SV2A as a Biomarker for Spinal Cord Injury: PET Imaging with [18F]SynVesT-1 in a Rat Model of Contusion

1049 Introduction: Traumatic spinal cord injury (SCI) is a devastating condition that involves long-term motor and sensory deficits, as well as cognitive and emotional dysfunctions. Current prognostication of SCI patients relies on anatomical MRI and clinical assessment of motor and sensory function. There are currently no effective treatments and diagnostic methods providing pathophysiological information for SCI. Thus, there is an unmet need to develop objective biomarkers to assess SCI progression and recovery, as well as to evaluate the efficacy of emerging therapies for SCI. Synaptic vesicle glycoprotein 2A (SV2A) is a synaptic vesicle protein that regulates action potential-dependent neurotransmitter release, and PET imaging of SV2A has been shown to have wide utility in measuring synaptic density changes in neuropsychiatric diseases1. Importantly, altered synaptic density has been found in the mouse spinal cord after SCI2-4. In this study, we explored a novel diagnostic PET imaging method using our recently developed SV2A tracer [18F]SynVesT-1 in a rat model of sub-acute SCI. Methods: A T7 moderate contusion injury (10 g, 25 mm) in female Sprague-Dawley rats (sham vs. SCI, n=4/group) was produced with the MASCIS impactor. The rats received special care and daily assessment of locomotor performance. At 10-11 days post injury, the rats underwent PET/CT scans on an Inveon PET/CT scanner. Emission data were acquired for 60 min after intravenous injection of [18F]SynVesT-1, followed by a 10-min CT scan. One day post imaging, the rats were sacrificed and the spinal cord was collected and evenly divided into six fractions of ~3 mm each (F1-6) and protein extracted for western blotting. PET images were reconstructed with attenuation and scatter correction. PET and CT image fusion and analysis were performed using Inveon Workplace to generate standardized uptake values (SUVs). Four regions of interest (ROIs 1 to 4) representing spinal cord regions of C1-C7, Th1-Th5, Th6-Th10, and Th11-L2 were drawn by referencing the individual CT. The SUVs of ROIs were normalized by the SUVs of ROI-1(C1-C7) to generate SUV ratio (SUVR). SUVRs from 0-10 min and 30-60 min post tracer injection were chosen for the evaluation of blood flow and SV2A levels, respectively. Results: Upon injury, SCI rats lost motor function on both hind legs compared to sham control. At day 10-11 post injury, there is no visible lesion or scar observed on spinal cords of SCI rats compared to sham controls. [18F]SynVesT-1 exhibits high specific uptake in the spinal cord as determined by blockade with the SV2A ligand levetiracetam (n=4). Interestingly, the level of SV2A in sham ROI-3 (Th6-10) corresponding to F3-4 happen to be lower than the other regions of spinal cords (Fig. 1A). Nevertheless, the SUVRs of both 0-10 min and 30-60 min show 30-35% decrease in ROI-3 in SCI rats compared to sham rats (Fig. 1B), which is consistent with the decreased SV2A protein expression as detected by western blotting (Fig. 1C-E). Importantly, the decrease of SV2A is also seen in lower thoracic and lumbar regions of SCI rats compared to sham controls, suggesting that a local contusion injury can induce synaptic loss caudal to the lesion epicenter (Fig. 1B, E). Conclusions: Using [18F]SynVesT-1 and PET/CT imaging, we successfully detected synapse loss in the sub-acute contusion SCI rat model. Our results indicate that the quantification of synaptic density by SV2A PET may provide an objective measure of SCI, and thus a valuable tool to evaluate novel therapeutics in clinical trials.