Alzheimer’s Disease: Independent Evidence on Tests and Drugs

Table of contents

• Evidence summary
• What Alzheimer's disease is
• Symptoms and red flags
• Causes and risk factors
• Diagnosis: assessment, biomarkers and blood tests
• Treatments that work, with an evidence table
• Medicines: tablets and amyloid antibodies
• Prevention and lifestyle claims, graded
• Supplements and brain-health claims, graded
• Risks and side effects
• Who to see and doses studied
• NICE and regulator positions
• Follow the money
• Related research
• Frequently asked questions
• Sources and funding notes

Evidence summary

Independent evidence and credibility scorecard

Independence tiers: 1 = regulator or government body; 2 = academic or Cochrane team with no declared commercial funding for the work (or a regulator with indirect industry-fee funding); 3 = independent authors with declared industry consulting ties, or a charity with a mission stake; 4 = manufacturer-funded or manufacturer-authored. Credibility runs from A (highest) to D. Our sources cluster in the US and UK, with Finland and Italy contributing one source each, so the verdict reflects Anglo-American research culture and funders.

What Alzheimer's disease is

Alzheimer's disease is the most common cause of dementia. The World Health Organization estimates that about 57 million people had dementia worldwide in 2021, and that Alzheimer's disease accounts for roughly 60% to 70% of cases. It also states plainly that there is no cure for dementia (WHO fact sheet, updated July 2026). The NHS describes the disease as linked to an abnormal build-up of proteins in and around brain cells, called amyloid plaques and tau tangles, with falling levels of the messenger chemical acetylcholine and gradual shrinkage of brain regions that handle memory. The process begins many years before symptoms (NHS: Causes).

The word "dementia" is an umbrella for the effects on thinking and daily life. Alzheimer's disease is one cause, alongside vascular dementia, dementia with Lewy bodies and others. The distinction matters for treatment: NICE recommends different medicines for Lewy body dementia and vascular dementia than for Alzheimer's disease (NG97 text reproduced by an NHS trust). In 2024 an Alzheimer's Association workgroup proposed revised criteria that define and stage the disease by biology (biomarkers), not only by symptoms (Jack et al. 2024). That approach is influential in research and drug development, and many authors on that paper list grants or ties with diagnostic and drug companies. We did not review criticisms of the biological definition.

Legal status by country. In the US, the FDA has granted full approval to lecanemab (Leqembi) and approved donanemab (Kisunla) (FDA, July 2023; Pharmaceutical Technology, July 2024). In Great Britain, the MHRA licensed lecanemab in August 2024 and donanemab a few months later, but NICE has not recommended either for NHS use (Alzheimer Europe).

Symptoms and red flags

The main early sign is memory lapses: forgetting recent conversations, misplacing things, repeating questions, struggling for words, poor judgement, and increased anxiety or agitation. Later symptoms include hallucinations and delusions that may come and go, aggression or restlessness, difficulty swallowing, weight loss, incontinence, loss of speech and, in severe stages, a need for full-time care (NHS: Symptoms).

Causes and risk factors

Age is the biggest risk factor; the NHS says risk roughly doubles every 5 years after 65, and early-onset cases can begin from around age 40. A family history adds a small amount of risk overall, while a single inherited gene in a few families causes much higher risk. Down's syndrome raises risk. Severe head injury, smoking, obesity, diabetes, high blood pressure, high cholesterol, hearing loss, untreated depression, social isolation and inactivity are also linked to risk, though several are associations rather than proven causes (NHS: Causes).

The 2024 Lancet Commission on dementia lists 14 potentially modifiable risk factors. Twelve are carried over: low education, hearing loss, hypertension, smoking, obesity, depression, physical inactivity, diabetes, excessive alcohol, traumatic brain injury, air pollution and social isolation. Two are new: higher LDL cholesterol in midlife and untreated vision loss in late life. The Commission estimates that 45% of dementia could potentially be prevented if all 14 were eliminated, with LDL cholesterol contributing 7% and vision loss 2% (Alzheimer Europe summary; Lancet record). Treat the 45% as a model-based ceiling, not a promise. It assumes every association is causal and that every factor could be removed. The Lancet record shows several authors have consulting ties to Lilly, Eisai, Roche, Biogen and TauRx Pharmaceuticals.

The WHO fact sheet lists a similar set of factors and stresses that age is the strongest, while dementia is not an inevitable part of ageing (WHO).

Diagnosis: assessment, biomarkers and blood tests

The standard route

Diagnosis usually starts with a GP, who asks about symptoms, examines you, reviews medicines and may order blood tests to rule out other causes such as thyroid or nutritional problems. Simple memory and thinking tasks follow. If the picture is unclear, a referral to a memory clinic (old age psychiatrist, geriatrician or neurologist) may include a CT or MRI scan to rule out other conditions and look for signs of Alzheimer's-related damage. The NHS states that there is no simple, reliable test for Alzheimer's disease, and that scored memory tests are never used alone (NHS: Diagnosis).

Blood tests: what the FDA has cleared (verified)

• May 2025: the FDA cleared the Lumipulse G pTau217/β-Amyloid 1-42 Plasma Ratio from Fujirebio Diagnostics for adults aged 55 and older with signs and symptoms, in specialised care settings. In a study of 499 samples, 91.7% of positive results and 97.3% of negative results matched amyloid PET or spinal fluid; fewer than 20% were indeterminate. The FDA states the test "is not intended as a screening or stand-alone diagnostic test" (FDA, 16 May 2025).
• October 2025: Roche's Elecsys pTau181 was cleared as an aid to initial assessment in primary care for people aged 55 and over. Roche reported a negative predictive value of 97.9% in a 312-person study meant to mimic primary care. Predictive values depend on how common the disease is in the group tested, and Roche says the test was developed with Eli Lilly (Practical Neurology).
• August 2026: Roche's Elecsys pTau217 was cleared as an aid in assessing amyloid pathology for adults aged 55 and older with cognitive decline. It reports positive, intermediate or negative results, and is not meant to predict dementia or monitor treatment. The Alzheimer's Association describes it as the fourth blood test cleared by the FDA (Practical Neurology; Alzheimer's Association). We did not find the accuracy figures in the sources we could open.

An Alzheimer's Association guideline suggests that, in specialist care for people who already have cognitive impairment, a triage blood test should reach at least 90% sensitivity and 75% specificity, and a confirmatory test at least 90% on both. Positive triage results need confirming by PET or spinal fluid, and the authors warn that many commercial tests do not meet these thresholds (Alzheimer's Association guideline).

What this means in plain English: a blood test can help show whether amyloid is probably present in someone who already has symptoms. It is not a way to predict dementia in healthy people, and a positive amyloid result does not by itself mean someone has dementia. These tests also matter commercially: confirming amyloid is a gateway to antibody treatment. Roche, which makes the p-tau181 and p-tau217 tests with Lilly, is a partner of the company that sells donanemab.

UK position. We could not confirm any NICE recommendation or NHS routine rollout of these blood tests. A £5 million Blood Biomarker Challenge by Alzheimer's Research UK, the Alzheimer's Society and NIHR was announced in November 2023, with experts saying blood tests could be on the NHS within five years (PMLiVE).

Treatments that work, with an evidence table

Medicines: tablets and amyloid antibodies

Cholinesterase inhibitors (donepezil, galantamine, rivastigmine)

These raise acetylcholine in the brain. The NHS says the three work about equally well, that rivastigmine also comes as a patch, and that only specialists usually start them. Common side effects are nausea, vomiting and loss of appetite, usually settling after about 2 weeks (NHS: Treatment). In the donepezil review, 24% of people on the drug withdrew early versus 20% on placebo, and adverse events were reported by 72% versus 65% (Birks 2018). A 23 mg dose gave no extra benefit over 10 mg, with more side effects. NICE says treatment should normally start with the lowest acquisition cost drug, and not to stop these medicines because of disease severity alone (NG97 text).

Memantine

Memantine blocks the effect of excess glutamate. NICE recommends it for severe Alzheimer's, and for moderate disease if cholinesterase inhibitors are not tolerated or are contraindicated, and suggests adding it for people on a cholinesterase inhibitor with moderate (consider) or severe (offer) disease (NG97 text). Cochrane found that in mild Alzheimer's it may increase discontinuation because of adverse events (RR 2.12, 95% CI 1.03 to 4.39) (McShane 2019). The NHS lists headaches, dizziness and constipation as common, usually temporary effects.

Lecanemab (Leqembi) and donanemab (Kisunla)

These are monoclonal antibodies given by intravenous infusion that bind amyloid and help remove plaques. They are intended for mild cognitive impairment or mild dementia with confirmed amyloid.

• Lecanemab (funded by Eisai and Biogen): 1,795 people, 18 months, 10 mg/kg every 2 weeks. CDR-SB worsened by 1.21 versus 1.66 on placebo (difference −0.45, 95% CI −0.67 to −0.23). ADAS-cog14 difference −1.44 (−2.27 to −0.61). Brain amyloid fell by 59 centiloids. ARIA with swelling occurred in 12.6% and infusion reactions in 26.4% (van Dyck 2023).
• Donanemab (sponsored by Eli Lilly): 1,736 people, infusions every 4 weeks for up to 72 weeks, stopped early if plaque cleared. Combined-population CDR-SB difference −0.7 (95% CI −0.95 to −0.45); ARIA with swelling in 24.0% versus 2.1%, 52 symptomatic; infusion reactions 8.7%; three treatment-related deaths versus one (Sims 2023).
• Independent class review: the Cochrane review (17 trials, 20,342 people, all industry-funded) found trivial effects on cognition and dementia severity at 18 months and small effects on functional ability at best, concluding that amyloid removal does not seem to bring clinically meaningful benefit. It estimated 107 more people per 1,000 with ARIA-E, but no extra serious adverse events or deaths at 18 months (Nonino 2026).
• The counter-argument: critics say the Cochrane review combined seven antibodies, most of which failed, with the two that worked, and reported that more than 30 experts asked for a retraction (Alzheimer's Society blog; HealthEd). We could not open the Alzheimer's Society page. These are reported by the secondary summaries we could read.

What the numbers mean: a 0.45-point difference on an 18-point scale is under half a point. Over 18 months the placebo group worsened by 1.66 points, and lecanemab by 1.21. NICE judged the effects small. Whether such a difference is noticeable to a patient or family is the central disagreement in the field.

Prevention and lifestyle claims, graded

Multidomain programmes: FINGER and US POINTER

FINGER randomised 1,260 at-risk Finnish adults aged 60 to 77 to two years of diet, exercise, cognitive training and vascular monitoring or general health advice. The intervention group gained slightly more on a cognitive test composite: a difference of 0.022 per year (95% CI 0.002 to 0.042). Adverse events were more common with the intervention (7% versus 1%), mainly musculoskeletal pain (Ngandu 2015). US POINTER randomised 2,111 older adults to a structured or a self-guided programme for two years. Both groups improved; the structured group gained 0.029 SD per year more (95% CI 0.008 to 0.050), with benefits similar in APOE ε4 carriers and non-carriers (Baker 2025). The study's lead funder says it will follow participants for four years to look at dementia risk, so dementia prevention itself is not yet shown (Alzheimer's Association).

Blood pressure

In SPRINT MIND, treating to a systolic target under 120 mm Hg rather than under 140 mm Hg lowered the rate of mild cognitive impairment (HR 0.81, 95% CI 0.69 to 0.95) but the dementia difference was not statistically significant (HR 0.83, 95% CI 0.67 to 1.04). The authors noted the trial stopped early and may have been underpowered (SPRINT MIND). See our guide to high blood pressure prevention.

Hearing

The ACHIEVE trial of hearing aids and counselling found no difference in 3-year cognitive change overall. A prespecified analysis suggested a benefit in the higher-risk subgroup, but that is a hypothesis to test, not a conclusion (Lin 2023). Hearing correction still has other benefits, and hearing loss is on the Lancet list.

Drugs given to slow Alzheimer's that NICE advises against

NG97 recommends not offering diabetes drugs, blood-pressure drugs, statins or NSAIDs specifically to slow the progression of Alzheimer's, other than in a trial (NG97 text). Our guides on high cholesterol and diabetes prevention cover those conditions in their own right.

Supplements and brain-health claims, graded

• Ginkgo biloba: the GEM trial gave 120 mg of extract twice daily to 3,069 volunteers aged 75 or older for a median of 6.1 years. Dementia rates were 3.3 versus 2.9 per 100 person-years (HR 1.12, 95% CI 0.94 to 1.33) with no effect on progression from MCI (DeKosky 2008). A Cochrane review of 36 treatment trials found the evidence of benefit "inconsistent and unreliable" and noted publication bias could not be excluded (Birks 2009). Grade: strong evidence of no prevention benefit. We found no existing Pure City Research page on ginkgo.
• Omega-3 fish oil: the Cochrane review of three trials (4,080 cognitively healthy people randomised; the mini-mental state result pools two trials, 3,221 people) found no effect on mini-mental state score (−0.07, 95% CI −0.25 to 0.10) and no trial measured dementia (Sydenham 2012). The review is dated; newer trials may exist that we did not review. See our omega-3 guide. Grade: moderate evidence of no benefit.
• Vitamin D: 5-year supplementation at 1,600 or 3,200 IU/day did not clearly change dementia incidence in 2,492 largely vitamin D-sufficient Finns, but with only 45 cases the trial could not rule out a benefit or harm (Lönnroos 2025). See our vitamin D guide. Grade: insufficient.
• Multivitamin and cocoa flavanols: COSMOS-Mind found a daily multivitamin-mineral raised a global cognition score by 0.07 SD over three years and cocoa extract did nothing. The authors call it a first signal needing confirmation (Baker 2023). Grade: weak.
• Vitamin E, ginseng and herbal formulations: NICE says do not offer them to treat dementia (NG97 text).
• Other "brain health" products: for lion's mane and creatine, see lion's mane and creatine. For the wider supplement picture, see our supplements overview. We did not find an independent trial showing any supplement slows Alzheimer's disease.

Risks and side effects

• Cholinesterase inhibitors: nausea, vomiting, appetite loss; more withdrawals and adverse events than placebo (Birks 2018).
• Memantine: headache, dizziness, constipation; Cochrane found more dropouts from adverse events in mild disease (McShane 2019).
• Antipsychotics for distress: the NHS says coping strategies come first and, if drugs are used, risperidone at the lowest dose for the shortest time because of serious side effects (NHS: Treatment).
• Interactions and cognitive side effects: NG97 asks clinicians to watch for anticholinergic burden from commonly prescribed medicines, which can worsen thinking (NG97 text). Ask a pharmacist to review medicines before and during treatment.
• Diagnostic risk: the FDA lists false positives (inappropriate diagnosis and treatment) and false negatives (more testing, delayed treatment) as risks of the blood test (FDA).

Who should avoid or be cautious. People with two ApoE ε4 copies and people on blood thinners need particular care with lecanemab (FDA). Antibodies were studied only in early disease (MCI or mild dementia), so the trial data do not cover moderate or severe disease. We did not find sourced guidance on pregnancy or breastfeeding, which is not relevant to most people with Alzheimer's.

Who to see and doses studied

Start with a GP, who can arrange initial checks, then a memory clinic for specialist assessment. Cholinesterase inhibitors and memantine are generally started by specialists, though a GP may prescribe on a specialist's advice (NHS: Treatment). Antibodies are specialist, hospital-based treatments.

Doses studied (not advice): donepezil 5 mg or 10 mg a day in most trials, 23 mg slow-release in two (Cochrane); memantine 20 mg a day or 28 mg extended release in the analysed trials (Cochrane); lecanemab 10 mg/kg every 2 weeks (CLARITY AD); donanemab every 4 weeks for up to 72 weeks (TRAILBLAZER-ALZ 2); ginkgo extract 120 mg twice daily (GEM). Do not start, stop or combine any of these without a clinician.

Related conditions that affect memory and mood. Depression can resemble or accompany dementia; see depression in older adults. Poor sleep and sleep apnoea also matter; see sleep disorders in older adults.

NICE and regulator positions

• NICE NG97 (dementia guideline): the text reproduced by an NHS trust recommends donepezil, galantamine and rivastigmine as options for mild to moderate Alzheimer's; memantine for severe, or moderate if the first group are not tolerated; consider or offer adding memantine for people on a cholinesterase inhibitor with moderate or severe disease; and says not to offer acupuncture, ginseng, vitamin E or herbal formulations (NG97 text). That document is dated December 2018; we could not check NICE's site for later updates.
• NICE on antibodies: in the June 2025 final draft, NICE did not recommend lecanemab or donanemab for routine use or managed access. It found real but small effects, a high treatment burden and cost per QALY well above the NHS threshold. It reported CDR-SB reductions of 27% to 33% for lecanemab and 23% to 29% for donanemab (Alzheimer Europe, 19 June 2025). Both manufacturers appealed; on 20 March 2026 appeal panels sent the appraisals back to the committee to reconsider carer utilities, unpaid care costs, infusion costs and (for donanemab) long-term data. A third draft went to consultation from 31 March to 28 April 2026, with a committee meeting scheduled for 10 June 2026 (Pharmaceutical Journal). We found no report of the outcome of that meeting. Check NICE directly for the current status.
• FDA: lecanemab converted from accelerated to traditional approval in July 2023, with an ARIA boxed warning (FDA); donanemab was approved in July 2024 (Pharmaceutical Technology).

Follow the money: who makes it and who funded the evidence

Who makes and sells the products

• Lecanemab (Leqembi): developed and sold by Eisai with Biogen; the pivotal CLARITY AD trial was funded by both (NEJM record). At US launch, Eisai gave a wholesale price of about $26,500 per year for an average 75 kg patient (Medthority, January 2023).
• Donanemab (Kisunla): developed and sold by Eli Lilly. A vial costs $695.65 in the US, about $32,000 for a 12-month course, according to one trade report (Pharmaceutical Technology).
• Blood tests: Fujirebio Diagnostics makes the Lumipulse ratio test (FDA). A search summary of company releases describes Fujirebio as part of Japan's H.U. Group Holdings; we did not open an ownership filing. Roche makes the Elecsys p-tau181 and p-tau217 tests, both developed with Eli Lilly (Practical Neurology).
• Older tablets and memantine: we did not research manufacturers or prices. NICE advises starting a cholinesterase inhibitor with the lowest acquisition-cost option (NG97 text).

Who funded the evidence

• Company-funded: CLARITY AD (Eisai, Biogen) and TRAILBLAZER-ALZ 2 (Eli Lilly; authors include employees and shareholders) (CLARITY; TRAILBLAZER). The Cochrane review notes all 17 amyloid-antibody trials it included were pharmaceutical-industry funded (Nonino 2026). In the donepezil review, 17 of 30 trials were industry-funded and 9 had no funding information (Birks 2018).
• Public and charity-funded: the Cochrane antibody review (Italian regional and national health funds), the memantine review (UK Department of Health), FINGER (Academy of Finland and charities), GEM (US NIA and NCCIH), SPRINT MIND and ACHIEVE (US NIH), and COSMOS-Mind (NIA) (see the scorecard and source notes). US POINTER is led and mainly funded by the Alzheimer's Association, which says it has invested nearly $50 million, with the NIA funding add-on studies (Alzheimer's Association).
• Scale of public spending: a Senate proposal in July 2025 would have brought annual NIH Alzheimer's and dementia research funding to $3.9 billion (Alzheimer's Association, advocacy page), which implies a current level of roughly $3.8 billion. We did not find an NIH budget document to confirm the exact figure.
• The regulators and reviewers: about 70% of the FDA's human drugs programme in FY2025 ($1.704 billion of $2.429 billion) came from industry user fees (CRS fact sheet). Cochrane policy bars its groups from taking money from commercial sponsors with an interest in the topic (Cochrane policy, June 2025).

Researcher and panel ties

• The Lancet Commission record lists authors with consulting fees or advisory roles at Lilly, Eisai, Roche, Biogen and TauRx (PubMed record).
• The record for the 2024 revised diagnostic criteria lists grants from Roche, Lilly, Eisai, Biogen, Fujirebio and other diagnostic and drug companies among its authors, and one author is employed by Takeda and is a former Lilly employee (Jack 2024 record). That does not make the criteria wrong, but diagnostic definitions influence who is eligible for treatment.
• Several US POINTER authors disclose grants or advisory roles with Eisai, Lilly, Biogen, Nestlé and others (PubMed record).

Documented integrity events in the field

• Aduhelm: a December 2022 report by two US House committees described the FDA's approval of Biogen's aducanumab as "rife with irregularities" and examined its $56,000-a-year price (Jurist). Biogen discontinued the drug in 2024 and said the decision was not related to safety or efficacy concerns (Radiology Business, seen in search results only). Biogen said it stands by the integrity of its actions.
• Retracted landmark paper: Nature retracted the 2006 paper on an amyloid assembly called Aβ*56 in 2024 (retraction note). The retraction was reported to follow image manipulation concerns; we did not open the notice text.
• NIH misconduct finding: in September 2024 the NIH reported it had found falsification and/or fabrication involving reuse and relabelling of figure panels in two publications by a senior NIA neuroscience division director, who no longer held that post (NIH statement reproduced by MLO).
• Cassava Sciences: in September 2024 the company agreed to a $40 million SEC penalty over disclosure of its 2020 simufilam trial results, without admitting or denying the allegations (Cassava SEC filing). The drug later failed phase 3 and development was discontinued (Alzheimer Europe, seen in search results only).

What this means: the only large trials of the newest drugs were run by the people who sell them, and the strongest independent look at the whole antibody class is contested. That does not make the drugs ineffective; it means no one with nothing to gain has yet shown a benefit that patients clearly feel. The prevention and older-drug evidence is mostly publicly or charity funded, but it shows small effects. Where an advocacy charity funds and leads a trial (US POINTER), its interest in positive messages deserves visibility. The independent bodies we lean on most, NICE and Cochrane, have reasons to be cautious about cost and hype, and both rely on company data for the antibodies.

Related research

For wider, evidence-graded context, see these Pure City Research guides:

• Depression in older adults
• Sleep disorders in older adults
• Obstructive sleep apnoea
• High blood pressure: prevention and management
• High cholesterol
• Type 2 diabetes: prevention and management
• Stroke
• Exercise and social support for mental health
• Smoking and heart disease
• Alcohol and heart disease
• Omega-3 guide
• Vitamin D guide
• Lion's mane
• Stress, anxiety and depression supplements

Frequently asked questions

Sources and funding notes

What we could not verify: NICE's website blocked automated access, so we could not read NG97 or the appraisal pages directly. NG97 wording comes from a December 2018 NHS trust document that reproduces it, and the NICE position on the antibodies comes from news reports, so later NICE changes may be missed. We found no report of the outcome of NICE's 10 June 2026 committee meeting. We could not open the Cochrane Library page for the antibody review, the Alzheimer's Society response, the Lancet Commission full text or the Alzheimer's Association guideline's funding statement, so the Commission's own funding is unverified. The critics' arguments against the Cochrane review come from news and charity summaries. We did not read independent evaluations of the Roche tests or the accuracy figures for the August 2026 Roche p-tau217 clearance. Headquarters countries for Eisai, Biogen, Lilly, Roche and Fujirebio are stated from general company knowledge or search summaries, not from filings we opened, and the Fujirebio and H.U. Group relationship comes from a search summary. We did not identify the manufacturers or prices of generic tablets, and the NIH funding total comes from advocacy material. We did not review other prevention trials such as multivitamin or B-vitamin trials beyond COSMOS-Mind, or newer omega-3 trials. No existing Pure City Research article covers lecanemab or donanemab, ginkgo or this condition, so we did not link a separate drug article.

1. NHS: Alzheimer's disease, diagnosis — UK government health information; reviewed 4 July 2024.
2. NHS: Alzheimer's disease, symptoms — UK government health information.
3. NHS: Alzheimer's disease, causes — UK government health information.
4. NHS: Alzheimer's disease, treatment — UK government health information.
5. WHO. Dementia fact sheet (updated July 2026) — intergovernmental, Switzerland; funded by member states and donors (not detailed here).
6. NICE NG97. Dementia: assessment, management and support — UK guideline; page not opened.
7. Essex Partnership University NHS Foundation Trust. Formulary section 7, management of dementia (December 2018) — NHS trust document reproducing NICE TA217 and NG97 recommendations; used for NG97 wording.
8. Alzheimer Europe. NICE final draft guidance on donanemab and lecanemab (19 June 2025) — European patient-organisation news; funding not reviewed.
9. The Pharmaceutical Journal. NICE opens consultation on donanemab and lecanemab (31 March 2026) — Royal Pharmaceutical Society journal.
10. Journal of Dementia Care. NICE publish final draft guidance (3 July 2025) — UK trade journal; quotes NICE's reasons.
11. FDA. Clears first blood test used in diagnosing Alzheimer's disease (16 May 2025) — US regulator; about 70% of human drugs programme funded by user fees.
12. FDA. Converts Leqembi to traditional approval (6 July 2023) — US regulator.
13. Practical Neurology. Blood-based p-tau181 testing cleared for primary care (November 2025) — trade press relaying a Roche statement.
14. Practical Neurology. FDA clears the Elecsys pTau217 blood test (August 2026) — trade press relaying a Roche release.
15. Alzheimer's Association. FDA clearance of Elecsys pTau217 (24 August 2026) — US charity; funded by donations and other sources (not reviewed).
16. Alzheimer's Association clinical practice guideline: blood-based biomarkers in specialised care — US charity guideline; funding statement not seen.
17. Jack CR et al. Revised criteria for diagnosis and staging of Alzheimer's disease. Alzheimers Dement 2024 — US; author grants and ties with Roche, Lilly, Eisai, Biogen, Fujirebio and others.
18. Birks JS, Harvey RJ. Donepezil for dementia due to Alzheimer's disease. Cochrane 2018 — UK; authors declare none; 17 of 30 included trials industry-funded.
19. McShane R et al. Memantine for dementia. Cochrane 2019 — UK; UK Department of Health grant.
20. van Dyck CH et al. Lecanemab in early Alzheimer's disease. NEJM 2023 — multinational; funded by Eisai and Biogen.
21. Sims JR et al. Donanemab in early symptomatic Alzheimer disease (TRAILBLAZER-ALZ 2). JAMA 2023 — multinational; Eli Lilly sponsor, many authors Lilly employees.
22. Nonino F et al. Amyloid-beta-targeting monoclonal antibodies for MCI or mild dementia due to Alzheimer's disease. Cochrane 2026 — Italy; Regione Emilia-Romagna and Italian Ministry of Health.
23. Alzheimer's Society. Response to claims that anti-amyloid drugs are ineffective — UK charity; page blocked, seen in search results only.
24. HealthEd. Cochrane finds Alzheimer's drugs ineffective, as experts slam the review — Australian clinical education publisher; seen in search results only.
25. Cochrane Hong Kong. Anti-amyloid Alzheimer's drugs show no clinically meaningful effect — Cochrane affiliate; summary of the review.
26. Livingston G et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet 2024 — UK-led; NIA grant listed; author consulting ties to Lilly, Eisai, Roche, Biogen, TauRx.
27. Alzheimer Europe. 2024 Lancet Commission identifies 14 modifiable risk factors — summary of the Commission.
28. Ngandu T et al. FINGER trial. Lancet 2015 — Finland; Academy of Finland, Alzheimer's Association, Novo Nordisk Foundation and other funders.
29. Baker LD et al. US POINTER trial. JAMA 2025 — US; Alzheimer's Association lead funder; NIA add-on studies; author ties with Eisai, Lilly, Biogen, Nestlé.
30. Alzheimer's Association. US POINTER study results page — funder's own summary; says it invested nearly $50 million.
31. SPRINT MIND Investigators. Intensive vs standard blood pressure control on probable dementia. JAMA 2019 — US; NIH grants.
32. Lin FR et al. ACHIEVE hearing intervention trial. Lancet 2023 — US; National Institutes of Health.
33. Woods B et al. Cognitive stimulation for people with dementia. Cochrane 2023 — UK; authors declare CST manual royalties and training fees.
34. DeKosky ST et al. Ginkgo biloba for prevention of dementia (GEM). JAMA 2008 — US; NIA and NCCIH grants.
35. Birks J, Grimley Evans J. Ginkgo biloba for cognitive impairment and dementia. Cochrane 2009 — UK; no conflicts known.
36. Sydenham E et al. Omega-3 for the prevention of cognitive decline and dementia. Cochrane 2012 — UK; MRC grant.
37. Lönnroos E et al. Vitamin D3 and incident dementia, Finnish Vitamin D Trial. J Gerontol A 2025 — Finland; Academy of Finland and Finnish foundations; some author industry grants disclosed.
38. Baker LD et al. Cocoa extract and multivitamin on cognition (COSMOS-Mind). Alzheimers Dement 2023 — US; NIA grants.
39. Pharmaceutical Technology. Lilly's donanemab wins FDA approval (July 2024) — trade press; price figures.
40. Medthority. Eisai's approach to US pricing for Leqembi (January 2023) — healthcare-professional news site; relays Eisai's statements.
41. PMLiVE. £5m project to bring Alzheimer's blood tests to the NHS (November 2023) — trade press; Alzheimer's Research UK, Alzheimer's Society and NIHR.
42. Congressional Research Service. FDA budget fact sheet (June 2026) — US legislative-branch research service; FDA funding mix.
43. Cochrane. Conflict of interest policy for Cochrane groups (June 2025) — Cochrane's own policy; bars commercial-sponsor funding.
44. Alzheimer's Association. Senate appropriations committee approves increase for Alzheimer's research (July 2025) — advocacy charity; used for NIH funding scale.
45. Jurist. US House committee report challenges FDA approval of Alzheimer's drug (December 2022) — US legal news.
46. Radiology Business. Biogen discontinues Aduhelm — trade press; seen in search results only.
47. Retraction Note: A specific amyloid-β protein assembly in the brain impairs memory. Nature 2024 — journal retraction notice; text not opened.
48. MLO. Statement by NIH on research misconduct findings (September 2024) — trade press reproducing the NIH statement.
49. Cassava Sciences. SEC filing on settlement (September 2024) — company filing; company's own account of the SEC settlement.
50. Alzheimer Europe. Cassava Sciences' experimental AD drug fails phase 3 study — patient-organisation news; seen in search results only.

Educational research reviewed 6 October 2026. This article is not medical advice; talk to a qualified clinician or pharmacist before starting, stopping or combining treatments.