Uncovering the Mechanism of APOE4-Driven ARIA

ABSTRACT Amyloid beta (Aβ) plaques are a pathological hallmark of Alzheimer’s disease, and have been brought to the forefront by the landmark approval of anti-Aβ antibodies, which are the first disease modifying treatments for Alzheimer’s disease (AD). Despite the excitement surrounding the reductions in amyloid lowering, and modest delays in cognitive decline afforded by approved Aβ antibodies, amyloid related imaging abnormalities (ARIA) are a key adverse effect of these therapeutic approaches. ARIA refers to MRI findings of vascular edematous effusions (ARIA-E) and hemosiderin deposition (ARIA-H) during clinical screening MRIs that are associated with severe side effects and even death. Importantly, ARIA has been seen at a substantially increased rate in carriers of the APOE4 allele. In phase III clinical trials for Leqembi™ (lecanemab), carriers of the E4 allele carried a 3- fold higher risk of developing ARIA, with this increasing to tenfold in E4 homozygotes. Similar findings were seen in phase III trials of Kisulna™ (donanemab), and in earlier trials of Aduhelm™ (aducanumab). Most importantly, APOE genotype is the strongest genetic risk factor for late onset AD (LOAD), with heterozygous carriers of the E4 allele 2-3x more likely to develop AD, while homozygous carriers face 10-15x higher risk of developing this devastating neurodegenerative disorder. Together, this means that E4+ individuals, which represents 25% of the US population, which are in greatest need of anti-amyloid antibody therapy, are significantly at risk for developing ARIA. With recent large-scale approvals by federal regulators, the usage of these drugs is set to rapidly increase worldwide, thus highlighting a critical need to understand the role E4+ plays in pathogenesis of ARIA. Despite this, the mechanism(s) by which E4+ dramatically increases ARIA pathogenesis remain unknown, and no E4+ pre-clinical models exist. Two proposed theories on the pathogenesis of ARIA include: 1) cerebral amyloid angiopathy (CAA) impeded plaque clearance; or 2) immune mediated localized inflammation, though evidence to support either of these theories has yet to be established. In the current proposal, we aim to leverage a mouse model of APOE4-associated ARIA-like symptoms to test the overall hypothesis that APOE4 drives increased aberrant immune activation in both the CNS and periphery due to anti-amyloid treatment, which leads to blood- brain barrier breakdown, thus producing ARIA. Based on our preliminary studies and overarching hypothesis described above, we propose the following Aims to uncover targetable mechanism(s) of E4-driven increases in ARIA risk: Aim 1: Establish, characterize and describe a reliable mouse model of human APOE4-driven immune response to anti-amyloid antibody therapy. Aim 2: Determine the effects of APOE on immune cell activation following anti-amyloid antibody therapy.