Peripheral Blood Brain-Derived Neurotrophic Factor as a Biomarker of Alzheimer's Disease: Are There Methodological Biases?

Abstract

Mounting evidence that alterations in brain-derived neurotrophic factor (BDNF) levels and signaling may be involved in the etiopathogenesis of Alzheimer's disease (AD) has suggested that its blood levels could be used as a biomarker of the disease. However, higher, lower, or unchanged circulating BDNF levels have all been described in AD patients compared to healthy controls. Although the reasons for such different findings are unclear, methodological issues are likely to be involved. The heterogeneity of participant recruitment criteria and the lack of control of variables that influence circulating BDNF levels regardless of dementia (depressive symptoms, medications, lifestyle, lack of overlap between serum and plasma, and experimental aspects) are likely to bias result and prevent study comparability. The present work reviews a broad panel of factors, whose close control could help reduce the inconsistency of study findings, and offers practical advice on their management. Research directed at elucidating the weight of each of these variables and at standardizing analytical methodologies is urgently needed.

Keywords: Alzheimer’s disease; Biomarker; Blood; Brain-derived neurotrophic factor; Confounding variables.

Fig. 1

Sketch of brain-derived neurotrophic factor (BDNF) functions. BDNF, nerve growth factor, neurotrophin-3, and neurotrophin-4/5 belong to the neurotrophin family of closely related proteins. BDNF is synthesized as pre-proBDNF and converted to proBDNF by removal of the signal peptide in the endoplasmic reticulum (ER); proBDNF is then transformed to mature BDNF (mBDNF) by intracellular (in trans-Golgi network (TGN) and secretory granules (SG)) and extracellular cleavage events. Pro and mature BDNF are both active molecules playing different key roles: proBDNF through the p75^NTR/sortilin complex and mBDNF through its receptor TrkB. Blue circles, proBDNF; red circles, mBDNF; LTD, long-term depression; LTP, long-term potentiation; tPA, tissue plasminogen activator

Fig. 2

Representation of the degree of control exerted on the various factors. The heatmap outlines the level of control (indicated by the values of colors) that is exerted on each factor that may introduce a bias in the studies that have tested blood BDNF as a possible biomarker of AD. Colors range from noncontrolled (white) to completely controlled (dark blue) on the basis of the following categories: AD staging, (1) not considered; (2) defined by FAS (Functional Assessment Staging) score; (3) defined by MMSE (Mini Mental Stage Examination) score; (4) defined by CDR (Clinical Dementia Rating) score; age/gender, (1) differences between healthy and AD cohorts not adjusted; (2) healthy and AD cohorts without differences or with differences between healthy and AD cohorts adjusted; lifestyle, (1) not evaluated; (2) evaluation of one factor; (3) evaluation of two factors; (4) evaluation of three factors; (5) evaluation of four factors; depressive symptoms, (1) not evaluated; (2) exclusion of patients with a clinical diagnosis of depression or assessment of their weight; (3) evaluation of depressive symptoms; (4) exclusion of patients with a clinical diagnosis of depression or assessment of their weight and evaluation of depressive symptoms; medications, (1) not evaluated; (2) evaluation of psychoactive medications in AD patients; (3) evaluation of psychoactive and nonpsychoactive medications in AD patients; (4) evaluation of psychoactive medications in AD patients and healthy subjects; (5) evaluation of psychoactive and nonpsychoactive medications in AD patients and healthy subjects; experimental aspects, (1) not evaluated; (2) evaluation of data normality; (3) control of time of blood sampling; (4) evaluation of data normality and control of time of blood sampling; (5) evaluation of data normality, control of time of blood sampling, and information about storage duration