spec sheet8 rows
luvadaxistat (TAK-831 / NBI-1065844); a selective DAAO inhibitor that raises D-serine to enhance NMDA-receptor co-agonism for schizophrenia cognition, with strong preclinical plasticity data but underwhelming phase 2 results.
- selective DAAO inhibition raises D-serine to enhance NMDA-receptor co-agonism and plasticity
- improved cognition and social deficits in schizophrenia-relevant animal models, alone and with an antipsychotic
- enhanced long-term potentiation with chronic dosing (a direct plasticity readout)
- improved cerebellar-dependent learning, fitting emerging cerebellar theories of schizophrenia
- targets cognitive impairment in schizophrenia, an area poorly served by current antipsychotics
Overview
a selective DAAO inhibitor (luvadaxistat / NBI-1065844) that raises D-serine to enhance NMDA co-agonism and plasticity for schizophrenia cognition; strong preclinical data, including cerebellar learning, but underwhelming phase 2 cognition results.
- it works by stopping the breakdown of D-serine, the co-agonist the NMDA receptor needs to fire properly.
- the enzyme it blocks, DAAO, is unusually concentrated in the cerebellum, and the drug improved cerebellar learning tasks.
- like TAK-653 and TAK-041, it moved from Takeda to Neurocrine (as NBI-1065844).
- it helped social and cognitive deficits in animals but did not move negative symptoms in the clinic.
Mechanism
TAK-831 (luvadaxistat, NBI-1065844) is a potent, selective inhibitor of D-amino acid oxidase (DAAO/DAO), the enzyme that degrades D-serine. D-serine is an obligatory co- at the subtype of receptor: the receptor needs glutamate plus a co-agonist (glycine or D-serine) at the GluN1 subunit to open efficiently. by inhibiting DAAO, luvadaxistat raises D-serine in brain, plasma and cerebrospinal fluid, increasing NMDA co-agonist occupancy and enhancing NMDA-receptor function, a rational counter to the NMDA-receptor-hypofunction hypothesis of schizophrenia, in which daily D-serine dosing itself has shown positive signals. according to PubMed, luvadaxistat significantly increased D-serine levels in the rodent brain, was efficacious in cognition tests and in a translational model of cognitive impairment in schizophrenia (both alone and alongside a typical antipsychotic), and with chronic dosing shifted the dose-response leftward and modulated , consistent with enhanced NMDA-dependent plasticity; it also improved a cerebellar-dependent associative-learning task, notable because DAAO is highly expressed in the cerebellum, an area of growing interest in schizophrenia (Fradley et al., 2023, Neurochem Res, https://doi.org/10.1007/s11064-023-03956-2). the same report candidly notes the translational gap: while luvadaxistat ameliorated sociability deficits in two negative-symptom-type animal tests, it failed to show effects on negative-symptom endpoints in clinical trials. clinically it was developed (as TAK-831, then Neurocrine's NBI-1065844) for cognitive impairment associated with schizophrenia through the phase 2 INTERACT program, where the cognition results were ultimately underwhelming. the honest read: a mechanistically elegant, well-characterized DAAO inhibitor that reliably raises D-serine and enhances NMDA-dependent plasticity in animals, but whose clinical benefit for schizophrenia cognition and negative symptoms has not convincingly materialized.
receptor fingerprint
D-amino acid oxidase (DAAO / DAO)potent, selective inhibitor; blocks the enzyme that degrades D-serine, raising brain, plasma and CSF D-serine
glycine/D-serine co- site (downstream)higher D-serine increases co-agonist occupancy, enhancing NMDA receptor function and long-term potentiation
Safetyrisks and cautions, not medical advice
clinical safety information is limited to the trial setting; the compound reached phase 2 and was generally advanced as an oral agent, but detailed tolerability is not fully summarized in the preclinical-focused source here. its main disappointment is efficacy, not a specific safety signal. it is an investigational drug that has not shown convincing clinical benefit, is not approved, and is not a supplement; grey-market sourcing is unverified and inadvisable.
History
TAK-831 originated in Takeda's neuroscience program as a DAAO inhibitor built to raise D-serine and correct NMDA hypofunction in schizophrenia. it acquired the generic name luvadaxistat and, along with TAK-653 and TAK-041, was licensed to Neurocrine Biosciences as NBI-1065844. it advanced through the phase 2 INTERACT program for cognitive impairment associated with schizophrenia; the cognition results underwhelmed and it failed to move negative-symptom endpoints, tempering the trajectory despite strong preclinical data.
Reputation
luvadaxistat is regarded as one of the most advanced and cleanly characterized DAAO inhibitors, and as a serious clinical test of the D-serine/NMDA-co-agonism approach to schizophrenia cognition. its reputation is a familiar one for this target: mechanistically compelling with good animal data, but clinically muted, which has left the DAAO-inhibitor strategy intriguing yet unproven.
Subjective profileweighing the evidence above
The mechanism is elegant and the preclinical plasticity data is strong, but Phase 2 in schizophrenia cognition did not deliver, which is the part that counts. There is no approved product, and grey-market material means an unverified compound with a disappointing efficacy record behind it.
Resources
This entry is here for reference.
Research
1 listed here; entry last updated August 2026
Reviews
My notesprivate to this device
FAQ
what is TAK-831 (luvadaxistat)?
it is a potent, selective inhibitor of D-amino acid oxidase (DAAO), the enzyme that breaks down D-serine. D-serine is a co-agonist the NMDA receptor needs to work properly, so blocking DAAO raises D-serine and boosts NMDA-receptor function, an approach aimed at the NMDA-hypofunction thought to underlie schizophrenia (Fradley et al., 2023, Neurochem Res, https://doi.org/10.1007/s11064-023-03956-2).
what is it developed for?
cognitive impairment associated with schizophrenia (CIAS) and negative symptoms, an area where existing antipsychotics do little. the logic is that restoring NMDA co-agonist tone could improve learning and memory circuits (Fradley et al., 2023, Neurochem Res, https://doi.org/10.1007/s11064-023-03956-2).
does it work?
the clinical picture is mixed. in the phase 2 INTERACT program its record was disappointing on the main cognition goal, but preclinically luvadaxistat robustly raised D-serine, improved cognition and social deficits in schizophrenia-relevant animal tests, and enhanced long-term potentiation (a plasticity readout) with chronic dosing (Fradley et al., 2023, Neurochem Res, https://doi.org/10.1007/s11064-023-03956-2). it notably failed to move negative-symptom endpoints in its own clinical tests in that report.
how does raising D-serine help the NMDA receptor?
the NMDA receptor will not open fully on glutamate alone; it also needs a co-agonist (glycine or D-serine) bound at a separate site. DAAO normally degrades D-serine, keeping levels in check. inhibiting DAAO lets D-serine accumulate, increasing co-agonist occupancy and making NMDA receptors more responsive, which supports synaptic plasticity and learning.
where does the cerebellum come in?
DAAO is highly expressed in the cerebellum, and luvadaxistat improved performance on a cerebellar-dependent associative-learning task in animals (Fradley et al., 2023, Neurochem Res, https://doi.org/10.1007/s11064-023-03956-2). that fits growing interest in the cerebellum's role in schizophrenia, beyond the usual cortical-striatal focus.
who develops it now?
it originated at Takeda as TAK-831 and, like TAK-653 and TAK-041, moved to Neurocrine Biosciences, where it carries the code NBI-1065844. it also has the generic name luvadaxistat.
can i buy it as a nootropic?
no. it is an investigational drug whose main cognition trials underwhelmed; it is not approved and not a supplement. it is documented here as the leading DAAO-inhibitor probe of the D-serine/NMDA cognitive hypothesis, which is a genuinely interesting mechanism even if the clinical results so far are muted.