Tear Fluid Biomarkers For Predicting The Severity And Pathologic Burden In Alzheimer’S Disease

Cerebrospinal fluid levels and Positron emission tomography (PET) imaging of amyloid-β (Aβ) and tau deposits are increasingly used as the pathologic biomarkers of Alzheimer’s disease (AD). However, they are restricted in use for the population screening because of their high cost and invasiveness. To overcome these limitations, we investigated novel tear biomarkers in patients with AD and mild cognitive impairment (MCI) using a proteomics in-depth analysis.

Prospective case-control study

A total of 234 participants were recruited, and 42 cognitively unimpaired (CU), 47 Aβ-positive MCI (MCI+) and Aβ-positive AD (AD+), and 56 AD+ participants were finally enrolled in the study. Of them, 42 subjects for the tear biomarker discovery cohort and separately 103 subjects for the biomarker validation cohort, including cognitively unimpaired (CU) participants and patients with Aβ-positive MCI (MCI+) and Aβ-positive AD (AD+). We performed global proteome profiling and individual tear analysis, as well as multiple regression and correlation analysis with neuroimaging biomarkers.

A total of 3,350 tear proteins were identified and eight candidate biomarkers were selected among 22 quantifiable proteins within an individual’s tear fluid. Of those, glutamine synthetase was positively associated with cortical Aβ burden, while lipocalin-1 and plastin-2 were negatively associated with Aβ and tau burden. The composite of biomarkers showed an area under the curve of more than 0.8 in discriminating MCI+ and AD+ from CU, and MCI+ from AD+, respectively.

We could identify novel biomarkers and quantify targeted proteins in a small tear fluid volume obtained from individuals with clinicopathologically confirmed AD. Tear protein biomarkers have not only predictive and discriminative utility in AD diagnosis, but also reflect the underlying AD-specific neurodegenerative changes.