Sodium Oligomannate for Health & Longevity

Evidence Review created on 09/25/2026 using AI4L + evidence-mcp / Grok 4

Also known as: GV-971, GV971, Jiuqi Yi, Oligomannate, Sodium Oligomannurarate, Nine Phase One

Motivation

Sodium oligomannate is an oral mixture of short sugar chains taken from brown seaweed. It was developed as a prescription drug for mild to moderate Alzheimer’s disease. A health and longevity interest comes from a proposed action on gut microbes and brain inflammation. It is not a supplement and not a general vitality drug.

China approved it for a limited time in late 2019 after one large trial reported a benefit on a cognitive test. The same trial did not show a matching gain in daily function, and the company that sold the drug also ran the research. A larger international trial later stopped early. In 2025 the Chinese approval was not renewed, production was reported halted, and another company said it would take the program forward.

This review examines what sodium oligomannate is, how it is thought to act, what human trials show for thinking scores and for harm, and what loss of marketing approval means for access. It keeps evidence in people with Alzheimer’s disease separate from use in people without dementia.

Benefits - Risks - Protocol - Conclusion

High-level overviews that discuss sodium oligomannate by name, outside systematic reviews and outside the dedicated reference sites below.

Alzforum’s 2019 report explains the conditional approval in China, the seaweed-sugar origin, and why specialists questioned a cognitive-score result that was not matched by daily function.

A short regulatory review of how this marine-algae sugar reached conditional approval in China for mild to moderate Alzheimer’s disease, including dose and development timeline.

An Alzheimer’s Drug Discovery Foundation summary covers the China cognitive-score trial, the lack of testing in people without dementia, and unwanted effects listed at approval.

Three overviews met the bar for a high-level discussion that is not a systematic review. No relevant article or episode was found from Rhonda Patrick, Peter Attia, Andrew Huberman, Chris Kresser, Life Extension Magazine, or Lifespan.io.

Grokipedia

Grokipedia’s article outlines this marine-oligosaccharide drug, the gut-microbiota account, China’s conditional approval, and the 2025 non-renewal.

Examine

Examine’s page identifies sodium oligomannate as a marine-algae oligosaccharide aimed at Alzheimer’s disease and notes that research is still ongoing. The entry itself is brief.

ConsumerLab

No ConsumerLab article on sodium oligomannate was found. ConsumerLab does not typically cover prescription medications, and sodium oligomannate is a prescription drug rather than a tested supplement.

Systematic Reviews

Network meta-analyses and one drug-class meta-analysis report sodium oligomannate alongside other Alzheimer’s drugs on thinking scores, daily function, and adverse events.

A network analysis found this drug better than placebo on a cognitive score, with no clear rise in adverse events.

Pools this drug with older Alzheimer’s drugs. Daily function and a clinician’s overall rating did not beat placebo, and withdrawal was not higher.

Ranks newer Alzheimer’s drugs. The cognitive-score estimate favored this drug, but the interval still crossed no difference.

Includes this drug among nine Alzheimer’s treatments. No drug showed statistically robust superiority over placebo on the primary cognitive outcomes.

Mechanism of Action

Sodium oligomannate is a mixture of short acidic sugar chains from the brown alga Ecklonia kurome. It is not aimed at one receptor. Most of an oral dose stays in the intestine. Bioavailability (the share that reaches the blood) was 0.6–1.6% in rats and 4.5–9.3% in dogs, passing between gut-lining cells (Lu et al., 2022; authors included Shanghai Green Valley Pharmaceutical). In healthy volunteers the half-life (time for the blood level to fall by half) was about 11 hours, with no age or sex difference (Wang et al., 2020). Absorbed drug stays mostly free in blood, with little tissue entry. In rats, brain-fluid exposure was about 5% of blood exposure, and the absorbed share left in urine. Much of the dose is never recovered and may be used by gut microbes. No cytochrome P450 route (liver enzymes that clear many drugs) is established.

Two accounts come from work tied to the maker, Shanghai Green Valley Pharmaceutical, which employed coauthors. One is that the drug remodels the gut microbiota (the intestinal microbial community), lowers the amino acids phenylalanine and isoleucine, and reduces entry of Th1 cells (an inflammatory immune-cell type) into the brain (Wang et al., 2019). The other is that the sugar binds amyloid-beta (a protein fragment that clumps in Alzheimer’s disease) and blocks fibrils (fiber-like clumps) from forming in lab tests (Zhou et al., 2023). A microbiome researcher argued that the mouse microbial shifts did not support an anti-inflammatory gut effect and might worsen imbalance (Shugart, 2019).

Historical Context & Evolution

Sodium oligomannate was developed in China by Shanghai Green Valley Pharmaceutical, with Meiyu Geng and the Shanghai Institute of Materia Medica, as an oral Alzheimer’s drug. Early lab work treated the sugar chains as binders of amyloid clumps. By 2019 a company-linked paper had shifted the account toward gut microbes and brain inflammation. China’s regulator granted conditional approval on 2 November 2019 for mild to moderate disease, to improve cognitive function. In 2022 insurance listing cut the price sharply. Antibody infusions cost far more, so payers have a reason to favor a cheap oral drug if they accept the evidence, and another reason to follow regulators who never approved it.

Approval rested on one China-only phase 3 trial (a large randomized trial meant to test benefit and harm) that met a cognitive-score endpoint and missed daily function, behavior, and a clinician’s overall rating (Xiao et al., 2021). Participants were not taking cholinesterase inhibitors (drugs such as donepezil that boost a memory-related chemical signal). Outside commentators argued that the score gap depended on an unusual late drop in the comparison group (Shugart, 2019). A global phase 3 trial opened in 2020 and remains suspended, with the sponsor citing the pandemic rather than a safety stop (GREEN MEMORY). The Chinese registration was not renewed in 2025, production was reported stopped, and Fosun Pharma, which would hold a commercial stake, announced a deal that December to pursue the drug. The cognitive-score result and the criticisms of it both remain.

Expected Benefits

High 🟩 🟩 🟩

No benefit reaches High: that class needs a human clinical endpoint shown in more than one trial, and the only trial with a significant cognitive-score primary result is the single phase 3 study.

Medium 🟩 🟩

Cognitive test scores in mild-to-moderate Alzheimer’s disease

A phase 3 trial in mild-to-moderate Alzheimer’s disease, sponsored by Shanghai Green Valley Pharmaceutical, found a better score on the Alzheimer’s Disease Assessment Scale-cognitive subscale (a memory-and-thinking test; higher is worse) after 36 weeks on 900 mg/day than on placebo (Xiao et al., 2021). Daily function, behavior, and a clinician’s overall rating did not differ, and people were not taking cholinesterase inhibitors. A smaller 24-week trial, about 80 people per arm, missed this result (Wang et al., 2020). A 2026 pooled analysis still favored the drug (Song et al., 2026).

Magnitude: 2.15-point greater improvement versus placebo at 36 weeks (95% confidence interval, the range compatible with the data, −3.07 to −1.23) in the phase 3 trial.

Low 🟩

Speculative 🟨

Cognitive aging without dementia

No trial has tested sodium oligomannate for thinking or memory in people without dementia. The idea rests only on mouse work on gut microbes and brain inflammation, not on a human outcome.

Benefit-Modifying Factors

  • APOE ε4 (apolipoprotein E epsilon 4, a gene variant that raises Alzheimer’s risk): The phase 3 gap was present in both carriers and non-carriers (Xiao et al., 2021).

  • Baseline cognition: The gap was larger at lower Mini-Mental State Examination scores (a 30-point bedside test): about 4.6 points in the most impaired band of the phase 3 trial versus about 1.7 points in the mildest band.

  • Baseline biomarkers: No blood, spinal-fluid, or imaging marker has been shown to change the cognitive-score gap. The phase 3 trial did not require an amyloid test at entry.

  • Sex: The phase 3 analysis reported a cognitive-score difference in both women and men. A mouse study found sex-specific shifts in gut microbes and amyloid (Bosch et al., 2024), with no matching human dose split.

  • Existing dementia: Trials enrolled mild-to-moderate Alzheimer’s disease, not normal cognition. Very low behavior scores at the start left little room for a mood change on that scale.

  • Age: The phase 3 gap was reported both under and over 65 years. Mean age was about 70. The global protocol enrolled ages 50–85, so adults older than that were outside the studied range.

Potential Risks & Side Effects

High 🟥 🟥 🟥

No risk reaches High: that class needs a human adverse event shown in more than one trial, and the lipid and nose-and-throat differences come from the single phase 3 safety table.

Medium 🟥 🟥

Higher blood lipids

In the phase 3 trial, hyperlipidemia (high blood lipids) was recorded more often on sodium oligomannate than on placebo. Overall adverse events, events judged related to the drug, and serious events were similar between groups. Two deaths in the drug group, from metastatic lung cancer and brain stem encephalitis (inflammation of the brain stem), were judged unrelated. A later meta-analysis did not find higher withdrawal than with placebo (Chen et al., 2024). The lipid count comes from one trial sponsored by Shanghai Green Valley Pharmaceutical (Xiao et al., 2021).

Magnitude: 7.1% of the drug group (29 of 407) versus 3.4% of the placebo group (14 of 410) in that 36-week trial.

Nasopharyngitis

Nasopharyngitis (a nose-and-throat cold) was also listed more often on drug than on placebo in the same safety table. Urinary infections and other upper-airway infections were not higher on drug. The paper did not print a separate statistical test for this one event. It is a mild, usually self-limited symptom rather than a reason the trial was stopped (Xiao et al., 2021).

Magnitude: 7.4% of the drug group (30 of 407) versus 5.6% of the placebo group (23 of 410).

Dry mouth

An approval safety summary (577 on drug, 495 on placebo) listed dry mouth more often on sodium oligomannate than on placebo (Syed, 2020). The phase 3 list of events at or above 5% did not include it.

Magnitude: 1.0% on drug versus 0.4% on placebo in that summary.

Blood in the urine

That approval summary also listed hematuria (blood in the urine) more often on drug than on placebo (Syed, 2020). The phase 3 list of events at or above 5% did not include it.

Magnitude: 1.0% on drug versus 0.2% on placebo in that summary.

Liver laboratory readings

That summary listed higher alanine aminotransferase (a liver enzyme), aspartate aminotransferase (another liver enzyme), and bilirubin (a bile pigment) on drug than on placebo (Syed, 2020). The phase 3 list of events at or above 5% did not include them.

Magnitude: Alanine aminotransferase 1.9% versus 0.4%; aspartate aminotransferase 1.7% versus 0.8%; bilirubin 1.2% versus 0.2%.

Low 🟥

Speculative 🟨

Risk-Modifying Factors

  • Drug-metabolizing genes: No cytochrome P450 variant (a liver-enzyme gene) has been linked to side effects. The drug is not described as cleared by those enzymes.

  • Baseline lipids: A high-lipid event matters more when fasting lipids are already high, because the phase 3 trial recorded more of those events on drug than on placebo.

  • Sex: Early testing found no sex difference in blood levels. Published trials do not report a consistent gap in adverse-event rates between women and men.

  • Liver disease and recent serious illness: The global protocol excluded inadequate liver function and major illness in the prior 6 months (GREEN MEMORY). Harm in those groups is not established.

  • Age: Enrollment centered on about 50–85 years. Adverse-event patterns in frail adults older than that range were not reported.

Key Interactions & Contraindications

  • Cholinesterase inhibitors (donepezil, rivastigmine, galantamine) and memantine: Severity: unknown. The China trial did not allow these drugs, so combined benefit or harm is not known. No dose change has been published.

  • Broad-spectrum antibiotics (ciprofloxacin, amoxicillin-clavulanate): Severity: caution. They can change gut microbes, the proposed target. Consequence: a possible loss of effect, not a documented toxic reaction. No dose separation is established.

  • Statins (atorvastatin, rosuvastatin; drugs that lower blood cholesterol): Severity: monitor. Consequence: the lipid signal may be masked or added to the statin effect. A repeat fasting lipid panel is the practical check.

  • Over-the-counter pain drugs (ibuprofen, naproxen): Severity: none identified. Trials have not reported a bleeding or kidney interaction. No extra step beyond ordinary use is established.

  • Gut or lipid supplements (probiotics, red yeast rice): Severity: caution. Consequence: a theoretical change in the gut target or in lipid readings. No interaction study was found, and dose separation has not been tested.

Populations who should avoid Sodium Oligomannate:

  • Known hypersensitivity to sodium oligomannate or the capsule contents. Severity: absolute contraindication. Consequence: allergic reaction. No desensitization step is described.

  • Inadequate liver function, excluded in the global protocol, with no published Child-Pugh class (a liver-failure grade) as a cutoff. Severity: caution. Consequence: uncharacterized liver risk. No reduced dose is defined.

  • Resting heart rate under 50 beats per minute, also excluded in that protocol. Severity: caution. Consequence: heart-rhythm safety was not studied. No mitigating dose is defined.

Risk Mitigation Strategies

  • Lipid panel: A fasting lipid panel at baseline and about every 3–6 months addresses the higher hyperlipidemia rate, 7.1% versus 3.4%, in the phase 3 trial.

  • Liver enzymes: A repeat alanine aminotransferase draw with the lipid panel, about every 3–6 months, addresses higher liver-enzyme readings in the approval safety summary.

  • Pause if lipids climb: A sustained rise in low-density lipoprotein cholesterol (the artery-related cholesterol fraction) is a reason to stop and recheck, which limits ongoing lipid exposure.

  • Match the trial’s other drugs: Staying off cholinesterase inhibitors and memantine matches the trial that produced the cognitive-score result and avoids an unmeasured combination.

  • Stay at 900 mg/day: The studied dose was 450 mg twice daily. A higher daily amount was not the efficacy regimen and adds exposure without a shown extra benefit.

Therapeutic Protocol

  • Studied dose: 450 mg orally twice daily (three 150 mg capsules each time; 900 mg/day), morning and evening. This was the phase 3 regimen and the former Chinese labeled dose.

  • No second schedule: No published longevity-clinic regimen uses another dose. The evidence is this schedule in mild-to-moderate Alzheimer’s disease, not in healthy adults.

  • Time of day: Morning and evening doses, about 12 hours apart, fit a half-life of about 11 hours. Trials did not require a fast.

  • Half-life: About 11 hours in healthy volunteers. A single daily dose was not the schedule used in the efficacy trial.

  • Split dosing: The regimen is split across the day. One 900 mg dose was not the phase 3 schedule.

  • Genes: No gene-based dose change is described. Apolipoprotein E epsilon 4 status did not remove the cognitive-score difference.

  • Sex: No sex-specific dose. Early drug-level testing found no difference between women and men.

  • Age: Studied mainly from about 50 to 85 years. No lower dose for older age was defined.

  • Baseline cognition: Larger score gaps appeared at lower starting cognitive scores. People with normal scores were not the trial population.

  • Baseline biomarkers: No blood or imaging marker is used to set the dose. The phase 3 trial did not require an amyloid test at entry.

  • Other illness: Inadequate liver function and recent unstable illness were exclusions. No adjusted dose exists for those conditions.

Discontinuation & Cycling

  • How long: The positive trial used 36 continuous weeks. Former marketing for Alzheimer’s disease was ongoing use, not a brief course. Published extension results are lacking.

  • Withdrawal: No withdrawal syndrome is described after stopping. Reports do not describe rebound confusion or a required taper.

  • Taper: No step-down schedule has been published. Stopping in the trials was not described as a gradual reduction.

  • Cycling: No on-off cycle has been tested as a way to keep an effect.

  • One drug: This is a single agent, so stopping does not require sequencing several drugs apart.

Sourcing and Quality

  • Product: A prescription capsule of 150 mg, formerly made by Shanghai Green Valley Pharmaceutical and sold as GV-971 (Jiuqi Yi). It is not a dietary supplement.

  • What to check: The studied article was that manufacturer’s 150 mg capsule. Supplement-style third-party seals do not apply to this prescription product.

  • Who made it: Green Valley was the sole producer of the trial drug. In December 2025 Fosun Pharma, a firm with a commercial stake, announced a deal to acquire a majority share.

  • Access now: Conditional approval in China was not renewed in 2025, and production was reported halted. There is no U.S. Food and Drug Administration approval. Powders sold online are not the studied capsule.

Practical Considerations

  • Time to effect: The phase 3 cognitive-score gap was present at week 4 and still present at week 36. Daily function did not separate over that span.

  • Pitfalls: Reading a test-score change as a gain in independence, using a dementia-trial dose in people without dementia, and assuming pharmacies still stock the capsule.

  • Regulatory status: Former conditional approval in China only. Re-registration was not granted in 2025. The global phase 3 registry record remains suspended.

  • Cost and access: While reimbursed in China, a box cost on the order of a few hundred yuan, and several boxes were used each month. Supply, not list price, is now the barrier.

Interaction with Foundational Habits

  • Sleep: Direction: none shown. Trials did not report worse or better sleep. The evening dose was part of the twice-daily schedule, with no published insomnia signal and no described sleep mechanism.

  • Nutrition: Direction: indirect. The leading account acts through gut microbes and their amino acids, so a large diet shift or an antibiotic course could in theory change that target. Trials did not require dosing away from meals.

  • Exercise: Direction: none shown. No study tested workout timing, strength, or endurance. The measured difference was a cognitive score, not a training result.

  • Stress management: Direction: none shown. Behavior scores did not favor the drug over placebo, so a calming effect is not supported. No cortisol study was found.

Monitoring Protocol & Defining Success

Before a first dose, a repeated cognitive test, a fasting lipid panel, a kidney-filtration estimate, and liver enzymes describe the start. The cognitive test is what the trials moved. Lipids matter because high blood lipids were more common on the drug. Kidney filtration matters because the absorbed fraction leaves through the kidney in animals, and no human dose change for low filtration is published. The cognitive test and lipids are repeated at about 4 weeks, when the score gap first appeared, then every 3–6 months. Kidney and liver tests can use that same cadence unless a result is abnormal. No drug-specific blood target exists, so the comparison is with the person’s own baseline and the ranges below.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
Fasting LDL-C and triglycerides LDL-C under 100 mg/dL (under 70 mg/dL if artery disease is present); triglycerides under 100 mg/dL High blood lipids were more common on drug than on placebo LDL-C is low-density lipoprotein cholesterol (the artery-related fraction). Conventional triglyceride limits often sit near 150 mg/dL. A fasting morning sample is used. If a lipid drug is already in use, change is tracked from that person’s baseline.
eGFR Over 90 mL/min/1.73 m², or stable versus the person’s own baseline if no oligomannate-specific target is set The absorbed fraction is cleared by the kidney in animal studies eGFR is estimated glomerular filtration rate (a blood estimate of kidney filtering). Conventional concern often starts under 60. No human dose table by eGFR was published.
ALT Under 25 U/L, or stable versus baseline where no drug-specific target exists The approval safety summary listed higher readings on drug than on placebo ALT is alanine aminotransferase (a liver enzyme). Conventional upper limits are often near 40 U/L. It is paired with the lipid draw. That summary recorded 1.9% versus 0.4%, below the phase 3 table’s 5% cutoff.
Cognitive score (MMSE or ADAS-Cog) No fixed blood-style optimum; change from the person’s baseline on the same test is followed This is the outcome that moved in the phase 3 trial MMSE is the Mini-Mental State Examination (30-point bedside test; higher is better). ADAS-Cog is the Alzheimer’s Disease Assessment Scale-cognitive subscale (higher is worse). Not a fasting test. The same form is used each time.

Qualitative markers, noted by the person and by someone who sees them often:

  • Recent-memory slips during ordinary tasks
  • Handling of medications, money, or appointments
  • Mood and irritability across the week
  • New nose-and-throat irritation or a sustained change in stamina
  • New blood in the urine

Emerging Research

  • Suspended global trial: GREEN MEMORY, NCT04520412 is a suspended phase 3 study, planned for about 2,046 people, with thinking and global-impression scores at 48–52 weeks. No results are posted. A restart would test whether the China cognitive-score finding holds elsewhere.

  • Long-term safety: NCT05058040 is a phase 4 safety study of about 2,500 people, listed as active and not recruiting, tracking adverse reactions through 48 and 96 weeks. A lipid or other signal here would change the harm picture.

  • Later efficacy study: NCT05908695 is listed as a recruiting phase 4 efficacy and safety study, about 1,312 people, with primary completion estimated in 2029. It is the registry record closest to a confirmatory repeat of the cognitive-score trial.

  • After stroke: NCT05545605 is recruiting about 116 people to test prevention of cognitive impairment after stroke. A clear result either way would widen or narrow use beyond Alzheimer’s disease.

  • Findings that could weaken the case: A new trial that misses the cognitive score, as the phase 2 study did (Wang et al., 2020), or a safety study that shows lipids rising with longer use, would weaken the single phase 3 reading (Xiao et al., 2021).

Conclusion

Sodium oligomannate is an oral sugar-chain drug from brown seaweed, made for people who already have mild to moderate Alzheimer’s disease. The main human signal is a better score on a standard memory-and-thinking test in one large trial run by the company that made the drug. That signal did not bring a clear gain in daily independence, behavior, or a clinician’s overall rating, and an earlier mid-size trial missed its main thinking result. People in the large trial were not on the older Alzheimer’s drugs used in many countries. Study authors included the seller’s employees.

Harm looks limited in the trials on record. Unwanted effects were about as common as with an inactive comparison. High blood lipids, and a nose-and-throat cold, were recorded more often on the drug. Serious events were similar, and deaths in the drug group were judged unrelated. None of those harm findings is a repeated excess across trials.

For longer healthy life, the gap is who was studied. There is no human test in people without dementia. Mouse work on gut microbes and brain inflammation is the whole basis for that idea. China’s approval was not renewed in 2025, production was reported stopped, and another company has said it wants to carry the program. Insurers and health systems have a financial reason to prefer a low-cost oral drug over costly antibody infusions, and a separate reason to follow regulators who have not approved this one.

Top - Benefits - Risks - Protocol