ARIA by clinical diagnosis and baseline amyloid PET levels: a subgroup analysis of clarity AD

<h3>Background</h3> Amyloid-related imaging abnormalities due to oedema (ARIA-E) or haemosiderin deposits (ARIA-H) have been associated with the use of monoclonal antibodies, such as lecanemab, for the treatment of Alzheimer’s disease (AD). Here, we evaluate the incidence of ARIA in key clinical and biomarker subgroups of the Clarity AD trial. <h3>Methods</h3> Clarity AD was a double-blind, placebo-controlled, Phase 3 clinical trial evaluating lecanemab in early AD. ARIA was assessed in the following subgroups: by baseline amyloid PET centiloid tertiles (low:≤68.185; middle:&gt;68.185–101.245; and high:&gt;101.245) and clinical diagnosis (mild cognitive impairment (MCI) and mild AD (mAD)). Results were also evaluated by apolipoprotein E (<i>ApoE4)</i> status within subgroups. <h3>Results</h3> ARIA-E rates in subgroups stratified by baseline centiloid tertiles for lecanemab vs placebo were 8.5% vs 3.3% (low), 15.3% vs 1.6% (middle), and 13.4% vs 0.9% (high). No trends were observed for ARIA-H. The rates of ARIA-E and ARIA-H were similar between <i>ApoE4</i> groups regardless of baseline centiloids. In the clinical subgroups, ARIA-E rates were 12.7% vs 1.4% (MCI) and 12.4% vs 2.1% (mAD) for lecanemab and placebo, respectively. <h3>Conclusions</h3> ARIA-E frequency in Clarity AD was consistent when evaluated by baseline amyloid PET tertile centiloid levels and by clinical subgroup. <i>ApoE4</i> status had no effect.