Impact of Clinical Subgroup and Baseline Amyloid PET Centiloid Levels on Occurrence of ARIA in the Lecanemab Phase 3 Clarity AD Study in Early Alzheimer’s Disease
Marwan Sabbagh1, David Li2, Shobha Dhadda2, Steven Hersch2, Luigi Giorgi2, Andre Matta3, Michael Irizarry4, Lynn Kramer2
1Barrow Neurological Institute, 2Eisai Inc., 3Eisai Co., Ltd., 4Eisai, Inc.
Objective:
To evaluate the frequency of amyloid-related imaging abnormalities due to edema (ARIA-E) and hemosiderin deposition (ARIA-H) in key subgroups of the Clarity AD trial, including by baseline amyloid PET centiloid levels and by clinical subgroup (mild cognitive impairment [MCI] and mild Alzheimer’s disease [mAD]).
Background:
In multiple early Alzheimer’s disease clinical trials, lecanemab, an approved amyloid beta-directed antibody that selectively targets protofibrils, has been shown to be well tolerated, with an increase in ARIA-E and ARIA-H relative to placebo. This presentation reports ARIA occurrences in key clinical and biomarker subgroups.
Design/Methods:
Clarity AD was a 18-month, phase 3 study in which eligible patients are randomized 1:1 across 2 treatment groups (placebo and lecanemab 10 mg/kg biweekly). ARIA occurrence was monitored throughout the study by central reading of magnetic resonance imaging. ARIA occurrence was evaluated in the following subgroups: by baseline centiloid tertiles (low:≤68.185, middle:>68.185-101.245, and highest:>101.245) and clinical subgroup (MCI and mAD).
Results:
Overall, 1107 patients were MCI (lecanemab:552; placebo:555) and 688 patients were mAD (lecanemab:346; placebo:342). When evaluating ARIA-E in subgroups based on baseline centiloid tertiles, the event rates were 8.5% and 3.3% for the low tertile, 15.3% and 1.6% in the middle tertile, and 13.4% and 0.9% in the highest tertile for the lecanemab and placebo groups, respectively. No trends were observed for ARIA-H. ApoE4 zygosity did not impact results across subgroups. In the clinical subgroup analysis, ARIA-E rates were 12.7% and 1.4% for MCI patients and 12.4% and 2.1% for mild AD patients in the placebo and lecanemab groups, respectively. The event rates of ARIA-H were similar between patients with MCI and mild AD.
Conclusions:
The frequency of ARIA-E in the Clarity AD trial was similar when evaluated by baseline amyloid PET tertile centiloid levels and by clinical subgroup. No trends were observed for ARIA-H.