenoic Acid Supplementation on the Immune System during Cancer Chemotherapy: Perspectives from Current Clinical Evidence. Advances in Nutrition, 16(8), 100464. DOI: https://doi.org/10.1016/j.advnut.2025.100464.

Noroozi, R. et al. (2026). Serum metabolomic-lipidomic signatures track long-term neurological performance in multiple sclerosis. Mol. Neurodegener. Adv., 2(1), DOI: https://doi.org/10.1186/s44477-025-00008-7.

Sousa, S. et al. (2024). Factors impacting the microbial production of eicosapentaenoic acid. Appl Microbiol Biotechnol, 108(1), 368. DOI: https://doi.org/10.1007/s00253-024-13209-z.

Spector, A. and Kim, H. (2019). Emergence of omega-3 fatty acids in biomedical research. Prostaglandins, Leukotrienes and Essential Fatty Acids, 140, 47-50. DOI: https://doi.org/10.1016/j.plefa.2018.11.017.

Sun, G. et al. (2024). Fatty acid desaturase insertion-deletion polymorphism rs66698963 predicts colorectal polyp prevention by the n-3 fatty acid eicosapentaenoic acid: a secondary analysis of the seAFOod polyp prevention trial. Am J Clin Nutr, 120(2), 360-368. DOI: https://doi.org/10.1016/j.ajcnut.2024.06.004.

Watabe, S. et al. (2024). Daily Consumption of α-Linolenic Acid Increases Conversion Efficiency to Eicosapentaenoic Acid in Mice. Nutrients, 16(10), 407. DOI: https://doi.org/10.3390/nu16101407.

Metabolomics, with its ability to track the behavior of small molecules, has a unique place in the 5P transformation, providing a measure of the impact of not only genetics, but also environment and lifestyle:

• Metabolomics enables preventive medicine through early, often reversible metabolic shifts that precede the onset of chronic disease
• Metabolomics enables predictive medicine through robust biomarker signatures already harnessed to foretell drug response
• Metabolomics enables precision medicine through stratification and metabotypes that group patients based on the molecular presentation of their disease,
• Metabolomics enables population based medicine by providing insight into environmental variance necessary to step into the post-genomic era
• Metabolomics enables participatory medicine through patient centric care including remote at-home sampling and actionable biochemical information.

This multifaceted perspective provided by metabolomics remains to be included in the routine toolkit of most medical practitioners, but when it does, it will drive a massive transformation in the way that we approach health.

Driving the adoption of metabolomics through 5P medicine

Here I introduce a simple but powerful model called Screen | Leverage | Translate. This 3-step framework explains how any scientist can harness the information contained in metabolomics and turn it into the solution to their specific 5P medicine problem.

If you are familiar with my book, the STORY principle, you’ll notice that Screen | Leverage | Translate follows a similar direction to the principle I describe there, planning an experiment, executing it, and excavating the gems out of your dataset before carving it into the answer you need for your research. Screen | Leverage | Translate is a streamlined version of the STORY principle, adapted to specific applications in 5P medicine.

Step 1 | Screen

This is the data acquisition step, where you’ll analyze the samples required for your experiment. Going back to the STORY principle, there is an unspoken understanding that by the time you reach sample measurement, you’ve already put in the work to determine your research question and how metabolomics will help answer it.

This doesn’t mean you know the results, rather that you’ve accounted for any factors that may predictably confound your findings. With a measure as sensitive to the environment as metabolomics, it is crucial to have these factors in check, in order to make the most of each study. Today, enough is known about these confounders to help you avoid interference by, e.g. sex, age, fasting time, diet, exposures, and more. If unsure, you can always refer back to the Safety Check step in my book or ask your metabolomics collaborators for support in planning your experiment.

A few questions to ask in preparation for this screening step relate to the method used for metabolomics measurement. Every method has benefits and limitations. Understanding the form of answer you need to your research question will turn certain benefits into requirements, and others into “nice-to-haves”. The method that provides your requirements without hindering the leveraging of the results is the one best suited for your study.

In our pathway spotlight below, I will describe an example focused on polyamine metabolites and their role in immunotherapy response.

Step 2 | Leverage

During your preparation step, you’ll also have planned the tools to be used to analyze and leverage your data. These days, with data and study design complexity increasing, the more automated the tool, the better.

Among omics, metabolomics is unique in the way that its datapoints connect to each other and to health. Unlike other omics, no direct line can be drawn between a given metabolite and a gene of origin. This makes the application of genomics-derived tools impossible, which is why the metabolomics community has had to develop its own data analysis tools.

Metabolites are most often grouped and analyzed around metabolic pathways. This work demands a deep knowledge of these mechanisms, that is condensed in databases like Wikipathways, Reactome and KEGG; databases that are constantly growing, as our knowledge of metabolism in different contexts expands.

Context is crucial.

Depending on the matrix you are studying (blood, urine, feces, tissue from an organ, cell model…), different metabolic pathways are at work – interacting, influencing each other, balancing each other.

The optimal way to leverage metabolomics is intricately linked to the context of the study. Species, matrix, disease context, study design – when combined, these factors make each experiment unique. But knowledge from previous experiments is a great way to start leveraging your own. I’ll introduce the one we have developed in our software in our pathway spotlight example.

Step 3 | Translate

Translation is where every omic meets its end application; where the well-oiled machine of data acquisition needs to slow down and level with what awaits it on the other side.

For applications where a routine solution is the expected outcome, translation is just the creative part before another well-oiled machine is created; e.g. a biomarker panel for routine disease screening, a biomarker for drug response prediction, a bioinformatics pipeline combining metabolomics with other omics and medical endpoints.

Seen through this lens, translating metabolomics is the thrilling part. It is the moment of truth, where the results of the previous steps are put to the test: will they hold up in a follow-up study? Because indeed, after crafting the solution to your original research question, you’ll need to verify its power in one or more validation studies.

This is when research begins to bring value to society; how we truly leverage knowledge to improve lives. Three years after its publication, I still refer to the paper by Tintelnot et al. on pancreatic cancer patient response to chemotherapy. Using metabolomics, this team identified a microbial metabolite, 3-indole acetic acid (3-IAA), as higher at baseline in patients who responded to treatment. Many studies end there, but not this time. What followed was a set of experiments in an animal model of pancreatic cancer, testing the impact of the patients’ microbiomes in fecal matter transplantation (FMT) experiments, an evaluation of the effect of direct supplementation with 3-IAA and other metabolites, and painting the picture of how 3-IAA modulated the immune system of the mice to improve chemotherapy response.

These are the kinds of studies I love to read: where omics are leveraged to start crafting the solutions to a specific problem. This is why translation is called “translation”. It takes work and creativity to transform the tabular results provided by omics into the life-sized solutions needed in the clinics.

In the pathway spotlight below, I explore how metabolites, specifically polyamines, may bring similar solutions for immunotherapy response. My goal here is to walk you through bits of published science to understand how you could leverage and translate your next metabolomics experiment to accelerate its adoption in the future medical landscape.

Pathway spotlight | Polyamine acetylation as a driver of immunotherapy response

Metabolomics profiling creates a broad picture of the metabolome. Depending on the level of expertise of scientists, they may enter this step with a fully blank mind or with an idea of what they think they will see. This second case is ideal, as it can also use some of the results as a form of positive control for the phenotypes that they are studying.
Here we will work under the assumption that broad metabolic profiling identified differences in polyamines in a cohort of patients receiving immunotherapy treatment for cancer. Since this pathway is relevant in several types of cancer, I will not specify here and rather detail the disease context each time that a study is discussed in this spotlight

Step 1 | Screen
In a study of patients undergoing anti CD19 CAR T-cell therapy for relapsed or refractory large B cell lymphoma, Fahrmann et al. found that acetylated polyamines, specifically acetylspermidine and diacetylspermidine, were elevated in non‑responders and strongly linked to poor treatment durability (Fahrmann et al. 2022).
The first thing any scientist would do is to research what these metabolites are known for.

Polyamines – putrescine, spermidine, and spermine – are essential molecules for cell growth, gene regulation, and immune function. Cancer cells exploit this pathway aggressively. A key regulator of this system is spermidine/spermine N¹ acetyltransferase (SSAT), the enzyme responsible for polyamine acetylation and export.

Mechanistic investigations show that overexpression of SSAT dramatically increases flux through the polyamine pathway, triggering both acetylation and compensatory upregulation of synthesis. This “futile cycle” results in a large, constantly replenished polyamine pool (Kramer et al. 2008). In addition, acetylated polyamines accumulate in several cancers, including breast, prostate, and lung tumors, contributing to an immunosuppressive microenvironment that blunts therapeutic efficacy. (Thomas & Thomas, 2007, Kramer et al. 2008)

This information is valuable and suggests that these metabolic differences may link to differences in treatment response. However, literature research can be a long and confusing endeavor. This is why the whole community strives to automate leveraging and interpreting metabolomics results. In the following step, we’ll review how MetaboINDICATOR can facilitate the first steps of this work.

Step 2 | Leverage
Enough data has been generated by now to inform on the expected metabolic profile of at least part of an experiment and better understand its context. This is what MetaboINDICATOR was designed for. This module of the WebIDQ software contains hundreds of sums and ratios of metabolites hand-picked from the literature and distributed across disease- and mechanism-related categories.
The most open-ended way to use this module is to compute any indicator for which the metabolites are quantified in the sample. This is my recommendation to maximize the chance to learn from other fields; although it is also possible to select only the groups of indicators of interest to the study, for example indicators related to cancer.

After statistical analysis of these new datapoints, data interpretation can begin. MetaboINDICATOR provides references to the literature where each sum or ratio has been previously described, thus immediately expanding the perspective for application.

Let’s take the example of the ratio of acetylated to non-acetylated polyamines. Experimental work in cancer models shows that this ratio correlates strongly with SSAT enzyme expression, providing a functional handle on acetylation dynamics (Kramer et al. 2008). While the ratio of product to substrate can be used as a proxi for enzymatic activity, interpretation based on the biological context is key. What is most interesting here though is the correlation of these metabolite levels with immunotherapy response.

For someone with little to no experience in metabolomics, having this automated access to a relevant indicator and literature attached to it can save hours of literature research and put them directly on the right track for their interpretation, where the precise context of their research is going to be a strong driver of their thinking process towards translation.

Step 3 | Translate
For this step, I’ll go back to the 5P medicine concept and pick examples for predictive and precision medicine applications of metabolomics.

Predictive Medicine: Forecasting immunotherapy success
Recent clinical research demonstrates that polyamine related metabolites are predictive of patient response. Elevated plasma acetylated polyamines are associated with poor response to CAR T-cell therapy in relapsed/refractory large B cell lymphoma. A 6 marker metabolite panel (including acetylspermidine and diacetylspermidine) was validated across cohorts as a predictor of non durable response (Fahrmann et al. 2022). This was the outcome of the very first study I introduced in step 1.

Identifying a biomarker signature is one of the best-known applications of metabolomics, and with strong results in validation cohorts, this can lead to deep impact in how medicine will be practiced in the field. But I also wanted to include an example focused on new drug therapy development.

Precision Medicine: Polyamine blockade therapy
Targeting polyamine metabolism doesn’t stop at prediction; it also offers therapeutic opportunity.
A landmark study introduced polyamine blockade therapy, a dual strategy combining inhibition of polyamine synthesis (DFMO) with a polyamine transport inhibitor (Trimer PTI). Polyamine blockade therapy significantly reduced tumor growth more effectively than either inhibitor alone (Alexander et al. 2017). The anti tumor effect is T cell dependent, featuring increased cytotoxic CD8⁺ T cells and reduced immunosuppressive cell populations. By reversing the immunosuppressive effects of high polyamine levels, this approach exemplifies how metabolomics informed targets can directly shape therapeutic innovation.

This pathway spotlight illustrates how metabolomics, guided by the Screen | Leverage | Translate framework, enables deep biological insight and clinical translation. Polyamine metabolism is just one pathway among many, yet its story captures the essence of what 5P medicine strives for: using precise biochemical information to create predictive, preventive, and precise health solutions, from individual to population health.

As you explore your own datasets, I encourage you to adopt this structured approach. In particular, don’t stop at the screening step. Dig further. Leverage your data to gain a broader understanding of the metabolic changes in your system, and devise the next experiments that will yield the truly transformative solutions to transform medicine.

This year’s Arizona Metabolomics & T32 Symposium was hosted by Dr. Haiwei Gu and Dr. Corrie Whisner at Arizona State University in Phoenix, AZ. This one-day event highlighted a variety of emerging research topics and applications in the field of metabolomics, systems biology, and health.

The event featured a wide range of topics, from substance exposure in adolescent development, gut microbiome mechanisms, food insecurity studies, sitting time and cardiovascular disease health risk, metabolic biomarker discovery, and pathway analysis.

This conference marked the first joined attendance of a metabolomics event by biocrates and PreOmics in the US, since the creation of Biognosys Group. Here are the perspectives of two of our colleagues who attended the meeting together.

Throughout the day, the breadth and quality of presentations underscored just how versatile and far-reaching metabolomics has become. Several talks focused on the growing intersection between metabolomics and gut microbiome research – covering topics such as the health benefits of acetate and short-chain fatty acid supplementation, the impact of food insecurity on microbial stability, and how specific bacterial interactions shape glucose metabolism. Other presentations highlighted novel applications, including how prolonged sitting and inactivity can rapidly alter metabolic profiles, and how metabolomics can be leveraged to assess and even predict biological age. Discussions on metabolite identification and normalization further emphasized the field’s ongoing need for rigorous, reproducible analytical standards.

In addition to the excellent science presented at the conference, it was a pleasure to showcase our combined technologies at a shared table, which provided a unique opportunity to engage with metabolomics researchers and customers, about their metabolomics, but also proteomics strategies. We had the opportunity to engage with scientists regarding biocrates’ standardized metabolomics and lipidomics kits, including analytical methods tailored to their mass spectrometry instruments, and workflow management software. We could also expand the discussion towards PreOmics’ low to high throughput standardized kit and workflow solution for bottom-up proteomic sample preparation. Allowing us to showcase top market technologies to both experienced analytical chemists and newcomers to liquid chromatography mass spectrometry-based proteomics.

Our joint presence highlighted the power of combining metabolomics and proteomics to deliver richer, more connected insights that drive innovation across health and life science research.

Overall, the symposium served as a vibrant platform for advancing metabolomics research, strengthening regional collaboration, and fostering innovations aimed at improving health outcomes across diverse areas of study.

• History & Evolution
• Biosynthesis & dietary uptake
• Dimethylglycine and cardiology
• Dimethylglycine and diabetes
• Dimethylglycine and neurology
• Dimethylglycine and immunity
• Dimethylglycine and oncology
• Dimethylglycine and 5P medicine
• References

History & Evolution

1943: discovery | 1950-1960: description as endurance enhancer | 1974: marketed as supplement

Dimethylglycine (DMG) was first described in 1943 (Cupp et al. 2003). It gained attention in the 1950s-1960s through studies conducted in the Soviet Union claiming improved oxygen use, reduced fatigue, and enhanced physical performance (Wolfsegger et al. 2021). This led to its widespread use among Russian athletes and cosmonauts. During this period, DMG was often incorrectly marketed as vitamin B15 or pangamic acid, in which it was sometimes included (Cupp et al. 2003).
DMG entered the U.S. supplement market in 1974, and similar performance claims appeared in American reports in 1975 (Wolfsegger et al. 2021). In 1982, the Food and Drug Administration (FDA) banned pangamic acid due to inconsistent composition and unsupported health benefits, but DMG itself remained legal (Cupp et al. 2003). Today, DMG is recognized not as a vitamin but as a natural metabolite in one carbon metabolism.

Biosynthesis vs. dietary uptake

DMG is a naturally occurring amino acid derivative present in the cells of all plants and animals. Small amounts are obtained through the diet, particularly from legumes, grains, and meat (Yao et al. 2022; Cupp et al. 2003). Dietary DMG that is not immediately metabolized in the liver enters the circulation and is delivered to peripheral tissues (Svingen et al. 2013)
In humans, most DMG arises from endogenous metabolism rather than direct dietary intake (Cupp et al. 2003). DMG is formed when choline is metabolized along the choline–betaine–glycine pathway. A key step in this path is the remethylation of homocysteine mainly in liver and kidney tissue: the enzyme betaine homocysteine methyltransferase (BHMT) transfers a methyl group from betaine to homocysteine, generating methionine and DMG (McGregor et al. 2001). Notably, DMG inhibits BHMT via negative a feedback loop (Svingen et al. 2013).
Once formed, DMG is rapidly oxidized in mitochondria by dimethylglycine dehydrogenase (DMGDH) to produce sarcosine, linking DMG metabolism directly to the folate dependent one carbon cycle (McAndrew et al. 2008). Under normal physiological conditions, circulating DMG levels remain low. A small fraction of unmetabolized DMG is excreted in urine (Svingen et al. 2013). Elevated DMG occurs when DMGDH function is impaired (McAndrew et al. 2008) or when BHMT activity is altered (McGregor et al. 2001).

Dimethylglycine and cardiology

Observational cohort studies consistently link plasma DMG levels with cardiovascular outcomes. In patients with recent acute coronary syndrome, higher baseline DMG was associated with increased risk of acute myocardial infarction, heart failure, and death over ~2.5 years (Lever et al. 2012). Similarly, in a large cohort with stable angina pectoris, DMG correlated with traditional coronary artery disease risk factors and independently predicted incident myocardial infarction during ~4.6 years of follow up (Svingen et al. 2013).
Furthermore, two cohorts of patients with suspected stable angina pectoris or acute myocardial infarction showed that elevated DMG was associated with higher all-cause and cardiovascular mortality, with stronger effects in acute myocardial infarction. DMG improved risk prediction in both groups (Svingen et al. 2015).
In pregnant women carrying fetuses with confirmed congenital heart disease, decreased maternal serum DMG emerged as a potential prenatal diagnostic marker (Xie et al. 2025). Mechanistically, high plasma DMG levels may be linked to the regulation of lipid and energy metabolism.
Experimental evidence suggests that activation of peroxisome proliferator activated receptor alpha (PPARα) reduces transcription of enzymes involved in DMG catabolism, potentially contributing to elevated circulating DMG concentrations (Svingen et al. 2015). However, the exact biological mechanisms underlying the observed associations between DMG and cardiovascular outcomes remain to be fully elucidated.

Dimethylglycine and diabetes

Observational human data indicate that low plasma DMG is associated with higher blood glucose. A genome-wide-association study identified a strong signal at the DMGDH locus that associated with lower DMG and higher insulin, HOMA IR and incident diabetes risk. This suggests that DMG deficiency may contribute to diabetes development and that DMGDH inhibition or DMG supplementation could be explored therapeutically (Magnusson et al. 2015). Additional metabolomics work shows DMG levels differ between normal glucose tolerance and impaired glucose regulation, positioning DMG as a risk associated marker in early dysglycemia (Liu et al. 2024).
Beyond these findings, maternal metabolic studies indicate that mothers carrying twins with low plasma DMG/betaine ratios have a decreased risk of gestational diabetes, suggesting that DMG-betaine balance may modulate glucose metabolism during pregnancy (Gong et al. 2021). Although all human evidence is observational and mechanistic or interventional studies are missing. The metabolic link becomes even clearer in the context of liver disease. The same ratio, this time assessed directly in hepatic tissue, proved highly relevant in a study of metabolic dysfunction associated steatotic liver disease (MASLD) in mice. The impaired DMG/betaine ratio was closely linked to disrupted one carbon remethylation (Pacana et al. 2015). These liver findings support the idea that altered DMG metabolism is part of a broader metabolic dysfunction relevant to both hepatic disease and diabetes.

Dimethylglycine and neurology

DMG supplements are often advertised as supporting cognitive function, mood, memory, or neurotransmitter synthesis, but scientific evidence for neurological benefits remains limited. Preclinical data show some promising mechanistic effects: in a BTBR mouse model of autism spectrum disorder (ASD), supplementation with DMG and B vitamins reduced ASD like behaviors, potentially through multifactor effects including decreased oxidative stress and inflammation, improved gut microbiota and gut permeability, altered body composition, reduced hepatic steatosis, and enhanced mitochondrial function in both liver and brain (Cimmino et al. 2025). However, these findings did not translate into clinical benefit. A double blind, placebo controlled trial in children with autism or pervasive developmental disorder reported no behavioral improvements in 33 of 37 participants receiving DMG (Kern et al. 2001). Similarly, a randomized, placebo controlled pilot study in patients with progressive multiple sclerosis found no effects of DMG on disability, fatigue, or cognitive and motor performance (Wolfsegger et al. 2021). Recent data strengthen the mechanistic rationale for DMG in neurodegeneration, particularly Alzheimer’s disease (AD). DMG was identified as the most effective of twelve metabolites in restoring catalase function, a key antioxidant enzyme impaired in AD, by preventing its aggregation, enhancing activity, stabilizing its native fold, and reducing amyloid like fibrils. Together, these effects suggest that DMG may help counter oxidative stress, a major driver of AD pathology (Devi et al. 2026). However, despite these mechanistic insights, human studies, including interventional trials or metabolomics guided supplementation studies, are still lacking, and no clinical evidence currently supports neurological benefits of DMG.

Dimethylglycine and immunity

DMG shows consistent immunomodulatory effects in animal studies, but human evidence remains limited. In a rabbit model, DMG supplementation markedly increased antibody titers to influenza and Salmonella typhi vaccines and enhanced lymphocyte proliferation, indicating stronger humoral and cellular immunity (Reap et al. 1990). A feline study also reported immunological shifts after DMG-containing supplementation, with reduced neutrophils and increased lymphocytes, possibly reflecting changes in B cell or NK cell activity (Shahril Agus et al. 2025). DMG may further influence immune tolerance: in a rat model of recurrent implantation failure, reduced DMG was linked to lower Treg cell levels and impaired endometrial receptivity, whereas supplementation helped restore immune balance (Liu et al. 2026).
In humans, only one small double blind trial from 1981 showed increased antibody responses and higher leukocyte inhibition factor activity after DMG intake, suggesting enhanced immunity (Graber et al. 1981). However, this finding has never been replicated, and no modern clinical trials confirm immune boosting effects. Thus, while animal data are supportive, evidence in humans remains preliminary and inconclusive.

Dimethylglycine and oncology

Evidence on an influence of DMG in cancer studies is mixed and appears highly context dependent. Large cohort studies report no association between circulating DMG levels and cancer risk, including two nested case-control studies on pancreatic cancer in Asian populations (Huang et al. 2017) and colorectal cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort (Nitter et al. 2014). In contrast, DMG is elevated in several tumor related settings. Urinary DMG was significantly increased in hepatocellular carcinoma (HCC) in a West African cohort and correlated with clinical stage, suggesting potential utility for surveillance (Ladep et al. 2014). Higher DMG has also been detected in esophageal tumor margins (Jianyong et al. 2017) and in fecal samples of colorectal cancer patients (Lin et al. 2016). One population study in individuals with hypertension reported a positive association between serum DMG and overall cancer risk (Zhang et al. 2023).
These divergent findings may reflect differences in population characteristics, disease stage, or sample types. Mechanistically, DMG participates in one carbon metabolism, supporting amino acid and nucleotide synthesis and cellular methylation reactions. When primary dietary methyl donors (choline, methionine, creatine) are insufficient, DMG can act as a secondary methyl donor. Because dysregulated methylation is a hallmark of cancer, shifts in DMG availability may influence tumor biology, although causal evidence is still limited (Zhang et al. 2023).

Dimethylglycine and 5P medicine

DMG is a molecular entry point into individualized, systems-oriented healthcare aligning with the principles of 5P medicine. Altered DMG levels have been linked to cardiometabolic risk and conditions such as myocardial infarction, diabetes (Magnusson et al. 2015), and metabolic dysfunction-associated liver disease (Pacana et al. 2015). Hence, DMG is positioned as potential early predictive marker even though causal evidence is still missing. Disease specific and matrix dependent patterns, like elevated urinary DMG in hepatocellular carcinoma (Ladep et al. 2014), increased fecal DMG in colorectal cancer (Lin et al. 2016), or reduced maternal DMG in pregnancies with congenital heart disease (Xie et al. 2025), highlight its relevance for more precise diagnostics or stratification. Because DMG metabolism is shaped by nutrition (Yao et al. 2022; Cupp et al. 2003), liver and kidney function (McGregor et al. 2001), genetics (McAndrew et al. 2008), and microbiome (Cimmino et al. 2025; Liu et al. 2024; Wang et al. 2022), individual interpretation is essential. Overall, DMG reflects interconnected pathways such as methylation, mitochondrial function, and cardiometabolic regulation, making it a valuable indicator of metabolic stress and early dysregulation.

References

Cimmino, F. et al.: Autism spectrum disorders and nutritional interventions: dimethylglycine and B-vitamins effects on behaviour, inflammation, microbiota and mitochondria in liver and brain synapses (2025) Biomedicine & pharmacotherapy | https://doi.org/10.1016/j.biopha.2025.118477.

Cupp, M.J. et al.: Dimethylglycine (N,N-Dimethylglycine) (2003) | https://doi.org/10.1007/978-1-59259-303-3_9.

Devi, A.P. et al.: Dimethylglycine as a Potent Modulator of Catalase Stability and Activity in Alzheimer’s Disease (2026) Biophysica | https://doi.org/10.3390/biophysica6010002.

Gong, X. et al.: Maternal Plasma Betaine in Middle Pregnancy Was Associated with Decreased Risk of GDM in Twin Pregnancy: A Cohort Study (2021) Diabetes, metabolic syndrome and obesity : targets and therapy | https://doi.org/10.2147/DMSO.S312334.

Graber, C.D. et al.: Immunomodulating properties of dimethylglycine in humans (1981) The Journal of Infectious Diseases | https://doi.org/10.1093/infdis/143.1.101.

Huang, J. et al.: Abstract 2273: Serum choline, methionine, betaine, dimethylglycine, and trimethylamine-N-oxide in relation to pancreatic cancer risk in two nested case-control studies in Asian populations (2017) Cancer Research | https://doi.org/10.1158/1538-7445.AM2017-2273.

Jianyong, Z. et al.: Rapid discrimination of human oesophageal squamous cell carcinoma by mass spectrometry based on differences in amino acid metabolism (2017) Scientific Reports | https://doi.org/10.1038/s41598-017-03375-8.

Kern, J.K. et al.: Effectiveness of N,N-dimethylglycine in autism and pervasive developmental disorder (2001) Journal of child neurology | https://doi.org/10.1177/088307380101600303.

Ladep, N.G. et al.: Discovery and validation of urinary metabotypes for the diagnosis of hepatocellular carcinoma in West Africans (2014) Hepatology | https://doi.org/10.1002/hep.27264.

Lever, M. et al.: Betaine and secondary events in an acute coronary syndrome cohort (2012) PLOS ONE | https://doi.org/10.1371/journal.pone.0037883.

Lin, Y. et al.: NMR-based fecal metabolomics fingerprinting as predictors of earlier diagnosis in patients with colorectal cancer (2016) Oncotarget | https://doi.org/10.18632/oncotarget.8762.

Liu, F.-T. et al.: Endometrial microbiota-dimethylglycine-Treg cell axis affects endometrial receptivity in recurrent implantation failure (2026) Science China. Life sciences | https://doi.org/10.1007/s11427-025-3138-x.

Liu, Y. et al.: Changes in Isoleucine, Sarcosine, and Dimethylglycine During OGTT as Risk Factors for Diabetes (2024) The Journal of Clinical Endocrinology & Metabolism | https://doi.org/10.1210/clinem/dgae018.

Magnusson, M. et al.: Dimethylglycine Deficiency and the Development of Diabetes (2015) Diabetes | https://doi.org/10.2337/db14-1863.

McAndrew, R.P. et al.: Molecular basis of dimethylglycine dehydrogenase deficiency associated with pathogenic variant H109R (2008) Journal of inherited metabolic disease | https://doi.org/10.1007/s10545-008-0999-2.

McGregor, D.O. et al.: Dimethylglycine accumulates in uremia and predicts elevated plasma homocysteine concentrations (2001) Kidney International | https://doi.org/10.1046/j.1523-1755.2001.00743.x.

Nitter, M. et al.: Plasma methionine, choline, betaine, and dimethylglycine in relation to colorectal cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC) (2014) Annals of oncology : official journal of the European Society for Medical Oncology | https://doi.org/10.1093/annonc/mdu185.

Pacana, T. et al.: Dysregulated Hepatic Methionine Metabolism Drives Homocysteine Elevation in Diet-Induced Nonalcoholic Fatty Liver Disease (2015) PLOS ONE | https://doi.org/10.1371/journal.pone.0136822.

Reap, E.A. et al.: Stimulation of the immune response by dimethylglycine, a nontoxic metabolite (1990) The Journal of laboratory and clinical medicine | PMID: 1691258.

Shahril Agus, S.A. et al.: Effect of N, N-Dimethylglycine (DMG) Supplementation on Haematological Parameters and Frequency of CD4+ and CD8+ T Cells in Cats Post-vaccination (2025) Pertanika Journal of Tropical Agricultural Science | https://doi.org/10.47836/pjtas.48.3.12.

Svingen, G.F.T. et al.: Plasma dimethylglycine and risk of incident acute myocardial infarction in patients with stable angina pectoris (2013) Arteriosclerosis, thrombosis, and vascular biology | https://doi.org/10.1161/ATVBAHA.113.301714.

Svingen, G.F.T. et al.: Elevated plasma dimethylglycine is a risk marker of mortality in patients with coronary heart disease (2015) European Journal of Preventive Cardiology | https://doi.org/10.1177/2047487314529351.

Wang, Z. et al.: Heat stress-induced intestinal barrier damage and dimethylglycine alleviates via improving the metabolism function of microbiota gut brain axis (2022) Ecotoxicology and Environmental Safety | https://doi.org/10.1016/j.ecoenv.2022.114053.

Wolfsegger, T. et al.: N,N-Dimethylglycine in patients with progressive multiple sclerosis: result of a pilot double-blind, placebo, controlled randomized clinical trial (2021) Neurological research and practice | https://doi.org/10.1186/s42466-021-00126-z.

Xie, B. et al.: Identification of serum N,N-dimethylglycine as a potential biomarker for prenatal diagnosis of congenital heart disease using 1HNMR and UPLC-MS/MS metabonomics (2025) Analytical and Bioanalytical Chemistry | https://doi.org/10.1007/s00216-025-06084-8.

Yao, H. et al.: Effects of dietary dimethylglycine supplementation on laying performance, egg quality, and tissue index of hens during late laying period (2022) Poultry Science | https://doi.org/10.1016/j.psj.2021.101610.

Zhang, H. et al.: The association of serum dimethylglycine with the risk of cancer (2023) | https://doi.org/10.21203/rs.3.rs-3371052/v1.

• History & Evolution
• Biosynthesis & dietary uptake
• Fructose and the microbiome
• Fructose and cardiometabolism
• Fructose and neurology
• Fructose and oncology
• Fructose and 5P medicine
• References

History & Evolution

1847: isolation | 1880s: stereochemistry | 1956: clinics | 1970s: industry | 2004: health concerns

Fructose entered the world of chemistry in 1847, when Augustin Pierre Dubrunfaut isolated it from sucrose (Rödel 1974). For decades it was also called levulose because its solution rotates plane polarized light to the left. In the 1880s-1890s, Emil Fischer revolutionized carbohydrate science by mapping relationships among glucose, fructose, and mannose and introducing Fischer projections, which fixed the stereochemical language we still use (Horton 2013). A clinical turn came in 1956-1963, when physicians characterized hereditary fructose intolerance (HFI), a recessive disorder of aldolase B, and explained the syndrome’s biochemistry (Chambers et al. 1956). The role of fructose in the food industry became prominent in the early 1970s after work on glucose isomerase brought about commercial high fructose corn syrup (HFCS) (Horton 2013). In 2004, after decades of increasing intake of HFCS sweetened beverages, their link with U.S. obesity trends was brought to light (Bray et al. 2004), catalyzing an enduring debate about the ethics of adding fructose and other sweeteners to our foods.

When it does make it into our plates and glasses, fructose has advantageous properties compared to glucose: it is generally ~1.2 to 1.7 times sweeter than sucrose (Fontvieille et al. 1989), thus requiring less fructose for the same sensation. However, this advantage is most pronounced in cold matrices. Classic psychophysical work shows fructose’s relative sweetness climbs at low temperatures and drops when heated (Green et al. 2015). The mechanistic study of the fate of fructose in the body further added fuel to the polemic, with the discovery that fructose is a substrate for hepatic lipogenesis.

Biosynthesis vs. dietary uptake

Fructose is a naturally occurring monosaccharide that can be produced endogenously in humans via the polyol pathway, where glucose is converted to sorbitol and then to fructose (Delannoy et al. 2025). In diet, fructose is abundant in fruits, honey, and some vegetables and is specifically present in processed foods as part of sucrose or in the form of HFCS (Flores Monar et al. 2025).
After ingestion, fructose is absorbed in the small intestine primarily via facilitated diffusion through the GLUT5 transporter located on the apical membrane of enterocytes (Taylor et al. 2021). Once inside the enterocyte, fructose is transported across the basolateral membrane into the portal circulation by GLUT2 (Jung et al. 2022). From the portal vein, fructose is delivered mainly to the liver, which serves as the central site of fructose metabolism.

Within the liver, fructose undergoes rapid phosphorylation by ketohexokinase (KHK) to form fructose-1-phosphate (F1P) (Flores Monar et al. 2025; Jung et al. 2022). Consequently, fructose-1-phosphate is cleaved by aldolase B into dihydroxyacetone phosphate (DHAP) and glyceraldehyde, intermediates which enter glycolysis or gluconeogenesis. A substantial fraction of these metabolites is directed toward de novo lipogenesis, contributing to triglyceride synthesis. A low flux of intermediates is directed towards the hexosamine biosynthetic pathway, supporting the formation of sugar nucleotides required for proper protein glycosylation. This essential regulatory process controls protein folding, receptor signaling and immune function (Paneque et al. 2023).

Unlike glucose, fructose metabolism does not stimulate insulin secretion or leptin production, which has implications for energy homeostasis and lipid accumulation (Taylor et al. 2021; Flores Monar et al. 2025). Under normal physiological conditions, almost all absorbed fructose is processed during first-pass metabolism, so urinary excretion is negligible (Macdonald et al. 1978; Jung et al. 2022). However, in cases of fructose malabsorption or hereditary fructose intolerance, unmetabolized fructose may appear in urine and cause gastrointestinal symptoms (Simões et al. 2025; Febbraio et al. 2021).

Fructose and the microbiome

When the small intestine’s first-pass capacity to absorb fructose is exceeded, particularly with high loads of free fructose from HFCS-sweetened foods and beverages, unabsorbed fructose reaches the distal gut and becomes directly available to the microbiota (Jung et al. 2022; Febbraio et al. 2021). Chronic high free-fructose intake shifts microbial communities toward simple-sugar-degrading bacteria and away from key fiber-degrading taxa (Simões et al. 2025). By contrast, fructose in whole fruits is delivered within a fiber-rich matrix that slows its arrival in the distal gut and supports saccharolytic commensal bacteria (Simões et al. 2025; Febbraio et al. 2021). Microbial fermentation of unabsorbed fructose generates gases typical of malabsorption and produces short-chain fatty acids (SCFAs) (Febbraio et al. 2021; Simões et al. 2025). Among SCFAs, acetate is particularly relevant because it enters the portal circulation and serves as a substrate for hepatic acetyl-CoA, thereby fueling de novo lipogenesis and linking microbial activity to liver fat accumulation.

In contrast, butyrate generally exerts anti-inflammatory and barrier-supporting effects in the colon. High-fructose diets also shift microbial metabolism toward cardiometabolic risk-associated metabolites such as trimethylamine N-oxide (TMAO) in mice (Jung et al. 2022).
In the same luminal environment, unabsorbed fructose can non-enzymatically be combined to dietary peptides and incretins through fructosylation. These fructose-derived advanced glycation end products (FruAGE) activate receptors for advanced glycation end products (RAGE) and amplify inflammatory signaling locally and systemically (DeChristopher 2024). As dysbiosis progresses, SCFA imbalance and microbial activity reduce gut barrier integrity, leading to a “leaky gut” with increased permeability. Translocated lipopolysaccharide (LPS) and other microbial components then reach the liver, activate toll-like receptor (TLR) 4 on Kupffer cells, trigger tumor-necrosis factor (TNF)-driven inflammatory cascades, and propagate hepatic injury and metabolic dysfunction (Febbraio et al. 2021; Simões et al. 2025). Together, these processes link chronic high fructose exposure, microbial dysbiosis, impaired barrier function, and immune activation to a broad spectrum of metabolic and inflammatory diseases across the gut-organ axis.

Fructose and cardiometabolism

Over the past four decades, HFCS in processed foods and beverages has markedly increased dietary fructose exposure beyond normal physiological handling. This change parallels rising rates of obesity, type 2 diabetes, metabolic dysfunction-associated steatotic liver disease (MASLD) and cardiovascular disease, implicating industrial sweeteners as important contributors to cardiometabolic risk (Febbraio et al. 2021; DeChristopher 2024; Jung et al. 2022). Unlike fructose in whole fruits and honey, HFCS provides large amounts of free fructose in liquid form, bypassing the fiber matrix that normally slows absorption (Flores Monar et al. 2025; Jung et al. 2022). When intake is high, small-intestinal first-pass clearance saturates and excess fructose reaches the liver in substantial amounts (Jung et al. 2022; Febbraio et al. 2021).
In the liver, KHK-driven fructose phosphorylation bypasses phosphofructokinase and rapidly generates fructose-1-phosphate. This unregulated flux promotes lipogenesis while depleting adenosine-triphosphate (ATP) and trapping inorganic phosphate, activating adenosine-monophosphate (AMP) deaminase (AMPD) and increasing uric acid production. Elevated uric acid together with reactive oxygen species has been shown to promote endothelial dysfunction, inflammation and a pro-steatotic, pro-inflammatory hepatic milieu in animal models (Febbraio et al. 2021; Jung et al. 2022; Flores Monar et al. 2025). At the same time, fructose weakly stimulates insulin and satiety hormones such as glucagon-like peptide (GLP)-1, lessening satiation and fostering increased energy intake and positive energy balance (Flores Monar et al. 2025).

These mechanisms collectively drive hepatic steatosis, dyslipidemia and insulin resistance, core features of metabolic syndrome (DeChristopher 2024; Febbraio et al. 2021). Epidemiological data further link chronic high-fructose intake to hypertension, hyperuricemia and elevated cardiovascular risk independently of body mass index (Febbraio et al. 2021; Jung et al. 2022; DeChristopher 2024). Although the relative contribution of HFCS versus sucrose remains debated (Feinman et al. 2013) decades of excessive fructose consumption, particularly from sugar-sweetened beverages, represent a key factor in the growing burden of cardiometabolic disease (DeChristopher 2024; Flores Monar et al. 2025; Jung et al. 2022).

Fructose and neurology

In the hypothalamus of mice, chronic fructose exposure alters energy sensing and shifts neuropeptide signaling toward orexigenic, or appetite-stimulating pathways, thus directly linking high-fructose intake to positive energy balance and the metabolic changes in humans described above (Payant et al. 2021). The developing brain is especially vulnerable: in mice, early-life high-fructose diets impair microglial phagocytic activity, and reduce clearance of synapses and apoptotic neurons disrupts circuit refinement, leading to lasting deficits in learning and emotional regulation (Zelenka 2025). In the mature rodent brain, the hippocampus emerges as a key target of fructose-induced injury, with early mitochondrial dysfunction, oxidative stress, neuroinflammation, and impaired neuronal insulin signaling. Some changes persist after dietary normalization and result in long-term memory impairment (Flores Monar et al. 2025; DeChristopher 2024).

Within this framework, Alzheimer’s disease (AD) has been hypothesized to be driven, at least in part, by overactivation of cerebral fructose metabolism, largely through endogenous fructose production via the polyol pathway. In neuronal models, fructose metabolism via KHK C causes energy depletion and uric acid generation, leading to mitochondrial oxidative stress, impaired mitophagy, reduced oxidative phosphorylation, cerebral glucose hypometabolism, and insulin resistance. Chronic energy failure promotes uric acid-driven neuroinflammation via NF-κB and TLR4 in rodents and humans (Shao et al. 2016). Disruption of plasticity-related pathways causes synaptic dysfunction in rodents. Amyloid/tau pathology might come secondary to oxidative stress and impaired protein repair also in humans. These processes might ultimately lead to neuronal dysfunction (Johnson et al. 2020).

High sugar and high-fructose intake, high glycemic load, excess salt, alcohol, obesity, diabetes, aging, and traumatic brain injury may all enhance endogenous fructose production by inducing aldose reductase and fructokinase in the brain. Consistently, the levels of sorbitol and fructose, as well as the activity of AMPD2 and production of ammonia are all increased in the brains of patients with AD. In both animal and human studies, high sugar intake was linked to cognitive decline, hippocampal inflammation, and reduced brain volume (Johnson et al. 2020; Flores Monar et al. 2025).

Fructose and oncology

High fructose intake has been associated with an increased risk of several cancers, with particularly strong epidemiological links to pancreatic cancer. Many tumor types overexpress the fructose transporter GLUT5, and elevated GLUT5 together with KHK expression indicates active fructose utilization by cancer cells (Jung et al. 2022; Febbraio et al. 2021). At the level of host tissues and the tumor microenvironment, fructose acts prominently along the gut-liver axis. Mouse models revealed that dietary fructose enhances epithelial cell survival and drives villus elongation in the intestine. The expanded absorptive surface area leads to increasing nutrient uptake and adiposity in the context of a Western diet (Taylor et al. 2021; Febbraio et al. 2021).
In the liver, fructose-induced fibrosis and inflammation promote an immunosuppressive milieu that weakens CD8⁺ T cell-mediated immunosurveillance and favors hepatocellular carcinoma development (Febbraio et al. 2021).

Fructose can also drive tumor growth through cell non-autonomous fructolysis in animal models: KHK-C-expressing hepatocytes convert fructose into circulating lysophosphatidylcholines (LPCs) that are taken up by cancer cells to support phosphatidylcholine synthesis and membrane biogenesis, with fructose supplementation enhancing tumor growth even without overt weight gain or insulin resistance (Fowle-Grider et al. 2024). In parallel, many cancers undergo intrinsic metabolic reprogramming to exploit fructose. Under glucose-poor, fructose-rich conditions, cancer cells can use fructose as an alternative carbon source to sustain central carbon metabolism, de novo lipid and nucleic acid synthesis, and the Warburg effect, with aberrant fructose utilization activating oncogenic mechanistic target of rapamycin complex (mTORC) 1 signaling and suppressing anti-tumor immunity (Ting 2024; Zhao et al. 2025). Different tumors deploy these pathways in distinct ways, for example, pancreatic cancer channels fructose primarily into nucleotide synthesis via the non-oxidative pentose phosphate pathway, whereas hepatocellular carcinoma may rely on acetate derived from microbial fructose metabolism. Inhibition of fructose metabolism in such fructose-adapted cancer cells reduces malignancy (Ting 2024).

Fructose and 5P medicine

High intake of HFCS over long periods is a major, yet modifiable, risk factor for metabolic, neurological, and oncological conditions (Febbraio et al. 2021; Jung et al. 2022; DeChristopher 2024). Because diet is a central driver, fructose is an instructive example of how individuals can participate in shaping their health. Fructose from whole fruits, delivered with fiber and antioxidants, shows neutral or beneficial effects on metabolic biomarkers, whereas fructose from sugar-sweetened beverages and juices correlates with higher inflammatory markers, dyslipidemia, and intrahepatic fat (DeChristopher 2024; Jung et al. 2022). Experimental data indicate that HFCS-like, fructose-rich sweeteners induce reward-related and metabolic disturbances at lower exposure than sucrose, suggesting stronger reinforcing and pathological effects when fructose/glucose ratios are very high (DeChristopher 2024; Levy et al. 2015). Nonetheless, the most robust strategy for obesity, diabetes, and metabolic syndrome remains a general reduction of total carbohydrate intake, not merely swapping fructose for glucose (Feinman et al. 2013).

When symptoms are present precision medicine gains importance. In inflammatory bowel disease and fructose malabsorption, several strategies improve gastrointestinal symptoms and breath hydrogen in clinical studies. These include fructose-restricted and low-FODMAP diets, microbiota-directed therapies (probiotics, fecal microbiota transplantation), and oral xylose isomerase, which converts fructose to glucose (Simões et al. 2025). Regarding the gut-liver axis, KHK inhibitors, which are tested in clinical trials, lower de novo lipogenesis and improve metabolic markers (Jung et al. 2022). Epidemiological data also suggests that targeting brain KHK C or AMPD2, together with uric acid-lowering strategies, could enable to reduce the risk of dementia in the population (Johnson et al. 2020).

References

Bray, G.A. et al.: Consumption of high-fructose corn syrup in beverages may play a role in the epidemic of obesity (2004) The American Journal of Clinical Nutrition | https://doi.org/10.1093/ajcn/79.4.537.

Chambers, R.A. et al.: Idiosyncrasy to fructose (1956) Lancet (London, England) | https://doi.org/10.1016/S0140-6736(56)92196-1.

DeChristopher, L.R.: 40 years of adding more fructose to high fructose corn syrup than is safe, through the lens of malabsorption and altered gut health-gateways to chronic disease (2024) Nutrition Journal | https://doi.org/10.1186/s12937-024-00919-3.

Delannoy, P. et al.: Aldose reductase, fructose and fat production in the liver (2025) Biochemical Journal | https://doi.org/10.1042/BCJ20240748.

Febbraio, M.A. et al.: “Sweet death”: Fructose as a metabolic toxin that targets the gut-liver axis (2021) Cell Metabolism | https://doi.org/10.1016/j.cmet.2021.09.004.

Feinman, R.D. et al.: Fructose in perspective (2013) Nutrition & Metabolism | https://doi.org/10.1186/1743-7075-10-45.

Flores Monar, G.V. et al.: Mindful Eating: A Deep Insight Into Fructose Metabolism and Its Effects on Appetite Regulation and Brain Function (2025) Journal of Nutrition and Metabolism | https://doi.org/10.1155/jnme/5571686.

Fontvieille, A.M. et al.: Relative sweetness of fructose compared with sucrose in healthy and diabetic subjects (1989) Diabetes care | https://doi.org/10.2337/diacare.12.7.481.

Fowle-Grider, R. et al.: Dietary fructose enhances tumour growth indirectly via interorgan lipid transfer (2024) Nature | https://doi.org/10.1038/s41586-024-08258-3.

Green, B.G. et al.: Temperature Affects Human Sweet Taste via At Least Two Mechanisms (2015) Chemical senses | https://doi.org/10.1093/chemse/bjv021.

Horton, D.: Carbohydrate Nomenclature | https://doi.org/10.1007/978-3-642-16712-6_67.

Johnson, R.J. et al.: Cerebral Fructose Metabolism as a Potential Mechanism Driving Alzheimer’s Disease (2020) Frontiers in Aging Neuroscience | https://doi.org/10.3389/fnagi.2020.560865.

Jung, S. et al.: Dietary Fructose and Fructose-Induced Pathologies (2022) Annual Review of Nutrition | https://doi.org/10.1146/annurev-nutr-062220-025831.

Levy, A. et al.: Fructose:glucose ratios–a study of sugar self-administration and associated neural and physiological responses in the rat (2015) Nutrients | https://doi.org/10.3390/nu7053869.

Macdonald, I. et al.: Some effects, in man, of varying the load of glucose, sucrose, fructose, or sorbitol on various metabolites in blood (1978) The American Journal of Clinical Nutrition | https://doi.org/10.1093/ajcn/31.8.1305.

Paneque, A. et al.: The Hexosamine Biosynthesis Pathway: Regulation and Function (2023) Genes | https://doi.org/10.3390/genes14040933.

Payant, M.A. et al.: Neural mechanisms underlying the role of fructose in overfeeding (2021) Neuroscience and biobehavioral reviews | https://doi.org/10.1016/j.neubiorev.2021.06.034.

Rödel, W.: J. S. Fruton: Molecules and Life – Historical Essays on the Interplay of Chemistry and Biology. 579 Seiten. Wiley‐Interscience, New York, London, Sydney, Toronto 1972. Preis: 8,95 £ (1974) Food / Nahrung | https://doi.org/10.1002/food.19740180423.

Shao, X. et al.: Uric Acid Induces Cognitive Dysfunction through Hippocampal Inflammation in Rodents and Humans (2016) The Journal of neuroscience: the official journal of the Society for Neuroscience | https://doi.org/10.1523/JNEUROSCI.1480-16.2016.

Simões, C.D. et al.: Fructose Malabsorption, Gut Microbiota and Clinical Consequences: A Narrative Review of the Current Evidence (2025) Life | https://doi.org/10.3390/life15111720.

Taylor, S.R. et al.: Dietary fructose improves intestinal cell survival and nutrient absorption (2021) Nature | https://doi.org/10.1038/s41586-021-03827-2.

Ting, K.K.Y.: Fructose-induced metabolic reprogramming of cancer cells (2024) Frontiers in Immunology | https://doi.org/10.3389/fimmu.2024.1375461.

Zelenka, L.: Fructose-induced anxiety (2025) Nature Neuroscience | https://doi.org/10.1038/s41593-025-02019-9.

Zhao, Q. et al.: Targeting fructose metabolism for cancer therapy (2025) Cancer Letters | https://doi.org/10.1016/j.canlet.2025.217914.

• Foreword
• How metabolomics is improving healthcare: 6 must-read studies from 2025
• Hepatology: functional detox capacity beyond fibrosis
• Cardiometabolic health: anticipating disease before symptoms
• Cohorts and mGWAS: from genetic signals to functional meaning
• Oncology: detecting cancer before its manifestation
• Neuropsychiatry and microbiome: modulating behavior through microbial metabolism
• Nanomedicine: designing safer therapies through metabolomics & lipidomics
• The next steps for metabolomics in 5P medicine

Foreword by Alice Limonciel, Chief Scientific Officer at biocrates

After decades of development, we are on the cusp of integrating metabolomics into medical practice. Numerous examples already exist in clinical settings, the result of the dedicated labor of passionate scientists and clinicians who recognized the potential of this omic and applied it across all areas of medicine. However, the broad adoption of metabolomics on a scale comparable to what we now see with genomics requires the development of robust, transferable, and scalable technology, which has been the mission of biocrates for the past 20 years.

In 2025, we chose to showcase the wide-ranging potential of metabolomics for all aspects of 5P medicine, from preventing chronic disease in a single individual through personalized strategies to enabling multiomic analyses in large cohort studies.

For this article, Franziska Hörburger selected six studies published in 2025 by biocrates’ community of users. These examples pave the way for the imminent implementation of metabolomics beyond the research lab and into clinical practice and our everyday lives. They span multiple regions, therapeutic areas, and dimensions of the future implementation of metabolomics in medicine and drug development.
These scientists are part of a community of early adopters of metabolomics, a technology that will transform how we understand and practice medicine. May this article inspire you and your team to join this community in 2026!

How metabolomics is improving healthcare: 6 must read studies from 2025

5P medicine – preventive, predictive, precision, population-based, and participatory – represents a paradigm shift in healthcare. It moves away from reactive treatment toward proactive, patient-centric strategies built on molecular insights. At its core, 5P medicine leverages high-quality standardized technologies to capture biology in unprecedented detail. It enables clinicians and researchers to predict disease risk, personalize interventions, while engaging patients in their individual health journey and at the population scale.

Among the molecular technologies shaping modern medicine, metabolomics stands out. While genomics can predict disease risk, it offers only a static view, like a snapshot of predisposition. Metabolomics, in contrast, captures the biochemical fingerprints of life, reflecting the dynamic interplay of genes, environment, lifestyle, microbiome, and pharmacological influences. This real-time perspective makes metabolomics indispensable for understanding complex chronic diseases, where genetic information alone cannot explain onset, progression, or therapeutic response.

Figure 1: Molecular health beyond genetic predisposition

When metabolomics is combined with other omics technologies, such as genomics or proteomics, the picture becomes even richer. Proteomics adds information about enzyme abundance and signaling networks, complementing metabolomics’ readout of pathway activity and flux. Together, these layers create a detailed system-level view of health and pathology, connecting genetic predisposition to molecular function and clinical phenotype. This integrated approach transforms omics from isolated data streams into actionable insights, connecting molecular complexity and medical decision-making based on the 5P concept.

Applying metabolomics within the 5P framework can be summarized in three steps. First, screen samples using broad metabolomics and lipidomics profiling. Second, leverage data to uncover biological meaning. This involves interpreting metabolite patterns, sums, and ratios, and linking them to pathways and literature. Third, translate insights into solutions: providing predictive biomarkers, metabotypes, risk scores, and decision-support tools that transform medicine from reactive to proactive.

Figure 2: Workflow for applying metabolomics in the 5P framework

This vision is not theoretical; it is already happening. Across biological matrices, continents and disciplines, researchers and clinicians are using biocrates’ technology to deliver actionable insights in fields as varied as hepatology, oncology, neuropsychiatry, cardiometabolic health, population studies, and nanomedicine.

Here we review six publications from 2025 in high-impact journals that illustrate how one standardized platform can drive breakthroughs aligned with the principles of 5P medicine.

Hepatology: functional detox capacity beyond fibrosis

Sugimoto et al.: Hepatic stellate cells control liver zonation, size and functions via R-spondin 3. Nature (2025), 640(8059):752–761 | https://doi.org/10.1038/s41586-025-08677-w Figure under creative commons license CC BY 4.0.

Sugimoto and colleagues uncovered how hepatic stellate cells orchestrate liver zonation and detoxification through the signaling molecule R-spondin 3 (RSPO3), a key regulator of the WNT pathway. When RSPO3 is lost, hepatocyte zonation collapses and regenerative capacity declines. Beyond structural changes, RSPO3 profoundly influences detoxification by modulating cytochrome P450 activity, which in turn alters circulating metabolite profiles. Liver tissue of RSPO3-deficient mice featured striking shifts in bile acid composition, particularly taurocholic, tauromuricholic, and taurochenodeoxycholic acids.

Furthermore, changes in steroid metabolism, lipid oxidation, and xenobiotic accumulation have been revealed. These metabolomic signatures predict functional liver capacity, drug metabolism potential, and ultimately toxicity risk. By identifying RSPO3 as both a prognostic and mechanistic marker, this work opens the door to early intervention, personalized risk stratification, and tailored therapeutic approaches for liver fibrosis, particularly in alcoholic liver disease and metabolic dysfunction associated liver disease.

Cardiometabolic health: anticipating disease before symptoms

Zieleniewska et al.: Preclinical Atherosclerosis and Prediabetes: A Cross-Sectional Metabolic Assessment In Apparently Healthy Population. Cardiovascular Diabetology (2025), 24(1), 280 | https://doi.org/10.1186/s12933-025-02841-2 Figure under creative commons license CC BY-NC-ND 4.0.

Cardiovascular disease and diabetes often develop silently over years, making early detection critical. The metabolic foundation of preclinical atherosclerosis compared to prediabetes was explored in 447 participants from the Bialystok PLUS cohort.

The analysis uncovered distinct and shared metabolic signatures in plasma for both conditions. Prediabetes exerted a broader impact on amino acid metabolism, lipid signaling and enzymatic activities than atherosclerosis. Glutamic acid, lactic acid, and alanine were strongly associated with prediabetes, indicating dysglycemia. Atherosclerosis was linked to lipid remodeling patterns captured by MetaboINDICATORs, including the ratio of polyunsaturated (PUFA)-lysophosphatidylcholines versus saturated fatty acids, the sum of steroid hormones, and cholesteryl ester (CE) classes such as monounsaturated CEs and long-chain fatty acids CEs. Trimethylamine N-oxide (TMAO) emerged as a unique link between prediabetes and its interaction with vascular pathology. At the same time, glutaminase activity, assessed through the glutamate/glutamine ratio, stood out as a robust shared predictor of both conditions. Metabolite set enrichment analysis observed converging disturbances in glutathione and folate metabolism, mitochondrial function, redox regulation and inflammation.

Cohorts and mGWAS: from genetic signals to functional meaning

Kodate et al.: Simulating metabolic pathways to enhance interpretations of metabolome genome-wide association studies. Scientific Reports (2025), 15(1), 17035 | https://doi.org/10.1038/s41598-025-01634-7 Figure under creative commons license CC BY-NC-ND 4.0.

Metabolome-genome-wide association studies (mGWAS) link genetic variation to metabolite concentrations in large cohorts. It provides great predictive power of risk models and enables rational intervention based on individual metabolic architecture. However, this powerful approach has some limitations: observed associations may reflect indirect effects through unmeasured metabolites, and the biological significance of many variants remains uncertain. To overcome these challenges, Kodate and colleagues combined mGWAS with mechanistic metabolic simulations, creating a framework that moves beyond correlation to causation.

By systematically adjusting enzyme reaction rates to mimic genetic variants, the team simulated their impact on plasma metabolite levels and validated most variant-metabolite pairs identified by mGWAS. For example, homocysteine was confirmed as a metabolite strongly influenced by methylenetetrahydrofolate reductase (MTHFR) activity. Both mGWAS and simulation agreed that reduced MTHFR activity increases homocysteine levels, reinforcing its role in folate and methionine metabolism. These simulations also revealed additional fluctuations that mGWAS had missed, suggesting that some associations could gain significance with larger sample sizes. Importantly, the study categorized enzymes into three tiers based on their influence on metabolite concentrations, highlighting variants with minimal biological impact and prioritizing those with strong functional relevance. This distinction is critical for guiding preventive strategies and therapeutic development.

Oncology: detecting cancer before its manifestation

Schulze et al.: Metabolomic liquid biopsy dynamics predict early-stage HCC and actionable candidates of human hepatocarcinogenesis. JHEP Reports (2025), 7(5):101340 | https://doi.org/10.1016/j.jhepr.2025.101340. Figure under creative commons license CC BY 4.0.

Hepatocellular carcinoma (HCC) develops through progressive metabolic reprogramming that begins long before tumors become radiologically or clinically detectable. In a global cohort of 654 patients, serum metabolome profiling captured these early, system-level alterations and predicted malignant transformation before overt tumor manifestation.

Across chronic liver disease, cirrhosis, initial HCC, and advanced HCC, amino acid-, lipid-, and nucleotide-related pathways were systematically deregulated, with aspartic acid, glutamic acid, taurine, and hypoxanthine emerging as key markers. In a phase II biomarker case-control study, a blood-based metabolite signature achieved an area under the curve (AUC) of 94% for distinguishing early-stage HCC from cirrhotic controls, with independent validation in an external cohort. Multiomics integration links these circulating markers to enzymatic nodes such as RRM2, GMPS, and BCAT1 – targets for precision oncology. By providing a validated, minimal-invasive liquid biopsy that outperforms current surveillance tools, serum metabolomics enables predictive identification of cancer risk, chemoprevention strategies, and personalized monitoring.

Neuropsychiatry and microbiome: modulating behavior through microbial metabolism

Park et al.: Gut microbiota and brain-resident CD4+ T cells shape behavioral outcomes in autism spectrum disorder. Nature Communications (2025), 16(1), 1–17 | https://doi.org/10.1038/s41467-025-61544-0 Figure under creative commons license CC BY 4.0.

Autism spectrum disorder (ASD) emerges from complex interactions between neurodevelopment, immune regulation, and the gut microbiome. Metabolites serve as critical messengers of this gut-immune-brain axis, influencing neuroinflammation and neurotransmitter flux. In a recent study, the absence of gut microbiota in male mice ameliorated ASD-associated behaviors and reduced inflammatory brain-resident CD4⁺ T cells, while depletion of these T cells further mitigated neuroinflammation and behavioral abnormalities. Fecal metabolomics in a mouse model of ASD revealed several microbial and metabolic regulators of ASD, particularly those affecting the glutamate/gamma-amino-butyric acid (GABA) ratio and neurotoxic intermediates such as 3-hydroxyglutaric acid.

While GABA levels remained stable, the glutamate/GABA ratio was significantly elevated in ASD mice treated with a broad-spectrum antibiotic cocktail (vancomycin, neomycin, metronidazole), a group that also showed enrichment of Lactobacillus species compared to neurotypical controls. Strikingly, beneficial microbiota, derived from healthy mice or administered as probiotics, reversed this imbalance. These findings underscore how metabolites from live bacteria can drive or mitigate ASD-like behaviors by altering excitatory/inhibitory signaling and immune tone. Ultimately, the study demonstrates that gut microbiota can override genetic predisposition in ASD, highlighting a powerful opportunity for metabolomics-informed interventions that rebalance neuroactive metabolites, suppress neuroinflammation, and improve behavioral outcomes.

Nanomedicine: designing safer therapies through metabolomics & lipidomics

Shaw et al.: Inflammatory disease progression shapes nanoparticle biomolecular corona-mediated immune activation profiles. Nature Communications (2025),16(1), 924 | https://doi.org/10.1038/s41467-025-56210-4 Figure under creative commons license CC BY 4.0.

Polymeric nanoparticles (NPs) are engineered to carry, protect, and deliver bioactive molecules or modulate biological responses. Their biological identity, the biomolecular corona, is not fixed by formulation alone but is dynamically shaped by the host environment. Multiomics analysis showed that, during acute systemic inflammation, plasma proteins, lipids, and metabolites change profoundly.

As a result, nanoparticle coronas are reshaped. They feature elevated levels of clotting factors, inflammatory proteins, cytoskeletal components, and lipids such as phosphatidylcholines, sphingomyelins, lysophosphatidylcholines, and fatty acids. These molecular signatures reflect heightened inflammatory activity and trigger immune pathways like TLR4/MyD88/NF-κB. This activation leads to the release of pro-inflammatory cytokines, including TNFα and IL-6. Together, these findings show how metabolic variability determines nanoparticle-based therapeutic efficacy and toxicity risk. The concept of a “personalized biomolecular corona” underscores the need to design nanomedicines that account for patient-specific metabolic states. Incorporating metabolomic profiling into nanoparticle development helps anticipate immune responses, optimize timing, and improve safety.

The next steps for metabolomics in 5P medicine

What unites these global studies performed in various species and matrices beyond their drive to bring medicine to a higher level, is their use of the metabolomics kit technology developed by biocrates.
Across a wide range of applications, our standardized kits provide the reproducible and quality-controlled methods that enable multiomics integration, cohort comparability, and regulatory-compliant workflows.

While 2025 saw the broad application of our MxP® Quant 500 and MxP® Quant 500 XL kits, 2026 will be the year of the MxP® Quant 1000 kit. Our broadest panel to date, this kit expands quantitative analysis to up to 1,233 metabolites from 49 biochemical classes, showing coverage comparable to untargeted metabolomics approaches, yet with the reproducibility and sensitivity of a targeted workflow.

To follow our next steps, make sure to register for our monthly newsletter.

For additional insights also explore a curated selection of 2025 publications from our Biognosys Group partners, Biognosys and PreOmics.

• How it started | The 5P medicine framework
• How metabolites tell 5P stories
• From mechanisms to applications
• Feedback from the scientific community
• 5P medicine in our podcast | The Metabolomist
• The path ahead

biocrates’ mission is to make metabolomics accessible to facilitate breakthroughs in biomedical research. In 2025, we explored this mission through a fresh lens: that of ‘5P medicine’, which builds on insights drawn from omics to rethink how medicine is practiced. We looked at how the 5P framework brings together preventive, predictive, precision, population-based and participatory medicine to nudge healthcare from reactive to truly proactive and patient-centric.

This approach is especially powerful for complex chronic diseases, where genetics alone can indicate risk but cannot track how this risk evolves with lifestyle, environment, or aging. Metabolomics helps fill that gap.

We started the year hoping to showcase the myriad applications of metabolomics in life science research. What we didn’t expect was the amazing feedback from the scientific community and the new collaborations that presented themselves along the way.

Here, we look back at the milestones of 2025, from launching our most comprehensive kit to date to joining Biognosys Group, alongside a few takeaways from our most impactful activities in 2025.

How it started | The 5P medicine framework

Our starting point was a metabolomist’s view of the 5P medicine overview (see below), which shows how preventive, predictive, precision, population-based and participatory medicine come together in one framework. Each ‘P’ connects to specific research components uniquely supported by metabolomics, including risk scores, biomarkers, stratification, environmental variance and at-home sampling, and we see why metabolomics is such a natural fit for this approach.

Building on that, we launched a five-part blog series written by Alice Limonciel , to look more closely at the role of metabolomics in 5P medicine:

“Preventive medicine” showed how metabolomics detects early metabolic shifts long before symptoms appear, enabling risk stratification, targeted lifestyle interventions and progress monitoring.

“Predictive medicine” focused on building metabolic signatures that forecast disease progression and treatment response, especially when metabolomics is combined with other omics in multiomics studies.

“Precision medicine” explored how quantitative metabolite panels reveal hidden subgroups and enable personalized prognostic/diagnostics and treatment optimization.

“Population-based medicine” scaled up to cohorts and biobanks, connecting metabolomics to exposomics while refining stratification methods through metabotyping, leveraging the power of molecular phenotypes.

“Participatory medicine” looked at at-home sampling and reference ranges for quantitative metabolomics that make medicine accessible to larger parts of the population, often at lower costs and for preventive and monitoring applications.

These articles formed the red thread of our guide to applying metabolomics in 5P medicine. You’ll find more on these themes throughout our articles, webinars, events and conference reports.

How metabolites tell 5P stories

In our “metabolite of the month” series, we look at one molecule each month and ask what it reveals about health. This year, we added a new section focused on the metabolite’s role in 5P medicine. Here are a few examples:

Methylmalonic acid (MMA)
Recognized as the most specific functional marker of vitamin B12 status, MMA holds promise for preventive and predictive medicine, with tight links to mitochondrial function, the gut microbiome and the nervous system.

Itaconic acid
Linking immune response and inflammation across obesity, cardiovascular, neurological and oncologic disease, itaconic acid is a powerful metabolite, acting as an early biomarker of inflammation and therapeutic response for preventive and predictive strategies. It’s also a promising precision target for future interventions.

Bilirubin
More than a waste product, bilirubin is a predictor of disease susceptibility, a target for preventive strategies in oxidative stress-driven disorders and a potential precision marker to guide personalized oncology and immunotherapy decisions.

p-cresol glucuronide (pCG)
Linking gut ecology to renal and systemic outcomes, pCG is measurable non-invasively in urine and blood for predictive and precision stratification, and is modifiable through diet and microbiome interventions to support preventive and participatory care.

From mechanisms to applications

While a single metabolite can play a role in many applications, some stories are better told with multiple protagonists. When using omics to describe the molecular presentation of a disease or the intricate mechanisms that support life, the focus needs to be on how groups of molecules interact. This is what we turned to in the articles below, which look at how metabolomics and lipidomics turn pathway insights into concrete clinical questions.

In “Energy metabolism in cancer – Mechanisms, plasticity and applications”, Gordian Adam, explored how metabolomics helps reveal the mechanisms driving cancer’s metabolic plasticity, from the Warburg effect and oncometabolites like lactate and 2-hydroxyglutarate to redox balance, lipid metabolism and the tumor microenvironment – view key takeaways here.

In “Peroxisome biogenesis disorders and the lipidome”, Franziska Hörburger reported from the International Conference on the Bioscience of Lipids (ICBL) 2025 that “when peroxisomes falter, lipids tell the story.” Early lipid remodeling – involving increased very long-chain fatty acids, decreased plasmalogens and decreased docosahexaenoic acid (DHA)-rich lipids – often precedes inflammation and tissue damage. These profiles are emerging as promising early markers of disease and show how targeted lipidomics can map changes and connect them to clinical phenotypes.

Feedback from the scientific community

Our 5P focus also shaped how we talked about biocrates at conferences, events and webinars.

At the American Society for Mass Spectrometry (ASMS) 2025, we launched the MxP® Quant 1000 kit , our most comprehensive kit to date, and really took the pulse of the community on the 5P concept. From the modular application of metabolomics and lipidomics to the wide-ranging applications of the results through the five pillars of 5P medicine, it all seemed to resonate really well – take a look at the highlights!

At Metabolomics 2025, we saw how strongly the community already lives the 5P medicine concept – from at-home sampling for skin microbiopsies for participatory and preventive care to metabolomics and lipidomics applications driving predictive and precision insights. For us, it confirmed that solutions like the MxP® Quant 1000 kit and insightful data interpretation are exactly what’s needed to bring omics closer to everyday practice.

5P medicine webinar

In October, Sabine Bahn and Oliver Bathe, together with Alice Limonciel, highlighted how metabolomics is transforming neurology and oncology research within the framework of 5P medicine.

Metabolomics India 2025 webinar

Every year, we put the Indian metabolomics community in the spotlight. In this year’s virtual conference, our many speakers highlighted how discoveries made with metabolomics and multiomics will be leveraged by medicine in the near future.

Whether at in-person events or through your feedback after our webinars, you’ve shown us that the 5P vision resonates widely – and that standardized and robust metabolomics solutions are an essential step in the future contribution of metabolomics into everyday medical decision-making.

5P medicine in our podcast | The Metabolomist

Every season of our podcast has a specific focus and this year, of course, our theme was 5P medicine. This allowed us to dive into vastly different fields of research, including plant metabolism, exposomics, microbial metabolism, obesity, software development, method transfer and publication.

Here is a taster of what you’ll hear in our six episodes in 2025:

Cristina Legido-Quigley discusses the relevance of lipidomics in obesity and neurology, monitoring dietary interventions with lipidomics and the importance of sex differences to precision medicine.

Tomáš Pluskal discusses the beauty and complexity of plant metabolism, sustainable ways to harness these molecules and pathways for medical applications, plus the development of one of the most widely used software tools in metabolomics.

Gary Miller discusses the synergies of exposomics and metabolomics, the next frontier in multiomics integration and how exposomics promises to bolster all 5Ps and the future of medicine.

This episode features exclusive interviews collected by Alice at the 2025 Metabolomics Society conference in Prague, Czech Republic. In these recordings, we discover the small molecules and lipids that make the hearts of ten Metabolomists beat for their research.

Anne Bendt discusses why metabolomics is a disrupting technology in the field of clinical chemistry, the most critical steps needed for its implementation in the clinics and the fascinating lipidomes of bacteria.

María Eugenia Monge discusses why building community matters in science, how to navigate the publishing world when establishing new methods in your lab and the region-specific implications of the implementation of metabolomics in the clinics.

The path ahead

If you’ve been following our activities over the last year, you’ll also know that in 2025 biocrates was acquired by Bruker and joined a new structure within this company: Biognosys Group.

Within this group, we bring expertise in metabolomics, lipidomics and mass spectrometry (MS) workflow standardization that will be paired with the expertise of other partners in proteomics and MS-based technologies. This new horizon promises to further enable the inclusion of metabolomics into broader workflows and applications that will surely support the contribution of omics to the 5P medicine framework.

To follow our next steps, make sure to register for our monthly newsletter.

• History & Evolution
• Biosynthesis & dietary uptake
• Homovanillic acid and the microbiome
• Homovanillic acid and neurology
• Homovanillic acid and oncology
• Homovanillic acid and 5P medicine
• References

History & Evolution

1950s: discovery | 1963: detected in brain and cerebrospinal fluid (Andén, N. et al., 1963) | 1985: metabolic pathway identified (Westerink, B., 1985)

Homovanillic acid (HVA) is a monocarboxylic acid best known for its role as a major dopamine metabolite. The name “homovanillic” refers to the compound’s relationship to vanillic acid, a similar phenolic acid, with the prefix homo- indicating the addition of a methylene group (–CH2–) between the aromatic ring and the carboxyl group.

Interest in HVA began in the 1950s and 60s, when early chromatography studies detected HVA in the urine of mammals and identified it as a marker of dopamine turnover in the brain (Armstrong, M. et al., 1956) (Williams, C. et al., 1960). In 1963, Andén et al. showed that HVA was present in the brain and cerebrospinal fluid (CSF), and introduced a fluorometric method for its detection (Andén, N. et al., 1963). Following the dopamine thread, researchers in the following decades focused on the associations between HVA, dopamine and neurological and psychiatric disorders, including Parkinson’s disease (PD), schizophrenia, depression and, more recently, neuroblastoma. Current research is directed at improving HVA’s diagnostic and prognostic utility as a biomarker in these conditions and refining techniques for its detection in different bodily fluids.

Although primarily found in humans and other mammals, as a phenolic compound HVA is also detectable in some bacteria and plants (Marhuenda-Muñoz, M. et al., 2019). It has been identified in beer, suggesting it can be produced during the fermentation process by microorganisms (Montanari, L. et al., 1999).

Biosynthesis vs. dietary uptake

HVA forms in the body as the final breakdown product of dopamine, through a two-step enzymatic process (Buleandră, M. et al., 2025). First, monoamine oxidase (MAO) breaks down dopamine into 3,4-dihydroxyphenylacetic acid (DOPAC) in dopaminergic neurons in the brain. Then, catechol-O-methyltransferase (COMT) methylates DOPAC and converts it to HVA. An alternative pathway involves COMT acting on dopamine to form 3-methoxytyramine, which MAO then breaks down to form HVA. HVA can also be biosynthesized from homovanillin through the action of the enzyme aldehyde dehydrogenase.

HVA enters the circulation from both the brain and peripheral tissues such as the lungs, liver and skeletal muscle, and is mainly excreted via the kidneys (Lambert, G. et al., 1993).

Clinically, HVA can be measured in urine, plasma and CSF. Because of the distance between site of production and site of measurement, all three matrices can be vulnerable to confounding factors (Amin, F. et al., 1992). In healthy adults, CSF concentrations typically fall in the range of hundreds of nanomoles per liter (Blennow, K. et al., 1993). Reference ranges for urinary HVA are between 0.8 to 35.0 µmol/mmol of creatinine (Newcastle Hospital). Plasma HVA has a half-life of around one hour, and may be affected by diet for several hours (Köhnke, M. et al., 2003). Fasting plasma levels vary by sex and age, and differences between females and males hold regardless of menopause and hormone treatments in transgender patients (Giltay, E. et al., 2005).

HVA may also form through dietary and microbial pathways. Dietary flavonols commonly found in tomatoes, onions, and tea, can lead to significantly elevated levels of urinary HVA (Combet, E. et al., 2011). Likewise, the microbial digestion of hydroxytyrosol (found in olive oil) can also lead to elevated levels of HVA in humans (Tuck, K. et al., 2002).

Homovanillic acid and the microbiome

Research shows links between HVA and gut microbiota. Zhao et al. found that gut bacterial species, including Bifidobacterium longum and Roseburia intestinalis, were depleted in patients with depression (Zhao, M. et al, 2024). Using metabolomics, they showed that B. longum could directly produce HVA in the gut with the substrates of mouse fodder and tyrosine. R. intestinalis promoted the growth of B. longum, in turn increasing HVA levels in the gut. R. faecis and Eubacterium rectale were also strongly correlated with HVA levels.

Another recent metabolomics study by Gątarek et al. investigated the relationship between gut-derived metabolites, including HVA, and Parkinson’s disease (Gątarek, P. et al., 2025). Here, urinary metabolite profiles indicated elevated levels of HVA and succinic acid , and reduced levels of trimethylamine N-oxide in PD patients compared to controls. These findings highlight how altered gut microbiota may influence HVA levels and in turn affect host dopamine catabolism.

Homovanillic acid and neurology

As an indicator of dopaminergic activity, HVA has long been studied as a marker in neurological and psychiatric diseases in which dopamine plays a role. In PD, HVA shows different patterns depending on the biological matrix analyzed. While urinary HVA may increase in PD patients compared to controls, CSF HVA levels are significantly lower in patients with PD, particularly those with akinesia, indicating reduced dopaminergic activity (Davidson, D. et al., 1977). HVA concentration increases with levodopa therapy, though this is not always associated with clinical improvement (Jiménez-Jiménez, F. et al., 2014).

Altered HVA has also been described in mood and cognitive disorders. CSF HVA levels are reduced in dementia with Lewy bodies and Alzheimer’s disease, compared to healthy subjects (Morimoto, S. et al., 2017). In a mouse model, administering HVA was found to improve symptoms of depression by inhibiting synaptic autophagy (Zhao, M. et al, 2024).

Autism spectrum disorder (ASD) has been linked to elevated urinary HVA using metabolomics (Gevi, F. et al., 2020). Higher HVA concentrations have also been associated with greater symptom severity, including agitation, stereotyped behaviors and reduced spontaneous activity (Kaluzna-Czaplinska, J. et al., 2010). Supplementation with vitamin B6, which has been found to be lacking in children with ASD, appears to reduce both HVA concentrations and neurological symptoms (Gątarek, P. et al., 2025). HVA has also been studied in relation to attention-deficit/hyperactivity disorder (ADHD), though findings are inconclusive (Predescu, E. et al., 2024).

In schizophrenia, plasma and CSF HVA often track with symptom severity, but the relationship is complicated. A 2021 review notes that lower CSF HVA often correlates with more severe symptoms and a poorer prognosis (Gasnier, M. et al., 2021). Other work indicates a positive correlation between HVA concentrations and symptom severity (Mazure, C. et al., 1991). Antipsychotic treatment typically causes an initial increase in HVA, followed by a gradual decrease (Gasnier, M. et al., 2021).

Homovanillic acid and oncology

HVA is a marker of catecholamine-secreting tumors, such as neuroblastoma, pheochromocytoma and other neural crest tumors. These tumors secrete excess catecholamines like dopamine, which is then broken down into HVA. Therefore, HVA levels can be a useful indicator of the tumor’s activity. Elevated urinary HVA and vanillylmandelic acid (VMA) are detected in children with neuroblastoma, making them standard tools for diagnosis, monitoring, and prognosis (Sadilkova, K. et al., 2013). A low VMA to HVA ratio is associated with more aggressive disease (Zambrano, E. and Reyes-Múgica, M., 2002).

Infant screening programs based on HVA/VMA were introduced in Japan and elsewhere in the 1980s, but later abandoned after large trials showed no survival benefit and significant overdiagnosis (Tajiri, T. et al., 2009).

While HVA remains clinically relevant in the diagnosis of neuroblastoma, a scoring system using urinary 3-methoxytyramine sulfate (3-MTS) and vanillactic acid has recently been found to be more sensitive than one based on HVA and VMA (Amano, H. et al., 2024).

Homovanillic acid and 5P medicine

HVA’s relationship to host behavior, genetics and microbial activity makes it highly relevant to 5P medicine. As a marker of dopaminergic function and treatment efficacy, HVA may help to refine and personalize different treatment strategies for patients. Advances in metabolomics now allow rapid, high-sensitivity quantification of HVA and vanillylmandelic acid in urine, streamlining diagnostic processes for a whole range of conditions (Pandya, V. and Frank, E., 2022).

At the population level, genome-wide association studies (GWAS) have identified genetic and epidemiological patterns of HVA (Luykx, J. et al., 2014). Multiomics studies are providing further insights, for example, into the causal genetic relationship between HVA and conditions such as eating disorders and schizophrenia (Wen, J. et al., 2025).

Outside the clinic, HVA also plays a role in public health. HVA and related metabolites are used in wastewater-based epidemiology, where their excretion rates correlate with catchment size and provide population-wide biomarkers for exposure and health monitoring (Pandopulos, A. et al., 2021).

References

Amano, H. et al. (2024). Scoring system for diagnosis and pretreatment risk assessment of neuroblastoma using urinary biomarker combinations. Cancer Sci, 115(5), 1634–1645 | https://doi.org/10.1111/cas.16116.

Amin, F. et al. (1992). Homovanillic Acid Measurement in Clinical Research: A Review of Methodology. Schizophrenia Bulletin, 18(1), 123-48 | https://doi.org/10.1093/schbul/18.1.123.

Andén, N. et al. (1963). On the occurrence of homovanillic acid in brain and cerebrospinal fluid and its determination by a fluorometric method. Life Sciences, 2(7), 448-458 | https://doi.org/10.1016/0024-3205(63)90132-2.

Armstrong, M. et al. (1956). The Phenolic Acids of Human Urine: Paper Chromatography of Phenolic Acids. J Biol Chem, 218(1), 293-303 | https://doi.org/10.1016/S0021-9258(18)65893-4.

Blennow, K. et al. (1993). Cerebrospinal fluid monoamine metabolites in 114 healthy individuals 18-88 years of age. Eur Neuropsychopharmacol, 3(1), 55-61 | https://doi.org/10.1016/0924-977x(93)90295-w.

Buleandră, M. et al. (2025). Electrochemical Study and Determination of Homovanillic Acid, the Final Metabolite of Dopamine, Using an Unmodified Disposable Electrode. Molecules, 30(2), 369 | https://doi.org/10.3390/molecules30020369.

Combet, E. et al. (2011). Dietary flavonols contribute to false-positive elevation of homovanillic acid, a marker of catecholamine-secreting tumors. Clinica Chimica Acta, 412(1-2), 165-169 | https://doi.org/10.1016/j.cca.2010.09.037.

Davidson, D. et al. (1977). CSF studies on the relationship between dopamine and 5-hydroxytryptamine in Parkinsonism and other movement disorders. J Neurol Neurosurg Psychiatry, 40(12), 1136–1141 | https://doi.org/10.1136/jnnp.40.12.1136.

Gątarek, P. et al. (2025). The Determination of Trimethylamine N-Oxide and Organic Acids Connected with Gut Microbiota in the Urine of Parkinson’s Disease Patients: A Pilot Study. Int J Mol Sci, 26(10), 4575 | https://doi.org/10.3390/ijms26104575.

Gasnier, M. et al. (2021). A New Look on an Old Issue: Comprehensive Review of Neurotransmitter Studies in Cerebrospinal Fluid of Patients with Schizophrenia and Antipsychotic Effect on Monoamine’s Metabolism. Clin Psychopharmacol Neurosci, 19(3), 395–410 | https://doi.org/10.9758/cpn.2021.19.3.395.

Gevi, F. et al. (2020). A metabolomics approach to investigate urine levels of neurotransmitters and related metabolites in autistic children. Biochimica et Biophysica Acta (BBA) – Molecular Basis of Disease, 1866(10), 165859 | https://doi.org/10.1016/j.bbadis.2020.165859.

Giltay, E. et al. (2005). The sex difference of plasma homovanillic acid is unaffected by cross-sex hormone administration in transsexual subjects. Journal of Endocrinology, 187(1) | https://doi.org/10.1677/joe.1.06307.

Jiménez-Jiménez, F. et al. (2014). Cerebrospinal fluid biochemical studies in patients with Parkinson’s disease: toward a potential search for biomarkers for this disease. Frontiers in Cellular Neuroscience, 8 | https://doi.org/10.3389/fncel.2014.00369.

Köhnke, M. et al. (2003). Plasma Homovanillic Acid: A Significant Association with Alcoholism is Independent of a Functional Polymorphism of the Human Catechol-O-Methyltransferase Gene. Neuropsychopharmacology, 28, 1004–1010 | https://doi.org/10.1038/sj.npp.1300107. Retrieved from Neuropsychopharmacology volume 28, pages1004–1010 (2003).

Kaluzna-Czaplinska, J. et al. (2010). Determination of homovanillic acid and vanillylmandelic acid in urine of autistic children by gas chromatography/mass spectrometry. Medical Science Monitor, 16(9), CR445-50.

Lambert, G. et al. (1993). Regional homovanillic acid production in humans. Life Sciences, 53(1), 63-75 | https://doi.org/10.1016/0024-3205(93)90612-7.

Luykx, J. et al. (2014). Genome-wide association study of monoamine metabolite levels in human cerebrospinal fluid. Mol Psychiatry, 19(2), 228-34 | https://doi.org/10.1038/mp.2012.183.

Marhuenda-Muñoz, M. et al. (2019). Microbial Phenolic Metabolites: Which Molecules Actually Have an Effect on Human Health? Nutrients, 11(11), 2725 | https://doi.org/10.3390/nu11112725.

Mazure, C. et al. (1991). Plasma Free Homovanillic Acid (HVA) as a Predictor of Clinical Response in Acute Psychosis. Biol Psychiatry, 30(5), 475-82 | https://doi.org/10.1016/0006-3223(91)90309-a.

Montanari, L. et al. (1999). Organic and Phenolic Acids in Beer. LWT – Food Science and Technology, 32(8), 535-539 | https://doi.org/10.1006/fstl.1999.0593.

Morimoto, S. et al. (2017). Homovanillic acid and 5-hydroxyindole acetic acid as biomarkers for dementia with Lewy bodies and coincident Alzheimer’s disease: An autopsy-confirmed study. PLoS ONE, 12(2), e0171524 | https://doi.org/10.1371/journal.pone.0171524.

Pandopulos, A. et al. (2021). Application of catecholamine metabolites as endogenous population biomarkers for wastewater-based epidemiology. Science of The Total Environment, 763, 142992 | https://doi.org/10.1016/j.scitotenv.2020.142992.

Pandya, V. and Frank, E. (2022). A Simple, Fast, and Reliable LC-MS/MS Method for the Measurement of Homovanillic Acid and Vanillylmandelic Acid in Urine Specimens. Methods Mol Biol, 2546, 175-183 | https://doi.org/10.1007/978-1-0716-2565-1_16.

Predescu, E. et al. (2024). Metabolomic Markers in Attention-Deficit/Hyperactivity Disorder (ADHD) among Children and Adolescents—A Systematic Review. Int J Mol Sci, 25(8), 4385 | https://doi.org/10.3390/ijms25084385.

Sadilkova, K. et al. (2013). Analysis of vanillylmandelic acid and homovanillic acid by UPLC–MS/MS in serum for diagnostic testing for neuroblastoma. Clinica Chimica Acta, 424, 253-257 | https://doi.org/10.1016/j.cca.2013.06.024.

Tajiri, T. et al. (2009). Risks and benefits of ending of mass screening for neuroblastoma at 6 months of age in Japan. Journal of Pediatric Surgery, 44(12), 2253-2257 | https://doi.org/10.1016/j.jpedsurg.2009.07.050.

Tuck, K. et al. (2002). Structural characterization of the metabolites of hydroxytyrosol, the principal phenolic component in olive oil, in rats. J Agric Food Chem, 50(8), 2404-9 | https://doi.org/10.1021/jf011264n.

Wen, J. et al. (2025). From Clinic to Mechanisms: Multi-Omics Provide New Insights into Cerebrospinal Fluid Metabolites and the Spectrum of Psychiatric Disorders. Molecular Neurobiology, 62, 9120–9132 | https://doi.org/10.1007/s12035-025-04773-0.

Williams, C. et al. (1960). In vivo alteration of the pathways of dopamine metabolism. American Journal of Physiology, 199(4), 722-726 | https://doi.org/10.1152/ajplegacy.1960.199.4.722.

Zambrano, E. and Reyes-Múgica, M. (2002). Hormonal activity may predict aggressive behavior in neuroblastoma. Pediatr Dev Pathol, 5(2), 190-9 | https://doi.org/10.1007/s10024001-0145-8.

Zhao, M. et al. (2024). Gut bacteria-driven homovanillic acid alleviates depression by modulating synaptic integrity. Cell Metabolism, 36(5), 1000-1012.e6 | https://doi.org/10.1016/j.cmet.2024.03.010.

• History & Evolution
• Biosynthesis & dietary uptake
• p-cresol glucuronide and nephrology
• p-cresol glucuronide and cardiometabolic disease
• p-cresol glucuronide and neurology
• p-cresol glucuronide and cancer
• p-cresol glucuronide and 5P medicine
• References

History & Evolution

Early 1990s: first studies on p-cresol

p-cresol glucuronide is a modified form of the gut-microbial metabolite p-cresol, synthesized in the liver to promote excretion through the urine. p-cresol is derived from bacterial proteolytic fermentation of tyrosine (Diether et al. 2019). Early uremia research often (mis)attributed circulating toxicity to unconjugated p-cresol, but improved analytics overturned this view. The current paradigm is that gut-derived p-cresol is rapidly conjugated across the colonic mucosa and in the liver, yielding chiefly p-cresol sulfate (pCS) and a smaller pool of p-cresol glucuronide (pCG), with free p-cresol usually undetectable (Soulage et al. 2022). The notion of unconjugated p-cresol as a uremic toxin is a “historical artefact” (Soulage et al. 2022) arising from acid or heat deproteinization that hydrolyzed conjugates during sample preparation.

Because research centered on pCS, pCG received comparatively little attention. A practical reason is that authentic pCG only became readily available as a commercial reference standard in the late 2000s (Koppe et al. 2017), slowing targeted quantification and mechanistic work until vendors supplied it. Despite higher total concentrations of pCS, differing albumin affinities leave pCS and pCG with comparably sized free fractions in serum (Liabeuf et al. 2013). Over time, pCG has moved from an overlooked “detox conjugate” to a recognized and valuable measure linking microbiome and host health.

Biosynthesis vs. dietary uptake

There appears to be no significant dietary pCG intake. However, diet shapes pCG levels indirectly by modulating colonic proteolysis. More total protein and aromatic amino acids increase substrate load, while fiber (Salmean et al. 2015), resistant starch (Snelson et al. 2024), and certain prebiotics (Meijers et al. 2010) steer carbon toward short-chain fatty acids (SCFA) and away from tyrosine fermentation (Snelson et al. 2024).

In the gut, tyrosine is first converted to 4-hydroxyphenylacetic acid and then decarboxylated to p-cresol, after which p-cresol crosses the mucosa and enters the portal circulation. During first-pass metabolism in the colon and the liver, phase II enzymes rapidly conjugate p-cresol: sulfotransferases yield pCS (Rong et al. 2021; Stachulski et al. 2023) and UDP-glucuronosyltransferases (UGTs) yield pCG (Rong et al. 2020). In the bloodstream, renal organic anion transporters OAT1/3 take up pCG and pCS for tubular secretion, leading to excretion through the urine. When kidney function or transporter capacity is reduced, both conjugates accumulate systemically (Wu et al. 2017), supporting their classification as uremic toxins.

p-cresol glucuronide and nephrology

In nephrology, pCG is clinically relevant: as glomerular filtration and tubular secretory capacity fall, pCG accumulates in serum, reaching even higher levels in hemodialysis (Peters et al. 2023; Liabeuf et al. 2013). Accordingly, pCG is considered a microbiome-derived uremic toxin linking gut ecology with renal outcomes.

Chronic kidney disease (CKD) reduces gut microbial diversity and induces compositional shifts (Peters et al. 2023). In a prospective cohort, pCG associated with greater estimated glomerular filtration rate (eGFR) loss over six years, showing that higher pCG tracks kidney decline (Peters et al. 2023). Furthermore, total and free pCG in serum rise with CKD severity, and higher pCG predicts overall and cardiovascular mortality, with risk prediction similar to pCS (Liabeuf et al. 2013). Beyond reflecting burden, pCG can perturb proximal tubular phenotype, inducing epithelial–mesenchymal transition markers, transporter dysregulation, and cellular stress in human renal proximal tubular cells. Notably, pCS did not affect cellular stress in the same system (Mutsaers et al. 2015).

Together, these lines of evidence support ongoing evaluation of pCG as a diagnostic and prognostic biomarker across CKD stages (Choudhary et al. 2025).

p-cresol glucuronide and cardiometabolic disease

Compared with its sulfate counterpart and the parent phenol, pCG appears less cytotoxic across human kidney, liver, and blood cell models; a pattern consistent with glucuronidation serving primarily as detoxification (Zhu et al. 2021; Bertarini et al. 2025). Nevertheless, its cardiometabolic relevance remains unsettled, with context-dependent findings that do not point uniformly toward a role as toxin or detoxicant.

In a prospective analysis in CKD patients, the total p-cresol burden (pCS + pCG) in serum associated with mortality and cardiovascular disease (CVD). Importantly the conjugation pattern mattered: a lower pCS/pCG ratio, so a relative shift toward glucuronidation, was independently associated with higher risks of mortality and CVD (Poesen et al. 2016). At the same time, metabolic effects diverged between conjugates, since in mouse and cellular models, pCG did not induce insulin resistance, whereas pCS did (Koppe et al. 2017). In liver models, the parent compound p-cresol induced pronounced oxidative stress, glutathione depletion, and necrotic cell death. In contrast, pCG showed minimal cytotoxicity in hepatocytes and behaved as a detoxification product rather than a toxin, making it unlikely to mediate the hepatic toxic effects attributed to p-cresol (Zhu et al. 2021).

Diet–microbiome evidence links pCG to hypertension and diabetes biology. In diabetic mice, uremic toxins, including pCG, were elevated in plasma. Resistant starch supplementation lowered pCG, strengthened the intestinal barrier and dampened renal inflammation as indicated by fewer neutrophils and lower complement activation. Finally, resistant starch reduced albuminuria, supporting a renoprotective effect of pCG in diabetes (Snelson et al. 2024).

In humans, low fiber intake associated with higher circulating pCG and higher blood pressure. In a randomized trial, SCFA-enriched fiber reduced both plasma pCG and blood pressure, consistent with a microbiota-mediated shift toward reduced tyrosine fermentation (Xu et al. 2025).

p-cresol glucuronide and neurology

As a host–microbe co-metabolite, pCG sits on the gut–brain axis. Although glucuronides are often considered inert (Bertarini et al. 2025), converging evidence suggests that pCG can be biologically active at the cerebral endothelium. In mice, pCG strengthened blood–brain barrier (BBB) integrity and reshaped the whole-brain transcriptome (Stachulski et al. 2023; Bertarini et al. 2025).

In human brain microvascular endothelial cells, pCG alone had little effect but prevented LPS-induced barrier permeability when co-applied, acting as a functional antagonist at toll-like receptor 4 (TLR4) (Stachulski et al. 2023). These findings support a context-dependent, neuroprotective role against pro-inflammatory stimuli under physiological conditions (Bertarini et al. 2025).

Population data are bidirectional: higher urinary p-cresol, pCS, and pCG have been reported in autistic children (Bertarini et al. 2025; Gabriele et al. 2014), whereas in children with CKD, pCG did not associate with neurological outcomes (Ebrahimi et al. 2025). In Parkinson’s disease cohorts, serum pCG was elevated alongside p-cresol and pCS in patients (Paul et al. 2023) and plasma pCG positively correlated with motor symptom severity (Chen et al. 2023). Evidences from Alzheimer’s disease (AD) research showed higher pCG levels in patients together with adverse brain aging and cognitive decline (Gordon et al. 2024).

p-cresol glucuronide and cancer

Metabolomics has contributed to placing pCG on the oncology map, not as a causal driver, but rather as a non-invasive urinary biomarker. In renal cell carcinoma, an independent validation cohort confirmed higher urinary pCG at diagnosis and still elevated one year post-nephrectomy (Oto et al. 2025). In colorectal cancer, urinary profiles of pCG across disease and recovery windows after surgery showed upregulation in disease stage 3-4 and downregulation in stage 0-2, suggesting stage-linked pCG variation (Fu et al. 2024; Liesenfeld et al. 2015). Similarly, a prospective study of bladder cancer (BC) patients validated urinary pCG as a diagnostic biomarker, and notably within non-muscle-invasive BC (NMIBC), as a staging biomarker (Oto et al. 2022).

p-cresol glucuronide and 5P medicine

Biomarkers predict treatment response, monitor progression, stratify risk, and enable precision care, serving several aims of 5P medicine. In blood, pCG levels rise with declining renal function and pCG has been suggested as diagnostic and prognostic marker across CKD stages (Choudhary et al. 2025).

Owing to its excretion fate, pCG is also a relevant marker in urine, a compelling non-invasive matrix for predictive and diagnostic applications in oncology (Liesenfeld et al. 2015; Oto et al. 2022). Urine reflects renal excretion, including short-term changes induced by diet or drugs, more rapidly than blood.

As a biomarker, pCG is also actionable. Diet and microbiome interventions, especially fermentable fibers/resistant starch, can lower circulating pCG levels (Snelson et al. 2024; Xu et al. 2025), supporting the preventive, predictive and participatory pillars of 5P medicine. Especially in combination with other uremic toxins like indoxyl sulfate or p-cresol sulfate , pCG can sharpen risk predictions, clarify mechanisms, and improve precision healthcare.

References

Bertarini, L. et al.: Para-Cresol and the Brain: Emerging Role in Neurodevelopmental and Neurodegenerative Disorders and Therapeutic Perspectives (2025) ACS pharmacology & translational science | https://doi.org/10.1021/acsptsci.5c00289.

Chen, S.-J. et al.: Plasma metabolites of aromatic amino acids associate with clinical severity and gut microbiota of Parkinson’s disease (2023) npj Parkinson’s Disease | https://doi.org/10.1038/s41531-023-00612-y.

Choudhary, B.B. et al.: Novel molecular biomarkers in kidney diseases: bridging the gap between early detection and clinical implementation (2025) Journal of Pharmacy and Pharmacology | https://doi.org/10.1093/jpp/rgaf096.

Diether, N.E. et al.: Microbial Fermentation of Dietary Protein: An Important Factor in Diet⁻Microbe⁻Host Interaction (2019) Microorganisms | https://doi.org/10.3390/microorganisms7010019.

Ebrahimi, M. et al.: Investigation of a targeted panel of gut microbiome-derived toxins in children with chronic kidney disease (2025) Pediatric Nephrology | https://doi.org/10.1007/s00467-024-06580-6.

Fu, C. et al.: The Potential of Metabolomics in Colorectal Cancer Prognosis (2024) Metabolites | https://doi.org/10.3390/metabo14120708.

Gabriele, S. et al.: Urinary p-cresol is elevated in young French children with autism spectrum disorder: a replication study (2014) Biomarkers | https://doi.org/10.3109/1354750X.2014.936911.

Gordon, S. et al.: Metabolites and MRI-Derived Markers of AD/ADRD Risk in a Puerto Rican Cohort (2024) Research square | https://doi.org/10.21203/rs.3.rs-3941791/v1.

Koppe, L. et al.: p-Cresyl glucuronide is a major metabolite of p-cresol in mouse: in contrast to p-cresyl sulphate, p-cresyl glucuronide fails to promote insulin resistance (2017) Nephrology Dialysis Transplantation | https://doi.org/10.1093/ndt/gfx089.

Liabeuf, S. et al.: Does p-cresylglucuronide have the same impact on mortality as other protein-bound uremic toxins? (2013) PLoS ONE | https://doi.org/10.1371/journal.pone.0067168.

Liesenfeld, D.B. et al.: Changes in urinary metabolic profiles of colorectal cancer patients enrolled in a prospective cohort study (ColoCare) (2015) Metabolomics | https://doi.org/10.1007/s11306-014-0758-3.

Meijers, B.K.I. et al.: p-Cresyl sulfate serum concentrations in haemodialysis patients are reduced by the prebiotic oligofructose-enriched inulin (2010) Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association – European Renal Association | https://doi.org/10.1093/ndt/gfp414.

Mutsaers, H.A.M. et al.: Proximal tubular efflux transporters involved in renal excretion of p-cresyl sulfate and p-cresyl glucuronide: Implications for chronic kidney disease pathophysiology (2015) Toxicology in vitro : an international journal published in association with BIBRA | https://doi.org/10.1016/j.tiv.2015.07.020.

Oto, J. et al.: LC-MS metabolomics of urine reveals distinct profiles for non-muscle-invasive and muscle-invasive bladder cancer (2022) World Journal of Urology | https://doi.org/10.1007/s00345-022-04136-7.

Oto, J. et al.: Validation of urine p-cresol glucuronide as renal cell carcinoma non-invasive biomarker (2025) Journal of Proteomics | https://doi.org/10.1016/j.jprot.2024.105357.

Paul, K.C. et al.: Untargeted serum metabolomics reveals novel metabolite associations and disruptions in amino acid and lipid metabolism in Parkinson’s disease (2023) Molecular Neurodegeneration | https://doi.org/10.1186/s13024-023-00694-5.

Peters, B.A. et al.: Association of the gut microbiome with kidney function and damage in the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) (2023) Gut Microbes | https://doi.org/10.1080/19490976.2023.2186685.

Poesen, R. et al.: Metabolism, Protein Binding, and Renal Clearance of Microbiota-Derived p-Cresol in Patients with CKD (2016) Clinical journal of the American Society of Nephrology : CJASN | https://doi.org/10.2215/CJN.00160116.

Rong, Y. et al.: Characterizations of Human UDP-Glucuronosyltransferase Enzymes in the Conjugation of p-Cresol (2020) Toxicological Sciences | https://doi.org/10.1093/toxsci/kfaa072.

Rong, Y. et al.: Characterization of human sulfotransferases catalyzing the formation of p-cresol sulfate and identification of mefenamic acid as a potent metabolism inhibitor and potential therapeutic agent for detoxification (2021) Toxicology and applied pharmacology | https://doi.org/10.1016/j.taap.2021.115553.

Salmean, Y.A. et al.: Fiber supplementation lowers plasma p-cresol in chronic kidney disease patients (2015) Journal of renal nutrition : the official journal of the Council on Renal Nutrition of the National Kidney Foundation | https://doi.org/10.1053/j.jrn.2014.09.002.

Snelson, M. et al.: Dietary resistant starch enhances immune health of the kidney in diabetes via promoting microbially-derived metabolites and dampening neutrophil recruitment (2024) Nutrition & Diabetes | https://doi.org/10.1038/s41387-024-00305-2.

Soulage, C.O. et al.: The very last dance of unconjugated p-cresol… historical artefact of uraemic research (2022) Nephrology Dialysis Transplantation | https://doi.org/10.1093/ndt/gfab325.

Stachulski, A.V. et al.: A host-gut microbial amino acid co-metabolite, p-cresol glucuronide, promotes blood-brain barrier integrity in vivo (2023) Tissue Barriers | https://doi.org/10.1080/21688370.2022.2073175.

Wu, W. et al.: Key Role for the Organic Anion Transporters, OAT1 and OAT3, in the in vivo Handling of Uremic Toxins and Solutes (2017) Scientific Reports | https://doi.org/10.1038/s41598-017-04949-2.

Xu, C. et al.: Lack of dietary fibre increases gut microbiome-derived uremic toxins that contribute to increased blood pressure (2025) medRxiv | https://doi.org/10.1101/2025.09.16.25335853.

Zhu, S. et al.: Effects of p-Cresol on Oxidative Stress, Glutathione Depletion, and Necrosis in HepaRG Cells: Comparisons to Other Uremic Toxins and the Role of p-Cresol Glucuronide Formation (2021) Pharmaceutics | https://doi.org/10.3390/pharmaceutics13060857.