INTRODUCTION: Alzheimer's disease (AD) is the most common cause of dementia worldwide and a cost-effective diagnostic biomarker is needed. This systematic review provides an overview of the current research on plasma amyloid beta (Aβ) as a biomarker of AD and explores the clinical implications of this line of research.

METHODS: PubMed was searched using the keywords plasma Aβ and AD from 2017 to 2021. Only clinical studies involving amyloid PET (aPET) or cerebrospinal fluid (CSF) biomarker analysis (or both) were included. A meta-analysis of CSF Aβ42/40 ratio, aPET and plasma Aβ42/40 ratio was conducted when possible.

RESULTS: A total of 17 articles were identified. Plasma Aβ42/40 ratio was inversely correlated with aPET positivity r = -0.48 (95% confidence interval (CI): -0.65--0.31). In numerous studies, plasma Aβ42/40 ratio was also found to be directly correlated with CSF Aβ42 and CSF Aβ42/40 ratio r = 0.50 (95% CI: 0.30-0.69). Three studies found plasma Aβ42 to be positively associated with aPET positivity and CSF Aβ42; however, four other studies found no significant association between these variables. Seven studies reported no significant association of plasma Aβ40 with aPET or CSF Aβ40.

CONCLUSION: Plasma Aβ42/40 ratio seems as a promising plasma biomarker as it significantly correlates inversely with aPET positivity and directly with CSF Aβ42 and CSF Aβ42/40 ratio. However, more research is warranted, including validation studies, longitudinally clinical studies, studies comparing measurement methods and studies of Aβ kinetics.