Scribe Therapeutics

Phase A — Understand the business

Lens 1 · Company Overview

Scribe Therapeutics is a clinical-stage, in-vivo genome-editing company spun out of Jennifer Doudna's UC Berkeley lab in 2017–2018, headquartered in Alameda, California. It is not a Cas9 company. Its platform, X-Editing (XE), is built on CRISPR-CasX — a distinct, smaller CRISPR effector (<1,000 amino acids vs. 1,200–1,400 for Cas9) that the founders re-engineered from a naturally occurring enzyme for activity, specificity and deliverability. Two things define the business model:

1. Platform-as-product, monetised through pharma partnerships. Scribe licenses XE to large pharma for named targets and keeps a wholly-owned cardiometabolic pipeline. Deals to date: Biogen (ALS + a 2nd neuro target; 2020, $15M upfront + up to $400M milestones), Sanofi (2022, expanded 2023; sickle cell + genomic diseases; eligible for >$1.2B milestones + high-single-digit-to-mid-teen royalties), and Prevail Therapeutics / Eli Lilly (2023; neuro/neuromuscular; eligible for >$1.5B milestones). This partner cash is why a Phase 1 biotech booked $51.2M revenue in 2025 (Lens 5).
2. A pivot to owned cardiometabolic assets built on epigenetic silencing. The internal pipeline (STX-1150/1200/1400) doesn't cut DNA — it uses CasX-derived "ELXR" molecules to install durable, reversible epigenetic repression of lipid genes (PCSK9, LPA, a triglyceride target). This is the value story the IPO is being sold on.

Customers/counterparties: pharma partners (Lilly, Sanofi, Biogen) on the platform side; the eventual end-customer for owned assets is the cardiovascular-prevention market (statin-inadequate / high-Lp(a) patients). Suppliers/enablers: LNP delivery (lipid-nanoparticle, liver-targeted) and CDMO manufacturing — standard for the field, not disclosed as proprietary. Contract structure: milestone-and-royalty (back-end-loaded, binary), not recurring revenue — the "$1.5B" headline numbers are eligible aggregates across all programs, not booked.

Lens 2 · Supply Chain (map the actual names)

Upstream → Scribe → end customer, every named stakeholder:

• Discovery inputs: UC Berkeley / Innovative Genomics Institute (origin of CasX + founder IP); large-scale protein + guide-RNA engineering done in-house; DeepXE, an AI guide-RNA prediction model, is the computational input layer.
• Delivery: liver-targeted lipid nanoparticles (LNP) carry the XE payload (STX-1150 uses LNP, dose <1.0 mg/kg). The neuro partner programs (Prevail/Lilly) use self-inactivating AAV (siXAAV) delivery. Scribe does not appear to own a differentiated delivery platform — this is a dependency/chokepoint (see Lens 13).
• Manufacturing: CDMO-based (not disclosed); typical single-source risk for LNP/plasmid supply.
• Downstream (platform): Biogen (neuro), Sanofi (hematology/genomic), Eli Lilly via Prevail (neuro/neuromuscular) are the paying counterparties — a strong, blue-chip buyer list.
• Downstream (owned): cardiovascular prevention — competes against the incumbents' distribution (Amgen Repatha, Novartis Leqvio) at commercialisation, years away.

Chokepoint verdict: delivery (LNP/AAV) is the single-source-style dependency and it is not owned. The editing enzyme is owned and IP-clean; the vehicle that gets it to the liver is the field's shared bottleneck.

Lens 3 · Competitive Advantages (moats)

Three candidate moats, ranked by durability:

1. IP-clean CasX enzyme (strongest). Because Scribe engineered its own CasX effectors rather than licensing Cas9, it sidesteps the Broad-vs-UC Cas9 patent war that entangles CRISPR Therapeutics, Intellia, Editas and Caribou. Owning the molecule end-to-end is a real, durable structural advantage — no royalty stack, no freedom-to-operate cloud. This is the best thing about the company.
2. No-double-strand-break epigenetic silencing (differentiated, unproven at scale). Binding-not-cutting avoids the permanent-edit / off-target-indel liabilities that dog base and prime editors, while claiming siRNA-beating durability (>18 months PCSK9 repression in NHPs after a single dose). The literature supports the concept (Chroma/nChroma HBV, hit-and-run epigenome editing in Nature 2024) but also flags DNMT3A-fusion off-target methylation as a known risk. Reversibility is a double-edged sword — safer, but a silencing state that can drift/reverse is a durability question in humans.
3. Platform validation via tier-1 partners (soft moat). Three of pharma's serious gene-medicine buyers (Lilly, Sanofi, Biogen) paying for XE is external validation and a cash engine — but partnerships are not a moat; they can lapse, and the milestone value is optional and back-loaded.

Bargaining power: low-to-moderate. Against partners, Scribe is the junior party (it needs their capital more than they need its enzyme, given competing platforms exist). Against the eventual patient market, pricing power depends entirely on beating a one-time-dose profile that Verve/Lilly may reach first.

Lens 4 · Segments

No reportable operating segments — single-segment, pre-commercial. Revenue is 100% collaboration/license revenue (milestones + research funding), not product sales. Geographic split n/a — partners are US/EU pharma; lead clinical trial runs in Australia (TGA jurisdiction) for speed. Segment "mix" that matters is pipeline mix: platform-partnered neuro/heme programs (someone else's P&L and risk) vs. wholly-owned cardiometabolic (Scribe's own risk and upside). The IPO reweights the story toward the owned lipid franchise.

---

Phase B — Measure performance (+clinical overlay: Lens 5 → pipeline; Lens 7 → catalysts/mechanism comps)

Lens 5 · Pipeline by phase (overlay swap for "Earnings Result")

The asset table is the company. All wholly-owned unless noted; status per S-1-era disclosures:

Financial snapshot (the reason this isn't a typical cash-bleed Phase 1 story) — from S-1 via secondary coverage:

• Collaboration revenue: $51.2M (FY2025), $27.4M (FY2024), $2.2M (Q1 2026) — 2025 nearly doubled on milestone hits.
• Net loss: $21.8M (FY2025), $47.8M (FY2024), $17.4M (Q1 2026) — the FY2025 loss is unusually small for a Phase 1 biotech because milestone revenue offset ~70% of spend. This flatters the run-rate: Q1 2026 revenue collapsed to $2.2M and the quarterly loss ran $17.4M — i.e., ex-milestones the true burn is ~$60–70M/yr ``.
• Accumulated deficit: $175.1M (as of 31 Mar 2026).
• Cash + investments: $49.7M (as of 31 Mar 2026).
• Going-concern: substantial-doubt language; capital needed beyond IPO proceeds.

Read: the collaboration model has genuinely subsidised the science (rare and creditable), but the milestone revenue is lumpy and non-recurring, cash is thin (<1 year at true burn), and even a fully-subscribed $75M raise + $49.7M ≈ $125M gets them to the STX-1150 readout and not much further — hence the going-concern flag and the need to IPO now.

Lens 6 · Founder / management signal trend (overlay: podcasts & PRs, not earnings calls)

No earnings calls yet. Signal comes from the cadence and framing of communications:

• Narrative has sharpened from "broad platform" → "cardiometabolic prevention company." Early messaging (2020–2022) spanned neuro/ophtho/muscle/metabolic/heme. 2026 messaging is disciplined and singular: "extending healthy lifespan through disease prevention," lipid-first (LDL-C, Lp(a), TG). That focus is the right instinct for a small company going public — but it's also a tell that the broad-platform promise didn't produce an owned clinical asset outside lipids.
• Milestone drumbeat is steady and real: Sanofi milestone (Jan 2025), Prevail/Lilly 1st then 2nd milestone (2025→2026), CIRM >$25M non-dilutive award (Jun 2026), Phase 1 clearance (May 2026), IPO filing (Jul 2026). A company hitting its marks.
• Recurring phrases: "CRISPR by Design," "durable / single-dose," "epigenetic silencing without permanent DNA changes," "potency and specificity." Dropped: the earlier breadth-of-indications language.

Lens 7 · Catalyst calendar + mechanism comps (overlay swap for "Comps")

Catalyst calendar (what de-risks or kills each program):

• H1 2027 — STX-1150 Phase 1 initial data (THE catalyst). First human proof that CasX epigenetic silencing lowers LDL-C durably and safely. Binary for the whole platform thesis.
• 2027 — STX-1200 (Lp(a)) FIH start. Lp(a) has no approved lowering therapy → higher differentiation if it works.
• 2028 — STX-1400 (TG) FIH start.
• Ongoing — undisclosed Lilly/Sanofi milestone triggers (each = cash + validation).
• IPO pricing itself (July 2026) — near-term catalyst; tests public appetite for pre-data, Phase 1, CRISPR biotech.

Mechanism comps (by target, not P/E) — the competitive reality on PCSK9:

The uncomfortable comp: on the exact same target (PCSK9), Verve — now inside Lilly after a ~$1.0B / $10.50-per-share acquisition — is ~2–3 years ahead with human data already at −53% LDL. Scribe's differentiator vs. Verve is reversibility/no-cut, not being first or best on the number. Comps for Lp(a)/TG (STX-1200/1400) are earlier and more open — that may be where the real differentiation lives.

Lens 8 · Stock-Price / valuation-event catalysts (private → funding & product events)

No public trading history. The value-marking events over the company's life:

• Series A $20M (a16z-led, 2018; Biogen pact) → Series B $100M (Avoro/Avoro-led + OrbiMed, a16z, Perceptive, T. Rowe, Wellington, RA Capital, Menlo; Mar 2021) → Sanofi $40M corporate-minority strategic (Jul 2023) → total ~$150M raised pre-IPO.
• Product/deal events that would have re-marked value: Biogen 2020, Sanofi 2022/2023, Lilly-Prevail 2023, Phase 1 clearance 2026, CIRM 2026, IPO filing Jul 2026.
• Pattern: value has been set by milestone + partnership news, not by clinical data — because there is no clinical data yet. That inverts at the H1 2027 readout, which will be the first true data-driven re-rating (up or down).

---

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

Lens 9 · Management

• Benjamin Oakes, PhD — Co-founder, President & CEO. UC Berkeley MCB PhD (2017, Doudna + Savage labs), the actual molecular engineer behind the enhanced-CasX work; IGI Entrepreneurial Fellow; 25+ publications/patents; MIT TR35, Endpoints 20-under-40. Archetype: scientist-founder, deep technical credibility, first-time public-company CEO. Skin in the game is real (founder equity) but this is his first time steering a listed biotech through the capital markets.
• Founding bench is elite: Jennifer Doudna (Nobel 2020, CRISPR co-inventor), Brett Staahl, David F. Savage — top-tier science pedigree, which is what attracted a16z/Avoro and the pharma partners.
• Commercial/finance bench is thin in the public record. Named execs surfaced are mostly VP-level ops (Controller, Program Management, Process Sciences, Talent) — no publicly prominent CFO/CMO name surfaced in this pass (the S-1 will name them; not re-fetched). Flag: a first-time-CEO scientist-founder taking a going-concern-flagged company public on a small raise puts a premium on the quality of the CFO/CMO and board — a gap this web-only read cannot fully close.
• Capital-allocation history: creditably capital-efficient — raised only ~$150M over 8 years yet reached the clinic, because they sold platform optionality to pharma to fund owned science. That is smart, disciplined allocation. The flip side: it kept the company small and cash-light, forcing an IPO into a pre-data window.
• Red flags: none material found (no related-party or promotional-behaviour hits in this pass); the going-concern language is a balance-sheet reality, not a governance red flag.

Lens 10 · Forensic Red Flags

Accounting / disclosure risk (from the reported S-1 shape) — all ``, unaudited by me:

• Revenue recognition is the #1 watch-item. Collaboration revenue is milestone/ratable and lumpy: $51.2M (2025) → $2.2M (Q1 2026). ASC 606 for these deals involves judgement on performance obligations and variable-consideration timing — the FY2025 near-breakeven optics are not a sustainable earnings base. Anyone valuing off the 2025 P&L is being misled by milestone timing.
• Going-concern (substantial doubt) explicitly disclosed — the honest flag: $49.7M cash vs. ~$60–70M true annual burn ``.
• Accumulated deficit $175.1M against ~$150M ever raised — consistent (deficit > raised is normal; funded partly by the ~$134M cumulative collaboration revenue).
• SBC / dilution: expect heavy stock-based comp and pre-IPO preferred → common conversion; the $75M raise is itself dilutive and likely insufficient (more raises to come). Not quantifiable without the S-1 cap table (not re-fetched).
• No inventory/receivables-outrunning-revenue issue (pre-commercial).

Regulatory findings (required sub-section). Per regulatory/regulatory-findings.md (Step 0): Scribe has no CIK and no SEC EDGAR enforcement history is searchable (it was private through the ingest) — total_sec_findings: 0. Web enforcement search ("Scribe Therapeutics" (FTC OR DOJ OR FDA OR settlement OR fine OR penalty) enforcement) and a targeted IP-litigation search returned no material enforcement actions, fines, consent decrees, or lawsuits naming Scribe. Notably, its CasX/IP-clean strategy was designed to avoid the Cas9 patent litigation that entangles peers — no evidence Scribe is a party to those disputes. 10-K Item 3 n/a (no 10-K). Conclusion: No material regulatory or legal findings — verified via SEC EDGAR EFTS (no CIK), web search, and IP-dispute search as of 2026-07-06. (Standard clinical/FDA/TGA trial risk applies but is not an enforcement finding.)

Science & exclusivity (+clinical add-on)

• Mechanism validation: epigenetic silencing durability is supported by independent literature (hit-and-run epigenome editing, Nature 2024; Chroma/nChroma HBV clinical entry) — the concept is de-risked; Scribe's specific CasX-ELXR embodiment in humans is not (H1 2027 is the test). Off-target methylation from DNMT-fusion editors is a documented class risk to watch in the Phase 1 safety data.
• IP estate: the crown jewel — engineered CasX enzymes + XE guide-RNA chemistry + DeepXE, owned outright, freedom-to-operate advantage over Cas9-licensing peers. Patent life on newly-engineered enzymes is long. This is the asset that survives even if STX-1150 fails.
• KOL/founder credibility: Doudna + Oakes = top of the field. Scientific risk is execution/translation, not credibility.

---

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

Lens 11 · Runway-to-catalyst + rough rNPV (overlay swap for "Forward Projection")

No EPS line for a pre-revenue biotech — the questions that matter are (a) does cash reach the next value-inflection, and (b) what is the lead asset worth risk-adjusted.

Runway-to-catalyst (the decisive one):

• Cash 31 Mar 2026: $49.7M. True burn ~$60–70M/yr ``. Undisclosed Lilly/Sanofi milestones + CIRM $25M (staged) extend it.
• IPO net proceeds ~$60–68M `` → pro-forma cash ~$110–115M.
• Does it reach the H1 2027 STX-1150 readout? Yes, comfortably — ~1.5+ years of runway to the single most important catalyst. Does it fund beyond that to STX-1200/1400 clinical proof? No — expect another raise (follow-on or a new partnership) in 2027, especially if the readout is good. This is the core +private/+clinical finding: the IPO is a bridge to one Phase 1 readout, not to self-sufficiency.

Rough rNPV of the lead (STX-1150, PCSK9) — illustrative, every input ``:

• Peak unadjusted sales if best-in-class one-and-done PCSK9: the PCSK9 category is real (Repatha ~$2.2B 2024; Leqvio ~$754M 2024 and growing). A durable single-dose entrant might address ~$1.5–2.5B peak `` — but only if it beats Verve/Lilly's profile and timing.
• PoS for a Phase 1, pre-human-data epigenetic asset in cardio: ~10–15% ``.
• rNPV lead asset ≈ peak $2.0B × ~3–4× sales-to-NPV heuristic × 12% PoS, discounted → very roughly $0.5–1.0B ``, extremely sensitive to the Verve-is-ahead discount. Treat as order-of-magnitude only — not a sourced valuation.
• Add optionality value of the platform + Lilly/Sanofi/Biogen milestone tails (each program a lottery ticket on up to $1.2–1.5B aggregate, PoS-weighted low).

No forecast.ts create — no committed base case in unattended breadth mode, and the honest scoreable binary is the H1 2027 readout, not an EPS line.

Lens 12 · Bull vs Bear

Bull case. Scribe is the rare CRISPR platform that (1) owns its editing enzyme and dodges the entire Cas9 IP war; (2) has a genuinely differentiated no-cut, reversible epigenetic-silencing modality with best-in-field durability data (>18mo PCSK9 in NHP); (3) has been validated and funded by three tier-1 pharmas (Lilly, Sanofi, Biogen) plus a16z/Avoro/T. Rowe/Wellington, keeping dilution low and reaching the clinic on ~$150M; (4) is attacking a proven, large cardio-prevention market with a "one-and-done" profile patients and payers want; and (5) has non-dilutive CIRM validation and a disciplined lipid-first focus. If STX-1150 shows durable, safe LDL lowering in H1 2027, this re-rates hard and becomes an obvious pharma acquisition target — exactly the Verve→Lilly playbook, and Lilly already knows Scribe intimately via Prevail. Contrarian bull: the market is fixated on Verve-is-ahead on PCSK9 and is undervaluing the reversibility/safety angle and the Lp(a) and TG programs, where the field is wide open and epigenetic silencing may be genuinely first-in-class.

Bear case (2–3 permanent-impairment risks). (1) Second-best on the flagship target. Verve/Lilly and YolTech already have human PCSK9 data at −53%/−52%; Scribe won't have human data until H1 2027. If the market decides one durable PCSK9 editor is enough, STX-1150 is a fast-follower into a target its own partner already owns a better asset for — a strategic dead-end. (2) Reversibility cuts both ways. The safety story (no permanent edit) is also a durability question: if human silencing drifts or partially reverses, the "one-and-done" pitch collapses into "redose like siRNA," and inclisiran already owns that niche. (3) Balance sheet. Going-concern flag, $49.7M cash, $75M raise into a pre-data window — a weak IPO or a soft readout forces dilutive down-rounds or a fire-sale partnership. Pre-mortem (18 months out, thesis broke): STX-1150 Phase 1 shows LDL lowering but a safety signal (off-target methylation / LNP tolerability) or insufficient durability; the stock (having IPO'd pre-data on hype) halves; cash dwindles; Lilly declines to expand and Scribe raises at a crushing discount or sells the platform for parts. Multiples: for a pre-data Phase 1 biotech there is no earnings multiple — valuation is pure narrative/PoS; a $75M raise implies a modest micro-cap that could be cheap if the readout hits and expensive if it's just Verve's shadow.

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

Dismantling the bull case:

• Revenue concentration is brutal and optional. ~$134M cumulative revenue is entirely pharma milestones from three partners — non-recurring, at the partners' discretion, and it cratered to $2.2M in Q1 2026. The "profitable-ish 2025" is a milestone-timing mirage; the real business burns ~$60–70M/yr with $49.7M in the bank. If Lilly or Sanofi walks, the cash engine and the validation story die together.
• The moat may be narrower than bulls think. Owning CasX is real, but the delivery vehicle (LNP/AAV) — the actual rate-limiter for in-vivo editing — is not owned and is the same commoditised bottleneck everyone fights. A better enzyme delivered by the same LNP as everyone else is a marginal, not structural, edge.
• The most dangerous competitor is Scribe's own partner. Eli Lilly bought Verve for ~$1.0B and has VERVE-102 years ahead on PCSK9 with −53% human data. Lilly has no reason to prioritise Scribe's PCSK9 asset over the one it just paid $1B for — it may keep Scribe only for the neuro/Prevail work and let STX-1150 wither. Bulls underestimate that Scribe's flagship indication is already spoken for inside its own biggest partner.
• Capital-allocation/incentive risk: IPO-ing a going-concern company pre-data, on a $75M raise, with a first-time-CEO founder, is a "raise now because you have to" event, not a position of strength. Insiders (a16z, Avoro, T. Rowe) taking a company public before human data is a liquidity-seeking tell.
• What single scenario permanently impairs it? STX-1150 shows a class-type off-target/safety signal or durability that fades — plausible for a first-in-human novel epigenetic modality, and it would indict the whole platform, not just one asset. Plausibility: moderate. If growth/PoS on the lead asset disappoints by 20–30%, the equity (pure-narrative valued) could fall 40–60%.

Lens 14 · Management Questions (ordered by information value)

1. STX-1150 Phase 1: what LDL-C reduction, durability, and off-target/safety threshold would you consider a clear success vs. a kill — and against Verve-102's −53% human benchmark, specifically?
2. How is durability of epigenetic silencing (vs. permanent editing) tracking in your longest NHP cohorts, and what is the reversal/drift risk in humans?
3. Given Lilly owns Verve's PCSK9 asset, what is your candid read on Lilly's commitment to your PCSK9 program vs. the Prevail neuro work — and does the Prevail deal restrict your cardio franchise?
4. Cash is $49.7M with a going-concern flag; what does IPO-plus-cash fund to, and when is the next raise or partnership realistically needed?
5. What is the true, ex-milestone annual operating burn, and how should investors model collaboration revenue that swung from $51.2M to $2.2M in a quarter?
6. Why IPO before human data rather than after — what does the public raise give you that another partnership couldn't?
7. Do you own or license your LNP/AAV delivery, and how differentiated is delivery vs. competitors who use the same GalNAc-LNP approach?
8. For STX-1200 (Lp(a)) and STX-1400 (TG) — where the field is more open — why not lead with those to differentiate from the crowded PCSK9 target?
9. What is the freedom-to-operate and patent-life picture on engineered CasX + ELXR + DeepXE, and how defensible against a fast-follower?
10. What are the remaining eligible milestones and booked-to-date cash from each of Biogen, Sanofi, and Prevail/Lilly?
11. What manufacturing/CDMO dependencies and single-source risks exist for STX-1150, and what's the CMC readiness for pivotal scale?
12. How do you compete on price/access at commercialisation against entrenched Repatha/Leqvio and a potentially-first Lilly one-dose editor?
13. Who are the CFO, CMO, and independent board members, and what public-company / cardio-development experience do they bring?
14. What is the regulatory path and endpoint strategy (surrogate LDL vs. CV-outcomes) — and does epigenetic silencing raise novel FDA questions vs. base editing?
15. If a large pharma offered to acquire Scribe pre-Phase-1-data, what value would justify staying independent?

---