Abstract P333: Plasma Proteins, Structural Brain Reserve, and Onset of Alzheimer’s Disease: Proteome-Wide Mendelian Randomization

Abstract

Introduction: While the emergence of amyloid-clearing treatments for Alzheimer’s disease is promising, the identification of additional therapeutic options for Alzheimer’s disease and related dementia syndromes (AD/ADRD) remains urgent. Plasma proteins represent a class of actionable drug targets AD/ADRD treatment. Yet, the role of the blood proteome in influencing structural brain reserve, and consequently, cognitive performance and the development of AD, remains elusive. Hypothesis: Plasma proteins may causally associate with changes in structural brain reserve and cognition phenotypes. Methods: We conducted proteome-wide Mendelian randomization (MR) to identify potentially causal associations of plasma proteins with changes in brain reserve, proxied by brain imaging-derived phenotypes (IDPs), cognitive performance, and AD onset. Genetic instruments for plasma proteins were previously developed by integrating summary data from five genome-wide association studies (GWASs) of plasma proteome. Summary statistics from large GWASs of 12 AD-associated IDPs (N=33,224), cognitive performance (N=257,841), and two AD cohorts (IGAP consortium, N=63,926, and FinnGen study, N=309,154) are obtained to perform two-sample MR. Causal estimates from two AD outcomes are meta-analyzed to improve statistical power. Correction for multiple testing and sensitivity analyses are performed. Results: Among 956 proteins with genetic instruments that satisfied MR assumptions, genetically determined levels of 14 plasma proteins were associated with at least one of the 10 AD-related IDPs at a false discovery rate of 5%. Six of these proteins were associated with white matter hyperintensities, an IDP known to be involved in cognitive decline. In addition, 26 plasma proteins were associated with cognitive performance, and four were causally linked to AD. Notably, four identified proteins were associated with structural brain reserve and either cognition or AD. Specifically, ADAM12, CREB3L4, and UROS were associated with distinct IDPs and cognitive performance, while TMEM106B, a protein that forms age-dependent amyloid filaments, was associated with increased hippocampal tail volume and AD. Conclusions: This study identified potential causal plasma proteins underlying structural brain reserve, cognitive performance, and AD, offering insights into the molecular etiology of these conditions and potential targets for drug development.