Provenance Record

Well-structured News piece (696 words, within the 400-1200 News range). Clear Overview / What We Know / What We Don't Know / Analysis arc. Technical depth appropriate for a rare-disease regulatory story; the molecular mechanism, the gait-speed endpoint, and the GFAP biomarker are tied together coherently without overstatement.

All 4 source snapshots read from disk and integrity-verified (uncompressed sha256 of every gunzipped file matches manifest exactly). source-0 (BioSpace topline, 200): confirms the verbatim 'first time an investigational medicine has shown a positive disease-modifying impact in AxD' quote, astrocytes, '1 per 1 to 3 million', NCT04849741, 54 participants, ages 1.5-53 ('between the ages of 1.5 and 53 years'), 13 sites / eight countries, 2:1 randomization, 60-week double-blind, 25 mg and 50 mg every 12 weeks, and the primary-endpoint quote verbatim (10MWT, week 61, mean difference 33.3%, p=0.0412). source-1 (BioSpace FDA Priority Review, 200): confirms NDA accepted for Priority Review, September 22, 2026 target action date (PDUFA), Breakthrough Therapy / Orphan Drug / Rare Pediatric Disease designations, antisense oligonucleotide inhibiting excess GFAP, the GFAP gene, the disease description, the Brett Monia CEO 'first and only treatment' quote verbatim, AND the verbatim secondary/exploratory-endpoint sentence ('adaptive function, communication, GI symptoms, sleep and seizures ... consistently favored zilganersen'). source-2 (BioSpace AAN 2026, 200): confirms plasma GFAP reduced 33.6% at Week 61 vs control (nominal p=0.003), the 21%-vs-0% 'much better' PGIC result, and the Amy Waldman lead-investigator quote verbatim. source-3 (FierceBiotech, 200 this fetch): confirms the single mechanism sentence cited to it ('bind to the [messenger RNA] ... prevent it from being turned into protein'). Spelling of zilganersen, Alexander disease, GFAP, and Ionis is verbatim across all four snapshots (Rule 7).

Every clinical specific traces to the snapshot it is cited to, with one cite-target nuance: the sentence 'Results across key secondary and exploratory endpoints evaluating adaptive function, communication, gastrointestinal symptoms, sleep and seizures also consistently favored zilganersen' is tagged with the AAN release URL (source-2), but its near-verbatim wording ('adaptive function, communication, GI symptoms, sleep and seizures') appears in the FDA Priority Review release (source-1), which IS in the article's sources array. source-2's own endpoint phrasing lists 'communication, swallowing, gastrointestinal and autonomic symptoms.' The claim is true and carried by a cited BioSpace source; only the link target is imprecise. Documented in the corrections record. The conflicting off-allowlist '2-53' age range (endaxd.org) was correctly dropped in favor of the source-grounded '1.5-53' (Rule 2).

Substantively strong, accurate, well-sourced clinical/regulatory News article. Every number, date, designation, and direct quote verifies verbatim against the cited BioSpace snapshots; the sourced 'first' claim and the September 22, 2026 PDUFA date both check out. The only blemish is a single clarification-level cite-target mismatch on the secondary-endpoint sentence, which a public corrections note covers honestly. APPROVE_WITH_CORRECTIONS.