Biomarkers for Cognitive Outcomes after Mechanical Thrombectomy for Emergent Large Vessel Occlusion (ELVO)

Stroke and Mechanical Thrombectomy: Emergent large vessel occlusion (ELVO) ischemic stroke continues to be a leading cause of death and disability globally,1 accounting for 30-40% of all ischemic stroke, or ~340,000 patients in the U.S. alone each year. Intravenous thrombolytic agents and mechanical thrombectomy (endovascular removal of the thrombus using suction and/or stent-retriever devices) have greatly improved the ability to restore blood flow during a stroke; in fact, thrombectomy improves patient outcomes even if initiated 24 hours after stroke onset.2-4 Despite receiving these effective interventions to re-establish blood flow, many patients still experience long-term cognitive and functional deficits. Therefore, while thrombectomy provides a significant and clinically important outcome benefit over thrombolysis alone for ELVO patients, there remains a clinical need for biomarkers to provide prognostic insight into neurological outcomes in order to guide individual patient treatment and rehabilitation. Parent Project (BACTRAC) Summary: Although ischemic stroke is a leading cause of death and disability in the US, there are no clinically approved biomarkers to predict prognosis or to tailor treatment. The increased use of mechanical thrombectomy (MT) has not only substantially improved clinical care, but has also created an opportunity to extract and study intracranial intraluminal serum contents: blood and clot from the peri-infarct vasculature. The Blood and Clot Thrombectomy Registry and Collaboration (BACTRAC; clinicaltrials.gov NCT03153683) at the University of Kentucky (UKY) collects blood samples distal and proximal to the thrombus during MT for each patient, as well as patient demographic data and diseaserelated clinical variables. We are currently analyzing these biological samples for proteomics and immune cell content, and have to date analyzed intracranial and systemic plasma proteins for 184 proteins in 26 of over 100 patients enrolled in the registry. Machine learning models are being employed to merge and analyze biological and demographic data sets, as well as to create two innovative algorithms to predict (a) the volume of infarct and (b) functional recovery after MT. The overall aim of BACTRAC is to generate clinically applicable predictive algorithms for thrombectomy patients using routinely available peri-infarct samples obtained during the procedure, and consequently, to create prognostic biosignatures that would guide patient treatment and rehabilitation. This is a prospectively enrolling, nonrandomized registry, in which blood samples and clinical data from stroke patients will be continuously procured. To our knowledge, this is the only tissue bank of its kind that permits the earliest possible glimpse of molecular events in the human brain during stroke, and requires no patient procedures outside of the standard of care. Proposed Pilot Project—Adding Cognitive and Affective Outcome Data: Currently, outcome data collected by BACTRAC is limited to the change in NIH Stroke Scale (NIHSS) and Montreal Cognitive Assessment (MOCA) scores from inpatient discharge to 90- and 180-day follow-up appointments with neurosurgery. Our facility recently established 180-day neurocognitive assessments through the neuropsychology service as clinical standard of care for MT patients to aid in treatment planning. The PI and primary mentor assembled a 90-minute to 2-hour test battery (see Approach section) to improve cognitive and affective domain coverage and sensitivity (over the mental status exam) while maintaining patient tolerability. Funds received as a part of the Early Career Pilot Study Award would support extraction of neuropsychological clinical data for entry into the BACTRAC database as cognitive and affective outcome data. The pilot data collected will enable us to address the following aims in pursuit of future extramural funding: Aim 1: To establish the utility of a brief neuropsychological battery to evaluate cognitive functioning after mechanical thrombectomy for ELVO. Hypothesis 1: Test battery results will improve over the MOCA and NIHSS in capturing cognitive effects of lesion laterality, infarct size, and duration of occlusion. Aim 2: To establish predictive associations between (a) protein, genetic, and arterial blood gas candidate biomarkers at time of stroke and (b) cognitive and affective outcomes six months post-stroke using a novel machine learning approach. Hypothesis 2: Candidate biomarkers will improve prediction of cognitive and affective outcomes over predictions based on injury characteristics and patient demographics alone.