New Alzheimer's Drugs Slow Decline a Little: What Trials and Regulators Found

Lecanemab and donanemab are amyloid-clearing antibody medicines shown in large trials to slow the decline of early Alzheimer’s disease. The slowing was real but small: less than one point on an 18-point dementia scale over about 18 months. It came with brain swelling or small bleeds, often silent and sometimes serious^[1][2]. Regulators in the US and the European Union have approved both drugs for a narrow group of people. NICE in England has so far judged them poor value for public money in draft guidance.

For people living with dementia and their families, news of these drugs arrived with a great deal of hope attached. This article sets out, at population level, what the two main trials found and did not show, how regulators and payers have responded, and why the harder question for most health systems is who can be diagnosed in time to be considered at all. It does not say whether any individual should or should not have these drugs; that is a decision for a person and their specialist team.

Both trials slowed decline, by less than one point on an 18-point scale

Both drugs are antibodies, given by infusion into a vein, that clear amyloid, a protein that builds up in the brain in Alzheimer’s disease. Both were tested in people at the earliest symptomatic stages, mild cognitive impairment or mild dementia due to Alzheimer’s disease, whose amyloid had been confirmed by a PET scan or a spinal fluid test.

Lecanemab (CLARITY AD). The trial enrolled 1,795 people aged 50 to 90, half given lecanemab every two weeks and half placebo, for 18 months^[1]. Its main measure was the Clinical Dementia Rating-Sum of Boxes (CDR-SB), which runs from 0 to 18, higher meaning more impairment. People started at about 3.2. After 18 months, scores had risen by 1.21 points with lecanemab and 1.66 with placebo, a difference of 0.45 points (95% confidence interval 0.23 to 0.67). Amyloid on PET fell sharply, and measures of memory, thinking and daily activities also favoured the drug. The authors described the result as “moderately less decline” than placebo^[1].

Donanemab (TRAILBLAZER-ALZ 2). This trial randomised 1,736 people at 277 centres in 8 countries, with a mean age of 73.0, and 76% completed it^[2]. Infusions were every four weeks, and people were switched to placebo once amyloid on PET had fallen below a pre-set level^[7]. Across everyone in the trial, CDR-SB scores (a secondary outcome here; the primary measure, the iADRS, also favoured the drug) rose by 1.72 points with donanemab and 2.42 with placebo after 76 weeks, a difference of 0.7 points. In the 68.1% with lower levels of tau, a second Alzheimer’s protein, the difference was 0.67 points^[2].

In both trials, people on the drug still declined. The drugs slowed that decline; they did not stop or reverse it.

Whether the difference matters to daily life is still disputed

The results were statistically clear. Whether a difference of 0.45 to 0.7 points on an 18-point scale is one that families would notice is a separate question, and experts disagree.

The US Food and Drug Administration (FDA) concluded that the lecanemab trial “verified clinical benefit”^[4]. A Cochrane systematic review published in 2026 reached a more cautious view. It pooled 17 randomised trials with 20,342 participants, testing seven amyloid-targeting antibodies, all funded by the pharmaceutical industry^[5]. It concluded that at 18 months the effect on memory and thinking and on dementia severity “is trivial, while on functional ability, it is small at best.” It also raised a concern about “functional unblinding”: side effects visible on scans may have let some participants and investigators guess who was receiving the drug, which can bias results^[5].

Only one of the review’s 17 trials tested lecanemab and one donanemab; most tested other antibodies in the same class. So its estimate describes the drug class, not these two drugs alone. Even so, it is an independent judgement that the benefits are modest.

Three further limits apply to the trials themselves:

• Duration. The main results cover about 18 months. NICE notes “a lack of evidence on the long-term effects”^[6].
• Who took part. In the donanemab trial, 91% of participants were White and 2% were Black or African American^[7].
• Stage. The drugs were tested only in early disease. Nothing in these trials applies to people with moderate or advanced dementia.

Brain swelling and small bleeds are common, mostly silent, and occasionally serious

The main safety concern with both drugs is amyloid-related imaging abnormalities, or ARIA: swelling or fluid in the brain (ARIA-E) and small bleeds (ARIA-H), seen on MRI scans. In the lecanemab trial, 12.6% of people on the drug had ARIA-E and 26.4% had infusion-related reactions^[1]. In the donanemab trial, 24.0% had ARIA-E, compared with 2.1% on placebo; 52 of those 205 cases caused symptoms. Infusion reactions affected 8.7% on donanemab and 0.5% on placebo. Three deaths in the donanemab group and one in the placebo group were judged to be related to treatment^[2].

Across the whole drug class, the Cochrane review estimated 107 more cases of ARIA-E per 1,000 people treated than with placebo, and 29 more cases of symptomatic ARIA-E per 1,000. At 18 months it did not find an increase in serious adverse events (high certainty) or in deaths overall^[5]. In one early US health-records study of 240 people treated with these antibodies, the hazard of ARIA was 4.65 times that of people on non-antibody treatment. Taking blood-thinning medicines at the same time was also associated with a higher hazard^[8].

Risk is not spread evenly. People who carry two copies of the APOE4 gene variant have more ARIA, including serious ARIA, than people with one copy or none; the FDA says APOE4 testing should be done before treatment begins, and the label carries a boxed warning about ARIA^[7]. That is why the European approvals exclude people with two copies, and why treatment involves repeated scheduled MRI scans: for lecanemab in the EU, before the start and before the 5th, 7th and 14th doses^[9]. In the EU, donanemab must not be used by people taking anticoagulants^[10].

Regulators approved the drugs for narrow groups; payers have been slower to fund them

Regulators ask whether a medicine’s benefits outweigh its risks. Health-technology-assessment bodies ask whether a public health system should pay for it, given what else the money could buy. For these drugs the answers have differed.

In Europe, both drugs are limited to people who have only one or no copy of APOE4. Each company must run a “controlled access programme” with national authorities, so that the drugs reach only that group^[9][10]. In the EU, lecanemab is stopped once a person progresses to moderate Alzheimer’s disease. Donanemab treatment should not exceed 18 months^[9][10].

NICE has been appraising both drugs since 2023, through draft guidance, a final draft in June 2025, appeals by both companies and a fresh round of draft guidance in March 2026^[11][12]. The 2026 draft for lecanemab said the benefit “is relatively small, but the cost for providing it is high (including fortnightly infusions in hospital and intensive monitoring for side effects)”, and that the most plausible cost-effectiveness estimate was “much higher than what NICE considers an acceptable use of NHS resources”^[6]. In late July 2026, NICE paused both evaluations because the companies and NHS England had begun commercial discussions. As of October 2026, no final guidance has been published^[11][12].

Only a small share of people with dementia would be eligible

These drugs are not for most people living with dementia. They are licensed only for early Alzheimer’s disease with confirmed amyloid, and the safety rules exclude many people with other health conditions or findings on brain scans. Studies that apply the rules to real populations give a consistent picture, although they count different things and are not directly comparable:

In the Minnesota study, common reasons for exclusion were other chronic conditions and findings on brain imaging^[3]. In Korea, 36.8% failed one memory-test cut-off, a criterion the authors said should be reconsidered where many older people have less formal education^[15]. The Gothenburg authors extrapolated to around 5.9 million potentially eligible people in Europe and 2.2 million in the US, a rough estimate from one city’s 70-year-olds^[13].

Eligibility can also differ between groups. Among 10,804 people screened at US sites for four early Alzheimer’s trials, Hispanic, Black and Asian participants had higher odds than non-Hispanic White participants of not meeting the amyloid criteria. For non-Hispanic Black participants, the odds were 3.75 times as high^[16]. Our study of ethnicity and dementia looks at these gaps in more detail.

The bigger bottleneck is diagnosis, not the drugs

To be considered for treatment, a person needs a timely specialist assessment, confirmation of amyloid by PET scan, lumbar puncture or blood test, then an infusion service and regular MRI. Many health systems were not built for this. In many countries, large numbers of people with dementia never receive a formal diagnosis, as our study of dementia diagnosis rates shows, and those who are referred often face long waits for memory assessment.

Modelling studies put numbers on the gap. One study of the NHS in England estimated an initial wait of 56 months in 2023, rising to 129 months in 2029 and then falling slowly to around 100 months, given the NHS’s current diagnostic capacity. Using a blood test as an extra triage step cut the modelled wait to around 17 to 25 months^[17]. A 2026 US model found current confirmatory testing so limited (under 1% of people on the diagnostic pathway have amyloid PET and 3.5% spinal fluid tests) that only about 0.3% of amyloid-positive people would receive a timely, accurate PET diagnosis^[18]. These are projections, not observed waits.

Blood tests may ease this. In May 2025 the FDA cleared the first blood test to aid diagnosis of Alzheimer’s disease, for adults aged 55 and over with symptoms. In its evaluation, 91.7% of positive results matched amyloid on PET or spinal fluid testing and 97.3% of negative results matched an absence of amyloid. Fewer than 20% of results were indeterminate, and the FDA says the test is “not intended as a screening or stand-alone diagnostic test”^[19]. A US model projects that adding a blood test would cut waits for specialist appointments by 35%^[20].

What the evidence asks of us

The benefit of these drugs is measurable but modest, the risks need careful monitoring, and only a minority of people living with dementia could be considered for treatment. Several things follow for health services, funders and policymakers.

• Fix diagnosis for everyone, not only for drug eligibility. The diagnostic bottleneck predates these drugs. A timely, accurate diagnosis matters for care planning, support and safety whatever treatment follows. Investment in memory services and validated blood tests would help people who will never be eligible, as well as those who might^[17][19].
• Make sure access is fair. Trial populations were mostly White, eligibility criteria screen out more people from some ethnic groups, and memory-test cut-offs may disadvantage people with less formal education^[7][15][16]. Health systems that offer these drugs should check who is being assessed and treated, and publish that information.
• Collect long-term evidence. NICE points to a lack of long-term evidence, and the Cochrane reviewers found that benefits and harms were inconsistently reported^[5][6]. The European controlled access programmes and real-world studies are a chance to measure benefits and harms in the people who actually receive these drugs, outside trial conditions^[8][9].
• Keep research broad. The Cochrane authors concluded that future research on disease-modifying treatments “should focus on other mechanisms of action”^[5]. Whatever view one takes of amyloid drugs, funders should not let a single approach crowd out others.
• Do not let new drugs displace care and support. For most people living with dementia, support after diagnosis, help for carers and good care matter most, and that is where most of the gap lies.

Language: this piece uses person-first language ("people living with dementia", "people with dementia"), as most dementia organisations and people with lived experience prefer, and uses clinical terms such as "Alzheimer's disease" or "mild cognitive impairment" only where a study, dataset or diagnosis refers to that specific condition.

Why we're publishing this

People living with dementia and their families deserve clear information about new treatments, without hype and without dismissal, including why a drug may not be offered where they live. This piece sets the trials, the funding decisions and the diagnostic gap side by side, so that dementia organisations can explain them accurately, and makes the case that the wider gaps in diagnosis and support affect far more people. More of our evidence is gathered in the Dementia & Health research library.

How to read this data

The trial figures come from the published abstracts of two industry-funded, placebo-controlled phase 3 trials. The trials differed in population, main outcome and length, so they should not be used to rank one drug against the other. The CDR-SB differences are averages across groups; they do not predict what will happen to any one person. The Cochrane review pools a whole class of antibodies, most no longer in use. Eligibility studies apply licensing or trial rules to particular cohorts or clinics and are not comparable with one another. Waiting-time and capacity figures come from models, not observed waits. Regulatory and NICE status is as shown on official pages in October 2026 and may change. Decisions in other countries, including Australia and Japan, could not be verified at source and are not included.

Use this data

Free to cite with attribution to Understanding Aging. Charities, carers' groups, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.

Important: This article summarises published research and statistics for general information. It is not medical, care or financial advice and cannot be used to diagnose, treat or predict anything about an individual. If you have concerns about your own or someone else's memory or health, please speak to a GP or another qualified health professional.

Sources

1. van Dyck CH, Swanson CJ, Aisen P, et al. Lecanemab in early Alzheimer’s disease. New England Journal of Medicine, 2023. europepmc.org/article/MED/36449413
2. Sims JR, Zimmer JA, Evans CD, et al. Donanemab in early symptomatic Alzheimer disease: the TRAILBLAZER-ALZ 2 randomized clinical trial. JAMA, 2023. europepmc.org/article/MED/37459141
3. Pittock RR, Aakre JA, Castillo AM, et al. Eligibility for anti-amyloid treatment in a population-based study of cognitive aging. Neurology, 2023. europepmc.org/article/MED/37586881
4. US Food and Drug Administration. FDA converts novel Alzheimer’s disease treatment to traditional approval. FDA, 2023. fda.gov/…/fda-converts-novel-alzheimers-disease-treatment-traditional-approval
5. Nonino F, Minozzi S, Sambati L, et al. Amyloid-beta-targeting monoclonal antibodies for people with mild cognitive impairment or mild dementia due to Alzheimer’s disease. Cochrane Database of Systematic Reviews, 2026. europepmc.org/article/MED/41985900
6. National Institute for Health and Care Excellence. Lecanemab for treating mild cognitive impairment or mild dementia caused by Alzheimer’s disease [ID4043]: draft guidance consultation 3. NICE, 2026. nice.org.uk/consultations/3281/1/recommendations
7. US Food and Drug Administration. FDA approves treatment for adults with Alzheimer’s disease (Kisunla). FDA, 2024. fda.gov/…/fda-approves-treatment-adults-alzheimers-disease
8. Dangpiaei S, Kamal KM, Shoair OA, Al-Mamun MA. Incidence of amyloid-related imaging abnormalities and health resource utilization in patients with Alzheimer disease receiving monoclonal antibody treatments: a real-world evidence study. Journal of Managed Care & Specialty Pharmacy, 2026. europepmc.org/article/MED/41885326
9. European Medicines Agency. Leqembi (lecanemab): EPAR overview. EMA, 2025. ema.europa.eu/en/medicines/human/EPAR/leqembi
10. European Medicines Agency. Kisunla (donanemab): EPAR overview. EMA, 2025. ema.europa.eu/en/medicines/human/EPAR/kisunla
11. National Institute for Health and Care Excellence. Lecanemab for treating mild cognitive impairment or mild dementia caused by Alzheimer’s disease [ID4043]: project information. NICE, 2026. nice.org.uk/guidance/indevelopment/gid-ta11220
12. National Institute for Health and Care Excellence. Donanemab for treating mild cognitive impairment or mild dementia caused by Alzheimer’s disease [ID6222]: project information. NICE, 2026. nice.org.uk/guidance/indevelopment/gid-ta11221
13. Dittrich A, Westman E, Shams S, et al. Proportion of community-dwelling individuals older than 70 years eligible for lecanemab initiation: the Gothenburg H70 Birth Cohort Study. Neurology, 2024. europepmc.org/article/MED/38593394
14. Rosenberg A, Solomon A, Bonnard A, et al. Preparing for the implementation of anti-amyloid therapies in Europe: assessing real-world eligibility for lecanemab and donanemab in a Swedish memory clinic. Journal of Prevention of Alzheimer’s Disease, 2026. europepmc.org/article/MED/41544372
15. Kang SH, Jeong JH, Pyun JM, et al. Eligibility for lecanemab treatment in the Republic of Korea: real-world data from memory clinics. Journal of Clinical Neurology, 2025. europepmc.org/article/MED/40308013
16. Grill JD, Flournoy C, Dhadda S, et al. Eligibility rates among racially and ethnically diverse US participants in phase 2 and phase 3 placebo-controlled, double-blind, randomized trials of lecanemab and elenbecestat in early Alzheimer disease. Annals of Neurology, 2024. europepmc.org/article/MED/37830926
17. Mattke S, Tang Y, Hanson M. Expected wait times for access to a disease-modifying Alzheimer’s treatment in England: a modelling study. Journal of Health Services Research & Policy, 2024. europepmc.org/article/MED/37931615
18. Roth S, Yan J, Patru MM, et al. The capacity for Alzheimer’s disease confirmatory testing in the United States: the current situation and simulations for future increase. Journal of Alzheimer’s Disease, 2026. europepmc.org/article/MED/41555795
19. US Food and Drug Administration. FDA clears first blood test used in diagnosing Alzheimer’s disease. FDA, 2025. fda.gov/…/fda-clears-first-blood-test-used-diagnosing-alzheimers-disease
20. Mattke S, Chen J, Hanson M, et al. Projected effect of a high performing blood test on wait times for determination of Alzheimer’s disease pathology in the United States. Journal of Alzheimer’s Disease, 2026. europepmc.org/article/MED/42461071