Plasma biomarkers for early detection of alzheimer's disease: a cross-sectional study in a Japanese cohort.

Background: Plasma biomarkers offer a promising alternative to amyloid beta (Aβ) positron emission tomography (PET) or cerebrospinal fluid (CSF) biomarkers for diagnosing Alzheimer's disease (AD). This cross-sectional study assessed the utility of multiple plasma biomarkers for diagnosing and staging AD in a Japanese cohort.

Methods: The assessed plasma biomarkers included Aβ42/40, phosphorylated tau (p-tau181 and p-tau217), glial fibrillary acidic protein (GFAP), and neurofilament light chain (NfL), individually and in combination. Aβ42/40 was measured using the HISCL^® platform, while all other biomarkers were measured using the Simoa^® platform. Participants were classified based on Aβ PET imaging and neuropsychological testing into healthy controls (HC), AD continuum (preclinical AD, mild cognitive impairment [AD-MCI], and mild dementia [AD-D]), and non-AD cognitive impairment (CI) groups. Receiver operating characteristic analyses were performed to predict the Aβ PET status, correlation with Centiloid (CL) values and cognitive scores, and biomarker comparisons across AD stages.

Results: Sixty-nine HC, 13 preclinical AD, 38 AD-MCI, 44 AD-D, and 79 non-AD CI participants were included. The area under the curves (AUCs) for predicting Aβ PET status were 0.937 (Aβ42/40), 0.926 (p-tau217), and 0.946 (p-tau217/Aβ42); results of pair-wise DeLong tests revealed no significant differences among these three metrics (all p > 0.05). In the cognitively normal group, the AUCs were 0.968 (Aβ42/40), 0.958 (p-tau217), and 0.979 (p-tau217/Aβ42), while in the cognitively impaired group, they were 0.919 (Aβ42/40), 0.893 (p-tau217), and 0.923 (p-tau217/Aβ42). Among HC and AD continuum participants, CL correlations were - 0.74 (Aβ42/40), 0.81 (p-tau217), and 0.83 (p-tau217/Aβ42). In the HC and AD continuum, Aβ42/40 levels showed a bimodal distribution (cutoff = 0.096), with a shift from high to low occurring at 19.3 CL, compared to the PET positivity threshold of 32.9 CL. P-tau217 exhibited a linear increase with disease progression. All biomarkers correlated strongly with logical memory scores.

Conclusions: Plasma biomarkers, Aβ42/40 and p-tau217, and particularly their ratio (p-tau217/Aβ42), show strong potential as Aβ PET alternatives for AD diagnosis. HISCL-based plasma Aβ42/40 detects Aβ accumulation earlier than Aβ PET visual reading threshold, underscoring its utility as an early diagnostic marker.