COPENHAGEN — Novo Nordisk’s Alzheimer’s trials of its drug semaglutide did not slow early-stage disease in more than 3,800 patients with mild cognitive impairment or early Alzheimer’s enrolled in the global EVOKE and EVOKE+ phase 3 program, prompting the company to nix an already planned one-year extension and sending its shares down by about 10 percent on Wednesday.
The once-a-day oral semaglutide pill, which is marketed for type 2 diabetes as Rybelsus, failed to beat a placebo on the Clinical Dementia Rating–Sum of Boxes (CDR-SB) scale across two years, even though it activated several Alzheimer ‘s-signaling biomarkers, according to a company statement.
EVOKE and EVOKE+ included model-confirmed Alzheimer’s patients between 55 and 85 years of age, randomizing 3,808 people to receive either semaglutide in oral form at a dose of 14 mg or placebo over an initial treatment period of 104 weeks, followed by a planned extension of up to an additional 52 weeks that will now be discontinued.
Martin Holst Lange, Novo Nordisk’s chief scientific officer and executive vice president for Research and Development, said in a statement that the company “owed it to society” to develop semaglutide despite a “low probability of success” and thanked “participants and caregivers who joined in the trials.”
CEO Mike Doustdar, who had called EVOKE a “lottery ticket” previously, told investors, “our efforts to slow the progression of Alzheimer’s disease are over,” while emphasizing management never saw a high probability of success.
Shares in Novo Nordisk tumbled about 9 percent to their lowest since mid-2021 in Copenhagen. U.S.-listed stock was down almost 10 percent in premarket trading, the Reuters report noted, deepening a year-long slide for the GLP-1 standard bearer.
Analysts at banks such as JPMorgan and Barclays call the semaglutide Alzheimer’s program an upside option rather than a core profit driver, with one describing a slide in shares of around 9% as an “overreaction” in a separate Reuters analysis, since few models assume meaningful dementia sales.
A design paper in the journal Alzheimer’s Research & Therapy earlier this year noted that EVOKE and EVOKE+ were powered to detect a drop of about 20% in CDR-SB, assuming placebo patients would worsen by roughly 2.2 points over 2 years.
In October, the Alzheimer’s Association informed families that if the results were promising, its EVOKE program would be able to fast-track semaglutide into dementia care as a potential GLP-1 treatment to dampen brain inflammation and protect neurons.
A BrightFocus Foundation explainer published in August also hinted that GLP-1 drugs like Ozempic and Wegovy might slow Alzheimer’s progression by flushing out toxic proteins, cooling neuroinflammation, and improving brain energy use, framing the EVOKE readout as a critical test of that hypothesis.
Some of those hopes were based on real-world data in people with diabetes and obesity. A study using electronic health records published in JAMA Network Open, which was summarized by the cardiovascular outlet TCTMD, found that semaglutide and tirzepatide use was associated with lower seven-year risks of dementia, stroke, and death compared to other glucose-lowering drugs.
Further observational studies — including work promoted by the Cleveland Clinic and a major U.S. veterans study described in the Guardian — even suggested GLP-1 receptor agonists might reduce the risk of dementia and dozens of other conditions tied to obesity, strengthening the case for directly testing them in early Alzheimer’s disease.
For now, however, declining topline EVOKE readouts will put a damper on the excitement for semaglutide in Alzheimer’s monotherapy — though some would argue, as Baird analyst Brian Skorney did previously, that new and improved GLP-1 agents could still be tested in Alzheimer’s down the line.
Advocates for Alzheimer’s emphasized that the work is not wasted. Dr. Joanne Pike, president and chief executive officer of the Alzheimer’s Association, said the results would still “add to our understanding” of the disease, while Alzheimer’s Research UK focused on combination approaches targeting multiple biological pathways.
Research entities like the Alzheimer’s Drug Discovery Foundation view semaglutide’s Alzheimer’s failure as a watershed, noting in a statement that the GLP-1s may ultimately become adjuncts in “cocktail” regimens rather than stand-alone disease-modifying therapeutics.
The overall Alzheimer’s pipeline is still chock-a-block. A review published earlier this year on Medscape provides a catalogue of dozens of non-amyloid mechanisms under investigation — including inflammation and vascular health, synaptic resilience, and others — emphasizing that the GLP-1s are now just one part of a much larger post-amyloid research strategy.
Novo stressed that semaglutide’s place in treatment for type 2 diabetes and obesity, and in reducing cardiovascular risk, remains intact, and that safety in EVOKE seemed consistent with its previous trials, but conceded that losing one executive described as a high-risk “lottery ticket” it held on the drug helping Alzheimer’s.
Top-line results from EVOKE and EVOKE+ will be available at the Clinical Trials in Alzheimer’s Disease (CTAD) meeting Dec. 3, with a full analysis due at the Alzheimer’s and Parkinson’s Diseases Conference (AD/PD) in March 2026 when researchers expect biomarker data and subgroup analyses to provide insights into why semaglutide’s hopes for people with Alzheimer’s failed to translate into clinical benefit.
