Alzheimer's disease — master index¶
Last curated: 2026-10-27 · status: literature sweep updated (10 pages curated, 10 pages draft pending independent review)
The condition in five sentences. Alzheimer's disease has been redefined over the last decade from a clinical syndrome into a biological one: the 2018 NIA-AA research framework classified people by amyloid, tau and neurodegeneration biomarkers rather than by symptoms, and the 2024 Alzheimer's Association revision carries that principle into diagnosis, defining AD as a process that begins with Alzheimer's neuropathologic change while people are still asymptomatic — a step the International Working Group explicitly declines to take, recommending that amyloid-positive cognitively normal people be described as at risk (Jack 2018, PMID 29653606; Jack 2024, PMID 38934362; Dubois 2024, PMID 39483064). The burden is large and grows largely through demography: GBD forecasts dementia cases rising from 57.4 million (95% UI 50.4–65.1) in 2019 to 152.8 million (130.8–175.9) in 2050 with age-standardised prevalence essentially flat, and increases concentrated in north Africa and the Middle East (367%) and eastern sub-Saharan Africa (357%) (GBD 2019 Dementia Forecasting Collaborators 2022, PMID 34998485). Anti-amyloid monoclonal antibodies became the first disease-modifying treatments: lecanemab slowed 18-month CDR-SB decline by 0.45 points (95% CI 0.23–0.67) with ARIA-E in 12.6%, and donanemab slowed iADRS decline by 3.25 points (1.88–4.62) in the low/medium-tau population with ARIA-E in 24.0% and three treatment-related deaths — effect sizes below the published minimal clinically important differences for the same instruments (van Dyck 2023, PMID 36449413; Sims 2023, PMID 37459141; Muir 2024, PMID 38561021). Diagnosis is being displaced from CSF and PET toward blood: plasma p-tau217 achieved pooled sensitivity 88.1% and specificity 88.7% across 113 studies, and a mass-spectrometry-based score reached about 90% diagnostic accuracy in Swedish primary care against physicians' 61% (Therriault 2025, PMID 40818474; Palmqvist 2024, PMID 39068545). The field's unresolved core is whether a biologically defined disease that usually occurs alongside vascular and Lewy pathology — 94% of autopsied older adults had at least one pathology and AD occurred in isolation in only 9%, explaining between 22% and 100% of any individual's cognitive loss — can be treated as a single target (Schneider 2007, PMID 17568013; Boyle 2018, PMID 29244218).
Start here: overview.md. Research frontier: OPEN-QUESTIONS.md. Growth history: LOG.md.
Companion condition: vascular dementia. The two are curated separately but share a substrate; see vascular-and-metabolic-contributions.md here, and mixed-pathology-and-alzheimer-overlap.md there once that condition is built (it is on the vascular-dementia canonical page list but not yet written). Neither page claims a clean separation the autopsy literature does not support.
At a glance¶
| Layer | Built content |
|---|---|
| Canonical wiki pages | 20/20 written; 10 curated, 10 draft after the 2026-10-27 sweep |
| Distinct live-resolved PMIDs in the condition | 405 (14 added in the 2026-10-27 sweep) |
| Distinct live-resolved NCT records | 60 (NCT02547818 added 2026-10-27; NCT00235716 is cited only in LOG.md) |
| Landmark paper notes | 6 |
| Guideline / regulatory records | 47 (40 journal-indexed, 7 web-based) |
| Patient organisations | 12 directly fetched, 1 search-index verified, plus 2 quarantined direct-fetch failures |
| Open-question records | 31 (30 open — OQ-27 narrowed by the 2026-09-03 audit — 1 resolved by the evoke/evoke+ publication) |
| Dots-not-yet-connected junctions | 2 |
Reading paths¶
| Need | Path |
|---|---|
| Rapid orientation | Overview → clinical presentation and staging → red flags |
| The definitional dispute | Diagnostic criteria and the biological definition → fluid biomarkers → guidelines |
| Mechanism | Amyloid biology → tau biology and spread → neuroinflammation and glia → genetics |
| Treatment decisions | Anti-amyloid immunotherapy → symptomatic and supportive therapy → neuropsychiatric symptoms → red flags |
| Measurement | Fluid biomarkers → imaging and neuropathology → clinical presentation and staging |
| Population and policy | Epidemiology and burden → risk reduction and prevention → care, caregiving and health systems |
| The shared border with vascular dementia | Vascular and metabolic contributions → vascular dementia → stroke |
| Lived experience | Patient experience and advocacy → literature/patient-voice/themes.md → care, caregiving and health systems |
| Research design | Clinical trials landscape → OPEN-QUESTIONS.md |
Pages¶
The canonical page list for this condition (per CONVENTIONS.md §5). Link only to these filenames.
| File | Scope | Status |
|---|---|---|
| overview.md | What AD is under the biological definition, burden, treatment era, map of the topic | curated |
| clinical-presentation-and-staging.md | Preclinical → MCI → dementia continuum, CDR/FAST staging, rates of progression, atypical variants | curated |
| diagnostic-criteria-and-biological-definition.md | NIA-AA 1984/2011/2018/2024, IWG counter-position, disclosure and its consequences | draft |
| epidemiology-and-burden.md | Incidence, prevalence, forecasts, sex ratio, geographic inequality, mortality, cost, under-ascertainment | curated |
| genetics.md | APOE dose effects, autosomal dominant disease, Down syndrome, GWAS, polygenic scores, testing ethics | curated |
| amyloid-biology.md | Aβ production, aggregation and clearance; the cascade hypothesis and the case against it | curated |
| tau-biology-and-spread.md | Tangles, Braak staging, propagation, tau-PET topography, tau-directed therapeutics | draft |
| neuroinflammation-and-glia.md | Microglial states, TREM2, astrocytes, complement, cause vs consequence | draft |
| vascular-and-metabolic-contributions.md | Mixed pathology, CAA, blood–brain-barrier failure, metabolic risk — the shared border | draft |
| fluid-biomarkers.md | CSF Aβ42/40 and p-tau, plasma p-tau217 and ratios, assays, cut-points, primary-care implementation | draft |
| imaging-and-neuropathology.md | Amyloid and tau PET, Centiloids, MRI, FDG-PET, the ABC autopsy score | draft |
| anti-amyloid-immunotherapy.md | Aducanumab, lecanemab, donanemab: effect sizes, meaningfulness, ARIA, eligibility, cost | draft |
| symptomatic-and-supportive-therapy.md | Cholinesterase inhibitors, memantine, discontinuation, non-drug therapy | curated |
| neuropsychiatric-symptoms.md | Agitation, psychosis, depression, apathy, sleep; antipsychotic mortality risk | curated |
| risk-reduction-and-prevention.md | Modifiable risk factors, PAFs, multidomain trials, blood pressure, hearing, education | curated |
| care-caregiving-and-health-systems.md | Diagnosis pathways, care models, caregiver burden, long-term care, cost, capacity | draft |
| guidelines.md | Criteria, appropriate-use documents, national guidance, regulatory decisions and their disagreements | draft |
| clinical-trials-landscape.md | Active and completed trials by mechanism and phase, with enrolment, stage and safety data | draft |
| red-flags-and-safety-concerns.md | ARIA and anticoagulation, misdiagnosis and mimics, driving, antipsychotic harms, unvalidated tests | curated |
| patient-experience-and-advocacy.md | Diagnosis and disclosure, stigma, caregiver voice, organisations, research priorities | curated |
Literature layer¶
| Resource | Contents | Status |
|---|---|---|
| BIBLIOGRAPHY.md | 405 live-resolved PubMed records, grouped by first major domain and cross-indexed to every use | draft after 2026-10-27 additions |
| notes/ | Six structured landmark notes: Braak 1991, Jack 2024, van Dyck 2023 (CLARITY AD), Sims 2023 (TRAILBLAZER-ALZ 2), Boyle 2018, Palmqvist 2024 | audited |
| guidelines/REGISTRY.md | 47 guideline, regulatory and reimbursement records with supersession chains, a nine-item disagreements table, identified gaps and a watch list | draft after 2026-10-20 additions |
| statistics/STATISTICS.md | Fourteen sections of quick-reference tables, including assay, imaging, care, safety, copathology, cost and global burden estimates | draft after 2026-10-20 addition |
| patient-voice/README.md | Method, ethics rules, interpretation guidance and coverage boundary | audited |
| patient-voice/organizations.md | 12 directly fetched organisations and 1 search-index-verified federation across six regions, plus an explicit quarantine of 2 direct-fetch failures | audited |
| patient-voice/themes.md | Fourteen aggregate themes, including LMIC support ecologies and positive dimensions of caregiving, plus three themes considered and not established | audited |
| patient-voice/sources.md | Annotated syntheses, primary qualitative studies, supporting quantitative sources, fetch record and stated coverage limits | audited |
Curation state¶
This condition passed an independent audit on 2026-08-31, underwent an evidence-breadth pass by Codex on 2026-09-03, and was independently re-audited by Claude the same day. The deepening pass added 56 records that the condition did not previously cite — none recycled from the existing pool — lifting the wiki corpus from 308 to 366 distinct records, i.e. from 15.4 to 18.3 distinct records per page. That is the measure the pass existed to move, and it moved faster than page length: distinct records per page rose 19% while mean page length rose 7% (143 to 154 lines). The pass did not pad pages with the condition's existing landmark set.
The audit re-fetched all 56 new records and all 3 newly used ClinicalTrials.gov records live, checked every new claim against its source, re-ran the searches behind the asserted absences, corrected eleven items (one stale absence, one wrong analyte attribution, three misdescribed study populations, two one-sided summaries, two softened overstatements, and sixteen misplaced section headings), and added two records of its own. At that audit, all 20 pages were curated.
The 2026-10-27 sweep added fourteen live-resolved PubMed records: ABN/RCPsych UK blood-biomarker guidance (recommendations unread; full text blocked), a Taiwanese hospital-caregiver guideline, Mayo plasma p-tau217 discordance (PMC full text), Whitehall II 12-year blood-biomarker prediction, a 37-study real-world lecanemab meta-analysis, the null ALZT-OP1 phase 3 (NCT02547818) and low-dose radiotherapy phase 2 trials, a Brazilian AT(N)-versus-AA-2024 reclassification study, Medicare dementia-detection trends, and five bibliography-only records. No Cochrane review appeared in the window. The five edited wiki pages were already draft; no curated page was touched.
The 2026-10-20 sweep added ten live-resolved PubMed records (Italian EPA2 appropriate-use update, French lecanemab eligibility, US utilisation/adherence, Tokyo real-world ARIA, oral TREM2 agonist phase 1, ARIC midlife p-tau181, capillary versus venous collection, GLP-1 class meta-analysis, and two Cochrane editorials on the Nonino review). It moved three 2026-10-13 bibliography-only records (Molina-Henry, Ataka, Alten) onto wiki pages from complete abstracts, added NCT06343636, and re-dated OQ-8, OQ-12, OQ-15, OQ-16, OQ-20, OQ-25, OQ-27, OQ-30 and OQ-31. Ten wiki pages are draft for independent review (eight edited today plus tau-biology and imaging from 2026-10-13); the other ten retain their earlier audit status. The dot table still contains two rows. No landmark deep note was written.
Seed anchors¶
The nine records resolved live while scoping this condition on 2026-08-31 were all re-verified at the start of the build session and all resolved with matching metadata. They are now superseded by BIBLIOGRAPHY.md and are retained here only as provenance: PMIDs 38934362, 29653606, 36449413, 37459141, 37280110, 40156286, 34998485, 39096926, 17568013.