Synapse-Related Serum and P300 Biomarkers Predict the Occurrence of Mild Cognitive Impairment in Depression

Abstract

Zongwei Xue,1,* Xiaoyan Zhu,2– 4,* Wenqi Wu,1 Yan Zhu,5 Yuhao Xu,1 Ming Yu1 1Department of Neurology, The Affiliated Hospital of Jiangsu University, Zhenjiang, Jiangsu, 212001, People’s Republic of China; 2Department of Neurology, People’s Hospital of Yangzhong City, Yangzhong, Jiangsu, 212299, People’s Republic of China; 3School of Clinical Medicine, Medical College of Yangzhou University, Yangzhou, Jiangsu, 225000, People’s Republic of China; 4Jiangsu Key Laboratory of Integrated Traditional Chinese and Western Medicine for Prevention and Treatment of Senile Diseases, Yangzhou, Jiangsu, 225000, People’s Republic of China; 5Department of Radiology, The Affiliated Hospital of Jiangsu University, Zhenjiang, Jiangsu, 212001, People’s Republic of China*These authors contributed equally to this workCorrespondence: Yuhao Xu; Ming Yu, Department of Neurology, The Affiliated Hospital of Jiangsu University, No. 438 Jiefang Road, Zhenjiang, 212001, People’s Republic of China, Tel [phone], Email [email protected]; [email protected]: Cognitive impairment is one of the common concomitant symptoms of depression. The aims of the present study were to predict the occurrence of mild cognitive impairment (MCI) in patients with depression.Methods: In this study, 217 patients with depression were recruited. Demographic data, serum indices and ERP indices from all participants were collected in the baseline period. The participants were followed for one year, and data from 200 patients were included in final analysis. Patients with depression were divided into those with MCI group (DWM group; n=145) and those without MCI (DWOM group; n=55). Data from the DWM group and the DWOM group were used to construct a logistic regression model, and a receiver operating characteristic (ROC) curve was drawn. Another 72 patients were used to validate the accuracy of our model.Results: Compared with DWOM individuals, DWM individuals were more likely to live alone (P< 0.05), had lower baseline serum levels of brain-derived neurotrophic factor (BDNF), fibroblast growth factor 2 (FGF2), and fibroblast growth factor 22 (FGF22) (P< 0.05), and exhibited higher baseline latencies of P300, mismatch negativity (MMN), and N200 (P< 0.05). Baseline serum BDNF and FGF22 levels, along with the P300 latency, were selected to construct the regression model using logistic regression. The regression equation was , and the combination of the 3 indices yielded an area under the ROC curve (AUC) of 0.790 and a predictive accuracy of 0.806.Conclusion: The logistic regression model and ROC curves based on serum BDNF and FGF22 levels and the P300 latency could provide a more effective means to predict the occurrence of MCI in patients with depression.Keywords: depression, mild cognitive impairment, brain-derived neurotrophic factor, fibroblast growth factor 22, P300, prediction

Keywords

• depression
• mild cognitive impairment
• brain-derived neurotrophic factor
• fibroblast growth factor 22
• p300
• prediction.