Serum brain-derived neurotrophic factor as diagnosis clue for Alzheimer's disease: A cross-sectional observational study in the elderly.

Abstract

Brain-derived neurotrophic factor (BDNF) has not been validated as a diagnostic marker for Alzheimer's disease (AD). To provide a different perspective, this study aimed to evaluate the relationship between serum levels of mature BDNF (mBDNF) and precursor BDNF (proBDNF) in AD and to investigate whether serum BDNF levels or the ratio of mBDNF levels to proBDNF levels (M/P) could be a valuable biomarker for determining the risk of AD in elderly individuals. A total of 126 subjects who met the inclusion criteria were assigned to either the AD group (n = 62) or the healthy control group (HC, n = 64) in this cross-sectional observationl study. Serum levels of mBDNF and proBDNF were measured using enzyme immunoassay kits. We analyzed the Mini-Mental State Examination (MMSE) scores from the two groups and examined the associations between AD and BDNF metabolism. The serum concentration of proBDNF was significantly higher in ADs (4140.937 pg/ml) than in HCs (2606.943 pg/ml; p < 0.01). The MMSE significantly correlated with proBDNF (p < 0.01, r = -0.686) and M/P (p < 0.01, r = 0.595) in all subjects. To determine the risk for AD, the area under the receiver operating characteristic curve was calculated, which was 0.896 (95% confidence interval 0.844-0.949) for proBDNF and 0.901 (95% 0.850-0.953) for proBDNF and M/P combined. We observed a correlation between low serum proBDNF levels and higher MMSE scores in AD. The most effective diagnostic strategy was the combination of proBDNF and M/P, whereas mBDNF levels performed poorly when we evaluated the predictive model.