O3‐14‐06: DISSECTION OF SYNAPTIC PATHWAYS THROUGH THE ANALYSIS OF BIOMARKERS IN THE CSF: A COMBINING TOOL FOR THE DIAGNOSIS OF ALZHEIMER'S DISEASE

In Alzheimer's disease (AD), synaptic degeneration is a central pathogenic event that occurs early during the course of disease and could be linked to cognitive symptoms. The synaptic proteins appear crucial for both synaptic transmission and cognitive function. Levels of synaptic proteins in the cerebrospinal fluid (CSF) might reflect synaptic dysfunction and could correlate with cognitive decline in AD. In a mono-centric retrospective cohort, 5 pre and post-synaptic biomarkers (GAP43, Neurogranin, SNAP25 total, SNAP 25aa40 and synaptotgamin) were evaluated in the CSF using either ELISA tests or mass spectromsssetry. We compared these synaptic protein levels in various groups of patients and studied their correlations with usual CSF AD biomarkers. We then compared their discriminatory power using ROC curves analysis and used backward stepwise logistic regression model to select which biomarkers were useful to discriminate AD from non-AD patients. This study included 241 patients (mean age 71± 9 years) : 79 AD cases, 31 MCI due to AD cases, 48 MCI non due to AD cases, 48 other dementias cases and 35 non neurodegenerative diseases cases considered as neurological controls. The CSF levels of all synaptic biomarkers were significantly higher in patients with MCI due to AD and AD-patients compared with neurological controls, patients with MCI non due to AD and patients with other dementias. Synaptic biomarkers were more strongly associated with CSF Tau than with CSF Aβ42 levels. GAP-43 and SNAP-25 had the higher power to discriminate AD from non-AD patients (area under ROC curve = 0.92 and 0.93 respectively). In backward stepwise selection models, CSF GAP-43 and SNAP-25 were both predictors of AD status, independently of CSF Aβ42 and CSF Tau. CSF synaptic proteins are promising biomarkers to monitor synaptic loss and function in AD including at a very early stage of the evolution. They are complementary markers to the usual CSF AD biomarkers. These biomarkers should be useful to monitor synaptic integrity and to explore the pathophysiology directly in patients with AD.