CTAD Presentation Lays Insights Into Disappointing Phase 3 EVOKE Trial of GLP-1 Semaglutide in Alzheimer Disease

Full details behind the pivotal phase 3 EVOKE (NCT0477396) and EVOKE+ trials (NCT4777409) testing semaglutide (Ozempic; Novo Nordisk), a glucagon-like peptide-1 (GLP-1) receptor agonist, in patients with Alzheimer disease (AD) were presented at the 18th Clinical Trials on Alzheimer’s Disease (CTAD) conference, held December 1-4 in San Diego, California.

All told, treatment with the weight loss medication did not result in significantly slowed disease progression, despite patients showing improvements in plasma high sensitivity C-reactive protein (hs-CRP) and AD-relevant biomarkers. Encouragingly, safety and tolerability for oral semaglutide in participants with early AD was consistent with studies in other indications.

EVOKE (n = 1855) and EVOKE+ (n = 1953), considered 1 of the largest AD trials of a GLP-1 receptor agonist (RA), included 3808 adults aged 55 to 85 years with mild cognitive impairment (MCI) or mild dementia because of AD. Coming into the study, patients had a Clinical Dementia Rating (CDR) Global score of 0.5, were amyloid-positive, and were allowed to continue or initiate stable approved AD medications, including recently marketed monoclonal antibodies.

In the trial, patients were randomly assigned to either once-daily semaglutide 14 mg or placebo for an 8-week titration period, followed by 104 weeks of treatment plus a 52-week extension. The efficacy and exploratory results were presented by Jeffrey Cummings, MD, ScD, a research professor with the department of brain health at University of Nevada, Las Vegas Kirk Kerkorian School of Medicine.

Efficacy Results

All told, semaglutide failed to distinguish itself from placebo on the primary end point of change in CDR-Sum of Boxes (CDR-SB) over a 104-week stretch, or 2-year period. Overall, there was an estimated –0.06-point (95% CI, –0.48 to 0.36) difference between oral semaglutide (2.2) and placebo (2.2; P = .7727) in EVOKE during that time. For EVOKE+, which included patients with vascular pathology, there was a 0.15-point difference (95% CI, –0.24 to 0.54) between semaglutide (2.1) and placebo (2.0), which was also not significant (P = .4604).

Coming into the study, 46.7% of patients were heterozygous APOE4 carriers and 12.3% were homozygous APOE4 carriers. In addition, 13.6% of patients had type 2 diabetes, and relatively small number of patients had concurrent significant small-vessel pathology (1.4%). Notably, close to 60% of patients came into the study on a concomitant AD medication, with donepezil (36.3%) and memantine (11.9%) as the most used medications.

Secondary outcomes, which included Alzheimer’s Disease Cooperative Study Activities of Daily Living Scale for MCI (ADCS-ADL-MCI) score, there was no significant difference between treatment groups. Across EVOKE and EVOKE+, investigators recorded ADCS-ADL-MCI treatment differences of 0.43 (oral semaglutide: –6.6; placebo: –7.1) and –0.21 (oral semaglutide: –6.5; placebo: –6.3), respectively, further confirming semaglutide’s lack of impact on slowing functional decline.

Overall, oral semaglutide did not delay time to progression to dementia in participants with MCI across both trials. Using an event-analysis, which pooled both studies (semaglutide: n = 1369; placebo: n = 1377), the investigators reported a hazard ratio (HR) of 0.96 (95% CI, 0.86-1.06). Similarly, there was no beneficial effects on cognition and function, using scales like Montreal Cognitive Assessment, ADAS-Cog13, MMSE, and ADCOMS.

Exploratory Results

In EVOKE and EVOKE+, there was a substudy that included 199 patients, 98 of which were on oral semaglutide and 101 patients on placebo. In the semaglutide cohort, there were several notable CSF biomarkers that demonstrated nominally significant reductions of around 10%. These included AD biomarkers like pTau181 (treatment ratio, 0.92), pTau217 (0.91), npT181 (0.90), and npTau205 (0.91), as well as other markers of neuroinflammation (YKL-40; 0.93) and neurodegeneration (total tau: 0.93; neurogranin: 0.92).

Across other exploratory blood-based biomarkers, there was a significant increase of around 5% in neurofilament light in the EVOKE+ trial (estimated treatment ratio, 1.05; 95% CI, 1.01-1.08), as well as a significant increase of around 4% in glial fibrillary acidic protein (GFAP) in both trials. Coming into the study, approximately two-third of patients had eMTBR-tau243 below lower limit of quantification. Following semaglutide treatment, investigators recorded significant decreases in hsCRP, with estimated treatment ratios of 0.76 (95% CI, 0.64-0.90) in EVOKE and 0.71 (95% CI, 0.62-0.82) in EVOKE+.

Safety Results

The pooled EVOKE and EVOKE+ safety results were presented by Filip Knop, MD, PhD, senior vice president and chief medical officer at Novo Nordisk. In terms of safety, adverse events (AEs) occurred were found in 91.2% (n = 1896) in those on semaglutide vs 84.8% of those on placebo. There were no treatment differences between the two groups on serious AEs (20.4% vs 23.8%) and severe AEs (12.1% vs 13.2%). The AEs occurring in more than 10% of patients on semaglutide were body weight decreases (36.5%), decreased appetite (33.1%), nausea (24.3%), diarrhea (14.5%), vomiting (12.3%), and COVID-19 (10.9%). All of these were mild to moderate in nature, and in line with the safety profile of semaglutide.

Across both trials, the mean body weight change from baseline at week 104 was –5.8% (–4.3 kg) for oral semaglutide versus increases of +0.6% (+0.2 kg) for placebo. The most notable change was found in those with body mass index of more than 30, with percent changes of –8.6%.

EVOKE and EVOKE+ were notable in that it enrolled patients in the earliest symptomatic stages of AD–those with MCI or mild dementia confirmed with amyloid pathology. The trial leveraged robust randomization, long treatment durations, and biomarker-rich end points, including MRI, plasma/cerebrospinal fluid (CSF) markers, and cognitive composites, offering one of the most comprehensive looks at GLP-1 based neuroprotection to date.

Phase 3 development of semaglutide in AD was supported by converging evidence from preclinical models, real-world datasets, and post hoc cardiovascular trial analyses, including a target trial emulation of electronic health records from more than 1 million patients with type 2 diabetes showing a 40% to 70% lower risk of first-time AD diagnosis with semaglutide compared with other antidiabetic medications, including other GLP-1 receptor agonists.²

Over 3 years of follow-up, semaglutide was associated with significantly reduced AD risk, with a hazard ratio (HR) of 0.33 (95% CI, 0.21-0.51) versus insulin and 0.59 (95% CI, 0.37-0.95) versus other GLP-1RAs. In addition, among adults aged 60 years and older (mean age 67.9), the 3-year incidence of first-time AD diagnosis remained twice that of the broader population (0.33% vs 0.16%), reinforcing the potential relevance of these findings.

Based on the negative data presented at CTAD, Novo decided to discontinue the 1-year extension period of the trials. Additional data from the substudy is also expected to be presented at CTAD 2025.

Click here for more CTAD 2025 coverage.

REFERENCES
1. Cummings J, Johannsen P, Atri A, et al. evoke and evoke+: Two phase 3 randomised placebo-controlled trials of semaglutide in participants with early-stage Alzheimer’s disease (NCT04777396 and NCT04777409). Presented at: 2025 CTAD Conference; December 1-4; San Diego, CA.
2. Wang W, Wang Q, Qi X, et al. Associations of semaglutide with first-time diagnosis of Alzheimer's disease in patients with type 2 diabetes: Target trial emulation using nationwide real-world data in the US. Alzheimers Dement. 2024;20(12):8661-8672. doi:10.1002/alz.14313