ERIS - Effective Reserve In Stroke

Massachusetts General Hospital, Boston MA

Investigators

Linked publications & trials

Abstract

PROJECT SUMMARY/ABSTRACT Stroke and dementia are two of the leading contributors to age-related disease burden and long-term disability. With aging populations in the US and worldwide, the prevalence of stroke increases and it becomes imperative to prevent related long-term functional disability from cerebrovascular disease and pathology. Understanding the determinants of post-stroke outcome that may lead to functional independence is of great clinical significance and essential to developing targeted treatment options. In the field of neurodegenerative diseases, the concepts of structural and functional reserve have been invoked to explain better than expected cognitive outcomes. Reserve describes the brain’s capacity to compensate for pathology. Specifically in stroke, elucidating such protective mechanisms has major practical implications for understanding outcome, however, few studies have investigated these concepts in stroke populations. Structural reserve generally aims to quantify the maximum brain reserve of a person, while pre-existing disease burden is not accounted for. The remaining, effective reserve, however, more likely reflects the brain’s ability to compensate for a sudden vascular event. The overall goal of this project is to create a translational approach to estimate the effective reserve from the low-resolution clinical imaging data of patients with acute ischemic stroke as they are acquired in the emergency department, and expand the concept by incorporating longitudinal and spatial lesion information to help improve outcome modeling. Aim 1 will investigate the concept of effective reserve and its feasibility in patients with acute ischemic stroke. We leverage two large-scale, retrospective stroke cohorts with clinical imaging available (single site: N=453, multi-site: N=912). To ensure the translational nature of effective reserve, we will enhance existing and develop new deep-learning enabled image processing pipelines to extract the required quantitative imaging biomarkers. These biomarkers will then be used to model functional outcome, measured as modified Rankin Scale score, ~90 days post-stroke. Aim 2 will develop longitudinal assessment methodologies for effective reserve and investigate how a drop in effective reserve post-stroke impacts stroke outcome. Aim 3 will further enhance this principle by harnessing spatial information of the stroke lesion and determining its impact on the brain’s ability to compensate for pathology. An important and unique feature of this project lies in its direct translational potential, as it is assessed on stroke neuroimaging data acquired in the emergency department to answer important questions about the brain’s ability to compensate for the acute vascular event and its focus on protective mechanisms in the brain relating to brain health. To achieve these goals, this project leverages unique data and multidisciplinary expertise, to create a clinically available, longitudinal, and spatially specific biomarker that enhances real-time stroke prognostication, and which can guide individualized patient care to improve outcomes.