NSI-189 (Amdiglurax): The Complete Guide

NSI-189 (INN: amdiglurax; later development code ALTO-100) is a small-molecule investigational drug, not a peptide. The molecule is a synthetic organic heterocycle — specifically a (4-benzylpiperazin-1-yl)-[2-(3-methylbutylamino)pyridin-3-yl]methanone — with no peptide bonds. The existing StackTrax encyclopedia describes NSI-189 as a "pyrazolo-pyrazine derivative"; that's wrong. The molecule is a benzylpiperazine–pyridine methanone with no pyrazolopyrazine core. Free base molecular formula C₂₂H₃₀N₄O, MW 366.5 g/mol, CAS 1270138-40-3.

It was identified by Neuralstem, Inc. through an in vitro phenotypic screen of ~10,269 compounds against a stable line of human fetal hippocampal neural stem cells, scoring on neurogenic proliferation. Lead chemist was Karl Johe (Neuralstem CSO). "NSI" = NeuralStem Inc.; "189" is the library entry.

The sponsor history matters. Neuralstem was renamed Seneca Biopharma (Oct 2019), reverse-merged into Palisade Bio (Apr 2021), and then Palisade sold the asset (2021) for up to $4.9M cash + milestones to Alto Neuroscience, which re-coded it as ALTO-100. The INN amdiglurax was assigned post-Alto.

Regulatory status

• FDA: never approved. No NDA filed. Not on Drugs@FDA. Not a controlled substance.
• EMA, MHRA, PMDA, Health Canada: no approvals.
• WADA 2026: not specifically named on the Prohibited List. As an unapproved investigational substance, S0 (non-approved substances) catch-all applies. Athletes should consult their governing body.

NSI-189 was identified in a phenotypic screen — scored on a cellular outcome (NSC proliferation in hippocampal culture), not by binding to a defined target. In rodent models, oral NSI-189 has been associated with increased dentate-gyrus neurogenesis, increased hippocampal volume on MRI, enhanced LTP in hippocampal slices, and behavioral improvements in stress, stroke, Angelman syndrome, and diabetic neuropathy models. Downstream signaling reported: TrkB and Akt phosphorylation increase (Liu 2019, PMID 30408487); BDNF and SCF protein levels rise in stroke-model brain tissue (Tajiri 2017, PMID 28181668).

What's not established: the receptor / direct molecular target hasn't been publicly identified. No published study reports an X-ray co-crystal structure, a confirmed primary binding partner, or a knockout that abolishes NSI-189 activity. Statements that NSI-189 "activates TrkB" should be read as "increases TrkB activity downstream," not as direct TrkB agonism.

What NSI-189 is not: not an SSRI/SNRI-type reuptake inhibitor (no SERT or NET activity); not a stimulant (no DAT inhibition or amphetamine-like activity); not a meaningful-affinity monoamine receptor agonist or antagonist; not an NMDA antagonist (mechanistically distinct from ketamine and esketamine); not an AngIV/HGF system modulator (mechanistically distinct from dihexa).

The audit-defensible mechanism statement: NSI-189 is a small-molecule investigational compound originally identified by phenotypic screening for hippocampal neural-stem-cell proliferation in vitro. In rodent models it has been associated with increased dentate-gyrus neurogenesis, hippocampal volume, and synaptic plasticity, with downstream activation of TrkB/Akt signaling. The direct molecular target remains uncharacterized in the public literature.

NSI-189 has been tested in two registration-grade MDD trials and missed the primary endpoint in both — plus an Alto Phase 2b in 2024 that also missed. There has never been a positive pivotal trial.

Phase 1b — Fava 2016 (signal-finding only). Fava M, Johe K, Ereshefsky L, et al. Mol Psychiatry 2016 (PMID 26643541, erratum PMID 27528461). Double-blind, randomized, placebo-controlled, multiple-dose-escalation in inpatient MDD; n = 24; 28 days dosing. Doses 40 mg QD, 40 mg BID, 40 mg TID (40 / 80 / 120 mg/day) vs placebo. Primary read: safety, tolerability, PK — no serious adverse events. PK: half-life ~17.4–20.5 h, AUC dose-proportional, steady state by 96–120 h, compatible with once-daily dosing. Exploratory reductions on MADRS, HAM-D, CGI-S, CGI-I, SDQ, CPFQ vs placebo on multiple scales; SDQ/CPFQ effects persisted post-dosing. Caveat: n=24 is signal-finding, not confirmatory; sponsor-funded; Karl Johe is a Neuralstem author.

Phase 2 — Papakostas 2020 (failed primary endpoint). Papakostas GI, Johe K, Hand H, et al. Mol Psychiatry 2020 (PMID 30626911). 12-week double-blind, randomized, placebo-controlled, sequential parallel comparison design (SPCD); n = 220 outpatients with MDD; doses 40 mg/day, 80 mg/day, or placebo (monotherapy). Primary endpoint MADRS change from baseline. Result: negative. Pooled mean MADRS difference vs placebo: 40 mg arm −1.8 (p = 0.22); 80 mg arm −1.4 (p = 0.34). HAM-D17 also didn't separate. Secondary self-rated SDQ and CPFQ showed nominal advantages for 40 mg/day — exploratory in the absence of primary-endpoint significance. Safety: well tolerated; no new safety signals.

Phase 2 post-hoc — Johe 2020. Johe K, Hand H, Heintz N, et al. Ann Clin Psychiatry 2020 (PMID 32722729). Re-analysis of the same 220-patient dataset with severity dichotomization. 80 mg/day showed effect in moderately depressed subgroup (MADRS <30) but not severely depressed. Methodological caveat: post-hoc dichotomized subgroup analysis with the sponsor's CSO as first author — hypothesis-generating only; can't rescue a negative primary endpoint per FDA review standards.

Alto Neuroscience era (ALTO-100). 2023 Phase 2 (precision-psychiatry-stratified): Alto press releases reported a biomarker-defined responder subgroup outperforming non-responders on MADRS at week 6, no peer-reviewed publication located. 2024 Phase 2b: missed primary endpoint (MADRS vs placebo) per Alto IR disclosure October 2024.

There's no published controlled trial of NSI-189 in cognitively normal adults for nootropic / cognitive-enhancement use. All efficacy data come from MDD populations or rodent models. Marketing claims of cognitive enhancement in healthy users are extrapolations.

From Fava 2016 Phase 1b (PMID 26643541): oral (capsule); half-life 17.4–20.5 h; AUC dose-proportional across 40/80/120 mg/day; steady state by 96–120 h of repeat dosing; BBB penetration inferred from CNS pharmacodynamic effects (no quantitative human CSF or PET data published); trial doses 40 or 80 mg/day in Phase 2. Food effect, drug interactions, metabolism, and excretion are not fully characterized in published literature.

StackTrax doesn't endorse NSI-189 use — two MDD trials missed primary endpoints. The dosing below is community / vendor convention.

Parameter	Range
Dose	20–80 mg/day oral (community); trial doses were 40 and 80 mg/day
Cycle length	4–12 weeks (community); longest published trial was 12 weeks
Route	Oral (capsule, sometimes sublingual per vendor instructions)

No long-term human safety data exist beyond 12 weeks. There's been no Phase 3 of any kind, no real-world registry.

From the trial record: across Phase 1, Phase 1b, and Phase 2 (combined ~280 patients with up to 12 weeks of dosing), generally well tolerated with no serious adverse events. The trials failed on efficacy, not safety.

Common community and trial reports: mild irritability or paradoxical mood effects (not consistently in trial data); headache; mild GI effects; sleep disturbance with later-day dosing.

Theoretical / unaddressed: long-term safety beyond 12 weeks is uncharacterized; off-target effects can't be ruled out for an undefined molecular target; a neurogenesis-enhancer mechanism that drives neural-stem-cell proliferation is theoretically aligned with proliferative pathways but no long-term carcinogenicity data exist; pregnancy / breastfeeding not studied.

• Pregnancy / breastfeeding — no safety data.
• Active or prior cancer — theoretical caution given the proliferation-enhancer phenotype with an undefined molecular target.
• Active major depressive disorder, especially severe — the Phase 2 post-hoc concentrated benefit in moderate depression and found nothing in severe depression. Severe MDD users shouldn't rely on NSI-189 instead of evidence-based treatment.
• Suicidal ideation — not a substitute for evidence-based MDD treatment.
• Concurrent SSRIs / SNRIs / mood stabilizers — uncharacterized interactions.
• Athletes subject to anti-doping testing — consult your governing body. WADA S0 catch-all may apply.
• Anyone seeking an "approved antidepressant" — NSI-189 has never been approved.

Compound	Class / Mechanism	Difference
SSRIs / SNRIs	Serotonin / NE reuptake inhibitors	NSI-189 has no reuptake activity
Ketamine / esketamine	NMDA antagonist; rapid-acting	Ketamine works in hours; NSI-189 over weeks via phenotypic-neurogenic mechanism, no NMDA activity
Bupropion	NDRI / nicotinic modulator	NSI-189 isn't a stimulant
Tianeptine	µ-opioid agonist	NSI-189 has no opioid activity; no abuse potential
Cerebrolysin	Porcine-brain-derived peptide mixture	Mixture given IM; NSI-189 is a single small molecule oral
Semax / Selank	Russian heptapeptide nootropics	True peptides; intranasal; distinct mechanism
Dihexa	HGF/c-Met modulator	Different chemistry and target. Dihexa's foundational papers retracted; NSI-189's literature intact but trials failed.
BDNF / 7,8-DHF	Direct TrkB agonists	NSI-189's TrkB activation is downstream, not direct
Spadin / PE-22-28	TREK-1 K-channel blocker peptides	Peptide; ion-channel target
Modafinil / armodafinil	Wake-promoting (DAT / orexin)	NSI-189 isn't stimulant-like
Racetams	Putative cholinergic / AMPA modulators	Different chemistry; racetam efficacy itself controversial

ALTO-100 / amdiglurax is the same molecule under different names. The Alto Phase 2b 2024 missed primary endpoint — not a different drug, just a re-coding by a new sponsor.