An exploratory analysis of plasma biomarkers associated with cerebral amyloid angiopathy

Cerebral amyloid angiopathy (CAA) remains diagnostically challenging, particularly in asymptomatic individuals. While CAA often co-exists with Alzheimer‘s disease (AD), it may even have a direct impact on AD pathophysiology and the cognitive decline within the clinical course of AD. While fluid biomarkers are well established for AD pathology, reliable markers to characterize CAA are lacking. We analyzed two subsets of participants from the Alzheimer's Disease Neuroimaging Initiative with available plasma biomarker measurements from a 145-analyte multiplex immunoassay panel: one with T2*-weighted gradient-echo magnetic resonance imaging (MRI) data (n=21) and another with postmortem neuropathological data (n=24). We defined CAA as ≥2 lobar microbleeds on MRI or moderate-to-severe neocortical amyloid angiopathy on neuropathological examination. Plasma analytes were assessed twice per participant, one year apart, with the earlier sample obtained up to 6.6 years prior to either the first MRI or neuropathological examinations. In both cohorts, various markers related to inflammation, lipid metabolism, and cell adhesion were associated with CAA proxy measures. Specifically, both increased (Fas ligand receptor, Receptor for Advanced Glycosylation End-Products, Osteopontin, and Vascular Cell Adhesion Molecule-1) and decreased (Vitronectin, Endothelial Growth Factor) biomarker levels were associated with lobar microbleeds, while increased apolipoproteins (ApoAII, ApoCI, ApoCIII, ApoE, and clusterin) and decreased AXL were associated with CAA severity in neuropathology. Ratios between inversely associated markers enhanced correlation strength and differed between CAA and non-CAA. Given the small sample sizes in our exploratory analyses, larger studies are required to evaluate the discriminatory potential and clinical translatability of the identified biomarkers for CAA. • Plasma inflammatory, lipid, and cell adhesion markers are associated with CAA. • Plasma FASLG, RAGE, VCAM-1, and Osteopontin increase with lobar microbleed number. • Plasma Apolipoproteins increase with neuropathological CAA severity. • Ratios of analytes demonstrated stronger associations with CAA than individual analytes