Deep learning‐based MRI analysis reveals Lewy body co‐pathology accelerates brain aging in Alzheimer's disease

Abstract Alzheimer’s disease (AD) and Lewy body (LB) pathology frequently co-occur. Recent advances in cerebrospinal fluid (CSF) α-synuclein seed amplification assays (SAA) enable in vivo detection of LB pathology, offering new opportunities to elucidate its combined effects with AD on neurodegeneration. We trained a deep learning model on multi-cohort MRI scans from 4,355 cognitively unimpaired individuals to estimate brain age and applied it to 803 cognitively impaired participants, who were classified into four AD/LB pathology subgroups using the p-tau181/Aβ42 ratio to specify AD pathology and SAA status to determine LB pathology. The co-pathology subgroup (AD+LB+) exhibited the most accelerated brain aging, with saliency maps revealing more pronounced neurodegeneration, aligning with its steeper longitudinal atrophy in various neuroanatomical regions and corresponding cognitive deficits. These findings underscore LB pathology’s synergistic role in amplifying AD-related neurodegeneration, highlighting the importance of combined biomarker assays and targeted interventions for individuals harboring co-existing AD and LB pathology.