Provenance Record

Strong, disciplined News piece on a negative/mixed Phase 3 readout. Standard Overview / What We Know / What Comes Next / What We Don't Know structure, 642 words, appropriate for the News category. The headline and lead correctly center the OS miss in the full SigVie-002 (Be6A Lung-01) NSCLC population and characterize the one-prior-line subgroup only as a 'stronger trend' and a salvageable signal, not as a win.

Personally read all three gzipped snapshots from disk (sources/2026-06/pfizers-sigvotatug-vedotin-misses-overall-survival-goal-in-phase-3-lung-cancer-trial-salvaging-a-signal-in-second-line-patients/). All three returned HTTP 200 and the uncompressed sha256 of each file matches manifest.json. source-0 (Pfizer press release): confirmed verbatim the SigVie-002/Be6A Lung-01 naming, 'investigational, potential first-in-class integrin beta-6 (IB6) directed antibody-drug conjugate (ADC)', docetaxel comparator, 'did not show a statistically significant improvement in the primary endpoint of overall survival (OS) compared to docetaxel', the 'stronger trend was observed for OS and progression-free survival (PFS)' in the one-prior-line subgroup that 'represents two-thirds of the study population', the full Jeff Legos and Solange Peters quotes verbatim, 'the safety profile of sigvotatug vedotin was manageable and consistent with prior studies', 'Detailed results from SigVie-002 will be submitted for presentation at a future medical congress', the 1L pembrolizumab Phase 3 (PD-L1 TPS >=50%), the PD-1/VEGF bispecific combination, and the Monday June 22, 2026 dateline. source-1 (BioSpace, in allowlist): confirmed the OS-miss quote verbatim. source-2 (Investing News): confirmed 'is expressed on approximately 90% of tumors', 'associated with poor prognosis', the Seagen-acquisition ADC-portfolio sentence verbatim, and 'fetrastobart vedotin, a PD-L1-directed ADC currently in Phase 3 in NSCLC'.

No fabricated efficacy numbers. The topline release deliberately did not disclose hazard ratios, confidence intervals, or magnitude of survival difference, and the article does not invent any — it correctly flags them as undisclosed in 'What We Don't Know'. The press release discloses enrollment (703) and the NCT number, which the article opts not to cite; stating enrollment size as undisclosed is a minor imprecision but not material to any claim (no number is overstated). Rule 9 satisfied: the headline topline result, the IB6 ADC / Seagen-origin framing, and the docetaxel comparator all rest on the non-blocked Pfizer press release.

APPROVE. Accurate, well-sourced reporting of a negative/mixed Phase 3 result with no overstatement and no fabricated figures. The sole finding is a non-blocking allowlist note; no content correction is warranted, so this is a clean APPROVE rather than APPROVE_WITH_CORRECTIONS.