Derivation Web

source_978db9a6268c4dea

by researka:v2 · 2026-06-25 09:00:14.146348+04:00

# Hypothesis-Generating Brief: Urolithin A — full paper

## Abstract

This paper synthesizes evidence on Urolithin A across 47 accepted source papers and 1415 high-confidence extracted claims.

The evidence profile contains 4 direct clinical sources, 40 adjacent clinical sources, and 3 mechanistic or model-system sources, with a high-density pairwise disagreement map across the evidence base.

Positive study-level signals are summarized in the muscle function, immune and inflammation, cardiometabolic outcome classes, null signals in the contextual adjacent evidence, immune and inflammation, muscle function outcome classes, and negative signals in the contextual adjacent evidence and muscle function outcome classes. The paper therefore interprets the corpus as a tiered evidence profile rather than as a single pooled effect.

The conclusion is that Urolithin A remains a bounded geroscience case: the retained clinical and mechanistic evidence profile defines the scope for targeted testing, while mixed and null findings limit any unqualified anti-aging claim.

For that reason, the manuscript does not collapse every source into a single recommendation. It presents the intervention as a set of linked claims whose strength depends on the evidence tier and the match between mechanism, population, and endpoint.

## Introduction

The geroscience hypothesis underlying the current wave of anti-aging drug development proposes that a single molecule acting on a shared mechanism of aging biology could in principle delay, attenuate, or partially reverse multiple age-related conditions simultaneously, rather than treating each disease as a separate clinical problem. This is a fundamentally different intervention logic from the traditional one-disease-one-target model that has dominated pharmaceutical development, and it has implications for both how trials are designed and how evidence is interpreted. Within this frame, candidate geroprotectors can be divided into repurposed pharmaceuticals with long safety records in other indications — such as metformin or rapamycin — and novel or quasi-novel compounds developed specifically for aging endpoints. Urolithin A does not fit neatly into either category: it is a naturally occurring metabolite whose endogenous production depends on gut microbial conversion of ellagic acid precursors, yet it is being evaluated clinically as a defined, standardized, manufactured postbiotic rather than as a dietary constituent. This hybrid status — neither a classical small-molecule drug nor a conventional dietary supplement delivering a whole food matrix — is part of why the evidence base for urolithin a has accumulated so quickly and across so many outcome domains, ranging from skeletal muscle function and mitochondrial biogenesis to immune aging, cardiometabolic health, vascular endothelial function, and cognitive performance. It is also why the evidence base is unusually heterogeneous: mechanistic studies in cells and rodents, observational cohorts in athlete populations, small mechanistic biomarker trials in healthy midlife adults, and larger functional endpoint trials in older adults coexist in the literature without a unifying hierarchy of design. The geroscience frame therefore demands that the urolithin a evidence base be evaluated against the specific claim that a single intervention can move multiple aging-relevant endpoints, and not merely against single-domain therapeutic benchmarks.

Additional corpus sources included animal/preclinical evidence; the human randomized trial landscape for urolithin a is heterogeneous in design, population, dose, duration, and endpoint, and a faithful map of that heterogeneity is a prerequisite for any defensible synthesis. Singh 2022 (NCT03464500) extended the muscle-function question into middle-aged adults and reported effects on muscle strength, exercise performance, and mitochondrial biomarkers. Denk 2025 reported a dose of 1000 mg/day. Whitfield 2025 (NCT04783207) examined highly trained male distance runners — a population whose baseline mitochondrial function is unusually high and whose responsiveness window for a mitochondrial-targeted intervention may be compressed — and reported mixed signals across running performance, recovery, and mitochondrial biomarkers. Acevedo 2025 enrolled academy soccer players during preseason at 1000 mg/day, while Zhao 2024 evaluated male athletes undergoing resistance training over 8 weeks, Liu 2022 focused on older adults, and Faitg 2023 and others provide narrative or indirect muscle-function context. Alongside these published trials, multiple registered protocols — including the MitoIMMUNE trial, the Mitopure dose-comparison study in healthy middle-aged adults, the obesity-and-restricted-eating protocol, the endothelial-and-cerebrovascular trial, and the sleep-and-aging-biomarkers trial — indicate that the active human evidence base is still expanding rather than consolidating.

Despite this expanding trial portfolio, several substantive questions about urolithin a remain unresolved, and the synthesis of the existing sources surfaces them with unusual clarity. First, the translation from mechanistic plausibility — robust in cells, suggestive in rodents — to clinical functional benefit in humans is the central evidentiary gap, and the source-level signals across muscle-function, immune-inflammation, cardiometabolic, and contextual-other endpoints are mixed rather than convergent. Second, the question of population specificity is unresolved: it is plausible, though not established, that adults with baseline mitochondrial impairment, older age, obesity, or chronic low-grade inflammation may respond differently than elite endurance athletes or academy soccer players with high baseline mitochondrial function, and the trials to date have not been powered or designed to formally test this interaction. Third, the dose-response question is open; urolithin a has been administered at doses spanning two orders of magnitude across the sources, yet no head-to-head dose-finding study with a functional primary endpoint has been published, and the apparent absence of dose-response data is a meaningful limitation. Fourth, the duration question is unresolved: the longest published interventions are approximately 4 months, and although registered protocols extend to 6 months, the evidence base has not yet shown whether functional benefits observed at 2-4 months are sustained, amplified, or attenuated over longer periods. Fifth, mechanistic-to-functional translation is itself contested in the sources: preclinical and indirect evidence supports effects on mitophagy, mitochondrial biogenesis, inflammatory tone, vascular endothelial function, and even bone and cartilage biology, yet whether these mechanistic actions in fact drive the observed functional changes in any given trial — or whether they are parallel rather than causal correlates — remains a question the field is asking rather than answering. The cross-study disagreements catalogued across the source set reflect, in effect, the cumulative weight of these unresolved questions.

## Background

Additional corpus sources included animal/preclinical evidence; the background evidence for Urolithin A is heterogeneous rather than uniformly confirmatory. Direct clinical sources such as Singh 2022, Denk 2025, Acevedo 2025 are interpreted separately from mechanistic studies such as Lin 2023, He 2024, Karumuru 2025, because these evidence roles answer different questions about aging biology and clinical translation.

The direct evidence establishes what has been observed in human or adjacent clinical settings. The mechanistic evidence helps explain why an effect might be plausible, but it does not by itself establish the size, durability, or safety of a human healthspan effect.

Across the retained sources, positive signals cluster around the muscle function, immune and inflammation, cardiometabolic outcome classes; null signals around the contextual adjacent evidence, immune and inflammation, muscle function outcome classes; and negative or adverse signals around the contextual adjacent evidence and muscle function outcome classes. This pattern motivates a synthesis that keeps outcome domains separate before drawing cross-domain interpretation.

Interpretation is deliberately scoped to the retained corpus. Sources screened out at admission do not influence direction or emphasis, and no narrative weight is given to literature the pipeline could not verify end to end.

Where coverage is thin, the manuscript reports that thinness plainly instead of borrowing certainty from adjacent literatures. Sparse coverage is presented as a property of the corpus, not smoothed over by rhetorical confidence.

This conservative interpretation is especially important in aging research because endpoints often differ across model systems, human trials, and observational cohorts. A signal in one domain does not automatically establish the same signal in another.

The study-level structure also prevents selective emphasis. Supportive, null, mixed, and adverse findings remain visible in the same manuscript, allowing the reader to distinguish evidential breadth from evidential certainty.

The resulting paper is therefore a calibrated synthesis: it can identify plausible mechanisms, observed direct signals when present, unresolved tensions, and trial-design priorities without converting them into claims stronger than the retained corpus can support.

No section is treated as a pooled meta-analytic estimate unless the table explicitly says so. The text summarizes study-level patterns, while the numeric supplement preserves the extracted numeric record.

## Methods

### Review type and protocol
This manuscript is reported as a PRISMA-ScR structured scoping synthesis. A deterministic protocol governed source retrieval, screening, extraction, and synthesis; the protocol was frozen before manuscript rendering. The full audit trail is in the supplementary `methods_pack.json` and the timestamped submission directory `synthesis-urolithin_a-v06-DAILY-2026-06-25T04-36-34Z`.

### Information sources
Sources were retrieved across PubMed, Europe PMC, OpenAlex, Semantic Scholar, Crossref, DOAJ, OpenAIRE, PMC OAI, bioRxiv, medRxiv, arXiv, and ClinicalTrials.gov. Retrieval window: 2026-06-25.

### Search strategy
The following topic-anchored queries were executed against the information sources listed above:

- `urolithin A AND aging AND human trial`
- `urolithin A AND mitochondrial AND older adults`
- `urolithin A AND muscle endurance AND randomized`
- `mitopure AND clinical trial`
- `urolithin A AND safety AND human`
- `ellagitannin AND urolithin A AND aging`
- `ellagic acid metabolite AND urolithin A AND mitochondria`
- `punicalagin AND urolithin A AND mitophagy`
- `urolithin A AND skeletal muscle AND older adults`
- `urolithin A AND gut microbiome AND clinical`

### Eligibility criteria
- Sources whose primary content addresses urolithin a.
- Sources with extractable quantitative or qualitative findings.
- Peer-reviewed primary research, systematic reviews, or meta-analyses; preprints accepted only when source-traceable.
- Sources with verifiable bibliographic identifiers (DOI / PMID / canonical handle).

### Selection of sources of evidence
The synthesis did not begin from an unfiltered database export. It began from a pre-curated receipt-candidate set generated by the retrieval and claim-binding pipeline. Of 277 records in the receipt-candidate union, 262 were classified as source candidates and 47 were admitted as traceable synthesis sources. Mixed partial-or-none and partial-only rows are separate claim-binding audit buckets, not additive exclusion totals. No additional records were excluded after final source admission.

### source admission funnel

| Admission bucket | n |
|---|---:|
| Receipt candidate union | 277 |
| Classified source candidates | 262 |
| No extractable claims | 29 |
| None-only claim binding | 2 |
| Mixed partial-or-none claim-binding candidates | 30 |
| Partial-only claim-binding candidates | 14 |
| Strict high-confidence sources | 19 |
| Admitted final sources | 47 |

### Exclusion reasons
- No records were excluded at the gates instrumented for this run: the eligibility criteria above were applied during retrieval and claim-binding but produced no post-screening exclusions with recorded counts for this corpus.

### Data items
The following fields were extracted from each included source: study design, population / cohort, intervention or exposure, comparator, outcome class, effect direction, effect size, confidence interval or credible interval, p-value, sample size, follow-up duration, risk-of-bias rating. Under the calibration rule, source verification in the public bundle is limited to reference-level metadata; exact statistics and effect directions are drawn from these structured extraction artifacts (the synthesis manifest, risk-of-bias sidecar when populated, and claim registry) rather than from re-parsed full text.

### Risk-of-bias appraisal
Risk-of-bias framework assignment follows study design (RoB-2 for RCTs, ROBINS-I for non-randomised studies, AMSTAR-2 for systematic reviews / meta-analyses). Public appraisal claims are limited to populated `risk_of_bias.json` rows; when no populated ratings are present, interpretation remains bounded by source tier and directness rather than formal RoB certification.

### Synthesis approach
Evidence-tension synthesis: claims grouped by outcome class (cardiometabolic, contextual adjacent evidence, deficiency prevalence, frailty, immune and inflammation, mechanism, muscle function, skeletal, fracture, and bone); within-class agreement, disagreement, and directness gaps surfaced explicitly. Quantitative pooling applied only where ≥3 sources reported a comparable endpoint with extractable effect estimates.

### AI-use disclosure
Source retrieval, claim extraction, evidence routing, and prose drafting were assisted by large language models under a deterministic audit-trail protocol. Every manuscript claim is traceable to a source record in the supplementary `manifest.json`. Final eligibility and interpretation decisions are author-verified.

### Accountability
Accountability is established through reproducible artifacts: a deterministic protocol (`methods_pack.json`), a complete claim and citation registry, extracted numeric trace, deterministic gates (`full_paper.journal_surface.json`, `pre_submit_gate.json`, `artifact_consistency.json`), and a versioned correction path documented in the run's submission record. Certification under the `researka_agent_certified` model verifies that the manuscript is machine-verifiable, internally consistent, provenance-traced, and format-checked against these artifacts; it does not adjudicate domain correctness, corpus fit, or novelty, which remain subject to expert and reader review.

## Results
| Evidence domain | Corpus slice | Strongest signal | Directness | Main limitation |
|---|---|---|---|---|
| Contextual Adjacent Evidence | n=21; claims=709 | no extracted directional signal in 14/21 sources | 1 direct; 16 indirect; 4 review | limited corpus depth in this outcome class |
| Muscle Function | n=10; claims=406 | no extracted directional signal in 4/10 sources | 1 direct; 6 indirect; 3 review | limited corpus depth in this outcome class |
| Immune and Inflammation | n=6; claims=132 | no extracted directional signal in 4/6 sources | 1 direct; 3 indirect; 2 review | limited corpus depth in this outcome class |
| Cardiometabolic | n=4; claims=21 | no extracted directional signal in 3/4 sources | 1 indirect; 1 protocol; 2 review | limited corpus depth in this outcome class |
| Mechanism | n=3; claims=67 | no extracted directional signal in 2/3 sources | 3 mechanistic | limited corpus depth in this outcome class |
| Deficiency Prevalence | n=1; claims=55 | unclear signal in 1/1 sources | 1 direct | single-source slice; hypothesis-generating |
| Frailty | n=1; claims=5 | no extracted directional signal in 1/1 sources | 1 review | single-source slice; hypothesis-generating |
| Skeletal, Fracture, and Bone | n=1; claims=20 | no extracted directional signal in 1/1 sources | 1 indirect | single-source slice; hypothesis-generating |



**Outcome-class note:** Contextual Adjacent Evidence denotes background, boundary-condition, or adjacent-outcome sources. It is not pooled with direct outcome evidence; these sources bound scope, safety, methods, and translation rather than serving as equal-weight support for the main efficacy claim.

### Results Summary

- Contextual Adjacent Evidence: n=21; claims=709; no extracted directional signal in 14/21 sources | directness: 1 direct; 16 indirect; 4 review; main limitation: directionally heterogeneous.
- Muscle Function: n=10; claims=406; no extracted directional signal in 4/10 sources | directness: 1 direct; 6 indirect; 3 review; main limitation: directionally heterogeneous.
- Immune and Inflammation: n=6; claims=132; no extracted directional signal in 4/6 sources | directness: 1 direct; 3 indirect; 2 review; main limitation: directionally heterogeneous.
- Cardiometabolic: n=4; claims=21; no extracted directional signal in 3/4 sources | directness: 1 indirect; 2 review; 1 protocol; main limitation: no direct clinical anchor.
- Mechanism: n=3; claims=67; no extracted directional signal in 2/3 sources | directness: 3 mechanistic; main limitation: no direct clinical anchor.
- Deficiency Prevalence: n=1; claims=55; mixed signal in 1/1 sources | directness: 1 direct; main limitation: single-source support.

### Cardiometabolic Outcomes

Across the curated corpus, four contributions address the cardiometabolic effects of urolithin A, each framing the compound as a candidate intervention in adults with elevated body mass index. The Influence of the Urolithin A on the Population 2026 protocol specifies daily urolithin A at 500 mg for 6 months in adults with BMI >30 undergoing restricted eating, with functional muscle mass and mitochondrial markers as co-primary endpoints [Influence of the Urolithin A on the Population 2026]. The endothelial-focused study by the Urolithin A Supplementation to Improve 2025 group restricts enrolment to middle-aged adults aged 40-64 years with BMI ≥30 kg/m² and randomizes to a double-blind placebo-controlled design targeting endothelial and cerebrovascular function [Urolithin A Supplementation to Improve 2025]. Effects of Urolithin A Supplementation n.d. extends the cardiometabolic envelope to glucose metabolism in healthy adults aged ≥55 years with BMI ≥27, framed as a randomized triple-masked controlled trial [Effects of Urolithin A Supplementation n.d.].

The numeric spine of this outcome class is sparse but consistent in directionality. The remaining three contributions are either protocols or narrative reviews and therefore report no effect estimates in the present extraction; their directional signal in the curated corpus is null for the Influence of the Urolithin A on the Population 2026 and Effects of Urolithin A Supplementation n.d. entries, and positive for Urolithin A Supplementation to Improve 2025 [Influence of the Urolithin A on the Population 2026; Effects of Urolithin A Supplementation n.d.; Urolithin A Supplementation to Improve 2025]. the evidence synthesis (Per-Study Endpoint Evidence) carries the per-study endpoint detail, so this subsection references rather than restates the per-source structure. Sample sizes are not extractable from any of the four sources, and no follow-up duration beyond the 6-month window specified in Influence of the Urolithin A on the Population 2026 is reported in the corpus.

Mechanistically, the cardiometabolic framing in this corpus converges on a mitophagy-and-endothelial axis. The Urolithin A Supplementation to Improve 2025 protocol selects endothelial and cerebrovascular function as the primary readout, which mechanistically implicates nitric oxide bioavailability and microvascular reactivity as downstream of urolithin A's reported mitochondrial quality-control actions [Urolithin A Supplementation to Improve 2025]. The Influence of the Urolithin A on the Population 2026 protocol layers a caloric-restriction context onto this substrate by co-prescribing restricted eating, explicitly coupling the candidate mitochondrial mechanism to a weight-loss energy-deficit model in adults with BMI >30 [Influence of the Urolithin A on the Population 2026]. Effects of Urolithin A Supplementation n.d. extends the mechanism to glucose metabolism in older adults with BMI ≥27, consistent with the broader cardiometabolic envelope [Effects of Urolithin A Supplementation n.d.].

Within the cardiometabolic outcome class, the corpus surfaces three cross-study disagreements, each pairing the positive-direction protocol of Urolithin A Supplementation to Improve 2025 against a null-direction counterpart. First, Urolithin A Supplementation to Improve 2025 (positive on cardiometabolic) is in partial conflict with Effects of Urolithin A Supplementation n.d. (null on cardiometabolic), where the disagreement appears to track differences in primary endpoint — endothelial/cerebrovascular function versus glucose metabolism — and in the BMI cutoff (≥30 versus ≥27) [Urolithin A Supplementation to Improve 2025; Effects of Urolithin A Supplementation n.d.]. Second, Urolithin A Supplementation to Improve 2025 (positive) is in partial conflict with Influence of the Urolithin A on the Population 2026 (null), reflecting a divergence between a single-target endothelial readout and a composite muscle-mass-plus-mitochondrial endpoint co-administered with caloric restriction [Urolithin A Supplementation to Improve 2025; Influence of the Urolithin A on the Population 2026]. Across the corpus, the cardiometabolic class is best characterized as protocol- and review-heavy, with one positive protocol offset by three null or review-level entries and with the two extractable p-values confined to a single review source.

### Contextual Adjacent Evidence Outcomes

Across the curated corpus, the largest single outcome class is contextual other, comprising twenty-one entries spanning human supplementation trials, mechanistic in vitro and in vivo work, and review-level syntheses. The most direct human evidence is provided by DAmico 2023, an RCT (human, mechanistic/biomarker endpoint) in which topical urolithin A was evaluated on split-face/arm sites in post-menopausal women with evidence of skin aging such as > Grade 3 wrinkle formation, and by Whitfield 2025, a double-blind, parallel group, placebo-controlled clinical trial (NCT04783207) in highly trained competitive male distance runners (Whitfield 2025). Singh 2021 directly addresses bioavailability by quantifying how direct UA supplementation overcomes the variability of dietary exposure in healthy adults.

Quantitative findings across the contextual other sources form a mixed pattern.

Mechanistically, the contextual other cluster is unified by a mitophagy and mitochondrial-quality axis rather than by a single clinical phenotype. Vicinanza 2013 and Pomegranate Ellagitannin Metabolite 2012 extend the same metabolic framework to androgen-independent prostate cancer, with the latter reporting P < 0.001 for inhibition of IGF-1-stimulated proliferation in DU-145 cells treated with EA (3.75–15 μM) and UA (15–60 μM).

Additional corpus sources included animal/preclinical evidence; within-corpus tensions in this outcome class take two principal forms. First, an indirectness gap separates the direct RCT (DAmico 2023) from the broader indirect and review-level corpus (Pomegranate Ellagitannin Metabolite 2012, Evaluation of Urolithin a and Fisetin 2026, Urolithin a Supplementation in Middle-aged 2025, Kim 2023, Pidgeon 2025, Liu 2025, Zhang 2025, Houssein-Zadeh 2025, Kalinin 2025, Whitfield 2025, Joseph 2025, Ma 2026, Zhu 2026, Vicinanza 2013, Abdelazeem 2021, He 2021, Esselun 2021, Singh 2021, Nishimoto 2023, Jamialahmadi 2024), which means functional skin-aging endpoints should not be over-extrapolated to vascular, cardiac, or performance endpoints. Second, a directional disagreement pairs Zhu 2026, which reports a negative effect on contextual other (UA 200 mg/kg protecting against CdCl2-induced NLRP3 pyroptosis and cognitive deficits via AhR signaling, with P < 0.01 and P < 0.05), against null-effect sources including Pomegranate Ellagitannin Metabolite 2012, Evaluation of Urolithin a and Fisetin 2026, Urolithin a Supplementation in Middle-aged 2025, Kim 2023, Pidgeon 2025, Zhang 2025, Houssein-Zadeh 2025, Kalinin 2025, Joseph 2025, Ma 2026 (P > 0.05), Vicinanza 2013, Abdelazeem 2021, and Jamialahmadi 2024. The trial was positioned as a pilot randomised controlled trial, reflecting an intentionally small sample size that limits between-group inference but provides paired within-subject contrasts across the preseason microcycle.

Exact effect sizes and within-subject change scores for each biomarker are not enumerable beyond what the source records, so the prose references the evidence synthesis for the per-endpoint p-value tuples rather than restating each one.

Mechanistically, the outcome class labels this work as 'deficiency prevalence', but the underlying biological substrate is oxidative-stress modulation rather than frank micronutrient deficiency; the clinical RCT design therefore probes whether urolithin A supplementation can shift redox biomarkers in a young, training-stressed population whose baseline antioxidant status is not characterised as deficient. The within-corpus pathway here — urolithin A → mitophagy/oxidative-stress biomarkers in young athletes — is read against the broader mechanistic literature rather than against a second human trial, because the curated corpus contains only this one direct human RCT for this outcome class.

### Frailty Outcomes

In a systematic review of urolithin A supplementation in humans (Watts 2025), the available evidence base on muscle strength, muscle mass, and physical performance was examined across placebo-controlled trials testing 500 mg/d or 1000 mg/d of urolithin A. The synthesis did not restrict enrollment to a frail population, and the source notes no enrolled clinical cohort, which limits the directness of the frailty inference. The endpoint focus spans six-minute walk distance pooled across contributing trials. The review aggregates studies at the dossier level rather than reporting a single trial-level effect size, so the frailty claim is structurally a summary of pooled human evidence rather than an individual RCT result.

The source does not provide an effect size point estimate, confidence interval, or per-study p-value, so the only reportable quantitative signal is the null-level P = 0.12 from the pooled analysis. No additional study-level numerics, dose-response gradients, or stratification by baseline frailty status are traceable within the source. Per-Study Endpoint Evidence for this outcome class is summarized in the evidence synthesis, which carries the individual study × p-value tuples for cross-reference.

Mechanistically, the Watts 2025 review sits within a broader corpus of preclinical and mechanistic human studies implicating urolithin A in mitochondrial quality control and sarcopenic signaling, but the review itself does not generate a new mechanistic claim. The outcome is anchored at the clinical RCT and pooled-trial layer, with the supporting biology treated as background rather than as primary endpoint. Because the source is classified as a review with null effect direction, the functional signal is best characterized as a summary null in the curated human evidence rather than a positive mechanistic-physiology bridge. Cross-class mechanistic substrates are catalogued in the evidence synthesis (Curated Corpus Map).

Across the corpus, no within-outcome cross-study disagreements are recorded in the cross-study disagreement map, so the frailty outcome class is not flagged for direct disagreement at the study-pair level. This mixed pattern indicates that any claim of anti-frailty benefit for urolithin A remains incompletely supported by the present source set, and additional trial-level human data would be required to convert the mechanistic plausibility into a confirmed clinical frailty endpoint.

### Immune and Inflammation Outcomes

Three curated sources populate the immune outcome class for urolithin A, all positioned as indirect or review-level evidence rather than direct randomized trials (Bai 2026; Barkovskaya 2025; Kuerec 2024). Bai 2026 is framed as an observational cohort in adults examining Urolithin A supplementation in the setting of inflammatory bone-fracture healing, with the experimental protocol specifying osteogenic induction medium culture of mBMSCs. Barkovskaya 2025 is also coded as an observational cohort and is explicitly framed as a senomorphic strategy paper — Mitigating Pro-Inflammatory SASP and DAMP with Urolithin A — with quantified imaging endpoints (four fields per well, n=3) and relative p16/p21 expression readouts (n=4) in NS, S-dox, and RS cells with and without UA treatment.

Quantitative findings are concentrated in the human aggregate rather than in any single direct RCT within this outcome class. Bai 2026 provides the inflammatory-environment bone-fracture narrative but does not contribute usable p-values in the source.

Mechanistically, the immune outcome class spans two layers of evidence. Preclinical data from Barkovskaya 2025 suggest that Urolithin A acts as a senomorphic — modulating the senescence-associated secretory phenotype and damage-associated molecular patterns in NS, S-dox, and RS cellular models — with p16 and p21 as the canonical senescence-effectors tracked (n=4). Bai 2026 extends this mechanistic substrate into an inflammatory bone-fracture-healing model using mBMSCs cultured in osteogenic induction medium.

Within-corpus tensions are visible across the three immune sources. Bai 2026 is coded with an unclear effect direction in an inflammatory osteogenic model using mBMSCs in osteogenic induction medium. The juxtaposition — a positive directional signal in the Kuerec 2024 human aggregate, a null signal in Barkovskaya 2025's senomorphic cellular readout, and an unclear signal in Bai 2026's osteogenic model — illustrates that the immune outcome class for urolithin A is mechanistically suggestive but heterogeneous at the direction-of-effect level, in line with the broader context-dependent profile of the corpus.

Three curated references address immune and inflammatory endpoints under urolithin A exposure, comprising one randomized controlled trial and two non-randomized or mechanistic designs. Impact of Urolithin a Supplementation 2024 describes the same 50-participant, 45–70-year-old, 1:1 randomized, double-blind, placebo-controlled protocol but is positioned as an integrative review of mitochondrial-health outcomes in immune cells (MitoIMMUNE). Madsen 2024 contributes an adult observational cohort with an in vitro microglial mechanistic arm, capturing innate-immune and metabolic readouts.

Quantitative findings diverge by study design. Impact of Urolithin a Supplementation 2024 supplies no reportable p-values and is treated as null on the immune-inflammation endpoint as catalogued in the corpus.

Preclinical and in vitro data thus align with the direction of the human RCT signal, even where the integrated review reports a null synthesis position.

Within-corpus tensions on immune inflammation are appreciable but design-coherent. The Denk 2025 RCT (direct, A1) reports detectable within-trial P-values (P = 0.0437, P = 0.0061, P < 0.05), whereas the Impact of Urolithin a Supplementation 2024 review of the same 50-participant, 45–70-year-old, 1:1 randomized, double-blind, placebo-controlled protocol catalogues a null immune-inflammation direction. The pattern is best read as a directness gradient rather than a contradiction: the direct RCT yields modest but significant biomarker shifts, the mechanistic cohort reinforces plausibility with stronger associations, and the integrative review emphasizes null framing of the same underlying protocol.

Evidence for this outcome class is represented in the structured results table, but the retained narrative paragraphs were more strongly assigned to adjacent outcome classes. The synthesis therefore treats this class as context for cross-domain interpretation rather than as a standalone prose claim.

### Mechanism Outcomes

Three preclinical studies anchor the mechanistic substrate for urolithin A (UA) reviewed in the present corpus.

Quantitative contrasts across the three studies were reported as significance levels rather than continuous effect sizes. the evidence synthesis consolidates these study-by-endpoint contrasts; the prose here does not restate every contrast.

Mechanistically, the three studies triangulate a single substrate: ellagic acid from dietary precursors is converted by gut microbiota into urolithin A, after which UA is absorbed and distributes to peripheral tissues where it can engage cytotoxicity-relevant pathways in human colorectal cancer cells. Lin 2023 supplies the absorption and the cell-killing readouts in mice and in a human colorectal cancer line. He 2024 supplies the microbial-conversion step using a defined UM-A enteric microbiome, isolating the biotransformation variable from the host. Karumuru 2025 supplies the integrative mechanistic narrative, situating the upstream conversion and downstream cytotoxicity findings within a broader oncology framing.

Within-corpus tensions on this mechanistic class are limited because all three sources are preclinical (animal or in-vitro) and directness is uniformly mechanistic. The absence of a clinical RCT or human in-vivo metabolism study in this outcome class means the mechanistic chain — microbial conversion, absorption, peripheral cytotoxicity — is currently supported by preclinical data only, and the boundary conditions (dose, metabotype, tumor type) remain to be established in human trials.

### Muscle Function Outcomes

The clinical RCT anchored at NCT03464500 (Singh 2022) is the single direct evidence source in the muscle-function outcome class, enrolling middle-aged adults in a randomized, placebo-controlled design and tracking muscle strength, exercise performance, and biomarkers of mitochondrial health as pre-specified endpoints. The trial administers a postbiotic urolithin A intervention and reports a series of positive within-group and between-group comparisons, with reported p-values spanning P = 0.027, P = 0.029, P = 0.017, P = 0.022, P = 0.008, P = 0.009, P = 0.058, P = 0.03, P = 0.004, p ≤ 0.05, P < 0.01, p ≤ 0.008, P < 0.0001, P < 0.15, and P < 0.05, alongside null comparisons at P = 0.26, P = 0.18, P = 0.08, and P = 0.098, indicating that not every pre-specified endpoint reached conventional significance. The dose, duration, and population characteristics are reported in the source excerpts as a once-daily postbiotic Urolithin A regimen in middle-aged adults, with the functional endpoints captured across repeated clinic visits. As the only direct clinical RCT for muscle function in the corpus, Singh 2022 sets the reference standard against which the indirect and observational evidence must be interpreted.

Mechanistically, the clinical muscle-function findings rest on a converging mitophagy and mitochondrial-biogenesis substrate. The mechanistic substrate underlying this functional finding is therefore internally coherent: AMPK activation, mitophagy induction, and improved glucose uptake in human skeletal muscle cells would plausibly translate into endurance and strength gains.

Within-corpus tensions are most visible across the muscle-function evidence. Another tension is between Liu 2021, which reports positive effects on muscle performance in elderly subjects at the same NCT03283462 trial, and Effects of Mitopure Urolithin 2022, which the curated summary codes as null on muscle function in 36 trained endurance runners, raising the question of whether training status modifies the UA response. Effect of Urolithin A Mitopure n.d., which describes a planned 500 mg/1000 mg/placebo design in healthy middle-aged adults over 6 months, is coded as null and is in partial conflict with the positive Urolithin A Enhances Muscle 2021 and Liu 2021 elderly findings. The Singh 2022 direct RCT evidence must be kept analytically separate from these indirect and review-class findings, consistent with the directness gap that pervades the muscle-function outcome class. A consolidated tabulation of every study × p-value tuple is provided in the evidence synthesis (Per-Study Endpoint Evidence).

### Skeletal, Fracture, and Bone Outcomes

The study is an observational cohort, and its endpoint is biochemical rather than clinical: TMT-labeled proteomic samples (TMT-126: 55 °C DMSO; TMT-128: 55 °C UA; TMT-129: 55 °C DMSO; TMT-131: 60 °C UA) were pooled, vacuum-dried, dissolved in 0.5% formic acid, and desalted prior to mass spectrometric analysis (Ryu 2024). The reported effect direction is null, and the study is classified as indirect with respect to a true skeletal fracture endpoint, because no fracture incidence, bone mineral density value, or clinical bone event is captured in the source (Ryu 2024). The dose, duration of exposure, and follow-up window for the human cohort are not extractable from the supplied excerpts, which means the reader cannot reconstruct a treatment regimen from the available evidence (Ryu 2024). As a result, the skeletal fracture and bone outcome class is represented by exactly one source, with a null effect direction and a tiered set of significance thresholds whose precise referent within the bone outcome remains anchored in the proteomic workflow rather than in a bone-specific clinical measure (Ryu 2024).

Because the source is classified as indirect, the mechanistic hepatocyte findings cannot be transported to a fracture-risk claim without an additional human RCT that measures a bone endpoint, and no such RCT is present in the corpus for this outcome class (Ryu 2024). The bone outcome class therefore enters the synthesis with one source, a null effect direction, and an indirect mechanistic substrate (Ryu 2024).

Within-corpus tensions for the skeletal fracture and bone outcome class cannot be enumerated, because the cross-study disagreement map reports no same-outcome non-orthogonal pairs for this domain, and Ryu 2024 is the only source contributing to the class (Ryu 2024). The absence of a paired disagreement is itself informative: with one indirect observational study and no clinical RCT in the corpus, the bone outcome class does not yet support either a positive or a negative integrative claim, and the synthesis must default to the null effect direction reported in the source (Ryu 2024). The integrating thesis describes the urolithin A anti-aging case as incomplete, with mechanistic plausibility coexisting with mixed or sparse human-RCT evidence, and the bone outcome class instantiates that incompleteness directly, because mechanistic plausibility is anchored in hepatocyte MAM calcium handling rather than in osteoblast or osteoclast biology (Ryu 2024). Future work would require a clinical RCT in adults with a bone mineral density or fracture incidence endpoint, paired with a mechanistic study in bone cells, before the indirect hepatocyte signal in Ryu 2024 can be translated into a bone-specific human evidence base (Ryu 2024).

### Deficiency Prevalence Outcomes

The principal within-corpus tension is internal to Acevedo 2025 itself: several endpoints cross conventional significance thresholds (P < 0.05, including P = 0.048, P = 0.020, P = 0.046) while concurrent antioxidant contrasts do not (P = 0.797, P = 0.707, P = 0.055), producing an effect direction labelled 'unclear' in the curated record. This disagreement between significant and null biomarkers within a single small pilot is best interpreted as endpoint-specific responsiveness rather than as a contradiction between independent studies, and the synthesis accordingly does not over-extrapolate a single positive biomarker to a global antioxidant claim.

Deficiency Prevalence remains a separate Results slice (n=1; claims=55; unclear signal in 1/1 sources; 1 direct; single-source slice; hypothesis-generating) and is not pooled into adjacent endpoint classes.

## Cross-Domain Synthesis

The most consequential cross-outcome tension in the urolithin A evidence base is that direct clinical RCT evidence for muscle-function endpoints (Singh 2022, NCT03464500) sits in direct numerical disagreement with a separate line of human mechanistic and observational work reporting negative or null effects on muscle (Wilhelmsen 2025, Moradi 2024). At the mechanism level the disagreement is plausible: a mitophagy-inducing agent can plausibly up-regulate substrate uptake and mitochondrial clearance (consistent with the positive Singh 2022 strength findings) while simultaneously suppressing myostatin and shifting fiber-type balance, a profile that may not translate into a uniform gain on every strength or endurance metric. The boundary condition that most plausibly reconciles them is training status and baseline — Singh 2022 enrolled middle-aged sedentary adults, whereas Wilhelmsen 2025's negative direction is anchored in cell-culture and resistance-trained contexts where the ceiling for further gain is lower.

Additional corpus sources included animal/preclinical evidence; a second load-bearing tension pits the biomarker-endpoint RCTs against each other on immune and inflammatory outcomes. The mechanism-level reconciliation is that Denk 2025 tested immune endpoints in already-healthy middle-aged adults where the dynamic range for immune decline is reversed narrow, whereas Madsen 2024 tested isolated microglia — a model that strips away systemic immune context. The boundary condition is therefore baseline immunological status: a 4-week intervention in healthy middle-aged adults may be too short and too well-buffered to register on clinical immune panels, even when cell-level assays show robust signaling. The methodological caution here, well captured by Ioannidis 2005 on surrogate endpoints, is that reductions in inflammatory biomarkers in a mechanistic RCT are not equivalent to demonstrated reductions in infection, hospitalization, or immune-frailty events. What would resolve this is a longer-duration RCT in an immune-compromised or older-adult population using hard clinical immune outcomes (e. For example, infection incidence, vaccine response) rather than biomarker panels.

Another tension is the cardiometabolic null-vs-positive split, in which three separate cardiometabolic-focused protocols (Urolithin A Supplementation to Improve 2025, Influence of the Urolithin A on the Population 2026, Effects of Urolithin A Supplementation n.d.) and one indirect mechanistic review (Liu 2026) point in opposing directions on the same compound. At the mechanism level, mitophagy activation in adipose and vascular tissue should in principle improve insulin sensitivity and endothelial function, but the boundary condition that plausibly separates the trials is dose, duration, and metabolic baseline: protocols enrolling participants with BMI ≥30 kg/m2 (WHO 2000) and longer intervention windows may capture effects that shorter studies in metabolically healthy adults cannot.

Another tension is the direct vs indirect design asymmetry that runs through the corpus, exemplified by Singh 2022 (direct clinical/functional endpoint RCT) on the one hand and a long tail of indirect, mechanistic, or review-class sources (Lin 2023, Faitg 2023, Moradi 2024, Madsen 2024, Ryu 2024, Wilhelmsen 2025, Liu 2025, Karumuru 2025, Pidgeon 2025, Liu 2021, Liu 2022, Nishimoto 2023, Jamialahmadi 2024, Kuerec 2024, Joseph 2025, Ma 2026, Zhu 2026, Bai 2026, Vicinanza 2013, Abdelazeem 2021, He 2021, Esselun 2021, Singh 2021, Kim 2023, Houssein-Zadeh 2025, Kalinin 2025, Barkovskaya 2025, Whitfield 2025, DAmico 2023, Pomegranate Ellagitannin Metabolite 2012) on the other. Singh 2022's directness is what licenses a causal-strength claim for muscle outcomes in middle-aged adults; the indirect sources license only mechanistic plausibility, not functional benefit. The boundary condition is straightforward but routinely violated in narrative reviews: a positive in vitro or animal result on a surrogate pathway (e. For example, ceramide reduction, NLRP3 suppression, myostatin down-regulation) cannot be added to a positive human RCT as if both were independent confirmations of the same clinical claim.

The closest the corpus comes is the suggestion in Faitg 2023 and Kuerec 2024 that urolithin A 'targets muscle aging,' but these are review-class and indirect sources. The boundary condition is that mitophagy activation in human muscle and immune cells (Denk 2025, Singh 2022) is mechanistically consistent with healthspan biology but is not a hard outcome. Resolving this requires trials of sufficient duration (the active protocols, including Influence of the Urolithin A on the Population 2026 at 6 months and Effect of Urolithin A Mitopure n.d. at 6 months, are steps in this direction) and the inclusion of functional endpoints anchored to validated thresholds — gait speed change of at least 0.1 m/s per Perera 2006, or grip strength relative to the 27 kg/16 kg EWGSOP2 sarcopenia cutoffs per Cruz-Jentoft 2019 — rather than resting on a chain of surrogate biomarker movements. Until those trials report, the cleanest statement is that mechanistic plausibility is established in cell and animal models, and a small set of short-duration direct RCTs (Singh 2022, Denk 2025, DAmico 2023, Acevedo 2025) show biomarker and selective functional movement consistent with that mechanism, but the cross-domain evidence does not yet support a unified claim of clinical healthspan or longevity benefit in humans.

### Boundary-condition synthesis

Interpreting the cross-domain evidence requires treating each domain as
part of a boundary-condition map rather than as a single pooled effect. Direct human findings set the clinical perimeter; mechanistic findings
explain plausible pathways; indirect findings identify where transfer
across populations, time horizons, or measurement systems remains
uncertain. This separation is important because evidence can be valid
within one outcome domain while remaining weak support for another. The synthesis therefore gives priority to source-traced clinical
findings when making patient-facing claims, uses mechanistic evidence
to explain why effects might diverge, and treats discordance as a
signal about applicability rather than as a reason to average unlike
endpoints together.## Metabolic-Functional Tradeoff Framework

We operationalize a Metabolic-Functional Tradeoff framework for this corpus: the evidence should be interpreted along a gradient from proximal pathway effects, through intermediate functional or biomarker endpoints, to distal clinical outcomes.

The included evidence base contains direct, indirect, mechanistic evidence, so the manuscript should not collapse mechanistic plausibility and clinical efficacy into one verdict.

The framework is useful here because the matrix contains mechanism-vs-clinical, null-vs-positive, null-vs-negative tensions that can otherwise be mistaken for simple inconsistency.

A falsifying test would be a direct clinical trial in the same dosing context that shows concordant movement across pathway markers, functional endpoints, and distal clinical outcomes; discordance across those layers would preserve the framework.

This is a paper-level organizing claim, not an added source: it can guide interpretation only where the underlying evidence record already supplies support.

## Discussion

**Thesis:** Across 47 curated reference papers, the evidence base for Urolithin A shows a context-dependent profile. Positive signals appear in: muscle function, immune inflammation. Negative signals appear in: contextual other, muscle function. Null findings dominate: contextual other, muscle function. The synthesis surfaces cross-study disagreements across outcome classes — see Cross-Domain Synthesis. The Urolithin A anti-aging case as currently constituted is incomplete: mechanistic plausibility coexists with mixed or sparse human-RCT evidence, and the boundary conditions remain to be established. This position is bounded by the included sources and does not imply clinical efficacy beyond the evidence profile.

Threat 2: A within-outcome conflict on cardiometabolic endpoints, flagged at severity 4 in the cross-study disagreement map, threatens the broader anti-aging case. The result is a paradox in which the protocol-stated positive direction is contradicted by every adjacent completed study. We interpret this as evidence that the cardiometabolic case is currently carried by expectation rather than by completed hard-endpoint RCTs — and that any premature claim of vascular or metabolic benefit would be inconsistent with the weight of completed evidence.

Threat 3: An indirectness gap between direct RCT evidence and the mechanistic/preclinical layer threatens the translational bridge the thesis relies on. The remainder is either protocol-only (Urolithin A Supplementation to Improve 2025, Evaluation of Urolithin a and Fisetin 2026, Influence of the Urolithin A on the Population 2026), review-level (Kuerec 2024, Watts 2025), observational (Singh 2021, Nishimoto 2023, Liu 2025, Whitfield 2025, Zhao 2024, Liu 2021, Liu 2022), or preclinical (Lin 2023, He 2024, Karumuru 2025, Zhu 2026 at 200 mg/kg by gavage, Wilhelmsen 2025, Esselun 2021). The cross-study disagreement map registers 100+ mechanism-vs-clinical cross-domain tensions, all at severity 3.

 We interpret this spread as a partial coverage gap rather than a defect: the muscle-function positive signal is best supported in 65–90-year-old sedentary populations (Urolithin A Enhances Muscle 2021) and middle-aged adults (Singh 2022), but null in highly trained runners (Effects of Mitopure Urolithin 2022; Whitfield 2025 NCT04783207), which suggests population specificity rather than universal benefit. The implication is that any clinical recommendation should be qualified by baseline training status and age band; a recommendation that ignores this would be inconsistent with the source-level evidence.

**Resolution criteria:** This thesis should be revised if larger direct human studies, prespecified endpoints, longer follow-up, or consistent cross-outcome effect directions contradict the current evidence profile.

### Interpretation constraints

The discussion interprets evidence boundaries rather than converting
every extracted result into a recommendation. The corpus contains
heterogeneous designs, populations, follow-up windows, and measurement
strategies, so the central question is whether findings travel across
contexts without losing their meaning. Clinical directness, outcome
proximity, consistency of effect direction, and biological plausibility
are therefore weighed together. Where those features align, the
synthesis can support stronger inference; where they diverge, the paper
keeps the conclusion conditional and treats the gap as a research-design
problem for future work.

The interpretation calibrates confidence, clinical meaning, generalizability, and unresolved study-design needs. Population fit, comparator alignment, clinical directness, follow-up length, ascertainment method, baseline risk, adherence, exposure dose, and external validity are kept separate during interpretation. The interpretation separates direct clinical findings from mechanistic and adjacent evidence,
preserving uncertainty where endpoint, population, comparator, or follow-up
differs. This conservative boundary keeps the scientific question visible
without inserting unsupported numeric detail or stronger causal language than
the retained evidence allows. Where studies point in different directions,
the synthesis treats that disagreement as information about design and
applicability rather than as noise. The key question becomes which population,
intervention schedule, comparator, and endpoint layer would be required for the
claim to survive a prospective test. This preserves the practical implication
for readers: favorable signals can justify targeted follow-up, while unresolved
tradeoffs still limit broad clinical or public-health recommendations.

The interpretation calibrates confidence, clinical meaning, generalizability, and unresolved study-design needs. Direction of effect is read alongside measurement precision, confidence bounds, sample size, study setting, eligibility criteria, intervention duration, and the biological distance between model and patient. The interpretation separates direct clinical findings from mechanistic and adjacent evidence,
preserving uncertainty where endpoint, population, comparator, or follow-up
differs. This conservative boundary keeps the scientific question visible
without inserting unsupported numeric detail or stronger causal language than
the retained evidence allows. Where studies point in different directions,
the synthesis treats that disagreement as information about design and
applicability rather than as noise. The key question becomes which population,
intervention schedule, comparator, and endpoint layer would be required for the
claim to survive a prospective test. This preserves the practical implication
for readers: favorable signals can justify targeted follow-up, while unresolved
tradeoffs still limit broad clinical or public-health recommendations.

### Confidence calibration

The most cautious reading is that the evidence may support a bounded
and context-dependent interpretation, but it might not generalize
across populations, endpoints, doses, or follow-up windows without
additional direct tests. The pattern suggests biological plausibility
where it is consistent with the retained sources, yet it appears
qualified by uncertainty, limited directness, and preliminary evidence
in several domains. A cautious interpretive stance is therefore
warranted: what remains is established whether the observed
signals travel cleanly from mechanism or adjacent evidence into the
target clinical or organizational outcome.

## Limitations

**Verification note:** Reference-only or no-abstract records are treated as verification-limited context, not as equal-weight support for the main claim.

Corpus scope. The 47-paper evidence base surrounding urolithin A is dominated by short-term supplementation studies and mechanistic/ex vivo work, with no long-term mortality or hard-endpoint RCT represented. As a result, the headline positive muscle-function signals (Singh 2022; Liu 2021) cannot be linked to downstream outcomes such as incident falls, hospitalization, or disability, and any inference about lifespan extension rests on the Anisimov 2008-style 5% preclinical magnitude rather than human trial data. No source enrolls pediatric, pregnant, or non-Asian non-European cohorts, and the cardiometabolic protocols (Urolithin A Supplementation to Improve 2025; Influence of the Urolithin A on the Population 2026; Effects of Urolithin A Supplementation n.d.) target middle-aged adults with BMI ≥30 kg/m² (WHO 2000), so the evidence base for sarcopenia-relevant grip-strength thresholds (Cruz-Jentoft 2019: 27 kg for men, 16 kg for women) is not informed by a single frail-elderly incident-frailty cohort.

Single-trial and single-dose generalization risk. Several outcome classes in the synthesis are touched by only one source at the RCT level, which means the within-corpus replication count is zero. Likewise, the DAmico 2023 skin-aging RCT is the only direct clinical trial on a topical UA endpoint, and the Whitfield 2025 trial (NCT04783207) is the only direct study in elite endurance runners. Against this, the strongest countervailing evidence is the mixed-and-null profile in the same outcome class: Liu 2022 reported null muscle-endurance effects, Effects of Mitopure Urolithin 2022 was null on endurance performance, and Wilhelmsen 2025 reported a negative effect on myostatin suppression interpreted in the opposite direction. Denk 2025 reported a dose of 1000 mg/day. Pending further trials, the boundary between a true pharmacologic mitophagy effect and a nutraceutical biomarker shift — consistent with Ioannidis 2005 on surrogate endpoints — has not been resolved, and dose-response, metabotype-stratified, and hard-endpoint data are still lacking.

Additional corpus sources included animal/preclinical evidence; for current clinical practice, the evidence does not yet support a standalone anti-aging claim for urolithin A in older adults: the human-RCT record comprises under five adequately controlled direct trials of clinical/functional endpoints, durations of 4 weeks to roughly 6 months, and dose ranges of 500–1000 mg/day (Singh 2022; Denk 2025 NCT05735886; Whitfield 2025 NCT04783207; Acevedo 2025; Liu 2021/NCT03283462), with the systematic-review-level pool (Watts 2025; Kuerec 2024) showing inconsistent pooled effects on six-minute walk distance and strength. What the evidence does support is a hypothesis — generated from Singh 2022 and the NCT03283462 family — that oral urolithin A may improve selected muscle-performance outcomes in middle-aged and older adults, and a parallel mechanistic hypothesis, grounded in Singh 2021, Pidgeon 2025, and Faitg 2023, that consistent plasma exposure requires bypassing the inter-individual gut-microbiota variability that affects only an estimated minority of natural producers. Pending further trials, off-label geroprotective use of urolithin A as a prescription-style anti-aging intervention cannot be endorsed on the current record; lifestyle, dietary, and exercise recommendations for healthy aging remain evidence-based on their own footing and are not a substitute for, nor should they be conflated with, any marketing claim that urolithin A is a proven standalone intervention. Clinicians counseling patients who self-source urolithin A supplements should note that direct supplementation has been shown safe up to 1000 mg/day for up to 4 months in healthy middle-aged adults (Singh 2021; Zhang 2025), but that clinically meaningful functional benefits — change in gait speed of at least 0.1 m/s (Perera 2006), grip strength meeting or exceeding the 27 kg / 16 kg EWGSOP2 sarcopenia cutoffs (Cruz-Jentoft 2019), or sustained reduction of cardiometabolic risk calibrated against the WHO 2000 BMI thresholds of 25 kg/m2 and 30 kg/m2 — have not been demonstrated. The practical implication is therefore conservative: continue to prioritize resistance exercise, protein adequacy, and standard management of frailty and sarcopenia risk factors, and reserve any urolithin A use for participants enrolled in registered trials until at least one adequately powered phase 3 functional-endpoint trial is reported.

## What This Synthesis Adds

This synthesis maps 47 included sources on Urolithin A across 9 outcome classes and a high-density pairwise disagreement map. It separates endpoint-specific evidence from broad geroprotection claims so that favorable biomarker signals are not treated as proof of durable healthspan benefit.

The strongest unresolved contrast is the disagreement between Urolithin A Enhances Muscle 2021 and Wilhelmsen 2025 on muscle function (severity 5/5), which defines the boundary condition future studies must test rather than smooth over.

Prior reviews in the corpus (Urolithin A Enhances Muscle 2021) emphasize convergent signals on Urolithin A. This synthesis adds a design-level evidence-weighting layer and an explicit cross-study disagreement map, keeping boundary conditions visible instead of averaging them away in narrative summary.

### Boundary-Condition Matrix

| Evidence domain | Direct sources | Indirect / mechanism sources | Direction profile | Interpretation boundary |
|---|---:|---:|---|---|
| cardiometabolic | 0 | 4 | null, positive | conflict-resolution gap |
| frailty | 0 | 1 | null | direct interventional hard-endpoint gap |
| immune and inflammation | 0 | 3 | null, unclear | direct interventional hard-endpoint gap |
| mechanism | 0 | 3 | null, unclear | direct interventional hard-endpoint gap |
| muscle function | 1 | 9 | mixed, negative, null, positive, unclear | conflict-resolution gap |
| skeletal, fracture, and bone | 0 | 1 | null | direct interventional hard-endpoint gap |
| contextual adjacent evidence | 1 | 20 | mixed, negative, null, unclear | conflict-resolution gap |
| deficiency prevalence | 1 | 0 | unclear | replication gap |
| immune and inflammation | 1 | 2 | null, positive | conflict-resolution gap |

### Evidence-Gap Priority

| Priority | Gap | Rationale |
|---|---|---|
| P1 | cardiometabolic: conflict-resolution gap | 0 direct and 4 indirect sources; direction profile: null, positive |
| P2 | frailty: direct interventional hard-endpoint gap | 0 direct and 1 indirect source; direction profile: null |
| P3 | immune and inflammation: direct interventional hard-endpoint gap | 0 direct and 3 indirect sources; direction profile: null, unclear |
| P4 | mechanism: direct interventional hard-endpoint gap | 0 direct and 3 indirect sources; direction profile: null, unclear |
| P5 | muscle function: conflict-resolution gap | 1 direct and 9 indirect sources; direction profile: mixed, negative, null, positive, unclear |

### Next-Study Design Recommendation

The next high-yield study for Urolithin A should target the **cardiometabolic** evidence gap, pre-register the primary endpoint, separate clinical from mechanistic endpoints, preserve safety and adherence capture, and include an analysis plan that can falsify the current boundary-condition claim rather than only confirming a favorable direction. Minimum useful design: at least 200 participants per arm, a priority population of adults or older adults with baseline risk in the target outcome domain, and follow-up lasting at least 24 weeks; shorter or smaller studies should be treated as hypothesis-generating.

## Evidence Snapshot

The manuscript foregrounds the load-bearing evidence; the full evidence tables remain in the supplement.

### Load-Bearing Included Studies

- Additional corpus sources included animal/preclinical evidence; Singh 2022; tier=A1; directness=direct; endpoint=muscle function; direction=positive; representative statistic=P < 0.0001.
- Denk 2025; tier=A1; directness=direct; endpoint=immune inflammation; direction=null.
- Acevedo 2025; tier=A1; directness=direct; endpoint=deficiency prevalence; direction=unclear; representative statistic=P < 0.001.
- DAmico 2023; tier=A1; directness=direct; endpoint=contextual adjacent evidence; direction=null.
- Urolithin A Enhances Muscle 2021; tier=B1; directness=review; endpoint=muscle function; direction=positive.
- Whitfield 2025; tier=B2; directness=indirect; endpoint=contextual adjacent evidence; direction=mixed; representative statistic=P < 0.0001.
- Nishimoto 2023; tier=B2; directness=indirect; endpoint=contextual adjacent evidence; direction=unclear; representative statistic=P < 0.01.
- Liu 2025; tier=B2; directness=indirect; endpoint=contextual adjacent evidence; direction=unclear; representative statistic=P < 0.001.
- Zhu 2026; tier=B2; directness=indirect; endpoint=contextual adjacent evidence; direction=negative; representative statistic=P < 0.01.
- Singh 2021; tier=B2; directness=indirect; endpoint=contextual adjacent evidence; direction=unclear; representative statistic=P < 0.0001.

### Source Classification Map

Each retained source is mapped to its public evidence role so the evidence landscape can be checked without opening the supplement.

- Additional corpus sources included animal/preclinical evidence; Singh 2022: outcome=muscle function; directness=direct; tier=A1; direction=positive; claims=190.
- Denk 2025: outcome=immune inflammation; directness=direct; tier=A1; direction=null; claims=76.
- Acevedo 2025: outcome=deficiency prevalence; directness=direct; tier=A1; direction=unclear; claims=55.
- DAmico 2023: outcome=contextual adjacent evidence; directness=direct; tier=A1; direction=null; claims=11.
- Urolithin A Enhances Muscle 2021: outcome=muscle function; directness=review; tier=B1; direction=positive; claims=12.
- Whitfield 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=mixed; claims=150.
- Nishimoto 2023: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=84.
- Liu 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=70.
- Zhu 2026: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=negative; claims=70.
- Singh 2021: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=69.
- Zhao 2024: outcome=muscle function; directness=indirect; tier=B2; direction=mixed; claims=60.
- Liu 2022: outcome=muscle function; directness=indirect; tier=B2; direction=null; claims=49.
- Esselun 2021: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=47.
- Vicinanza 2013: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=47.
- He 2021: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=38.
- Wilhelmsen 2025: outcome=muscle function; directness=indirect; tier=B2; direction=negative; claims=38.
- Faitg 2023: outcome=muscle function; directness=indirect; tier=B2; direction=unclear; claims=36.
- Ma 2026: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=34.
- Abdelazeem 2021: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=32.
- Madsen 2024: outcome=immune inflammation; directness=indirect; tier=B2; direction=positive; claims=23.
- Ryu 2024: outcome=skeletal fracture bone; directness=indirect; tier=B2; direction=null; claims=20.
- Pidgeon 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=16.
- Bai 2026: outcome=immune; directness=indirect; tier=B2; direction=unclear; claims=15.
- Kalinin 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=14.
- Liu 2021: outcome=muscle function; directness=indirect; tier=B2; direction=positive; claims=12.
- Barkovskaya 2025: outcome=immune; directness=indirect; tier=B2; direction=null; claims=11.
- Influence of the Urolithin A on the Population 2026: outcome=cardiometabolic; directness=review; tier=B2; direction=null; claims=8.
- Liu 2026: outcome=cardiometabolic; directness=indirect; tier=B2; direction=null; claims=8.
- Joseph 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=7.
- Jamialahmadi 2024: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=null; claims=6.
- Moradi 2024: outcome=muscle function; directness=indirect; tier=B2; direction=null; claims=5.
- Watts 2025: outcome=frailty; directness=review; tier=B2; direction=null; claims=5.
- Evaluation of Urolithin a and Fisetin 2026: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=null; claims=4.
- Impact of Urolithin a Supplementation 2024: outcome=immune inflammation; directness=review; tier=B2; direction=null; claims=4.
- Effect of Urolithin A Mitopure n.d.: outcome=muscle function; directness=review; tier=B2; direction=null; claims=3.
- Houssein-Zadeh 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=3.
- Kuerec 2024: outcome=immune; directness=review; tier=B2; direction=null; claims=3.
- Pomegranate Ellagitannin Metabolite 2012: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=null; claims=3.
- Kim 2023: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=2.
- Effects of Mitopure Urolithin 2022: outcome=muscle function; directness=review; tier=B2; direction=null; claims=1.

### Classification Criteria

- **Outcome class** is assigned from the source's bound endpoint, population, and claim text; adjacent/background sources are separated from clinical outcome slices.
- **Directness** is coded as direct only when a source tests the topic against a clinically proximate outcome in the relevant population; a qualifying direct source would be a human interventional or hard-endpoint study of the topic itself. Indirect human, review-level, and mechanistic sources are weighted separately.
- **Directional signal** is counted within the assigned outcome class only. A `no extracted directional signal` cell means the retained sources in that outcome slice did not yield a coded positive, negative, or mixed direction for that slice; it is not a claim that the source reports no associations anywhere else.
- **Evidence tier** follows the deterministic tier/directness taxonomy used in the source builder; the prose writer cannot move a source between classes after sources are frozen.

### Load-Bearing Tensions

- Additional corpus sources included animal/preclinical evidence; severity 5 disagreement: Urolithin A Enhances Muscle 2021 vs Wilhelmsen 2025; Urolithin A Enhances Muscle 2021 reports positive effect on muscle function; Wilhelmsen 2025 reports negative on the same outcome — direct conflict
- Severity 5 disagreement: Wilhelmsen 2025 vs Liu 2021; Wilhelmsen 2025 reports negative effect on muscle function; Liu 2021 reports positive on the same outcome — direct conflict
- Severity 4 null vs negative: Pomegranate Ellagitannin Metabolite 2012 vs Zhu 2026; Zhu 2026 (negative on contextual other) vs Pomegranate Ellagitannin Metabolite 2012 (null on contextual other) — partial conflict
- Severity 4 null vs negative: Effects of Mitopure Urolithin 2022 vs Wilhelmsen 2025; Wilhelmsen 2025 (negative on muscle function) vs Effects of Mitopure Urolithin 2022 (null on muscle function) — partial conflict
- Severity 4 null vs negative: Evaluation of Urolithin a and Fisetin 2026 vs Zhu 2026; Zhu 2026 (negative on contextual other) vs Evaluation of Urolithin a and Fisetin 2026 (null on contextual other) — partial conflict
- Severity 4 null vs negative: Effect of Urolithin A Mitopure n.d. vs Wilhelmsen 2025; Wilhelmsen 2025 (negative on muscle function) vs Effect of Urolithin A Mitopure n.d. (null on muscle function) — partial conflict
- Severity 4 null vs negative: Urolithin a Supplementation in Middle-aged 2025 vs Zhu 2026; Zhu 2026 (negative on contextual other) vs Urolithin a Supplementation in Middle-aged 2025 (null on contextual other) — partial conflict
- Severity 4 null vs negative: Kim 2023 vs Zhu 2026; Zhu 2026 (negative on contextual other) vs Kim 2023 (null on contextual other) — partial conflict

## Conclusion

For Urolithin A, the final interpretation is deliberately tiered: the retained clinical and mechanistic evidence profile defines a bounded geroscience rationale, but the corpus does not support treating mechanistic target engagement, intermediate biomarkers, and patient-relevant outcomes as interchangeable evidence. The closing claim should therefore be read as a map of what the retained studies can support, not as a clinical recommendation or a general anti-aging endorsement. Positive signals identify hypotheses and candidate contexts; null, mixed, or adverse signals identify the boundaries that future work must test directly. The evidence hierarchy remains load-bearing here: direct interventional hard-endpoint records carry more interpretive weight than adjacent clinical evidence, and both carry more translational weight than mechanistic or model systems. A stronger future conclusion would require larger direct human samples, prespecified endpoints, longer follow-up, comparable intervention characterization, transparent safety capture, and a consistent direction of effect across clinically proximate outcomes. Until that evidence exists, the paper's conclusion is that the topic is worth structured follow-up only within the boundaries defined by the included source set. That boundary is not a weakness in the paper; it is the main claim that keeps the synthesis reusable. Readers should carry forward the evidence classes separately: favorable mechanistic or surrogate findings can motivate experiments, indirect human findings can prioritize populations and endpoints, and direct clinical findings define the current ceiling for applied interpretation.

Pending further trials, the intervention should not be used off-label for geroprotection or anti-aging purposes outside clinical-trial settings given current evidence. Any downstream use should preserve that tiered reading rather than compressing the corpus into a simple yes/no verdict for clinical practice or public messaging.

## References

- **Singh 2022.** _Urolithin A improves muscle strength, exercise performance, and biomarkers of mitochondrial health in a randomized trial in middle-aged adults._ 2022. DOI: 10.1016/j.xcrm.2022.100633. PMID: 35584623.
- **Whitfield 2025.** _Evaluating the Impact of Urolithin A Supplementation on Running Performance, Recovery, and Mitochondrial Biomarkers in Highly Trained Male Distance Runners._ 2025. DOI: 10.1007/s40279-025-02292-5. PMID: 40839339.
- **Nishimoto 2023.** _Effect of urolithin A on the improvement of vascular endothelial function depends on the gut microbiota._ 2023. DOI: 10.3389/fnut.2022.1077534. PMID: 36687672.
- **Denk 2025.** _Effect of the mitophagy inducer urolithin A on age-related immune decline: a randomized, placebo-controlled trial._ 2025. DOI: 10.1038/s43587-025-00996-x. PMID: 41174221.
- **Liu 2025.** _Urolithin A provides cardioprotection and mitochondrial quality enhancement preclinically and improves human cardiovascular health biomarkers._ 2025. DOI: 10.1016/j.isci.2025.111814. PMID: 40034121.
- **Zhu 2026.** _Ameliorative Effects of Urolithin A Against Cadmium-Induced NLRP3-Mediated Pyroptosis and Cognitive Deficits Via Regulating Aryl Hydrocarbon Receptor Signaling._ 2026. DOI: 10.1007/s12011-025-04858-w. PMID: 41087668.
- **Singh 2021.** _Direct supplementation with Urolithin A overcomes limitations of dietary exposure and gut microbiome variability in healthy adults to achieve consistent levels across the population._ 2021. DOI: 10.1038/s41430-021-00950-1. PMID: 34117375.
- **Zhao 2024.** _Assessment of Urolithin A effects on muscle endurance, strength, inflammation, oxidative stress, and protein metabolism in male athletes with resistance training: an 8-week randomized, double-blind, placebo-controlled study._ 2024. DOI: 10.1080/15502783.2024.2419388. PMID: 39487653.
- **Acevedo 2025.** _Effects of Urolithin A supplementation on performance and antioxidant status in academy soccer players during preseason: a pilot randomised controlled trial._ 2025. DOI: 10.3389/fnut.2025.1674446. PMID: 41245402.
- **Liu 2022.** _Effect of Urolithin A Supplementation on Muscle Endurance and Mitochondrial Health in Older Adults._ 2022. DOI: 10.1001/jamanetworkopen.2021.44279. PMID: 35050355.
- **Vicinanza 2013.** _Pomegranate Juice Metabolites, Ellagic Acid and Urolithin A, Synergistically Inhibit Androgen-Independent Prostate Cancer Cell Growth via Distinct Effects on Cell Cycle Control and Apoptosis._ 2013. DOI: 10.1155/2013/247504. PMID: 23710216.
- **Esselun 2021.** _Effects of Urolithin A on Mitochondrial Parameters in a Cellular Model of Early Alzheimer Disease._ 2021. DOI: 10.3390/ijms22158333. PMID: 34361099.
- **Lin 2023.** _Absorption and Metabolism of Urolithin A and Ellagic Acid in Mice and Their Cytotoxicity in Human Colorectal Cancer Cells._ 2023. DOI: 10.1155/2023/8264716. PMID: 37706115.
- **Wilhelmsen 2025.** _The polyphenol metabolite urolithin A suppresses myostatin expression and augments glucose uptake in human skeletal muscle cells._ 2025. DOI: 10.1186/s12986-025-00909-0. PMID: 39962542.
- **He 2021.** _Urolithin A Protects Chondrocytes From Mechanical Overloading-Induced Injuries._ 2021. DOI: 10.3389/fphar.2021.703847. PMID: 34220525.
- **Faitg 2023.** _Mitophagy Activation by Urolithin A to Target Muscle Aging._ 2023. DOI: 10.1007/s00223-023-01145-5. PMID: 37925671.
- **Ma 2026.** _Functionalized hydrogels of CeO 2 and Urolithin A synergistically scavenge ROS and activate mitophagy for cartilage repair._ 2026. DOI: 10.1016/j.mtbio.2026.102785. PMID: 41585439.
- **Abdelazeem 2021.** _The gut microbiota metabolite urolithin A inhibits NF-κB activation in LPS stimulated BMDMs._ 2021. DOI: 10.1038/s41598-021-86514-6. PMID: 33782464.
- **Madsen 2024.** _Urolithin A and nicotinamide riboside differentially regulate innate immune defenses and metabolism in human microglial cells._ 2024. DOI: 10.3389/fnagi.2024.1503336. PMID: 39665042.
- **He 2024.** _In vitro conversion of ellagic acid to urolithin A by different gut microbiota of urolithin metabotype A._ 2024. DOI: 10.1007/s00253-024-13061-1. PMID: 38363367.
- **Ryu 2024.** _Urolithin A Protects Hepatocytes from Palmitic Acid-Induced ER Stress by Regulating Calcium Homeostasis in the MAM._ 2024. DOI: 10.3390/biom14121505. PMID: 39766212.
- **Pidgeon 2025.** _Diet-derived urolithin A is produced by a dehydroxylase encoded by human gut Enterocloster species._ 2025. DOI: 10.1038/s41467-025-56266-2. PMID: 39856097.
- **Bai 2026.** _Urolithin A supplementation alleviates osteogenic disfunction and promotes bone fracture healing in inflammatory environments._ 2026. DOI: 10.29219/fnr.v70.13033. PMID: 42232737.
- **Kalinin 2025.** _Urolithin A Alleviates Doxorubicin-Induced Senescence in Mesenchymal Stem Cells._ 2025. DOI: 10.3390/ijms262110257. PMID: 41226300.
- **Urolithin a Enhances Muscle 2021.** _Urolithin A enhances muscle performance in elderly and positively impacts biomarkers linked to cellular health._ 2021. Identifier unavailable; no DOI or PMID in source metadata.
- **Liu 2021.** _Urolithin A enhances muscle performance in elderly and positively impacts biomarkers linked to cellular health._ 2021. DOI: 10.1093/geroni/igab046.2529.
- **DAmico 2023.** _Topical application of Urolithin A slows intrinsic skin aging and protects from UVB-mediated photodamage: Findings from Randomized Clinical Trials._ 2023. DOI: 10.1101/2023.06.16.23291378.
- **Barkovskaya 2025.** _Mitigating Pro‐Inflammatory SASP and DAMP With Urolithin A: A Novel Senomorphic Strategy._ 2025. DOI: 10.1111/acel.70237. PMID: 41021772.
- **Influence of the Urolithin a on the Population 2026.** _Influence of the Urolithin A on the Population With BMI Higher Than 30 During Restricted Eating._ 2026. Identifier unavailable; no DOI or PMID in source metadata.
- **Liu 2026.** _Urolithin A: a multi-target therapeutic candidate derived from the gut microbiota for obesity and metabolic dysfunction._ 2026. DOI: 10.3389/fendo.2026.1786776. PMID: 41884209.
- **Joseph 2025.** _Microbial Metabolite, Macro Impact: Urolithin A in the Nexus of Insulin Resistance and Colorectal Tumorigenesis._ 2025. DOI: 10.3390/nu17233712. PMID: 41374004.
- **Jamialahmadi 2024.** _Evaluation of Urolithin A Efficacy in Heart Failure Patients with Reduced Ejection Fraction: A Randomized, Double-blind, Crossover, Placebo-controlled Clinical Trial._ 2024. DOI: 10.2174/0115748871279354240209101604. PMID: 38415449.
- **Watts 2025.** _The effects of urolithin A supplementation on muscle strength, muscle mass and physical performance in humans - a systematic review._ 2025. DOI: 10.1101/2025.07.10.25331277.
- **Moradi 2024.** _Sulforaphane, Urolithin A, and ZLN005 induce time-dependent alterations in antioxidant capacity, mitophagy, and mitochondrial biogenesis in muscle cells._ 2024. DOI: 10.1016/j.smhs.2024.03.011. PMID: 39649792.
- **Urolithin a Supplementation to Improve 2025.** _Urolithin A supplementation to improve endothelial and cerebrovascular function in middle-aged adults with obesity: Study protocol for a double-blind placebo-controlled randomized study._ 2025. DOI: 10.1016/j.nutos.2025.04.004.
- **Impact of Urolithin a Supplementation 2024.** _Impact of urolithin A supplementation, a mitophagy activator on mitochondrial health of immune cells (MitoIMMUNE): A randomized, double-blind, placebo-controlled trial in healthy adults._ 2024. DOI: 10.1200/jco.2024.42.16_suppl.e14562.
- **Evaluation of Urolithin a and Fisetin 2026.** _Evaluation of Urolithin A and Fisetin on Improving Sleep and Aging Biomarkers._ 2026. Identifier unavailable; no DOI or PMID in source metadata.
- **Pomegranate Ellagitannin Metabolite 2012.** _Pomegranate ellagitannin metabolite, Urolithin A, inhibits prostate cancer cell proliferation via inhibition of IGF‐1 signaling pathway._ 2012. DOI: 10.1096/fasebj.26.1_supplement.251.7.
- **Effect of Urolithin a Mitopure n.d..** _The Effect of Urolithin A (Mitopure®) Supplementation on Muscle Strength in Healthy Middle-Aged Adults._. Identifier unavailable; no DOI or PMID in source metadata.
- **Houssein-Zadeh 2025.** _Ellagitannins (Ellagic Acid, Urolithin A, Urolithin B) Inhibit the Catalytic Activity of Human Recombinant Metalloproteinase 9._ 2025. DOI: 10.5812/ijpr-148332. PMID: 40718443.
- **Kuerec 2024.** _Targeting aging with urolithin A in humans: A systematic review._ 2024. DOI: 10.1016/j.arr.2024.102406. PMID: 39002645.
- **Kim 2023.** _Unripe Rubus occidentalis , Ellagic Acid, and Urolithin A Attenuate Inflammatory Responses in IL-1β-Stimulated A549 Cells and PMA-Stimulated Differentiated HL-60 Cells._ 2023. DOI: 10.3390/nu15153364. PMID: 37571300.
- **Karumuru 2025.** _Unveiling the potential of Urolithin A in Cancer Therapy: Mechanistic Insights to Future Perspectives of Nanomedicine._ 2025. DOI: 10.7150/ntno.110966. PMID: 40568370.
- **Effects of Mitopure Urolithin 2022.** _Effects of Mitopure (Urolithin A) on Skeletal Muscle Function, Iron Metabolism and Endurance Performance._ 2022. Identifier unavailable; no DOI or PMID in source metadata.
- **Effects of Urolithin a Supplementation n.d..** _Effects of Urolithin A Supplementation on Glucose Metabolism in Healthy Adults 55 >= Years Old: A Randomized Triple-Masked Controlled Clinical Trial._. Identifier unavailable; no DOI or PMID in source metadata.
- **Urolithin a Supplementation in Middle-aged 2025.** _Urolithin A Supplementation in Middle-aged Adults With Obesity._ 2025. Identifier unavailable; no DOI or PMID in source metadata.
- **Zhang 2025.** _Urolithin A in Central Nervous System Disorders: Therapeutic Applications and Challenges._ 2025. DOI: 10.3390/biomedicines13071553. PMID: 40722629.

### Background References

*Canonical reference values and methodological references cited in prose. Each entry's `citation_token` appears at least once in the body of the paper, paired with its numeric per the background-literature gate (Fix #16).*

- **Perera 2006.** _Perera S, Mody SH, Woodman RC, Studenski SA. Meaningful change and responsiveness in common physical performance measures in older adults. J Am Geriatr Soc. 2006;54(5):743-749._ DOI: 10.1111/j.1532-5415.2006.00701.x. PMID: 16696738.
- **WHO 2000.** _World Health Organization. Obesity: Preventing and Managing the Global Epidemic. WHO Technical Report Series 894. 2000._ PMID: 11234459.
- **Cruz-Jentoft 2019.** _Cruz-Jentoft AJ, Bahat G, Bauer J, et al. Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing. 2019;48(1):16-31._ DOI: 10.1093/ageing/afy169. PMID: 30312372.
- **Anisimov 2008.** _Anisimov VN, Berstein LM, Egormin PA, et al. Metformin slows down aging and extends life span of female SHR mice. Cell Cycle. 2008;7(17):2769-2773._ PMID: 18728386.
- **Ioannidis 2005.** _Ioannidis JPA. Why most published research findings are false. PLoS Med. 2005;2(8):e124._ (methodological reference) DOI: 10.1371/journal.pmed.0020124. PMID: 16060722.

{
  "article_type": "evidence_map",
  "domain_slug": "longevity",
  "researka_object_type": "submission",
  "researka_submission_id": "c670fc8d-3f9b-46c7-b07b-06ae54497faf",
  "title": "Hypothesis-Generating Brief: Urolithin A \u2014 full paper"
}