A Molecular Biomarker Panel for the Diagnosis of Vascular Cognitive Impairment and Dementia: A Multi-Cohort Study

Abstract

Introduction

Vascular Cognitive Impairment and Dementia (VCID) are a heterogenous group of disorders arising from cerebrovascular insufficiency to the brain. There has been limited progress in identifying blood biomarkers for its diagnosis. As several salient pathways are involved in VCID including endothelial dysfunction, neurovascular unit dysfunction, cardiac dysfunction and neuroinflammation, a targeted panel of biomarkers were measured in human plasma samples using multiplex and single plex assays.

Methods

Plasma samples were obtained from nine multinational cohorts (Singapore, USA and Australia) after phenotyping was performed using data from neuroimaging markers such as white matter hyperintensities, neuropsychological assessment results (MoCA or MMSE) and diagnosis. Over 1000 cases and controls from the following cohorts- Sydney Memory and Ageing Study, Older Australian Twins Study, Sydney Stroke Study, National Alzheimer’s Coordinating Centres, The Australian Imaging, Biomarker and Lifestyle Flagship Study of Ageing, MarkVCID, The Biomarkers and Consortium Study and Harmonisation cohort will be analysed. Plasma levels of 16 protein biomarkers -D-dimer, YKL40, VE- cadherin, ICAM-1, VCAM-1, Interleukin 6 and 18, Lipocalin 2, Matrix metalloproteinases-2 and 9, Lipoprotein-associated phospholipase A2, Placental growth factor 1, N-terminal (NT)- pro hormone BNP, TIMP metallopeptidase inhibitor 1, Vascular endothelial growth factor and Platelet-derived growth factor receptor ligand were measured using a two- plate design by multiplexing using the Luminex instrument platform (Bio-rad, California, USA). Myelin Basic Protein, Oligodendrocyte myelin glycoprotein and platelet-derived growth factor receptor beta were measured by ELISA assays. GFAP and NfL measurements were provided by respective study coordinators where available.

Results

Preliminary results from 4 out of 9 cohort show significant differences = for 8 of the plasma proteins: VE-cadherin (p < 0.001), NT-proBNP (p = 0.01), VEGF-A (p = 0.01), YKL-40 (P<0.001), D-dimer (p = 0.01), ICAM-1 (p = 0.002) NGAL (p = 0.001) and TIMP-1 (p<0.001) were found to be significantly elevated in cases vs controls. A preliminary analysis of claudin-5 and occludin found no significant differences between cases and controls (p >0.05).

Conclusions

Our findings suggest that the nine biomarkers may indeed be upregulated in VCID. Further longitudinal analyses and measurements using more sensitive assays like SIMOA may improve the prognosis of these markers in prediction and monitoring.