Bold claim: Novo Nordisk is defending its decision after GLP-1 Alzheimer's trials failed, arguing the scientific question was worth pursuing despite setbacks. This is the core tension driving the discussion.
Novo Nordisk, the Danish pharmaceutical company, launched pivotal trials in 2020 to evaluate semaglutide, its GLP-1 diabetes medication, as a potential treatment for Alzheimer’s disease. The decision followed a cascade of supporting evidence from human, animal, and real-world data, but the company later acknowledged criticisms that the study design had flaws. The trials did not demonstrate a statistically significant slowing of cognitive decline in participants treated with the drug.
Nevertheless, Novo Nordisk maintains that pursuing an answer to the underlying scientific question was the correct course. Peter Johannsen, the company’s international medical vice president, stated at the Clinical Trials in Alzheimer’s Disease meeting in San Diego that, while the trials failed to meet their primary goals, the research was still justified from a scientific standpoint.
Novo’s analysis now relies on consolidated data indicating that the GLP-1 hormone participates in neurotransmission and has multiple effects across the brain. Still, Johannsen cautions that Alzheimer’s disease is inherently complex. He emphasized that its pathology involves various factors, including different genetic signatures, and that many aspects remain poorly understood despite the established role of amyloid plaques.
Two 2-year studies tested Rybelsus, Novo’s oral GLP-1 diabetes medication, against placebo in nearly 4,000 individuals with Alzheimer’s disease. Novo plans to share initial findings on Wednesday, with full results slated for a separate medical meeting in March. A brief press release last week indicated that the studies did not meet their predefined endpoints.
Regarding cognitive outcomes observed in diabetes patients taking GLP-1s, Johannsen noted retrospective analyses suggesting benefits after approximately a year of treatment, with improvements potentially continuing with longer use. He pointed out that many of these studies did not clearly distinguish Alzheimer’s disease from other dementia types. Some real-world evidence relied on clinical diagnoses rather than precise biomarkers such as amyloid imaging.
Current estimates from the Alzheimer’s Association place the proportion of dementia cases attributable to Alzheimer’s at about 60%, with remaining cases arising from vascular and other etiologies. Johannsen highlighted potential biases in real-world analyses. He explained that GLP-1–treated diabetes patients often have access to endocrinologists and may come from higher socioeconomic groups, factors that could influence health outcomes and dementia diagnosis timelines. Additionally, these patients are more likely to have better glycemic and metabolic control, which might delay progression to dementia or the point at which diagnosis occurs.
The story continues as Novo Nordisk prepares to present the early results of the two-year trials, while the broader scientific community weighs the implications for GLP-1 therapies in neurodegenerative disease. Thoughts on whether a scientifically sound negative result should deter exploration of GLP-1 pathways, or whether alternative trial designs might better capture potential benefits, are welcome in the comments.
