On March 19, at the AD/PD 2026 International Conference on Alzheimer’s and Parkinson’s Diseases and Related Neurological Disorders, during a symposium focused on oral semaglutide in early Alzheimer’s disease, which includes mild cognitive impairment (MCI) and mild Alzheimer’s, Novo Nordisk presented full data from the EVOKE (NCT04777396) and EVOKE+ (NCT04777409) trials for the first time. Top line results for these trials were announced in November 2025. Semaglutide is a glucagon-like peptide-1 receptor agonist (GLP-1RA). Since GLP-1RAs are currently a hot topic in the pharma industry, as they have made waves in the metabolic space, there was high anticipation for their potential to affect a neurodegenerative disease such as Alzheimer’s. Therefore, the failure of oral semaglutide in the two EVOKE trials has been hugely disappointing for the Alzheimer’s and neurodegenerative field.
The results of the EVOKE and EVOKE+ trials were comprehensive in showing oral semaglutide’s lack of clinical efficacy for cognition or function in patients with early Alzheimer’s. In both trials, there was no difference between oral semaglutide and placebo at the primary endpoint of change from baseline in Clinical Dementia Rating – Sum of Boxes (CDR-SB) score at week 104 (two years), or at the secondary endpoint of change in Alzheimer’s Disease Cooperative Study – Activities of Daily Living Scale for MCI (ADCS-ADL-MCI) score from baseline to week 104. The same trend was seen for the other secondary endpoints, such as Mini-Mental State Examination (MMSE) score and Alzheimer’s Disease Assessment Scale–Cognitive Subscale (ADAS-Cog) score. A pooled analysis across both trials also showed that oral semaglutide did not delay progression from MCI to mild Alzheimer’s, as there was no difference between the oral semaglutide and placebo groups. A further exploratory analysis of pre-defined subgroups, including sex, age, race, body mass index (BMI), and type 2 diabetes (T2D) status, also failed to provide any strong signals of efficacy for a particular subgroup of early Alzheimer’s patients.
Go deeper with GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
Despite the conclusive negative efficacy results, there was a significant impact on biomarkers in the EVOKE trials. In patients who were treated with oral semaglutide, there was a significant reduction in cerebrospinal fluid (CSF) biomarkers of Alzheimer’s pathology, such as p-tau181, at week 78 compared with the placebo group. Despite the significance of this finding, the reduction was only around 10%. Experts agreed that this reduction, in combination with the negative clinical outcomes in the trials, was too small to provide any real benefit for patients. However, experts noted that this finding does raise the need for more investigation to better understand the standard amyloid- and tau-based biomarkers in the context of non-amyloid and non-tau therapies.
Semaglutide was being evaluated in Alzheimer’s following consistent real-world and observational evidence from multiple studies showing that patients receiving GLP-1RAs for obesity and T2D had a reduced risk of all-cause dementia, as well as AD specifically. Furthermore, two small clinical trials (NCT01469351, NCT01843075) had shown the possible benefits of GLP-1RAs in mild-to-moderate Alzheimer’s, although no clinical conclusions could be drawn because of their sample sizes and study designs. It is therefore surprising that the EVOKE trials were so comprehensively negative. However, there were some key differences between the real-world evidence (RWE) and the clinical trials. Notably, the RWE was looking at patients with metabolic pathologies (T2D and obesity), whereas regulators specified that patients had to have confirmed the pathology for an Alzheimer’s trial, creating two very different patient populations. Additionally, the RWE was looking at a time to diagnosis of Alzheimer’s as a kind of endpoint, whereas the endpoint for the EVOKE clinical trials was measuring the slowing of disease progression in patients with already-established Alzheimer’s, creating a different question about the effects of semaglutide.
When the top-line results were first presented in November 2025, Novo Nordisk announced that the one-year extension period in both trials would be discontinued and that the program would be terminated. Despite the failure of the EVOKE trials, panellists at the symposium did not believe that this should mark the conclusion of investigations into the use of GLP-1RAs in AD. Areas of research remain for evaluating semaglutide in a patient population that has comorbid Alzheimer’s and metabolic disorders / vascular pathology, or as a preventive therapy. Although interest in the field remains, any further progress for GLP-1RAs will require significant time and investment to run further clinical trials for a mechanism of action / drug class that now has a high-profile failure behind it.
