DT-109 for Health & Longevity

Evidence Review created on 09/09/2026 using AI4L / Opus 5

Also known as: Gly-Gly-L-Leu, Gly-Gly-Leu, GGL, DT109, Diapin

Motivation

DT-109 is an experimental compound built from three linked amino acids: two units of glycine and one of leucine. Glycine is one of the smallest building blocks of protein, and people with fatty liver disease, poor blood sugar control, or narrowed arteries consistently carry less of it in their blood. That observation led researchers to ask whether restoring glycine, in a form the body absorbs efficiently, might correct something upstream of those conditions rather than manage them one at a time.

The compound came out of a university laboratory and has since been administered orally to mice and to monkeys fed rich diets. Reported effects reach across the liver, the arteries and blood sugar control, which is unusual for a single small molecule. It has never been given to people in any published study, and it is sold nowhere as a medicine or as a supplement.

This review examines what the animal and laboratory record actually shows about DT-109, where that record is thin, single-sourced or internally contradictory, and what remains unknown about its behavior and safety in humans.

Benefits - Risks - Protocol - Conclusion

Recommended Reading

High-level sources that explain what DT-109 is, what has been shown about it, and who stands behind the work.

• Glycine-based treatment ameliorates NAFLD by modulating fatty acid oxidation, glutathione synthesis, and the gut microbiome - Rom et al., 2020

  The paper that introduced DT-109 for nonalcoholic fatty liver disease (NAFLD, fat build-up in the liver). Its authors patented the compound and founded the university start-up that licensed it.

• Ammonia in the crosshairs: microbial targets for metabolic dysfunction-associated steatohepatitis prevention - Leone & Kennedy, 2026

  An independent commentary, from authors outside the developing consortium, on how DT-109 lowers bacterial ammonia in metabolic dysfunction-associated steatohepatitis (the inflamed, scarring stage of fatty liver disease).

• Drug candidate successfully treats atherosclerosis, fatty liver disease in large mammals - Noah Fromson

  A plain-language account of the monkey artery-plaque results, carrying the disclosure that the university, the investigators and their start-up hold a financial interest in the compound.

• Glycine supplementation could help improve NAFLD - D Dye

  A plain-language Life Extension summary of the founding Michigan study, explaining why low circulating glycine matters in fatty liver disease and how DT-109 outperformed glycine and leucine in mice.

Only four items are listed rather than five. Six papers are indexed under the name DT-109, alongside one earlier report of the same compound under its original name Diapin, and no eligible blog post, podcast episode, lecture or independent expert commentary beyond the four above could be found. Of the six priority expert platforms, only Life Extension has covered the compound; web searches and direct on-site searches of foundmyfitness.com, peterattiamd.com, hubermanlab.com, chriskresser.com and lifespan.io returned nothing on DT-109. The list has deliberately not been padded with catalogue pages or press-release rewrites.

Grokipedia

No Grokipedia article on DT-109 exists.

Examine

No Examine article on DT-109 exists. Examine covers dietary supplements, nutrients and food compounds; DT-109 is an unapproved investigational compound that is neither sold as a supplement nor prescribed as a medication, so it falls outside the site’s coverage.

ConsumerLab

No ConsumerLab article on DT-109 exists. ConsumerLab tests and reviews commercially sold supplements and nutrition products; DT-109 is an unapproved investigational compound with no consumer product to test, so it falls outside the site’s coverage.

Systematic Reviews

No systematic reviews or meta-analyses for DT-109 were found on PubMed as of September 9, 2026. Both sides of the trade-off are therefore unrepresented: there is no systematic review or meta-analysis of the claimed liver and vascular benefits, and none covering the principal risk, which is the complete absence of human safety data.

Mechanism of Action

DT-109 is Gly-Gly-L-Leu, a three-amino-acid chain that works as a delivery vehicle for glycine rather than a receptor ligand. Glycine is the final, rate-limiting substrate for building glutathione, the cell’s main internal antioxidant. Carbon-tracing in liver cells shows DT-109 supplies the glycine that unblocks this step. In treated livers, genes for fatty acid oxidation — the pathway that burns fat — rise under PPARα (its master switch), while NF-κB (a master switch for inflammatory genes) and TGFβ/SMAD signaling (the scar-formation pathway) fall. In artery wall cells it lowers NLRP3 (an immune sensor that triggers inflammation) and RUNX2 (a switch that turns muscle cells into bone-like calcifying cells). In the gut it reduces ammonia-producing bacteria and restores barrier integrity. Glucose lowering runs by a separate route: oral dosing raised insulin and the gut hormone glucagon-like peptide-1 in diabetic mice.

Two mechanistic readings compete. The originating work reports peak concentrations in mouse plasma and liver roughly 30 minutes after oral administration, implying systemic delivery. A later study from the same consortium found the compound below the limit of detection in plasma and concluded its action is largely confined to the gut. Which reading holds decides whether this is a systemic metabolic agent or a gut-acting one.

Pharmacological properties are largely uncharacterized: no half-life is published in any species, there is no receptor selectivity, tissue distribution rests on the conflicting measurements above, and metabolism is presumed to be hydrolysis by intestinal peptidases to glycine and leucine, with no cytochrome P450 (CYP) drug-metabolizing enzymes implicated.

Historical Context & Evolution

DT-109’s ancestor was not a liver drug. The same laboratory first described the tripeptide, then called Diapin, as a glucose-lowering agent for type 2 diabetes, chosen because it lowered blood glucose in mice more effectively than free glycine, free leucine, or the two-amino-acid combinations. Its move into liver disease came from a separate line of evidence: circulating glycine is repeatedly found low in people with fatty liver disease, diabetes and coronary disease, and liver tissue from those patients shows suppressed AGXT1 (the liver enzyme that makes glycine from glyoxylate). Deleting that enzyme in mice worsened diet-induced liver injury, which reframed low glycine from a passive marker into a candidate cause.

The interest of the health-optimization field followed from a third thread. Glutathione falls with age, glycine is its limiting ingredient, and glycine supplementation extends lifespan in mice, so a more potent glycine carrier looked like a longevity tool rather than only a liver drug.

What has changed since is the standard of proof demanded. Earlier antioxidant programs reached large human trials and produced null or harmful results, and a series of liver drug candidates failed in phase 3 after convincing rodent data. Both experiences pushed this field toward nonhuman primate models before human testing, which is why DT-109’s monkey studies exist at all. That shift is a change in method, not a settled verdict on whether raising glycine helps humans.

Expected Benefits

High 🟩 🟩 🟩

No benefit reaches High: every reported outcome is a rodent, nonhuman primate or cell-culture measurement, and no human clinical endpoint or validated human surrogate has been measured for DT-109 in more than one trial — or in any trial.

Medium 🟩 🟩

No benefit reaches Medium either: the class of evidence needed here is a single human trial or a consistent human observational dataset, and neither exists for DT-109; the entire evidence base is animal and in-vitro work.

Low 🟩

Speculative 🟨

Reduction of Liver Fat and Steatohepatitis

In mice with diet-induced steatohepatitis and in monkeys, oral DT-109 lowered liver fat, liver enzymes and biopsy activity score. The evidence is entirely preclinical; no human imaging or biopsy data exist.

Attenuation of Liver Fibrosis

Collagen staining showed less scarring in treated mice, and fibrosis progression was prevented in monkeys, through reduced TGFβ signaling. No human fibrosis measurement has been made.

Reduction of Arterial Plaque Burden

In cholesterol-fed monkeys, five months of oral DT-109 cut atherosclerosis (plaque build-up in artery walls) in the aorta and coronary arteries. One species, one study, males only.

Reduction of Arterial Calcification

Treated monkeys showed less calcium deposition in the aortic arch and thoracic aorta, and cultured smooth muscle cells calcified less. All data are animal and cell-culture only.

Lower Blood Glucose

Oral DT-109 lowered glucose more than equivalent glycine or leucine in mouse tolerance tests, and cut non-fasting glucose. No glucose result has been published in monkeys or in people.

Improved Blood Lipid Profile ⚠️ Conflicted

Mice on DT-109 had lower total and LDL (low-density lipoprotein) cholesterol; monkeys showed only transient falls that disappeared over five months. Net reading: a lipid effect is unproven beyond rodents.

Increased Glutathione and Lower Oxidative Stress

Carbon-tracing in hepatocytes and macrophages shows DT-109 supplies the glycine limiting glutathione synthesis. Glutathione and oxidation markers are unvalidated surrogates, not outcomes.

Restoration of the Gut Barrier and Lower Gut Ammonia

In monkeys and mice, DT-109 reduced ammonia-producing Clostridium perfringens, lowered intestinal ammonia and restored barrier integrity. No human stool, ileal or barrier data have been reported.

Reduced Fat Mass with Preserved Lean Mass ⚠️ Conflicted

Mice on the higher dose lost body fat while keeping lean mass; monkeys showed no body-weight change. Net reading: a body-composition effect appears in rodents only.

Benefit-Modifying Factors

• Baseline glycine and glutathione status: The animal effects all run through relieving a glycine shortage. Where circulating glycine and whole-blood glutathione are already normal, the substrate step is not limiting and less change is expected.

• Genetic variation in glycine handling: Variants in AGXT (the enzyme that makes glycine from glyoxylate) and GLDC (the enzyme that breaks glycine down) set how much of a dose becomes usable glycine rather than oxalate.

• Fatty liver genotype: Carriers of PNPLA3 I148M (a variant that impairs liver fat export and raises steatohepatitis risk) start from a worse liver phenotype and, on the mechanism, have more room for improvement.

• Sex: Unmeasured rather than absent. Every animal study used males only, chosen to avoid estrogen-driven variability in fat metabolism, so no claim about female benefit rests on data.

• Pre-existing metabolic disease: Benefits were largest in animals with established steatohepatitis or diet-induced atherosclerosis; healthy chow-fed mice still showed glucose and lipid shifts, so disease amplifies rather than gates the effect.

• Age: Glutathione synthesis declines across the lifespan, which widens the gap a glycine carrier could close; against that, older kidneys clear an oxalate load less well. No aged animal has been dosed.

Potential Risks & Side Effects

High 🟥 🟥 🟥

No risk reaches High: the class of evidence required is a documented human adverse event or a validated human safety surrogate shown in more than one trial, and DT-109 has never been administered to a human in any published study.

Medium 🟥 🟥

No risk reaches Medium either: that would require a single human trial or consistent human observational safety data, and neither exists; every risk below is mechanistic, animal-derived, or extrapolated from the free amino acids the compound carries.

Low 🟥

Speculative 🟨

Absence of Any Human Safety Data

No phase 1 study, tolerability report or toxicology package for DT-109 is in the public record. Every safety statement below is an inference from rodents, monkeys and free amino acids, not an observation.

Oxalate Load and Kidney Stone Risk

Glycine feeds the glyoxylate-to-oxalate route, and a projected gram-scale daily dose delivers two glycine units per molecule. Urinary oxalate has never been measured with DT-109 in any species.

Gastrointestinal Intolerance

Gram-scale oral amino acid loads commonly cause nausea, bloating and loose stools, and the gut-localized action reported for DT-109 concentrates exposure there. No tolerability data exist at any human dose.

Sedation and Lowered Core Temperature

Glycine is an inhibitory neurotransmitter that lowers core body temperature and promotes sleepiness at gram doses. Daytime dosing could therefore impair alertness; this has never been tested with the tripeptide.

Leucine-Driven mTOR Activation

Each dose delivers leucine, the strongest amino acid activator of mTOR (a cellular growth-signaling hub). Chronic mTOR activation shortens lifespan in model organisms, which sits awkwardly with a longevity rationale. Untested for DT-109.

Nitrogen Load in Liver or Kidney Impairment

A peptide load adds nitrogen that must be disposed of through the urea cycle, the opposite direction from the compound’s gut ammonia-lowering effect. No study has dosed an animal with impaired liver or kidney function.

Long-Term Suppression of Resident Gut Clostridia

DT-109 suppresses Clostridium perfringens and related taxa. The consequences of chronically depressing a resident genus over years are unknown; all microbiome observations span three to five months.

Blunted Adaptive Signaling from Sustained Glutathione Elevation

Raising glutathione dampens the reactive oxygen species signals that drive adaptation to exercise and other stressors. Whether the sustained glutathione induction reported in animals carries this cost has not been examined.

Risk-Modifying Factors

• Kidney function: The oxalate route is cleared renally. A reduced estimated glomerular filtration rate concentrates the risk of calcium oxalate crystal formation, making impaired kidneys the single most important modifier of this compound’s risk.

• Genetic variation in glyoxylate handling: AGXT mutation carriers (primary hyperoxaluria type 1, in which glyoxylate is diverted to oxalate instead of glycine) convert a glycine load toward oxalate far more readily than non-carriers.

• Baseline biomarkers: Raised 24-hour urinary oxalate, raised plasma ammonia, or a low estimated glomerular filtration rate each mark a person for whom the nitrogen and oxalate load carries more consequence.

• Sex: Unknown by design. Male-only animal studies leave female-specific risks, including any interaction with estrogen-driven lipid handling, entirely unexamined.

• Pre-existing conditions: Cirrhosis (advanced liver scarring) and urea cycle disorders impair nitrogen disposal; a stone history marks oxalate susceptibility; treatment with clozapine introduces a documented glycine interaction.

• Age: Kidney clearance, urea cycle capacity and intestinal peptidase activity all decline with age, so the same dose represents a larger metabolic load at 70 than at 40.

Key Interactions & Contraindications

• Glucose-lowering agents (metformin, glipizide, insulin, empagliflozin): Caution — additive glucose lowering risks hypoglycemia (blood sugar falling too low), since DT-109 outperformed equivalent glycine in mouse glucose tests. Mitigation: continuous glucose monitoring and downward titration of the existing agent.

• Clozapine and other antipsychotics (olanzapine, risperidone): Caution — glycine acts at the glycine site of the NMDA receptor (a brain receptor for the signal glutamate); added glycine reduced clozapine’s benefit while adding it to olanzapine and risperidone improved negative symptoms. Mitigation: psychiatric review first.

• Sedatives and sleep agents (zolpidem, benzodiazepines such as lorazepam): Caution — glycine’s inhibitory signaling and temperature drop add to these agents, causing excess drowsiness. Mitigation: dose separation by several hours, and avoidance of tasks needing alertness until tolerance is established.

• Lipid-lowering drugs (atorvastatin, ezetimibe): Monitor — any lipid effect would be additive, and the animal lipid signal is inconsistent. Mitigation: repetition of the lipid panel at four and twelve weeks rather than assumption of either added benefit or interference.

• High-dose vitamin C (1 g/day or more) and other over-the-counter oxalate precursors: Caution — vitamin C and a glycine load both feed the glyoxylate-to-oxalate route, raising urinary oxalate and stone risk. Mitigation: a cap of 500 mg/day on vitamin C.

• Sedating over-the-counter antihistamines (diphenhydramine, doxylamine) and melatonin: Caution — additive sedation and lowered core temperature, expected to be mild and fully reversible. Mitigation: dose separation by several hours.

• Glycine, collagen and gelatin supplements: Monitor — these supply the same amino acid, so combining them multiplies the glycine load and its oxalate and sedation consequences with no known added benefit. Mitigation: counting total glycine from all sources.

• N-acetylcysteine: Additive by design — cysteine is the other limiting ingredient of glutathione, so the pair raises glutathione more than either alone. The consequence is intended rather than adverse. Mitigation: none beyond the standard panel.

• Berberine and other glucose-lowering supplements: Caution — additive glucose lowering with the same hypoglycemia consequence as prescription agents. Mitigation: staggered introduction by at least four weeks so any glucose fall can be attributed to one agent.

• Leucine, branched-chain and essential amino acid blends: Monitor — each DT-109 dose already delivers leucine, and the combined mTOR activation is the additive effect of concern. Mitigation: subtraction of the leucine already supplied from any separate intake.

• Glucagon-like peptide-1 receptor agonists (semaglutide, tirzepatide): Monitor — both target the same liver endpoint by different routes, and DT-109 changed neither body weight nor glucagon-like peptide-1 in monkeys, so overlap is unlikely. Mitigation: none identified.

• Prolonged fasting and ketogenic diets: Monitor — both raise fatty acid oxidation through the same gene switch DT-109 activates, and the combined effect is untested. Mitigation: introduction of one variable at a time.

Populations who should avoid DT-109:

• Anyone with a history of calcium oxalate kidney stones, or with primary hyperoxaluria of any type
• Chronic kidney disease with an estimated glomerular filtration rate below 30 mL/min/1.73 m²
• Cirrhosis of Child-Pugh Class B or C, or any diagnosed urea cycle disorder
• Anyone taking clozapine
• Pregnancy and lactation
• Anyone under 18 years of age

Risk Mitigation Strategies

• Dose escalation from a fraction of the projected dose: Extrapolated human doses land near 3 g daily; starting at 500 mg and rising over four weeks limits the gastrointestinal intolerance and sedation that large amino acid loads produce.

• Kidney and oxalate surveillance: Estimated glomerular filtration rate, serum creatinine and a 24-hour urinary oxalate collection before starting and every three months address the stone and kidney injury risk created by the glycine-to-oxalate route.

• Fluid intake targeting 2–2.5 L of urine daily: Dilute urine is the single most effective countermeasure to calcium oxalate stone formation, which is the most concrete of the mechanistic risks identified above.

• Glucose monitoring through the first eight weeks: Continuous or twice-daily fingerstick monitoring catches the additive hypoglycemia risk when DT-109 is combined with metformin, insulin or a sulfonylurea (a drug class that pushes the pancreas to release insulin).

• Evening dosing: Confining intake to the two hours before bed converts glycine’s drowsiness and core temperature drop from a daytime hazard into a neutral or favorable one.

• Identity and purity verification of any sourced material: Mass spectrometry identity, high-performance liquid chromatography purity above 98%, and confirmation of the L-leucine isomer address the contamination and wrong-isomer risks of research-grade supply.

• Liver panel at baseline, four weeks and twelve weeks: Alanine aminotransferase and aspartate aminotransferase catch idiosyncratic liver injury, which has never been excluded because no human has been dosed.

Therapeutic Protocol

• No established human protocol: No clinician or clinic uses DT-109. It has never been administered to a human in a published study, so nothing below is a protocol in practice — only what the animal record implies.

• Originating group: The compound was developed by the Chen laboratory at the University of Michigan and licensed to its start-up, Diapin Therapeutics. Every dosing decision in the literature traces back to that single group.

• Animal dosing on record: Mice received 0.125–1 mg per gram of body weight daily orally; cynomolgus monkeys received 150 mg/kg daily orally for five months while remaining on a cholesterol-rich diet.

• Arithmetic human-equivalent extrapolation: Applying the standard mouse conversion factor of 12.3 to the effective 0.5 mg/g dose gives roughly 40 mg/kg daily, and the monkey factor of 3.1 gives roughly 48 mg/kg — about 3 g daily at 70 kg.

• Competing approach, plain glycine: Doses of 3–15 g daily appear in human studies of glycemic control and sleep. This is the cheap comparator DT-109 was designed to beat, and no head-to-head human comparison has been run.

• Competing approach, glycine with N-acetylcysteine: The Baylor College of Medicine group led by Sekhar pairs the two glutathione precursors in older adults, an approach that has human randomized trial data which DT-109 lacks.

• Best time of day: Not established. The animal work dosed in the morning, while the sedative property of a glycine load argues for evening. No timing comparison exists in any species.

• Half-life: Unpublished everywhere. Mouse plasma and liver concentrations peaked about 30 minutes after an oral dose, with hepatic glutathione rising at 60–120 minutes, implying short exposure and at least daily dosing.

• Single versus split dosing: Animal studies used one daily dose. The short time to peak and the gram-scale load argue for splitting, which would also reduce gastrointestinal intolerance; neither pattern has been tested in humans.

• Genetic polymorphisms: AGXT variant carriers face a larger oxalate diversion from the same dose. PNPLA3 I148M carriers, whose liver fat export is impaired, are the group the mechanism predicts would respond most.

• Sex-based differences: Unknown. Every mouse and monkey study used males only, deliberately, to avoid estrogen-driven variability in fat metabolism, leaving female response and female dosing entirely unmeasured.

• Age considerations: Glutathione synthesis and kidney clearance both fall with age, so older adults have more headroom for the intended effect and less margin for the oxalate load. No aged animal has been studied.

• Baseline biomarkers: Low fasting glycine, low whole-blood glutathione, raised liver enzymes and raised liver fat define the state the animal work modelled. Normal values leave less room for a measurable change.

• Pre-existing conditions: Effects were largest in animals with established steatohepatitis or diet-induced atherosclerosis, but glucose and lipid changes also appeared in healthy chow-fed mice, so disease is not a prerequisite.

Discontinuation & Cycling

• Intended duration: Framed as chronic, disease-modifying use. Animal dosing ran continuously for three to five months with no washout, and no study has tested whether liver or arterial gains persist after stopping.

• Withdrawal effects: None reported and none predicted mechanistically. Glycine and leucine are ordinary dietary amino acids, and no receptor downregulation, rebound or dependence signal has been described for the tripeptide.

• Tapering: No taper is indicated by any published finding. The only consideration on stopping is that a substrate-driven glutathione rise reverses within days once the substrate is withdrawn.

• Cycling: Untested. Nothing in the mechanism predicts tolerance, and the reported effects are substrate-driven rather than receptor-driven, so continuous rather than cyclical exposure is what the animal data model.

Sourcing and Quality

• Not commercially available in any legitimate form: DT-109 is an unapproved investigational compound. It is not a licensed medicine, not a lawful dietary supplement ingredient, and no pharmacy or supplement brand can supply it for human use.

• Research-chemical suppliers are the only real source: Catalogue houses sell milligram quantities labelled for laboratory use only, at a price per gram that makes gram-scale daily dosing absurd, and with no pharmaceutical manufacturing obligations.

• Grey-market peptide vendors carry the practical risk: Compounds that appear in prominent papers reliably surface in unregulated peptide shops, where identity, purity, endotoxin content and sterility are unverified and mislabelling is common.

• What a credible certificate of analysis (COA) shows: Batch-specific mass spectrometry identity, high-performance liquid chromatography purity above 98%, residual solvent and heavy metal limits, and endotoxin testing, issued by a laboratory independent of the seller.

• Isomer identity is the specific failure mode: The active compound is Gly-Gly-L-Leu. Its mirror image Gly-Gly-D-Leu, designated DT-110, performed no better than plain glycine in animal glucose testing, and routine purity assays do not distinguish the two.

• No compounding route exists: Compounding pharmacies may prepare approved drugs in altered forms; they cannot lawfully compound an unapproved new chemical entity, so no prescription or compounded pathway is available.

Practical Considerations

• Time to effect: Hepatic glutathione rose within one to two hours of a single animal dose. Glucose and lipid changes took weeks, and liver histology and arterial plaque changes required three to five months of continuous dosing.

• Common pitfalls: Treating DT-109 as interchangeable with plain glycine, scaling animal milligram-per-gram doses directly to human bodyweight, and assuming that catalogue purity on a research chemical implies human-grade safety.

• Regulatory status: Unapproved everywhere. It holds no marketing authorization from the U.S. Food and Drug Administration (FDA) or the European Medicines Agency (EMA), and no clinical trial registration appears in any public registry.

• Cost and accessibility: Research-grade material costs hundreds of dollars for quantities far below a single projected daily dose, which places a genuine multi-gram daily regimen out of reach regardless of willingness to pay.

• Payer and funder incentives diverge: Insurers and national health systems have no reason to fund a patented tripeptide when bulk glycine costs cents per gram, while research money flows to the patentable molecule — a structural bias in which comparison ever gets run.

Interaction with Foundational Habits

• Sleep: Direct and potentiating. Glycine lowers core body temperature and shortens time to sleep onset in human studies, and each DT-109 molecule carries two glycine units. Practically, this argues for dosing within two hours of bedtime and against dosing before tasks requiring alertness.

• Nutrition: Direct and overlapping. Ordinary dietary protein already supplies glycine — collagen and gelatin heavily so — which means habitual intake sets the baseline any dose is added to. Oxalate-rich foods such as spinach, rhubarb and almonds compound the one metabolic risk a glycine load creates.

• Exercise: Indirect and plausibly blunting. Raising glutathione dampens the reactive oxygen species burst that signals training adaptation, the mechanism by which high-dose antioxidants have blunted endurance gains. The leucine content pulls the other way by supporting muscle protein synthesis; the net effect is unmeasured.

• Stress management: Indirect. Glycine is an inhibitory neurotransmitter and glutathione buffers the oxidative component of chronic stress, so a calming effect is mechanistically plausible. No study has measured cortisol, heart rate variability or any other stress endpoint with DT-109.

Monitoring Protocol & Defining Success

Because no human protocol exists, monitoring here is framed around the specific ways this compound could plausibly help or harm, not around an approved label. A baseline panel taken before any exposure would establish the starting point for the outcomes the animal work claims — liver enzymes, liver fat by imaging, fasting glucose and insulin, a full lipid panel — and for the two organs most exposed to a gram-scale amino acid load: kidney filtration and urinary oxalate. Ongoing testing would reasonably run at four weeks, twelve weeks, then every three to six months, with kidney markers repeated sooner if any stone symptom appears. Because DT-109 has never been dosed in humans, no target response has been defined, and the honest reading of any single result is a change from the individual’s own baseline.

Biomarker	Optimal Functional Range	Why Measure It?	Context/Notes
ALT	10–26 U/L (men), 8–22 U/L (women)	Liver cell injury; the endpoint DT-109 lowered in animals	ALT = alanine aminotransferase. Conventional labs call up to 40–55 U/L normal; the functional range is tighter. Draw fasting.
AST	10–26 U/L	Second liver enzyme; confirms an ALT signal	AST = aspartate aminotransferase. Also rises with muscle damage, so pair with creatine kinase after heavy training.
GGT	<20 U/L (men), <15 U/L (women)	Oxidative stress and bile duct strain; tracks glutathione demand	GGT = gamma-glutamyl transferase. Conventional upper limits reach 50–60 U/L. Alcohol raises it independently of liver fat.
Liver fat fraction	<5% by MRI-PDFF; CAP <248 dB/m	Direct measure of the primary animal endpoint	MRI-PDFF = magnetic resonance imaging proton density fat fraction; CAP = controlled attenuation parameter on transient elastography. MRI is the more reproducible of the two.
HbA1c	4.8–5.3%	Three-month average blood sugar; the compound’s original target	HbA1c = glycated hemoglobin. Conventional prediabetes cut-off is 5.7%. Falsely low when red cell lifespan is shortened.
Fasting insulin and HOMA-IR	Insulin 2–5 µIU/mL; HOMA-IR <1.0	Detects insulin resistance earlier than glucose alone	HOMA-IR = homeostatic model assessment of insulin resistance, calculated from fasting glucose and insulin. Requires a 10–12 hour fast.
ApoB and triglycerides	ApoB <80 mg/dL; triglycerides <80 mg/dL	The lipid claim is the most contested one in the animal record	ApoB = apolipoprotein B, one particle per atherogenic lipoprotein. Conventional panels report only LDL-C (low-density lipoprotein cholesterol). Best paired with a fasting lipid panel.
hs-CRP	<0.5 mg/L	Cheapest proxy for the anti-inflammatory claim	hs-CRP = high-sensitivity C-reactive protein. Conventional “low risk” is <1.0 mg/L. Invalid within two weeks of infection, injury or hard training.
eGFR and serum creatinine	eGFR >90 mL/min/1.73 m²	Kidney filtration, the organ exposed to the oxalate load	eGFR = estimated glomerular filtration rate. Cystatin C-based eGFR is less confounded by muscle mass in trained individuals.
24-hour urinary oxalate	<25 mg/24 h	Tracks the specific mechanistic stone risk of a glycine load	Collect on a normal diet. High-dose vitamin C and high-oxalate foods within the collection window inflate the result.
Plasma ammonia	15–45 µmol/L	Tests the gut-liver mechanism and the nitrogen-load safety question	Sample on ice and process within 15 minutes; tourniquet time and fist clenching raise it falsely. Best drawn fasting.
Whole-blood glutathione	No established target exists; the change from the individual’s own baseline is what can be tracked	The proposed proximate mechanism of every claimed benefit	Assay values differ widely between laboratories, so serial measurements must use the same laboratory and method.

Qualitative markers worth tracking alongside the panel:

• Energy and post-meal alertness — the subjective correlate of the glucose claim
• Sleep onset and depth — the most likely noticeable effect of a glycine load
• Digestive tolerance — bloating, stool consistency and nausea, the expected dose-limiting complaints
• Exercise recovery and session quality — the place a blunted adaptation signal would first show
• Cognitive clarity — daytime drowsiness would be the counterpart of the sleep effect

Emerging Research

• No registered DT-109 trial: A ClinicalTrials.gov search on 9 September 2026 returned no interventional or observational study of DT-109 in any indication. Investigators have publicly stated an intent to move to human testing, but no registration or regulatory filing is public.

• Dietary glycine in fatty liver disease: NCT07285135, a 60-participant Singapore General Hospital study of glycine supplementation in metabolic dysfunction-associated steatotic liver disease, not yet recruiting, tests the parent amino acid against the endpoint DT-109 targets. A null result would weaken the glycine-deficiency premise.

• Glycine with N-acetylcysteine in older adults: NCT02348762, a 16-participant phase 1 Baylor College of Medicine study, and the randomized trial that followed it, test whether restoring both glutathione precursors lifts glutathione, mitochondrial function and physical function in humans.

• Glutathione restoration in Alzheimer’s disease: NCT04740580, a 52-participant early-phase Baylor study now recruiting, extends the same strategy to brain metabolism and inflammation. It bears on whether raising glutathione produces clinical benefit at all.

• Findings that could weaken the case: The monkey artery study of Jia et al., 2025 found no significant lipid difference across five months, and the gut-axis study of Qu et al., 2026 could not detect the compound in plasma at all.

• Findings that could strengthen the case: A published human pharmacokinetic and tolerability study would settle the systemic-versus-gut question, and work in female animals would close the sex gap left by the male-only design of every study to date.

Conclusion

DT-109 is a small chain of three amino acids — two of glycine, one of leucine — designed to deliver glycine more efficiently than glycine itself. The case for it is that low glycine tracks with fatty liver disease, high blood sugar and diseased arteries, and that supplying glycine restores the body’s main internal antioxidant. In mice and in monkeys fed rich diets, the compound reduced liver fat, liver injury, arterial plaque and arterial calcium.

Every one of those findings sits at the lowest tier of evidence used in this review, because none of them has been measured in a human being. Nothing has been published on how the compound behaves in the human body, what dose is tolerated, or what it does over years. The plausible harms are unmeasured too: a daily gram-scale amino acid load raises questions about the kidneys and about the growth signaling that leucine switches on. The animal record also contradicts itself: the blood fat effect seen in mice did not hold in monkeys, and reports of where the compound goes after oral administration disagree.

The evidence also comes from a single source. The scientists who discovered it hold the patent and founded the company that licensed it, and their university shares in that interest; no independent group has repeated the work. There is a further asymmetry: plain glycine costs almost nothing and cannot be patented, so the direct comparison between it and the three-part compound carries no commercial incentive and has not been run.

Top - Benefits - Risks - Protocol