NSI-189 remains an experimental compound, not a validated treatment for depression or a proven way to improve memory in healthy adults. The original phase 2 trial and the subsequent phase 2b trial conducted under the name ALTO-100 did not meet their primary efficacy endpoint in major depression. Exploratory signals must be read alongside these results, not in their place.
This guide covers the available evidence and the status verified on October 19, 2026. It offers no purchasing advice, preparation instructions or dosing regimen. If depression or cognitive concerns are driving your search, a professional assessment remains the useful next step.
NSI-189, ALTO-100, amdiglurax: what are we talking about?
NSI-189 is not a peptide. The initial clinical publication describes a synthetic small molecule in the benzylpiperazine-aminopyridine family. A “peptides” category in a shop does not change its chemical nature.
The names overlap, but the results belong to specific trials:
| Name and documentary reference | What it refers to | What it does not mean |
|---|---|---|
| NSI-189, notably in phosphate form | Code used in the original trials published by Neuralstem and its collaborators | An already approved treatment |
| ALTO-100, program taken over by Alto | New development code for the compound | Fresh evidence of efficacy with every name change |
| Amdiglurax, a name listed in the FDA substance registry consulted in 2026 | Name of the substance also indexed as NSI-189 | FDA or European authorization |
The official Palisade SEC filing describes the transfer of assets in 2021 and states that NSI-189 is now known as ALTO-100. The FDA substance registry links amdiglurax to NSI-189. It explicitly notes that a UNII identifier does not imply any regulatory approval.
This continuity helps trace the publications. It does not mean that a commercially sold reagent shares the formulation, manufacturing or monitoring of a clinical trial product. ALTO-300 and ALTO-101 are separate programs, not synonyms to add to this list.
Animal neurogenesis and human cognition: two different kinds of evidence
Neurogenesis refers to the production of new neurons. Neuroplasticity is broader: it involves changes in how networks function or are organized. Neither concept automatically means “better memory.”
The development of NSI-189 drew on preclinical observations. Some foundational data from cells and mice are presented in the initial publication as internal data that had not been peer-reviewed. Animal studies were subsequently published, notably on models of cognitive deficits after irradiation.
These questions need to be kept separate:
- An observation in cells provides insight into a possible mechanism.
- A result in an animal with an experimentally induced deficit warrants further research.
- A cognitive test in a patient measures performance or a reported difficulty in that clinical context.
- A useful, lasting improvement in a healthy adult requires evidence of its own.
A change in volume on MRI would not directly count new neurons, either. In phase 1b, comparisons of hippocampal volume were not statistically significant. This trial therefore does not demonstrate an increase in human brain size, let alone a percentage gain in memory.
The biological terminology can be accurate while the sales pitch skips several steps. Our Dihexa guide · FR examines a different molecule and a different body of evidence: it is not a validated alternative.
Depression trials: populations, endpoints and results
The MADRS is a clinician-rated scale of depressive symptom severity. A questionnaire on patient-perceived cognition and a computerized test do not measure the same thing. The table preserves these distinctions.
The main clinical milestones are as follows:
| Program | Population and duration | Endpoint or objective | Key finding |
|---|---|---|---|
| NSI-189 phase 1b, published online in 2015 | 24 adults with major depression, treatment for 28 days, follow-up through day 84 | Primarily safety and pharmacokinetics, exploratory efficacy | Signals on some self-reported questionnaires. Too small to establish efficacy |
| NSI-189 phase 2, published online in 2019 | 220 adults with depression, 12 weeks, two-stage design | Primary endpoint: change in MADRS | No statistically significant superiority on the primary endpoint |
| ALTO-100 phase 2a, results announced in 2023 | Cohort of 133 patients with major depression within an eight-week open-label study | MADRS at six weeks and predictive cognitive profile | Signal reported in a biomarker-defined group, without a randomized placebo comparison |
| ALTO-100 phase 2b, results announced in 2024 | 301 adults with major depression, six-week double-blind period | MADRS in the subgroup predefined by a memory profile | Primary endpoint and key prespecified secondary analyses not met |
| ALTO-100 phase 2b in bipolar depression | Separate trial, as an add-on to mood-stabilizing treatment, still recruiting in the registry consulted | MADRS at six weeks in a predefined subgroup | No efficacy results available in the sources consulted as of that date |
Phase 1b: an early signal, with six people on placebo
The publication by Fava and colleagues assigned the 24 participants to active treatment (18) or placebo (six), across several cohorts. The study primarily addressed questions of tolerability and drug exposure. Efficacy analyses pooled the cohorts and remained exploratory.
Some self-reported questionnaires showed favorable results. The clinical measures and imaging are not enough to confirm a treatment’s efficacy or human neurogenesis. An encouraging result in this format helps prepare a larger trial.
Phase 2: the primary result remains negative
The publication by Papakostas and colleagues found no superiority on the MADRS. Some results were favorable on self-reported questionnaires and exploratory CogScreen measures, but not on Cogstate measures.
There were numerous cognitive tests, and the authors call for replication. Cherry-picking positive results does not turn the trial as a whole into proof of improved memory. The participants had depression; they were not healthy adults looking to learn faster.
ALTO-100: favorable open-label study, then failed confirmation
The January 2023 phase 2a announcement mainly compares patients with or without a given cognitive profile while receiving treatment. Validation in a held-out dataset can help test a predictor’s reproducibility. It does not replace a randomized placebo group.
The October 22, 2024 press release subsequently reports that phase 2b failed on its primary endpoint and its key secondary analyses. This is a sponsor source and must be identified as such.
Later subgroup analyses, discussed in company communications, supported continued development. The subgroup receiving the treatment as an add-on to an antidepressant was described as prespecified, but its result remained nonsignificant (p = 0.09). Subsequent selection based on the presence of the compound in the blood raises another interpretive issue: it is not equivalent to the result in the planned randomized population. These analyses do not turn the primary result into a positive one. A hypothesis can survive a negative trial, but it must then pass a new test.
Bipolar depression is a separate question
The registry entry NCT06656416, updated on October 13, 2026, indicates ongoing recruitment and an estimated primary completion date in September 2027. Its endpoint concerns depressive symptoms in bipolar disorder, alongside background treatment, not memory in healthy adults.
The sponsor was still announcing results for mid-2027 in its August 2026 update. These dates are forecasts from different documents, not results. For the reader, the current conclusion is simple: this trial does not yet provide a clinical answer you can use.
Known tolerability and unknowns outside trials
The earlier phase 2 trial reported no serious adverse events among participants receiving NSI-189. This provides information from limited follow-up in selected patients. It does not establish safety over several years or with freely chosen combinations of other substances.
The supplementary adverse-event table for this phase 2 trial reports headaches, drowsiness, dizziness, nausea and abnormal dreams, among other events. Events also occur with placebo. Frequencies depend on the group and trial stage: they are not a personal prediction.
In the 2024 phase 2b trial, the sponsor lists headaches, nausea and abnormal dreams among common events, at frequencies similar to placebo. Relatively favorable tolerability and demonstrated efficacy remain two different conclusions.
The practical limitations are as follows:
| Situation | Why the results are not enough |
|---|---|
| Prolonged use | The trials described do not adequately characterize several years of unsupervised use |
| Pregnancy, breastfeeding, minors | The data do not support extrapolating safety |
| Bipolar disorder or symptoms of activation | The bipolar program provides a framework for background treatment and monitoring. It does not justify personal experimentation |
| Multiple psychotropic drugs or nootropics | Interactions and the complete product are not sufficiently characterized for a do-it-yourself cocktail |
| Kidney disease, liver disease or another significant illness | Trial selection criteria exclude many complex situations |
| Reagent bought online | Identity, formulation and quality controls have not been shown to be equivalent to those of the investigational drug |
There’s no need to invent spectacular toxicity to advise against self-medication. Clinical benefit has not been established for this use, while uncertainties about the product and monitoring add to the pharmacological risks.
Status in France and risks of reagents sold online
Document check dated 19 October 2026: no matches for NSI-189, NSI189, ALTO-100, ALTO100 or amdiglurax were found in the medicinal product and composition files of the French Public Medicines Database, operated notably by ANSM. No matches were identified in the EMA list of medicines downloaded on the same day.
These checks therefore did not identify any authorised medicinal product applicable in France under these names. The French database and the EMA list have defined scopes: the EMA specifies that its website does not list all national authorisations. This check does not claim to examine every exceptional access pathway or every possible synonym.
An international name, an FDA identifier and a trial number do not amount to marketing authorisation. Permission to conduct a study is not permission to sell a treatment to the public.
The DGCCRF highlights the risks of health products sold outside authorised channels. For a reagent advertised as “research use only”, a purity certificate is not enough to establish pharmaceutical quality, a suitable formulation or a favourable benefit–risk balance for ingestion.
The absence of a medicinal product from these lists does not, on its own, establish a general criminal prohibition on possession or importation. Specific legal questions should be checked with ANSM or a qualified professional. This article provides neither suppliers nor instructions for use.
Depression or cognitive complaints: no substitute for an assessment
Memory difficulties can accompany depression, a sleep disorder, a medication effect or other problems. Immediately attributing them to a lack of neurogenesis may delay a useful assessment. Don’t change a prescribed treatment to try an experimental compound.
The Noopept guide lets you compare the quality of another body of evidence, not choose a validated substitute.
If you’re having suicidal thoughts or are worried about someone close to you, 3114 is available free of charge in France, 24 hours a day, seven days a week. If there is immediate danger, call 15 or 112. Searching for a compound must not delay that call.
The Verdict
Frequently asked questions
Is NSI-189 a peptide?
No. It is a synthetic small molecule. Earlier trials often use the name NSI-189 phosphate. A seller listing it among peptides does not provide reliable chemical identification through that category alone.
What does the phase 2 trial show?
In 220 adults with major depression, the primary MADRS endpoint was not met. Some questionnaires and exploratory tests showed favorable results, which need replication. The subsequent phase 2b trial under the name ALTO-100 also missed its primary endpoint.
Do the results prove a memory benefit in healthy adults?
No. The populations, endpoints and limitations of the analyses do not support that conclusion. A cognitive score in a patient with depression and beneficial neurogenesis in the human brain are two different things to demonstrate.
Is there an authorization valid in France?
No corresponding medicinal product was identified in the French and European sources checked on October 19, 2026. This dated check does not constitute comprehensive legal advice. A product being available online does not make it an authorized medicine.
Sources and methodology
The main clinical papers and the adverse events appendix were consulted, then registries and regulatory data were checked on October 19, 2026. Alto’s press releases are identified as information from the sponsor. Trial completion dates remain projections. This selection is neither a systematic review nor a medical consultation.
