FRAME: Fine-Resolution Asymmetric Migration Estimation.

Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by a cascade of pathological processes, from the accumulation of misfolded proteins such as β-amyloid (Aβ) and hyperphosphorylated tau (p-Tau) to the eventual development of neurodegeneration []. Despite decades of research efforts, a significant portion of AD pathogenesis remains uncertain. To understand the complex nature of AD, multiple levels of omics characteristics of the disease have been investigated [], lipidomics being one of them. Lipidomics is the systems-level analysis of lipids and factors that interact with lipids []. The involvement of lipids in AD pathogenesis has been suggested in many previous studies. In particular, alterations of phospholipid, plasmalogens, ceramide, ganglioside, and sulfatide levels in the brain have been observed [–]. Also, several recent studies have demonstrated alterations of blood lipidome profiles in AD [–]. Furthermore, in a recent lipidomics study using two large cohorts, the Australian Imaging, Biomarkers and Lifestyle (AIBL) study and the Alzheimer’s Disease Neuroimaging Initiative (ADNI), a total of 218 lipid species were identified as associated with prevalent or incident AD []. Advances in AD biomarker research have led to the shift of AD diagnosis from clinical syndrome to biological process. AD-related biomarkers can be grouped into those of β-amyloid, hyperphosphorylated tau, and neurodegeneration (A/T/N) []. In this regard, the concept of an A/T/N classification system was included in the 2018 National Institute of Aging and Alzheimer’s Association Research Framework [] and has been widely used in AD research since then. Central biomarkers in this study refer to the established A/T/N biomarkers measured by PET, CSF, and MRI, as defined by recent international consensus criteria [, ]. Abnormalities in these biomarkers are central to the biological classification of AD and are strongly predictive of subsequent cognitive decline and progression to dementia, as demonstrated in large-scale longitudinal studies [, ]. Recent clinical guidelines and major therapeutic trials now rely on these biomarkers for both disease definition and regulatory approval of new treatments. A few existing studies examining the associations between A/T/N biomarkers for AD and blood lipidome have employed techniques that have limited resolution of lipid species [–] and have focused primarily on cross-sectional associations. Here, we investigated associations of circulating lipidome with cross-sectional and longitudinal central A/T/N biomarkers for AD using a recently developed lipidomics platform covering 749 lipid species across 46 classes that focuses on lipid and lipid-like compounds utilizing chromatographic separation and quantitation. We examined the main effect and interactions of sex and ε4 carrier status on circulating lipids at the lipid species and lipid class levels. Finally, we identified network modules of correlated lipids and performed association analyses between the modules and cross-sectional and longitudinal A/T/N biomarkers for AD.