Impact of dietary intervention and supplementation on the physiological profile of individuals with down syndrome

Abstract

Down syndrome, or trisomy 21 (T21), is a genetic neurodevelopmental disorder (NDD) often associated with various health challenges. In this population, physiological alterations remain poorly recognized, including dysregulated interactions among systems that play important roles in maintaining homeostasis. Common challenges include nutritional deficiencies, impaired metabolic detoxification, a hypoxia-like state, and gut dysbiosis. This study investigates the impact of dietary interventions and nutrient supplementation on the physiological profiles of individuals with T21. Using a database from the “Integración Down” Institute in Mexico City, we analyzed a panel of 121 biochemical markers before and after a 10-month intervention period. The interventions comprised dietary restriction alone (D) and dietary restriction with nutrient supplementation (D + S), aimed at modifying the participants’ physiological profiles. Our findings demonstrate changes in several physiological markers following the interventions, mainly in response to D + S. In particular, we observed changes in dysbiosis markers related to bacterial and fungal overgrowth; the D + S protocol reduced biomarkers associated with fungal overgrowth. Biomarkers related to nutritional deficiency and a hypoxia-like state (described for T21) showed changes, especially in the D + S group. Increases in red‑blood‑cell indices and decreases in markers of metabolic pathway disruption suggest improved energy metabolism. This research highlights the potential of dietary interventions to ameliorate T21-associated comorbidities and improve overall health outcomes in this population. Further research is needed to explore the long-term effects and optimal strategies for personalized nutritional management in individuals with T21.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-32212-6.

Introduction

Neurodevelopmental disorders (NDD) present significant global challenges. Trisomy 21 (T21), is a prominent NDD, characterized by distinct cognitive, behavioral, and physiological characteristics. The underlying mechanisms of NDD, including T21, are not yet fully understood¹.

This syndrome, primarily caused by trisomy 21, is often associated with nutritional deficiencies², gut dysbiosis³, impaired metabolic detoxification processes and a hypoxia-like state⁴. These factors can contribute to increased comorbidities, which in turn impact quality of life⁵. Individuals with T21 frequently exhibit metabolic alterations indicative of nutritional deficits, which can impair cognitive function and exacerbate existing health issues². Moreover, individuals with T21 have been reported to excrete less hippuric acid after benzoic acid loading than typically developing controls, supporting the view that hepatic function is abnormal⁶. This underscores the importance of comprehensive nutritional assessments in T21 care. Oxidative stress and hypoxia-like state (low oxygen availability) are implicated in the pathophysiology of various T21-associated conditions, highlighting the need for targeted interventions to mitigate oxidative damage and potentially improve metabolic, cognitive, and physical health outcomes⁷. Furthermore, the interplay between dysbiosis, nutritional status, and oxidative stress suggests a crucial role for gut microbiota in modulating metabolic and inflammatory pathways often disrupted in this condition^5,8–10.

Various nutritional deficiencies significantly impact the health and well-being of T21 population^2,11. Iron deficiency, often leading to anemia and cognitive impairment, is particularly prevalent¹². Other common deficiencies include zinc, folate, selenium, and vitamin B12¹³, micronutrients essential for growth, development, immune function and oxidative balance¹⁴. Additionally, individuals with T21 may experience dietary sensitivities or intolerances, including celiac disease, gluten sensitivity, and malabsorption^2,15. The prevalence of celiac disease is estimated to be 5–12%, significantly higher than the 1% prevalence in the general population¹¹. Sensitivities to casein and gluten in T21 are subjects of ongoing research. Some studies suggest that a gluten-free, casein-free diet may benefit individuals with T21, potentially improving gastrointestinal symptoms, cognitive function, and overall quality of life^12,15. Addressing specific nutritional deficiencies requires a multidisciplinary approach. Comprehensive individual nutritional assessments, ongoing monitoring, and personalized dietary interventions are essential for optimizing nutritional status and minimizing adverse health outcomes.

This study investigates physiological alterations associated with T21, including gut dysbiosis, nutritional deficiencies, a hypoxia-like state, and impaired metabolic detoxification. The aim is to establish a comprehensive physiological profile of individuals with T21 by analyzing a panel of biochemical markers before and after implementing either dietary restriction alone or dietary restriction with supplementation. By identifying alterations in specific biochemical profiles and their responses to dietary interventions, this study proposes a translational framework for personalized dietary interventions in the T21 population. The goal is to support the development of targeted dietary restriction and nutrient supplementation strategies. These personalized interventions hold the potential to mitigate T21-related comorbidities and enhance the effectiveness of behavioral therapies. Ultimately, this research strives to contribute to an innovative, physiology-based approach to health care for individuals with T21.

Results

Biomarkers classification

Table 1 presents four functional categories of biomarkers: Dysbiosis, Hypoxia‑like state, Detoxification, and Nutritional deficiencies. Each biomarker is assigned a unique code consisting of the biomarker category and a number. The presence of biomarkers shared across categories does not imply identical interpretation; the meaning of a given biomarker can vary depending on the physiological system and its classification.

Table 1.

Biomarker classification.

Dysbiosis		Nutritional deficiencies
Dysbiosis code	Biomarker name	Nutritional deficiencies code	Biomarker name
DYS-1	3-4-Dihydroxy-phenyl-propionic acid (DHPPA)	ND-1	Glucose
DYS-2	3-3-Hydroxyphenyl-3-hydroxypropionic acid (HPHPA)	ND-2	Total cholesterol
DYS-3	4-Hydroxyphenilbenzoic acid	ND-3	HDL
DYS-4	4-Hydroxyphenylacetic acid	ND-4	LDL
DYS-5	4-Hydroxyhippuric acid	ND-5	Triglyceride
DYS-6	4-cresol (4-Methylphenol)	ND-6	Total homocysteine
DYS-7	2-Hydroxyphenylacetic acid	ND-7	Hemoglobin (Hb)
DYS-8	Hippuric acid	ND-8	Mean corpuscular volume (MCV)
DYS-9	3-indole-acetic acid	ND-9	Hematocrit (HCT)
DYS-10	Succinic acid	ND-10	Mean corpuscular hemoglobin (MCH)
DYS-11	Lactic acid	ND-11	Mean corpuscular hemoglobin concentration (MCHC)
DYS-12	Pyruvic acid	ND-12	Red blood cell (RBC)
DYS-13	Furan-2-5-dicarboxylic acid	ND-13	Red cell distribution width (RDW)
DYS-14	Furancarbonylglycine	ND-14	Fumaric acid
DYS-15	5-Hydroxymethyl-2-furoic acid	ND-15	2-oxoglutaric acid
DYS-16	Tricarballylic acid	ND-16	Lactic acid
DYS-17	Oxalic acid	ND-17	Pyruvic acid
DYS-18	Glyceric acid	ND-18	Adipic acid
DYS-19	Glycolic acid	ND-19	Ethylmalonic acid
DYS-20	Arabinose	ND-20	Methylsuccinic acid
DYS-21	Carboxycitric acid	ND-21	Sebacic acid
DYS-22	Citramalic acid	ND-22	Suberic acid
DYS-23	Tartaric acid	ND-23	Ascorbic acid (VitC)
DYS-24	3-Oxoglutaric acid	ND-24	Glutaric acid (VitB2)
DYS-25	Casomorphin	ND-25	3-hydroxy-3-methylglutaric acid (VitQ10)
DYS-26	Gluteomorphin	ND-26	Methylcitric acid (VitH)
DYS-27	Citric acid	ND-27	Methylmalonic acid (VitB12)
DYS-28	2-oxoglutaric acid	ND-28	N-acetylcysteine (NAC)

Detoxification		Hypoxia-like state
Detoxification code	Biomarker name	Hypoxia-like state code	Biomarker name
DE-1	Urea Nitrogen	HYP-1	Hemoglobin (Hb)
DE-2	Urea	HYP-2	Mean corpuscular volume (MCV)
DE-3	Creatinine	HYP-3	Hematocrit (HCT)
DE-4	Uric acid	HYP-4	Mean corpuscular hemoglobin (MCH)
DE-5	2-Hydroxyhippuric acid	HYP-5	Mean corpuscular hemoglobin concentration (MCHC)
DE-6	Orotic acid	HYP-6	Red cell distribution width (RDW)
DE-7	Pyroglutamic acid	HYP-7	Succinic acid
DE-8	Oxalic acid	HYP-8	Lactic acid
DE-9	Glyceric acid	HYP-9	Pyruvic acid
DE-10	Glycolic acid	HYP-10	Succinic acid

To evaluate the effect of each dietary modification, pre‑ and post‑intervention comparisons within the three study groups—Control (C), Restriction Diet (D), and Diet + Supplement (D + S)—were conducted using Wilcoxon tests. The tests revealed significant improvements in numerous parameters for the D and D + S groups, whereas no statistically significant differences were observed in the Control (C) group. To identify potentially meaningful yet non‑significant changes, effect sizes appropriate for non‑parametric data were also calculated. Consistently, larger effect sizes were observed in the D + S group.

Dysbiosis biomarkers

As shown in Fig. 1, several metabolites exhibited substantial alterations following the interventions. Among the most notable were glyceric acid, 3‑indole‑acetic acid, oxoglutaric acid, furancarbonylglycine, and carboxycitric acid, which showed significant changes in the D + S group (Wilcoxon p < 0.05).

Fig. 1.

Dysbiosis: Biomarkers affected by dietary interventions. (A) Volcano plot of dysbiosis biomarkers. This volcano plot summarizes statistical comparisons of dysbiosis biomarkers across experimental groups: Control (C), Restriction Diet (D), and Diet + Supplement (D + S). Each point represents a biomarker (see Table 1). The x-axis indicates effect size and was calculated using the Rank-Biserial Correlation. The y-axis represents statistical significance as –log₁₀(p-value) from the Wilcoxon signed-rank test. The horizontal blue dotted line indicates the significance threshold (p < 0.05). The different colors indicate effect size categories: orange (huge), yellow (very large), blue (large), light blue (moderate), and green (small). Only biomarkers with statistically significant differences and a moderate effect size are identified. (B) Box-and-whisker plots of biomarkers with greater relevance. These plots show significant changes between basal and post-intervention for scaled values. Groups are color-coded: blue (control), yellow (diet), and orange (diet + supplement), with lighter shades representing basal values. Boxes represent the interquartile range, whiskers show minimum and maximum values, and horizontal lines show the medians. The asterisks represent statistically significant changes between basal and post-intervention levels (*p < 0.05, **p < 0.01, Wilcoxon signed-rank test).

Additionally, markers such as DHPPA and tartaric acid showed appreciable reductions in their scaled values post‑intervention, despite not reaching conventional statistical significance thresholds (see the Supplementary Material). Their associated effect sizes suggest biological changes. These markers are primarily associated with changes in dietary intake and reductions in gut overgrowth^16,17.

Conversely, the Control (C) group did not show significant changes, with minimal variation across all dysbiosis markers, reinforcing that the observed changes were specific to the intervention protocols. The dietary interventions induced broad modifications across multiple physiological domains, as revealed by the analysis of scaled biochemical markers. Both the D group and the D + S group demonstrated distinct response patterns, with the latter generally exhibiting greater magnitude and consistency in the observed changes.

The significant changes in dysbiosis‑related metabolites—particularly furancarbonylglycine, glyceric acid, carboxycitric acid, and 3‑indole‑acetic acid—in the D + S group stand out as one of the clearest biochemical signatures of the intervention’s efficacy. Some of these compounds are well‑known byproducts of microbial fermentation, especially associated with the proliferation of yeast species (e.g., Aspergillus), which are frequently identified as dominant organisms in dysbiotic gut environments¹⁸. This imbalance could be implicated in gastrointestinal inflammation, impaired mucosal immunity, and systemic metabolic disturbances, particularly in individuals with neurodevelopmental conditions^19,20 A more detailed description is provided in Supplementary Table 1.

The magnitude of the reduction suggests not only a change in microbial load but also a potential restoration of intestinal eubiosis. This shift may reflect improvements in digestive function, barrier integrity, and immune surveillance, representing a pivotal mechanism by which the dietary intervention exerts systemic benefits^21,22.

Elevated levels of 3‑indole‑acetic acid (IAA) (Fig. 1b) may reflect increased microbial fermentation of tryptophan and suggest a shift toward a more tolerogenic and anti‑inflammatory gut environment. IAA is an indole derivative known to activate the aryl hydrocarbon receptor (AhR), a transcription factor involved in maintaining mucosal immunity, promoting epithelial barrier integrity, and regulating inflammation^23,24. In individuals with T21, who exhibit chronic low-grade inflammation and immune dysregulation, such shifts could represent a beneficial microbial adaptation enhancing immune homeostasis and reducing systemic oxidative stress²⁵.

A sustained decrease in urinary 2-oxoglutaric acid (α-ketoglutarate) and 3-oxoglutaric acid in the D + S group—absent in the others—may indicate a coordinated metabolic shift involving both mitochondrial and gut microbial functions. α-Ketoglutarate serves as a central metabolic hub within the tricarboxylic acid (TCA) cycle and acts as a signaling molecule that regulates nitrogen metabolism, redox balance, and carbon–nitrogen homeostasis²⁶. Moreover, α-ketoglutarate modulates gut microbial dynamics, contributing to protein metabolism and nitrogen recycling within the host–microbiota axis, particularly under dietary conditions that alter amino acid availability²⁷.

In conjunction with the decline in α-ketoglutarate, the reduction of urinary 3‑oxoglutaric acid may underscore a regulated metabolic response involving microbial–host interplay. In metabolomic profiling studies conducted in pediatric populations, including children with gastrointestinal or neurodevelopmental conditions, 3‑oxoglutaric acid has been consistently identified as a microbial fermentation byproduct linked to shifts in gut ecology^17,28.

On the other hand, the increase in urinary furancarbonylglycine observed in the D and D + S groups may reflect enhanced detoxification of furan‑derived compounds via Phase II glycine conjugation, rather than persistent fungal overgrowth. While direct data on this metabolite in neurodevelopmental cohorts are lacking, studies such as Vissoker et al.¹⁸ have documented elevations of related microbial byproducts (e.g. 3‑oxoglutaric acid) in children with autism and gastrointestinal symptoms, suggesting associations with gut microbial alterations.

Similarly, the increase in urinary carboxycitric acid detected in both intervention groups may signal a microbiota‑driven metabolic shift rather than a dysbiotic signature. This increase has been associated with fungal fermentation and altered citrate metabolism, previously reported in individuals with gastrointestinal imbalances²⁹. In this context, its elevation could reflect a modulated microbial response or compensatory mitochondrial overflow, potentially linked to changes in nutrient processing and microbial cross-feeding dynamics.

Taken together, these trends may reflect a harmonization of the gut metabolic environment, suggesting that interventions designed to restore microbial equilibrium could simultaneously alleviate oxidative and metabolic stress in tissues beyond the gastrointestinal tract. Future metabolomic profiling and gut microbiome sequencing could further elucidate the origin and systemic impact of these metabolites, solidifying their role as both biomarkers and potential therapeutic targets in T21.

From a mechanistic standpoint, these microbial and fungal metabolites could act as functional disruptors of mitochondrial oxidative phosphorylation and central nervous system signaling²¹. Their sustained presence may perpetuate a cycle of neurodegeneration, impaired redox homeostasis, and disrupted neurotransmitter turnover, all of which are altered in T21^21,30.

Several other metabolites that were reduced in this study—such as oxalic, tartaric, pyruvic, and lactic acid—have been reported to influence mitochondrial function by promoting the generation of reactive oxygen species (ROS) and a shift toward anaerobic metabolism. This increase in oxidative stress may compromise the activity of key enzymes, including succinate dehydrogenase, cytochrome c oxidase, and α‑ketoglutarate dehydrogenase, thereby contributing to impaired energy metabolism³¹. High levels of these compounds may indicate altered ATP production and trigger compensatory metabolic shifts, such as upregulation of glycolysis and anaerobic metabolism, which were indirectly reflected by baseline elevations in lactic and pyruvic acids in individuals with T21³². The observed post-intervention decreases in these byproducts, in parallel with microbial metabolite reductions, could I suggest a systemic recovery of mitochondrial efficiency.

Although these specific pathways were not directly measured, the shifts observed in microbial metabolites in this study support the idea that diet‑induced changes in the gut environment modify the biochemical pathways of the gut–brain axis. The partial normalization of these markers in the intervention groups highlights that dietary interventions may modulate these central mechanisms through peripheral biochemical reprogramming.

Nutritional deficiency biomarkers

In Fig. 2, the intervention groups showed clear improvements related to micronutrient status. Methylmalonic acid (MMA), a marker of vitamin B12 metabolism^33,34, decreased significantly in the D + S group. Similarly, suberic acid—whose elevated levels are typically associated with mitochondrial dysfunction—showed a marked reduction, suggesting potential improvements in mitochondrial fatty acid oxidation³⁵. Moreover, increases in hemoglobin (Hb), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC) and mean corpuscular volume (MCV) were observed.

Fig. 2.

Nutritional Deficiencies: Biomarkers affected by dietary interventions. (A) Volcano plot of nutritional biomarkers, summarizing statistical comparisons across experimental groups: Control (C), Diet (D), and Diet + Supplement (D + S). (B) Box-and-whisker plots of biomarkers with greater relevance. See the footnote of Fig. 1 for the plot´s description.

Other markers, such as N‑acetylcysteine (NAC) and pantothenic acid (vitamin B5), exhibited post‑intervention changes with moderate effect sizes, although without reaching statistical significance (Supplementary Table 2). These findings may suggest trends toward improved vitamin absorption and sulfur amino acid metabolism in response to supplementation. Notably, even the D group displayed modest but consistent improvements in several markers, including suberic and adipic acid, supporting the partial efficacy of dietary restriction alone. In contrast, the C group’s values remained mostly stable or slightly worsened in some cases (Supplementary Table 2).

The observed improvements in key nutritional and hematologic biomarkers (see statistical significance in Supplementary Table 6) among participants in the D + S group provide strong evidence supporting the efficacy of targeted metabolic interventions in individuals with T21. As shown by the volcano plot in Fig. 2, statistically significant improvements in Hb, MCH, MCHC, and MCV—together with a decrease in MMA—reflect restoration of erythropoiesis, enhanced micronutrient bioavailability, and systemic metabolic homeostasis.

Among these markers, MMA stands out as a sensitive indicator of intracellular vitamin B12 deficiency^36–39. Elevated MMA may result from impaired absorption, increased oxidative degradation, or functional enzymatic limitations^40,41. Their substantial reduction in the D + S group suggests improved vitamin B12 uptake and utilization, potentially supporting myelination, mitochondrial ATP production, and hematologic stability^42,43.

Reductions in RDW further corroborate this hematologic normalization, pointing to less variation in erythrocyte size and more efficient erythropoiesis^44,45. Given the frequent occurrence of ineffective hematopoiesis in T21, these findings may extend beyond hematology to clinical benefits such as improved oxygen transport, stamina, and cognitive alertness^4,46. Biochemical rebalancing through supplementation can produce systemic effects relevant to both physical and neurocognitive domains^46,47.

A significant post‑intervention decrease in suberic acid was observed in both the D and D + S groups, with minimal changes in controls. Suberic acid, a medium‑chain dicarboxylic acid, accumulates during impaired mitochondrial β‑oxidation and lipid peroxidation, often reflecting oxidative stress and metabolic inefficiency³⁵ Its reduction may indicate enhanced mitochondrial fatty‑acid oxidation capacity and improved redox balance, likely promoted by the dietary and microbiota‑modulating effects of the intervention. These findings suggest a potential improvement in energy metabolism, particularly in a population with mitochondrial vulnerability such as individuals with T21⁴⁸.

Notably, even markers that did not meet statistical significance thresholds, such as N‑acetylcysteine (NAC), showed consistent increases with moderate effect sizes (Supplementary Table 2). NAC—a precursor of glutathione—plays a pivotal role in neutralizing hydrogen peroxide, a byproduct of SOD1 triplication in T21^49,50. The upward trend in NAC, though not statistically significant, suggests an improved redox environment and may enhance detoxification as well as modulate glutamate and dopamine levels in the brain^51,52.

These insights are further supported by the observation that the D + S group consistently exhibited more pronounced biochemical changes across all markers compared with the D and C groups. This suggests that dietary intervention, while beneficial, may be insufficient to correct entrenched deficiencies or meet the elevated metabolic demands imposed by T21. Supplementation with bioavailable cofactors appears necessary to bypass gastrointestinal and enzymatic barriers to absorption and utilization.

Importantly, baseline data revealed a disproportionately high number of individuals with out‑of‑range values for MMA, Hb, and RDW, which shifted closer to reference ranges following the intervention (particularly in the D + S group), highlighting a movement toward physiological restoration rather than mere numerical improvement. Given the known consequences of these imbalances—including cognitive delay, anemia, and reduced immune competence—these findings have important clinical implications.

In parallel with improvements in redox and hematologic markers, a specific reduction in fasting plasma glucose was observed in the D + S group, while values in the C group remained largely unchanged. This decline may reflect a favorable shift in systemic glucose regulation and insulin sensitivity, potentially driven by enhanced micronutrient availability and modulation of gut microbial composition. Individuals with T21 are known to exhibit an increased risk of impaired glucose metabolism, insulin resistance, and mitochondrial dysfunction, which collectively contribute to altered energy utilization and elevated cardiometabolic risk^53,54. The observed reduction in blood glucose may represent an adaptive metabolic response to nutritional intervention, supporting improved carbohydrate handling and systemic energy homeostasis. These results underscore the need for proactive nutritional management even in the absence of overt deficiency symptoms.

Hypoxia-like state biomarkers

As illustrated in Fig. 3, normalization of oxygen‑transport indicators was evident, particularly in the D + S group. Red blood cell indices such as MCHC and RDW showed significant improvements. Decreases in lactic and pyruvic acids with large to very large effect sizes were also noted, suggesting a shift toward improved aerobic metabolism^55,56. Although these changes did not meet the statistical significance threshold, they indicate a consistent direction of metabolic adaptation that may reflect improved tissue oxygenation or mitochondrial function (see Supplementary Table 3)^32,57. Minimal or no changes were evident in the C group, and the D group exhibited moderate trends toward improvement in selected markers (e.g., MCH, MCHC, RDW), albeit less robust than in the D + S cohort (see the clinical significance of hypoxia‑like state biomarkers in Supplementary Table 3).

Fig. 3.

Hypoxia-like state: Biomarkers affected by dietary interventions. (A) Volcano plot of hypoxia-like state biomarkers. This volcano plot summarizes statistical comparisons of hypoxia-like state biomarkers across experimental groups: Control (C), Diet (D), and Diet + Supplement (D + S). (B) Box-and-whisker plots of biomarkers with greater relevance. See the footnote of Fig. 1 for the plot’s description.

The consistent reductions in anaerobic metabolites—most notably lactic and pyruvic acids—among participants in the D + S group strongly support the hypothesis of improved mitochondrial efficiency in individuals with T21. These metabolites, elevated at baseline, are hallmarks of incomplete oxidative metabolism and are typically upregulated in contexts of mitochondrial dysfunction and tissue‑level hypoxia^32,58. The post‑intervention decline, supported by large and very large effect sizes, suggests a shift away from glycolytic overdependence toward more effective engagement of aerobic metabolic pathways. This improvement likely reflects enhanced mitochondrial respiration and energy efficiency, reducing cellular reliance on anaerobic glycolysis and lactate production to meet ATP demands⁵⁹.

Lactic and pyruvic acid accumulation in T21 often stems from compromised activity of mitochondrial complexes I and IV, leading to a metabolic bottleneck that forces reliance on anaerobic glycolysis for energy^55,56. The observed biochemical shift suggests enhanced mitochondrial enzymatic activity, improved substrate oxidation, and more complete engagement of the tricarboxylic acid (TCA) cycle⁶⁰. Supplementation may have provided critical cofactors (such as NAD⁺, FAD, and lipoic acid) that help restore redox balance and improve the functionality of pyruvate dehydrogenase and electron transport chain components⁶¹. In turn, this would allow for more efficient conversion of pyruvate into acetyl-CoA and reduce its conversion into lactate, thereby promoting a more energy-yielding and oxygen-efficient metabolic profile⁵⁹.

The significant reduction in RDW indicates more uniform erythrocyte size, reducing variability that often reflects ineffective erythropoiesis or bone marrow stress^44,45. The elevation in MCHC suggests increased Hb concentration within red blood cells, enhancing their oxygen-carrying efficiency⁴⁴. Together, these effects not only reflect improved nutrient availability but also point to a more functional erythropoietic process, likely supported by enhanced mitochondrial ATP production within hematopoietic progenitors⁶². Improved red blood cell morphology and Hb saturation may ultimately contribute to better tissue oxygenation and metabolic performance⁶³.

Detoxification biomarkers

As shown in Fig. 4, detoxification‑related biomarkers were also modulated post‑intervention. Statistically significant changes were detected in glyceric and pyroglutamic acid levels within the D + S group. These markers are commonly associated with impaired detoxification or oxidative stress (Supplementary Table 4).

Fig. 4.

Detoxification biomarkers affected by dietary interventions. (A) Volcano plot of Detoxification biomarkers. This volcano plot summarizes statistical comparisons of detoxification biomarkers across experimental groups: Control (C), Diet (D), and Diet + Supplement (D + S). (B) Box-and-whisker plots of biomarkers with greater relevance. See the footnote of Fig. 1 for the plot’s description.

Moreover, metabolites such as 2‑hydroxyhippuric and glycolic acids exhibited moderate effect sizes and apparent reductions. Although not statistically significant, these trends may indicate early shifts in detoxification pathways. Similar patterns were also observed in the D group. By contrast, the C group showed no significant variation, underscoring the specific effects of the dietary strategies.

The significant decrease in urinary glyceric acid observed exclusively in the D + S group may reflect a reduction in fructose‑ and glycerol‑catabolic flux associated with microbial overactivity, rather than impaired metabolism. Glyceric acid is an intermediate of glycerate and fructose metabolism that tends to accumulate in the context of microbial fermentation or enzymatic inefficiency⁶⁴. In this controlled nutritional context, its downward shift could indicate improved metabolic handling of dietary substrates, reduced substrate overflow into microbial pathways, and enhanced enzymatic efficiency. These changes align with broader biochemical improvements observed in the D + S group.

Pyroglutamic acid is a marker of disrupted glutathione metabolism and a byproduct of the γ-glutamyl cycle⁶⁵. Its elevation suggests changes in glutathione synthesis or utilization, although elevated urinary levels of pyroglutamic acid are often associated with glutathione depletion, some studies suggest that, under certain conditions characterized by increased oxidative stress—such as sepsis—higher urinary pyroglutamate was not correlated with worse clinical outcomes⁶⁶. Further studies in populations with T21 are needed to clarify why the intervention in the D + S group increased urinary pyroglutamic acid levels in the D + S group.

Other organic acids involved in detoxification pathways—such as oxalic acid (see Supplementary Table 4)—showed consistent downward trends with large effect sizes, despite the absence of statistical significance. Oxalic acid is known for its nephrotoxic potential and mitochondria‑disrupting effects, and it often accumulates under conditions of gut dysbiosis, vitamin B6 deficiency, or impaired glyoxylate metabolism. The observed reduction in oxalic acid may indicate restored microbial balance and enhanced cofactor availability, thereby reducing renal burden and oxidative stress^67,68. This reinforces the value of interpreting both statistical and clinical indices when evaluating outcomes in heterogeneous populations such as individuals with T21.

Taken together, the biochemical data indicate consistent shifts across multiple physiological domains in response to the dietary interventions. The D + S group exhibited the most prominent changes in metabolite concentrations, with some markers reaching statistical significance and others showing moderate to very large effect sizes. The D group also showed trends toward improvement, albeit to a lesser extent. In contrast, minimal changes were observed in the C group.

These patterns were evident across dysbiosis‑associated metabolites; hematologic and nutritional markers; indicators of a hypoxia‑like state and oxidative stress; and detoxification‑related compounds. The consistency and distribution of these shifts across intervention arms suggest a distinct response profile that will be addressed further in the Discussion.

Discussion

The present study provides a comprehensive biochemical assessment of the systemic impact of dietary and supplemental interventions in individuals with T21. By targeting interconnected domains such as microbial balance, mitochondrial metabolism, redox regulation, hematologic integrity, and detoxification pathways, this intervention framework offers a multidisciplinary therapeutic strategy for a population characterized by chronic metabolic vulnerability. In contrast to reductionist approaches that focus on isolated endpoints, our findings underscore the relevance of integrated nutritional strategies for modulating the complex network of physiological disturbances in T21.

In T21, gut dysbiosis is no longer viewed as a secondary phenomenon but rather as an integral pathophysiological axis intersecting with the immune, endocrine, and central nervous systems^69,70. Chromosome 21 overexpression leads to alterations in genes encoding interferon receptors, proinflammatory cytokines, and oxidative stress modulators such as SOD1⁷¹, creating an immunometabolic environment that is both permissive to microbial imbalance and vulnerable to its consequences. This biological background predisposes individuals with T21 to chronic low‑grade inflammation, increased gut permeability, and impaired tolerance to dietary and microbial antigens⁷². As a result, the gut becomes a central organ of dysfunction, capable of perpetuating systemic inflammation, modulating neurodevelopment through the gut–brain axis, and even altering endocrine rhythms via microbial metabolites and immune signaling^73,74. The findings of this study reinforce the growing view that gastrointestinal modulation should not be ancillary to T21 care but should instead be considered a primary target for therapeutic intervention^23,72.

Furthermore, casomorphin (an opioid peptide derived from incomplete digestion of casein) was notably elevated at baseline in both intervention and control groups (see Supplementary Material). This peptide can cross the gut barrier and the blood–brain barrier, bind µ‑opioid receptors, and modulate behavior, learning, and social cognition⁷⁵. Its persistent elevation in the C group and modest reductions in the D + S group support the hypothesis that gut‑derived exorphins may be particularly relevant to the neurobehavioral phenotype of T21.

The µ‑opioid receptor system is involved not only in nociception but also in social bonding, language processing, and stress responses^76,77. Chronic stimulation of this system by dietary exorphins has been proposed to contribute to symptoms such as irritability, antisocial behaviors, and apathy—clinical features observed in a subset of individuals with T21—which may underlie behavioral patterns resembling those seen in autism spectrum disorder [81]. The dietary interventions employed in this study, particularly those excluding casein, may have reduced the intestinal production or systemic absorption of this peptide. Although the reductions in our study were not statistically significant, some clinical reports report behavioral improvement after casein exclusion diets in populations with neurodevelopmental disorders⁷⁸.

Thus, post-intervention values demonstrate a clear trend toward established biochemical reference ranges, indicating partial metabolic normalization and may underlie broader, multi-systemic improvements, suggesting a promising therapeutic avenue for neurodevelopmental disorders⁷⁹.

The data presented here illuminate the multifactorial roots of nutritional dysregulation in T21, including malabsorption, oxidative overload, and mitochondrial stress. The success of targeted nutritional interventions—particularly those involving B vitamins and antioxidant support—suggests a viable strategy for restoring systemic homeostasis and enhancing functional outcomes. Future research should focus on linking these biochemical changes to cognitive, developmental, and quality‑of‑life measures in T21, as well as testing their durability through long‑term follow‑up.

The reduction in anaerobic metabolites—most notably lactic and pyruvic acids—among participants in the D + S group strongly supports the hypothesis of improved mitochondrial efficiency in individuals with T21. These metabolites, elevated at baseline, are hallmarks of incomplete oxidative metabolism and are typically upregulated in contexts of mitochondrial dysfunction and tissue‑level hypoxia^32,58. The post‑intervention decline, supported by large rank‑biserial effect sizes, suggests a shift away from glycolytic overdependence toward more effective engagement of aerobic metabolic pathways. This improvement likely reflects enhanced mitochondrial respiration and energy efficiency, reducing cellular reliance on anaerobic glycolysis and lactate production to meet ATP demands⁵⁹.

Interestingly, even hypoxia‑like state markers that did not reach statistical significance—such as HCT—showed directional improvements with moderate effect sizes, reinforcing a coherent physiological narrative. These patterns, while underpowered statistically, align with the broader trend of hematologic normalization and support the hypothesis that mitochondrial and nutritional interventions may enhance erythropoietic quality and efficiency. Taken together, these results reflect not isolated changes but a coordinated biological adaptation across red‑blood‑cell indices, suggestive of systemic metabolic restoration.

Given the inherently elevated metabolic stress and reactive oxygen species (ROS) burden in T21, any modulation of mitochondrial function has far‑reaching implications. Mitochondrial dysfunction is a central driver of cellular aging, neurodegeneration, and immune dysregulation in T21^56,80. Improvements in oxidative metabolism may reduce compensatory ROS overproduction, decrease inflammation, and support more sustainable energy generation⁸¹. Such systemic shifts may ultimately manifest as better physical endurance, less fatigue, and improved neurocognitive functioning⁸².

Moreover, the comparative analysis across groups strengthens this conclusion: the D + S group exhibited the most robust and consistent metabolic improvements, whereas the D group showed partial trends and the C group remained largely unchanged or worsened. This dose–response relationship highlights the necessity of biochemical precision in intervention strategies for T21, in which nutritional insufficiency is often multifactorial and compounded by the effects of gene overexpression. The data suggest that while dietary improvement lays the foundation, targeted supplementation is essential to overcome endogenous metabolic barriers and elicit measurable physiological benefits.

These findings support the hypothesis that individuals with T21 experience impaired detoxification and antioxidant buffering, which can be partially restored through integrated nutritional interventions. Future studies should explore the impact of these metabolic changes on clinical manifestations such as susceptibility to infections, inflammatory markers, renal function, and cognitive fatigue. Integrating redox biomarkers, mitochondrial assays, and measures of immune reactivity into future clinical trials will be essential to determine the full systemic impact of improving detoxification in this uniquely vulnerable population.

Many biomarkers demonstrated meaningful changes without reaching statistical significance, likely due to sample‑size limitations or high inter‑individual variability characteristic of T21 populations. However, when interpreted alongside effect sizes and change indices, several of these trends reveal a high degree of physiological relevance.

Importantly, baseline biomarker distributions revealed that a substantial proportion of participants with T21 had values far outside standard reference intervals, as established in the Harriet Lane Handbook⁸³. Biomarkers such as arabinose, oxalic acid, methylmalonic acid, and RDW exceeded normal thresholds in more than 80% of participants before the intervention. This pervasive dysregulation across metabolic, hematologic, and detoxification pathways reflects the widespread systemic challenges faced by individuals with T21 and provides a strong rationale for targeted intervention.

Post‑intervention shifts in these outlier distributions—particularly in the D + S group—indicate more than statistical significance; they suggest clinical normalization. The proportion of participants whose values moved closer to standard physiological ranges increased markedly, demonstrating the capacity of the intervention to modulate systemic dysregulation rather than merely adjust isolated biochemical end points.

In conclusion, the findings of this study suggest that integrated dietary interventions—especially when combined with targeted supplementation—can produce multidimensional physiological improvements in individuals with T21. These effects span the gut–microbiota–brain axis, hematologic function, mitochondrial respiration, and detoxification pathways. Each of these systems is compromised to varying degrees in T21 due to gene overexpression, oxidative stress, and immune dysregulation, making them ideal targets for metabolic restoration.

While the relatively small sample size, coupled with a not‑entirely random assignment of groups, may limit generalizability and statistical power in some domains, the convergence of significant, near‑significant, and biologically relevant trends across independent physiological systems provides strong internal validity and is consistent with prior reports. The consistent dose–response effect observed between the D and D + S groups further reinforces the central role of biochemical supplementation in overcoming the metabolic inertia inherent to T21.

Taken together, these results lay the foundation for a new paradigm in T21 care—one that embraces metabolic precision and integrative dietary strategies to reduce systemic fragility, and potentially mitigate the early onset of age‑associated decline. Future studies should include longitudinal follow‑up, functional and cognitive end points, and larger sample sizes to confirm the long‑term impact of these interventions and to refine them into scalable clinical tools.

Methods

This investigation utilizes a database of individuals diagnosed with DS from the “Integración Down” institute in Mexico City. The database includes baseline assessments and data collected after a 10-month nutritional intervention. The intervention involved dietary modifications, specifically dietary restrictions, and/or nutrient supplementation aimed at modifying the intestinal microenvironment and the participants’ physiological profiles.

Ethics

This study, conducted in collaboration with the private “Integración Down” institute specializing in the care of children with T21, adhered to Good Clinical Practice guidelines and the Declaration of Helsinki. All participants agreed to these ethical standards. Parents or caregivers provided informed consent after receiving detailed information about the protocol interventionsThe study protocol received approval from the Bioethics and Research Committee of the Hospital Hispano in Guadalajara, Mexico (CONBIOETICA-14-CEI-001-20210428, CEI 000002 Protocol: 43). Clinical trial registration number NCT07165509 on September 3, 2025.

Eligibility

Eligible participants were individuals with Down syndrome (Trisomy 21) confirmed by karyotype, aged 2–28 years, and residents of Mexico City. Participants were allocated to three parallel groups with balanced group sizes. Group assignment was proportional to the total sample; partially random assignment was repeated until age and sex were equally represented.

Nutritional intervention protocol

• Control (C) group: Maintained their regular diet without modifications or supplementation. Evaluations were performed at the same time points as in the intervention arms.

• Diet (D) group: Followed a restricted diet eliminating:

  1. Gluten: all wheat derivatives
  2. Casein: all dairy products
  3. Refined sugar
  4. Soy
  5. Artificial colors
  6. Chemical additives

• Diet restriction + nutritional supplementation (D + S) group: In addition to the above dietary restrictions, participants received a proprietary nutritional supplement T-2 (Nutrisem™). The dosing regimen was 23 g per dose, twice daily, each dissolved in 250 mL of liquid, with flavor selected according to participant preference. The first dose was taken upon waking—ideally 30–60 min before the first meal—and the second dose in the afternoon, approximately 2 h before nighttime sleep, irrespective of dinner timing. The formulation included isomaltulose (Palatinosa™), short-chain fructo-oligosaccharides, sc-FOS (NutraFlora™), lactoferrin (Bioferrina™), ground flaxseed (BevGrad™), pyridoxine hydrochloride (vitamin B6), whey protein, and additional vitamins, minerals, fatty acids, and slowly assimilated carbohydrates.

Both pre‑ and post‑intervention (10 months later) assessments evaluated general health status through clinical and biochemical analyses.

Physiological status assessment

Routine clinical biochemical analyses (blood count, blood biochemistry, C-reactive protein, homocysteine, cortisol, and coprological analysis) were performed by Olarte & Akle Bacteriologos^® in Mexico City, Mexico, following current quality standards. Urine organic acids and peptides analyses were conducted by Great Plains Laboratory Inc.^® (actually MosaicDX’s) as shown in Table 2.

Table 2.

Clinical studies and biomarkers performed.

Clinical determinations		Biomarkers
1.	Blood count	21
2.	Blood biochemistry	12
3.	Coprological	17
4.	Organic acids	69
5.	Opioid peptides	2
	Total	121

Data analysis

For each individual, 121 biomarkers were obtained. To identify dysbiosis, nutritional deficiencies, hypoxia-like state, and detoxification issues, a subset of these biomarkers was selected based on a literature review and then classified.

Classification based on function

The biomarkers were classified into four physiological categories: dysbiosis, nutritional deficiencies, hypoxia-like state, and detoxification, based on a review of the literature describing their function^17,84.

Data scaling

Clinical studies, particularly those with small sample sizes but numerous variables, often involve data with high variability and dimensionality. This heterogeneity—compounded by differences in units, interpretation, and scale (as illustrated for the arabinose biomarker in Fig. 5)—can complicate statistical comparisons. Data normalization enhances the efficiency of analysis and modeling in such cases⁸⁵.

Fig. 5.

(A–D) Age‑ and sex‑specific reference ranges and scaling for the arabinose biomarker. Panels A–C display unscaled clinical measurements and their reference ranges stratified by age and sex: A female participants < 13 years; B male participants < 13 years; C male participants ≥ 13 years. Because reference limits differ by age and sex, measurements are plotted separately. Panel D shows the same biomarker after scaling each value to its age‑ and sex‑specific reference range (to the 0,1 interval), allowing direct comparison across all participants.

Given the observed variation in baseline values by age and sex, each biological parameter was normalized to the average of its reference values to obtain comparable, dimensionless quantities (Fig. 5A and C). Using clinical reference values for normalization enables the use of scaled data without the need for stratification by sex or age, as illustrated for arabinose in Fig. 5D. This normalization approach has been previously validated for data processing in physiological networks^86,87.

Normal reference ranges for each marker, used for data scaling, were obtained from established medical literature. To facilitate comparison, baseline and post-intervention biomarker values were scaled from 0.0 to 1.0, where 0.5 is the average of the reference values. The biomarker was scaled (Eib) according to the following function:

1

i individual, b biomarker determined, E_ib scaled value of the biomarker calculated based on reference values, V_ib Value of the biomarker obtained from the determinations, Min_b Lower limit of the biomarker reference value, Max_b Upper limit of the biomarker reference value

Data screening

The total study population comprised 22 individuals. Analyses were conducted on 21 participants with complete pre‑ and post‑intervention data across all evaluations; one participant with incomplete clinical and behavioral assessment sets was excluded from the statistical analysis. Figure 6 highlights the specific clinical assessments included in the evaluation set; descriptive characteristics are provided in Table 3.

Fig. 6.

Twenty‑two participants from the “Integración Down” Institute were enrolled and allocated to three groups: Control (C; n = 6; no dietary modifications or supplementation), Restriction Diet (D; n = 7), and Restriction Diet + Nutritional Supplementation (D + S; n = 9). All participants underwent baseline biomedical and behavioral assessments. Attrition occurred during the 10‑month protocol as depicted in the diagram. At study completion, only one participant remained in the Control group. All individuals had initial evaluations (biomedical determinations and behavioral battery). During the protocol, the sample in each group decreased as shown in each stage. At the end of the study (10 months later), only 1 participant was retained from the control group.

Table 3.

Characteristics of population and groups.

Characteristics	Control	Diet	Diet + Supplement
N	5	7	9
Age	2–23 14.8 ± 9.6	7–26 14.14 ± 7	5–28 14.56 ± 7
Body mass index (kg/m2)	24.02 ± 7.33	20.88 ± 5.76	21.04 ± 6.34
Male	4	3	6
Female	1	4	3

Statistical test and correlation

Three groups were considered for the statistical analysis. A non‑parametric approach was employed to assess the impact of each intervention scheme across the evaluated domains. The normality of each variable was assessed using the Shapiro–Wilk test. Given the predominantly non‑normal distributions, non‑parametric tests were applied. The Wilcoxon signed‑rank test was used to compare paired pre‑ and post‑intervention measurements for each biomarker. Differences between distributions were evaluated with the Kolmogorov–Smirnov test. Additionally, the rank‑biserial correlation coefficient (r) was calculated as a measure of effect size.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

EIC, APT and RHV designed the study. RHV, EIC, IHP and KIPH performed the acquisition and analysis of data. AMG, KIPH, IHP and EIC performed the statistical analysis. APT, EIC, IHP, KIPH contributed in the interpretation. All authors participated in writing and discussion of the manuscript. All authors read and approved the final version of the manuscript.

Funding

This research was funded with both national and international resources coming from governmental institutions such as CONACyT (Consejo Nacional de Ciencia y Tecnología) and private resources from Palsgaard Industri de México S. de R.L. de C.V, the sponsors were not involved in the study design, collection, analysis and interpretation of data. Public grant numbers: CONACYT/PEI 2011/PROINNOVA/157118 and CONACYT/PEI 2013/PROINNOVA/196483.

Data availability

The datasets used and analyzed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Informed consent

Informed consent was obtained from all subjects involved in the study.

Institutional review board statement

All procedures were conducted with strict adherence to the ethical standards set forth in the Declaration of Helsinki and were revised and approved by the Research Ethics Committee of the Hospital Hispano in Guadalajara, Mexico (CONBIOÉTICA-14-CEI-001-20210428).