Fore Biotherapeutics · Regulatory · plixorafenib (PLX120-03 / FORTE)
Consolidated Citation Register & Traceability Matrix
Scope: all sources across the five IND-summary documents
Prepared: 11 Aug 2026
Status: Working draft — reconciliation pass
Classification: Internal — Confidential
About this format — traceability matrix for CTD 2.5.1
Under ICH M4E(R2), every claim in the Module 2.5 Clinical Overview must be traceable to source. §2.5.1 is short and argumentative by design, so it cannot display its own evidence base — which means the corpus behind it has to be reconcilable on demand. This register is that reconciliation.
One row per distinct source, with: its identifier, which document(s) cite it, which 2.5.1 claim it supports, and verification status. Sources that support no 2.5.1 claim are marked Unassigned rather than dropped — an unassigned source is either a gap in the argument or a source that belongs in 2.7.3/2.7.4, and both need a decision.
⚠ Decision support — not medical or regulatory advice. Verification status is carried verbatim from the source documents: ✓ verified against a primary source; ⚠ real but requires a manual primary-source pull before external or investor-facing use. Medical + regulatory HITL review required before any external use.
0 · What the reconciliation found
The 2.5.1 draft was built from four derived documents, not from the corpus. The DLGNT & PXA report carries 96 citation-bearing elements — 78 listed reference entries across §3 and §5.1–5.6, plus 9 gap-table rows and 9 detail modals with exact-source citations — resolving to 59 distinct machine-readable identifiers (40 PMIDs, 16 NCT numbers, 3 DOIs) plus agency documents and conference abstracts that carry none. The first 2.5.1 draft carried 2 of those 59. Across all four derived documents, 3.
Some compression is correct — 2.5.1 is an argument, not a bibliography. What was not correct: §5's second-pass material (regulatory designations, 16 trial registrations, conference readouts, and the 26-item competitive & resistance landscape) never entered the pipeline at all, so §2.5.1.3.3 and §2.5.1.5 were written from a subset while better evidence sat unused in the corpus. Those two sections have been rebuilt.
0.1 · Material corrections this pass produced
| Item | First 2.5.1 draft said | Corpus says — and what now stands |
|---|
| Tovorafenib efficacy |
ORR 67%, mDOR 16.6 mo (single figure, Kilburn Nat Med) |
The approval figures are ORR 51%, mDOR 13.8 mo (FDA accelerated approval, 23 Apr 2024, FIREFLY-1). The 67%/16.6-mo pair is a different cut of the same trial. Quoting the higher pair as "the benchmark" overstates the bar and is checkable against the label. Both now shown with their sources. |
| Plixorafenib regulatory status |
Absent entirely |
Fast Track (Sep 2022) · Orphan Drug (Mar 2023, primary brain & CNS malignancies) · Breakthrough Therapy (1 Apr 2026, adult BRAF-V600E high-grade glioma) — reported as the first BTD for a targeted therapy in HGG. This is the single strongest regulatory fact in the corpus and it was missing from the rationale. |
| Regulatory template |
Only the tovorafenib accelerated-approval precedent |
FDA tissue-agnostic BRAF-V600E solid-tumour approval (dab+tram, 22 Jun 2022) — the actual histology-agnostic template for a molecularly-defined programme, and the closest structural precedent for FORTE. Was absent. |
| DLGNT unmet need |
Argued from outcome data (5-yr PFS 15.9%) |
Corpus adds a white-space finding: no interventional RAF/MEK trial names DLGNT in eligibility (only a terminated CSF-ctDNA study, NCT05934630), while PXA is named across dab+tram/MEK trials. Stronger and more concrete than the survival argument. |
| DLGNT targeted-therapy evidence |
Not addressed |
Campanelli 2025: trametinib responses in DLGNT are real but not durable — all progressed <2 years. Directly supports the need for a different mechanism rather than more MEK inhibition. |
| Competitive set |
Two approved agents only |
Adds the CNS-penetrant competitor cohort: belvarafenib CNS Ph2 (NCT07688356) and PF-07799933 (Pfizer, brain-penetrant pan-mutant BRAF, same paradox-breaker thesis, in-brain responses reported). Plixorafenib sits in a three-agent CNS cohort, not a two-agent field. |
| Plixorafenib CNS denominators |
"6 of 9 evaluable" |
The corpus holds multiple cuts with different denominators — SNO 2023 CTNI-76: 6/10 (60%), mPFS 34.1 mo; SNO 2024 TRLS-02: n=113, V600E ORR 67%, mDOR 13.9 mo. See §3 — this must be resolved before any external use. |
| PXA guideline position |
"No approval covers PXA" |
Still true, but incomplete: Rudà 2022 EANO/EURACAN/SNO guideline positions BRAF/MEK inhibitors for BRAF-altered PXA. Guideline support without approval is a stronger and more precise statement of the gap. |
1 · The register
Claim codes: C1 disease definition & classification · C2 epidemiology & natural history · C3 current therapy & limitations · C4 unmet need · C5 scientific rationale / mechanism · C6 regulatory precedent · C7 competitive landscape · U unassigned to any 2.5.1 claim. Documents: R DLGNT/PXA report · G GBM & eGBM overview · K GBM readiness · P pilocytic astrocytoma · O 2.5.1 rationale.
1.1 · Disease background — DLGNT
| Source | Identifier | Docs | Claim | Status & note |
|---|
| Rodriguez FJ, et al. Acta Neuropathol 2012;124(5):627–41 | PMID 22941225 | R, O | C1 C2 | ✓ Entity description; n=36, 24M:12F; anaplastic progression 8/36 |
| Deng MY, et al. Acta Neuropathol 2018;136(2):239–53 | PMID 29766299 | R, O | C2 | ✓ MC-1 vs MC-2; 5-yr OS 100% vs 43% |
| Mikkelsen MK, et al. Acta Neuropathol 2025;150(1):18 | PMID 40788548 | R, O | C2 C4 | ✓ n=30; 5-yr OS 83.3%±8.8 vs PFS 15.9%±8.0; KIAA1549::BRAF 96% |
| Chiang J, et al. Acta Neuropathol 2022;144(6):1185–87 | PMID 36175668 | R | C2 | ⚠ 1q gain adverse — contested; not independent of Mikkelsen 2025 |
| Jiang H, et al. Front Oncol 2022;12:970076 | PMID 36185310 | R | C2 | ✓ n=145/43 studies; M:F 1.7/1 — the pooled sex-ratio anchor |
| Campanelli 2025, Acta Neuropathol Commun 13:208 | DOI 10.1186/s40478-025-02100-1 | R | C3 C4 | ✓ Newly promoted this pass. Trametinib responses real but all progressed <2 y |
| Chun BM, et al. The Oncologist 2025;30(5):oyaf093 | PMID 40377441 | R, O | C5 | ✓ Case report — tovorafenib response in KIAA1549::BRAF DLGNT. Most on-point targeted precedent; single case |
| Lee C, et al. Cancers 2026;18(6):912 | DOI 10.3390/cancers18060912 | R | U | ✓ Pooled IPD review — unassigned; resection/PFS-OS data could support C3 |
| Manoharan N, ANC 2021;9(1):147 · Messiaen J, Acta Neuropathol 2022;144(3):585 · Castellano-Damaso, Front Oncol 2024;14:1381354 · Selt F, J Neurooncol 2020;149(3):499 | PMID 34493325 · 36107235 · 38846974 · 33026636 | R | U | ✓ Genomics, PDX model, MEK-inhibitor durability. Unassigned — Selt/Castellano-Damaso support C3 (off-treatment rebound) |
| de los Reyes-Nabhan 2024 · Witten 2024 · Torres-Rey 2023 · De Carli 2026 · Giantini Larsen 2023 (SNO METB-07) | PMID 38404404 · 39073721 · 38034149 · Cancers 18(4):549 | R | U | ✓ Adult-onset, morphology, MMR, CSF cfDNA (KIAA1549::BRAF 5/7). Unassigned |
1.2 · Disease background — PXA
| Source | Identifier | Docs | Claim | Status & note |
|---|
| Louis DN, et al. Neuro-Oncol 2021;23(8):1231–51 | PMID 34185076 | R, G, K, O | C1 | ✓ WHO 2021 — the classification anchor across four documents |
| Giannini C, et al. Cancer 1999;85(9):2033–45 | PMID 10223246 | R, O | C2 | ✓ Mean age 26±16 (n=71) |
| Phillips JJ, et al. Brain Pathol 2019;29(1):85–96 | PMID 30051528 | R | C1 | ✓ CDKN2A/B 18/19 (95%) |
| Dampier CH, et al. Neuro-Oncol Adv 2025;7(1):vdaf089 | DOI 10.1093/noajnl/vdaf089 | R, O | C1 | ✓ n=469 — V600E 86%, CDKN2A/B del 87%, TERT 22% |
| Vaubel R, et al. Brain Pathol 2021;31(1):20–32 | PMID 33368753 | R, O | C2 | ✓ ~21% anaplastic at diagnosis; 5-yr OS 80.8% vs 47.6% |
| Ida CM, et al. Brain Pathol 2015;25(5):575–86 | PMID 25318587 | R, O | C2 | ✓ Anaplasia 33/74 (44.6%) |
| Vizcaino 2024 | PMID 39042515 | R | C5 | ✓ CDKN2A homozygous deletion >90% of PXA — anchors the CDK4/6-combination rationale |
| Vaubel 2018 | PMID 28181325 | R | C5 | ✓ CDKN2A/B the dominant recurrent CNA |
| Rudà 2022 EANO/EURACAN/SNO guideline | PMID 35908833 | R | C3 C4 | ✓ Newly promoted. Guideline positions BRAF/MEK inhibitors for BRAF-altered PXA — guideline support without approval |
| Zuo 2023 · Sullivan 2024 | PMID 37448517 | R | C2 | ✓ High-grade PXA 5-yr OS 51.5%; grade-3 aggressive behaviour |
| Schindler 2011 · Weber 2007 · Ma 2018 · Mahajan 2022 | PMID 21274720 · 16909113 · 30240866 · 35031693 | R | U | ✓ / ⚠ V600E frequency, original CDKN2A description, TERT. Mahajan "26.3" unverifiable — do not cite |
1.3 · Disease background — GBM, eGBM & pilocytic astrocytoma
| Source | Identifier | Docs | Claim | Status & note |
|---|
| Ostrom QT, et al. CBTRUS. Neuro-Oncol 2021;23(Suppl 2) | PMID 34608945 | G, K, O | C2 | ✓ Incidence 3.23/100,000; median age 65; M:F 1.6:1. Pre-WHO-2021 entity — caveat required |
| Guo X, et al. Front Oncol 2023;13:1200815 | PMID 37483487 | G, K, O | C1 C2 | ✓ n=191; TERTp 74.7%, EGFR amp 57.5%, +7/−10 94.8%; mOS 12.6 mo |
| Behling F, et al. Diagn Pathol 2016;11:55 | PMID 27350555 | G, K, O | C1 C5 | ✓ GBM V600E 3/312 (0.96%); all three epithelioid |
| Kleinschmidt-DeMasters BK, et al. Am J Surg Pathol 2013;37(5):685–98 | PMID 23552385 | G, O | C1 C5 | ✓ eGBM V600E 7/13 (~54%); 0/9 giant-cell, 0/2 rhabdoid |
| Wang S, et al. Int J Clin Exp Pathol 2020;13(7):1529–39 | PMID 32782671 | G, O | C2 C4 | ✓ n=33; median age 36; mOS 10 mo; 5-yr 0%. Its 100% V600E is an outlier — do not quote |
| Korshunov A, et al. Brain Pathol 2018 | PMID 28990704 | G, K, O | C1 | ⚠ Read via abstract only. eGBM does not cluster separately — governs all eGBM framing |
| McNulty SN, et al. Sci Rep 2021;11:19999 | PMID 34625582 | G, K | U | ✓ Explicitly no eGBM cases; n=5 BRAF, one V600E; survival from n=3. Unassigned — cannot support a 2.5.1 claim |
| Lassaletta A, et al. J Clin Oncol 2017;35(25):2934–41 | DOI 10.1200/JCO.2016.71.8726 | P, O | C2 C4 | ✓ n=510; 10-yr PFS 27% (12.1–41.9) vs 60.2% (53.3–67.1); GTR recurrence one-third |
| Mistry M, et al. J Clin Oncol 2015;33(9):1015–22 | — | P | C2 | ✓ Transformation 26/866 (2.9%); V600E 39% and CDKN2A del 57% of secondary HGG |
| Ater JL, et al. J Clin Oncol 2012;30(21):2641–47 (COG A9952) | — | P, O | C3 | ✓ 5-yr EFS 39%±4% vs 52%±5%, P = 0.10 — not significant |
| PDQ Childhood Astrocytomas (NCI), upd. 14 Apr 2025 | — | P, O | C2 C3 | ✓ Verified verbatim. Its three V600E findings trace to three different PDQ refs [24][26][27]; [27] primary unidentified |
| de Blank P, et al. Curr Opin Pediatr 2019;31(1):21–27 | — | P, O | C2 | ✓ 10-yr OS >85–90% — a pediatric LGG figure, not PA-specific |
| StatPearls, Glioblastoma Multiforme | NBK558954 | G, O | C2 | ⚠ Review-grade. Sole source for GBM localization, imaging and 5-yr survival. Replace before external use |
1.4 · Regulatory precedent — entire block was absent from the first 2.5.1 draft
| Item | Fact | Claim | Status & note |
|---|
| Plixorafenib — Fast Track | Sep 2022, BRAF Class 1/2 | C6 | ✓ |
| Plixorafenib — Orphan Drug | Mar 2023, primary brain & CNS malignancies | C6 | ✓ |
| Plixorafenib — Breakthrough Therapy | 1 Apr 2026, adult BRAF-V600E high-grade glioma — reported as first BTD for a targeted therapy in HGG | C6 C5 | ✓ Strongest regulatory fact in the corpus. ⚠ Strategic flag: BTD is adult HGG while the programme is positioned pediatric |
| Plixorafenib — designation gaps | No Rare-Pediatric-Disease designation; no EMA designation found | C6 | ⚠ Flag — inconsistent with a pediatric positioning |
| Tovorafenib (OJEMDA) — FDA | Accelerated approval 23 Apr 2024; pediatric LGG with BRAF fusion/rearrangement or V600; ORR 51%, mDOR 13.8 mo | C3 C6 C7 | ✓ The approval figures — use these as "the bar," not the 67%/16.6-mo cut |
| Tovorafenib — EMA | Conditional MA 22 Apr 2026 (Ipsen); confirmatory FIREFLY-2 | C6 C7 | ✓ |
| Dab+tram — FDA pediatric LGG | 16 Mar 2023, BRAF-V600E; ORR 46.6% vs 10.8% | C3 C6 C7 | ✓ |
| Dab+tram — FDA tissue-agnostic | 22 Jun 2022, BRAF-V600E solid tumours | C6 | ✓ The histology-agnostic template — the closest structural precedent for FORTE. Was absent from the rationale |
| Dab+tram — EMA | Finlee/Spexotras, 15 Nov 2023, with an explicit pediatric HGG V600E indication | C6 C7 | ✓ |
| Pathway precedents | RACE for Children Act (BRAF on FDA pediatric molecular-target list); tissue-agnostic framework; EMA PIP + conditional-MA criteria | C6 | ✓ |
| Rare-Pediatric-Disease PRV | Programme authorisation sunset ~Dec 2024; reauthorisation pending | C6 | ⚠ Verify current status before relying on a voucher in any plan |
| Vemurafenib | No CNS/glioma indication (melanoma/ECD only) | C7 | ✓ |
1.5 · Clinical-trial landscape — 16 NCT registrations, none carried into the first draft
| Programme | Registrations | Claim | Status & note |
|---|
| Plixorafenib | FORTE NCT05503797 (Ph2 pivotal; Sub-A BRAF-fusion CNS, Sub-B V600E glioma/glioneuronal, ≥8 y) · NCT02428712 (Ph1/2a, completed) · NCT06610682 (± retifanlimab, CSF-ctDNA endpoint) | C5 C6 | ✓ DLGNT and PXA are captured molecularly, not by histology name — relevant to how the label could read |
| Tovorafenib | FIREFLY-1 NCT04775485 · FIREFLY-2/LOGGIC NCT05566795 (Ph3 front-line vs chemo) · NCT07206849 (HGG/DIPG) · NCT06381570 · NCT05465174 · NCT05828069 · NCT05760586 | C7 | ✓ FIREFLY-2 moves tovorafenib front-line — the competitive trajectory, not just current label |
| Dab+tram | NCT02684058 (PXA explicitly named in eligibility) · NCT03919071 (HGG; anaplastic PXA named) · NCT03975829 | C3 C7 | ✓ Qualifies the "no approval covers PXA" claim — PXA patients are already enrollable elsewhere |
| CNS competitors | Belvarafenib NCT07688356 (Ph2, BRAF-altered brain tumours) · PF-07799933 NCT05355701 (± binimetinib) | C7 | ⚠ Belvarafenib trial investigator-initiated, 2025-registered. Closest CNS competitors |
| DLGNT white space | No interventional RAF/MEK trial names DLGNT in eligibility; only terminated NCT05934630 (CSF-ctDNA) | C4 | ✓ Promoted to the unmet-need argument this pass |
1.6 · Efficacy evidence — plixorafenib & comparators
| Readout | Result | Claim | Status & note |
|---|
| Plixorafenib SNO 2023 (CTNI-76) | V600+ primary CNS n=10 — ORR 60% (6/10); HGG 67%; mPFS 34.1 mo | C5 | ✓ PMC10639671. mPFS 34.1 mo is the most striking published figure |
| Plixorafenib SNO 2024 (TRLS-02) | n=113 incl. 5 children; V600E ORR 67%, mDOR 13.9 mo; no pediatric DLTs | C5 | ✓ PMC11182979 |
| de la Fuente 2024, Neuro-Oncology 26(Suppl 4) | CNS cut — 9 evaluable MAPKi-naïve V600 PCNST | C5 | ✓ See §3 — denominators differ across readouts |
| de la Fuente 2023, JCO 41:3006 | V600 MAPKi-naïve PR 39%, mDOR 32 mo | C5 | ✓ |
| Rosen E, JCO 2025;43:TPS2091 | FORTE design | C6 | ✓ Trial-in-progress |
| Kilburn LB, Nat Med 2024;30(1):207–17 | FIREFLY-1 — ORR 67%, mDOR 16.6 mo | C7 | ✓ PMID 37978284. Different cut from the 51%/13.8-mo approval figures — never mix |
| Singh S, Clin Cancer Res 2025;31(8):1383–89 | FDA approval summary, tovorafenib | C6 C7 | ✓ PMID 39808502 |
| Bouffet E, NEJM 2023;389(12):1108–20 | Pivotal RCT, dab+tram vs chemo, pediatric LGG V600 | C3 C7 | ✓ PMID 37733309. Inactive against fusions |
| Wen PY, Lancet Oncol 2022;23(1):53–64 (ROAR) | Dab+tram in V600E low- and high-grade glioma | C3 C7 | ✓ PMID 34838156 — the adult glioma comparator |
| Hargrave DR, JCO 2023;41(33):5174–83 | R/R V600 pediatric HGG — ORR 56%, mDOR 22.2 mo | C7 | ✓ PMID 37643378 |
| Kaley T, JCO 2018;36(35):3477–84 (VE-BASKET) | Vemurafenib in BRAF-V600 gliomas | C3 | ✓ PMID 30351999 |
| Brown NF 2017 · Chamberlain 2013 | Dabrafenib in V600 anaplastic PXA; salvage BRAF inhibitors in recurrent PXA | C3 C4 | ✓ PMID 27781490 · 23756727. The only PXA-specific targeted evidence — case-series level |
| Tovorafenib SNO 2024 (CTNI-09) · Nysom 2025 | Drug-holiday: 24/26 stayed off treatment; optic-pathway subgroup | C7 | ✓ PMID 39700439. Durability-after-stopping is a competitive differentiator to answer |
| ASCO 2022 LBA2002 · AACR 2025 CT247 · AACR 2026 CT154 · ASCO 2026 abstr 10031 | LGG debut; FORTE design; V600 CRC ORR 8%; pediatric MAPKi long-term FU | C7 | ⚠ Numbers unverified — do not quote externally without a primary pull |
1.7 · Mechanism & resistance
| Item | Content | Claim | Status |
|---|
| Class mechanism | Type I (dabrafenib, vemurafenib — V600 monomer only, paradoxically activates fusions) → type II / pan-RAF (tovorafenib — dimers + V600) → paradox-breaker (plixorafenib — BRAF-dimer-selective, fusions + V600, monotherapy) | C5 | ✓ The through-line of the entire rationale |
| Pan-RAF competitor pack | Exarafenib/KIN-2787, naporafenib/LXH254, belvarafenib, lifirafenib — anchored in NRAS melanoma, not CNS, and mostly need a MEK partner | C7 | ✓ Supports the CNS-monotherapy differentiation |
| PF-07799933 (Pfizer) | Brain-penetrant pan-mutant BRAF on the same paradox-breaker thesis; in-brain responses in RAF-refractory patients | C7 | ✓ Yaeger, Cancer Discov 2024, PMC11372368. Principal watch-item |
| Resistance mechanisms | Paradoxical MAPK activation (type II solves) · CDKN2A/B loss → CDK4/6 combinations · MAPK/dimer reactivation · PK/CNS exposure gating durability | C5 | ✓ CDKN2A/B loss links directly to PXA's >90% deletion rate (§1.2) |
| Positioning | Three-agent CNS-penetrant cohort (tovorafenib, plixorafenib, PF-07799933); moat = dimer-selective fusion + V600 coverage as monotherapy with proven in-brain exposure | C5 C7 | ✓ Best evidenced by CSF-ctDNA study NCT06610682 |
2 · Unassigned sources — decisions required
Sources in the corpus that currently support no 2.5.1 claim. Each is either a gap in the argument or belongs in 2.7.3/2.7.4 — neither is a reason to leave it undecided.
| Source group | Recommended disposition | Reason |
|---|
| Lee 2026 pooled IPD review (DLGNT) | Promote to C3 | Resection-vs-outcome data would strengthen the surgical-limitation argument for DLGNT |
| Selt 2020 · Castellano-Damaso 2024 (MEK inhibitors) | Promote to C3 | High on-therapy control with off-treatment rebound — a limitation of current targeted options, and it corroborates Campanelli 2025 |
| Messiaen 2022 (DLGNT PDX/cell line) | Move to Module 4 / nonclinical | Preclinical model, not clinical rationale |
| Giantini Larsen 2023 (CSF cfDNA, KIAA1549::BRAF 5/7) | Promote to C5 | Supports CNS-compartment biomarker feasibility alongside NCT06610682 |
| McNulty 2021 | Retire from 2.5.1 | n=5 BRAF / one V600E / survival from n=3, and explicitly no eGBM cases. Cannot bear a rationale claim |
| Schindler 2011 · Weber 2007 · Ma 2018 | Retain as background only | Foundational frequency papers; superseded by Dampier 2025 for citable figures |
| Mahajan 2022 | Do not cite | The "26.3" figure could not be verified; paywalled and absent from the abstract |
| Witten 2024 · Torres-Rey 2023 · De Carli 2026 · de los Reyes-Nabhan 2024 | Retain for 2.7 / disease background | Morphologic and adult-onset detail below the resolution of 2.5.1 |
3 · Contested & unresolved — must clear before submission use
| Issue | Severity | Detail & required action |
|---|
| Plixorafenib CNS denominators | High | The corpus carries at least three cuts: 6/10 (60%) SNO 2023 CTNI-76 · 9 evaluable de la Fuente 2024 CNS abstract · n=113, ORR 67% SNO 2024 TRLS-02. Which population any quoted ORR refers to must be fixed and stated with its denominator. A reviewer comparing two Fore documents that quote different "67%"s will ask. |
| Tovorafenib benchmark figure | High | 51% / 13.8 mo (label) vs 67% / 16.6 mo (Kilburn). Use the approval figures when describing the competitive bar; cite Kilburn only with its cut identified. |
| BTD is adult HGG | Strategic | Breakthrough Therapy designation is for adult BRAF-V600E HGG, while the programme is positioned pediatric and FORTE enrolls ≥8 y. No Rare-Pediatric-Disease designation found. Positioning and designation strategy should be reconciled deliberately. |
| DLGNT 1q gain | Medium | Chiang 2022 vs Mikkelsen 2025 — not independent (shared senior authorship; up to ~12 of 30 patients may overlap). Do not present as two independent cohorts. |
| eGBM entity status | Medium | Not a WHO 2021 entity; merges with PXA on methylation. Any eGBM subgroup needs an operational definition distinguishing it from anaplastic PXA. |
| "No approval covers PXA" | Medium | True, but PXA is explicitly named in dab+tram trial eligibility (NCT02684058, NCT03919071) and guideline-supported (Rudà 2022). Restate as "guideline-supported and trial-accessible but not approved." |
| GBM localization / imaging; PA imaging, incidence, sex | Medium | Review-grade or absent. Replace with primary/registry sources. |
| PDQ ref [27] | Low | Diencephalic 22% vs 52% traceable only through PDQ. Identify the primary. |
| ⚠-flagged conference items | Blocking for external use | AACR 2025 CT247, AACR 2026 CT154, ASCO 2026 abstr 10031, belvarafenib registration, PRV sunset status. Agency and publisher pages blocked automated fetch — manual pull required. |
| Embargo control | Blocking | The SNO-2026 CNS reinterpretation is embargoed and must not enter 2.5.1 or any external text. Published abstracts only. |
Method & limits. Sources were extracted programmatically from all five documents (reference lists, gap-table rows, detail modals) and reconciled by identifier — PMID, PMCID, DOI, NCT — then mapped by hand to the 2.5.1 claim each supports. Counts in §0 are machine-derived: 96 citation-bearing elements, 78 listed reference entries, 59 distinct identifiers. Verification marks (✓ / ⚠) are carried verbatim from the source documents and were not re-verified in this pass; ⚠ items remain blocked for external use on the original grounds (agency and conference publisher pages blocked automated fetch). This register records provenance and traceability — it does not re-adjudicate the underlying science. Where two documents state different figures for the same result, both are shown with their cut identified rather than reconciled to one number.