Examination of plasma YKL‐40 as a clinical biomarker for Alzheimer’s disease

Abstract

AbstractBackgroundBlood biomarkers represent a potential non‐invasive and cost‐effective strategy for the clinical detection and monitoring of neurodegenerative conditions including Alzheimer’s disease (AD). YKL‐40 is a secreted glycoprotein associated with inflammatory processes, and elevated cerebrospinal fluid levels have been observed in AD and other neurodegenerative conditions. Fewer studies have examined plasma YKL‐40. We examined relationships between plasma YKL‐40 and clinical indices of AD in a large cross‐sectional sample from a clinical cohort registry.MethodThe sample included 569 participants from the Boston University Alzheimer’s Disease Research Center, including participants with normal cognition (NC; n = 235), mild cognitive impairment due to suspected AD (MCI; n = 181), and AD dementia (n = 153). Blood plasma samples were analyzed using standard ELISA methods; values were log transformed and standardized into z‐scores. Cross‐sectional analyses investigated the relationships between YKL‐40 and clinical indices of functioning including diagnostic status, dementia severity (on the Clinical Dementia Rating scale [CDR]), and neuropsychological test performance. Age, sex, education, race, and APOE ε4 carrier status were included as covariates in all analyses.ResultNo significant differences in plasma YKL‐40 were observed between participants with NC, MCI, and AD dementia after accounting for covariates. The comparison between MCI and dementia was marginally significant (p = .07), with an adjusted mean difference of Z = 0.19. For participants with dementia (CDR scores 1‐3), non‐significant trends were observed for relationships between plasma YKL‐40 values and dementia severity, including for Sum of Boxes (Pearson partial correlation: r = .14, p = .10) and Total Score (ordinal logistic regression: OR = 1.38; 95% CI: [0.98,1.93]), p = .06. Modest but significant correlations were observed between YKL‐40 and performance on 4/12 neuropsychological tests including Animal Fluency (r = ‐.10), Vegetable Fluency (r = ‐.09), and Neuropsychological Assessment Battery List Learning Trials 1‐3 (r = ‐.10) and Short Delay (r = ‐.12).ConclusionPlasma YKL‐40 biomarker levels were modestly related to neuropsychological assessments of language and memory but not significantly related to cognitive diagnosis of AD or dementia severity. This biomarker may reflect underlying inflammatory processes in the context of AD. However, the use of plasma YKL‐40 to assess the presence and severity of AD in a larger cross‐sectional sample size, and using a multiple biomarker approach, warrant further study.