Sarcosine for Health & Longevity

Evidence Review created on 08/14/2026 using AI4L / Grok 4.5

Also known as: N-Methylglycine, Methylglycine, N-Methylaminoacetic Acid

Motivation

Sarcosine is a small amino acid the body already makes when it handles glycine and methyl groups.
It is sold as a powder or chewable tablet and is used by people seeking cognitive performance, mood stability, or help with age-related muscle loss.
It works mainly by raising glycine at brain synapses, which increases activity of a major brain signaling receptor involved in learning and mood.

Interest began in psychiatry, where add-on trials tested it for leftover schizophrenia symptoms and for depression.
A separate laboratory paper linked the same molecule to prostate-cancer invasion, which still frames the safety debate.
An aging study later reported that blood levels fall with age-related muscle loss and that giving sarcosine to mice supported muscle regeneration.

This review examines the human evidence for oral sarcosine as a health and longevity intervention.
It covers proposed benefits for mood, leftover schizophrenia symptoms, and muscle aging, the conflicting prostate findings, dosing used in trials, and what remains unknown.

Benefits - Risks - Protocol - Conclusion

High-level sources that introduce sarcosine’s psychiatric, metabolic, and aging literature without substituting for the systematic-review section below.

No dedicated sarcosine content was found from Rhonda Patrick, Peter Attia, Andrew Huberman, Chris Kresser, Life Extension Magazine, or Lifespan.io.

Grokipedia

  • Sarcosine

    A compact encyclopedia entry covering chemistry, one-carbon metabolism (the pathway that builds and recycles methyl groups), glycine-transporter blockade, schizophrenia trials, and the prostate-cancer biomarker debate.

Examine

  • Sarcosine

    Examine’s primary sarcosine page summarizing glycine-transporter blockade, schizophrenia and depression uses, a 30 mg/kg dose rule, and the prostate-biomarker caveat.

ConsumerLab

No ConsumerLab article for sarcosine was found.

Systematic Reviews

Five systematic reviews and meta-analyses that quantify oral sarcosine or place it among other NMDA modulators, with effect sizes given as the standardized mean difference (a unit-free contrast of group averages).

No systematic review of oral-sarcosine cancer risk or long-term supplementation safety was found on PubMed as of 14 August 2026.

Mechanism of Action

Sarcosine (N-methylglycine) is an intermediate of one-carbon metabolism.
Glycine N-methyltransferase (GNMT, the enzyme that transfers a methyl group from S-adenosylmethionine to glycine) produces it; mitochondrial sarcosine dehydrogenase (SARDH) converts it back to glycine and a one-carbon unit.
Oral sarcosine also blocks glycine transporter 1 (GlyT1, the protein that clears glycine from synapses), raising synaptic glycine and D-serine.
Extra glycine occupies the glycine site on the NMDA receptor, which is required for synaptic plasticity.
This is the main proposed explanation for mood, psychotic-symptom, and cognitive effects.

A competing account treats sarcosine as a weak direct co-agonist at the same glycine site and as a modulator of NMDA-receptor surface trafficking.
A third line of work, from aging biology, is independent of GlyT1: circulating sarcosine falls with age and sarcopenia, and in mice it polarizes macrophages toward an anti-inflammatory state via general control nonderepressible 2 (GCN2, an amino-acid-sensing kinase), supporting muscle regeneration and adipose thermogenesis (heat production from fat).

Pharmacologically, oral sarcosine is water-soluble and well absorbed.
In a small add-on study, plasma kinetics were roughly linear, with time to peak concentration of about 1.5–2.5 hours and an elimination half-life of about 1 hour (Amiaz et al., 2015).
Tissue distribution is broad (muscle, liver, kidney, brain).
Metabolism is mainly SARDH-dependent oxidative demethylation rather than cytochrome P450 oxidation; no enzyme-selective drug–drug interaction profile has been mapped.
Selectivity is greater for GlyT1 than for glycine transporter 2, with additional weak NMDA-site activity.

Historical Context & Evolution

Justus von Liebig isolated sarcosine from meat hydrolysates in the nineteenth century; the name comes from the Greek sarkos (“flesh”).
For most of the twentieth century it was an industrial intermediate rather than a health compound.
The health use began with the NMDA-hypofunction account of schizophrenia.
In 2004, Tsai et al. reported that 2 g/day added to stable antipsychotics improved positive (hallucinations or delusions), negative, and cognitive ratings in a six-week double-blind trial.
A Taiwanese program then compared sarcosine with D-serine (Lane et al., 2005), tested monotherapy (Lane et al., 2008), and showed no added benefit on clozapine (Lane et al., 2006).
Huang et al., 2013 later ran the first depression trial against citalopram.

A parallel 2009 Nature metabolomic paper (Sreekumar et al.) reported rising tissue and urine sarcosine with prostate-cancer progression and showed that adding sarcosine made benign prostate cells more invasive in culture.
Later epidemiology did not confirm a simple “higher blood sarcosine, higher cancer” story; a large Norwegian nested study found a modestly lower risk at higher serum levels (de Vogel et al., 2014).
Pharmaceutical GlyT1 inhibitors followed the same target: bitopertin failed late-stage negative-symptom trials (Bugarski-Kirola et al., 2017), and iclepertin failed the 2025 CONNEX cognition program (Keefe et al., 2025).
Those null results sit beside smaller academic sarcosine trials and keep the cognition claim unsettled.
In 2025, Liu et al. moved the molecule into aging biology by linking lower plasma levels to human sarcopenia and showing muscle and fat effects in mice.

Expected Benefits

Medium 🟩 🟩

Residual Psychotic and Negative Symptoms ⚠️ Conflicted

Add-on oral sarcosine (2 g/day) has been tested mainly on background antipsychotics.
One meta-analysis of seven double-blind trials found a moderate overall-symptom gain (standardized mean difference 0.51).
A later six-trial analysis found no overall effect, with a smaller signal only in chronic, non-clozapine illness.
Adding sarcosine to clozapine showed no benefit.
The discrepancy likely reflects small samples and mixed illness phase.

Magnitude: Standardized mean difference 0.51 for overall symptoms versus comparators in a seven-trial meta-analysis (Chang et al., 2020); a six-trial add-on analysis found no overall effect (Marchi et al., 2021).

Depressive Symptom Reduction

Two randomized trials report mood improvement.
Huang et al. compared sarcosine with citalopram for six weeks in 40 people with major depression and found larger rating-scale gains, faster remission, and fewer dropouts.
Padhan et al. added sarcosine or placebo to a serotonin-reuptake inhibitor for eight weeks (n=60) and saw a greater Montgomery–Åsberg Depression Rating Scale drop.
Both samples were small and short.
Proposed mechanisms are NMDA-receptor facilitation and possible anti-inflammatory effects.

Magnitude: Add-on to a serotonin-reuptake inhibitor reduced Montgomery–Åsberg scores by 4.6 points more than placebo over eight weeks (−13.3 versus −8.7) (Padhan et al., 2024; Huang et al., 2013).

Low 🟩

Cognitive Performance

A four-trial meta-analysis in schizophrenia found a small, non-significant cognitive effect (standardized mean difference 0.27).
A 12-week trial improved cognition only when 2 g/day was paired with sodium benzoate.
No adequate trials exist in healthy or older adults.

Magnitude: Standardized mean difference 0.27 versus comparators (not statistically significant) across four schizophrenia trials (Chang et al., 2020; Lin et al., 2017).

Obsessive-Compulsive Symptoms

A 10-week open-label study in 26 outpatients used 0.5–2 g/day.
Mean Yale–Brown Obsessive Compulsive Scale scores fell about 20%, and 32% met a 35% response threshold.
Drug-naive participants appeared to do better.
There is no placebo-controlled trial.

Magnitude: Mean Yale–Brown score fell 19.8% (27.6 to 22.7) over 10 weeks in an uncontrolled sample (Wu et al., 2011).

Speculative 🟨

Muscle Mass Maintenance With Aging

In two cohorts (n=1,013), plasma sarcosine was lower with aging and sarcopenia (Liu et al., 2025).
Mice given sarcosine showed muscle regeneration via anti-inflammatory macrophages.
No human supplementation trial has tested muscle outcomes.

Stroke Recovery Association

A multicenter observational study linked higher plasma sarcosine after ischemic stroke with fewer adverse outcomes (Sun et al., 2025).
This describes endogenous levels, not oral supplementation.

Benefit-Modifying Factors

  • Background antipsychotic: Benefit in schizophrenia trials appears with older and newer antipsychotics other than clozapine; add-on to clozapine was indistinguishable from placebo (Lane et al., 2006).

  • Illness phase and treatment resistance: Meta-analytic signals concentrate in chronic, clinically stable, non-refractory illness; acute-phase and treatment-resistant subgroups are weaker or null (Marchi et al., 2021; Fornaro et al., 2025).

  • Baseline plasma sarcosine: Lower circulating levels track with older age and sarcopenia; whether people in that low range gain more from oral dosing is untested (Liu et al., 2025).

  • Sex: One schizophrenia meta-analysis noted larger effects as the female share of the sample rose, but this was not a prespecified sex analysis (Chang et al., 2020).

  • Age: Psychiatric trials enrolled mostly young and middle-aged adults.
    The muscle-aging signal is observational in older humans and interventional only in mice.

  • Prior treatment history: In the open-label obsessive-compulsive series, drug-naive participants improved more than those already on serotonin-reuptake inhibitors (Wu et al., 2011).

  • One-carbon genetics: Variants in GNMT or SARDH could change how much oral sarcosine accumulates; this has not been tested as a dosing biomarker.

Potential Risks & Side Effects

Medium 🟥 🟥

Prostate Cancer Biology ⚠️ Conflicted

Sreekumar et al. found rising tissue and urine sarcosine with prostate-cancer progression and showed that exogenous sarcosine induced an invasive phenotype in benign prostate cells.
A later 3,000-case nested study in the JANUS cohort found a modestly lower incident-cancer risk at higher serum sarcosine.
Tissue studies after 2009 have not consistently confirmed sarcosine as a marker of aggressiveness.
Short psychiatric trials were not designed to detect cancer.

Magnitude: Highest versus lowest serum-sarcosine quintile had odds ratio (relative odds versus the lowest group) 0.86 (0.72–1.01) for incident prostate cancer in JANUS, while cell work showed invasion after exogenous sarcosine (Sreekumar et al., 2009; de Vogel et al., 2014).

Low 🟥

Mood Elevation and Hypomania (Milder Elevated Mood)

Two published cases describe hypomania (a milder elevated-mood state), with reduced sleep need and irritability, after 2 g/day was added to an antidepressant plus an antipsychotic.
One met formal criteria for drug-induced hypomania.
Controlled trials did not report this signal but largely excluded bipolar illness.

Magnitude: Not quantified in available studies. Only two published case reports describe hypomania or activation after 2 g/day was added to an antidepressant plus antipsychotic (Strzelecki et al., 2014; Strzelecki et al., 2015).

Headache

Sarcosine has been described as well tolerated in randomized schizophrenia trials, with no excess of serious adverse events versus placebo.
In the open-label obsessive-compulsive series, one of 26 participants withdrew because of transient headache.

Magnitude: One of 26 participants withdrew for transient headache in the open-label obsessive-compulsive study; schizophrenia trials reported no excess of significant adverse events versus placebo (Wu et al., 2011; Tsai et al., 2004).

Speculative 🟨

NMDA Overactivation

Sarcosine facilitates NMDA-receptor function, so a theoretical risk is excitotoxicity (nerve-cell injury from overstimulation), seizures, or worsening of NMDA-blockade conditions.
Trials have not shown that signal; the basis is mechanistic only.

Risk-Modifying Factors

  • Sex and prostate status: The invasion biology and biomarker debate apply to people with a prostate.
    Baseline prostate-specific antigen (PSA, a blood marker of prostate tissue activity) and personal or family cancer history change how that debate is weighted.

  • SARDH deficiency: Rare sarcosine dehydrogenase deficiency (sarcosinemia) already raises circulating sarcosine; extra oral intake would add to that load.

  • Bipolar spectrum and mixed mood: Case-level activation occurred on combined serotonergic and glutamatergic treatment; bipolar history raises the chance that activation is not benign.

  • Age: Older men carry both a higher prostate-cancer base rate and the observational muscle-aging signal, so the same dose sits on a different risk–benefit line.

  • Baseline one-carbon markers: High homocysteine or low folate/cobalamin marks a stressed methyl-group pool; how that changes oral-sarcosine risk is unknown.

  • Concurrent serotonergic drugs: Both published activation cases involved a serotonin-reuptake or serotonin-norepinephrine reuptake inhibitor plus an antipsychotic.

Key Interactions & Contraindications

  • Clozapine (Clozaril): No added symptom benefit in a six-week randomized add-on trial.
    Severity: no efficacy interaction rather than a toxicity warning; combination is not supported (Lane et al., 2006).

  • Serotonin-reuptake inhibitors (citalopram, escitalopram) and serotonin-norepinephrine reuptake inhibitors (venlafaxine): Used together in depression trials, but case reports describe hypomania or activation.
    Severity: caution; monitor sleep and mood, and consider a lower sarcosine dose.

  • Other NMDA facilitators (D-serine, glycine, sodium benzoate, D-cycloserine): Additive glycine-site tone is plausible.
    Severity: caution; one trial paired sarcosine 2 g with benzoate 1 g and reported cognitive gain without a new safety signal (Lin et al., 2017).

  • NMDA blockers (ketamine, esketamine, memantine, dextromethorphan): Opposing actions on the same receptor class.
    Severity: caution; timing separation and close observation if both are used.

  • Antipsychotics other than clozapine (risperidone, olanzapine, quetiapine): The usual trial background.
    Severity: monitor; generally tolerated at 2 g/day.

  • One-carbon supplements (folate, cobalamin, choline, betaine, glycine): Share the same methyl-group pathway.
    Severity: monitor; theoretical additive effects on methylation and glycine tone.

Populations who should avoid Sarcosine:

  • Active or recently treated prostate cancer, or a rising PSA without a completed work-up
  • Current mania or untreated bipolar I disorder
  • Known SARDH deficiency (sarcosinemia)
  • Pregnancy or lactation (no adequate human data)

Risk Mitigation Strategies

  • Lower starting dose with an antidepressant: Published activation cases involved 2 g/day added to a serotonin-reuptake inhibitor; a 1 g/day start is the dose used after those reports to reduce hypomania risk.

  • Morning or early-day dosing: Case reports of reduced sleep need led later writers to place the dose in the morning or midday rather than at bedtime.

  • Prostate surveillance in men: Protocols that take the invasion-biology debate seriously use a baseline PSA, a repeat at 8–12 weeks, and then every 6–12 months.

  • No add-on to clozapine for efficacy: The randomized clozapine add-on trial found no symptom benefit, so that combination is not used for efficacy.

  • Dose cut for emerging hypomania: In the published cases, the response to elevated mood, reduced sleep need, or irritability was to stop sarcosine or step from 2 g to 1 g/day.

  • Identity versus detergent derivatives: The intended material is labeled sarcosine or N-methylglycine, not sodium lauroyl sarcosinate, a detergent rather than an amino acid.

Therapeutic Protocol

  • Taiwan fixed-dose regimen (Tsai/Lane): Most schizophrenia trials used 2 g by mouth once daily for six weeks on a stable antipsychotic (Tsai et al., 2004).

  • Flexible psychiatric titration: Depression and obsessive-compulsive studies used 0.5–2 g/day, often starting at 500 mg and rising as tolerated (Wu et al., 2011).

  • Examine weight-based rule: Examine states 30 mg/kg/day, about 2.0–2.7 g for a 70–90 kg adult, taken daily.

  • Activation-adjusted dose (Strzelecki): If drive or irritability appears at 2 g, the published case response was a cut to 1 g/day.

  • Time of day: Morning or midday dosing is the practical default because of the short half-life and the case-level reduction in sleep need.

  • Half-life and splitting: Elimination half-life is about 1 hour and peak levels occur at 1.5–2.5 hours (Amiaz et al., 2015).
    Trials still used once-daily dosing; split doses are untested.

  • Higher-dose pharmacokinetics: An open-label study included 4 g/day and reported linear kinetics without a new safety signal (Amiaz et al., 2015); efficacy trials stayed at 2 g.

  • Genetics: No protocol uses GNMT, SARDH, or NMDA-receptor genotypes.
    SARDH deficiency is a reason not to dose rather than a titration cue.

  • Sex: No separate male/female dose is established; one meta-analysis only noted a larger effect as female representation rose.

  • Age: Psychiatric protocols were developed in young and middle-aged adults.
    Older users have no separate schedule; prostate monitoring becomes more relevant.

  • Baseline biomarkers: Lower plasma sarcosine tracks with sarcopenia but is not yet a start/stop rule.
    Residual psychiatric symptoms, not a laboratory cut-point, selected trial participants.

  • Pre-existing conditions: Clozapine-treated psychosis, active bipolar mania, and known prostate cancer sit outside the protocols that showed benefit.

Discontinuation & Cycling

  • Duration of use: Psychiatric trials lasted 6–12 weeks; one metabolic-safety trial ran 6 months (Strzelecki et al., 2015).
    Lifelong use for longevity has not been studied.

  • Withdrawal: No discontinuation syndrome is described.
    Effects on mood or drive would be expected to fade over hours to days given the 1-hour half-life.

  • Tapering: A formal taper is not required by pharmacokinetics.
    If activation occurred, stopping or stepping from 2 g to 1 g was sufficient in the published cases.

  • Cycling: No evidence that time off restores efficacy.
    Cycling has not been tested for tolerance or for prostate-related risk.

  • After a hypomanic episode: The published sequence was dose reduction of either sarcosine or the companion antidepressant, not a planned off-cycle.

Sourcing and Quality

  • Identity of the compound: The intended material is sarcosine (N-methylglycine), a water-soluble crystalline amino acid, not N-acyl sarcosinate surfactants used in soaps and toothpaste.

  • Form: Trials used oral powder or capsules at gram doses.
    Consumer products include 500 mg chewable tablets and bulk powder; liquids are less common.

  • Third-party testing: No ConsumerLab review exists.
    Certificates of analysis for identity, heavy metals, and microbial limits are the available quality signal.

  • Purity grade: Look for a stated assay (typically ≥98%) and a single-ingredient label.
    “Sarcosinate,” “sarkosyl,” or lauroyl/cocoyl prefixes indicate a different chemical.

  • Brands: Specialty cognitive-supplement suppliers (for example Nootropics Depot tablets) are the usual retail source.
    There is no compounding-pharmacy standard and no United States Pharmacopeia monograph for supplemental use.

  • Food sources: Egg yolk, meats, and some legumes contain small amounts that do not approach the 1–2 g trial doses.

Practical Considerations

  • Time to effect: Psychiatric rating-scale changes in trials appeared over 2–6 weeks.
    Some open-label reports describe drive or mood shifts within 1–2 weeks.

  • Common pitfalls: Using a sarcosinate detergent by mistake; combining several NMDA facilitators at once; ignoring sleep or mood activation; treating a 6-week schizophrenia trial as evidence for healthy-adult cognition.

  • Regulatory status: Sold in the United States as a dietary supplement, not as a Food and Drug Administration–approved drug.
    Disease claims are not authorized; psychiatric use in trials was investigational.

  • Cost and access: Gram-dose powder is inexpensive versus billed NMDA drugs such as esketamine and is widely sold online. Insurers have little reason to fund or prefer the unpatentable powder, a possible bias in guidelines and research funding.

  • Taste and use: The powder is sweetish and highly water-soluble, so it mixes in water; chewable 500 mg tablets are used when powder is inconvenient.

Interaction with Foundational Habits

  • Sleep: Direct and potentially disruptive.
    Case reports describe reduced sleep need and nighttime activation, consistent with glutamatergic arousal.
    Morning dosing and a sleep log are the practical counters; bedtime dosing is a poor default.

  • Nutrition: Direct overlap with one-carbon metabolism.
    Folate, cobalamin, choline, betaine, and glycine feed the same pool.
    Sarcosine is not a substitute for those nutrients, and high-dose glycine is a different intervention (much larger grams-per-day amounts).

  • Exercise: Indirect and untested in humans.
    Mouse data suggest support for muscle regeneration and adipose thermogenesis after sarcosine exposure.
    No trial has timed a dose around training or measured hypertrophy.

  • Stress management: Indirect via mood and NMDA tone.
    Depression trials reported rating-scale gains; activation cases show the opposite pole.
    People using breath work or meditation for arousal control may notice either calmer mood or extra drive.

Monitoring Protocol & Defining Success

Before starting, blood tests and a prostate evaluation in men document one-carbon status and hidden prostate disease, because oral sarcosine sits on a path studied as a cancer biomarker.
Mood, sleep, and attention are logged for two weeks so later change is judged against personal baseline.
Repeat laboratories at four to eight weeks, then every six to twelve months if use continues, check that PSA has not risen and that folate, cobalamin, and homocysteine stay in a functional range.
Cognitive tasks and strength measures can be repeated on the same cadence.
There is no established target blood level for supplemental sarcosine; tracking uses change from the individual’s starting point plus absence of mood activation or a rising PSA.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
PSA (men) Prefer <1.0 ng/mL if previously low; investigate any clear rise from personal baseline Flags prostate tissue activity in the setting of the invasion-biology debate Conventional cut-point is often <4.0 ng/mL, which is too coarse here; fasting not required; pair with digital rectal exam if indicated
Homocysteine 6–8 µmol/L Reads the one-carbon / methylation pool that produces and clears sarcosine Conventional “normal” extends to ~15 µmol/L; fasting preferred; interpret with folate and cobalamin
Folate (serum or RBC) Serum >15 ng/mL or red-cell folate in the upper laboratory third Cofactor for the one-carbon unit released when SARDH demethylates sarcosine RBC = red blood cell; conventional lower bounds are lower; not fasting-dependent
Cobalamin (vitamin B12) >500 pg/mL Partner of folate in methyl-group recycling Conventional lower bound is often 200–300 pg/mL; pair with methylmalonic acid if low-normal
Plasma glycine No established supplement target; track change from personal baseline Sarcosine is both a glycine product and a glycine precursor Fasting amino-acid panel; not a start/stop rule
Plasma sarcosine (if available) No established target; Liu et al. reported lower levels with aging and sarcopenia Documents exposure and the aging-metabolite signal Research-grade assay; not a clinical decision cut-point

Qualitative markers:

  • Sleep duration and sleep need versus the two-week baseline
  • Mood stability, irritability, and any reduced need for sleep
  • Cognitive clarity, working memory, and processing speed on a repeated task
  • Drive and inner tension (early activation clues)
  • Training recovery and subjective muscle function in older users

Emerging Research

  • Aging muscle signal: Liu et al. (2025) linked lower plasma sarcosine to human sarcopenia and showed mouse muscle-regeneration effects via GCN2 and macrophages (Liu et al., 2025).
    A human supplementation trial would strengthen or drop this longevity use.

  • Pharmaceutical GlyT1 failure: Iclepertin missed cognition and function endpoints in three phase 3 trials (n=1,835) at 10 mg/day for 26 weeks (NCT04846868, NCT04846881, NCT04860830; Keefe et al., 2025).
    That weakens the class-level cognition claim.

  • Completed depression add-on: A 60-person phase 4 trial of add-on sarcosine in major depression is completed and published (NCT04975100; Padhan et al., 2024).
    Replication in larger, longer samples would move the mood signal.

  • Endogenous stroke association: Higher plasma sarcosine after ischemic stroke tracked with fewer adverse outcomes in a multicenter cohort (Sun et al., 2025).
    This is not an oral-dosing result.

  • No active oral-sarcosine longevity trial: A ClinicalTrials.gov search on 2026-08-14 found no recruiting supplementation study aimed at healthy aging, cognition in unimpaired adults, or muscle.

Conclusion

Sarcosine is a small, naturally occurring relative of glycine that can be taken by mouth.
It raises glycine at brain synapses and also sits in the chemistry that builds and recycles methyl groups.
The strongest human data come from short add-on trials in schizophrenia and from two small depression trials.
Those studies show a possible improvement in leftover schizophrenia symptoms, reduced motivation or flattened mood, and depressive scores, with a weaker and less consistent signal for thinking speed and memory.
Two research summaries of the schizophrenia trials do not fully agree, and the samples are small.
An aging study found lower blood levels in people with age-related muscle loss and muscle benefits in mice, but that is not a human treatment trial.

Safety in weeks-to-months psychiatric trials has been favorable, with no clear excess of serious adverse events versus placebo.
The main unresolved concern is prostate biology: laboratory work raised the possibility that extra sarcosine can make prostate cells more invasive, while a large blood-based cohort found a modestly lower cancer risk at higher natural levels.
Case reports describe elevated mood and reduced sleep when sarcosine is combined with antidepressants.
Most psychiatric evidence comes from academic groups, not companies that sell the powder.
Payers reimburse patented drugs for the same receptor, not this powder, which can steer large trials.
Drugs that block the same protein that clears glycine in the brain have not shown a thinking benefit in large late-stage trials, which keeps that claim tentative.

Top - Benefits - Risks - Protocol