Provenance Record

Clear, well-structured News piece (750 words, within the 400-1200 range). Sectioned Overview / What We Know / Why It Matters / What We Don't Know. Efficacy figures are presented with their statistical qualifiers (LS mean vs observed mean, p-values) rather than cherry-picked, and the cross-trial vosoritide comparison is explicitly framed as non-head-to-head. Appropriate technical depth for a clinical-trial story.

All 7 source snapshots read from disk (sources/2026-07/bridgebios-oral-infigratinib-adds-21-cmyear-of-growth-in-achondroplasia-phase-3-published-in-nejm/). (1) source-0.html.gz clinicaltrials.gov/study/NCT06164951 — DID NOT confirm: the committed snapshot is an empty JavaScript SPA shell (title only 'ClinicalTrials.gov', no study data; the three '114' hits are Unicode font-range strings, not the enrollment count). The registry-attributed specifics (114 participants; official title with 'Phase 3, Multicenter, Double-Blind, Randomized, Placebo-Controlled ... Children 3 to <18 Years of Age'; QED Therapeutics as lead sponsor; sprinkle-capsule intervention; primary outcome = change from baseline in annualized height velocity) were therefore verified via last-resort query to the ClinicalTrials.gov v2 JSON API (https://clinicaltrials.gov/api/v2/studies/NCT06164951), which returned an EXACT match on all of them (officialTitle verbatim, enrollmentInfo.count 114, leadSponsor 'QED Therapeutics, a BridgeBio company', intervention 'Daily doses of oral Infigratinib (sprinkle capsules)', primaryOutcome 'Change from baseline (BL) in annualized height velocity (cm/year)', minimumAge '3 Years' / maximumAge '17 Years'). (2) source-1.html.gz quiverquant.com — CONFIRMED: NEJM publication + ICCBH 2026 late-breaking presentation, +1.74 LS mean / +2.10 observed cm/yr, PROPEL 3 met primary endpoint of change from baseline in annualized height velocity, arm span +0.37 SD, -0.05 in children ages 3 to 8 years. (3) source-2.html.gz biotechreality.com — CONFIRMED: +1.74/+2.10 cm/yr, selective oral small-molecule FGFR1-3 inhibitor, no adverse events associated with FGFR1/FGFR2 inhibition, Kyowa Kirin exclusive Japan license February 2024, NDA Q3 2026 with early-to-mid 2027 launch and EMA MAA. (4) source-3.html.gz clinicaltrialvanguard.com — CONFIRMED: +2.1 cm/yr headline, 'largest mean height velocity increase versus placebo recorded in any Phase 3 achondroplasia study', sitting-height-to-standing-height ratio LS mean -0.05 (p<0.05), vosoritide 1.57 cm/yr over placebo at 52 weeks (attributed to vosoritide's Lancet Phase 3), NDA Q3 2026 / US launch early-mid 2027. (5) source-4.html.gz stocktitan.net — CONFIRMED: +1.74/+2.10 cm/yr, arm span +0.37 SD, height Z-score +0.41 SD at Week 52, body proportionality -0.05 in children <8 (p<0.05), 3 cases (4%) hyperphosphatemia mild/transient/asymptomatic, no discontinuations or serious drug-related AEs, hypochondroplasia, and the verbatim Savarirayan quote 'The publication of our pivotal trial data (PROPEL 3) in the New England Journal of Medicine is a defining milestone for the field of skeletal dysplasia'. (6) source-5.html.gz medlineplus.gov — CONFIRMED: FGFR3 variants make protein overly active, 1 in 15,000 to 40,000 newborns, most common short-limbed dwarfism, adult male avg 131 cm (4 ft 4 in) / female 124 cm (4 ft 1 in), spinal stenosis, foramen magnum compression, sleep apnea, hydrocephalus. (7) source-6.html.gz biospace.com — CONFIRMED: Voxzogo (vosoritide) FDA approval initially granted November 2021, analog of C-type natriuretic peptide.

Every efficacy and safety figure (+2.10 observed / +1.74 LS mean cm/yr, +0.37 SD arm span, +0.41 SD height Z-score, -0.05 body proportionality in children <8, 3 cases/4% hyperphosphatemia) traces to at least one snapshot and cross-corroborates across StockTitan, Quiver, Biotech Reality, and Clinical Trial Vanguard, all of which reproduce the BridgeBio company release. Registry specifics confirmed exactly against the ClinicalTrials.gov v2 API. Direct quote is verbatim. No fabricated specifics, no hallucinated quotes. The unverified '0.25 mg/kg' dose figure the Journalist flagged as dropped is correctly absent from the article body (grep count 0); the API confirms 0.25 mg/kg/day is the real registered starting dose, so its omission is conservative and harmless.

APPROVE. A clean, well-sourced, neutral clinical-trial story. All efficacy figures, trial specifics, the NEJM publication, and the safety profile are verified; the cross-trial vosoritide figure is correctly framed as non-head-to-head; the dropped 0.25 mg/kg dose stayed out; and no load-bearing claim rests solely on a bot-blocked source (the registry facts were independently confirmed via the ClinicalTrials.gov API). The only warning is off-allowlist provenance, which is an editorial judgment call rather than a factual error — a corrections note cannot honestly 'correct' a fact that is right, so APPROVE (not APPROVE_WITH_CORRECTIONS) is the appropriate verdict.