Precision BioSciences

Phase A — Understand the business

Lens 1 · Company Overview

Precision BioSciences is a Durham, NC clinical-stage biotech built on ARCUS — a proprietary genome-editing platform derived from a natural homing endonuclease (I-CreI), licensed originally from Duke University (April 2006). After a multi-year strategic retrenchment, the company is now "solely focused" on wholly-owned in vivo gene editing, having divested or out-licensed its entire cell-therapy / allogeneic CAR-T franchise.

Pipeline (the company):

• PBGENE-HBV (lead) — wholly-owned, in the global first-in-human ELIMINATE-B Phase 1/2a trial for chronic hepatitis B. Designed as a potentially curative therapy that eliminates cccDNA (covalently closed circular DNA), the persistent viral reservoir and "sole source of viral replication." Described as "the only clinical-stage program that targets the elimination of cccDNA". FDA Fast Track granted April 2025. Trial sites in Hong Kong, New Zealand and elsewhere.
• PBGENE-DMD — wholly-owned, for Duchenne muscular dystrophy. Uses two complementary ARCUS nucleases in a single AAV to excise exons 45–55, addressing ~60% of DMD patients. IND cleared (early 2026); the FUNCTION-DMD Phase 1/2 trial is activating. Carries Rare Pediatric Disease designation (Jun 2025), Orphan Drug (Jul 2025), Fast-Track (Feb 2026), and is PRV-eligible — a sellable priority-review voucher worth ~$100M historically.
• PBGENE-3243 — m.3243 mitochondrial disease; development paused to prioritize the two leads.
• iECURE-OTC (ECUR-506) — partnered (iECURE) ARCUS gene-insertion program for neonatal OTC deficiency; the partner's OTC-HOPE study reported a "complete clinical response" in the first patient at the lowest dose (Jan 2025). DTIL earns milestones + mid-single-to-low-double-digit royalties.

Revenue model: DTIL has never commercialized a product and has never generated product revenue. Reported "revenue" is upfront + milestone recognition from out-licensing deals (Novartis, Imugene, TG Therapeutics, Caribou, iECURE). It is funded by collaboration payments, its IPO, PIPEs, convertibles and an at-the-market (ATM) facility. 68 full-time employees ("Precisioneers") at Dec 31 2025, 66 at Mar 31 2026 — a deliberately tiny, R&D-only footprint.

Lens 2 · Supply Chain (→ CDMO / manufacturing)

For a gene-editing company the "supply chain" is the manufacturing and delivery stack, not a goods chain:

• Upstream IP: core ARCUS meganuclease IP licensed from Duke University (12 issued US patents in the foundational family; Duke royalty + milestone obligations apply).
• Delivery: PBGENE-HBV uses lipid nanoparticle (LNP) delivery to the liver (the differentiator: first in vivo editor to use repeat LNP administration). PBGENE-DMD uses AAV delivery to muscle. Both rely on third-party CMOs/CDMOs for cGMP manufacture of mRNA/LNP and AAV — a named single-point dependency the 10-K flags as a risk ("manufacture of clinical trial material produced in compliance with cGMPs").
• Downstream: no commercial channel exists — pre-approval. For partnered assets (azer-cel, ECUR-506) the partner (Imugene, TG Therapeutics, iECURE) owns manufacturing and commercialization.

Chokepoint: AAV and LNP cGMP capacity is the genuine bottleneck for the whole field; DTIL outsources it and is small enough that it competes for slots against far larger sponsors. No vertical manufacturing moat.

Lens 3 · Competitive Advantages (moats)

The moat is the editing mechanism and its IP, not scale.

• Single-component editor. ARCUS is "the only single-component editor" — one engineered protein that both recognizes and cuts the target, vs CRISPR's two-component Cas9 + guide-RNA system. Small enough to fit an insertion template into the same AAV — enabling in vivo gene insertion, not just disruption. This is genuinely differentiated technology.
• cccDNA elimination. PBGENE-HBV is positioned as the only clinical program designed to eliminate the HBV reservoir rather than suppress it — a categorical, not incremental, differentiation if it converts to durable cure (see Lens 7 competitive read).
• IP estate: 46 issued US patents (as of FY24), 29 in vivo gene-editing patent families + 16 immunotherapy families, with key claims running to ~2042. Real, but narrower than the broad-CRISPR estates of Intellia/Beam/CRISPR Tx.
• Bargaining power: weak. A $189M micro-cap with one Phase-1 asset has little leverage over CDMOs, payers, or partners — evidenced by Novartis walking away (Lens 5) and DTIL waiving earned milestones to keep a Servier deal alive.

Verdict on moat: a real technical moat (mechanism + IP) sitting on top of an extremely weak commercial/financial moat. The science is differentiated; the company is fragile.

Lens 4 · Segments

No product or geographic segments exist — single reportable segment, no product revenue. The only meaningful "segmentation" is the source of recognized collaboration revenue, which is lumpy and shrinking as deals wind down:

Revenue halved YoY (FY25 vs FY24) purely on deal mechanics — it is not a demand signal and should not be modeled as recurring. With Novartis terminated, the collaboration-revenue base is now thin.

---

Phase B — Measure performance (clinical-stage swap)

Lens 5 → Pipeline by phase + latest result (swap: Earnings Result)

The pipeline table is the earnings result.

Provenance: trial design/timing; readout cadence.

The de-risking event — EASL Congress, 27–30 May 2026 (Barcelona), late-breaker. New liver-biopsy data from ELIMINATE-B, presented by Prof. Man-Fung Yuen (Univ. of Hong Kong):

• First-ever clinical evidence of a therapeutic eliminating and inactivating cccDNA in liver biopsies of treated CHB patients.
• 1-log (10×) reduction in cccDNA-derived transcripts; in the <1% of cccDNA remaining, indels permanently knocked out polymerase function.
• 100% of patients (n=15) experienced substantial HBsAg declines across all dose levels; 100% pgRNA loss in patients with detectable baseline pgRNA (pgRNA established as the cccDNA-elimination biomarker).
• "Clear therapeutic window," manageable safety across cohorts; 13 participants had completed 30+ administrations of repeat-dose LNP — the first in vivo editor to do so.

Financial "print" (Q1 2026): revenue $10.8M, total opex $19.9M, net loss $18.4M. FY2025: revenue $34.3M, R&D $54.2M, G&A $32.2M, total opex $86.4M, operating loss $52.1M, net loss from continuing ops $46.6M (FY24 showed income of $7.2M only because of the $50M Novartis upfront). Accumulated deficit $546.5M at Mar 31 2026 (from $528.2M at YE25).

Lens 6 → Calls / management cadence (sentiment trend)

No transcripts on the shelf (transcripts=0); read from filings + the Q1-2026 business update. The clear, consistent message across the FY25 10-K and Q1-26 10-Q is a deliberate narrowing to a single thesis: "we are now solely focused on" wholly-owned in vivo editing. Recurring emphasis: "potentially curative," "elimination of cccDNA," "first and only," cash runway "through 2028." The tone is that of a company that has shed its sprawl (cell therapy gone), concentrated its capital on two assets, and is leaning hard on the HBV mechanism story. The risk in the sentiment is its single-asset dependence — the narrative has no fallback if HBV durability disappoints. Note: no earnings-call transcripts in transcripts/; refresh should ingest the Q1-26 and next call.

Lens 7 → Catalyst calendar + mechanism comps (swap: Comps)

A. Catalyst calendar (what de-risks or kills, and when):

B. Mechanism comps — by target/approach, NOT by P/E (pre-revenue):

The comp story: every HBV competitor suppresses; only DTIL eliminates cccDNA — a categorical edge IF it converts to durable off-treatment cure. But bepirovirsen has already shown a 19% functional cure and is at the FDA now; DTIL must clear that bar from Phase 1. Against editing peers, DTIL at ~$189M is ~1/10th of Intellia and ~1/20th of Beam — a fraction that reflects single-asset risk, micro-cap fragility, and a 1-for-30 reverse-split history, but also leaves enormous room if HBV proves out. Multiples (EV/Sales, P/E) are n/a — pre-revenue; do not fabricate.

Lens 8 → Stock-Price Catalysts (be-early / event pattern)

DTIL trades on binary scientific events and dilution, full stop — there is no fundamental floor.

• 52-week range $3.53 → $8.82; stock roughly doubled off the low into the EASL readout and trades near the high (~$7.29).
• 27 May 2026 EASL biopsy data = the dominant up-catalyst; H.C. Wainwright (Patrick Trucchio) reiterated Buy, $60 PT on the data.
• Oct 31 2025 Novartis termination notice (effective Jan 30 2026) = a structural negative — lost the hemoglobinopathy option value and future milestones.
• Feb 13 2024 1-for-30 reverse split to cure a sub-$1 Nasdaq bid-price deficiency — the canonical micro-cap distress tell.
• Pattern: clinical data >> everything; financings and dilution are the recurring down-catalyst. This is a trade-the-readout name, not a compounder.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO: Michael Amoroso (President & CEO since Oct 2021) — a commercial/operations executive with cell-and-gene-therapy leadership background (research, clinical, regulatory, commercial). He is the operator who executed the pivot out of cell therapy and into concentrated in vivo editing.
• Scientific founders still in seat: Derek Jantz, Ph.D. (CSO) and Jeff Smith, Ph.D. (CRO) — co-inventors of ARCUS. Founder-scientist continuity is a genuine positive for a platform company; the people who built the technology still run the science.
• Skin in the game: thin. CEO directly owns ~0.93% (~243,392 shares + ~135,593 RSUs as of Feb 2026), ~$0.5M. Form-4 activity is routine RSU-vest tax sales, not signal selling. Low insider ownership is typical for a post-reverse-split micro-cap but means limited alignment leverage.
• Capital-allocation history: mixed-to-poor, but improving in discipline. The Novartis ($50M upfront, 2022) and TG/Imugene deals brought in cash and de-risked the cell-therapy exit. But the record also includes a destroyed share base (1-for-30 reverse split), a $528M accumulated deficit with nothing commercialized, waived milestone payments to preserve a Servier relationship, and an impaired/written-down note receivable to "Elo" ($5.4M loss). The current strategy — kill PBGENE-3243, partner out everything non-core, run lean (66 FTEs) — is the right discipline for a cash-constrained biotech, late.
• Archetype: professional-manager CEO + founder-scientists. Appropriate for a platform pivoting to focus; the question is execution speed against the cash clock.

Lens 10 · Forensic Red Flags (→ trial-design / going-concern / dilution)

Accounting/forensic:

• Revenue recognition is the area to watch — 100% of revenue is collaboration upfront/milestone recognition under ASC 606, inherently judgment-laden and lumpy (FY25 $34.3M almost entirely Novartis run-off + one Imugene milestone; Q1'26 $10.8M largely TG license-rights). Management "constrained all remaining variable consideration" on Imugene milestones given uncertainty — conservative, a good sign, not a flag.
• Going concern: financials prepared on a going-concern basis; management states cash "will be sufficient to fund operating expenses and capital expenditure requirements through 2028". The shelf language stops short of a substantial-doubt qualification — meaningful for a micro-cap — but the cushion is thin and explicitly assumption-dependent ("we could use our capital resources sooner").
• Note-receivable impairment (Elo): written down to $0, $5.4M net loss; Elo missed interest in Dec 2025 — a small but real capital-allocation blemish.
• SBC / dilution: ongoing ATM issuance (4.4M shares sold via ATM in 2025) plus the structural dilution of a perpetually-financing pre-revenue biotech. Share count already rose 24.7M → 25.8M in one quarter.

Regulatory findings (read from regulatory/regulatory-findings.md):

• SEC Litigation Releases: none found for Precision BioSciences since 2021-06-30.
• AAERs: none.
• Non-SEC (FTC/DOJ/FDA/etc.): web search returned no material enforcement actions, consent decrees, fines or penalties against the company. FDA interactions are favorable (Fast Track ×2, Orphan, RPD designations), not enforcement.
• 10-K Item 3 (Legal Proceedings): no material litigation disclosed beyond ordinary course.
• Conclusion: No material regulatory or legal findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-K Item 3 as of 2026-06-30. The real "regulatory risk" here is clinical-trial / FDA-approval risk, not enforcement risk.

---

Phase D — Project & stress-test

Lens 11 → rNPV + runway-to-catalyst (swap: Forward Projection)

No EPS line — pre-revenue. The right frame is risk-adjusted NPV of the lead asset against cash runway. Every input below is `` with arithmetic shown; treat as scenario scaffolding, not a model.

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

• Cash $125.8M (Mar 31 2026). Quarterly net burn ~$18M. → ~7 quarters ≈ runway into ~H2 2027–2028, consistent with management's "through 2028" claim given continued ATM top-ups. Runway reaches the value-inflection catalysts (HBV expansion-phase data + DMD initial data, both 2026) — the single most important fact for a clinical micro-cap. ✔

Lead-asset rNPV (PBGENE-HBV), illustrative:

• CHB 7MM therapeutics market ~$3.2B by 2034, >65% functional-cure-driven; ~5M prevalent / ~1.8M diagnosed across 7MM.
• Bull: peak sales $1.0B × PoS 20% (post-Ph1 mechanism-validated cure candidate) × ~0.5 cumulative discount ≈ $100M risk-adjusted contribution. Against an EV of ~$63M ($189M cap − $126M cash) the lead asset alone could justify the price if cure-durability reads out — before any DMD or PRV value.
• Bear: HBsAg decline does not convert to durable off-treatment functional cure → PoS collapses toward single digits → rNPV << current EV.
• DMD optionality + PRV: a positive DMD signal plus a saleable Priority Review Voucher (~$100M comp) is unmodeled upside.

No forecast.ts logged (--watchlist rule + pre-revenue → no EPS line to score; the right Brier forecast would be a binary on HBV durable functional cure, to be logged at the next data readout). No forecast.ts create run.

Lens 12 · Bull vs Bear

Bull. ARCUS is genuinely differentiated single-component editing, and PBGENE-HBV just produced the first clinical evidence anywhere of cccDNA elimination in human liver biopsies — 100% HBsAg decline, 100% pgRNA loss, manageable safety over 30+ repeat doses. If "elimination" converts to a durable, off-treatment functional cure, DTIL owns a categorically better mechanism than the suppressive siRNA/ASO field — in a market heading to $3B+ that will be >65% functional-cure drugs. At a ~$63M enterprise value with cash into 2028 and two 2026 catalysts (HBV expansion, DMD initial data), the asymmetry is extreme: analysts carry a Strong-Buy with a $60 top target (~8× current). A platform validation could also reopen partnering at far better terms than the Novartis exit.

Bear. This is a single-asset, post-reverse-split micro-cap that bleeds ~$18M/quarter, has a $546M accumulated deficit, and just lost Novartis. HBsAg decline and a 1-log cccDNA cut are not a functional cure — and GSK's bepirovirsen is already at the FDA with a demonstrated 19% functional cure and a Q3-2026 decision. DTIL must beat an approved bar from Phase 1, with n=15 and no expansion data yet. AAV/LNP delivery safety overhangs the whole field after Sarepta's Elevidys deaths and FDA distribution halt. Cash "through 2028" leans on continued ATM dilution; the share count is already climbing.

Pre-mortem (18 months out, thesis broke): PBGENE-HBV's HBsAg declines proved transient off-treatment / failed to clear a functional-cure bar in expansion; an LNP/AAV safety event (or field contagion from a competitor) spooked the FDA; bepirovirsen approved and anchored the standard of care; DTIL diluted heavily at a depressed price to stay alive. The stock round-trips to the $3s.

Are multiples too high? No multiple exists (pre-revenue). The EV is low in absolute terms — the question is binary asset value, not over-valuation.

Contrarian view (what the market refuses to see): the market is treating DTIL as a generic broke micro-cap (1/20th of Beam) while it holds the only clinical asset attacking the HBV reservoir at its root — the biopsy data quietly reframed it from "story stock" to "mechanism-validated cure candidate," and most allocators haven't repriced the optionality.

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

Dismantling the bull case:

• Revenue is a mirage — 100% lumpy collaboration recognition, now shrinking with Novartis gone; there is no business, only a science project burning $70M+/yr.
• The "cure" claim outruns the data. On-treatment HBsAg decline ≠ durable off-treatment functional cure. The bar that matters (sustained HBsAg loss 24+ weeks after stopping) has not been shown, n=15, no expansion cohort. cccDNA "elimination" is a 1-log reduction with <1% residual — impressive mechanistically, unproven clinically.
• A large, funded, ahead competitor: GSK/Ionis bepirovirsen — Breakthrough Therapy, priority review, 19% cure, decision Q3-2026. DTIL is years behind a soon-to-be-approved drug and must out-cure it from Phase 1.
• Delivery safety is a field-level landmine — Sarepta's AAV DMD deaths, FDA distribution suspension and boxed warning poison sentiment for all in vivo genetic medicines, including DTIL's AAV-delivered DMD program.
• Capital allocation: 1-for-30 reverse split, $546M cumulative deficit, waived milestones, an impaired related note receivable, perpetual ATM dilution. Thin insider ownership (~0.93%).
• What must hold for the price: that HBV durability reads out positive and the company funds to that proof without a punitive raise. Knock HBV PoS from ~20% to ~5% and the rNPV falls well below today's ~$63M EV.
• Most-dangerous-competitor bulls underestimate: not another editor — it's bepirovirsen as "good-enough" functional cure arriving first and defining the market before DTIL exits Phase 1.
• Single scenario that permanently impairs: an LNP-delivery serious-safety event in ELIMINATE-B — it would simultaneously kill the lead asset and tar the platform.

Lens 14 · Management Questions (ordered by information value)

1. ELIMINATE-B showed HBsAg decline on-treatment — what is your evidence, and timeline, for durable functional cure measured ≥24 weeks off all treatment, the bar bepirovirsen is being judged on?
2. What specific therapeutic-index / safety threshold triggers the expansion-phase go decision, and when in 2026 do you expect it?
3. With ~$126M and ~$18M/quarter burn, how far does cash actually reach without further ATM issuance, and what dilution is assumed in the "through 2028" statement?
4. How does PBGENE-HBV's profile position against bepirovirsen if GSK is approved in late 2026 — combination, monotherapy, or salvage?
5. After the Elevidys AAV deaths and FDA halt, what delivery-safety de-risking distinguishes your AAV PBGENE-DMD program?
6. What drove the Novartis termination, and what does it imply for the value of the reverted sickle-cell/beta-thal options?
7. What are the minimum efficacy/safety criteria for the year-end-2026 PBGENE-DMD initial readout to be considered a success?
8. Do you intend to monetize the PBGENE-DMD Priority Review Voucher, and what does that do to runway?
9. What is your CDMO capacity and single-source exposure for LNP and AAV at clinical and (eventual) commercial scale?
10. Which non-core assets remain to partner or out-license, and on what timeline, to extend runway non-dilutively?
11. What is the repeat-dose ceiling for LNP — immunogenicity or tolerability limits across the 30+ administrations to date?
12. Beyond HBV and DMD, what is the next wholly-owned ARCUS indication, and when does it enter the clinic?
13. How do you think about insider ownership and alignment given current levels post-reverse-split?
14. What IP claims most directly protect the cccDNA-elimination approach specifically, and through what year?
15. Under what circumstances would you consider a whole-company sale or reverse merger rather than continuing to finance independently?

---