Update on Alzheimer’s disease treatments

SPECIAL REPORT 25.15

Update on Alzheimer’s disease treatments

The year 2025 will not become a positive milestone for European Alzheimer’s patients. This updated summary of major events should be a convincing advocacy.

In November 2024, the EMA’s Committee for Medicinal Products for Human Use (CHMP) issued a positive opinion recommending marketing authorisation for lecanemab (brand name Lequembi) from Eisai/Bavarian Nordic/Biogen. In early 2025, the European Commission (EC) asked CHMP to re-examine new safety information. On 15 April, the EC adopted an EU-wide marketing authorisation for the treatment of early Alzheimer’s disease. This authorisation was accompanied by risk-minimisation measures, including strict controlled access, MRI monitoring for amyloid-related imaging abnormalities (ARIA) and specific risk-management obligations. With these restrictions the commercial success of Lequembi is in strong jeopardy in Europe. For instance, on 9 September, the French Haute Autorité de Santé (HAS) refused the “accès précoce” (early access) request for the drug stating that the benefit-risk balance was not sufficient to grant early access. Even more so, on 7 November, HAS issued an unfavourable opinion on reimbursement given uncertainties about long-term efficacy and major safety concerns.

In March 2025, CHMP initially issued a negative opinion on donanemab (brand name Kisunla) from Eli Lilly but reversed it after re-examination and issued a positive recommendation in July. The EC granted EU-wide marketing authorisation for the drug for early symptomatic Alzheimer’s disease on 25 September. The approval limits use to patients with one or no copies of the APOE4 gene and includes safety measures to manage ARIA abnormalities like the ones for Lequembi. It is too early to get any indication on its commercial impact.

On 25 November, Novo Nordisk reported that semaglutide (a GLP-1 agonist) an approved drug for treatment of diabetes and obesity failed in two pivotal phase 3 clinical trials (EVOKE and EVOKE+) in Alzheimer’s disease. At the 18th
Clinical Trials on Alzheimer's Disease (CTAD) conference, held December 1-4 in San Diego, Novo Nordisk released the full details of the trials.

The studies and the results.

EVOKE (n = 1855) and EVOKE+ (n =1953) are considered as one of the largest Alzheimer’s disease trials including 3808 adults aged 55 to 85 years old with mild cognitive impairment (MCI). The randomly assigned subjects with MCI received either once a day 14 mg of semaglutide or placebo for an eight-week titration period followed by 104 weeks of treatment plus a 52-week extension. Semaglutide failed to show any superiority over placebo on the primary end point in CDR-Sum of Boxes (CDR-SB). Secondary outcomes, which include Alzheimer Disease Cooperative Study of Daily Living Scale for MCI (ADCS-ADL-MCI) confirmed that semaglutide had no impact on slowing functional decline. Overall semaglutide did not delay time to progression to dementia. In both trials there was a sub study that included 199 patients, 98 on oral semaglutide and 101 on placebo.

In the semaglutide cohort, one could identify several notable CSF biomarkers that demonstrated low but significant reductions: pTau181 (8%), pTau217 (9%), npT181 (10%), npTau205 (9%) as well YKL-40 (7%), a neuroinflammation and neurodegeneration marker (total tau 7% and neurogranin 8%). Based on these results Novo Nordisk decided to discontinue the one-year extension period of the trials.“This is not what we had hoped for”said Dr. Peter Johannsen, Novo Nordisk’s International Medical Vice-President of Obesity and Cardiometabolism. It would have taken figures closer to 20% or 25% to expect translation into observable clinical benefit.

“This is not what we had hoped for”
— Dr. Peter Johannsen, Novo Nordisk VP

Experts in the field, who were not involved in the study, commented the results.

• The possibility that the oral low dose of 14 mg/day was too low, lower than what is used for other indications.
• The fact that the trial was monitoring CDR-SB, a scale used to slow progression of the disease by clearing amyloid plaques, possibly not the best measure for this mechanism of action.
• The possibility that the selected participants in the trials have been treated too late in their disease. “For a metabolic-acting drug, this may simply be too late” said Professor Paul Edison from Imperial College London.
• The fact that investigators recorded significant decreases in hsCRP (24% in EVOKE) and (29% in EVOKE+) is a clear demonstration that the “systemic inflammation fire was extinguished”. 

Despite these little discouraging results, Novo Nordisk with external experts still believes that GLP-1 agonists have the potential to help treat Alzheimer’s disease and are now considering additional fundamental preclinical work.

"For a metabolic-acting drug, this may simply be too late"
— Professor Paul Edison, Imperial College London

To the best of your knowledge there are no large ongoing phase 3 studies where results are expected in 2026.

On the tau front news are not exciting anymore. Lilly’s anti-tau candidate (LY3372689 or ceperognastat) had not met any endpoints and was stopped in 2024. Earlier this year Asceneuron discontinued work on ASN51, an orally active anti-tau drug and Biogen did the same with BIIB113.

In April, Roche had announced that in a phase 1b/2a study, trotinemab, a beta amyloid antibody engineered to cross the blood-brain barrier, achieved significant reductions in amyloid levels, with 81% of participants reaching below a critical threshold after 28 weeks, without significant side effects. Results of a first pivotal study could become available in 2027.

All over, unfortunately, 2025 did not provide anything very promising for the 55 million Alzheimer's patients and their caregivers.

This is a test

^{This document has been prepared by Jean-Claude Muller and is provided for information purposes only. The information contained herein has been obtained from sources believed to be reliable but is not warranted to be accurate or complete. The views presented are those of the author at the time of writing and are subject to change. Jean-Claude Muller has no obligation to update these opinions or the information presented.}