Cellectis

Phase A — Understand the business

Lens 1 · Company Overview

Cellectis is a 25-year-old, Paris-based, clinical-stage gene-editing company building allogeneic ("off-the-shelf") CAR-T cell therapies for blood cancers, plus a platform-licensing business. The core thesis: instead of re-engineering each patient's own T-cells (autologous — the approved Kymriah/Yescarta/Breyanzi model), Cellectis edits T-cells from healthy donors using its proprietary TALEN nuclease + PulseAgile electroporation, knocking out the T-cell receptor (to avoid graft-versus-host disease) and CD52 (to survive lymphodepletion), to mass-produce cryopreserved doses — "potentially hundreds of doses per batch". It manufactures in-house in Paris and Raleigh, NC.

Two business lines: (1) self-owned UCART pipeline (the value driver — lasme-cel, eti-cel) and (2) IP/platform licensing that funds it — collaborations with AstraZeneca, Allogene, Servier and Iovance "validated our technology platform". There is no product revenue: "To date, we have not generated any revenues from therapeutic product sales". 2025 "revenue" of $72.9M is ~99% AstraZeneca collaboration accounting.

• Customers/partners: AstraZeneca (AZ Ireland / AZ JRCA — the dominant counterparty), Allogene, Servier, Iovance, Genzyme (alemtuzumab supply for CLLS52).
• Suppliers: Life Technologies/Thermo Fisher and University of Minnesota (in-licensed foundational IP — "significant annual payment and royalty expenses" ); CDMOs for some manufacturing; Genzyme for alemtuzumab.
• Competitors: Allogene (partner and rival), Caribou Biosciences, Precision BioSciences, Fate Therapeutics (iPSC), plus the approved autologous incumbents (Novartis, Gilead/Kite, BMS) it must beat on data.
• Payment structure: milestone-and-royalty (lumpy, non-recurring), not recurring revenue. Concentration is extreme — see Lens 13.
• Employees: 229 (185 in R&D), 151 in France / 78 in US.
• Dual-listed: Euronext Growth Paris (since 2007) + Nasdaq ADS (since 2015).

Lens 2 · Supply Chain (→ manufacturing / CDMO, per +clinical overlay)

Map: in-licensed IP (Thermo/LTC, U. Minnesota) → Cellectis gene-editing platform (TALEN + PulseAgile) → in-house cGMP manufacturing (Paris + Raleigh, NC) → clinical trial sites (MD Anderson et al.) → (future) patients/payers. Named stakeholders along the chain:

• Upstream IP (chokepoint): Cellectis does not own all its foundational tools outright — it licenses key nuclease/cloning patents in from Life Technologies Corp (Thermo Fisher) under 2014 agreements (Halle Patent Therapeutic/Research Licenses, GeneArt/Seamless Cloning) and from the University of Minnesota. This is the single most dangerous link — see Lens 10 (LTC purported to terminate these in April 2026).
• Lymphodepletion input: alemtuzumab supplied by Genzyme (Sanofi), used as Cellectis's own coded product CLLS52 in the BALLI-01/NATHALI-01 protocols — a single-source biologic dependency baked into the regimen.
• Manufacturing (a genuine asset, vertically integrated): in-house facilities in Paris and Raleigh, North Carolina (Site Head Steven Doares, ex-Biogen). Cellectis pitches in-house manufacturing as a moat — its "Process 2" (in-house) lasme-cel showed a higher response rate (67% vs 50%) than the external-CDMO "Process 1". Vertical manufacturing for allogeneic CAR-T is genuinely differentiated.
• Downstream: clinical sites (MD Anderson dosed first patient Nov 2019); eventual commercial path requires building sales/distribution from scratch.

Chokepoint verdict: the binding constraint is upstream IP licensing, not physical inputs — a feature unusual for a "manufacturing" story.

Lens 3 · Competitive Advantages (moats)

• IP estate (partial moat): Cellectis is the originator of therapeutic TALEN; it licenses TALEN out to Thermo (research/applied markets) and to Iovance (TIL editing). Patent families cover composition-of-matter, manufacture and use of UCART22 (WO2018…) and UCART20x22 (WO2022…). But the moat is leaky at the root — its own therapeutic freedom-to-operate depends on in-licensed third-party patents now under attack (Lens 10). Owning the brand "TALEN®" ≠ owning unencumbered FTO.
• Manufacturing know-how (real, durable): 25 years of genome-engineering + in-house cGMP allogeneic production at scale is hard to replicate; the Process-2 potency uplift is evidence it matters.
• Validation-by-partner (signaling moat): AstraZeneca took a 31% equity stake and a 10-product collaboration; that is a strong third-party vote that the platform works. Servier, Allogene and Iovance deals reinforce it.
• Bargaining power: weak. As a sub-$250M-cap, pre-revenue, cash-consuming biotech facing a 31%-owner partner (AZ) with board control and a foundational-IP licensor (Thermo) suing it, Cellectis is the price-taker in nearly every relationship. Patients/payers don't yet exist as a counterparty (no approved product). The moat is scientific/manufacturing, not commercial or financial.
• The structural challenge (Lens 12/13): the entire allogeneic-CAR-T category's moat question is durability of response — off-the-shelf cells historically persist less than autologous, so the moat only becomes real if the data shows competitive durability. Cellectis's median OS of 14.8 months in MRD-negative responders is the proof-point to watch.

Lens 4 · Segments

Cellectis reports as effectively one segment (immuno-oncology cell therapy R&D); the former Plants segment (Calyxt) was deconsolidated in 2023. Revenue split (no geographic segmentation disclosed; collaboration revenue follows partner accounting, mostly US/UK/EU counterparties):

[All figures research-layer.] The trend is accelerating reported revenue — but this is an artifact of AstraZeneca collaboration-accounting (recognition of Research-Plan performance obligations under the AZ JRCA), not commercial traction. It will be lumpy and can reverse (Q1 2026 revenue fell to $5.8M from $10.7M a year earlier ). Treat the "segment" as a single binary asset story (lasme-cel), with collaboration cash as runway-extension, not a P&L.

---

Phase B — Measure performance (+clinical overlay: Lens 5 → Pipeline-by-phase; Lens 7 → Catalyst calendar + mechanism comps; Lens 8 carried as funding/data events)

Lens 5 · Pipeline by Phase (swapped in for "Earnings Result")

The asset table is the company. Cellectis is pre-revenue; what matters is each program's phase, next readout, and probability of success.

[Pipeline status: research-layer 20-F for structure; web (Cellectis PRs, EHA 2026, Q1-2026 release) for the 2026 catalyst dates and RMAT.]

Lead-asset clinical data — lasme-cel BALLI-01 (the crux):

• Oct 2025 (per 20-F): ORR 68% with Process-2 (n=22); 83% at RP2D (n=12); 100% in target Phase-2 population (n=9); median OS 14.8 months in MRD-negative CR/CRi responders; in target Ph2 pop, 100% became transplant-eligible, 78% proceeded to transplant; among 11 patients who failed all 3 standard targeted therapies (inotuzumab, blinatumomab, CD19 CAR-T), 8 responded / 7 reached MRD-negative.
• Final Phase 1 at EHA 2026 (June 2026): 100% ORR (7/7) in the target cohort, CR/CRi 57% (4/7), 75% of responders MRD-negative; safety manageable — grade ≥3 CRS 4%, ICANS 4%, IEC-HS 2%, all resolved. Confirmed BALLI-01 pivotal Phase 2 open; first interim Q4 2026.
• eti-cel NATHALI-01 (Dec 2025): ORR 88%, CR 63% (n=8) at current dose.

Financial snapshot (context, not the thesis):

The 2025 net loss widened to $(67.6)M despite higher revenue — driven by a $(34.9)M net financial loss (mostly FX losses on USD cash as USD weakened vs EUR, plus a $(14.7)M mark on EIB warrants). Underlying operating loss actually improved to $(33.1)M as collaboration revenue offset opex. R&D at $93.5M is ~128% of revenue — the spend that the binary readout has to justify.

Lens 6 · Earnings Calls / management messaging (sentiment trend)

No transcripts on the shelf (transcripts=0); read from PRs and the 20-F. Management's narrative arc has clearly shifted from "platform/IP-licensing company hunting validation" (2022-23) to "late-stage allogeneic CAR-T company with a pivotal trial" — the Jan-2026 strategy PR explicitly frames 2025 as "a transformational year… transitioned to a late-stage development allogeneic CAR-T company". Recurring themes: (1) "first allogeneic CAR-T in a pivotal trial for r/r B-ALL" (the lead claim, reinforced by June-2026 RMAT ); (2) AstraZeneca partnership as validation + funding; (3) in-house manufacturing as differentiation; (4) new optionality (TALEM epigenetic platform). What they've stopped emphasizing: the Servier relationship (now adversarial — arbitration terminated UCART19 Dec 2025 ) and the broader autologous-vs-allogeneic debate. Tone is confident/promotional, as is typical for a small-cap binary biotech approaching a catalyst — discount accordingly.

Lens 7 · Catalyst Calendar + Mechanism Comps (swapped in for P/E comps)

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

Mechanism comps (allogeneic CAR-T peers, by approach — NOT by P/E, which is meaningless pre-revenue):

[Market caps: web, MacroTrends/StockAnalysis/WallStreetZen, ~June 2026. Traditional P/E and EV/EBIT are n/a and not meaningful: every name is pre-revenue and loss-making. The honest comp is catalyst-and-cash, not multiples.] Read-through: Cellectis trades below Allogene despite arguably more-advanced lead-asset data and RMAT — partly the Thermo IP overhang, partly micro-cap neglect. Caribou's 17.1-month PFS print is the benchmark off-the-shelf bulls cite that the category can match autologous.

Lens 8 · Stock-Price Catalysts (5-yr pattern → funding/data events)

For a pre-revenue biotech the tape reacts to clinical data, regulatory designations, partner deals, and financings — not earnings. The 52-week range alone ($1.35 → $5.48, now ~$3.11 ) shows ~4x peak-to-trough volatility typical of binary-catalyst names. Pattern of what moves CLLS:

• Partner/equity events (up): AstraZeneca's Nov-2023 $80M initial + May-2024 $140M preferred investments (total $220M) were transformational re-ratings.
• Regulatory designations (up): FDA RMAT (June 2026), fast-track/ODD/RPDD grants.
• Clinical data (both ways): EHA/ASH data prints; the Oct-2021 industry-wide FDA clinical hold on Allogene's allo-CAR-T (chromosomal-abnormality scare) hit the whole category.
• Litigation (down): Servier arbitration; the April-2026 Thermo/LTC termination+arbitration is a fresh overhang.
• Macro/FX: as a EUR-functional company holding USD cash, FX swings move reported net income materially ($(34.9)M financial loss in 2025) but are non-operating noise. The market reacts most violently to lead-asset data — making the Q4-2026 BALLI-01 interim the dominant scheduled catalyst.

---

Phase C — Judge people & books

Lens 9 · Management

• André Choulika, Ph.D. (CEO & Co-founder, age 61): co-invented nuclease-based genome editing as a Harvard/Boston Children's post-doc; founded Cellectis in 1999 and has run it ever since (Chevalier de la Légion d'Honneur; board of Institut Pasteur). Founder-scientist archetype — deep credibility on the science, 25-year tenure, ~3.61% beneficial ownership (2.68M shares). The risk of the archetype: founder attachment to a platform through a quarter-century of cash burn without a commercial product.
• David Sourdive, Ph.D. (Co-founder, Deputy CEO, EVP Manufacturing): École Polytechnique, ex-French MoD biolab; ~2.41% ownership; runs the CMC/manufacturing differentiator.
• Arthur Stril (CFO & CBO, age 37): École Normale Supérieure/Cambridge, French Corps des Mines, ex-EU DG-Competition (pharma M&A); CFO since Jan 2025 — the dealmaker who structures the partnerships that fund the company.
• Adrian Kilcoyne, M.D. (CMO, since Aug 2024): ex-Celularity (allo CAR-T/NK), ex-AstraZeneca/Celgene — a recent, relevant late-stage clinical hire as the company goes pivotal.
• Board / skin in the game: Chairman Jean-Pierre Garnier (ex-CEO GSK; built/sold Actelion to J&J for $30B) is genuine pharma pedigree. AstraZeneca controls two board seats (Marc Dunoyer — CEO Alexion/ex-AZ CFO; Tyrell Rivers — AZ Corporate Development) as long as it holds ≥40%; one seat at ≥20%. Bpifrance's Laurent Arthaud sits on the board (state-backed holder).
• Capital allocation: has funded a quarter-century of losses via equity, grants, French R&D tax credits, EIB debt and partner payments. Smart moves: monetizing the platform (Thermo/Iovance licensing-out, the AZ mega-deal), non-dilutive EIB/Bpifrance funding, and deconsolidating Calyxt. Weak spots: serial dilution (weighted basic shares 57M→90M→100M, 2023→2025 ); ROE/ROIC structurally negative (it's pre-revenue). Total exec+director cash comp $5.6M FY2025 — reasonable for the size.
• Red flags (governance): founder-CEO is also a long-tenured insider; CEO Choulika and General Counsel Terrier are domestic partners (disclosed); AZ's board control + 31% stake creates an asymmetric-influence dynamic (could be a takeout or an overhang). No related-party self-dealing flagged.

Lens 10 · Forensic Red Flags

Acting as a forensic analyst on IFRS statements:

• Revenue recognition is the #1 watch-item. ~99% of "revenue" is AstraZeneca collaboration accounting — recognition of performance obligations under "Research Plans" and milestone/upfront allocation, not cash sales. Deferred-revenue/contract-liability swings are huge and reverse: change in deferred revenue was +$59.1M (2023), +$5.7M (2024), −$28.9M (2025). Reported revenue can look like growth one year and collapse the next purely on obligation timing — Q1-2026 revenue already fell ~46% YoY. Do not extrapolate the "revenue" line.
• Financial result distorts net income. The 2025 net loss is dominated by $(34.9)M non-operating financial loss (FX on USD cash + EIB-warrant fair-value marks) — a EUR-functional reporting artifact, not operating deterioration. Conversely 2024's net financial gain of +$22.8M flattered that year. Look at operating loss ($(33.1)M, improving) for the real trend.
• Cash vs. earnings divergence: 2025 operating cash flow $(39.4)M vs. operating loss $(33.1)M — broadly consistent; the swing from +$23.0M (2024, boosted by $57M of the AZ subsequent-investment proceeds allocated to operating) to $(39.4)M (2025) is partner-cash timing, not a quality-of-earnings red flag once understood.
• SBC: $6.1M stock-based comp in 2025 (up from $3.2M in 2024 on higher fair value) — modest relative to losses; not flattering a non-GAAP story (there is no non-GAAP profit to flatter).
• Dilution / warrant overhang: EIB warrants (Tranches A/B/C, ~4.3M shares, PIK interest at 6-8% capitalizing the €40M loan principal to ~$72.1M owed) + serial equity raises + AZ convertible preferred = structural dilution machinery.
• Going concern — NUANCE (do not misread): Cellectis's own statements are prepared on a going-concern basis; it explicitly states $61.5M cash + $144.8M deposits "will be sufficient to fund its operations into the second half of 2027". The "substantial doubt about going concern" language in the 20-F refers to Cibus (the merged former-Calyxt entity whose Roseville lease Cellectis still guarantees — a contingent ~$19.9M exposure), not Cellectis. Cellectis itself is not a going-concern risk on current guidance.

Regulatory findings (required sub-section):

• SEC Litigation Releases: None. "No LR found for this company in the search period".
• SEC AAERs: None. "No AAER found for this company".
• Non-SEC enforcement (web search): No FTC/DOJ/FDA/SEC enforcement actions found against Cellectis. The material legal exposures are commercial arbitrations, not regulator enforcement:
• Life Technologies Corp / Thermo Fisher (NEW, material): On April 20, 2026, LTC purported to terminate the 2014 license agreements granting Cellectis non-exclusive rights under foundational patents (Halle Patent Therapeutic/Research, GeneArt/Seamless Cloning) and commenced AAA arbitration, alleging Cellectis underpaid sublicense royalties. Cellectis calls the termination "invalid" and the claims "without merit." This is the most dangerous active item — it touches in-licensed IP underpinning the therapeutic platform.
• Servier arbitration: On Dec 15, 2025, an arbitral tribunal (Paris) ordered partial termination of the Servier License Agreement w.r.t. UCART19 V1 / ALLO-501, requiring good-faith negotiation of a possible direct Allogene license. Net effect: lost control of the oldest CD19 program (already deprioritized).
• Item 3 (Legal Proceedings): The 20-F (report date 2025-12-31, filed before the April-2026 LTC action) discloses the Servier arbitration and ordinary-course matters; product-liability insurance maintained. Net: no accounting/securities-fraud history; the risk is contractual/IP litigation, and the Thermo/LTC arbitration is fresh and foundational.

---

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

Lens 11 · rNPV + Runway-to-Catalyst (swapped in for EPS projection)

No EPS model — Cellectis won't have product revenue before ~2029+ (BLA targeted 2028 ). The two questions that matter:

(1) Does cash reach the next value-inflection catalyst? — YES, comfortably. $188M cash+deposits at Q1-2026, guided into Q4 2027; operating burn ~$33-40M/yr; modest debt (€40M EIB PIK + ~$4M PGE). The Q4-2026 BALLI-01 Phase 2 interim lands well inside the runway with ~$120M+ still on the balance sheet — i.e. Cellectis can deliver its single most important catalyst without forced dilution. That is the strongest single fact in the bull case.

(2) What is the lead asset worth? — rNPV sketch (all, illustrative, NOT a forecast):

• lasme-cel addressable: r/r B-ALL 3L+ is small — ~6,100 US ALL cases/yr, only a fraction relapsed/refractory and eligible. Call it a few-thousand-patient niche; a first allogeneic entrant with RMAT + orphan/RPDD (priority-review-voucher optionality) could plausibly support peak sales of a few hundred $M if approved and durable.
• Apply a B-ALL-niche PoS for a pivotal-Phase-2-stage RMAT asset of ~30-40% and a high biotech discount rate (~12-15%): a single-asset rNPV in the few-hundred-$M range is defensible — i.e. the ~$233M market cap is roughly pricing lasme-cel near risk-adjusted fair value with AstraZeneca's 10-product collaboration, eti-cel, TALEM, and the Allogene/Servier royalty stubs as unpriced optionality. The market is paying little for the platform beyond the lead asset.
• Brier forecast (logged conceptually, not via forecast.ts — skipped per --watchlist rules): "lasme-cel BALLI-01 pivotal Phase 2 first interim (Q4 2026) reports an efficacy result the company characterizes as positive / supportive of continuation — p ≈ 0.60".

Lens 12 · Bull vs Bear

Bull case. Cellectis is the first allogeneic CAR-T in a pivotal trial for r/r B-ALL, now FDA RMAT-designated (June 2026) — a credible shot at being the first approved off-the-shelf CAR-T, in a niche with orphan + rare-pediatric-disease designations (PRV optionality). The Phase-1 target-cohort data is striking: 100% ORR, 56% CR/CRi, ~80% MRD-negative, 100% bridged-to-transplant. It is fully funded through the make-or-break Q4-2026 readout ($188M, runway Q4 2027) — so the catalyst hits without dilution. AstraZeneca owns 31%, holds board seats, and is co-developing up to 10 products ($80-253M milestones each + royalties) — a strategic anchor that is also a plausible acquirer. In-house manufacturing is a real differentiator (Process-2 potency uplift). Free options on top: eti-cel (88% ORR early), the TALEM epigenetic platform (>90% silencing), Servier/Allogene royalty stubs. At ~$233M cap with ~$125M EV, the downside is partly cash-backed and the platform is nearly free. Average analyst target ~$7.25-7.60 vs ~$3.11 implies ~2x+ upside if the thesis holds.

Bear case (permanent-impairment risks).

1. The Q4-2026 Phase 2 interim disappoints or allogeneic durability fails. The category's structural Achilles heel is that off-the-shelf cells persist less than autologous; a Phase-1 n=7-9 "100% ORR" can regress hard at n=40. A bad interim is a 50%+ down-day and resets the whole thesis. This is the single most likely thesis-killer.
2. The Thermo/LTC IP arbitration (April 2026) impairs foundational freedom-to-operate. If LTC's termination of the in-licensed patents is upheld, Cellectis's therapeutic platform could face a royalty re-pricing or FTO challenge at the root — an existential overhang no amount of clinical data fixes.
3. Dilution beyond 2027 + micro-cap neglect. Even with a good readout, BLA→launch (2028+) needs more capital; a sub-$250M, serially diluting, EUR-reporting Nasdaq micro-cap with 4 analysts is structurally illiquid and re-rating-resistant.

Pre-mortem (18 months out, thesis broke): The BALLI-01 Phase-2 interim showed CR/durability below the autologous bar; the Street concluded allogeneic-CD22 doesn't persist; the Thermo arbitration forced a royalty give-up; AstraZeneca slowed program nominations; the stock round-tripped to its $1.35 52-week low and Cellectis raised dilutive equity into weakness.

Multiples too high? No — there are no meaningful multiples (pre-revenue). The question is whether ~$125M EV over-/under-prices a single-asset rNPV plus optionality; on balance it looks roughly fair-to-cheap if the readout is even modestly positive, expensive if it isn't.

Contrarian view (what the market refuses to see): The market lumps Cellectis with the "allogeneic CAR-T is dead / industry moved on" narrative and ignores that (a) it's the most regulatory-advanced allo asset in B-ALL with RMAT, (b) it's fully funded through its catalyst unlike many peers, and (c) it has a 31%-owner Big-Pharma partner who could simply buy it. The bear consensus may be over-discounting a name that is one good Q4-2026 print from a violent re-rate — or correctly pricing a category in secular decline. It is genuinely a coin-flip with a funded balance sheet, which is exactly why it's a watch, not a hold.

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

Dismantling the bull case:

• Revenue is an accounting mirage. ~99% is AstraZeneca obligation-recognition; it fell ~46% YoY in Q1-2026. Anyone valuing CLLS on "75% revenue growth" is being fooled by collaboration accounting. There is zero product revenue and won't be before ~2029.
• Concentration is total. The entire commercial counterparty base is essentially one partner, AstraZeneca, which also owns 31% and controls board seats — Cellectis is a price-taker with negotiating leverage of roughly zero. If AZ slows or walks, the funding-and-validation story evaporates.
• The moat is leased, not owned. Cellectis sues and is sued over its own foundational tools — the April-2026 Thermo/LTC arbitration attacks the in-licensed patents underpinning the therapeutic platform. A "platform company" that doesn't fully own its platform's FTO is structurally fragile.
• Most dangerous competitor bulls underestimate: not Allogene — Caribou (CRISPR chRDNA, 17.1-mo PFS print beating Breyanzi/Yescarta ) and, more broadly, the approved autologous incumbents whose data Cellectis must beat on durability, where allogeneic historically loses.
• Capital-allocation reality: 25 years, ~$1B+ cumulative burn, serial dilution (basic shares 57M→100M in two years ), still pre-product. The Servier relationship already soured into arbitration.
• What must hold for the price: that the Q4-2026 Phase-2 interim confirms durable, competitive efficacy at n=40 and the Thermo IP fight is contained and AZ keeps funding. If lead-asset efficacy/durability disappoints by 20-30%, the rNPV — and most of the market cap above cash — is impaired, and the stock revisits $1-2.
• Single permanent-impairment scenario (plausible): Phase-2 durability washes out + Thermo wins the arbitration → platform FTO compromised and lead asset de-risked downward simultaneously. Not the base case, but far from tail.

Lens 14 · Management Questions (ordered by information value)

1. At the Q4-2026 BALLI-01 Phase-2 interim, what durability/MRD-negativity and CR threshold at n=40 would you consider a clear success vs. a signal to stop — and how does it compare to the autologous CD19/CD22 benchmark?
2. On the April-2026 Life Technologies/Thermo arbitration — which specific patents are at issue, what is your therapeutic freedom-to-operate if LTC's termination is upheld, and what is the realistic worst-case royalty/financial exposure?
3. What is the honest probability and timeline that lasme-cel reaches BLA submission in 2028, and what could push it right?
4. How durable are the lasme-cel responses beyond 6-12 months — what does the latest persistence/OS data say, given allogeneic CAR-T's historical durability gap vs autologous?
5. Does the $188M / Q4-2027 runway carry you through BLA submission, or will you need to raise — and would you prefer dilutive equity, AstraZeneca milestones, or a partnering deal?
6. What would trigger AstraZeneca to exercise more of the 10 collaboration options (or to acquire Cellectis outright), and what is the current state of program nominations under the JRCA?
7. Is the in-house Raleigh manufacturing scalable and cost-competitive at commercial volumes, and what is your projected cost-of-goods per dose vs autologous CAR-T?
8. How real is the TALEM epigenetic platform as a value driver vs a research curiosity — do you intend to partner it, and on what timeline?
9. After the Servier termination, what are the expected economics from the Allogene-run cema-cel (ALPHA3) and ALLO-316 programs, and how do you account for them?
10. What is the regulatory path and reimbursement strategy if lasme-cel is approved into a small B-ALL niche — is the addressable population large enough to support a standalone commercial build?
11. How do you intend to manage dilution through commercialization given the micro-cap, serially diluting history?
12. What is your competitive read on Caribou/Allogene/Fate, and where does TALEN-edited allogeneic CD22 win or lose vs CRISPR/iPSC approaches?
13. With AstraZeneca controlling 31% and two board seats, how do you protect minority shareholders' interests in a potential take-private?
14. What are the eti-cel (NATHALI-01) next steps after the Q4-2026 full Phase-1 dataset — does it advance to pivotal, and on what funding?
15. What is the single risk on the horizon you worry about most that the Street isn't pricing?

---