Identification of novel candidate autoantibodies in Alzheimer's disease

Background and purpose: Accumulated failures in Alzheimer's disease (AD) clinical trials have highlighted an urgent need to identify additional biomarkers involved in AD. Recently, mounting evidence reported that autoantibodies are ubiquitous in human sera. However, it is unknown whether autoantibodies are upregulated in amyloid-tau biomarker-confirmed AD.

Methods: A total of 40 subjects with mild dementia (Clinical Dementia Rating = 1) were stratified into AD (n = 16) and non-AD (n = 24) groups according to their cerebrospinal fluid levels of tau and Aβ₄₂ . Their sera were collected and analyzed using a microarray containing > 1600 potential human autoantigens. Autoantibodies that were present exclusively in the AD group were identified and selected using the penetrance-based fold change method with the following criteria: penetrance fold change_(AD) ≥ 2, frequency_(AD) ≥ 15% and frequency_(non-AD) = 0%.

Results: All controls and samples passed the quality control criteria and were further used for biomarker analysis. Six autoantibodies with elevated responses to the following autoantigens were found exclusively in the AD group: nucleosome assembly protein 1-like 3 (31.3%, 5/16 subjects) and microtubule-associated protein 4, pantothenic acid kinase 3, phosphoinositide-3-kinase regulatory subunit 1, protein tyrosine phosphatase type IVA member 1 and SRY (sex-determining region Y)-box 15 (all 18.8%, 3/16 subjects).

Conclusions: Although some identified autoantigens are linked to AD and cognitive dysfunction, the increased autoantibody levels have not been reported in AD. Autoantibodies may provide deeper insights into the pathogenesis of AD and serve as diagnostic biomarkers; their corresponding antigens can be further studied to assess their potential as therapeutic targets.