NMRA-335140

Navacaprant (NMRA-335140) is Neumora Therapeutics' once-daily oral selective kappa-opioid receptor (KOR) antagonist, previously positioned as a non-monoaminergic alternative to SSRIs for major depressive disorder (MDD). As of June 2026, the program is dead. All three Phase 3 KOASTAL studies missed their primary endpoint, the change from baseline to Week 6 in MADRS (Montgomery-Åsberg Depression Rating Scale, a 60-point clinician-rated depression scale where a clinically meaningful drug-versus-placebo improvement is typically ~3 points or more). KOASTAL-1 (NCT06029426) failed in January 2025 with an identical -12.5 MADRS change in both drug and placebo arms (p=0.993) [1][2]. KOASTAL-2 (NCT06058039) and KOASTAL-3 (NCT06058013) reported topline failures in June 2026 with least-squares mean drug-vs-placebo differences of -0.3 and +0.7 MADRS points respectively (drug numerically worse than placebo in KOASTAL-3) [3][4][11]. Neumora discontinued navacaprant on the KOASTAL-2/-3 readout, announced a ~35% workforce reduction, and pivoted to NMRA-511 (Alzheimer's agitation), NMRA-898 (schizophrenia), and NMRA-215 (cardiometabolic) [11]. The class itself appears broken: Johnson & Johnson's aticaprant, the most advanced competing KOR antagonist, also failed its Phase 3 VENTURA program in March 2025 [6]. There is no remaining registrational path for this asset.

DISCONTINUED. Novel small molecule, never approved anywhere, and no longer in active development. The Phase 3 KOASTAL program enrolled adults with moderate-to-severe MDD across three placebo-controlled studies plus a 52-week open-label extension (OLE, NCT06029439) [5]. KOASTAL-1 (n=383) read out in January 2025 and missed [1][2]. KOASTAL-2 (n=430) and KOASTAL-3 (n=422) read out in June 2026 and both missed primary and key secondary endpoints [11]. Neumora has stated that it is discontinuing navacaprant; no NDA (New Drug Application) submission will occur. The OLE generated long-term safety data showing no new safety signals, but is moot for registration purposes given the efficacy failures. Neumora retained no breakthrough therapy or fast track designation. The company will publish full datasets at a future medical meeting and is redirecting resources to its three remaining pipeline assets.

Kappa-opioid receptors (KOR, gene OPRK1, UniProt P41145) sit on neurons in brain circuits that handle stress, mood, and reward [8]. Their natural ligand, dynorphin, is essentially the brain's bad-mood opioid: stress drives dynorphin up, dynorphin activates KOR, and KOR activation cranks down dopamine release in the nucleus accumbens, the brain region that registers pleasure and motivation. Animals with chronic stress show high dynorphin tone, low reward, and behaviors that look like depression. Block KOR, lift the brake on dopamine, restore the capacity to feel pleasure. That was the entire pitch, and it was a real pitch: it bypasses the serotonin system entirely, so in theory it should help patients who don't respond to SSRIs and should specifically attack anhedonia, the loss-of-pleasure symptom SSRIs handle poorly. Important non-expert clarification: navacaprant is an antagonist (it blocks the receptor) which is the mechanistic opposite of how addictive opioids like morphine or fentanyl work. Those drugs are agonists at mu-opioid (not kappa) receptors, which is what drives euphoria and dependence. KOR antagonism does not produce euphoria, has no abuse liability signal across the published clinical record, and has no DEA scheduling. The preclinical case for KOR antagonism in depression was solid; the clinical case has now collapsed. Navacaprant's Phase 2 study had reported a MADRS signal in moderate-to-severe MDD that was largest in the anhedonic subgroup [7], but none of the three Phase 3 KOASTAL studies reproduced it. Aticaprant, J&J's KOR antagonist with similar pharmacology, also failed Phase 3 [6]. That is a class-level efficacy failure, not a navacaprant-specific issue.

Three twin Phase 3 studies powered for change from baseline in MADRS at Week 6 versus placebo, the standard FDA-acceptable endpoint for an antidepressant. KOASTAL-1 (NCT06029426, n=383): -12.5 MADRS change in both navacaprant 80mg and placebo arms, p=0.993; the key secondary endpoint on the Snaith-Hamilton Pleasure Scale (SHAPS, an anhedonia-specific measure) also missed [1][2]. A post hoc female-only subgroup showed -14.0 versus -11.4 (p=0.072), nominally positive but not statistically significant and not pre-specified. KOASTAL-2 (NCT06058039, n=430): drug-versus-placebo least-squares mean difference of -0.3 MADRS points at Week 6. KOASTAL-3 (NCT06058013, n=422): +0.7 MADRS points, with the drug arm numerically worse than placebo [11]. The 6-week primary endpoint window is short for any non-monoaminergic mechanism and is brutal in MDD trials because placebo response in modern depression studies often reaches 10-13 MADRS points, swallowing real drug effects of 2-3 points. The KOASTAL-1 placebo arm fell -12.5 points, exactly within that high-placebo range. But the KOASTAL-2/-3 numerical separations are so small (within 1 point) that this is not a placebo-response artifact: the drug has no clinically meaningful antidepressant effect in this design. Comparator was placebo, not active control. No pre-specified anhedonia-enriched subgroup analysis has been disclosed as registration-relevant.

The model gives this drug a 40% chance of eventually being approved. That figure starts from the historical approval rate for Phase 3 drugs in this area, roughly 51%, then adjusts based on ten specific facts about the trial and sponsor. The estimate rises because the trial uses a non-randomized design and open-label blinding, but falls because the sponsor has a thin approval record and earlier-phase results were weak. The remaining facts were close to average for this stage, so they left the number roughly where it started.

Efficacy risk is fully realized: navacaprant failed all three Phase 3 studies and the program is terminated [1][11]. The most advanced same-mechanism competitor (J&J aticaprant) also failed Phase 3 in March 2025 across the VENTURA program [6]. Safety risk turned out to be modest and is no longer relevant to navacaprant specifically: across the KOASTAL studies and the OLE, no new safety signals emerged, suicidal ideation was not elevated versus placebo, and the historical KOR-antagonist concerns about pruritus and sleep disruption did not materialize as program-limiting issues [11]. Execution and survival risk for Neumora the company is now the live question. Neumora was effectively a single-asset biotech (navacaprant was the lead and only Phase 3 program). The company reported $147.1 million in cash as of March 31, 2026, has announced a 35% workforce reduction with ~$10 million annualized cost savings, and projects cash runway into the third quarter of 2027 [11]. Market cap stood at approximately $252 million as of mid-June 2026, down from a post-IPO peak well above $2 billion and reflecting the cumulative impact of all three KOASTAL failures [12]. Commercial risk is moot for navacaprant. For investors evaluating Neumora's pipeline pivot, the relevant risks are pre-clinical-to-Phase-1 risk on NMRA-511, NMRA-898, and NMRA-215, none of which has Phase 3 data.

Dead asset. Not worth watching on its own merits. The investment story has fully shifted to Neumora's three pipeline candidates (NMRA-511 in Alzheimer's-related agitation, NMRA-898 in schizophrenia, NMRA-215 in cardiometabolic disease), none of which are yet in late-stage trials. Cash runway into Q3 2027 buys Neumora time to advance one or more of those to a meaningful readout, but at current ~$252M market cap, the equity is priced for either a transformative early-stage win or a slow decline [11][12]. For the broader MDD treatment landscape: navacaprant was targeting first-line moderate-to-severe MDD, not treatment-resistant depression (TRD). The TRD competitive landscape is where the recent novel-mechanism wins have happened, with esketamine (Spravato, J&J) and dextromethorphan-bupropion (Auvelity, Axsome) already approved for adults inadequately responding to standard antidepressants. The first-line MDD niche where KOASTAL enrolled is dominated by cheap generic SSRIs and SNRIs, which is why even a positive readout would have faced steep payer pushback. The honest read on the mechanism: KOR antagonism for depression has now failed across two well-resourced sponsors and at least eight Phase 3 studies. Future KOR-targeting programs in MDD need a fundamentally different approach (biomarker-enriched anhedonia selection, longer duration, or combination with monoaminergic agents) to be worth running. The bipolar II depression Phase 2 (NCT06429722, n=18, completed) is too small to be a meaningful proof-of-concept; it is a signal-seeking pilot, not a powered efficacy study, and Neumora has not announced any continuation in that indication [10].