uniQure

Phase A — Understand the business

Lens 1 · Company Overview

uniQure N.V. is an Amsterdam-headquartered (Nasdaq: QURE), Dutch-incorporated clinical-stage AAV gene-therapy company — "seeking to deliver… single treatments with potentially curative results" for rare, devastating diseases . It is a *pipeline*, not a P&L: 2025 total revenue was **$16.1M**, essentially all a royalty on a drug it no longer owns, against a **$199.0M** net loss . Founded 1998 as Amsterdam Molecular Therapeutics (AMT); reorganized into uniQure B.V. in 2012; IPO'd Feb 2014 ``.

What it actually is in mid-2026: a single-asset bet on AMT-130 for Huntington's disease (HD). AMT-130 is an AAV5 vector carrying a miRNA (the proprietary miQURE gene-silencing platform) designed to lower the huntingtin gene and the toxic exon-1 protein fragment. It is a one-time, surgically-delivered (intrastriatal, MRI-guided) therapy. If approved it would be the first disease-modifying therapy for HD — a disease with ~30,000 diagnosed U.S. patients and no drug that slows the underlying neurodegeneration ``.

The rest of the pipeline (support cast): AMT-260 (temporal-lobe epilepsy / MTLE, Phase I/IIa), AMT-191 (Fabry disease, Phase I/II), and — as of Sept 2025 — a discontinued AMT-162 (SOD1-ALS, killed after a dose-limiting toxicity/SAE) ``.

Revenue model & key terms. The only recurring revenue is a royalty on HEMGENIX (etranacogene dezaparvovec, gene therapy for Hemophilia B) — the world's first approved AAV gene therapy for Hemophilia B, list price ~$3.5M/patient. Critically, uniQure sold the exclusive global rights to HEMGENIX to CSL Behring in 2021 (the "License Sale") for large upfront/milestone payments, and then sold a portion of the future HEMGENIX royalty stream to HemB SPV, L.P. in May 2023 for ~$370.1M (net) up front . So the "Hemgenix royalty base" is **monetized and partly encumbered** — the ~$15.9M 2025 license revenue is real but is being offset against a royalty-financing liability that accrues **~$54M/yr of non-cash interest expense** (see Lens 10). uniQure is an *agent*, not a principal, in HEMGENIX manufacturing (title passes Genezen→CSL directly). In April 2026 the parties agreed to **terminate** the residual supply/CSA obligations once contracted batches ship (mid-2026), leaving uniQure with a clean royalty-only relationship .

Customers / suppliers / competitors: Customer — CSL Behring (the only revenue counterparty; AR of $3.6M at Q1'26). Supplier — Genezen (CDMO; bought uniQure's Lexington, MA plant in the July 2024 "Lexington Transaction"; uniQure holds a $14.9M convertible note receivable from Genezen). Competitors — in HD specifically: Roche/Ionis (tominersen), Wave Life Sciences (WVE-003), PTC Therapeutics (PTC518), Prilenia (pridopidine); broadly named in the 10-K: Novartis, Roche, Wave, Alnylam, Regeneron, Skyhawk Therapeutics ``.

Lens 2 · Supply Chain

Map: AAV vector manufacturing (CDMO) → uniQure clinical/commercial supply → surgical administration → patient.

• Upstream / manufacturing chokepoint — Genezen (single source). uniQure divested its own commercial manufacturing (Lexington, MA) to Genezen in July 2024 and now depends on Genezen as CDMO for both HEMGENIX (on CSL's behalf) and, prospectively, AMT-130 supply. This is a genuine single-source dependency and a concentration of credit risk: uniQure carries $12.3M of prepaid Genezen services + a $14.9M Genezen convertible note (due Sept 2029) ``. For a company whose entire value is one AAV product, outsourcing the AAV supply chain to a third party it also finances is a structural fragility.
• Midstream — process validation / CMC. The single largest 2025 cost increase inside AMT-130 was $19.4M of BLA-prep + manufacturing process-validation / process-performance-qualification (PPQ) spend ``. AAV commercial-scale CMC is historically where gene therapies stumble (comparability, potency assays) — and it is now on Genezen's shop floor, not uniQure's.
• Administration chokepoint — neurosurgery. AMT-130 is delivered by MRI-guided stereotactic intraparenchymal (intrastriatal) infusion — a specialized neurosurgical procedure, not an IV. This caps the realistic number of dosing centers and the launch slope even if approved: the most common adverse events in the trial were "related to the administration procedure" ``.
• Inputs: plasmids, HEK293/insect-cell production, fill-finish — standard AAV inputs, not disclosed as constrained.
• End customer: for HEMGENIX, CSL Behring (then hemophilia treatment centers). For AMT-130, HD specialty/neurology centers, reimbursed by payers at gene-therapy pricing (~$2M+; see Lens 3/11).

Verdict on the chain: thin and externalized. One CDMO, one surgical modality, one meaningful royalty counterparty (CSL). Names present, single-source flags real.

Lens 3 · Competitive Advantages (moats)

• Asset/data moat — the strongest one, and it's specific. AMT-130 is the only HD program with 3-year controlled clinical outcomes showing statistically significant slowing of disease progression (75% on cUHDRS, 60% on TFC — Lens 5) and, uniquely, RMAT designation — the first ever granted in Huntington's disease — plus Breakthrough Therapy and Fast Track ``. A first-mover, disease-modifying, one-time therapy in a disease with zero disease-modifying options is a formidable position if the label happens.
• Platform / IP moat — miQURE. The gene-silencing miQURE platform (miRNA in AAV) is proprietary and reused across AMT-130/AMT-260. But platform value in gene therapy has repeatedly proven softer than bulls hope — the moat is the clinical de-risking of the lead asset, not the platform abstraction.
• Switching costs / durability. A one-time curative-intent therapy has near-total switching costs per patient (you don't re-dose) but that cuts both ways: it also means no recurring revenue per patient and a finite prevalence pool that gets "used up."
• Bargaining power. Weak vs. CSL historically (uniQure sold HEMGENIX outright and then sold the royalty — a company selling its crown jewel twice is not negotiating from strength). Stronger vs. future HD payers if it is the only approved disease-modifier.
• Modality edge vs. rivals. AMT-130's competitive differentiation is single-administration + durable exon-1-inclusive silencing vs. the chronic re-dosing of ASOs (tominersen, WVE-003) and oral splicers (PTC518). But rivals have safety/convenience counter-arguments — ASOs are titratable and reversible; AMT-130 is irreversible brain surgery.

Commercial-layer files for genomics (bottlenecks.md, positioning.md, supply-chain.md) are missing on disk ``, so moat framing here is filing- + web-derived rather than KB-compiled.

Lens 4 · Segments

uniQure has no product segments — it is single-development-stage with one royalty line. The only meaningful "segmentation" is R&D spend by program, which is the truest map of where the company is placing its bets ``:

Geography: operations split Netherlands (Amsterdam HQ) + France (uniQure France SAS, source of the AMT-260 program and its EUR 160M / ~$184M contingent milestone tail) + U.S. (Lexington divested). Revenue is effectively 100% U.S.-derived (CSL/HEMGENIX royalty). No geographic revenue split is meaningful pre-commercial.

The segment story in one line: the company deliberately narrowed from a broad AAV platform to a HD-first, two-support-asset pipeline, killing SOD1-ALS and shrinking preclinical — a focusing move, not a growth story.

---

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

Lens 5 · Pipeline by phase (replaces "Earnings Result")

The asset table is the company. Data cutoffs and status per the filings + the June-2026 update:

AMT-130 — the 36-month high-dose efficacy data (data cutoff 30 Jun 2025; n=12 high-dose patients at 36mo vs propensity-matched Enroll-HD external control n=940) ``:

• cUHDRS (primary): 75% slowing of progression, p=0.003 — mean change −0.38 (treated) vs −1.52 (control). Primary endpoint met.
• TFC (key secondary): 60% slowing, p=0.033 — −0.36 vs −0.88. Met.
• SDMT: 88% slowing, p=0.057 (miss on strict 0.05). SWRT: 113% slowing, nominal p=0.002. TMS: 59% slowing, nominal p=0.174 (not significant).
• CSF NfL: −8.2% mean reduction at 36mo (supportive neurodegeneration biomarker).
• Dose-dependent (high-dose favorable vs variable low-dose); "generally well-tolerated"; most AEs procedure-related.

This is a genuinely strong, statistically significant, controlled dataset in a disease that has buried every prior drug (tominersen Ph3 failed 2021; pridopidine mixed). The caveats bulls must own: n=12 at the 36-month timepoint, an external (not randomized) control, and endpoints that mix hit (cUHDRS/TFC) and miss (SDMT/TMS). That is exactly why the FDA wobbled.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the research-layer shelf (transcripts/ is empty) . Sentiment reconstructed from filings + dated PRs :

• Nov 2024 – Apr 2025 (three Type B meetings): confident — FDA agreed external-control data could be the primary BLA basis under accelerated approval, and that cUHDRS could be an intermediate endpoint / NfL supportive. Management tone: launch-preparation mode (hiring commercial staff, SG&A rising).
• Sept 2025 (topline data): triumphant — "positive topline," pivotal framing, pre-BLA scheduled. This is the window (Sept 24–Oct 31 2025) the securities class action targets (Lens 10).
• Dec 2025 → Mar 2026 (the crash): the tone in the 10-K/10-Q is defensive and disclosure-heavy — FDA "unlikely to provide the primary evidence" (Dec), then "cannot agree… strongly recommended a randomized, double-blind, sham-surgery-controlled study" (Mar). Management pivots to "continue engaging" + a UK MHRA path as a hedge.
• Jun 2026 (redemption): CEO Matt Kapusta: "The FDA has agreed that our current clinical data can support a near-term BLA submission and has committed to work expeditiously with us to align on the design of the required confirmatory study." ``.

Recurring phrases: "single treatments with potentially curative results," "first disease-modifying therapy for Huntington's," "accelerated approval pathway." What changed: the confidence around the regulatory path swung violently three times in ~18 months — the clinical data never changed, the FDA's read of it did. That volatility is the investment risk profile.

Lens 7 · Catalyst calendar + mechanism comps (replaces P/E comps)

Catalyst calendar (what de-risks or kills the thesis, and when):

Mechanism comps — HD competitive set (by target/modality, not P/E):

**Valuation comps — gene-therapy peers (market cap, , mid-2026):** QURE **~$3.08B** (~63.07M sh, late Jun 2026) ; Krystal Biotech (KRYS) ~$11.06B (commercial, profitable — the "what good looks like" comp) ; Sarepta (SRPT) **~$2.22B** (post-Elevidys troubles) ; Rocket Pharma (RCKT) ~$0.38B (approved but tiny) ``. Traditional multiples (EV/Sales, P/E, ROE) are n/a / not meaningful for a pre-revenue developer; the honest comp is clinical/regulatory de-risking vs. market cap, and on that axis QURE is priced richly vs. RCKT/SRPT and cheaply vs. KRYS — i.e., priced as likely-approvable, not-yet-commercial.

Lens 8 · Stock-Price Catalysts (the >5% movers)

This name trades on binary regulatory/clinical headlines, not fundamentals ``:

• Sept 2025 — positive 3-yr topline: sharp rally; the stock and pre-BLA optimism drove the follow-on offering at $47.50 (5.79M sh + pre-funded warrants), ~$404M aggregate net proceeds ``.
• Oct-Dec 2025 — FDA "unlikely"/"failed product": collapse. FDA reportedly called AMT-130 a "failed product" late 2025 ``. Triggered the securities class action.
• Mar 2026 — FDA "cannot agree," wants randomized sham study: further leg down / overhang.
• 17 Jun 2026 — FDA reverses, accepts 3-yr data for accelerated-approval BLA: stock +~70% intraday . "The real catalyst… was the FDA changing its stance… that single regulatory pivot removed a long-running overhang." uniQure immediately raised **~$259M gross at $45.50** into the strength .
• Net: up ~240% over the trailing year to ~$48.85 ``.

What the tape reveals: the market reacts almost exclusively to FDA posture on AMT-130. Efficacy data is necessary but not sufficient — the same 3-year dataset produced both a "failed product" and an "acceptable BLA basis." This is a regulatory-headline stock, and management has skillfully sold equity on every up-leg.

---

Phase C — Judge people & books (+clinical overlay: Science & exclusivity added)

Lens 9 · Management

• CEO — Matthew Kapusta (also Director). Long-tenured; formerly CFO before taking the CEO seat — a finance-first leader, which shows in the capital strategy: he sold HEMGENIX to CSL (2021), sold the HEMGENIX royalty to HemB SPV (2023, ~$370M), divested the Lexington plant to Genezen (2024), and raised ~$404M (Sept 2025) + ~$259M (Jun 2026) of equity — serial monetization to fund a single high-conviction asset. ``
• CFO — Christian Klemt. CLO — Jeannette Potts.
• Board: Madhavan Balachandran, Robert Gut, Rachelle Jacques, Jack Kaye, David Meek, Jeremy Springhorn ``. David Meek (ex-Ipsen/FerGene CEO) and Rachelle Jacques (ex-Enzyvant/Alexion) bring rare-disease commercial credibility — useful for an HD launch.
• Skin in the game: directors/officers/5%+ holders beneficially owned ~47.1% of share capital at 31 Dec 2025 . But *insider* ownership specifically is ~4.9%, institutional ~96.6% . Two insider sells in the past three months, no insider buys `` — CEO Kapusta adopted a 10b5-1 plan (332,576 sh) on 5 Oct 2025, days after the Sept topline and before the Dec FDA reversal — optically uncomfortable given the class action window is Sept 24–Oct 31 2025 (Lens 10).
• Capital-allocation verdict: aggressive and shareholder-dilutive but arguably correct for the situation. Kapusta has kept a binary company solvent to H2 2029 without a partner by repeatedly selling equity at highs and monetizing non-core assets. The flip side: shareholders have been diluted heavily (share count grew via two large raises), and the company has twice sold its only commercial-stage economics (HEMGENIX product + royalty) — you're left owning a pure clinical option, by design.
• Archetype: professional-manager / financial-engineer, not scientist-founder. For a de-risking-toward-BLA stage, that's a reasonable fit — the science is set; the game now is regulatory navigation + balance-sheet management, which is his wheelhouse.

Lens 10 · Forensic Red Flags

Every figure `` unless noted.

• Going concern: clean, but engineered. Financials prepared on a going-concern basis; cash + investments $586.6M at 31 Mar 2026 ($622.5M at YE2025), runway into H2 2029 ``. No going-concern doubt flagged. But the runway is a product of serial equity sales, and the 10-Q's own liquidity language is heavily caveated ("assumptions that may prove to be wrong… could exhaust our available capital resources sooner").
• The royalty-financing liability is the single biggest accounting subtlety. The May-2023 sale of HEMGENIX royalties to HemB SPV for ~$370M was booked as debt-like financing, not a sale — it generates non-cash interest expense of ~$54.1M (2025) / ~$50.9M (2024) / $12.6M in Q1'26 alone ``. This means: (a) reported "license revenue" from HEMGENIX is largely swallowed by interest on the money already taken against it, and (b) net loss is inflated by a large non-cash charge. Cash flow diverges from GAAP earnings favorably here (the interest is non-cash) — a rare case where the accounting looks worse than the cash reality on this line. Watch that the liability's amortization assumptions (royalty forecast) don't get revised.
• Contingent consideration (uniQure France / AMT-260): EUR 160M (~$184M) of milestone payments owed to former uniQure France shareholders, payable 2030-2034, up to 25% settleable in shares; fair value EUR 15.9M ($18.7M) at YE2025 and swings through R&D expense ($6.2M loss in 2025) as probabilities/discount rates change ``. A soft, judgment-heavy liability that could balloon if AMT-260 advances.
• SBC: ~$17.7M total in 2025 ($7.6M R&D + $10.1M SG&A) — moderate for a biotech, down YoY on restructurings ``. Not flattering non-GAAP aggressively (uniQure reports GAAP net loss).
• FX noise: Dutch entity → large FX swings ($26.1M gain 2025 vs $10.5M loss 2024) on euro-denominated debt/cash — non-operating volatility, not a red flag but a reason net loss is noisy.
• Internal controls: management concluded ICFR effective at YE2025 (auditor attested); a new ERP went live Jan 2026 (disclosed control-environment change, not a weakness) ``.

Regulatory findings (required sub-section):

• SEC Litigation Releases / AAERs: None. "No LR found… No AAER found for this company" via SEC EDGAR EFTS (LR + AAER), period 2021-07-06 to 2026-07-06 ``.
• Item 3 Legal Proceedings (company's own disclosure): A securities class action. Christopher Scocco v. uniQure N.V., et al., Case No. 1:26-cv-01124, filed 10 Feb 2026, SDNY, against the company and certain officers, under §10(b)/§20(a) + Rule 10b-5, on behalf of investors who bought QURE between Sept 24 and Oct 31, 2025 — alleging false/misleading statements about the AMT-130 Phase I/II study and BLA-filing timing. uniQure "intends to vigorously defend"; loss not estimable ``. This is the material legal risk — it directly targets the Sept-2025 "positive topline" messaging that preceded the Dec-2025 FDA reversal, and it overlaps the CEO's 5-Oct-2025 10b5-1 plan adoption. The June-2026 FDA re-approval of the pathway helps the narrative defense but does not moot the class-period conduct.
• Non-SEC enforcement (FTC/DOJ/FDA/etc.): Web search surfaces no FTC/DOJ/CFPB enforcement action; FDA interactions are ordinary-course regulatory (the AMT-130 saga), not enforcement ``.
• Summary: No SEC/accounting-enforcement history; one active securities class action tied to the AMT-130 disclosure whipsaw is the sole material legal finding.

Science & exclusivity (+clinical add):

• Mechanism validation: HTT-lowering is a validated hypothesis (genetic causality is unambiguous in HD); the open question was always whether lowering slows clinical progression — AMT-130's 3-yr cUHDRS/TFC significance is the first controlled human evidence that it can.
• IP / exclusivity: miQURE platform + AMT-130 composition/method patents (patent estate not on the shelf — patents/ empty). Note the St. Jude license underlying HEMGENIX hFIX patents: majority of U.S. rights expire 2028, European (and one U.S.) expired 2025 `` — relevant to the HEMGENIX royalty tail, not AMT-130. For AMT-130, RMAT/Breakthrough/orphan-style regulatory exclusivity + biologics data exclusivity would anchor the moat post-approval.
• KOL/reimbursement: HD is a well-defined, KOL-concentrated rare disease; payer path at ~$2M+ gene-therapy pricing is plausible for a one-time disease-modifier but will face outcomes-based-contract pressure.

---

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

Lens 11 · rNPV + runway-to-catalyst (replaces EPS projection)

No EPS forecast — uniQure will not have meaningful product revenue for years, and I will not fabricate an EPS line. The right frame is (a) does cash reach the value-inflection catalyst, and (b) a scenario-weighted rNPV of AMT-130.

Runway to catalyst — the question that matters:

• Quarterly operating cash burn ≈ $38.2M (Q1'26 net cash used in ops) . Annualized ≈ **$150-160M** .
• Cash + investments $586.6M (31 Mar 2026), plus ~$259M gross raised Jun 2026 → ~$800M+ pro-forma + ~$259M gross less fees].
• Company-stated runway: into H2 2029 (pre-June raise) ``; the June raise extends it further. Runway comfortably clears the Q3-2026 BLA, a ~2027 approval decision, and the start of the confirmatory study. Runway is not the binary here — regulatory outcome is.
• Optional +$100M Hercules term-loan draw if BLA approved before 15 Jun 2027 ``.

rNPV of AMT-130 (all inputs ``, shown):

• Addressable: ~30,000 diagnosed U.S. HD patients ``; assume ~40-50k US+EU diagnosed, realistically ~15-25k treatable (staged, eligible, surgically appropriate).
• Price ~$2.0M/patient ``.
• Peak penetration: bull ~40-60% of treatable pool over years → analyst peak-sales range $2-4B . Base case, haircut for surgical-capacity + payer friction + competition: **~$1.0-1.5B peak** .
• PoS to accelerated approval: ~45-55% `` — FDA re-engaged + RMAT, but confirmatory study un-designed and external-control precedent is contentious.
• Discount 12-13%; single-asset, single-country-launch risk.
• Illustrative base rNPV ≈ current ~$3B market cap `` — i.e., the stock is roughly fair-to-full on a risk-adjusted base case; it is cheap only if you underwrite approval at >60% and peak sales at the high end, and expensive if the BLA slips or the confirmatory study terms are onerous. The ~240% YoY move has already harvested most of the "FDA-reverses" optionality.

Brier forecast to log (for calibration; not created in --watchlist per skill): "QURE — FDA accepts the AMT-130 BLA for review (filing acceptance), by 30 Jun 2027, p≈0.70" and "AMT-130 receives FDA accelerated approval by 31 Dec 2027, p≈0.45." (Logged conceptually; forecast.ts create skipped in breadth mode.)

Lens 12 · Bull vs Bear

Bull case. uniQure owns the only HD asset with statistically significant, 3-year, controlled evidence of disease slowing (75% cUHDRS, p=0.003), the first-ever HD RMAT, and — after the June-2026 reversal — an FDA that will accept its data for an accelerated-approval BLA filed Q3 2026. In a disease with 30,000 US patients and zero disease-modifying options, first-mover approval of a one-time therapy priced ~$2M+ supports $2-4B peak sales. Cash to H2 2029+ removes financing risk through the catalyst. Rare-disease-savvy board (Meek, Jacques) can execute a specialty launch. If approved, this re-rates toward the Krystal ($11B) end of the gene-therapy comp set, not the Rocket ($0.4B) end.

Bear case (permanent-impairment risks).

1. The data may not survive the confirmatory bar. Accelerated approval is conditional on a confirmatory study uniQure hasn't designed or started; if that study (concurrent-SOC or otherwise) fails or reads out negative, approval is withdrawn — and the external-control Phase I/II (n=12 at 36mo) is exactly the kind of dataset that a randomized study can contradict. The FDA already called it a "failed product" once.
2. The stock has priced the win. Up ~240% YoY to ~$3B on the regulatory posture change, not new efficacy — the asymmetry that existed at the trough is largely gone. A BLA slip, refusal-to-file, or harsh confirmatory terms resets it.
3. Single-asset, single-modality, single-CDMO concentration — plus irreversible brain surgery as the delivery, capping launch slope and inviting oral/ASO competitors (PTC518 oral, WVE-003 allele-selective) to erode the pool over time.

Pre-mortem (18 months out, thesis broke): The Q3-2026 BLA was filed but the FDA refused to file it or issued a CRL demanding the randomized confirmatory data before approval; the class action advanced on the Sept-2025 messaging; a fresh equity raise at a much lower price re-diluted holders; the stock round-tripped to pre-June-2026 levels. The 75%/60% numbers were real but "not enough patients, not randomized" won inside FDA on second thought.

Are multiples too high? There are no earnings multiples; on rNPV the base case ≈ the market cap, so not obviously mispriced either way — the debate is entirely PoS and peak-sales, not valuation mechanics.

Contrarian view (what the market may be refusing to see): The market is treating June-2026 as "de-risked to approval," but the FDA only agreed the data can support a filing — it has not agreed to approve, and it explicitly wants a confirmatory study whose design is unsettled. The actual remaining binary (filing-acceptance → approval → confirmatory readout) is being under-priced as a formality after a 70% relief rally. Conversely, the deeper contrarian long is manufacturing/commercial execution: if AMT-130 is approved, the bottleneck won't be demand, it'll be neurosurgical dosing capacity and CDMO scale — an underappreciated upside constraint on the ramp that also means the TAM converts slowly (good for durability, bad for near-term revenue vs. the $2-4B headlines).

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case:

• Revenue concentration = zero real revenue. The only revenue (~$16M) is a HEMGENIX royalty on a drug uniQure sold twice (product to CSL, royalty to HemB SPV) — and the St. Jude hFIX patents underpinning it start expiring 2028. There is no commercial business to fall back on; this is a pure option on one surgical brain therapy.
• The moat is a regulatory coin-flip, and the coin has landed on both sides. The identical 3-year dataset was, within seven months, an "acceptable BLA basis" (Apr 2025), a "failed product" (late 2025), a "cannot agree" (Mar 2026), and "acceptable" again (Jun 2026). A moat that depends on which way the FDA is leaning this quarter is not a moat.
• Most dangerous competitor bulls underestimate: PTC Therapeutics (PTC518) — an oral, daily HTT-lowerer. If a pill slows HD, patients and payers will resist irreversible MRI-guided intracranial surgery. AMT-130's "one-time" advantage is also its ceiling: finite prevalence, used up once, no chronic revenue, and surgery caps throughput.
• Capital allocation red flags: serial dilution (two large raises in <12 months), a CEO 10b5-1 plan adopted 5 Oct 2025 straddling the class-action window, twice-sold crown-jewel economics, and a single-source CDMO (Genezen) that uniQure also finances (prepaids + convertible note) — related-party-adjacent supplier entanglement.
• What must hold for today's price: BLA filed Q3 2026 and accepted and approved on accelerated basis and a confirmatory study that ultimately confirms — a chain of ~4 conditional gates, each <100%, compounding to well below the ~"approval is coming" mood.
• If growth/PoS disappoints 20-30%: halve peak-sales or PoS and the base rNPV falls meaningfully below the current ~$3B — the stock is only "cheap" under bull inputs.
• Single scenario that permanently impairs: a CRL or a failed confirmatory study on AMT-130 — there is no second asset material enough to catch the company (AMT-162 already dead, AMT-191 on a DLT pause, AMT-260 early). This is a one-asset company wearing a platform's clothes.

Lens 14 · Management Questions (ordered by information value)

1. What exactly did the FDA agree at the June-2026 Type B meeting — data to support a filing, or a path to approval — and what specifically must be true in the BLA for it to be accepted for review (not refused-to-file)?
2. What is the confirmatory-study design the FDA will accept — concurrent standard-of-care control vs sham — and its size, endpoints, and timeline? Can it enroll and read out before an accelerated-approval withdrawal risk?
3. Given the FDA reversed twice, what changed at the agency (personnel, policy, review division posture) between "failed product" and "acceptable BLA basis," and how durable is that?
4. With n=12 high-dose patients at 36 months and an external control, what is your statistical answer to the "too few patients, not randomized" critique that the FDA raised in March 2026?
5. What is the realistic annual dosing capacity (qualified neurosurgical centers) at launch, and what does the revenue ramp look like given surgery, not demand, is the constraint?
6. Is AMT-130 manufacturing (Genezen) validated at commercial scale, and what is your contingency if your single CDMO — which you also finance — has a comparability or supply failure?
7. How should shareholders think about further dilution: after ~$404M (Sept 2025) + ~$259M (Jun 2026), what is the financing plan through approval, and will you draw the $100M Hercules facility?
8. What is the exposure and defense strategy on the Scocco securities class action, and does the June-2026 FDA reversal materially change the litigation calculus for the Sept-2025 class period?
9. On pricing: what is the target price per patient, and what outcomes-based/warranty structures are you prepared to offer payers for a one-time HD therapy?
10. How do you defend against oral (PTC518) and allele-selective (WVE-003) competitors on the "why undergo brain surgery" question over a 5-year horizon?
11. What is the plan for the HEMGENIX royalty as the St. Jude hFIX patents expire (EU 2025, US 2028) and the HemB SPV royalty-financing liability amortizes?
12. AMT-191 Fabry: what is the path off the DLT (Grade-3 liver enzyme) dosing pause, and is there a dose that preserves the strong ERT-withdrawal efficacy without the toxicity?
13. AMT-260 MTLE: what initial-efficacy signal would justify the EUR 160M contingent-milestone commitment to advance to Phase III?
14. What ex-US regulatory strategy (UK MHRA Q3 2026, EU EMA) de-risks a single-agency dependence on the FDA?
15. What is the long-term corporate strategy if AMT-130 is approved — build a standalone HD commercial franchise, or is uniQure a partnership/acquisition candidate at that point?

---