Neumora Therapeutics

Phase A — Understand the business

Company Overview

Neumora Therapeutics (Watertown, MA; incorporated Nov 2019; Nasdaq IPO 15 Sep 2023) is a clinical-stage, pre-revenue "precision neuroscience" biopharma building CNS therapies targeting novel mechanisms for prevalent brain/centrally-mediated diseases. The pitch was that genetics + biomarkers could de-risk neuropsychiatry the way precision medicine de-risked oncology. No products are approved; the company "has not generated any revenue from the sale of products".

The business model was a single-asset bet dressed as a platform: navacaprant (NMRA-140), a once-daily oral selective kappa-opioid-receptor (KOR) antagonist for major depressive disorder (MDD), was ~90% of the equity story. That asset is now dead — see Lens 5. As of 2026-06-15 the company is a restructured holdco around three earlier-stage programs (Lens 5/11) with ~35% of staff cut and ~$10M annualized savings.

Suppliers/inputs: contract CDMOs for drug substance/product (no internal manufacturing). "Customers" do not exist — there is no commercial product (our figures empty ). Key counterparties are licensors, not buyers: The Scripps Research Institute (TSRI — KOR/V1aR/OTR programs) and Amgen (in-licensed the CK1δ and GCase programs in Sep 2021).

Supply Chain (→ CDMO / licensing chain

For a pre-revenue biotech the "supply chain" is the discovery-to-IP-to-CDMO chain, named:

• Upstream IP / discovery: TSRI (The Scripps Research Institute) — worldwide exclusive license to the KOR ("navacaprant"), V1aR (NMRA-511) and OTR (oxytocin-receptor PAM) programs, originally via the 2015 TSRI License Agreement that NMRA inherited when it acquired BlackThorn Therapeutics in Sep 2020. Amgen — exclusive, sublicensable licenses (Sep 2021) to the CK1δ and GCase programs; Amgen retains internal-research use and a time-limited right of first negotiation on certain commercial rights.
• Midstream (NMRA): all research, development, manufacturing and (future) commercialization responsibility sits with Neumora under each license.
• Manufacturing: outsourced to third-party CDMOs (named suppliers not disclosed in the ingested sections; flagged as a single-source/third-party dependency risk in Item 1A).
• Downstream: none — no approved product, no payer/distribution chain yet. NMRA-511 (closest to a pivotal) would face the LTC/Alzheimer's-agitation payer path if it ever reaches approval.

Chokepoint: the chain is IP-and-trial-data-dependent, not input-dependent. The single point of failure was always the clinical readout — which just broke (Lens 5). Milestone obligations to TSRI/BlackThorn legacy holders are mostly satisfied (a $2.3M cash + 50,903-share navacaprant Phase-3-dosing milestone was paid; "no other contingent consideration related to the [BlackThorn] merger" remains as of 12/31/25).

Competitive Advantages (→ platform / IP moat

The claimed moat was the "precision neuroscience" platform + a differentiated KOR asset. Post-15-June that moat thesis is substantially impaired:

• Mechanism moat — broken. Navacaprant's pitch was 300-fold KOR-vs-MOR selectivity (vs ~30-fold for J&J's aticaprant) and a clean once-daily oral profile. Selectivity did not translate to efficacy. The entire selective-KOR-antagonist class has now failed late-stage in depression — both navacaprant (Neumora, 0/3 Phase 3) and aticaprant (J&J, Phase 3 fail). The platform's flagship validation is a class-wide negative.
• IP estate (durable, but now lower-value): 174 patents/applications — 33 issued U.S., 141 issued foreign — across the molecule families; navacaprant alone has 5 issued U.S. patents. Real, but IP on a dead indication is worth little; the value migrates to NMRA-511/898 estate.
• Bargaining power: weak in every direction. No commercial leverage over payers (no product); dependent on TSRI/Amgen licensors upstream; and with the stock at ~$1.51 and the lead asset dead, negotiating leverage in any partnering/M&A talk is poor — a distressed seller.

Net: the "platform" framing was a single-asset bet. With the asset gone, the durable advantages are a modest CNS IP estate and a discovery group that is now 35% smaller. Not a moat that protects earnings — there are no earnings.

Segments

Not applicable in the conventional sense — one reportable activity (R&D), no product revenue, no geographic or segment revenue split (our figures empty ). The economically meaningful "segments" are pipeline programs and their spend. From the 10-KForm 10-KA company’s audited annual report to the US regulator. The most complete thing it publishes. MD&A expense walk:

• R&D fell ~$24.9M YoY, driven by a $12.2M reduction in navacaprant program expense post-KOASTAL-1 completion, partially offset by +$1.2M (M4 PAM / NMRA-898) and +$0.9M (NMRA-511) clinical-cost increases plus +$5.2M preclinical/manufacturing.
• Acquired IPR&D of $5.0M (FY2025) tied to a Phase 1 milestone under the Vanderbilt license.
• G&A fell ~$2.4M YoY.

The trend even pre-failure was a deceleration of navacaprant spend and a tilt toward NMRA-511/898 — which, in hindsight, foreshadowed the pivot now forced by KOASTAL-2/-3.

---

Phase B — Measure performance (: Lens 5 → Pipeline-by-phase; Lens 7 → Catalyst calendar + mechanism comps)

Pipeline by phase

The asset table is the company. Post-15-June state:

The print that matters — KOASTAL-2 and -3 (topline 2026-06-15): both Phase 3 studies failed the primary endpoint (Δ MADRS at Week 6 vs placebo):

• KOASTAL-2: navacaprant 80mg (n=217) −12.2 vs placebo (n=213) −12.0; LSMD −0.3, p=0.813.
• KOASTAL-3: navacaprant 80mg (n=212) −10.1 vs placebo (n=210) −10.8; LSMD +0.7 (favoring placebo), p=0.480.
• Pre-specified "post-optimization" pool (426 patients enrolled after the early-2025 KOASTAL-1 fixes — the analysis bulls had hoped would rescue the program): nava (n=216) −12.1 vs placebo (n=210) −12.1; LSMD 0.0, p=0.976.

This is the cleanest possible failure: no efficacy on primary, no efficacy on the rescue analysis, drug numerically behind placebo in one arm. Combined with the Jan-2025 KOASTAL-1 miss (also failed MADRS + SHAPS ), navacaprant is 0-for-3 in Phase 3 MDD. The company discontinued it the same day.

Balance-sheet flags (filing-grounded):

• FY2025 net loss $236.9M (FY2024 $243.8M); operating cash burnCash burnHow much more cash goes out than comes in, per period. The clock on a company with no profits. $206.4M in FY2025.
• Cash & equivalents $182.5M at 12/31/25 → $147.1M at 3/31/26 (~$35.4M Q1 burn).
• Q1 2026 net loss $53.5M (vs $68.0M Q1 2025) — burn improving.
• Accumulated deficit $1,184.1M.
• No going-concern qualifier; runwayRunwayHow long the cash lasts at the current rate of spending. It shortens the moment spending rises, which is why a figure taken from a quiet quarter flatters. stated >12 months. Post-restructuring the company guides cash runway into Q3 2027. Estimated current cash ~$125–135M.

Market reaction: shares −51% on 2026-06-15 (prior close $1.78 → ~$0.87 indicated premarket). The market had already priced heavy skepticism (stock was a low-single-digit "lottery ticket" into the readout), yet still halved — confirming a real residual option value was being assigned to KOASTAL-2/-3. That option is now zero.

Earnings Calls / management messaging (sentiment trend)

No transcripts on disk (transcripts=0 ). From public messaging:

• Pre-failure (early 2026): management framed 2026 as a multi-catalyst year and explicitly guided "Phase 3 KOASTAL-2 and -3 topline data in Q2 2026" as the marquee event, having "made adjustments to the KOASTAL-2 and -3 studies" after KOASTAL-1. Tone: confident the optimizations would work.
• Post-failure (2026-06-15): CEO Paul Berns pivoted the narrative to the rest of the pipeline — "We remain excited about the best-in-class potential of our pipeline" and pointed to "anticipated catalysts over the next 12 months across three programs". Classic post-mortem reframe — survival + redirection, not contrition.

Sentiment shift: from "precision neuroscience, lead Phase 3" confidence → "diversified early pipeline + cost discipline + runway" damage-control. The thing they stopped saying: navacaprant.

Catalyst calendar + mechanism comps

Catalyst calendar (post-15-June):

Mechanism comps (comparables are by target/mechanism, not P/E — +clinical rule):

• KOR-antagonist (navacaprant's class): aticaprant (J&J, ex-Cerecor/BTRX sibling chemistry) — also failed Phase 3 in depression. Icalcaprant (others) — earlier-stage. The class is now a graveyard for MDD efficacy — the single most important comp read-across, and it is negative.
• M4 PAM (NMRA-898's class): the validated commercial comp is KarXT / Cobenfy (Bristol Myers Squibb, ex-Karuna — M1/M4 agonist, approved for schizophrenia) — proves the muscarinic thesis can win, but NMRA-898 is a Phase 1 M4-selective PAM entering a field BMS already commercialized and AbbVie/Cerevel (emraclidine) contested. Differentiation unproven.
• V1aR / AD-agitation (NMRA-511's class): the approved comp is Rexulti (brexpiprazole, Otsuka/Lundbeck) for Alzheimer's-agitation; Avanir/Otsuka AVP-786 history is mixed. NMRA-511's V1aR mechanism is differentiated but unproven beyond Phase 1/early-2.
• Traditional multiples: n/a and not meaningful (no revenue, no EBITDA, no EPS). Do not fabricate an EV/Sales for a pre-revenue name.

Stock-Price Catalysts (what actually moves NMRA)

The tape says NMRA is a pure binary-event stock:

• IPO 15 Sep 2023 — priced as a hot neuroscience IPO (ARCH/Amgen-backed).
• Jan 2025 — KOASTAL-1 fails → stock collapsed; by 30 Jun 2025 the non-affiliate float market value was only ~$63.3M. CEO change followed (Lens 9).
• 2026-06-15 — KOASTAL-2/-3 fail → −51% in a day.
• Analyst target cuts mid-June 2026: Mizuho $6→$4; Needham $8→$5; H.C. Wainwright $18→$7 (all still above the ~$1.5 market price — sell-side lagging the tape, as usual on busted biotech).

Pattern: this stock reacts to one thing — clinical readouts on the lead asset. Macro, rates, sector flows are noise here. With the lead asset gone, the next mover is the NMRA-511 Phase 2 path (Q4 2026 / 2027) — or a corporate-action headline (M&A, reverse splitReverse stock splitMerging several shares into one to lift the share price. It changes nothing about what the company is worth, and is usually done to keep a listing or to prepare a larger raise., financing).

---

Phase C — Judge people & books

Management

• Paul L. Berns — co-founder, Chairman & CEO (CEO since Feb 2025, when he replaced Henry Gosebruch days after the KOASTAL-1 failure). Berns is a veteran biotech operator/board professional (ARCH-orbit; prior CEO/board roles across multiple drug companies). Archetype: professional manager / serial biotech chair, not a scientist-founder. Track record is one of dealmaking and capital-formation more than of shipping an approved drug at NMRA.
• Henry Gosebruch — former President/CEO (Jul 2023–Feb 2025), ex-AbbVie CSO/dealmaker; exited after KOASTAL-1; has since moved on (Galapagos spinout). The C-suite was overhauled after the Phase 3 flop.
• Other named execs (per web, 2025): Joshua Pinto (President), Bill Aurora (COO/Chief Development Officer), Michael Milligan (CFO) — verify against the next proxy.
• Skin in the game / ownership: ARCH Venture Partners and Amgen are anchor holders; founder/insider economics meaningful (no our figures on disk to quantify — n/a, not sourced).
• Capital allocation: raised a large IPO (Sep 2023), spent >$1.18B cumulative to a dead lead asset and an early pipeline. The one defensible recent move is decisive cost action — killing navacaprant the day it failed and cutting 35% of staff rather than chasing a sunk-cost subgroup story. That is the right call, executed fast. But the scoreboard is brutal: ~$1.18B deficit, lead asset zero, equity down ~95%+ from IPO.

Assessment: competent, well-connected operators who allocated heavily behind a mechanism that the whole class failed to validate. Post-failure conduct (speed, discipline, runway preservation) is a point in their favor for what comes next; the historical capital-allocation record is value-destructive by necessity of the science.

Forensic Red Flags + Regulatory

Accounting: clean-looking for a pre-revenue biotech. Auditor Ernst & Young LLP, since 2020 (PCAOB ID 42) — no auditor turnover, no restatement, no error-correction or clawback flags in the 10-K cover checkboxes. No revenue → no revenue-recognition risk. The genuine accounting watch-items for this kind of name: (1) going-concern (currently fine — runway into Q3 2027, no qualifier ); (2) SBC dilution flattering non-GAAP (NMRA reports GAAP net losses, so less relevant); (3) goodwill/IPR&D impairment — expect a write-down tied to the discontinued navacaprant program in the Q2/FY2026 filings (the asset's carrying value should be impaired) — watch the Aug 2026 10-Q. No cash-vs-earnings divergence concern (it's a cash-burn shell, not an accrual-manipulation risk).

Regulatory / legal findings:

• SEC EDGAR EFTS: 0 Litigation Releases, 0 AAERs naming Neumora since 2021.
• Securities class action — ACTIVE. A putative class action (S.D.N.Y., Case No. 1:25-cv-01072), filed 2026-02-06 against the company, certain executive officers and IPO underwriters, alleges Securities Act violations — that the IPO offering documents "contained false and misleading statements and omitted material facts about the prospects of navacaprant." NMRA "do[es] not believe these allegations have merit and ha[s] moved to dismiss." Lead plaintiff (Victor Otcher) appointed 2025-11-19. The 0-for-3 navacaprant outcome will be used to amplify this complaint — a real, if modest-dollar (small-cap), overhang and D&O-insurance cost driver.
• Non-SEC (FTC/DOJ/FDA enforcement): none found.
• Verdict: No accounting fraud signal. One IPO-disclosure securities class action (the standard post-collapse suit), pending a motion to dismiss. Expect an IPR&D/program impairment in the next filing.

---

Phase D — Project & stress-test (: Lens 11 → rNPV + runway-to-catalyst)

rNPV + runway-to-catalyst

No EPS projection — pre-revenue, and a three-year EPS path would be fabrication. The +clinical question is: does cash reach a value-inflection catalyst, and what is the risk-adjusted value of what's left?

Runway-to-catalyst (the question that matters):

• Cash ~$125–135M less ~one quarter's reduced burn]. Company guides runway into Q3 2027. Post-restructuring burn ≈ $120–140M/yr minus ~$10M restructuring savings minus navacaprant Phase-3 spend rolling off — directionally lower].
• Does runway reach a catalyst? Yes, but the catalysts are weak. NMRA-898 Phase 1 (H2 2026) and NMRA-511 MAD/Phase-2b (Q4 2026 → 2027) all fall inside the Q3-2027 runway. So the company will not be forced to raise before producing some data — but none of these is a pivotal de-risking event of navacaprant's magnitude. The risk is a dilutive raise into a ~$1.5 stock if any early read disappoints, or a slow bleed toward a reverse split.

rNPV / sum-of-parts (every input ``):

• Cash floor: ~$125–135M ≈ ~$0.68–0.73/share on ~185.3M shares ].
• Pipeline rNPV: n/a — not reliably sourced. NMRA-511 (Phase 2 AD-agitation) and NMRA-898 (Phase 1 schizophrenia) are too early and too mechanism-uncertain to peg defensible peak-sales × PoS without inventing numbers. Directionally, the market's ~$280M cap at ~$1.51 minus ~$130M cash implies the market is assigning ~$140–155M of net option value to the remaining pipeline + platform. That is a modest, plausibly-still-generous figure for two early CNS shots after a flagship class failure.
• Brier forecast (per skill): the next scoreable binary is NMRA-511 — "NMRA-511 advances into a Phase 2b dose-ranging study by 31 Dec 2026" (a go/no-go proxy). Not logging via our model (watchlist-loop rule — only log on genuine commitment; this is a watch item, not a position).

Bull vs Bear

Bull case. (1) Below-/near-cash optionality — at ~$280M cap on ~$130M cash, you're paying ~$150M for two early CNS programs + a CNS IP estate + ARCH/Amgen pedigree; if NMRA-511 (V1aR, AD-agitation — a real unmet need with a validated commercial precedent in Rexulti) shows a clean Phase 2 signal, the stock is a multi-bagger off a tiny base. (2) Fast, disciplined restructuring preserves runway into Q3 2027 — no forced raise before data. (3) CNS/late-stage assets are hot M&A targets — a busted-but-cash-rich CNS shell with NMRA-511 could be acquired or reverse-merged. (4) Sell-side targets ($4–7) sit well above the ~$1.5 price.

Bear case. (1) The lead asset is dead and the entire KOR class failed — this is not a delay, it's mechanism invalidation; the "precision neuroscience platform" thesis took a class-wide hit. (2) What's left is early and unremarkable — NMRA-898 enters a schizophrenia M4 field BMS already won (Cobenfy) and AbbVie contests; NMRA-511 is Phase 2 with a mixed-history mechanism (V1aR/AVP-786 lineage); NMRA-215 is preclinical and off-CNS-thesis. (3) Dilution / value-trap risk — ~$1.5 stock + multi-year runway + early pipeline = high odds of a dilutive raise or reverse split before any pivotal value-inflection. (4) Litigation overhang — the IPO securities class action gains ammunition from the 0/3 outcome.

Pre-mortem (18 months out, thesis broke): NMRA-898 Phase 1 was unremarkable, NMRA-511's higher-dose MAD/Phase-2b showed no clear signal, the cash crept toward the Q3-2027 wall, and management did a dilutive raise or reverse split at a sub-$1 price — the stock is a sub-$100M-cap zombie or got taken out for scraps. Most plausible path.

Are multiples too high? No multiple to assess. The relevant question is whether ~$150M of option value for two early CNS programs post-flagship-failure is too high — arguably yes, given mechanism uncertainty and dilution risk, but not egregiously so given the cash backstop.

Contrarian view (what the market may be missing): the market is treating NMRA as a left-for-dead lottery ticket — but the cash-vs-cap gap is narrow enough, and the CNS-M&A bid strong enough, that the downside may be more floored than the chart suggests (cash ~$0.70/share + takeout optionality). The asymmetry is "probably zero-to-the-pipeline, but a real cash/M&A floor." That is a trade, not an investment.

Devil's Advocate (short-seller)

Dismantling even the residual bull case:

• Structural break already happened. The way this company was supposed to make money — navacaprant in a huge MDD market — is gone, and the mechanism (selective KOR antagonism) failed across the class. You can't "fix" that with study optimizations; they already tried (KOASTAL-2/-3 were the optimized studies, p=0.976 in the rescue pool).
• Concentration → now diffusion into weakness. Revenue was always zero; the concentration risk was clinical, and it broke. What replaces it is three earlier, individually-low-PoS shots, none de-risked.
• Most dangerous competitor bulls underestimate: for NMRA-898, BMS (Cobenfy) and AbbVie/Cerevel have already taken the muscarinic-schizophrenia ground — a Phase 1 M4-selective PAM is years and hundreds of millions behind, with differentiation unproven.
• Worst capital allocation: ~$1.18B accumulated deficit for a zero-revenue, dead-lead-asset outcome. Even granting the science was unknowable, the scale of spend behind one mechanism was the bet that broke.
• What must hold for ~$1.5: that NMRA-511 produces a credible Phase 2 signal and the company avoids heavily dilutive financing and the M&A bid materializes. Stack three "ands" on early CNS assets and the base case is value erosion.
• If a remaining program disappoints by 20–30% on any read: with no revenue cushion, the equity re-rates straight toward cash value (~$0.70) or below (the market routinely values busted-pipeline shells below net cash on dilution fear).
• Single scenario that permanently impairs: NMRA-511 Phase 2b fails or stalls in 2027 → the company is a cash-burning shell with no clinical narrative → reverse split / wind-down / sub-cash takeout. Plausible — arguably the modal outcome.

Management Questions (ordered by information value)

1. After 0-for-3 in Phase 3 MDD and a class-wide KOR failure, what is the falsifiable go/no-go bar for NMRA-511 at the Q4-2026 MAD / Phase-2b decision — what data kills it?
2. Given runway only into Q3 2027, what is the financing plan — and will you commit to not doing a dilutive raise before a value-inflection read, or is an ATM already active?
3. Is the board running a strategic-alternatives / M&A process in parallel, and what would make you choose a sale or reverse merger over going-it-alone?
4. What impairment / write-down to navacaprant-related IPR&D and assets should investors expect in the Q2/FY2026 financials?
5. NMRA-898 enters a schizophrenia field BMS (Cobenfy) already commercialized — what is the specific differentiation of an M4-selective PAM that justifies the spend?
6. What did the KOASTAL-2/-3 failure teach you about the "precision neuroscience" platform — does the genetic/biomarker thesis survive, and where is it now validated?
7. Why is NMRA-215 (cardiometabolic/obesity) — off your CNS thesis — still funded post-restructuring rather than cut or partnered?
8. What is the post-restructuring quarterly cash-burn target, and what headcount/program scope does the Q3-2027 runway assume?
9. What are the terms of the Amgen (CK1δ/GCase) and TSRI licenses now — any milestone/royalty obligations or reversion triggers that bite if programs stall?
10. How do you retain a 35%-reduced scientific team and the discovery engine through a multi-year, low-stock-price rebuild?
11. What is the status and worst-case exposure of the S.D.N.Y. securities class action (1:25-cv-01072), and what is the D&O cost trajectory?
12. Would you consider returning capital / a structured liquidation if no program clears Phase 2 by 2027 — i.e., what protects the cash value for shareholders?
13. Which single remaining program has the clearest regulatory/approval path, and what is its realistic timeline to a pivotal?
14. What insider buying have you and the board done since 15 June at these prices — and if none, why should the market assign option value the insiders won't pay for?
15. Beyond the named four programs, is the in-licensed CK1δ/GCase (Amgen) neurodegeneration work alive, or effectively shelved?

---