Clinical outcomes for lecanemab, the FDA-approved anti-amyloid therapy for Alzheimer’s disease, remain consistent with initial trial data when administered in real-world settings. Researchers at Duke Health tracked over 230 patients between 2023 and 2025 to evaluate the drug’s safety, treatment persistence, and impact on disease progression.
The study, published August 7 in Neurology Open Access, indicates that approximately 80% of patients successfully maintained the treatment regimen for at least one year. Most adverse events observed during the study period resolved without requiring hospitalization or additional medical intervention, suggesting that the therapy is manageable within standard clinical environments.
Investigators utilized specialized MRI sequences to monitor for Amyloid Related Imaging Abnormalities, a known side effect involving potential brain swelling or bleeding. The data revealed that 24.3% of the patient cohort developed ARIA during the course of their treatment, a figure that mirrors the incidence rates reported in earlier, highly controlled clinical trials.
Genetic markers played a significant role in patient risk profiles, with APOE ε4 homozygotes showing a fourfold increase in the likelihood of developing ARIA. These events typically occurred in two distinct peaks at approximately 10 weeks and 25 weeks after the initiation of therapy, highlighting the critical importance of sustained, long-term radiographic surveillance for high-risk patients.
Serious adverse events were reported in 31% of the patient population, though researchers noted that some complications were attributable to comorbid conditions rather than the medication itself. Common side effects included infusion reactions, falls, and strokes, which necessitated careful clinical oversight and multi-disciplinary coordination among neurology and internal medicine teams.
The research team found that current diagnostic tools, including standard blood and imaging tests, remain insufficient for reliably predicting individual susceptibility to ARIA. However, combining MRI measures of vascular changes with pTau/AB42 biomarker levels showed potential for identifying low-risk candidates, offering a pathway toward more refined patient selection criteria.
To enhance safety protocols, Duke Health piloted an artificial intelligence-driven imaging tool designed to detect subtle ARIA changes on brain scans with greater consistency. This technology aims to provide more reliable monitoring as the center expands its experience with anti-amyloid treatments, potentially reducing the burden on radiologists and improving the accuracy of longitudinal assessments.
Andrew Liu, M.D., corresponding author and associate professor at the Duke University School of Medicine, emphasized that the findings underscore the necessity of rigorous patient selection. He noted that the primary goal remains the mitigation of risk while extending the window of patient independence, effectively buying time for individuals facing a progressive neurodegenerative condition.
P. Murali Doraiswamy, MBBS, a study co-author and professor of psychiatry and behavioral sciences, highlighted the role of cerebral amyloid angiopathy in the development of ARIA. He identified this underlying mechanism as a critical focus for future research into patient safety and the development of standardized criteria for when it is safe to resume treatment after an adverse event.
Heather Whitson, M.D., co-director of the Duke & UNC Alzheimer’s Disease Research Center, stated that the decision to pursue anti-amyloid therapy requires a highly individualized approach. She emphasized that the integration of real-world evidence and advanced imaging tools is essential for personalized clinical decision-making, ensuring that patients and their families are fully informed of the potential risks and benefits.
The medical community continues to monitor these patients to determine the long-term efficacy of slowing cognitive decline in diverse populations. Future research will likely focus on developing more precise biomarkers to identify which patients are at the highest risk for serious complications before beginning treatment, thereby optimizing the therapeutic index of anti-amyloid interventions.
As Duke Health has now treated more than 400 patients, the institution has established itself as a primary center for gathering longitudinal data on these therapies. The ongoing curation of a biorepository containing blood and cerebrospinal fluid samples provides a foundation for future studies aimed at identifying the molecular signatures of treatment response and adverse event susceptibility.
