Real‑World Study Finds 80% of Alzheimer’s Patients Tolerate New Drug Even With ARIA Risks
New therapy lecanemab clears Alzheimer’s brain protein, boosting patient outcomes in latest study.
Real‑world evidence supports the safety of lecanemab for early Alzheimer’s disease
A recent Duke Health analysis of more than 230 individuals receiving lecanemab – the FDA‑approved anti‑amyloid antibody – shows that outcomes observed in controlled trials are being replicated in routine clinical practice.
How the investigation was conducted
The cohort comprised patients diagnosed with early Alzheimer’s disease or mild cognitive impairment who began lecanemab therapy at Duke between 2023 and 2025. Researchers recorded adverse events, performed serial MRI examinations using specialized sequences to detect microhemorrhages, administered cognitive assessments, and tracked treatment continuity for up to 396 days.
Retention and adverse‑event profile
Approximately 79 % of participants remained on lecanemab for at least one year, while the remaining 21 % discontinued primarily because of side effects. Serious adverse events occurred in about one‑third of the sample, most frequently infusion‑related reactions, falls or strokes, although several of these incidents were linked to pre‑existing conditions rather than the medication itself.
Frequency and risk factors for amyloid‑related imaging abnormalities
Overall, 24.3 % of patients experienced ARIA, a spectrum of brain swelling or bleeding visible on MRI. Individuals homozygous for the APOE ε4 allele were roughly four times more likely to develop ARIA, with no apparent gender difference. ARIA episodes could persist for up to 30 weeks after treatment initiation, showing two incidence peaks near weeks 10 and 25. Importantly, ARIA status did not correlate with changes in cognitive performance after 12 months.
Current clinical, laboratory, and imaging markers – whether evaluated alone or in combination – failed to reliably forecast ARIA onset. Among the tested approaches, an MRI‑derived vascular metric paired with blood concentrations of the pTau/AB42 ratio showed the greatest potential for identifying low‑risk patients.
Artificial‑intelligence imaging tool bolsters monitoring
As part of the safety protocol, investigators trialed an AI‑driven system designed to flag subtle ARIA changes on brain scans more promptly and consistently. The team plans to broaden the application of this technology to enhance patient surveillance.
Expert commentary on risk management
Andrew “Andy” Liu, associate professor of neurology and pathology at Duke University School of Medicine, emphasized that careful patient selection and vigilant monitoring enable most individuals to stay on therapy without requiring additional medication or hospitalization. Co‑author P. Murali Doraiswamy highlighted that cerebral amyloid angiopathy – present in over half of Alzheimer’s patients – likely underlies ARIA and underscored the urgency of developing superior predictive tests.
Heather Whitson, co‑author and co‑director of the Duke & UNC Alzheimer’s Disease Research Center, reminded clinicians that treatment decisions must be tailored to each patient’s circumstances.
Patient story illustrates practical impact
Sally Osmer, a 74‑year‑old diagnosed with early Alzheimer’s disease, consented to lecanemab shortly after her diagnosis. She began infusions within a month and has now completed more than a year of therapy without serious side effects. Osmer reported that the medication has allowed her to stay active in daily life and maintain close contact with her grandchildren, providing a sense of optimism about the future.
Implications and future research pathways
Although lecanemab does not cure Alzheimer’s disease, slowing disease progression can preserve independence and improve quality of life for patients like Osmer while newer treatments emerge. Liu’s laboratory is assembling a biorepository of blood and cerebrospinal fluid samples from treated individuals to pursue more accurate biomarkers for cerebral amyloid angiopathy and ARIA.
The study, published in Neurology Open Access, was funded in part by the Ann B. Bussel Award and a Duke‑UNC NIH grant. Several authors disclosed relationships with pharmaceutical companies, including the manufacturer of lecanemab.
This article has been fact checked for accuracy, with information verified against reputable sources. Learn more about us and our editorial process.
Last reviewed on .
Article history
- Latest version
Reference(s)
- Liu, Andy J.., et al. “Clinical Practice Outcomes With Lecanemab for Alzheimer Disease.” Neurology Open Access, vol. 2, no. 3, September 1, 2026 Ovid Technologies (Wolters Kluwer Health), doi: 10.1212/WN9.0000000000000121. <https://doi.org/10.1212/WN9.0000000000000121>.
Cite this page:
- Posted by Asif Iqbal