Two major clinical trials evaluating the potential of the weight-loss drug semaglutide to treat Alzheimer’s disease have concluded without showing the desired clinical benefit. The manufacturer, Novo Nordisk, reported that patients taking the medication did not experience a slower decline in memory or thinking skills compared to those taking a placebo. While the drug successfully altered certain biological markers associated with the disease, these biological changes did not result in a measurable preservation of cognitive function for the participants.
The investigation into semaglutide was driven by a growing body of evidence linking metabolic health to brain function. Semaglutide belongs to a class of medications known as GLP-1 receptor agonists. These drugs are primarily prescribed to manage type 2 diabetes and to assist with weight loss under brand names such as Ozempic and Wegovy.
The hormone GLP-1, or glucagon-like peptide-1, helps the body regulate blood sugar levels. However, receptors for this hormone are also found in the brain. Activation of these receptors has been shown to reduce inflammation and improve the health of blood vessels in animal studies. This led researchers to hypothesize that the drug might protect brain cells from the damage caused by Alzheimer’s.
Observational data further supported this hypothesis. Researchers had noted that patients with diabetes who were treated with GLP-1 drugs seemed to develop dementia at lower rates than those who were not. This correlation sparked hope that the medication could be repurposed to treat neurodegeneration directly.
Novo Nordisk, a Danish healthcare company with a long history in diabetes care, initiated the Phase 3 “evoke” and “evoke+” trials to test this theory rigorously. These were large-scale, international studies designed to provide a definitive answer regarding the drug’s efficacy for Alzheimer’s.
The trials enrolled a combined total of 3,808 adults. The participants ranged in age from 55 to 85 years old. All subjects had been diagnosed with either mild cognitive impairment or mild dementia due to Alzheimer’s disease.
Additionally, all participants had confirmed amyloid positivity. This means they had a buildup of amyloid plaques in the brain, which is a hallmark physical sign of Alzheimer’s pathology. This inclusion criterion ensured that the study focused specifically on Alzheimer’s and not other forms of dementia.
The study design was randomized and double-blind. This is considered the gold standard for clinical research. Half of the participants received a daily oral dose of 14 milligrams of semaglutide. The other half received a placebo pill that looked identical but contained no active medication. Neither the patients nor the doctors knew who was receiving which treatment until the trial concluded.
The researchers monitored the participants for over three years. The primary analysis focused on changes observed over a 104-week treatment period. To measure the drug’s effectiveness, the study used a tool called the Clinical Dementia Rating – Sum of Boxes, or CDR-SB.
The CDR-SB is a standardized interview process. Clinicians assess patients and interview their caregivers to rate performance in six areas. These areas include memory, orientation, judgment, community affairs, home and hobbies, and personal care. A higher score indicates more severe impairment.
The goal was to see if the group taking semaglutide had a lower increase in their CDR-SB scores compared to the placebo group over the two years. A lower increase would indicate that the disease was progressing more slowly.
On November 24, 2025, Novo Nordisk announced that this primary goal was not met. The statistical analysis showed no significant difference in the rate of decline between the two groups. The disease progressed at roughly the same speed regardless of whether the patient took semaglutide or the placebo.
Despite the lack of clinical benefit, the study did reveal that semaglutide had an effect on the body. The company noted that the drug led to improvements in “Alzheimer’s disease-related biomarkers.” Biomarkers are measurable indicators of what is happening inside the body, such as levels of specific proteins in the blood or spinal fluid.
This creates a complicated picture where the biology of the patient appeared to respond to the drug, but their mental, functional state did not improve. This disconnect highlights the difficulty of translating biological changes into meaningful improvements in a patient’s daily life.
Experts responding to the news through the Science Media Centre emphasized that while the results are a setback for treatment, they do not necessarily rule out the drug’s potential for prevention. The timing of the intervention may be the deciding factor.
Ivan Koychev, a Clinical Associate Professor in Neuropsychiatry at Imperial College London, commented on this distinction. He noted that the earlier data that inspired the trial pointed toward a preventive signal rather than a curative one.
“In large real-world datasets, people exposed to GLP-1 receptor agonists over many years appear to have a lower risk of ever developing dementia,” Koychev said in a statement to the Science Media Centre. He explained that this suggests the drug influences the earliest stages of the disease process rather than established neurodegeneration.
Koychev further elaborated on why the trial might have failed despite the biomarker improvements. By the time a person has mild cognitive impairment, significant damage to the brain’s neural networks has already occurred.
“At this point in the disease course, even meaningful shifts in fluid biomarkers may not translate into measurable improvements in cognition or daily functioning over a two-year trial,” Koychev stated.
This sentiment was echoed by Prof. Tara Spires-Jones, Director of the Centre for Discovery Brain Sciences at the University of Edinburgh. She described the results as disappointing but noted the biomarker data offers a specific insight.
“Treatment did improve Alzheimer’s-related biomarkers leaving open a tiny window of hope that in future this drug might be effective if used earlier as a preventative strategy,” Prof. Spires-Jones told the Science Media Centre.
The safety profile of the drug in these trials was consistent with its known effects. Semaglutide is widely used and has been studied extensively. The trial data indicated it was well-tolerated by the older population in the study.
However, the lack of efficacy in slowing symptom progression means Novo Nordisk will discontinue the extension period of the trials. The company stated it would present the full data at upcoming medical conferences in late 2025 and early 2026.
The failure of the evoke trials reinforces the complexity of Alzheimer’s disease. It is not a condition driven by a single mechanism. Consequently, treating it may require targeting multiple biological pathways simultaneously.
Susan Kohlhaas, Executive Director of Research and Partnerships at Alzheimer’s Research UK, highlighted this necessity. She pointed out that no single approach is likely to be sufficient.
“The field now needs to focus on understanding those processes in much greater detail and developing treatments that can be used together to tackle the disease from multiple angles,” Kohlhaas said in her comments to the Science Media Centre.
Professor Paul Morgan, Director of the UK Dementia Research Institute Cardiff, also weighed in on the duration of the study relative to the disease’s development. He noted that the impact of risk factors like obesity and diabetes accumulates over decades.
“In the trials, subjects were treated for two years, while the impacts of obesity and diabetes on dementia likely accumulate over many years,” Morgan noted via the Science Media Centre. He expressed hope that the widespread use of these drugs in the general population might still reveal long-term protective effects against developing Alzheimer’s.
For now, the immediate hope that oral semaglutide could serve as a treatment for those already showing symptoms of Alzheimer’s has been dampened. The focus of research may now shift toward earlier intervention.
The theory is that protecting the brain’s metabolic and vascular health must happen before neurons die and cognitive connections are severed. Once the structural damage of dementia is established, metabolic drugs alone appear insufficient to reverse or halt the decline.
Fiona Carragher, Chief Policy and Research Officer at the Alzheimer’s Society, reminded the public that negative results are still a vital part of the scientific process. They help refine where researchers should look next.
“No trial is wasted,” Carragher said to the Science Media Centre. “Every investigation helps us develop better drugs and design better trials in the future.”
The scientific community will now await the detailed presentation of the data. Researchers will analyze the specific biomarkers that changed. This analysis will help determine if GLP-1 drugs should remain a part of the conversation for Alzheimer’s prevention, even if they are not the solution for treatment.
