NBI-1065845

Neurocrine Biosciences is running a five-trial Phase 3 program for osavampator (NBI-1065845, formerly TAK-653), an oral AMPA receptor positive allosteric modulator licensed from Takeda in 2020, as an adjunctive treatment for major depressive disorder patients still symptomatic on standard antidepressants [1][2][8]. The program includes three replicate acute efficacy trials (each ~200 patients), a maintenance trial (~550 patients), and a long-term safety study (~850 patients), targeting roughly 2,000 patients in total across the program [2][3][4][5][6]. The Phase 2 SAVITRI study (n=183) showed the 3 mg once-daily dose produced a placebo-adjusted MADRS improvement of 7.5 points at Day 56 (p=0.0016), one of the largest acute MADRS deltas reported in an adjunctive MDD trial [10]. If two of the three acute Phase 3 trials replicate that signal on the Day 56 MADRS endpoint, this would be the first AMPA-mechanism antidepressant ever approved.

Osavampator (NBI-1065845, formerly TAK-653) is a novel compound. It has never been approved anywhere for any indication. Neurocrine licensed it from Takeda in a 2020 deal worth up to ~$2 billion (including $120 million upfront, up to $495 million in development milestones, up to $1.4 billion in commercial milestones, and double-digit royalties on net sales) covering osavampator and several other early- to mid-stage psychiatry assets [8][11]. In January 2025 the parties amended the agreement: Neurocrine took exclusive rights ex-Japan, Takeda regained Japan, and the structure shifted from a 50:50 profit share option to a royalty-bearing license in each region [11]. Neurocrine advanced the compound to Phase 3 after the Phase 2 SAVITRI readout in MDD patients with inadequate response to first-line therapy [10]. No FDA breakthrough therapy or fast track designation has been publicly disclosed for the program, which is notable given that Spravato (esketamine) received breakthrough designation for treatment-resistant depression. The Phase 3 program includes three identically designed acute efficacy trials (NCT06786624, NCT06911112, NCT06963021), each enrolling roughly 200 patients with primary readout at Day 56 measured by change from baseline in the Montgomery-Asberg Depression Rating Scale (MADRS) [2][3][4]. A maintenance trial (NCT07196501, n=550) measures time to relapse, and a long-term safety study (NCT06966401, n=850) collects open-label adverse event data [5][6]. All five trials are currently recruiting per ClinicalTrials.gov. Neurocrine has not publicly committed to a regulatory submission date, but the program structure (multiple replicate acute trials plus maintenance and safety) is what a sponsor builds when planning a full NDA package once at least two acute trials hit. First acute readouts are most plausibly 2027 given current enrollment status.

Glutamate is the brain's main excitatory neurotransmitter, the chemical signal that tells most neurons in the cortex and hippocampus to fire. AMPA receptors are the proteins on neurons that catch glutamate and let positively charged ions flow into the cell, triggering it to fire. Think of AMPA receptors as the brain's fast 'go' signal, distinct from NMDA receptors which control longer-lasting changes in synaptic strength. A positive allosteric modulator (PAM) does not activate the receptor on its own. It binds to a different site on the receptor and makes it more responsive when glutamate arrives. So osavampator does not flood the brain with stimulation, it amplifies the brain's own glutamate signaling [1]. Phase 3 uses a 3 mg oral once-daily dose, the higher of the two doses that showed efficacy in Phase 2 [10]. Why might that help depression? The dominant hypothesis for the past decade is that depression involves weakened synaptic connections in mood-regulating brain circuits, and that restoring glutamate signaling, particularly through AMPA receptors, drives the rapid antidepressant effects observed with ketamine. Ketamine blocks NMDA receptors, but the downstream antidepressant effect runs through AMPA receptor activation in preclinical models. Osavampator attempts to hit the AMPA node directly without ketamine's dissociative side effects or abuse potential. The biology is reasonable. Validation outside the ketamine and esketamine programs is weak, because the AMPA-PAM mechanism has not produced an approved CNS drug despite multiple prior programs across psychiatric and neurodegenerative indications (see Risks for specific failed programs). The case rests on the Phase 2 SAVITRI result, which was strong enough on top-line MADRS numbers to commit several hundred million dollars of Phase 3 spend.

The three Phase 3 acute efficacy trials (NCT06786624, NCT06911112, NCT06963021) are placebo-controlled, randomized, and adjunctive [2][3][4]. Adjunctive means patients stay on their current antidepressant and add either osavampator or placebo on top, rather than replacing existing therapy. The target population is MDD patients with inadequate response to current treatment. Primary endpoint is change from baseline in total MADRS score at Day 56. MADRS is a 10-item clinician-administered scale that runs 0 to 60, with higher scores indicating more severe depression; a moderately depressed patient typically scores in the 20s to low 30s at baseline. With n=200 per trial and roughly 600 patients on active drug across the three replicates, Neurocrine has powered the program to absorb noise from any single trial. Day 56 is a reasonable timepoint, antidepressant trials typically read out at 6 to 8 weeks because that captures the response window for serotonergic agents and gives glutamatergic effects time to manifest. The maintenance trial (NCT07196501, n=550) uses the FDA-preferred randomized withdrawal design: responders to open-label osavampator are randomized to continue drug or switch to placebo, and time to relapse is the primary endpoint [5]. This is the standard package for a chronic indication maintenance claim. The long-term safety study (NCT06966401, n=850) is an open-label extension generating chronic exposure safety data needed for label [6]. Concerns: n=200 per acute trial is on the lower end for MDD Phase 3, where placebo response rates routinely run 35 to 45 percent and can swallow modest effect sizes. Running three replicate trials suggests Neurocrine knows this and is buying redundancy: two positive trials make the regulatory case even if one washes out.

About 9% of drugs at this stage are estimated to eventually receive approval. That figure starts from the historical approval rate for similar Phase 3 drugs (around 51%), then adjusts based on ten specific facts about this trial and its sponsor. The estimate is pulled down mainly by heavier-than-usual blinding, a weak sponsor approval record, limited earlier-phase results, and a randomized design. The remaining factors fall close to average for this stage, so they don't shift the number much in either direction.

Efficacy: this is the dominant risk. MDD Phase 3 is a graveyard. AMPA receptor positive allosteric modulators specifically have not delivered an approved drug. Named prior failures include: mibampator (LY-451395, Eli Lilly), which reached Phase 2 in Alzheimer's agitation/aggression and was discontinued without advancing [12]; farampator (CX-691 / ORG-24448, Cortex/Organon/Schering-Plough), which had its Phase 2 short-term MDD trial suspended in 2008 pending cardiac safety follow-up and never resumed, and which in healthy elderly volunteers showed memory impairment at the dose studied [13]; and pesampator (BIIB-104), which reached Phase 2 for schizophrenia-related cognition before being deprioritized. The full Phase 2 SAVITRI data package is not yet in peer-reviewed publication, so dose-response, response curves, and any subgroup heterogeneity remain partly opaque to outside readers. Safety: AMPA receptors mediate excitatory neurotransmission. Over-amplifying them carries seizure risk, particularly at higher doses. Several earlier AMPA modulator programs were terminated for seizure signals or cardiac safety questions (the farampator suspension is the cleanest precedent) [13]. The Phase 1 pharmacokinetic study with midazolam and oral contraceptives reported no clinically meaningful drug-drug interactions, which is useful for a depression drug given how often these patients are on multiple comedications including hormonal contraception [1]. Long-term cognitive and behavioral effects of chronic AMPA potentiation are not well characterized, the open-label extension will be the first systematic look [6]. Execution: enrollment across five trials is operationally heavy. MDD trials commonly slip on recruitment because eligible patients are often already on multiple agents and washout requirements can disqualify them. Commercial: even if approved, osavampator enters a crowded MDD market. Standard SSRIs and SNRIs are generic and cheap. Spravato (esketamine) is the established novel-mechanism option with payer infrastructure built around REMS-required administration. Zurzuvae (zuranolone) is the recent novel-mechanism oral entrant. Payers will require a clear efficacy delta and a clean chronic safety story before paying premium prices for another novel-mechanism antidepressant.

Worth watching. Neurocrine is a profitable, focused CNS company with approximately $2.86 billion in 2025 revenue, primarily from Ingrezza (valbenazine) for tardive dyskinesia, plus a growing Crenessity (crinecerfont) franchise [7]. They have the cash and the development experience to run this program properly. Osavampator appears to be among the most advanced AMPA PAM programs in clinical development globally, which means a positive Phase 3 readout would matter for the broader thesis that AMPA is a tractable depression target and would put pressure on Spravato and Zurzuvae positioning. A negative readout would effectively close the chapter on AMPA-mechanism depression drugs for the foreseeable future. Market context: roughly 21 million US adults experience MDD in a given year, and published meta-analyses indicate 50 to 70 percent fail to remit on first-line antidepressant therapy, putting the adjunctive-treatment addressable population in the high single-digit millions [14]. For revenue comparables, Spravato generated approximately $780 million in the first nine months of 2024 (up ~62 percent year over year) and is on a trajectory toward roughly $1.7 billion at annual run rate, demonstrating real payer willingness to pay for a novel-mechanism MDD asset despite REMS-required administration [15]. Zurzuvae, by contrast, produced only about $72 million in net sales in 2024 (its first full year, recognized as $36.1 million in collaboration revenue to Sage at the 50 percent share level), a more cautious launch trajectory for an oral novel-mechanism antidepressant [16]. The oral, once-daily profile of osavampator looks more like Zurzuvae's distribution model than Spravato's, but if the Phase 3 effect size approaches Phase 2 it could justify premium pricing. Specific signal to watch: first acute Phase 3 readout, most plausibly in 2027, on change from baseline in MADRS at Day 56. A placebo-adjusted separation above 2.5 points (on the 0-to-60 MADRS scale, this is near the lower bound of what is conventionally considered clinically distinguishable from placebo) with a clean safety profile would clear the typical FDA bar for an adjunctive antidepressant. A delta approaching the Phase 2 number (7.5 points) would be commercially differentiating. Below 1.5 points the asset is in trouble even if two replicates hit statistical significance, because payers will not pay premium prices for marginal effect sizes. Check back when Neurocrine publishes the Phase 2 SAVITRI data in a peer-reviewed journal or pre-announces the first Phase 3 database lock. Until then, the 35.6 percent PoS is a fair anchor and likely conservative given the Phase 2 effect size, the trial design is sound, and the biology is plausible without being proven.