REAL-WORLD SAFETY OF THERAPIES FOR ALZHEIMER’S DISEASE
Author(s)
Yuqian Lin, MSc1, Gad Marshall, MD2, Joy Shi, PhD1, Deborah Blacker, MD, ScD3, Sachin Shah, MD, MPH3, Vicki Fung, PhD1, Christine Ritchie, MD, MPH3, Joseph Newhouse, PhD4, John Hsu, MD1;
1Massachusetts General Hospital, Mongan Institute, Clinical Economics and Policy Analysis Group, Boston, MA, USA, 2Brigham and Women's Hospital, Boston, MA, USA, 3Massachusetts General Hospital, Boston, MA, USA, 4Harvard Medical School, Boston, MA, USA
1Massachusetts General Hospital, Mongan Institute, Clinical Economics and Policy Analysis Group, Boston, MA, USA, 2Brigham and Women's Hospital, Boston, MA, USA, 3Massachusetts General Hospital, Boston, MA, USA, 4Harvard Medical School, Boston, MA, USA
OBJECTIVES: Novel therapies, e.g., lecanemab, may slow cognitive decline among patients with early-stage Alzheimer’s Disease (AD) but pose safety risks, primarily for Amyloid-Related Imaging Abnormalities (ARIA), which are detected through MRI scans. We examined patterns of therapy initiation and use and the relationship between ARIA findings and therapy interruptions.
METHODS: Using electronic health records, we examined lecanemab use (12/2023-9/2025) through completion of the 18-month regimen (40 biweekly infusions and four routine MRI scans), death, or 12/2025. We examined schedule-interruptions (gap>17 days) and regimen-discontinuation (no infusion>=90 days), differentiating between interruptions that did versus did not involve MRI scans. We reviewed MRI radiology reports and clinical notes to determine ARIA status. We used logistic regression to examine the presence or type of interruption, and proportional hazard models to examine time to the first interruption.
RESULTS: Among 504 patients initiating therapy, 69% had >=1 infusion interruption (mean follow-up=13 months). For most (60%), the interruptions were associated with MRI scans; of these, 41% had ARIA. Interruptions not involving MRIs occurred later in treatment. Among all patients, 37% had ARIA (of which 79% were detected during a routine scan). Following the first ARIA detection, 27% delayed their next infusion (mean interruption duration=63 days) and 21% discontinued therapy. There were no hospitalizations for cerebral hemorrhage. Among the 98 patients with complete 18-month follow-up to date, 93% received ≥1 extra MRI. Those with MRI-associated schedule interruptions were less likely to complete the full regimen.
CONCLUSIONS: Most patients experienced infusion schedule interruptions, many developed ARIA (frequently identified during routinely scheduled scans), and none had cerebral hemorrhage. While nearly all patients received extra MRI scans to address potential symptoms, most ARIA findings first appeared during routine scans. Despite the frequent ARIA findings, lecanemab therapy in the real-world appears relatively safe with respect to hemorrhage.
METHODS: Using electronic health records, we examined lecanemab use (12/2023-9/2025) through completion of the 18-month regimen (40 biweekly infusions and four routine MRI scans), death, or 12/2025. We examined schedule-interruptions (gap>17 days) and regimen-discontinuation (no infusion>=90 days), differentiating between interruptions that did versus did not involve MRI scans. We reviewed MRI radiology reports and clinical notes to determine ARIA status. We used logistic regression to examine the presence or type of interruption, and proportional hazard models to examine time to the first interruption.
RESULTS: Among 504 patients initiating therapy, 69% had >=1 infusion interruption (mean follow-up=13 months). For most (60%), the interruptions were associated with MRI scans; of these, 41% had ARIA. Interruptions not involving MRIs occurred later in treatment. Among all patients, 37% had ARIA (of which 79% were detected during a routine scan). Following the first ARIA detection, 27% delayed their next infusion (mean interruption duration=63 days) and 21% discontinued therapy. There were no hospitalizations for cerebral hemorrhage. Among the 98 patients with complete 18-month follow-up to date, 93% received ≥1 extra MRI. Those with MRI-associated schedule interruptions were less likely to complete the full regimen.
CONCLUSIONS: Most patients experienced infusion schedule interruptions, many developed ARIA (frequently identified during routinely scheduled scans), and none had cerebral hemorrhage. While nearly all patients received extra MRI scans to address potential symptoms, most ARIA findings first appeared during routine scans. Despite the frequent ARIA findings, lecanemab therapy in the real-world appears relatively safe with respect to hemorrhage.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
EPH23
Topic
Epidemiology & Public Health
Topic Subcategory
Safety & Pharmacoepidemiology
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Neurological Disorders