Genomics
A one-asset, binary bet on a genuinely best-in-class disease-modifying Dravet drug whose pivotal read-out (mid-2027) lands inside a fully-funded runway — own it for the read-out, but size it like the 50/50 single-trial event it is.
Research
The verdict
A one-asset, binary bet on a genuinely best-in-class disease-modifying Dravet drug whose pivotal read-out (mid-2027) lands inside a fully-funded runway — own it for the read-out, but size it like the 50/50 single-trial event it is.
Stoke is a Bedford, MA late-stage clinical biotech (founded June 2014, IPO June 2019) building antisense oligonucleotide (ASO) medicines on a proprietary platform called TANGO — Targeted Augmentation of Nuclear Gene Output . The platform's trick is the opposite of most RNA medicine: instead of *knocking down* a bad transcript, TANGO *upregulates* a healthy one. It uses an ASO to block a "non-productive" splicing event on the wild-type copy of a gene, increasing the amount of correctly-spliced mRNA and thus protein. That makes it uniquely suited to **haploinsufficiency** diseases — where one mutated gene copy leaves ~50% of normal protein and causes disease .
The business is effectively one product plus a platform option:
. This asset *is* the company — it carried $16.4M of the $39.7M Q1-2026 R&D .Contract / revenue structure. Stoke has no product sales. All reported "revenue" is collaboration accounting under ASC 606 — recognition of upfront payments and development-cost reimbursement from two partners:
This is the cleanest possible framing for the bet: the equity is a call option on one Phase 3 read-out, de-risked financially by a major-pharma ex-US partner who validated the asset with $165M cash, and with the US commercial upside retained in-house.
For a synthetic-oligo company the "supply chain" is chemistry and contract manufacturing, not foundries. Stoke states its oligonucleotide drug substance is formulated in saline/buffered saline for intrathecal (lumbar-puncture), intravitreal, subcutaneous or IV delivery, manufactured "using common processes and readily available materials" via a network of contract manufacturers that can also scale to commercial supply ``. Named stakeholders along the chain:
. **In the US/Canada/Mexico, Stoke must build its own commercial + distribution from scratch** — it is "recruiting experienced commercial leaders from across the rare disease space" . SG&A is already rising for "commercial readiness" (+$3.4M YoY in Q1) ``.Chokepoint summary: the genuine single points of failure are (1) the Southampton platform license and (2) a single pivotal trial. Oligo CDMO supply is not the binding constraint.
Mechanistic moat (the real one). TANGO is a differentiated, patent-protected approach to a problem most modalities can't touch. Stoke's own framing — backed by the science — is that gene therapy, gene editing and modRNA are mostly built for recessive or gain-of-function diseases, and "other next-generation ASOs have generally had limited success in upregulating gene expression of haploinsufficiencies" because they target indirect mechanisms (microRNAs, lncRNAs) ``. Stoke calls itself a pioneer in disease-modifying haploinsufficiency therapy. If zorevunersen works, it proves the platform, not just the drug — and a platform that turns "+50% of a protein" into a repeatable design problem across CNS/eye genes is a durable franchise, not a one-drug story.
IP estate. Multi-national issued + pending claims on TANGO mechanisms and on composition-of-matter for oligos targeting specific TANGO splice elements across many genes, reinforced by trade secrets ``. Standard biotech patent-cliff risk applies (20-yr life, possible expiry near commercialization).
Regulatory moat. Breakthrough Therapy Designation (Dec 2024) gives enhanced FDA engagement and supports a rolling NDA . Orphan exclusivity (7-yr US / 10-yr EU) is the typical add-on for a disease this rare (not yet granted/disclosed in the filings — confirmation pending).
Bargaining power. Mixed. Over partners: Stoke negotiated a strong Biogen deal — retained North American economics, only gave up ex-territory, kept global development control. That signals real asset leverage. Over payers/customers (future): unproven; a disease-modifying orphan drug in a market with no approved disease-modifying option should command premium pricing and strong payer reception, but Stoke has zero commercial track record. Over suppliers: limited — beholden to the Southampton license and to CDMOs.
Net: the moat is scientific + IP, and it is real if the data hold. But a single-asset clinical biotech has no moat at all until the Phase 3 reads out — the moat is contingent on the very event the stock is betting on.
Per the 10-Q, Stoke is one reportable segment with no product revenue since inception ``. There is no product/geography revenue mix to break out — the "segment" disclosure exists only to satisfy ASC 280. The only meaningful disaggregation is R&D spend by program, which is the true map of where the company is investing:
| Program | Q1 2026 R&D ($K) | Q1 2025 R&D ($K) | Read |
|---|---|---|---|
| Zorevunersen (Dravet) | 16,440 | 16,930 | Flat — Phase 3 fully underway, steady-state |
| ADOA (STK-002) | 1,706 | 1,056 | +62% — ramping as OSPREY Phase 1 dosing begins |
| SYNGAP1 | 738 | 342 | +116% — Acadia-partnered, small |
| MECP2 | 0 | 147 | Zeroed — Acadia terminated Sept 2025 |
| Personnel / unallocated | 20,789 | 14,201 | +46% — headcount + commercial-readiness build |
| Total R&D | 39,673 | 32,676 | +21% YoY |
``
The trend is unambiguous: ~95% of direct program spend is zorevunersen, with a deliberate, early ramp on ADOA as the second shot on goal. The big jump is in unallocated/personnel — Stoke is staffing up for a US launch before it has data, a calculated bet on speed-to-market.
The asset table is the company. Status as of the Q1-2026 10-Q (filed 2026-05-07):
| Program | Indication | Modality / target | Phase | Next catalyst | PoS (est.) |
|---|---|---|---|---|---|
| Zorevunersen (STK-001) | Dravet syndrome | ASO upregulating Nav1.1 via wild-type SCN1A | Phase 3 (EMPEROR) | Final pt randomized June 2026; topline mid-2027 | ~55-65% `` |
| STK-002 | ADOA (optic atrophy) | ASO (OPA1 upregulation) | Phase 1 (OSPREY) | Dose-escalation data 2026→early 2027 | ~20-30% `` |
| SYNGAP1 | SYNGAP1-related NDD | ASO (Acadia 50/50) | Preclinical | IND-enabling | n/a |
| Undisclosed CNS/eye | various | TANGO ASO | Preclinical/discovery | — | n/a |
Zorevunersen — the evidence base (this is what underwrites the PoS). In May 2026 Stoke reported four-year longitudinal OLE data: 93% (75/81) of eligible patients continued into the open-label extension; durable reductions in major-motor seizure frequency on top of standard-of-care ASMs at year four, plus continuing improvements in cognition and behavior (Vineland-3) . >850 doses administered; some patients treated >5 years; generally well tolerated. The cognition/behavior signal is the differentiator — every approved Dravet drug treats seizures only; **none addresses the developmental/cognitive burden**, and Dravet's cognitive impairment is known to be *independent of seizure count* . A drug that moves Vineland is a genuinely new category.
Safety watch-item: elevated CSF protein lab values in ~94% of patients (59% classified as a treatment-emergent adverse event), but no serious/severe clinical manifestations associated ``. This is the single most likely safety question at the FDA — a near-universal lab finding, currently asymptomatic. Bulls read it as benign; a cautious reviewer will scrutinize long-term consequences.
EMPEROR design (the binary). Global pivotal Phase 3. US/UK/Japan cohort screening is closed; final patient randomized June 2026. ~150 patients randomized to zorevunersen vs. sham control (sham lumbar puncture). Topline data mid-2027, intended to be the final data for a rolling US NDA; potential US launch as early as late 2027 ``. A European cohort (≥20 pts) enrolls through Q3 2026 but is not in the US NDA package.
Financials around the pipeline (the "earnings" that exist):
. FY2025 total revenue was **$184.4M** .— crucially this **crosses the mid-2027 read-out and reaches a potential late-2027 launch** without a forced raise. Stoke did, however, raise **~$168.5M via ATM** in the recent window ($87.8M + $80.7M net), i.e. it is topping up opportunistically into strength.Market reaction context: the stock is up ~244% YoY `` on the Biogen validation + accelerating EMPEROR timeline — the print itself (a "loss" vs. a prior-year gain) is irrelevant to a clinical-stage name; the tape trades the catalyst path, not the P&L.
No transcripts on disk (transcripts/ empty). From the filings' own program narrative and public releases, the management cadence has shifted decisively toward execution and speed over the last several quarters ``:
Net sentiment: increasingly confident, execution-focused, pre-commercial. The risk in that confidence is that it front-loads spend (and expectations) ahead of an unhedgeable single read-out.
Multiples are not meaningful for a pre-revenue, one-asset name — EV/Sales and P/E are n/a — not applicable (no product revenue, GAAP losses). The relevant table is the catalyst calendar:
| Date | Event | Why it matters | Direction risk |
|---|---|---|---|
| June 2026 | EMPEROR final patient randomized | Locks the mid-2027 read-out clock | Low (procedural) — but any slip is a negative tell |
| Q3 2026 | European cohort (≥20 pts) enrollment complete | Ex-US data, supports Biogen filing | Low |
| 2026 → early 2027 | OSPREY (STK-002/ADOA) dose-escalation data | First clinical proof of platform's second asset | Medium — early safety/PD signal |
| H1 2027 | Rolling US NDA submission begins | Regulatory milestone, pre-data optics | Medium |
| Mid-2027 | EMPEROR topline (the binary) | Makes or breaks the company | Very high |
| Late 2027 | Potential US commercial launch | Inflection to product revenue | High (contingent on data) |
Mechanism / event comparables (the right peer lens for ASO/genetic medicine — by modality, not multiple):
n/a.). Others developing Dravet ASMs: **Ionis, Lundbeck, Xenon, Harmony Biosciences** .. The broader ASO market is ~$4.5B in 2026, ~7-8% CAGR . Specific peer EV/Sales and P/E: n/a (would require pulling each name's live financials; not done here, do not fabricate).The pattern over the last ~18 months is catalyst-driven, not fundamentals-driven — exactly as expected for a clinical-stage name ``:
What the market actually reacts to: (1) partnership/validation events, (2) trial de-risking (timeline, design, regulatory alignment), (3) OLE/clinical data updates. It is not trading on earnings or revenue. The dominant forward catalyst — mid-2027 EMPEROR topline — is the one that will move it ±50%+ in a single session. Everything between now and then is positioning.
. Former CFO/COO of Vertex Pharmaceuticals — i.e. he helped build one of the most successful rare-disease commercial machines in biotech (Kalydeco/Trikafta). This is a high-quality, commercially-credentialed operator installed precisely to run the launch. A Vertex CFO taking the top job at a one-asset pre-revenue biotech is itself a signal of conviction in the asset.. Kaye was previously **CMO of Sarepta** (Exondys 51) — a deep ASO/rare-disease clinician. He sold **45,996 shares in Sept 2025** post-departure — routine, not a red flag . The CEO swap from clinician→commercial-operator at the pivot point is strategically coherent, but a CEO transition immediately before a pivotal read-out is a non-trivial execution risk and a discontinuity in accountability.insider-transactions.csv). Institutional ownership is specialist-biotech heavy — Baker Bros, RTW, Redmile, Lynx1, Skorpios Trust alongside BlackRock/Vanguard/Fidelity; 281 institutional holders, ~66.6M shares held ``. A syndicate of the most respected biotech specialists is a meaningful quality endorsement.Red flags: none material. The watch-items are the CEO transition timing and the heavy pre-data commercial spend ramp (betting opex ahead of an unhedged read-out).
For a pre-revenue biotech the forensic surface is small and clean. There is no revenue-recognition aggression to hunt in the usual sense — but the ASC 606 collaboration accounting is where any judgment risk lives, so it gets scrutiny:
. Auditor does not attest to ICFR (smaller-reporting-company exemption), and management concluded **disclosure controls + ICFR effective, no material weakness** .Regulatory findings (required sub-section).
regulatory/regulatory-findings.md (SEC EDGAR EFTS search, period 2021-06-20 → 2026-06-20) ``.. The Q1-2026 10-Q repeats: not subject to any material legal proceedings .EPS projection is meaningless here (deep losses for the foreseeable future). The two questions that matter are (a) what is the lead asset worth risk-adjusted, and (b) does cash reach the value-inflection catalyst?
Runway-to-catalyst (the decisive one): YES, with margin. $411.0M liquidity at 3/31/26, ~$200M/yr burn → runway "into 2028" per management ``. The mid-2027 EMPEROR read-out and H1-2027 NDA both sit comfortably inside the runway, and cash reaches a potential late-2027 launch. Stoke does not need to raise to reach its binary catalyst — a major de-risking of the financing overhang that usually plagues clinical-stage names. (It will likely raise opportunistically anyway, as the ATM activity shows, but from a position of choice not necessity.)
rNPV of zorevunersen (lead asset), illustrative — every input labeled ``:
. ROW (Biogen territory) adds **tiered double-digit-to-high-teen royalties** on a comparable-or-larger ex-US population → another ~$0.3-0.6B to Stoke at peak ``.. The current **EV ~$1.4-1.7B** (market cap ~$1.83-2.08B less ~$0.4B net cash) therefore implies the market is pricing zorevunersen at roughly fair-to-slightly-cheap on a risk-adjusted basis, ascribing little to the platform/STK-002 optionality.Forecast log: per the --watchlist rules, no forecast.ts create run in breadth mode. The binary worth tracking (for a future committed forecast) is: "STOK — EMPEROR Phase 3 primary endpoint met, p≈0.60, resolves 2027-09-30." Logged here as the scoreable claim, not committed to the tracker.
Bull case. Stoke owns a genuinely best-in-class, potentially first-and-only disease-modifying therapy for a severe pediatric epilepsy with no disease-modifying option today. The differentiation isn't incremental — the cognition/behavior improvement (independent of seizure control) opens a category no ASM touches, and four years of durable OLE data make the mechanism look real, not a flash. A major pharma (Biogen) validated it with $165M cash and took ex-US, while Stoke kept the high-margin US economics and is fully funded through the read-out and into launch — removing the financing overhang that kills most clinical-stage names. A Vertex-pedigree CEO is building the commercial org now. If EMPEROR hits mid-2027, this is a $2-3B+ peak-sales US franchise plus ex-US royalties plus a validated platform (STK-002/ADOA and SYNGAP1 as free options) — a multi-bagger from a ~$1.8B cap. The contrarian point the market underrates: the platform, not just the drug — a positive read-out re-rates every TANGO program at once.
Bear case (permanent-impairment risks).
Pre-mortem (it's late 2027 and the thesis broke). Most likely failure path: EMPEROR's primary seizure endpoint misses statistical significance (sham-control effect larger than powered for, or effect size attenuates in a larger/cleaner population than the small OLE), the cognition data is relegated to "supportive/exploratory," the FDA balks at the CSF-protein signal, and the stock halves. Secondary path: data is positive but messy, NDA slips, the pre-built commercial org becomes a cash drag, and a dilutive raise follows.
Are multiples too high? No — there are no earnings multiples, and the rNPV (Lens 11) says the EV is roughly fair-to-cheap on a risk-adjusted basis. The risk is not overvaluation; it is binary outcome variance. You are not overpaying; you are taking a coin-flip-ish bet at a fair price.
Contrarian view. The market treats STOK as "another single-asset epilepsy biotech." What it underprices is that a positive read-out validates a repeatable protein-upregulation platform across a whole class of haploinsufficiency diseases — the second-order re-rate (STK-002, SYNGAP1, and the discovery engine) is largely unpriced.
Dismantling the bull case:
Not a stock anymore — a closed M&A. Lilly bought the whole company for $10.50/share cash (closed Jul 2025); the only live "position" is the $3.00 CVR, which pays only if VERVE-102 reaches a US Phase 3 dosing — market priced ~21% odds, a coin-flip dressed as a lottery ticket.
A David Liu-pedigreed prime-editing pioneer cornered into a defensive crouch — 25% RIF, founder-CEO out, lead CGD asset demoted to a partnering candidate, going-concern doubt at ~14 months of cash, and an existential Beam IP arbitration over one of its two surviving lead programs. At a ~$537M cap it is a deeply discounted binary call option on 2027 first-in-human liver data; bullish only for risk-seeking capital that can survive a dilution-or-bust 2026.