Plasma clusterin has been reported to be associated with the pathology, prevalence, severity, and rapid clinical progress of Alzheimer's disease (AD). However, whether plasma clusterin can be used as a biomarker of AD are inconsistent and even conflicting. We conducted this study to evaluate the potential of plasma clusterin as the biomarker of AD. PubMed, Embase, and Cochrane databases were systematically searched for studies on the relationship between plasma clusterin levels and AD diagnosis, risk and disease severity. We also compared the difference in cerebrospinal fluid (CSF) clusterin levels between AD and control groups. We converted and pooled data using standardized mean difference, Pearson linear regression model and the Cox regression model. A total of 17 articles and 7228 individuals, including 1936 AD were included. The quality ranged from moderate to high. There was no difference in plasma clusterin between AD and control groups (SMD= 0.19 [-0.10, 0.48], p=0.20). Plasma clusterin levels were not correlated with the risk (RR=1.03 [0.97-1.09], p=0.31), the MMSE scores (R=0.33 [-0.06, 0.71], p= 0.09), and the integrated neuropsychological measurements (R=0.21 [-0.20, 0.63], p=0.31) of AD. Additionally, there was no difference in CSF clusterin between AD and control groups (SMD=1.94 [ -0.49, 4.37], p=0.12). Our meta-analysis suggested no relationship between plasma clusterin levels and the diagnosis, risk, and disease severity of AD and no difference in the CSF clusterin between AD and the control groups. Overall, there is no evidence to support plasma clusterin as a biomarker of AD based on the pooled results. Copyright© Bentham Science Publishers; For any queries, please email at epub@benthamscience.net.