Normal gene editing cuts DNA. This switches a gene off and leaves it intact. Hepatitis B is simply the first disease it is being tried on.
30%likely
Single sourceNext check
Watching
Why it is different
The bet is on the method of switching genes off reversibly, not on the hepatitis drug that happens to be testing it first.
Why it matters
If it holds up in people it opens a whole class of medicines, far beyond this one disease.
Next check · 30 Jun 2027
Is a second-stage trial of TUNE-401 registered and recruiting patients on or before 30 Jun 2027?
How we scored it ·
from how similar cases went
If right, how big5/5How overlooked5/5Upside vs downside2/5Source trust3/5Independent sources1/5
The exact wording we score
On or before 31 Dec 2030, TUNE-401 (NCT06671093) reports in a controlled trial that at least 25% of the evaluable HBV cohort achieves confirmed HBsAg loss below assay quantification and sustains it for at least six months off therapy. Resolves YES only if every condition holds; NO on a lower response, non-durability, program termination, or no qualifying controlled readout by the deadline. No other asset can be substituted.
PROOF
Evidence and provenance
The links below are the evidence recorded when this forecast was scored. Each note states source quality or bias; a citation is not an endorsement.
Recorded sources · 4 citations / 1 independent origins
2026-07-28 | sell-side: no direct coverage in the evidence set; mainstream press: data largely absent outside company-release relays; priced market: no accessible public instrument identified; specialist: small-company abstract and one analyst review carry the entire human-data claim | E5
Revision history5 entries
Probabilities and interpretations are never silently overwritten. This is the append-only record of what changed and why.
The “closed” class set was not closed: it explicitly deferred enumerating other members. That creates a post-outcome rescue path. The claim returns to the only verified asset, TUNE-401, with an exact 25% cohort threshold and six-month off-treatment duration; P 35→30.
(3rd pass) — Claim raised from asset-level to class-level; P 30→35. The edge argument said in as many words that 'the platform read-through (not the HBV number) is the mispriced part', while the claim scored only TUNE-401's HBsAg number. The modality could have been validated by another sponsor and this signal would have logged a miss — a calibration penalty for having been right. The claim now sits where the belief sits, with a closed eligible set and criteria fixed at logging time so membership cannot be added retroactively to rescue it. P moves only 30→35 because TUNE-401 remains the sole verified member (~1.2 effective assets); enumerating the rest is an open item for the 2027-06-15 review. This does not repair the sourcing problem — still 1 independent origin, still 🔴, still the book's clearest research priority.
(2nd pass) — Reclassified refclass: phase3-attrition rather than corrgroup; these five drug bets share a base rate, not an event, so discounting them as correlated portfolio exposure was wrong. Assumption tag A2+ added: this signal is long 'announced dates hold' with less registry evidence than any other signal in the book — no Phase 2 registered as of 2026-07-28 against guidance of 'as early as late 2026' — and prices no schedule risk at all.
v2 rescore. P held at 30; no change found since 2026-07-23. The only registered TUNE-401 study remains the Phase 1b (NCT06671093, Recruiting, primary completion 2028-06-28 est., n=36); no Phase 2 record exists; no peer-reviewed publication. indepsources corrected 4→1 — the single most important sourcing fix in this pass. BioSpace (x2), the Tune press release and the InSilens analysis all trace to one company dataset that has never been peer-reviewed. Confidence drops 🟡→🔴 while materiality is unchanged, which is exactly the intended behaviour: this is now the book's clearest research priority, not a downgrade.
created from fresh research sweep. HIGH EDGE / LOW CONFIDENCE — flagged research priority; needs peer-reviewed data or a controlled readout before any conviction upgrade.
Method challenge
Adversarial review record
Admitted
Structured internal role review tied to this frozen claim. It is not independent human peer review. Independent specialist review not yet performed
clinical-domain specialist
biostatistics and trial-design reviewer
regulatory reviewer
forecasting-method reviewer
Strongest specialist challenge
The only registered study is a 36-participant open-label Phase 1b with safety as its primary outcome and estimated completion in 2028. Company-reported biomarker repression cannot be treated as controlled functional-cure evidence.
Outside view
The outside view for moving from early open-label HBV biomarker activity to durable controlled HBsAg loss is poor, especially in a cure-hard indication. The 30 percent call already requires an unusually large uplift from sponsor-reported durability.
Causal rival
Transient antigen suppression, assay variability, nucleoside-background therapy, selection of responsive participants, or non-durable epigenetic effects can reproduce the early biomarker story without generating off-treatment functional cure.
Measurement risk
The future qualifying trial must be controlled, pre-specify the evaluable cohort, use HBsAg below assay quantification, and document six months off therapy. Open-label dose cohorts or mean antigen decline cannot satisfy the literal threshold.
Residual risk
No qualifying controlled protocol is registered yet, and LNP re-dosing or immune risks remain unresolved. The first meaningful review gate is therefore protocol registration, not another sponsor conference update.
COLDTELL