ACD856

ACD856

ACD856 is an experimental oral drug designed to boost the brain's response to its own growth factors, specifically BDNF (brain-derived neurotrophic factor) and NGF (nerve growth factor). These are the two molecules your brain uses to keep neurons alive, grow new connections, and stay plastic. They drop with age, in depression, and in Alzheimer's, and the loss of that signaling is one of the things that makes a brain feel slower, less adaptable, and more fragile over time. ACD856 is being developed as a way to amplify what BDNF and NGF you already have, rather than replacing them.
It's currently in clinical development by Swedish biotech AlzeCure for Alzheimer's disease, depression, and other cognitive disorders. As of 2026, it has finished Phase 1 trials in healthy volunteers (safe, well-tolerated, oral doses of 10-90 mg crossed the blood-brain barrier and produced measurable changes on EEG) and just completed a Phase 1b higher-dose study, with Phase 2a in Alzheimer's patients funded by an EU grant and being prepared. There is no human efficacy data yet, and the compound is not commercially available, not approved for any indication, and not realistically obtainable as a research chemical (the molecular structure hasn't even been publicly disclosed). So this page is mostly a forward-looking primer on what it is, why it matters, and what to track as the trials read out, rather than a usage guide.

Deep-dive

ACD856 is a positive allosteric modulator (PAM) of the Trk family of receptors: TrkA, TrkB, and TrkC. These are the receptors that NGF, BDNF, and NT-3 bind to. A PAM doesn't activate the receptor on its own (or only weakly); instead it sits on a different site than the natural ligand and makes the receptor more responsive when the natural ligand is around. Practically, this means ACD856 only really does something where BDNF or NGF are already present and signaling. It amplifies the existing tone rather than overriding it, which is the main argument for why it might be safer than full agonists of these pathways (full TrkA agonists, for example, have historically caused pain and hyperalgesia by hitting nociceptive neurons too hard).
Mechanically, the original characterization paper showed ACD856 binds to the intracellular domain of TrkA and increases the maximum velocity of the kinase, meaning each receptor cycles through more catalytic events per unit time. In cellular assays it potentiates TrkA signaling with an EC50 around 382 nM and TrkB with an EC50 around 295 nM. It's pan-Trk, so it hits all three subtypes, plus it has weaker activity at IGF1R and FGFR1. Downstream, this drives the ERK1/2 and PI3K-Akt pathways, which are the workhorses of synaptic plasticity, neurite outgrowth, and neuron survival.
What it does in animal and cell models. The Parrado-Fernandez 2023 paper is the most thorough preclinical writeup. ACD856 increases BDNF protein expression in cortical neurons by about 30%, and in 21-month-old mice (rodent equivalent of an elderly human), 4 weeks of daily dosing significantly raised brain BDNF levels, while the same compound had no effect on BDNF in young healthy mice. This is consistent with how the system works: when BDNF tone is already optimal, there's nothing to amplify. The compound also protected cortical neurons from amyloid-beta toxicity and from energy deprivation (glucose and pyruvate withdrawal), and increased SNAP25, a presynaptic protein that's lost early in Alzheimer's. In the cognitive behavioral models, single doses reversed scopolamine-induced memory impairment and improved age-related memory decline in aged mice. With repeated dosing, the minimum effective dose dropped from 0.3 mg/kg to 0.1 mg/kg, suggesting the compound creates lasting plasticity changes that compound over time.
Antidepressant effects. This is one of the more interesting parts of the profile. In the forced swim test, a single course of ACD856 (5 days) produced antidepressant-like effects that lasted up to 7 days after the last dose, similar to what you see with ketamine in the same model. This is mechanistically plausible because BDNF/TrkB is the convergent pathway both ketamine and conventional antidepressants are thought to ultimately work through. Whether this translates to humans is still unknown, but it's why AlzeCure is positioning ACD856 as a potential depression drug as well, and the Phase 1b trial was specifically designed to push doses up to a depression-relevant range.
Human data so far. Three Phase 1 trials and one Phase 1b have been completed. The microdose IV study (six healthy men, 100 µg IV) established a half-life of about 20 hours. The single ascending oral dose study tested 1-150 mg in 56 healthy volunteers with rapid absorption (Tmax 30-45 min), near-complete bioavailability, and no serious adverse events. The multiple ascending dose study gave 10, 30, or 90 mg daily for 7 days to 24 volunteers, reached steady state by day 6, and showed dose-dependent CSF penetration confirming the drug crosses the blood-brain barrier (CSF concentrations ran 37-120% of unbound plasma levels). qEEG showed dose-dependent changes (increased relative theta power, decreased fast alpha and beta power), which is interpreted as evidence of central target engagement, although what those EEG shifts mean for cognitive performance in humans is still an open question. None of the studies measured cognition or mood as primary endpoints. The Phase 2a study in actual Alzheimer's patients is the first time we'll see whether the mechanism translates.
Limitations of the evidence. Pretty much all the efficacy data is preclinical. Animal cognition models are notoriously poor predictors of human outcomes in this space. The Trk-PAM concept is genuinely novel, which means there's no precedent in approved drugs to lean on. The chemical structure of ACD856 has not been publicly disclosed (only its predecessor ponazuril, a veterinary antiparasitic), so independent characterization is not possible. The drug being well-tolerated in healthy volunteers for 7 days at 90 mg tells you very little about chronic safety in older patients with comorbidities and polypharmacy. And the mechanism (chronically pushing on BDNF/TrkB signaling) has theoretical concerns that haven't been ruled out, like whether long-term TrkB potentiation could promote unwanted growth in cells where Trk receptors are also involved in proliferation.
Women. Both the SAD and MAD studies included women, though restricted to those of non-childbearing potential, which means most of the early human PK data is in men or postmenopausal women. There's no sex-specific PK or efficacy analysis published yet. Mechanistically, BDNF/TrkB signaling differs meaningfully between sexes: oestrogen upregulates TrkB phosphorylation in the hippocampus, and TrkB-mediated neuroprotection in female hippocampal neurons appears to be estrogen-receptor-alpha dependent in ways it isn't in males. This means the substrate ACD856 is amplifying may be quantitatively different in women, particularly across cycle phases and between pre- and postmenopausal women. Whether that translates to a different effective dose, response magnitude, or side effect profile in women is genuinely unknown and will need to be looked at in Phase 2.

Dosage:

This is mostly forward-looking because ACD856 isn't available outside clinical trials. For context only:
  • Doses tested in Phase 1 single-dose: 1-150 mg orally, with no dose-limiting toxicity reached
  • Doses tested in Phase 1 multiple-dose: 10, 30, or 90 mg once daily for 7 days, well-tolerated. CSF concentrations rose dose-dependently, with 90 mg producing the strongest qEEG signal
  • Higher doses: explored in the recently completed Phase 1b study, results pending publication
  • Predicted human pharmacologically active dose: AlzeCure's preclinical-to-human translation suggested the active dose generates an unbound Cmax of roughly 50-70 ng/mL, which corresponds to roughly a 30 mg single oral dose
  • Pharmacokinetics: rapid absorption (Tmax 30-45 min), near-complete oral bioavailability, half-life around 20 hours (allows once-daily dosing), food delays absorption but doesn't reduce overall bioavailability much. Steady state reached by day 6
  • What's not known yet: the optimal dose for cognitive or antidepressant effect in humans, whether women need different dosing, whether older adults need lower doses (likely yes given pharmacokinetic patterns in this age group), how long the antidepressant effect lasts after stopping treatment
  • No legitimate research-chemical supply exists and the molecular structure hasn't been publicly disclosed, so anything sold as ACD856 outside of a clinical trial is almost certainly not the real compound. Don't buy it

Here's what you can expect:

Honestly, nothing yet. There is no human efficacy data. In healthy volunteers given 10-90 mg for a week, the only measurable effects were on EEG (a shift toward more theta power and less fast alpha/beta), which is suggestive of central activity but not something you'd subjectively feel. No serious adverse events, no clear mood or cognitive changes were reported, though those weren't the primary endpoints.
If the preclinical results translate, the expected profile in patients with cognitive dysfunction or depression would be slow-onset improvement in memory and mood over weeks of daily dosing, with the antidepressant effect potentially lasting longer than the dosing window (the way ketamine does). This is informed speculation, not evidence. Phase 2a results in Alzheimer's patients are the first real readout we'll get and probably won't read out until 2027 at the earliest.
For people watching this space, the things to track are: whether Phase 2a shows any cognitive signal in mild-to-moderate Alzheimer's, whether AlzeCure or a partner runs a depression trial, and whether the drug behaves the same in older adults and women as it does in young healthy male volunteers.

Side effects & risks:

  • Phase 1 safety profile is clean so far. Across roughly 100 healthy volunteers dosed up to 150 mg single dose or 90 mg daily for a week, no serious adverse events have been reported. Most adverse events were mild, with headache and nausea the most common in the SAD study
  • Two participants in the 7-day MAD study had transient elevations in lipase and amylase, which are pancreatic enzymes. These returned to normal without intervention but are worth flagging
  • One subject in the MAD study had a 22 ms QT interval increase at 30 mg on day 7. This is below the 30 ms threshold that triggers regulatory concern, and no broader cardiac signal was seen, but cardiac safety is a real consideration for any chronic CNS drug and bears watching
  • Class-related concerns from compounds with similar mechanisms (BDNF/NGF enhancement) include increased stool frequency and looser stools, muscle aches, reduced appetite, weight loss, and eosinophilia, per AlzeCure's own review of the literature. None of these were observed at clinical significance in the trials so far
  • Theoretical concern about pain or hyperalgesia from TrkA activation. Full NGF/TrkA agonism causes pain in animals and humans, but PAMs don't seem to do this because they only amplify endogenous signaling rather than driving it independently. Phase 1 trials specifically monitored for pain symptoms and found none, but it's still a class-level issue to be aware of
  • Theoretical oncogenic concern. Trk receptors are involved in cell proliferation and Trk pathway dysregulation appears in some cancers. Whether chronic Trk PAM activity carries any oncogenic risk over years of dosing is unknown and won't be answered by current trials
  • No drug-drug interaction data outside of basic PK studies. Older patients with Alzheimer's are typically on multiple medications and how ACD856 interacts with cholinesterase inhibitors, anti-amyloid antibodies, antidepressants, or anticoagulants is currently unknown
  • No data in pregnancy, breastfeeding, or in anyone under 18 or over 65. Phase 1 trials excluded these populations
  • No reliable supply. As above, anything sold outside a trial is almost certainly counterfeit. The actual molecular structure hasn't been publicly disclosed, which makes this a particularly bad compound to source from grey-market vendors
ACD856 is an investigational drug, not approved or marketed in any country, and not legitimately available outside of registered clinical trials.