Diagnostic Performance of MRI‐based Alzheimer’s Disease Resemblance Atrophy Index and Plasma‐Based Biomarkers on Alzheimer’s Disease

Background Although positron emission tomography (PET) or cerebrospinal fluid measurements of amyloid and tau burden are available for the detection of Alzheimer’s disease pathology, such methods are invasive and not easily accessible. Recent studies show that Alzheimer’s Disease‐resemblance atrophy index (AD‐RAI), an MRI‐based machine learning‐derived biomarker, and plasma‐based biomarkers can be used as accurate biomarkers for Alzheimer’s disease pathology. We aim to evaluate the diagnostic metrics of AD‐RAI and plasma‐based biomarkers for detecting Alzheimer’s disease. Method We recruited 69 subjects from the CU‐SEEDS (The Chinese University of Hong Kong‐Screening for Early AlzhEimer’s DiseaSe) study who were stroke‐free and had different degrees of cognitive impairment: 8 cognitive unimpaired [CU], 26 with subjective cognitive decline [SCD], 21 with mild cognitive impairment [MCI] and 14 with dementia. All subjects underwent 11C‐ PIB and 18F‐T807 PET to measure pathological Aβ deposition (A+) and tau burden (T+). Subjects received structural MRI for deriving AD‐RAI. Plasma levels of Aβ40, Aβ42, total tau (t‐tau), phosphorylated tau at 181 (p‐tau181), and neurofilament light chain (NfL) were measured by Single Molecule Array (SiMoA) assays. Result Among 69 subjects (mean [SD] age, 67.7 [6.8] years; 28 men [40.6%]), 25 (36.23%) subjects were confirmed to be A+ and T+. AD‐RAI, plasma p‐tau 181, and plasma Aβ42 were associated with A+T+ after adjusting for age, gender, and education (p