CUE / cue-221-csu — CUE-221 for chronic spontaneous urticaria (long-lasting hives with no identifiable trigger)

Program analysis · bpiq_drug_id 20405 · prepared 2026-08 · USD · framework v5.10.3 · Coverage: CLEARED

Company context, financials, ownership, options and the full pipeline: ../company.md.

How to read this. Written for a reader who has not studied pharmacology or finance. Terms specific to this program are in the Glossary below; general terms are defined once in framework/04-glossary.md. Tags: [VERIFIED — source] / [UNVERIFIED] / [WEB ESTIMATE — source, date].

Glossary

Executive summary

• What it is (one sentence): CUE-221 is a laboratory-made antibody that mops up IgE — the antibody that drives allergy — and, unusually, also switches off the B cells that make new IgE, as a treatment for long-lasting hives that antihistamines do not control.
• The event and when (as disclosed): Topline results from a Phase 2 dose-ranging study in China run by Genesis Life Sciences, comparing CUE-221 against both placebo and an active comparator. The company’s own most recent wording, on 2026-08-14, is “Phase 2 data readout in Chronic Spontaneous Urticaria anticipated by the end of the third quarter of 2026” [VERIFIED — 8-K Exhibit 99.1, 2026-08-14]. BPIQ stores this as catalyst_date 2026-09-30 with catalyst_date_text “Q3 2026”; that is a quarter-end placeholder and not a disclosed day (rule 23).
• The main reason it could work: The mechanism is about as de-risked as an unapproved mechanism gets. Omalizumab, a weaker anti-IgE, is already approved for exactly this disease. CUE-221 binds IgE far more tightly and, in the one peer-reviewed paper on it, achieved rapid and durable free-IgE suppression lasting more than twelve weeks from a single dose, with symptom relief, in first-in-human chronic spontaneous urticaria patients [VERIFIED — J Clin Invest 2022, doi:10.1172/JCI157765].
• The main risk: Beating placebo is the easy half. The trial also carries an active comparator, and the field has already watched a better-binding anti-IgE — ligelizumab — beat omalizumab in Phase 2 and fail to in Phase 3. A result that clears placebo but merely ties the comparator is a technically positive readout that could still take the shares down. Layered on top: the trial is not registered on ClinicalTrials.gov, so its size, design and endpoints cannot be checked before the data lands, and Cue itself flags “potential challenges associated with clinical trials conducted in China and the company’s access to, and acceptability of, the data therefrom” as a risk factor [VERIFIED — 8-K Exhibit 99.1, 2026-08-14].
• What it means for the stock: This is the whole company. It is the only pipeline row with a pending catalyst, and the shares are 5.2× their own 52-week low on the strength of the in-licensing rather than on any data (../company.md C.2, C.4).

0. Program-tier coverage — CLEARED

Company-tier coverage is in ../company.md C.0 (CLEARED).

Why this is CLEARED and not PROVISIONAL. No mandatory row reads NOT CALLED. Open Targets reads BLOCKED with its verbatim error recorded and the dependent claim named, which 02-connectors.md permits. The unregistered pivotal trial is a data gap, not a coverage gap: the connector was called and answered honestly that the registry has no such record.

A. Scientific & clinical assessment

A.1 Mechanism of action and scientific rationale

What the drug is. CUE-221 is a humanised monoclonal antibody of the IgG1 class — a single, laboratory-made antibody protein, mass-produced from one cell line, engineered so the human immune system treats it as its own [VERIFIED — J Clin Invest 2022, doi:10.1172/JCI157765]. It binds to a specific stretch of the IgE molecule called the Cε3 domain, with a binding strength reported at 5.9 × 10⁻¹¹ (picomolar) [WEB ESTIMATE — Synapse/PatSnap compound record, read 2026-08-19].

How the disease works. In chronic spontaneous urticaria, IgE antibodies — a large share of them pointed at the patient’s own tissues rather than at anything external — load onto FcεRI receptors on mast cells sitting in the skin. When two of those loaded IgE molecules are cross-linked, the mast cell dumps its granules of histamine and other irritants into the surrounding tissue. That is a hive. Do it repeatedly for more than six weeks with no identifiable trigger and it is chronic spontaneous urticaria.

How the drug is meant to work. Two mechanisms, and the second is the entire commercial argument for this molecule:

1. Neutralising free IgE. CUE-221 grabs IgE out of the circulation before it can reach FcεRI, so fewer mast cells end up armed. This is what omalizumab does, and CUE-221 does it harder: the peer-reviewed characterisation reports it neutralises IgE and prevents FcεRI-mediated basophil activation and degranulation with “superior IgE-neutralizing activity to that of omalizumab” [VERIFIED — J Clin Invest 2022]. Third-party material puts the potency gap at roughly eightfold [WEB ESTIMATE — Synapse/PatSnap, read 2026-08-19; this figure was not confirmed in the primary paper and is reported as the third-party claim it is].
2. Switching off new IgE production. This is the differentiated half. CD23 is a second, lower-affinity IgE receptor found mainly on B cells, and cross-linking it tells the B cell to make less IgE. Every previous anti-IgE antibody had to avoid CD23 or was simply inert toward it. The paper’s central finding is that CUE-221 in free form “bound abundantly to CD23-occupied IgE and, in oligomeric mAb-IgE complex forms, freely engaged CD23, while ligelizumab reacted limitedly and omalizumab stayed inert toward CD23” — and that this translated into CUE-221 outperforming both in CD23-mediated downregulation of IgE production [VERIFIED — J Clin Invest 2022].

How well is the target validated? As well as any target in this document could be, and not by genetics. Omalizumab — a weaker antibody against the same molecule — has been approved for this exact disease since 2014 and generates roughly $1.2 billion a year in chronic-spontaneous-urticaria revenue [WEB ESTIMATE — search summary of Novartis fiscal-year 2024 disclosure, read 2026-08-19]. “Reducing IgE relieves chronic spontaneous urticaria” is not a hypothesis; it is the standard of care. The Open Targets block noted in section 0 therefore costs this analysis very little.

The exact scientific step this readout must prove. Not that anti-IgE works — that is settled. The readout must show that CUE-221’s extra potency and its CD23 mechanism convert into a clinically larger effect than the approved comparator, in patients, at a tolerable dose. Everything published on the CD23 mechanism to date is laboratory work plus animal models plus a single-dose first-in-human study; nothing has yet shown that suppressing new IgE synthesis produces a deeper or more durable clinical response than simply mopping up the IgE already there.

The honest scientific risk. Three, in order of size:

• Superior binding has already failed to convert once, in this exact disease. Ligelizumab binds IgE more tightly than omalizumab, beat omalizumab in a 2b, and then did not beat it across two Phase 3 trials enrolling over 2,000 patients between them, after which Novartis stopped the programme [VERIFIED — ClinicalTrials.gov NCT03580356, NCT03580369, NCT04210843]. The reason is believed to be a ceiling effect: once free IgE is suppressed far enough, suppressing it further adds nothing. If that ceiling is real, CUE-221’s eightfold potency edge buys nothing either, and only the CD23 mechanism can differentiate it.
• The CD23 mechanism is unproven in humans as a clinical benefit. It is well characterised biochemically and it is genuinely novel. It has never been shown to change a UAS7 score.
• The evidence base is one primary paper, written by the sponsor. Of nine PubMed hits, exactly one is primary research, and 27 of its 29 authors are employees of United BioPharma or its affiliates [VERIFIED — PubMed get_article_metadata PMID 35912861, 2026-08-19]. There is no independent replication of the CD23 finding.

A.2 Clinical development plan, timeline, feasibility, resourcing

Trials

Two things in that table deserve to be said plainly. First, every registered CUE-221 trial that started after 2022 sits at status UNKNOWN with no results posted, years past its primary completion date. NCT05298215 — the only registered Phase 2, and the only one with a clinical efficacy measure anywhere in it — reached its primary completion date on 2023-12-31 and has published nothing in the thirty-two months since. Neither has NCT04175704. That is not proof of a bad result; registries are frequently left un-updated by small sponsors, especially outside the United States. It is, however, the opposite of a clean development record, and it means there is no public efficacy number for this molecule anywhere except the single-dose findings summarised in the 2022 paper.

Second, the trial that reads out cannot be inspected. Its size, its randomisation, its comparator, its primary endpoint and its analysis timepoint are all undisclosed, in a study run by a private related party of the licensor, in a jurisdiction whose data Cue’s own risk factors flag as a concern. A reader cannot check whether it is powered to show what a positive headline would claim.

Feasibility matrix

Resourcing sufficiency. Yes, comfortably, for this event. Cue holds roughly $67 million pro forma and states a runway of at least twelve months from 2026-08-14; the catalyst lands about six weeks out (../company.md C.3). Cue is not even paying for this trial — Genesis Life Sciences is running it in the territory Cue does not own. Cue’s cost is the $15.0 million upfront it has already paid, plus milestones, plus whatever the global Phase 2b in food allergy costs once it starts. The resourcing question that matters is the next one: funding a global Phase 2b, and then a Phase 3 programme against a comparator, on roughly $67 million and a $36.8 million unused at-the-market facility. That will require another raise, but not before this readout.

A.3 Target label, target product profile, strategic questions, designations

A.3a Target label. Adults and adolescents with chronic spontaneous urticaria who remain symptomatic despite H1-antihistamine therapy — the same population omalizumab is approved in — pursued outside mainland China, Hong Kong, Macau and Taiwan, which Cue’s licence excludes [VERIFIED — 10-Q Note 8]. The company’s own stated commercial priority for the molecule is food allergy rather than urticaria; chronic spontaneous urticaria is where the data comes from [VERIFIED — 8-K Exhibit 99.1, 2026-05-14].

A.3b Target product profile

A.3c Strategic Go/No-Go questions. The asset’s next decision is whether to commit to a global Phase 2b, so the pre-Phase-II set below is the one that matches.

Pre-Phase-II (Go-to-Phase-II):

A.3d Regulatory designations. None disclosed [VERIFIED — Cue Biopharma press releases 2026-04-30 through 2026-08-18, and the 10-Q filed 2026-08-14, none of which announces or references any designation]. No Fast Track, no Breakthrough Therapy, no Orphan Drug, no Priority Review voucher. That is unsurprising: chronic spontaneous urticaria has an approved therapy, which rules out orphan status and raises the bar for the accelerated routes. The one regulatory step that has happened is an Investigational New Drug application submitted to the FDA on 2026-08-03 for the food-allergy indication — permission to begin a US trial, not a designation and not an approval [VERIFIED — BPIQ fetch_company_historical_catalysts; 8-K Exhibit 99.1, 2026-08-14].

A.4 Target product profile valuation matrix

Calibration examples, not asset claims: in oncology each incremental month of overall survival is worth roughly $150–300M in peak-sales potential; oral formulations command roughly 15–25% price premiums over injectables. The second of those cuts against this asset rather than for it — the principal oral competitor in this indication is remibrutinib.

A.5 Evidence quality and endpoints

Endpoints

A.5b Key opinion leaders.

Panel as of. 2026-08-19.

Investigators

The pivotal China study is not on ClinicalTrials.gov, so no investigator on the trial that produces this catalyst could be identified. search_investigators was instead run against the registered Taiwan trials. It returned one name in an administrative contact role, not an investigator role. The two people below are the clinical co-authors of the first-in-human report — the only individuals publicly identifiable as having run a CUE-221 trial. Both are recorded with the sponsor relationship their own byline discloses.

Independent voices

One voice qualified. It is a thin panel and the reason is worth stating: nobody outside the sponsor has published on CUE-221’s clinical performance, because no efficacy result has ever been published for anyone to comment on.

One notable omission, recorded deliberately. Marcus Maurer (Institute of Allergology, Charité — Universitätsmedizin Berlin) is the most-cited commentator in this indication and appears as senior author on two of the nine retrieved papers that mention CUE-221. He is not listed below, because he holds well-known relationships across competing chronic-spontaneous-urticaria sponsors and this sweep did not verify that disclosure set. Recording him as “independent” without having read his conflicts would be exactly the unevidenced claim rule 40 forbids.

Judgement

Dissent

Empty. endpoint_supported is UNKNOWN rather than SUPPORTIVE_CONTESTED, so no named disagreement is required, and inventing one nobody holds would be worse than leaving this blank.

B. Commercial assessment

B.0 Current treatment algorithm

What a patient with chronic spontaneous urticaria receives today, in order:

1. A second-generation H1-antihistamine at the standard dose — cetirizine, fexofenadine, bilastine and similar. Cheap, oral, safe, and enough for roughly half of patients.
2. The same antihistamine at up to four times the standard dose. This is the guideline-endorsed second step. Even after it, 40–60% of patients still do not achieve adequate control [WEB ESTIMATE — CSU treatment literature summary, read 2026-08-19].
3. Omalizumab, 300 mg subcutaneously every four weeks. The only approved biologic in this indication, and the only one recommended by international guidelines. Roughly a third to a half of these patients reach complete response; the rest do not [WEB ESTIMATE — ASTERIA/GLACIAL summary, read 2026-08-19].
4. Ciclosporin or another immunosuppressant, off-label, for omalizumab non-responders. Effective but poorly tolerated, with kidney and blood-pressure monitoring.

Where CUE-221 would fit: step 3, displacing omalizumab, or step 3b, taking the patients omalizumab fails. Those are two very different commercial propositions. Displacing omalizumab means beating it head-to-head — the bar ligelizumab could not clear — into a market where biosimilars are already arriving. Serving omalizumab failures is a smaller, better-defended niche, but it is also precisely where two non-anti-IgE mechanisms are heading: remibrutinib, an oral BTK inhibitor Novartis has taken through positive Phase 3, and barzolvolimab, Celldex’s anti-KIT antibody that depletes mast cells outright. Neither depends on IgE, so neither has the ceiling problem that sank ligelizumab.

B.1 Competitive landscape and positioning

Where this asset wins, and the single fact the thesis rests on. It wins in one place only: if suppressing new IgE synthesis through CD23 produces a deeper or longer-lasting response than mopping up existing IgE, and if that shows up as a bigger UAS7 improvement than omalizumab delivers. The single fact the thesis rests on is the CD23 engagement demonstrated in J Clin Invest 2022 — a real, precisely characterised, laboratory finding that has never been shown to change a patient outcome. Everything else about this asset is worse than its competitors: it is later, it is intravenous, its evidence base is one sponsor-authored paper, and it has no data in the territory its licensee actually owns.

Calibration example, not an asset claim: oncology first-movers in novel mechanisms have shown roughly 3.2× higher peak-sales potential but about 1.8× higher development risk. CUE-221 is not a first-mover on its mechanism; it is a fourth entrant on a target opened in 2014.

B.2 Addressable market

Launch markets. The United States first, then EU5 and Japan — the territory Cue’s licence covers. Mainland China, Hong Kong, Macau and Taiwan are excluded and belong to Ascendant, which is also where the only trial is running [VERIFIED — 10-Q Note 8].

B.3 Value and feasibility

B.3a Expected peak sales. Built bottom-up, in the Cue territory only (world excluding mainland China, Hong Kong, Macau and Taiwan), conditional on approval, and excluding food allergy entirely — food allergy is the larger prize the company is actually chasing, but no patient has been dosed and no estimate for it would be defensible.

No single figure is given, and none should be, because two of the three inputs — the price a differentiated anti-IgE can hold against a biosimilar, and the share it takes from two better-placed competitors — are unverified (rule 10). Every figure above is a global-territory-excluding-Greater-China number, at peak, conditional on approval, before the high single-digit to low double-digit royalty owed to Ascendant.

B.3b Feasibility

B.4 Product-development risk

Framing questions.

1. Does the development plan support the target product profile claims? Only partly. The target product profile claims superiority over omalizumab and, implicitly, less frequent dosing. A dose-ranging Phase 2 with an active comparator can suggest the first; nothing in the disclosed plan addresses the second, because no subcutaneous formulation has been disclosed, and the registered trials all dosed intravenously.
2. Will the identified risks affect the target product profile? Yes, two of them directly. The formulation gap attacks the “payer value” and “provider” rows outright. The unregistered, unspecified trial design means the efficacy row cannot be checked against the evidence that will be used to claim it.
3. If a risk cannot be mitigated, is the asset still differentiated from competitors? If the CD23 mechanism does not convert into a clinical benefit, then no. What remains is an intravenous anti-IgE arriving seven or more years after omalizumab’s biosimilars, against an oral competitor with completed positive Phase 3 data. Differentiation in that world is zero.

Readout

Sources

The modelled estimate. The standard arithmetic — a registry primary completion date plus the lag to database lock and analysis — cannot be computed here, because the ctgov row above is null. The arithmetic actually used is stated instead so it can be argued with: take the company’s own stated ceiling of 2026-09-30, and add the lag between a trial finishing and a licensee being able to announce it, which here includes a data-and-know-how transfer from Genesis Life Sciences to Cue that the licence prices as a separate $6.5 million milestone and that has not yet been paid [VERIFIED — 10-Q Note 8]. Applying the stated default lag of two to four months to that transfer step, rather than to a database lock, gives a plausible band of 2026-09 to 2026-12, and this is tagged [UNVERIFIED — modelled, default lag] because data/benchmarks/readout-lag.json holds no observation for a comparable trial (its observations array is empty). The tag is doing real work here: this is a stated default applied to a step it was not written for, on a trial nobody outside the sponsor can inspect.

Window

Basis. The only substantive source is the company’s own guidance, which names a quarter-end ceiling rather than a day or a month, so the precision is PERIOD and no amount of confidence in the company can raise it — that is a fact about what was disclosed, not about who disclosed it. Earliest is set at 2026-09-15 because on 2026-08-14 the chief executive said the data was expected “by the end of this quarter”, which admits a landing in the second half of September but effectively rules out the days immediately after that statement. Likeliest is 2026-09-30, the guided ceiling, because a company that reaffirms an end-of-quarter date twice in twelve days usually delivers at the edge of it. Latest is 2026-12-15 rather than 2026-09-30, because Cue does not run this trial and cannot announce data it has not received: the licensor’s data-and-know-how transfer is a separately priced, unpaid milestone, and a one-quarter slip is the ordinary consequence if it is slow. Confidence is MEDIUM and not HIGH because the guidance has been reaffirmed three times without slipping, which is genuinely reassuring, but the corroborating registry source that would normally support it does not exist.

Disagreement. CONSISTENT. The two sources that returned a value — BPIQ’s placeholder and the company’s own statement — both resolve to 2026-09-30, and they do so because BPIQ synthesized its placeholder from that same guidance. Note what that means: they agree because they are the same source read twice, not because two independent sources converged. The genuinely independent check — the registry — returned nothing.

Date slippage. Parsed from the BPIQ note field and the company’s own releases, oldest first.

Zero slips, and one tightening. Three dated statements, two transitions, and neither is a slip: “H2 2026” narrowed to “end of Q3 2026”, and “end of Q3 2026” was then reaffirmed unchanged. Guidance moving earlier is the opposite of the pattern 02-connectors.md warns about, and it is the single most encouraging fact in this section.

Attribution

Status. CLEAN

Conflicts

Empty, as CLEAN requires.

Computed, not authored: lib/clustering.mjs’s attributionFor was run over this ticker’s pipeline for bpiq_drug_id 20405 at CATALYST_CLUSTER_MIN_MONTHS = 6 and returned {"status":"CLEAN","conflicts":[]}. The reason is structural rather than lucky: bpiq_drug_id 20405 is the only row on CUE’s entire pipeline with has_catalyst: true (../company.md C.2). The other four rows carry no pending catalyst, so there is nothing on this ticker for this window to collide with. A price move around this readout is attributable to this program alone.

Note. Not required — CLEAN.

Market and timing for this event

• Plain takeaway. The whole company rides on this. The shares are 5.2× their 52-week low with no data behind that move, then 43% below the high they reached in June, and two specialist investors paid $33.21 six weeks ago for stock now at $26.10. The market has walked the price back from the licensing euphoria without waiting for the result.
• Months to this catalyst. About 0.9 months — roughly four weeks — from 2026-08-19 to readout.window.earliest of 2026-09-15. Said plainly: readout.precision is PERIOD, so this is the earliest edge of a window that could extend to 2026-12-15, not a countdown to a known date.
• Expected move around this event. ../company.md C.6 records the options chain as structurally unusable — no listed strike within $21 of spot — so no market-implied move can be read, and a bracket is given instead of a point estimate. Calibrated against framework/07-benchmarks.md: the Phase 2 class averages are +12% on a positive and −16% on a negative; immunology runs +8% / −16%; a placebo-controlled design carries a 29% amplitude and a head-to-head design 72%, and this trial is both placebo- and active-comparator-controlled, so it sits between the two; and a company under $1 billion in market capitalisation reacts about 9.5× more strongly at Phase 2 than one above it, which CUE is at roughly $190 million. Bracket: a move of 25% to 55% in absolute terms, in either direction [WEB ESTIMATE — IQVIA 2024]. This ticker’s own history supports the wide end rather than the class average: it moved +12.85% intraday on a Phase 1 poster (../company.md C.7).
• Nearest comparable past reaction. The 2024-11-08 row of ../company.md C.7 — Phase 1 CUE-101 data at SITC, +12.85% intraday. It is the closest analogue because it is the only row in that table that is a clinical data event with a clean press-feed check. It is not a good analogue in three ways, all of which point the same direction: it was a Phase 1 poster rather than a controlled Phase 2 readout, it was on an asset that carried a fraction of the company’s value rather than nearly all of it, and it predates the 1-for-30 reverse split and the entire change of business. A readout on the dominant asset of a single-asset company should move the shares considerably more than 12.85%.
• Materiality. Dominant — the word recorded for this program in ../company.md C.2. It is the only pipeline row with a pending catalyst, on the asset the company itself calls its lead, and the entire re-rating of the shares traces to its in-licensing. The stock-direction call below is sized on that: this is not a program that nudges the shares, it is the program that sets them.
• Date slippage. Zero slips across three dated statements, with one tightening from “H2 2026” to “end of Q3 2026”. See Readout, above.

Spot. $26.10, read 2026-08-19, cited from ../company.md C.1 and C.4.

On the published analyst targets: they are unusable and are not carried into any figure below. A search returned a $3.00 average from two analysts, a $115.00 target from one, and a $5.13 consensus from eight with a $10.00 high dated 2024 [WEB ESTIMATE — aggregator summaries, read 2026-08-19]. Those cannot all be about the same security, and they are not: CUE did a 1-for-30 reverse split on 2026-04-23, and the aggregators are mixing pre-split and post-split targets without adjustment. $5.13 pre-split is $153.90 post-split; $3.00 pre-split is $90.00. Rule 6 requires rejecting a broken third-party figure rather than passing it through, so no analyst target appears among the anchors below.

Scenario prices

Expected value. $28.70, or +10.0% against the $26.10 spot. Method: the 72% probability applied to the midpoint of each range — 0.72 × $35.00 + 0.28 × $12.50. This is arithmetic, not advice, and it is not a price target. It is worth reading alongside the ranges rather than instead of them: a +10% expected value assembled from a +34% positive midpoint and a −52% miss midpoint describes a position with a small positive edge and very large variance, not a comfortable one.

Run-up

Entry and exit

Predicted move and peak

Priority score drivers

Two of those rows need a sentence each. Priced-in-ness fell back to its neutral default of 50 because data/prices/CUE.json does not exist — lib/prices.mjs had nothing to compute a 52-week position from. The hand computation in ../company.md C.4 puts the price at 52.1% of its 52-week range, which would have scored 48; the difference is immaterial to the total, but the gap is recorded rather than papered over. Financing and clustering risk scored 10, the lowest value the OK runway bucket allows, because the company holds roughly $67 million against a catalyst four weeks away and the attribution reading is CLEAN — the two independent risks this driver combines are both close to absent here.

Priority score. Priority score 14 · formula_version 1.0.0

That is a low score, and the reason is entirely the date. Six of the seven drivers are respectable — a 72% outcome probability, a 74 on value uplift, a 61 on squeeze mechanics and a 10 on the negative driver. date_confidence scored 20, and lib/runup.mjs applies it as a multiplicative gate over the other six rather than as a seventh averaged term, precisely so that a program whose readout date is known only to a quarter cannot out-rank one whose date is known to a day. The score is therefore a statement about tradeability, not about the asset: as a scientific and commercial proposition CUE-221 ranks well, and as a timed run-up entry it does not, because nobody outside the sponsor knows what day the data lands. Read the 14 as the honest consequence of a PERIOD-precision window, not as a verdict on the program.

Settlement. Left null at lock time. Note additionally that no run-up settlement can be resolved for this ticker yet: runup.exit requires resolving the exit rule against data/prices/CUE.json, which does not exist, so exit is null and stays null until scripts/fetch-prices.mjs covers CUE.

Verdict

What I would do. Watch.

Why. The science is real and the setup is not. CUE-221’s CD23 mechanism is a genuine, precisely characterised advance on the only anti-IgE approved for this disease, and beating placebo is likely — modelled at 72%. But the trial that reads out has no ClinicalTrials.gov record at all, so its size, its design, its comparator and its primary endpoint cannot be checked before the number arrives; the only registered Phase 2 of this molecule used a laboratory measure rather than a symptom score as its primary endpoint, and never posted a result; and the bar that actually matters is beating the active comparator, which is exactly the bar ligelizumab cleared in Phase 2 and failed in Phase 3 on the same target in the same disease. A +10% expected value assembled from a +34% positive midpoint and a −52% miss midpoint is a small edge with very large variance, on a company where this readout is the only thing holding the price.

What would change this. Publication or disclosure of the China study’s protocol — its sample size, its randomisation, its comparator and its primary endpoint — before the topline lands. If the primary endpoint turns out to be UAS7 at week 12 in an adequately powered trial against omalizumab, this becomes a buy at these levels. If it turns out to be free IgE or another pharmacodynamic measure, a “met its primary endpoint” headline would carry almost no information about the asset’s commercial future, and the trade becomes a coin flip dressed as a result. Disclosure of a subcutaneous formulation would independently raise the whole case.

What to watch.

• Now to 2026-09-30 — any 8-K, press release or investor-conference remark that describes the China study’s design, size, comparator or primary endpoint. This is the single highest-value observable between now and the event.
• Now to 2026-09-30 — a ClinicalTrials.gov or Chinese-registry record appearing for the study, or results finally posting for NCT05298215 (primary completion 2023-12-31) or NCT04175704 (primary completion 2024-11-30). Either would materially change the clinical-operations risk reading in B.4.
• 2026-09-15 to 2026-12-15 — the topline itself. Read the comparator arm before the placebo arm.
• On the day of the topline — whether the release names a UAS7 figure at a stated timepoint with a p-value against both placebo and the active comparator, or only against placebo. The absence of a comparator number in a trial that has a comparator arm would be the tell.
• Within roughly 30 days of a positive topline — an 8-K recording the Ascendant Top-Up Obligation firing and a milestone payment from the $205.0 million development-and-regulatory tranche. Both are contractual consequences of “threshold data” and both dilute or spend.
• 2026-12-31 — the CUE-401 Investigational New Drug submission and Phase 1 start, guided to year-end. Not this program, but it is the residual value the miss scenario’s floor rests on.

Locked prediction

• Outcome-direction (will the readout succeed on its primary endpoint?): positive — probability 72%, band 58–83% [UNVERIFIED — modelled]. The reasoning: anti-IgE in antihistamine-refractory chronic spontaneous urticaria is not a mechanism that needs proving — omalizumab, a weaker antibody against the same target, was approved for this exact indication in 2014, and ligelizumab also beat placebo comfortably before failing against omalizumab. CUE-221 binds IgE more tightly than either and produced free-IgE suppression beyond twelve weeks from a single dose. Against that: the trial’s size and powering are undisclosed and cannot be checked; it is a dose-ranging study, which is designed to compare doses rather than to prove one; the only registered Phase 2 of this molecule enrolled 25 patients and never reported; and Cue’s own risk factors flag the acceptability of China-generated data. The band is 25 points wide because the design is unverifiable, which is a wider band than a comparably de-risked mechanism would usually carry. No third party has published a probability on this event — the analyst targets that exist mix pre- and post-reverse-split conventions and are rejected under rule 6, so there is no published figure to position against.
• Stock-direction (which way do the shares move?): up — confidence low — window 2026-09-15 to 2026-12-31, basis: opens at readout.window.earliest and closes two weeks after readout.window.latest, so the window covers the readout wherever inside its PERIOD precision it lands, plus the settling period after it. Materiality is dominant (../company.md C.2), and this call is sized on that: the only pending catalyst on the ticker, on the asset the company calls its lead, with the entire re-rating from a $4.98 52-week low traceable to its in-licensing. The call is up because the expected value is +10% against spot and the two agree. Confidence is low and not medium because the variance dominates the edge: the positive definition being scored spans both a clear comparator win and a comparator tie, and those two produce opposite share-price outcomes.
• Scenario prices: positive $24–$46 · miss $9–$16
• Expected value: $28.70, +10.0% against spot $26.10 (arithmetic, not advice)
• Run-up: entry $26.10 on 2026-08-19, exit rule T-5 trading days before readout.window.earliest — predicted move +3% to +22%, predicted peak +8% to +30% around 2026-09 — priority score 14, formula_version 1.0.0. The score is low because date_confidence gates it; see the Run-up section for why that is a statement about tradeability rather than about the asset.
• Settles on: Cue Biopharma, Ascendant Health Sciences or Genesis Life Sciences publicly reporting topline results from the China Phase 2 dose-ranging study of CUE-221 in chronic spontaneous urticaria. Positive means at least one CUE-221 dose arm is reported statistically superior to placebo (p < 0.05) on the study’s pre-specified primary efficacy endpoint, with no new dose-limiting safety signal disclosed and no statement that the programme is being discontinued. Primary source: a Cue Biopharma press release or SEC filing (8-K), or an Ascendant Health / Genesis Life Sciences release, whichever comes first.
• Locked: yes · Settled: no

Program data-quality flags

• The pivotal trial has no ClinicalTrials.gov record. Three separate search strategies confirmed it (by condition and intervention; by sponsor; by condition, country and phase). The readout block’s ctgov row is therefore null with a note, the modelled estimate cannot use the standard primary-completion-plus-lag arithmetic, and the trial’s size, arms, comparator, primary endpoint and analysis timepoint are all unverifiable before the result lands. This is the largest single gap in this document.
• Both readout sources that returned a value are the same source read twice. BPIQ synthesized its 2026-09-30 placeholder from the company’s own guidance, so disagreement: CONSISTENT records agreement between a statement and its own echo, not a convergence of independent sources.
• Open Targets is BLOCKED, verbatim Rate limit exceeded for client: global, on three attempts. No independent genetic target validation. The cost is small because target validation here rests on an approved drug against the same target in the same disease, which is stronger evidence than a genetic association.
• ChEMBL has no record of this molecule. compound_search on UB-221 returned a legitimate empty result while the same tool returned a full record for omalizumab on the same sweep. No selectivity or off-target data — of limited consequence for an antibody.
• Two of the three registered CUE-221 trials sit at status UNKNOWN, years past their primary completion dates, with no results posted. NCT05298215 (primary completion 2023-12-31) and NCT04175704 (primary completion 2024-11-30). No efficacy figure for this molecule has ever been published beyond the single-dose findings in the 2022 paper.
• The evidence base is one primary paper with 27 of 29 authors employed by the sponsor. The differentiating CD23 finding has never been independently replicated. PubMed’s authors-returns-nulls bug did not fire on this sweep; the author list was fully retrieved and checked.
• No conflict-of-interest statement could be read for any named person. The PubMed connector returns metadata and abstracts but not declaration-of-interest sections. Every conflicts_checked entry in A.5b says so explicitly and certifies the search rather than the absence.
• The formulation is intravenous in every registered trial and no subcutaneous formulation has been disclosed. This is not a data-quality flag about a connector; it is an unresolved fact about the asset that materially affects B.1, B.2 and B.4 and is recorded here because a reader could otherwise assume a subcutaneous product.
• data/prices/CUE.json does not exist, so the priced-in run-up driver fell back to its neutral default, and runup.exit cannot be resolved and stays null until the cache covers this ticker.
• Published analyst targets are unusable and were rejected under rule 6 rather than passed through: aggregators are mixing pre- and post-1-for-30-reverse-split figures, producing a spread from $3.00 to $115.00 that cannot describe one security.