Curation log — Alzheimer's disease¶

Newest entries first. Every content-changing session appends an entry: date, what changed, what was searched, follow-ups for next time. See CLAUDE.md.

2026-10-27 — Periodic literature sweep (Claude)¶

Searches and source boundary. The search manifest records every query string, its PubMed translation, the IDs returned, the Europe PMC records and the registry checks. I queried PubMed E-utilities for publications dated 2026-10-20 through 2026-10-27. That covered one broad AD query (195 records) and the same 12 topic searches as the last two sweeps, which gave 282 unique records to screen by title. Each topic also had a dated "Cochrane Database Syst Rev"[jour] search, plus an all-AD Cochrane search by publication date and another by Entrez date. All Cochrane searches returned zero records, so no Cochrane review appeared in this window. Europe PMC REST was queried across the same 12 topics and the preprint source (67 preprints). It returned 84 distinct records with no PMID, almost all preprints, and none was cited, so no DOI-only record was added. Three errata appeared in the window (PMID 42913447; PMID 42909823; PMID 42905858). None corrects a paper this condition cites. Every one of the 14 additions was checked against PubMed publication types and Crossref updated-by, and none carried a retraction or expression of concern. I skipped Unpaywall: LIFEBANK_CONTACT_EMAIL was unset, and no contact address was read or sent anywhere. Cochrane CENTRAL and SciELO/LILACS were not searched (HTTP 403). WHO ICTRP, EU CTIS, ChiCTR and CTRI were not queried this week, because previous sweeps found them unusable programmatically. An absence reported here is an absence in PubMed, Europe PMC, ClinicalTrials.gov and ISRCTN only.

Trials. I queried all 58 previously cited NCT IDs individually through ClinicalTrials.gov API v2, plus NCT06343636 and the newly cited NCT02547818. No status changed and no tracked trial posted first results. Three made routine registry updates with unchanged status: NCT06566170 (donanemab real-world study, still recruiting), NCT06957418 (Alzheimer's Tau Platform master protocol, recruiting) and NCT07167966 (AADvac1 regimen, enrolling by invitation). Diranersen remains ACTIVE_NOT_RECRUITING with no posted results; its primary completion was 2026-03-11 (NCT05399888). The windowed PubMed diranersen query and the VG-3927 query returned no record that the 2026-10-20 sweep had not already seen. ISRCTN (condition:Alzheimer, 103 records) had three records updated in the window. None is tracked on a wiki page, and the ELAD liraglutide update (ISRCTN89711766) does not change the published result on the trials page. The AD-SMART record has no update in the window; its BMJ news item (PMID 42914149) was not added.

Official and patient-organisation checks. NICE's lecanemab (GID-TA11220) and donanemab (GID-TA11221) appraisals were re-fetched: both are still "In development", publication date TBC. The Alzheimer's Association's latest visible news item is still dated 6 October 2026. ADI news returned HTTP 403. The Alzheimer's Society news page fetched, but no dated item could be parsed from it. No patient-voice material survived triage.

What survived triage. Fourteen records went into the bibliography (391 → 405). Nine reached a wiki page or the literature layer: - UK guidance (Habash-Bailey 2026, PMID 42914559). The Association of British Neurologists and the Royal College of Psychiatrists issued joint guidance on blood-biomarker testing. PubMed has no abstract and the publisher returned HTTP 403, so the guidance is registered and listed on guidelines.md with no recommendation attributed to it. It is also on the registry watch list and in OQ-12. - Taiwanese hospital-caregiver guideline (Shen 2026, PMID 42914691). Added as a registry row only, from the abstract. - p-tau217 discordance (Shir 2026, PMID 42821845), verified against the PMC full text. At Lumipulse cut-points, 8.5% of results were discordant; false negatives outnumbered false positives (5.6% vs 2.8%), and eGFR was independently associated with both error types. Added to fluid-biomarkers.md, STATISTICS and OQ-31. The page now sets this beside, rather than over, Arslan 2025's finding that eGFR adjustment did not improve prediction, because the two studies asked different questions. The full text gives some univariate ORs slightly differently from the abstract; only the age-adjusted estimates, which match in both, were used. - Whitehall II (Sabia 2026, PMID 42913444), abstract only. For 12-year all-cause dementia, PPV was 16.4–23.5% and NPV 92.3–94.7%. A risk-factor model reached C-index 0.765, rising to 0.813 with biomarkers. Added to fluid-biomarkers.md, STATISTICS and OQ-3. - Real-world lecanemab (Wu 2026, PMID 42907679), abstract only. A 37-study meta-analysis: any ARIA 15.1%, with US studies at 19.5% and Asian studies at 9.7%, and an APOE ε4 gradient. Added to anti-amyloid-immunotherapy.md and STATISTICS. - ALZT-OP1 (Elmaleh 2026, PMID 42920249; NCT02547818), abstract only. The phase 3 COGNITE trial of cromolyn plus ibuprofen missed its primary endpoint. Registry primary completion was 2020; primary results were published in 2026. - Low-dose whole-brain radiotherapy (Hong 2026, PMID 42906637), abstract only. A phase 2 sham-controlled trial whose primary endpoint was not met. Both trials were added to clinical-trials-landscape.md. - Brazilian reclassification study (Botelho 2026, PMID 42911077), abstract only. Biological AD rose from 27.4% under AT(N) 2018 to 49.3% under AA-2024, and only 38% of reclassified patients received a final clinical AD diagnosis. Added to diagnostic-criteria-and-biological-definition.md. - Medicare detection trends (Qian 2026, PMID 42923120), abstract only. Added to STATISTICS.

Five records are bibliography-only: Barcellos 2026 on causal education effects (PMID 42912790; the abstract has no effect sizes and the full text returned HTTP 403), van der Veere 2026 (microsimulation; PMID 42928321), Abu-Rumeileh 2026 (brain-derived tau; PMID 42918778), Oberstein 2026 (serum β-synuclein; PMID 42904923) and Lin 2026 (Taiwanese p-tau217 thresholds; PMID 42905283). The other ~265 windowed records were reviews, preclinical work, methods papers, letters, non-AD dementias or incremental cohorts. No landmark note was written.

Files changed. Wiki: fluid-biomarkers.md, anti-amyloid-immunotherapy.md, clinical-trials-landscape.md, diagnostic-criteria-and-biological-definition.md and guidelines.md, each with last-curated: 2026-10-27. All five were already draft, no curated page was edited, and the count stays at 10 curated / 10 draft. Literature layer: BIBLIOGRAPHY.md (405), guidelines/REGISTRY.md (rows 46–47, provenance note, a new 2.3 entry, a watch-list item), statistics/STATISTICS.md (five rows) and OPEN-QUESTIONS.md (OQ-3, OQ-12, OQ-31 sharpened; none resolved or newly opened). INDEX.md was reconciled, and its NCT count is now 60 because NCT00235716 is cited only in this log. Validation: every PMID cited outside this log is in the bibliography, no relative link is broken, and no file outside conditions/alzheimers-disease/ was touched.

Decisions. (1) Six of the nine wiki- or literature-level claims rest on complete PubMed abstracts, and each is labelled "abstract only" where it appears. Only Shir 2026 was checked against full text. (2) The ABN/RCPsych guidance was recorded but not synthesised, because attributing recommendations from a title alone would be inventing them. (3) The Korean aducanumab ARIA series (PMID 42908770), p-tau217 assay-harmonisation and stability papers, and the FDA subcutaneous-starting-dose news item (PMID 42911210; already covered by the 13 July 2026 FDA entry) were screened and not added.

Next sweep. Get the ABN/RCPsych full text (PMID 42914559), for example through Europe PMC once it is indexed or an institutional route, and synthesise its primary-care position into guidelines.md and OQ-12. Get full text for Sabia, Wu, Elmaleh and Botelho before treating their numbers as audited. Check whether the ALZT-OP1 registry posts results. Recheck NICE, Diranersen results posting (NCT05399888) and VG-3927 phase 2 registration. Re-run the dated absences in OQ-25 and OQ-27; this week's windowed OQ-27 qualitative query returned 0 PubMed records. The ten draft pages still need independent audit by a different engine.

2026-10-20 — Periodic literature sweep (Grok)¶

Searches and source boundary. The search manifest records query strings, PubMed translations, returned IDs, Europe PMC records and registry checks. PubMed E-utilities was queried for publications dated 2026-10-13 through 2026-10-20: one broad AD query (242 records), 12 topic searches spanning diagnosis, epidemiology, genetics, amyloid/tau, glia/vascular/metabolic disease, fluid biomarkers, imaging, anti-amyloid treatment, symptomatic/neuropsychiatric care, prevention, care/patient experience and guidelines/trials, plus a dated Cochrane Database of Systematic Reviews search for each topic. The windowed Cochrane search returned two editorials (PMID 42878653; PMID 42878647) about the already-catalogued Nonino 2026 review, and no new systematic review. Europe PMC REST was queried across the same 12 topics plus a tau-personalised-endpoint query; 109 distinct records lacked a PMID at retrieval (PMC full texts and preprints). Titles were screened; no DOI-only record was added. OpenAlex returned 68 works in the date window (used as a cross-check, not as a citation source). DOAJ was queried but its ranking is not date-restricted and was not used. Unpaywall was skipped; LIFEBANK_CONTACT_EMAIL was not set and no personal address was read or transmitted. Crossref updated-by and PubMed publication types were checked for every bibliography addition; none carried a retraction. One Expression of Concern in the window (PMID 42887664, on Chu 2015 GSAP) is not a paper this condition cites.

All 58 previously cited NCT IDs were queried individually through ClinicalTrials.gov API v2, plus NCT06343636 (VG-3927 phase 1: COMPLETED, 115 actual, no posted results). Status labels on the trials page matched the live records. Two tracked trials posted routine updates without status change: NCT06566170 last update 2026-10-16 (still recruiting, 6,250 estimated) and NCT07717411 last update 2026-10-13 (still recruiting, 1,600 estimated). Diranersen phase 2 remains ACTIVE_NOT_RECRUITING, primary completion 2026-03-11, no posted results (NCT05399888). Named-registry checks: ISRCTN condition:Alzheimer (103 records; none last-updated on or after 2026-10-13; AD-SMART ISRCTN17108793 last updated 2026-09-23); EU CTIS public API HTTP 403; ChiCTR English search returned 405 or an undecodable payload; CTRI landing page fetched but is a login/declaration form, not a result set; WHO ICTRP portal fetched (JS-driven, no programmatic results). Cochrane CENTRAL and SciELO/LILACS were not searched (HTTP 403). An absence found in PubMed and Europe PMC is an absence in those sources.

Official-source checks: NICE lecanemab and donanemab appraisals remain in development, expected publication TBC, no project documents after 6 August 2026; NG97 unchanged (last reviewed 24 October 2025); EMA Leqembi and Kisunla EPARs fetched, authorisation dates unchanged; FDA at-home starting-dose page current as of 13 July 2026; AWMF S3 page fetched (thin JS shell); WHO dementia fact sheet last dated 3 July 2026. Patient-organisation news: Alzheimer's Association latest visible item 6 October 2026; ADI 1 October 2026 (new CEO) and 29 September 2026 (World Alzheimer's Month); Alzheimer's Society latest 9 October 2026. None in the 13–20 October window survived patient-voice triage.

Follow-up searches from the 2026-10-13 log: Molina-Henry, Ataka and Alten now have complete abstracts (still no PMC full text) and were moved onto wiki pages from those abstracts. Tau-personalised endpoint PubMed query → 11 records; Europe PMC → 2 (Biel 2022 and a 2026 opinion piece); ClinicalTrials.gov outcome searches did not find a trial re-scoring randomised arms with tau-map-derived personal composites. OQ-27 windowed qualitative query → 0 records. OQ-25 insurance/employment/capacity query → 34 records, none a longitudinal outcomes study.

What survived triage. Most of the 324 unique windowed PubMed records were reviews, methods papers, animal work or incremental cohorts and were not added. Ten records were added to the bibliography. Wiki-level: Italian EPA2 appropriate-use update (Benussi 2026, PMID 42880054); French memory-clinic eligibility 9.7% / ~18,200 per year (Le Scouarnec 2026, PMID 42874425); US utilisation 0.2% with racial and vulnerability gradients (Reynolds 2026, PMID 42879726); Tokyo real-world ARIA-H OR 14.0 with baseline microbleeds (Kujuro 2026, PMID 42892274); oral TREM2 agonist VG-3927 phase 1 (Dejanovic 2026, PMID 42873892; NCT06343636); ARIC midlife p-tau181 and 11-year memory decline (Smith 2026, PMID 42886876); capillary versus venous p-tau217 r=0.23 (Ho 2026, PMID 42885968); GLP-1 RA seven-RCT meta-analysis, null on ADAS-Cog and CDR-SB (Siqueira 2026, PMID 42890838). Three prior bibliography-only records were promoted onto pages: Molina-Henry AHEAD %p-tau217 by race/ethnicity (PMID 42862590); Ataka primary-care blood-biomarker disclosure (PMID 42862138); Alten EPVS-CSO predicting first haemorrhagic lesion (PMID 42855813). Two Cochrane editorials were catalogued only. No landmark deep note: nothing here replaces a pivotal trial or a current guideline family.

Files changed. Ten records into literature/BIBLIOGRAPHY.md (391 total). Wiki: guidelines.md, anti-amyloid-immunotherapy.md, clinical-trials-landscape.md, neuroinflammation-and-glia.md, fluid-biomarkers.md, care-caregiving-and-health-systems.md, vascular-and-metabolic-contributions.md, diagnostic-criteria-and-biological-definition.md. Literature layer: REGISTRY.md (45 rows), STATISTICS.md, OPEN-QUESTIONS.md (OQ-8, OQ-12, OQ-15, OQ-16, OQ-20, OQ-25, OQ-27, OQ-30, OQ-31, D7). INDEX.md reconciled. Eight wiki pages edited today are draft with last-curated: 2026-10-20; tau-biology and imaging remain draft from 2026-10-13 (ten draft in total). No patient-voice file or note was changed.

Decisions. (1) Wiki claims for Molina-Henry, Ataka, Alten, Benussi, Le Scouarnec, Reynolds, Kujuro, Dejanovic and Smith are taken from complete PubMed abstracts; none of those records had a usable PMC full text in this session. (2) PubMed attached wrong DOIs to several windowed records (including the GLP-1 meta-analysis and the capillary-collection paper); DOIs used here were resolved through Crossref by title. (3) NICE was re-fetched rather than inferred from the previous sweep.

Next sweep. Recheck Diranersen results posting (NCT05399888). Recheck NICE final appraisal status. Fetch full text of Benussi EPA2, Le Scouarnec, Reynolds, Molina-Henry, Ataka, Alten, Ho and Smith before treating any of those numbers as audited. Watch for a VG-3927 phase 2 registration. Re-run the tau-map personalised-endpoint absence on completed randomised arms. OQ-25 and OQ-27 remain dated absences. Independent audit of the ten draft pages.

2026-10-13 — Periodic literature sweep (Codex)¶

Searches and source boundary. The search manifest records the exact query strings, PubMed translations, returned IDs, Europe PMC records and individual registry checks. PubMed E-utilities was queried for publications dated 2026-10-07 through 2026-10-13: one broad AD query (147 records), 12 topic searches spanning diagnosis, epidemiology, genetics, amyloid/tau, glia/vascular/metabolic disease, fluid biomarkers, imaging, anti-amyloid treatment, symptomatic/neuropsychiatric care, prevention, care/patient experience and guidelines/trials, plus a dated Cochrane Database of Systematic Reviews search for each topic (zero reviews in that six-day window). A separate 2026 Cochrane anti-amyloid query found one April review that the prior build had missed. Europe PMC REST was queried across the same 12 topics and publication window; 68 distinct records lacked a PMID at retrieval, including full-text PMC records and preprints. Their titles and full-text flags were screened; none without a PMID was cited, so no DOI-only record was added. Full texts of the older tau-PET follow-up papers and the Cochrane review were read through Europe PMC.

All 58 NCT IDs already cited by this condition were queried individually through ClinicalTrials.gov API v2. One tracked trial posted a routine update on 2026-10-07, but its active/not-recruiting status matched the wiki. No tracked trial had a first results posting since the page's 2026-09-03 curation. A full comparison also exposed four stale labels from registry updates posted in September: the preclinical trontinemab trial and the donanemab/RG6289 combination now recruit, while both symptomatic trontinemab trials are active but not recruiting; the latter also report actual enrolments of 1,101 and 1,060, rather than 800 estimated each (NCT07717411; NCT06996730; NCT07169578; NCT07170150). Diranersen's phase 2 remains active/not recruiting, with actual primary completion on 2026-03-11 and no posted results (NCT05399888). Two ClinicalTrials.gov endpoint searches, plus checks of the named ISRCTN and EU CTIS registries, found trials measuring tau PET alongside global cognitive composites, but did not establish use of a personalised tau-map-derived randomised endpoint; this is a scoped retrieval result, not a worldwide absence claim. WHO ICTRP's JS portal, Cochrane CENTRAL and SciELO/LILACS were not searched. The Cochrane review itself searched CENTRAL through August 2025, which does not make it a database searched in this session. The regional portability questions remain limited by the SciELO/LILACS HTTP 403 barrier.

Official-source checks covered the German DGN–DGPPN/AWMF S3 guideline PDF, the full text of the Italian RAPID consensus, WHO's 2026 Global Health Estimates announcement, NICE's lecanemab appraisal page (still in development) and its donanemab page (direct fetch returned 403), plus the Alzheimer’s Association and Alzheimer's Disease International news listings. The latest visible patient-organisation items were dated 2026-10-06 and 2026-10-02, respectively; no October 7–13 item survived patient-voice triage. The WHO announcement was published before the PubMed window but after the statistics file's last curation. Crossref updated-by and PubMed publication types were checked for all eight bibliography additions; none carried a retraction or expression-of-concern signal. Unpaywall was skipped; no contact email was read or transmitted.

What survived triage. The missed Cochrane review (Nonino 2026, PMID 41985900) synthesised 17 antibody trials and 20,342 participants. Its full-text Summary of Findings table reports a pooled 18-month CDR-SB difference equivalent to 0.29 points (low certainty), ADAS-Cog 0.85 points (moderate certainty), and 107 additional ARIA-E cases per 1,000 (moderate certainty). Its practice conclusion says the main clinical differences are “well below the minimal clinically important difference.” The pooled class estimate includes seven drugs, so it was set beside, rather than substituted for, the lecanemab and donanemab pivotal estimates. The earlier audit's three tau-PET follow-ups were verified against full text: Biel's personalised composites improved sensitivity to simulated treatment effects (PMID 36345046); Malpetti's regional tau PET predicted domain-specific decline in trial cohorts (PMID 41053874); and Chao's two-brain PET–autopsy comparison found medial temporal underestimation by flortaucipir (PMID 42108381). The German S3 guideline's weak anti-amyloid recommendation and its DEGAM dissent were added from the official version 6.1 PDF; the new Italian service-organisation consensus was added from full text (Padovani 2026, PMID 42850455). WHO's 2000–2023 cause-of-death ranking was added with an explicit dementia, not AD-only, label. Three other new papers were bibliography-only pending full text (PMID 42862590; PMID 42862138; PMID 42855813). No landmark deep note was created: the Cochrane synthesis materially changes the evidence summary, but does not replace drug-specific pivotal trials or establish a new treatment practice standard.

Files changed. Eight records went into literature/BIBLIOGRAPHY.md (381 total). The 2026 Cochrane synthesis was added to wiki/anti-amyloid-immunotherapy.md, literature/statistics/STATISTICS.md and OQ-1. Biel and Malpetti were added to wiki/tau-biology-and-spread.md and OQ-22; Chao to wiki/imaging-and-neuropathology.md and OQ-28. wiki/guidelines.md and literature/guidelines/REGISTRY.md gained the German guideline and Italian consensus (44 registry rows). wiki/clinical-trials-landscape.md gained four corrected status labels and two updated enrolments. STATISTICS.md gained the WHO row. INDEX.md was reconciled to 381 bibliography records, 44 guidance records, two dot rows, and five edited wiki pages at draft. All five edited wiki pages carry last-curated: 2026-10-13 and await independent review. No patient-voice file or note was changed.

Next sweep. Check for full text and subsequent corrections to the AHEAD 3-45 plasma %p-tau217 race/ethnicity study and the new primary-care biomarker-disclosure study; review the preclinical CAA imaging study before moving its abstract-only result into a wiki claim. Recheck the German guideline's implementation and DEGAM dissent, NICE's final appraisal status, Diranersen results posting, and whether any completed randomised trial has actually re-scored its arms using tau-map-derived personal cognitive endpoints. Audit the Cochrane table transcription and four draft pages independently.

2026-10-07 — Independent audit of the dot-table grounding pass (Claude, auditor; author of the changes: Codex)¶

Scope. Only the 2026-10-07 grounding pass on the "Dots not yet connected" table (14 rows → 2). I re-fetched every record live through PubMed E-utilities and checked its PubMed publication types and its Crossref updated-by field. I also re-ran both retained absence searches and broadened each one. Rows that carried citations before the pass were out of scope. None of the 14 did.

Audit check	Result
Retained rows checked	2 (D7, D11)
Citations in retained rows re-fetched	4/4 exist and are the papers named (PMID 31894103; PMID 39945003; PMID 39804128; PMID 39878968)
Records cited in deletion reasons re-fetched	6/6 exist and carry the figures attributed (PMID 41070364; PMID 41929981; PMID 41396614; PMID 32542468; PMID 42108381; PMID 40122980)
Retracted / Expression of Concern (PubMed) or Crossref `updated-by`	0 of 10. D-CARE has a PubMed erratum (PMID 40193097) that only adds omitted grant numbers, and Crossref does not record it. This is one more case of the two sources disagreeing.
Citations that failed verification (non-existent or wrong paper)	0
Figures that had drifted from the source	2, both in D11
Motivated anchoring (citation attached to a claim it does not really support)	1 (D7, Dot B)
Stale absences	1 (D7)
Rows repaired	2 (D7, D11)
Rows deleted by this audit	0
Deletions by the grounder upheld	12/12. One reason (D9) was partly difficulty, as noted below.

D7 — stale absence and motivated anchoring, repaired. The row's absence said no intervention study had tested tau-map-chosen cognitive endpoints, citing a 3-record query. A broader PubMed search ((Alzheimer*[tiab]) AND (tau PET[tiab] OR tau-PET[tiab] OR flortaucipir[tiab]) AND (individualized[tiab] OR individualised[tiab] OR personalized[tiab] OR personalised[tiab] OR patient-specific[tiab] OR subject-specific[tiab]) AND (endpoint*[tiab] OR outcome measure*[tiab] OR composite[tiab]) → 8 records) found the row's proposed junction already published. Biel 2022 (PMID 36345046) built tau-PET-informed personalised cognitive composites for "patient-specific cognitive endpoints in clinical trials". Those composites needed smaller samples to detect simulated intervention effects than memory or global composites in ADNI, though less strongly in the A05 validation cohort. Malpetti 2025 (PMID 41053874) showed in 731 amyloid-positive people with MCI or mild AD, including trial cohorts, that regional tau-PET patterns predict decline in specific cognitive domains. The original row's junction ("individualised, topography-matched cognitive endpoints") is therefore not untouched. By the grounder's own standard for deleting D12, it should have been caught. Dot B was also anchored on placebo's SUCRA ranks (75.7%, 79.5%). Those numbers do not show what the row needed, which is that trials are judged on global scales. They also left out the record's own finding that semorinemab beat placebo on MMSE and ADAS-Cog. I rewrote Dot B to report the record faithfully. The junction now names Biel and Malpetti and is narrowed to what remains: applying such composites to the real randomised arms of a completed trial. Searches for that narrower junction: PubMed Alzheimer*[tiab] AND ("personalized composite"[tiab] OR "personalised composite"[tiab] OR "personalized cognitive composite"[tiab] OR "individualized cognitive"[tiab] OR "personalized cognitive endpoint*"[tiab] OR "tau-informed"[tiab] OR "patient-specific cognitive"[tiab]) → 11 records. PubMed (Alzheimer*[tiab]) AND (semorinemab OR gosuranemab OR tilavonemab OR zagotenemab OR lecanemab OR donanemab [tiab]) AND (tau PET OR tau-PET OR flortaucipir [tiab]) AND (regional OR topograph* OR domain-specific [tiab]) → 3 records. Europe PMC TITLE_ABS:("tau PET" OR "tau-PET" OR flortaucipir) AND TITLE_ABS:("personalized cognitive" OR "personalised cognitive" OR "personalized composite" OR "individualized cognitive" OR "patient-specific cognitive" OR "tau-informed") → 2 records (Biel 2022, and a 2026 opinion piece arguing for tau-informed selection). None re-scored a randomised trial's arms with personalised composites. Dot A (La Joie 2020: 32 patients, 15 months, topography of atrophy at single-patient level) verified as written.

D11 — two drifted figures, repaired; absence upheld. 1. Woods 2023 — the row called communication and social interaction "the review's high-certainty outcome". The review also reports high-quality evidence for instrumental ADL, self-reported mood, anxiety and general behaviour, but only for slight benefit. The defensible claim is that this is the only high-quality outcome the review calls clinically relevant. The row also omitted that the ratings were made by staff or interviewers. Both corrected. SMD 0.53 (0.36–0.70), 5 trials, 702 participants: correct. 2. Reuben 2025 — the NPI-Q (9.8/9.5/10.1) and caregiver-strain (10.7/10.5/10.6) means were presented as "at 18 months". The abstract reports them as overall least-squares means. Corrected. The row also claimed D-CARE "did not measure communication or social interaction". The abstract names only one of the three secondary outcomes, so the row now says only that the abstract reports none. 3. Absence re-run. The grounder's PubMed query was re-run (0 records, as logged). Broadened: PubMed (dementia[tiab]) AND (care management OR collaborative care OR care coordination OR case management OR care navigation [tiab]) AND (social participation OR social engagement OR social interaction OR social health OR loneliness OR communication [tiab]) AND (randomized controlled trial[pt] OR randomi*[tiab]) → 11 records. Europe PMC TITLE_ABS:(dementia) AND TITLE_ABS:("care management" OR "collaborative care" OR "case management" OR "care coordination" OR "care navigation") AND TITLE_ABS:("social participation" OR "social engagement" OR "social interaction" OR "social health") AND TITLE_ABS:(randomised OR randomized) → 4 records. PubMed cognitive-stimulation-implementation query → 25 records. The care-model trials found (e.g. a Taiwanese multicomponent case-management RCT, the WHELD care-home programme) used behavioural, caregiver or quality-of-life outcomes, and none used communication or social participation as an outcome. The absence holds as a scoped title/abstract retrieval gap. 4. Dot A and Dot B are distinct claims: an intervention's effect on one outcome versus a care-model trial's choice of outcome. That is a genuine junction.

Deletions checked against their reasons. - D1 — upheld. Igarashi 2025 (PMID 41070364) simulated Japanese diagnostic flows with blood-biomarker triage and confirmation (maximum mean wait 6.4 months under the current workflow). Zupanic 2026 (PMID 41929981) modelled Slovenian capacity: 76,923 candidates, 64% made ineligible by the CSF bottleneck, and 7% with biomarker integration plus capacity enhancement. The proposed junction has been modelled. - D2, D3, D4, D5, D6, D8 — upheld. Each proposed design is stated almost word-for-word as the design or enabling work of OQ-15, OQ-16, OQ-9, OQ-20, OQ-8 and OQ-6 respectively (checked against the current text). A dot that restates its own OQ adds nothing to the novelty gate. - D9 — upheld, but the first reason is weak. "A diagnostic rule is a criterion, not an empirical finding" reflects difficulty rather than evidence. Dot B was groundable from records already in the condition: the 2025 LATE clinical criteria (Wolk 2025, PMID 39807681, re-fetched) explicitly allow possible LATE when amyloid is present but hippocampal neurodegeneration is disproportionate, beside Nelson 2019's ~25% community prevalence (PMID 31039256, re-fetched; its erratum PMID 31147685 is an unrelated correction notice). The second reason carries the deletion. Brown 2026 (PMID 41396614) classified 19.3–19.7% of amyloid-positive ADNI participants as "vulnerable" (CDR-SB greater than tau burden), with TDP-43 imaging signatures and α-synuclein positivity. It replicated this in 244 Penn-ADRC participants and applied the method to a treated cohort. Figures verified. Had the row been kept, it should have said that the LATE share of biomarker-positive "AD" has been approximated by tau–clinical mismatch but not attributed specifically to TDP-43 with autopsy validation, which is OQ-17's enabling work. - D10, D13 — upheld. D10 merges the designs of OQ-16 and OQ-26. In D13, the claim that "APOE4 is treated as an amyloid-clearance gene" was an unquantified characterisation, and its design is OQ-23's. - D12 — upheld. Pontecorvo 2020 (PMID 32542468): 3 subjects, cortical SUVr vs AT8 Ptau Pearson r=0.81. Chao 2026 (PMID 42108381): two whole brains, AT8–FTP Spearman rs=0.461 in the Braak VI case only. Chao 2026 also reports that FTP underestimates medial-temporal burden. The grounder omitted that, and it matters to OQ-28/OQ-21. - D14 — upheld. Claus 2025 (PMID 40122980): anticoagulant use 31.2% of 968 Rotterdam participants with MCI or early AD, and combined eligibility 8–15%. A grep of wiki/ and literature/ on 2026-10-07 found no atrial-fibrillation prevalence cited anywhere in the condition, confirming that Dot B was ungrounded.

Judgement. The table can now be trusted by the novelty gate, with one qualification. Both rows are grounded in verified records with correct figures, both sides of each are distinct, and both absences rest on dated, logged PubMed and Europe PMC title/abstract searches. D7 now claims a narrower gap than the grounder wrote. The qualification: D7's junction cites two records (PMID 36345046; PMID 41053874) that are not yet in BIBLIOGRAPHY.md or on any wiki page. That fell outside this pass's two-file edit scope. The deletions were sound, with D9's first reason the only one that leaned on difficulty. One failure pattern, an under-broad absence query, survived the grounder's pass in 1 of 2 retained rows, so a two-row table is small enough that the gate should still re-run D7's and D11's queries before relying on them.

Follow-ups. 1. Add Biel 2022 (PMID 36345046) and Malpetti 2025 (PMID 41053874) to BIBLIOGRAPHY.md and to wiki/tau-biology-and-spread.md (prognosis/endpoints). OQ-22's enabling work could cite Biel as the simulation already done. 2. Add Chao 2026's finding that FTP underestimates medial-temporal tau to OQ-28 or the imaging page. 3. INDEX.md still shows the former 14-row dot count (carried over from the grounder's note). Reconcile it at publication. 4. CENTRAL and SciELO were not searched (inaccessible, per SOURCES.md). No registry search was run for D7's narrower junction. An ongoing trial using tau-informed endpoints would show up in ClinicalTrials.gov outcome measures, so query them at the next sweep.

2026-10-07 — Grounding the uncited dot rows (Codex, grounder)¶

Outcome. Started with 14 uncited rows, rewrote 2 (D7 and D11), and deleted 12. The surviving IDs were not renumbered. Each PMID used in the new rows or in the published-junction findings below was fetched through live PubMed E-utilities in this session; its title, abstract, publication types and DOI were read, and its Crossref updated-by field was checked. None of those checked records was labelled Retracted Publication or Expression of Concern in PubMed or had a Crossref update. No NCT ID was added.

Deleted row	Reason
D1	The proposed diagnostic-flow junction has been modelled: a Japanese simulation examined blood-biomarker triage/confirmation, waiting times and lecanemab eligibility (PMID 41070364); a Slovenian capacity model also included plasma-biomarker triage (PMID 41929981). The row's universal capacity premise was broader than its evidence.
D2	Its proposed test of Boston CAA criteria against microbleed count is the design already stated in OQ-15, not an additional junction.
D3	Its proposed midlife prevention trial with biomarker endpoints repeats OQ-16; “every” multidomain trial enrolled people over 60 was not grounded as a universal claim.
D4	Its copathology-stratified anti-amyloid treatment analysis repeats OQ-9's design.
D5	Its protective-versus-harmful microglial-profile question repeats OQ-20; the ADAPT anti-inflammatory prevention result is a different intervention and does not supply a second TREM2-agonism finding.
D6	Its disclosure-protocol trial repeats OQ-8; “strongest” qualitative theme was not supported by a comparative figure.
D8	Its dedicated APOE4-homozygote pathway repeats OQ-6's design.
D9	Its diagnostic rule is a criterion, not a numbered empirical finding for Dot B. An amyloid-positive cohort has already analysed tau–clinical mismatch, TDP-43 imaging signatures and trajectories, including a treated cohort (PMID 41396614); a LATE-specific attributable fraction remains unresolved but this row did not establish a distinct, unexamined junction.
D10	Its LMIC-led prevention proposal combines the designs already stated in OQ-16 and OQ-26, without a separate testable cross-literature question.
D12	Quantitative tau PET-to-autopsy calibration has already been published: a 3-person regional study reported cortical PET–p-tau correlation r=0.81 (PMID 32542468), and a two-brain whole-brain histology study reported an AT8–PET correlation of 0.461 in one Braak-VI case (PMID 42108381). The specific Braak-V density distribution still needs validation, but “calibrate PET against density” is not an untouched junction.
D13	The vascular-target design repeats OQ-23, while “APOE4 is treated as an amyloid-clearance gene in therapeutic reasoning” had no quantified second finding in the cited condition evidence.
D14	The condition did not cite a measured atrial-fibrillation prevalence for the second side. A population-based eligibility study has already measured anticoagulant use in 31.2% of 968 people with MCI or early AD (PMID 40122980), so the proposed exclusion fraction has also been investigated at least in part.

PubMed searches behind the retained absence claims (2026-10-07; title/abstract field searches, so these are scoped retrieval results, not claims of no study anywhere):

D7: (Alzheimer*[tiab]) AND (tau PET[tiab] OR tau-PET[tiab]) AND (topography[tiab] OR regional[tiab]) AND (cognitive outcome*[tiab] OR cognitive endpoint*[tiab]) AND (intervention[tiab] OR trial[tiab]) → 3 records. Their abstracts described observational regional tau change, regional tau–cognition associations in an anti-amyloid trial cohort, and a rasagiline biomarker analysis; none tested participant-specific tau-map selection of a cognitive treatment endpoint.
D11: (dementia[tiab] OR Alzheimer*[tiab]) AND cognitive stimulation[tiab] AND (care model[tiab] OR care management[tiab] OR collaborative care[tiab]) AND (communication[tiab] OR social interaction[tiab] OR social participation[tiab]) → 0 records. A broader check omitting the final communication/social-outcome clause returned 7 records, none a trial combining cognitive stimulation with a care model and measuring those outcomes. This supports only a PubMed title/abstract retrieval gap.

Europe PMC cross-check (2026-10-07): D7 query TITLE_ABS:("tau PET" OR "tau-PET") AND TITLE_ABS:(topography OR regional) AND TITLE_ABS:("cognitive endpoint" OR "cognitive outcome") AND TITLE_ABS:(trial OR intervention) returned 0 records. D11 query TITLE_ABS:"cognitive stimulation" AND TITLE_ABS:("care management" OR "collaborative care" OR "care model") AND TITLE_ABS:(communication OR "social interaction" OR "social participation") returned 1 record, a descriptive technology-supported frailty/homecare project, not a dementia care-model outcome trial. These are similarly scoped title/abstract searches.

Searches that overturned proposed junctions: (Alzheimer*[tiab]) AND (plasma[tiab] OR blood[tiab]) AND (lecanemab[tiab] OR donanemab[tiab]) AND (capacity[tiab] OR demand[tiab] OR resources[tiab] OR MRI[tiab]) returned 29 PubMed records, including the Japanese and Slovenian models above. (Alzheimer*[tiab]) AND (tau PET[tiab] OR flortaucipir[tiab]) AND (quantitative tau[tiab] OR tau density[tiab]) AND (autopsy[tiab] OR postmortem[tiab]) returned 6, including direct PET-to-histology comparisons. The D14 eligibility search, Alzheimer* AND (lecanemab OR donanemab) AND (anticoagulation OR "atrial fibrillation") AND (eligibility OR economic OR cost), returned 8.

Unresolved limits. The surviving absence searches cover PubMed and Europe PMC titles/abstracts, not all registries or full texts; CENTRAL and SciELO were not searched. The quantitative LATE-attributable fraction in biomarker-positive people, Braak-V-specific PET-density calibration, and the incremental treatment exclusion due solely to anticoagulation remain open research questions, but their deleted rows could not establish the claimed two-sided junctions. The condition index still displays the former 14-row count; it was outside the two-file edit scope of this pass and should be reconciled at publication.

2026-09-03 — Independent audit of the evidence-breadth pass (Claude; author of the additions: Codex)¶

Scope. This audit targets only the material Codex added on 2026-09-03; the body checked by the two earlier audits was not re-litigated. The new material was isolated by diffing every page against the site build published on 2026-09-02, which predates the deepening pass, so "new" here means new to the condition, not merely new to a page. Every PubMed record was re-fetched live through E-utilities in this session and every ClinicalTrials.gov record through API v2; nothing below is quoted from memory.

Audit check	Result
Pages audited (all 20 left at `draft`)	20/20
Records added by the pass, live-fetched	56/56 resolved
Reference lines for those records checked against the live record (journal, year, volume, pages)	56/56 matched
Newly used NCT records live-fetched	3/3 resolved; asserted statuses matched
Claim-level checks of the new material	every added trial, meta-analysis, guideline and cohort result checked number by number against its abstract
Asserted absences re-run	4
Corrections made	11 content items + 16 structural
Records added by this audit	2 (PMID 40207637; PMID 41307609)
Pages promoted to `curated`	20/20

Citation breadth — the measure this pass existed to move.

Measure	Before (published 2026-09-02)	After the pass	After this audit
Distinct records cited on wiki pages	308	364	366
Distinct records per page (20 pages)	15.40	18.20	18.30
Mean distinct PMIDs per page	23.6	27.1	27.1
Mean page length (lines)	143	154	154

The pass widened the evidence base rather than padding it, and this is verifiable rather than asserted. Of the 70 page-level citation additions (56 distinct records, some legitimately used on two or three pages), every one was a record the condition did not previously cite anywhere — zero were recycled from the existing landmark pool onto a second page. Distinct records per page rose 19% while mean page length rose 7%. Each page gained between 1 and 6 new records sourced to that page's own topic (fluid biomarkers gained five assay-comparison studies, vascular gained five pericyte/CAA studies, care gained three care-model trials and a discharge-prescribing cohort), and the recurrence that does exist is confined to records that genuinely belong on two pages (New IDEAS on imaging and on diagnostic criteria; the two cost reviews on epidemiology and on care). No page was padded with the condition's existing landmark set, and no page is promoted on that basis.

No fabricated citation was found. All 56 identifiers exist, every reference line names the paper the record actually is, and every quoted effect size, confidence interval, sample size and percentage in the new material matches its abstract except where corrected below. The three newly used trial records were re-queried individually: NCT03887455 ACTIVE_NOT_RECRUITING (1,906 actual), NCT01760005 ACTIVE_NOT_RECRUITING, NCT06424236 TERMINATED with 73 actual — the last matching the 73 participants treated in the gantenerumab open-label extension.

Errors and overstatements corrected.

Wrong analyte attribution (overview). "A blood p-tau217 test spans assay AUCs from 0.642 to 0.947" attributed the bottom of the range to a p-tau217 assay. In Janelidze 2023 (PMID 36087307), 0.642 belongs to a p-tau181/p-tau231 immunoassay; the p-tau217 assays occupy the top of the range. Rewritten as "a blood p-tau test", naming which assays sit at each end.
Stale absence — the experience of anti-amyloid treatment. The patient-experience page and OQ-27 asserted that no qualitative evidence existed on anti-amyloid therapy. A live search on 2026-09-03 found two: 22 people with confirmed AD interviewed after discussing lecanemab with a clinician, reporting information sources, hope, and risk–benefit weighing shaped by family, insurance and trust (Parks 2025, PMID 40207637); and 22 Japanese care partners on donanemab and treat-to-clearance, reporting mental burden, relief at confirmed clearance and continuing worry after stopping (Katayama 2026, PMID 41307609). Both study people considering or supporting treatment. The page and OQ-27 now carry both, and the gap is narrowed and dated to what remains genuinely unstudied: the treatment period itself — infusion attendance, surveillance MRI, and the experience of an ARIA event.
Study population misdescribed (neuropsychiatric symptoms). The suvorexant CSF experiment was described as "a single dose in 12 cognitively unimpaired adults". It is a 38-person randomised trial (placebo 13, 10 mg 13, 20 mg 12) in adults aged 45–65; the quoted effects are the 20 mg arm against placebo (PMID 36897120). Corrected, and the phosphosites named as the record names them.
Cohort split conflated (imaging). Groot 2024's AUC increments were presented as coming from 448 people. They are discovery-cohort figures (n=331; 117 external validation). The AD-dementia base model was also given the all-cause base model's AUC: the base was 0.75 (0.69–0.82), not 0.71. Corrected, and the amyloid-PET increment that did not replicate was added, since "only tau PET replicated" is only meaningful beside what else was significant in discovery (PMID 38857029). The corresponding STATISTICS.md row was corrected too.
Pooled estimate attributed to the wrong denominator (imaging). The AI-imaging SROC-AUCs were attributed to "38 AI-imaging studies"; 38 were included but 28 moderate-to-high-quality studies were pooled (PMID 41061249). Corrected, and the P=0.02 for the intermodal difference added.
Population scope dropped (clinical presentation). Wilkosz's six trajectories come from 201 Caucasian patients with possible or probable AD (PMID 19781112). Labelled, consistent with the treatment the earlier audit gave the Mühlbauer and Katz populations.
One-sided summary (risk reduction). The MAPT blood-amyloid subgroup analysis was summarised as losing significance after multiplicity adjustment, omitting that the per-protocol positive-amyloid comparison (n=154) remained significant at 12 and 36 months after adjustment, and that the authors call the finding a trend requiring confirmation (PMID 37872582). Both results now stand, with the 60-month null in both analyses.
Number without the context that makes it meaningful (care). Post-discharge psychotropic initiation was framed as a safety problem at 6.6%, omitting that 63% of the same cohort was already on a psychotropic before admission and that the authors therefore call new prescribing "relatively uncommon" (PMID 36514208). The paragraph now leads with the baseline, and rests the concern where the paper rests it: 51% persistence beyond 90 days and delirium as the predictor.
Unsupported independence claim (vascular). CSF angiopoietin-2 was called "an independent vascular signal". The study does not test independence from sPDGFRβ — it reports correlation with it (PMID 38182581). Restated as a second barrier-related marker, with what the record actually reports.
Recommendation strength over-assigned (guidelines, registry). The Korean behavioural-symptom guideline grades antipsychotics as a conditional recommendation; the retrieved record states no strength grade for the citalopram advice (PMID 39944528). Both the wiki synthesis and registry row 42 now say so.
Duplicated figures across two pages (care, epidemiology). The same two cost reviews were printed with the same headline ranges in the same framing on both pages. The epidemiology page keeps the severity-graded ranges (its remit); the care page now carries the health-system-specific point — that formal accommodation shifts as much as 67.3% of the burden into direct cost — and cross-links rather than reprinting.

Structural defect fixed on 16 of 20 pages. Every new section was inserted immediately after an existing ## heading and before that heading's own content. The effect was that 16 pages had a heading with no body, and the older section's content appeared to belong to the new one — for example the tau page's therapeutics table sat under "Structure and network propagation", and the trigger-list table on the red-flags page sat under "Two safety transitions". Each new section was relocated to sit before the heading it had displaced, which also puts the new evidence ahead of the summary sections it was inserted into. No prose was changed by the move.

Absences re-run. Four were re-searched live on 2026-09-03. Three were upheld and are now dated to this session: no longitudinal study of insurance, employment or legal-capacity outcomes after a biological AD diagnosis (OQ-25 — the new clinician-documentation study closes only the narrower question of whether practice is uniform, and it is not); no qualitative study of the anti-amyloid treatment period (the narrowed OQ-27); and no study of downstream patient outcomes after direct-to-consumer blood testing. One failed, and is item 2 above.

Nothing verified was deleted. Diffing against the 2026-09-02 build confirms the pass was purely additive: the only non-added lines that differ are link paths rewritten by the site build, not content.

Patient-voice layer. The two records this audit added are filed in literature/patient-voice/sources.md as primary qualitative studies 21 and 22 with their limitations stated, the section-C items renumbered, and the coverage-limits note on the treatment-experience gap rewritten to say what is now covered and what is not.

Literature layer. The 56 records are all present in BIBLIOGRAPHY.md with correct metadata and cited-by paths; 373 records now, after this audit's two additions. STATISTICS.md's new sections match their sources (one row corrected, item 4). REGISTRY.md rows 40–42 (EFNS–ENS/EAN 2015, two Korean Dementia Association 2025 guidelines) match their records, with the correction at item 10; the coverage count of 42 is right. OQ-31 (assay portability) is well-founded on the two head-to-head comparisons it cites. Every wiki PMID appears in its page's reference list and in the bibliography; no cited record is missing from it.

Result. All 20 pages promoted from draft to curated. INDEX.md counts and the curation-state paragraph reconciled. CONDITIONS-ROADMAP.md already reads audited and was not changed.

Follow-ups.

The pass filed all 56 records in BIBLIOGRAPHY.md under a single session-provenance heading ("Evidence-breadth additions — 2026-09-03") with the tag 2026-09-03-deepening, rather than distributing them into the file's 17 topic sections with topical tags as CONVENTIONS §3 specifies. The records are complete and correctly cross-indexed, so nothing is lost; re-filing them by topic is a tidy-up for the next sweep.
clinical-trials-landscape.md reference lines 3 and 4 (PMID 39780249; PMID 41522368 — the evoke design and baseline-characteristics papers) are not cited anywhere in the page body. They predate this pass and were left alone; either cite them or drop them at the next sweep.
Two source records added to this knowledge base's running list of internally inconsistent abstracts: the Care Ecosystem record prints a point estimate that is not centred in its own interval, and the Thyrian record calls a caregiver-burden effect significant (P=0.045) while printing an interval that crosses zero. Both are reported here as the records print them, as with S-CitAD, Graff and DIADS-2. A standing note in CONVENTIONS is now overdue.
Diranersen's phase 2 (NCT05399888) passed its 2026-03-11 primary completion date and should be re-queried for results.

2026-09-03 — Page-by-page evidence-breadth deepening (Codex)¶

Objective and result. Extended rather than rebuilt all 20 canonical wiki pages. The condition-wide PubMed pool increased from 315 to 371 distinct records (+56; 17.8%), moving the condition-level density from 15.75 to 18.55 distinct records per page. Distinct PMIDs on individual pages now range from 21 to 38; every touched page is status: draft and dated 2026-09-03 for independent re-audit.

What was deepened. Added human Aβ-dimer and presenilin kinetic evidence; long-term lecanemab and gantenerumab extension results with design limitations; quantified collaborative-care and caregiver-component trials; trajectory heterogeneity and operational LATE criteria; management utility versus outcome utility for amyloid PET; multi-ethnic incidence trends and severity-specific costs; rare coding immune variants; tau-fold and connectivity evidence; complement/astrocyte/microglia division of labour; pericyte, barrier and CAA autopsy data; assay-specific plasma-biomarker accuracy; tau-PET prognosis and functional-imaging comparisons; graded symptomatic-treatment effect sizes; replicated and discordant brexpiprazole results; positive-versus-null lifestyle trials; diagnosis, discharge and driving safety transitions; and young-onset family experience. Controversies were preserved explicitly, with evidence on both sides where available.

Shared literature layer. Added every new record to literature/BIBLIOGRAPHY.md in the one-line house format with cited-by mappings. Added 15 rows to literature/statistics/STATISTICS.md covering assays, imaging, treatment, care, safety, copathology and cost. Added three graded documents to literature/guidelines/REGISTRY.md: EFNS–ENS/EAN combination therapy (weak recommendation despite significant effects) and the Korean Dementia Association 2025 cognitive-enhancer and behavioural-symptom guidelines (strong/moderate-evidence and conditional recommendations respectively). Sharpened OQ-17, OQ-23, OQ-25 and OQ-29, and added OQ-31 on assay portability.

Live searches and identifier integrity. Issued 54 topic-specific PubMed searches spanning all page domains, then fetched candidate abstracts and metadata through E-utilities. At validation, all 371/371 distinct PMIDs used anywhere in the condition resolved live; seven records omitted from an initial batched response were re-queried individually and all resolved. All 58/58 distinct NCT IDs used anywhere in the condition resolved live through ClinicalTrials.gov API v2. The three NCT records newly used in prose were also queried directly: NCT03887455 was ACTIVE_NOT_RECRUITING, NCT01760005 ACTIVE_NOT_RECRUITING, and NCT06424236 TERMINATED on 2026-09-03.

Deliberately left alone. Preserved all previously verified claims, tables, references and page structure. Did not inflate complete sections with generic reviews, create new pages or paper notes, alter patient-voice source files, or modify WHATS-NEW.md. The absence of patient-centred qualitative evidence on anti-amyloid infusion/MRI/ARIA experience remains open; the 2026-09-03 search found management-impact studies but no qualifying experience study. Actual longitudinal insurance, employment and legal-capacity outcomes also remain unlocated; the new clinician-interview study closes only the narrower documentation-practice question.

Validation. Every inline wiki PMID appears in that page's References section; no wiki PMID is absent from the master bibliography; all 20 page frontmatter blocks are draft; no new [unverified] marker was introduced; the index counts and statuses match the files. The next session should independently re-check the 56 additions claim-by-claim before promoting any page.

2026-08-31 — Independent audit of the deepening pass (Claude; author of the additions: Grok)¶

Scope. This audit targets the material added by Grok's substance-deepening pass, not the previously curated body that Codex checked earlier the same day. All 15 pages left at status: draft by that pass were audited. Every PubMed record was re-fetched live through E-utilities in this session and every ClinicalTrials.gov record through API v2; nothing below is quoted from memory.

Audit check	Result
Pages audited (all pages left at `draft`)	15/15
Distinct PMIDs cited on those pages, live-fetched	239/239 resolved
Reference lines cross-checked against the live record (journal + year)	393/393 matched
Distinct NCT records live-fetched	55/55 resolved; asserted statuses matched
Claim-level checks of the newly added material	every added trial, meta-analysis and cohort result checked number by number against its abstract
Newly asserted or load-bearing absences re-run	5
Corrections made	21
Records added by this audit	2 (PMID 9110909; PMID 41885326)
Pages promoted to `curated`	15/15 (condition now 20/20)

No fabricated citation was found. Every identifier the deepening pass introduced exists, and every reference line names the paper the record actually is. The INVOKE-2 numbers the previous log asked the auditor to re-check (week-96 CDR-SB LS differences −0.31, +0.13, −0.17, all P>0.05) match the published abstract exactly, as does NCT04592874 (COMPLETED, 356 actual). Walsh 2002 is confirmed as Nature 416:535-9. NCT01843075 was re-queried and is still UNKNOWN in the registry despite the 2026 publication; the page already says so.

Errors and overstatements corrected.

Husebo trial mis-described. The stepwise-pain-protocol row said "60 nursing homes". The trial randomised 60 nursing-home units within 18 nursing homes in western Norway (PMID 21765198). Corrected, and the 8-week treatment period and the null ADL/cognition results were added, since a 17% CMAI reduction means little without its duration.
Stale absence — vitamin E. The page claimed the TEAM-AD vitamin E result "has not been replicated at this dose in a non-VA population". A live search found the ADCS trial of selegiline and alpha-tocopherol: 341 patients, 2,000 IU/day, two years, non-VA, reporting a delay in a composite endpoint (median 670 vs 440 days, P=0.001) — but only in analyses adjusted for a baseline MMSE imbalance, with no significant unadjusted difference (Sano 1997, PMID 9110909). The text now carries both results and says why two positives at the same dose are not a replication of each other.
Overbroad absence — ChEIs and ARIA. "No retrieved study addresses this" was replaced with the study that does bear on it: a 2026 real-world cohort of 240 antibody-treated patients versus propensity-matched non-monoclonal-treated patients (ARIA hazard ratio 4.65, 95% CI 3.77–5.73; antithrombotic use 1.87, prior stroke 1.59, hyperlipidaemia 1.41) (Dangpiaei 2026, PMID 41885326). The remaining gap is now stated positively and dated: no retrieved study reports whether continuing a cholinesterase inhibitor during anti-amyloid therapy changes ARIA risk or outcome.
Two internally inconsistent source records reproduced as if sound. The Graff occupational-therapy record prints the patient Dqol interval as "0.6-.1" and the DIADS-2 record prints a CSDD interval of "1.65-4.05" that does not contain its own point estimate of 1.2. The deepening pass silently repaired both into plausible intervals (0.6–1.1 and −1.65 to 4.05). Neither repaired bound is in any record. Both are now reported as point estimates with the record defect disclosed — the same treatment the earlier audit gave the S-CitAD abstract.
EXPEDITION3 duration wrong. Given as 76 weeks on the anti-amyloid page and in STATISTICS.md; the trial's primary comparison is at week 80 (PMID 29365294). Corrected in both.
GRADE certainty inflated. Galantamine's effect on global function was described as "probably" improving; the Cochrane review rates CIBIC-plus as low-certainty ("may"). Corrected, and the high-certainty DAD and NPI estimates it omitted were added.
Goate 1991 overstated. Described as "the first demonstration that APP mutation is sufficient". The paper reports co-segregation in one autopsy-confirmed kindred plus a second unrelated family and claims only that some cases of AD could be caused by APP mutations. Rewritten to the paper's own strength. Sherrington 1995 was likewise described as "the more common early-onset locus", which that record does not establish; it now reports what the record says.
Barthélemy 2024 strengthened by one word. "Superior to CSF for tau-PET" in two places; the paper says "generally superior". Corrected on both pages.
Unverifiable study descriptors. "Single-centre" was asserted for the Graff and nabilone trials; neither record states site count. The nabilone record's actual recruitment setting (one long-term-care facility plus geriatric psychiatry clinics) replaced it, and the nabilone row gained the null CGIC result (47% vs 23%, P=0.09) that the positive scale results were quoted without.
Population scope silently dropped. The Mühlbauer Cochrane review covers Alzheimer's and vascular dementia (17 of 26 comparisons in AD specifically); Katz 1999 enrolled 73% AD, 15% vascular, 12% mixed; HARMONY enrolled psychosis in five dementia aetiologies. All three are now labelled on an Alzheimer's-specific page.
Speculation presented as inference. The INVOKE-2 ARIA-like MRI finding was glossed as "consistent with a phagocytic attack on vascular amyloid". The trial does not test that; it is now marked as a hypothesis. The minocycline adverse-effect list said "vestibular"; the record says dizziness.

Padding removed. Three passages quoted a number without the context that makes it meaningful, and are the reason this audit did not simply confirm the pass:

The sleep section quoted Herring's 28-minute suvorexant gain and then the Cochrane orexin-antagonist pooled estimate of 28.2 minutes as if the second corroborated the first. The Cochrane estimate derives from a single study of n=274 — that same trial. The text now says so.
The two-step plasma p-tau217 workflow was summarised as "88–92% accuracy while avoiding 61–86% of confirmatory tests", pairing the best of each range. Accuracy and test-avoidance move in opposite directions: 85.9% of CSF tests avoided at 88.2% accuracy, 61.2% avoided at 92.0%. Stated as the trade-off it is.
The music-therapy row called the social-behaviour advantage over other activities one of the "reliable" effects; it is low-certainty evidence from 4 studies and 84 participants, against a moderate-certainty null for agitation. Rewritten to the certainty grades.
One further miscalibration: donepezil and galantamine were said to sit "at or just inside" the ADAS-Cog MCID boundary. Against Muir's published thresholds both fall just short of the +3 mild-AD MCID while exceeding the lower bound of the +2 to +3 MCI band. Stated numerically instead.

Absences re-run and upheld. Four asserted absences survived a live re-search and were left standing, now dated to this session: no randomised trial has reduced dementia incidence through lifestyle modification (the retrieved multidomain literature is mediation analyses and cognitive-composite outcomes); the APA 2016 antipsychotic guideline has not been superseded (the only later record is a 2017 Focus reprint of the same document); no PART consensus record was invented — the tau page names primary age-related tauopathy only as one candidate explanation inside an open question; and the ChEI-during-antibody question above, now narrowed rather than removed.

Literature layer. BIBLIOGRAPHY.md 313 → 315 records with the two additions above; no orphan entries and no cited PMID missing from it. STATISTICS.md carried the EXPEDITION3 duration error and was corrected. REGISTRY.md row 39 (APA 2016) is accurate and honest about the record having no abstract. OQ-20 and D5 match the INVOKE-2 record. Headline counts in INDEX.md reconciled.

Result. All 15 deepened pages promoted from draft to curated; the condition is 20/20 curated. CONDITIONS-ROADMAP.md already reads audited and was not changed.

Follow-ups. (1) The Graff and DIADS-2 records join S-CitAD as source-record defects this knowledge base has now documented three times — worth a standing note in CONVENTIONS if a fourth appears. (2) Whether continuing a ChEI during anti-amyloid therapy changes ARIA risk is a clean, designable question and belongs in OPEN-QUESTIONS.md as a Tier 1 candidate at the next sweep. (3) Diranersen's phase 2 (NCT05399888) has a primary completion date of 2026-03-11 and should be re-queried for results.

2026-08-31 — Substance deepening (Grok; no structural restart)¶

Job. The condition was fully built and audited but thin (mean ~135 lines and ~20 distinct PMIDs per page). This pass extended existing pages; it did not delete verified content, did not rewrite from scratch, and did not promote anything (all touched pages are status: draft for a different engine to re-audit).

Pages deepened (15 of 20). Added landmark-trial effect sizes with CIs, competing guideline positions, quantified epidemiology, mechanism detail, and both-sides controversies.

Page	What was added	Deliberately left
`symptomatic-and-supportive-therapy.md`	Per-drug Cochrane (donepezil 2018, galantamine 2024, rivastigmine 2015) with GRADE; Tariot 2004 vs DOMINO on combination; TEAM-AD vitamin E; O'Regan discontinuation meta-analysis; Graff OT; music Cochrane 2025; LipiDiDiet 24/36-month; AAN/CCCDTD5 MCI positions	No restart of the AD2000/DOMINO synthesis
`neuropsychiatric-symptoms.md`	Mühlbauer 2021 Cochrane class effects; Katz 1999 risperidone; APA 2016 guideline; HARMONY pimavanserin; DXM-quinidine; nabilone; Husebo pain protocol; DIADS-2; McCleery 2020 sleep Cochrane vs melatonin meta-analysis	Did not inflate with unvalidated agents
`neuroinflammation-and-glia.md`	Ewers 2019 sTREM2; INVOKE-2/AL002 (null primary, ARIA-like MRI, NCT04592874); Heneka NLRP3; Howard 2020 minocycline	Did not pad with unselected CD33 review papers
`fluid-biomarkers.md`	Olsson 2016 CSF fold-changes; Barthélemy 2024 plasma vs FDA CSF; Brum 2023 two-step workflow; Karikari 2020 p-tau181; Salvadó 2026 plasma staging; 2025 CPG thresholds	Did not add every p-tau phospho-site paper
`tau-biology-and-spread.md`	Hutton 1998 MAPT; TANGO gosuranemab; Tauriel semorinemab (individual trial numbers behind the network meta-analysis)	PART consensus title search returned 0 hits; not invented
`amyloid-biology.md`	Glenner 1984; Goate 1991; Sherrington 1995; Walsh 2002 oligomers; EPOCH and APECS verubecestat with harm in prodromal disease	Did not rewrite the cascade controversy
`anti-amyloid-immunotherapy.md`	EXPEDITION3 solanezumab; TRAILBLAZER-ALZ phase 2 donanemab (iADRS 3.20 already present at n=257)	Did not re-argue meaningfulness
`risk-reduction-and-prevention.md`	Barnes 2011 and Norton 2014 PAF lineage; Framingham; Kivipelto CAIDE; Whitmer midlife obesity	Did not re-quote Livingston 2024 numbers (still no abstract)
`vascular-and-metabolic-contributions.md`	ELAD liraglutide (null primary); Whitmer midlife obesity	Did not treat observational GLP-1–dementia papers as causal
`diagnostic-criteria-and-biological-definition.md`	Jack 2010 hypothetical cascade as the AT(N) predecessor	Criteria table left intact
`imaging-and-neuropathology.md`	Plasma vs PET staging (Barthélemy, Salvadó) as the changing comparator for AUC documents	IDEAS pair left intact
`clinical-trials-landscape.md`	TANGO, Tauriel, ELAD, INVOKE-2, minocycline, EPOCH/APECS, HARMONY, AVP-923; live NCT statuses 2026-08-31	Pipeline-shape table not recomputed
`epidemiology-and-burden.md`	Framingham four-epoch incidence	GBD tables not rewritten
`guidelines.md`	APA 2016 antipsychotic guideline; Cochrane class effects as the evidence behind it	NICE/FDA/EMA block not re-fetched
`overview.md`	INVOKE-2 as the first TREM2-agonist RCT	Five-sentence field map left intact

Pages left alone (5 of 20). clinical-presentation-and-staging.md (26 PMIDs — already at the citation target), genetics.md (23 PMIDs, 154 lines), care-caregiving-and-health-systems.md (the D-CARE/REACH II/WHELD synthesis is already the controversy), red-flags-and-safety-concerns.md (24 PMIDs, 150 lines), patient-experience-and-advocacy.md (23 PMIDs, 149 lines). These were not padded.

Literature layer. Bibliography 265 → 313 live-resolved PMIDs. STATISTICS.md gained rows for Framingham, CSF fold-changes, plasma-vs-CSF, two-step workflow, EXPEDITION3, TRAILBLAZER-ALZ phase 2, INVOKE-2, ELAD, EPOCH/APECS, TANGO, per-drug ChEI Cochrane, TEAM-AD, atypical-antipsychotic SMDs, HARMONY, Husebo, Norton PAF, Kivipelto, Whitmer, and a 13th known-conflict on TREM2 agonism. REGISTRY.md gained the APA 2016 antipsychotic guideline (row 39). OPEN-QUESTIONS.md: OQ-20 narrowed after INVOKE-2; D5 updated. No new landmark notes.

Searches. ~70 PubMed E-utilities esearch queries plus batched efetch of 59 records; ClinicalTrials.gov API v2 for NCT04592874, NCT03352557, NCT03325556, NCT01584440, NCT05399888, NCT01739348, NCT01953601, NCT01900665, NCT03367403, NCT01843075, NCT03289143, NCT00235716. Title-exact searches were used when keyword searches failed. A PART-consensus title search returned 0 hits and was not back-filled from memory.

Decisions. (1) LipiDiDiet's 36-month completer analysis is reported as a selected-survivor result of a negative 24-month primary, not as a treatment for established dementia. (2) APA 2016 is catalogued from the live PubMed record; class/level tables were not reproduced because the record has no abstract. (3) evoke/evoke+ was already on the clinical-trials page from the audit; it was not re-litigated.

Follow-ups for the auditor. Re-fetch the 48 newly added PMIDs; check INVOKE-2 CDR-SB intervals against the full paper (abstract used); re-query NCT01843075 (registry status UNKNOWN despite a 2026 publication); confirm Walsh 2002 volume/pages (Nature 416:535-9 from live efetch).

2026-08-31 — Independent audit (Codex; author: Claude)¶

Scope and result. Codex independently audited all 20 files in wiki/ and all 13 Markdown artifacts in literature/ written by Claude. The audit checked 310 identifier records (265 distinct PubMed records and 45 distinct ClinicalTrials.gov records) and 1,445 contextual claim–citation pairs outside bibliography/reference lists. Every PMID was re-fetched through live PubMed E-utilities during this audit session; every NCT ID was re-fetched individually through ClinicalTrials.gov API v2. All 265 PMIDs existed, all 45 NCT IDs returned the same identifier, and the dated NCT statuses used in the text matched the live records. All citations resolved and matched the subject claimed after the corrections below. All 20 wiki pages were therefore promoted from draft to curated, and CONDITIONS-ROADMAP.md was changed from built to audited.

Audit check	Result
Wiki pages read and checked	20/20
Literature artifacts read and checked	13/13
Distinct PMIDs live-fetched	265/265 resolved
Distinct NCT records live-fetched	45/45 resolved and identifier-matched
Contextual claim–citation pairs checked	1,445
Bibliography coverage	265/265 cited PMIDs present; no orphan or missing record
Bare `[unverified]` markers	0
Pages promoted to `curated`	20/20

Errors found and fixed. Each item below records a content correction, not merely an audit observation.

Stale semaglutide absence. The build said evoke/evoke+ primary results were unavailable. A live PubMed search found the 2026 primary Lancet report (PMID 41865758). Both trials were null at week 104: CDR-SB differences −0.08 (95% CI −0.35 to 0.20; P=0.57) and +0.10 (−0.17 to 0.38; P=0.46). The clinical-trials page, statistics file, bibliography, guideline watch list and OQ-30 were updated; OQ-30 is now resolved rather than open.
Reversed WHELD effects. The neuropsychiatric page printed positive CMAI and NPI-NH point estimates beside negative confidence intervals. The point estimates were corrected to −4.27 and −4.55, respectively, matching PMID 29408901.
Unsafe S-CitAD transcription. PubMed's abstract for PMID 40133524 reports a risk-difference point estimate outside its own confidence interval. The inconsistent numeric estimate was removed from all summaries; the supported null primary conclusion and drug-related QT prolongation remain, with the abstract inconsistency disclosed.
False LMIC evidence absence. The build described LMIC qualitative evidence as almost absent. Live searching located five relevant records: Colombia (PMID 41717228), Iran (PMID 39097934), two Ugandan studies (PMID 37680685; PMID 33043153) and Malaysia (PMID 34368037). The patient-experience page, patient-voice sources/themes/method file, bibliography and OQ-26 now describe the small but real evidence base and narrow the remaining gap to cross-country, patient-led priority setting.
False positive-caregiving absence. The themes file said no second synthesis had been located. The audit found a 41-study integrative review (PMID 29128685) and a separate 17-study synthesis (PMID 30466499). Positive meaning, mutuality, competence and growth are now an established fourteenth theme, explicitly presented as coexisting with—not cancelling—caregiver burden.
False direct-to-consumer guideline absence. The registry said no guideline addressed direct-to-consumer AD blood tests. The 2025 Spanish Society of Neurology positioning document (PMID 40685136) explicitly recommends against such use. It was added to the guideline synthesis and registry; the remaining gap was narrowed to empirical downstream patient outcomes.
Overbroad anticholinergic-trial absence. The red-flags page said no deprescribing trial existed. The audit found AgeWell.de (PMID 40817636) and a 12,787-person high-risk-medication trial in people with dementia (PMID 39432286). The text now distinguishes trials of prescribing outcomes from the still-unanswered question of dementia incidence.
Other absolute absence language. Assertions that no disclosure trial and no agitation/psychosis trajectory trial existed were replaced after focused live PubMed re-queries with dated, outcome-specific evidence gaps: retrieved studies do not establish whether a structured diagnostic encounter changes the listed harms or whether short-term symptom control changes long-term progression.
Regulatory staleness. The guideline page and registry omitted EMA's Kisunla authorisation (24 September 2025), FDA's 2025 subcutaneous lecanemab maintenance and 13 July 2026 at-home starting-regimen approvals, and subsequent NICE appeal/reconsultation activity. Both NICE appraisals remain in development with publication TBC; their latest negative drafts are not final guidance. Current FDA, EMA and NICE pages were directly fetched on 2026-08-31 and the care-capacity discussion was updated to separate infusion demand from continuing biomarker, MRI and emergency-response requirements.
Stale web-access labels. NICE and FDA pages previously labelled 403/404 now fetched successfully and were relabelled. Twelve patient organisations were directly fetched; Alzheimer's Disease International remained HTTP 403 to command-line retrieval but its current official homepage/member directory was present in the live search index. ARDSI returned HTTP 406 and alzheimers.org.in failed DNS; both remain quarantined rather than treated as verified.
Missing and untidy citation wiring. The clinical-trials page gained the live-resolved PMIDs for CitAD (24549548), DIAN (22784036), intranasal insulin (32568367) and the pivotal antibody trials where their results were invoked. Duplicate reference numbering was corrected. The bibliography was expanded from 254 to 265 records and its cited-by paths and all headline counts were reconciled.
Overstatement in synthesis prose. “Burden grows almost entirely through demography” was softened to “largely”; headings and prose implying that amyloid initiation or immune causality were established across all AD were narrowed to what monogenic genetics and pathway association support; and a definitional claim was separated from prognostic evidence.

Remaining dated evidence gaps and access limitations. No bare [unverified] marker remains, and none of the following prevents page curation: (a) the 2026-08-31 PubMed query found no outcomes study of insurance, employment or legal-capacity consequences after a biological diagnosis (OQ-25); (b) the audit query found no qualitative study centred on receiving anti-amyloid treatment, surveillance MRI or ARIA (OQ-27); (c) direct-to-consumer blood-test recommendations exist, but no retrieved study measured downstream patient outcomes; (d) the S-CitAD abstract's internally inconsistent estimate remains a source-record defect, so the number is deliberately not reproduced; (e) the NICE appraisals remain unfinished; and (f) the two South Asian organisation domains above remain in quarantine, while ADI is search-index rather than direct-fetch verified.

Validation. Final checks found 265 unique PMIDs in both the condition corpus and bibliography, no cited PMID missing from the bibliography, 45 unique NCT IDs, no unresolved identifier, no [unverified] text, sequential numbered reference lists, and no broken relative Markdown links within the condition. No file outside conditions/alzheimers-disease/, CONDITIONS-ROADMAP.md, or the requested log/index metadata was modified.

2026-08-31 — Full build (Claude)¶

What was built. All 20 canonical wiki pages from the INDEX page plan, the complete literature layer, a replacement tiered open-questions agenda, and an updated INDEX. Every page is status: draft; nothing was promoted, because the model that wrote these pages must not audit them.

Deliverable	Result
Wiki pages	20/20 written from live literature; ~54,200 words; mean 2,708 words per page (stroke reference: 2,882)
Distinct PMIDs	254, every one resolved through live PubMed E-utilities in this session
Distinct NCT records	45, every one resolved through live ClinicalTrials.gov API v2 queries in this session
`literature/BIBLIOGRAPHY.md`	254 records in 17 topic sections, metadata regenerated from live `efetch` output rather than transcribed, cross-indexed to every citing file
`literature/notes/`	6 landmark notes: Braak 1991, Jack 2024 (revised criteria), van Dyck 2023 (CLARITY AD), Sims 2023 (TRAILBLAZER-ALZ 2), Boyle 2018 (person-specific pathologies), Palmqvist 2024 (blood biomarkers in primary care)
`literature/guidelines/REGISTRY.md`	36 records (31 journal-indexed, 5 web-based), supersession chains, a 7-row disagreements table, 5 identified gaps, a 7-item watch list
`literature/statistics/STATISTICS.md`	13 sections of quick-reference tables with source, year, population and method on every row; a 12-item known-conflicts section
`literature/patient-voice/`	Four files; 13 organisations verified by direct fetch across six regions; 12 aggregate themes each with ≥2 sources; 2 fetch failures quarantined
`OPEN-QUESTIONS.md`	Replaced the six seed questions with 30 questions (16 Tier 1, 14 Tier 2) and a 14-row "dots not yet connected" table
`INDEX.md`	At-a-glance counts, nine reading paths, page statuses, literature-layer table, and a prioritised list of what the audit should check first

Seed anchors re-verified. All nine anchor PMIDs recorded at seeding (38934362, 29653606, 36449413, 37459141, 37280110, 40156286, 34998485, 39096926, 17568013) were re-fetched at the start of this session; all nine resolved with matching author, journal, year and content. One caveat: Livingston 2024 (PMID 39096926) has no abstract in PubMed, so it is cited for framing only and the population-attributable-fraction numbers on the prevention page are taken from Stephan 2024 (PMID 38824956) instead.

Searches run. Roughly 90 PubMed E-utilities searches across every planned page topic, plus targeted efetch retrieval of about 200 abstracts; about 20 ClinicalTrials.gov v2 queries by intervention name, condition and NCT ID; and 17 direct web fetches of guideline, regulatory and patient-organisation pages. Search terms are recorded per domain in literature/patient-voice/README.md for the patient-voice layer and are reconstructable from the bibliography's topic grouping for the rest.

Deliberate scoping and framing decisions.

The AA-versus-IWG definitional dispute is presented as unresolved content, not adjudicated. Both positions are given with the progression data that support each, because the data genuinely support both readings at different biomarker profiles (amyloid alone: 11–20% five-year progression; amyloid plus tau: 57%).
The clinical-meaningfulness dispute over anti-amyloid effect sizes is likewise presented with numbers on both sides, including the explicit finding that trial effects fall below published MCIDs.
The shared border with vascular dementia is carried by vascular-and-metabolic-contributions.md, which opens with a note stating that the two conditions are curated separately, cross-links the reciprocal page, and explicitly declines to assert a clean separation. Mixed-pathology data are treated as central content rather than as a caveat.
wiki/overview.md was written fresh because the seed had no wiki/ files; the seed's cited overview paragraph in INDEX.md was preserved and extended rather than discarded.
Non-journal sources (guidelines, regulators, patient organisations) are cited in the repository's (Publisher — "Title", URL, accessed date) form, and every one records whether it was directly fetched or only search-result verified.
Forward links to vascular dementia are not hyperlinks. That condition is seeded and its wiki/ is empty, so links to mixed-pathology-and-alzheimer-overlap.md, cerebral-amyloid-angiopathy.md, cerebral-small-vessel-disease.md and nosology-and-diagnostic-criteria.md would have been broken. Those pages are named in prose, marked as planned, and the hyperlink points to the companion condition's INDEX instead. The audit or the vascular-dementia build should convert them to real links once those pages exist. A link check across the whole condition returns zero broken relative links.

What could not be verified, and is flagged rather than hidden.

NICE guidance pages (NG97 and the lecanemab/donanemab final draft guidance) returned HTTP 403 to direct fetch on 2026-08-31; the content used is search-result verified and labelled as such in wiki/guidelines.md and the registry.
FDA approval and labelling pages returned HTTP 404 to direct fetch through this session's proxy; the content used is search-result verified and labelled. The EMA Leqembi page fetched successfully.
evoke / evoke+ (semaglutide) primary results were not retrievable: both trials show COMPLETED status with a September 2025 primary completion date on ClinicalTrials.gov, but no primary-results publication was returned by PubMed searches. Recorded as OQ-30.
Two patient-organisation sites (ardsi.org, alzheimers.org.in) failed to fetch (connection reset; DNS failure) and are quarantined in literature/patient-voice/organizations.md. No South Asian organisation is therefore verified in this build.
Four asserted absences are recorded explicitly in OPEN-QUESTIONS.md so that the audit can re-run them: empirical work on insurance/employment/capacity consequences of a biological AD diagnosis (OQ-25); qualitative research on the experience of anti-amyloid treatment, infusion and ARIA (OQ-27); evoke primary results (OQ-30); and low- and middle-income-country patient-voice evidence (OQ-26).
Four reference lines initially carried article identifiers (afad128, afaf344, eaau5732, eabd1327) that had not been returned by the live records; these were removed mid-build and the references now carry only the verified volume.

Follow-ups for the next session.

Independent audit by a different engine — the immediate next step. Priorities are listed in INDEX.md under "Curation state": numeric transcription on the three highest-density pages, the four asserted absences, the three search-result-verified web records, NCT statuses, and whether the Lancet Commission can be given quantitative content.
Re-query evoke/evoke+ for results and close or update OQ-30.
Retry the two failed organisation fetches and add South Asian, Chinese, Middle Eastern and Spanish-language Latin American organisations; this is the largest single gap in the patient-voice layer.
Search for non-English national guidelines (German S3, Japanese, Chinese, Latin American) absent from the registry.
Re-verify all 45 NCT statuses; several trials have primary completion dates in 2026–2028 and will change status.
Once audited, add the condition to tools/build_frontend.py and publish.

2026-08-31 — Seeded¶

Created the condition scaffold (INDEX.md, OPEN-QUESTIONS.md, LOG.md, wiki/, literature/) per CLAUDE.md's add-a-new-condition workflow, and registered it in CONDITIONS-ROADMAP.md.
Defined the canonical 19-page list in INDEX.md (the canonical list for every condition after the two founding ones is its own INDEX Pages table, per CONVENTIONS.md §5). The list was designed against the shape of this condition's literature, not copied from another condition.
Ran live PubMed scoping searches and recorded the resolved anchor records in INDEX.md. Every PMID in the seed came from a live query in that session; none was written from memory.
Wrote six seed open questions, each tied to an anchor record, to be replaced by a full tiered agenda plus dots table at build time.

Deliberate scoping decision. Alzheimer's disease and vascular dementia are curated as two conditions with an explicit shared border rather than as one. They have distinct literatures, criteria sets and trial pipelines, but mixed pathology is the common case at autopsy, so each condition carries a dedicated overlap page cross-linking the other, and neither may assert a clean separation the pathology literature does not support.

Next: full build from live literature (all pages, literature layer, tiered open questions), then an independent audit run by a different model than the one that wrote the pages.