Probing Osteoarthritis Pathogenesis by Noninvasive Imaging of Cartilage Strain

点击翻译按钮获取中文摘要

PROJECT SUMMARY / ABSTRACT The objective of this proposal is to predict osteoarthritis (OA) pathogenesis in vivo using a novel noninvasive MRI-based method of measuring articular cartilage biomechanics. Recent advances in magnetic resonance imaging (MRI) have been introduced with exciting potential to diagnose and predict the progression of OA, the most common degenerative joint disease. MRI methods have sought to discover early changes in OA, when emerging disease-modifying interventions (e.g. cell implantation) may be most effective. OA pathophysiology often involves joint injury (e.g. ligament rupture) and a degenerative cascade of increased expression of inflammatory cytokines and enzymes. Moreover, the breakdown and loss of major macromolecules such as aggrecan and type II collagen leads to altered strains and material properties (e.g. moduli) within the tissue, suggesting MRI of cartilage biomechanics may be sensitive to degeneration. Unfortunately, noninvasive diagnosis of OA remains poor, especially in early disease stages, and several challenges remain, including the need for sensitive and specific imaging biomarkers that predict OA outcomes, and the need to relate imaging biomarkers to tissue function and biomechanics. In our original grant (AR063712), we pioneered dualMRI (displacements under applied loading by MRI) for cartilage biomechanics to monitor joint health. We discovered that dualMRI is robust to detect strain increases following controlled enzyme digestions or mechanical trauma to excised tissues, and in an in vivo time-course meniscectomy study in sheep. Compared to quantitative MRI (qMRI, e.g. T1ρ mapping), shear strains better correlated with OA severity in human cartilage. We also recently performed first-in-human in vivo and intra-tissue cartilage strain measures on a clinical 3 Tesla (T) MRI system. In this renewal application, we will establish a workflow to measure strains and moduli (i.e. elastography), and validate this workflow in multiple model systems. In humans, we will also identify biomechanics-based MRI metrics and biomarkers that predict time-course cartilage function and symptomatic pain following ligament reconstruction in a subset of patients. We will pursue three related specific aims. In Aim 1, we will establish a routine, clinical workflow for dualMRI measures of intra-tissue strain and properties. We will extend our existing dualMRI sequence to accelerate clinical measurement of strain within 15 minutes, and coupled to inverse modeling, automate measurement of in vivo elastography. In Aim 2, we will validate dualMRI intra-tissue strain and properties against gold-standard benchmarks, confirming reproducibility for in vivo time course analyses by quantifying numerous error metrics. In Aim 3, we will predict functional outcomes and cartilage health in patients following ligament reconstruction. We will determine the extent that MRI metrics at six months predict patient-reported outcomes and tissue health at one and two years post treatment. If successful, we will establish a routine method for functional assessment of cartilage, and support a new paradigm targeting cartilage biomechanics as specific indicators of joint damage and repair.