Amyloid-Related Imaging Abnormalities With Donanemab in Early Symptomatic Alzheimer Disease: Secondary Analysis of the TRAILBLAZER-ALZ and ALZ 2 Randomized Clinical Trials

Abstract

Importance: Amyloid-related imaging abnormalities (ARIA) are the major adverse event associated with amyloid-targeting immunotherapy. Identifying clinical features and individual risk factors for ARIA could facilitate effective prediction and prevention strategies.

Objective: To characterize ARIA in participants treated with donanemab.

Design, setting, and participants: These prespecified and post hoc exploratory analyses use data from the placebo-controlled portions of the TRAILBLAZER-ALZ and ALZ 2 randomized clinical trials, conducted from December 2017 to December 2020 and from June 2020 to April 2023, respectively. Additional analyses are included from a stand-alone open-label addendum conducted from August 2021 through August 2023. Participants in the placebo-controlled trials and the open-label addendum aged 60 to 85 years with early symptomatic Alzheimer disease and elevated amyloid levels were included. The placebo-controlled trials, but not the addendum, had tau inclusion criteria.

Interventions: Placebo-controlled trial participants were randomized 1:1 to receive placebo or donanemab, and all open-label participants received donanemab. Donanemab was administered every 4 weeks for up to 72 weeks.

Main outcomes and measures: The primary outcomes were the frequency, radiographic severity, seriousness, symptoms, timing relative to donanemab treatment, and risk factors for ARIA.

Results: Across 3030 total participants (placebo-controlled trials: 999 placebo participants, 984 donanemab participants; open-label addendum: 1047 donanemab participants), mean (SD) age was approximately 73.7 (6.0) years and 1684 participants (55.6%) were female. Frequencies of ARIA-edema/effusions (ARIA-E) and ARIA-microhemorrhages and hemosiderin deposition (ARIA-H) were higher with donanemab (24.4% and 31.3% in placebo-controlled trials, respectively; 19.8% and 27.2% in open-label addendum, respectively) than with placebo (1.9% and 13.0%, respectively). ARIA-E was mostly mild or moderate in severity. Serious ARIA-E was reported in 1.5% and symptomatic ARIA-E in 5.8% of donanemab-treated participants in the placebo-controlled trials. Symptoms most frequently reported with ARIA-E were headache and confusional state. In 58.3% of donanemab-treated participants with ARIA-E, the first event occurred by the third infusion (approximately month 3). Risk analysis demonstrated independent associations between ARIA-E and 6 baseline variables, including increased risk with APOE ε4 allele number, greater number of microhemorrhages, presence of cortical superficial siderosis, higher amyloid plaque, and elevated mean arterial pressure, and decreased risk with antihypertensive use.

Conclusions and relevance: ARIA is an adverse event associated with donanemab treatment that requires safety monitoring. Individual ARIA risk can be assessed by APOE ε4 status and baseline imaging findings.

Trial registrations: ClinicalTrials.gov Identifiers: NCT03367403 and NCT04437511.

Figure 1.. Number of Infusions to First Serious Amyloid-Related Imaging Abnormalities (ARIA) Event in Donanemab-Treated Participants

Number of infusions to first serious ARIA event in donanemab-treated participants in the integrated placebo-controlled trials and the open-label addendum by apolipoprotein E (APOE) ε4 carrier status (A) and ARIA type (B). Infusions were administered approximately every 4 weeks. ARIA-E indicates amyloid-related imaging abnormality–edema/effusions; ARIA-H, amyloid-related imaging abnormality–hemorrhage/hemosiderin deposition.

Figure 2.. Frequency of Amyloid-Related Imaging Abnormalities–Edema/Effusions (ARIA-E) and ARIA–Hemorrhage/Hemosiderin Deposition (ARIA-H) Based on Concomitant Antithrombotic Medication Use

Concomitant antithrombotic medication use, overall and by apolipoprotein E (APOE) ε4 status, in donanemab-treated participants in the integrated placebo-controlled trials and open-label addendum who experienced ARIA-E (A) or ARIA-H (B). The size of the bar indicates the percentage of participants with antithrombotic use at any time, and the shaded portion indicates use within 30 days prior to the ARIA event. ^aIncludes unknown APOE ε4 status. ^bNo. at any time in the trial, not necessarily prior to the ARIA event. ^cFor participants who experienced multiple ARIA events, ≥1 ARIA event had antithrombotic use within 30 days prior to the event.

Figure 3.. Association of Baseline Risk Factors With Amyloid-Related Imaging Abnormalities–Edema/Effusions (ARIA-E)

Analysis includes donanemab-treated participants in the integrated placebo-controlled trials and the open-label addendum. Forest plot of independent associations with ARIA-E identified in a post hoc analysis using machine learning approaches. Ranked from top, with the highest odds ratio for ARIA-E, to bottom, with lowest odds ratio. Odds ratios were obtained from a multiple logistic regression model adjusted for multiple covariates. This approach enables a thorough multivariable-adjusted evaluation of the association between the 6 key risk factors and ARIA-E events. APOE indicates apolipoprotein E; CL, Centiloids; PET, positron imaging tomography. ^aMean arterial pressure was estimated from systolic and diastolic blood pressure. ^bCerebellum used as reference region.