Identifying determinants of rapid structural and/or clinical progression in knee osteoarthritis by quantitative assessment of structural features on radiographs

Abstract Knee osteoarthritis (KOA) is a major public health problem that affects ~30 million US adults, with increased morbidity and premature mortality. KOA has been identified as one of the biggest unmet medical needs because there are no FDA-approved treatments available to prevent, slow or halt OA progression. Development and evaluation of potential OA treatments has been hampered in part by KOA typically being a slowly progressive disease, though progression is variable and some knees undergo rapid structural or clinical progression. Feasible high-throughput methods are needed to gain insights as to how to identify rapid progressors. We have developed a rapid, reproducible, and responsive quantitative software tool, Quantitative Radiographic Software Scores (QROS), to measure minimal joint space width (JSW), fixed JSW (fJSW), variance of 14 fJSW locations; degrees of knee alignment/ malalignment; and orientation and roughness of trabecular bone texture (TBT). The overarching goal of our research is to identify knees at high risk of rapid progression, and our core concept is that quantitative assessment of a novel combination of baseline structural features on radiographs will identify a group of knees with rapid structural and/or clinical progression in KOA. Our central hypothesis is that abnormalities of structural features on knee radiographs will be determinants of varying combinations of worsening rates of structural and/or clinical progression in KOA. Our approach takes advantage of a unique opportunity to innovatively assess radiographic features on ~10,500 individuals by combining three of the largest and most racially diverse longitudinal observational studies of individuals with or at risk of developing KOA: the Johnston County Osteoarthritis Project (JoCoOA), the MulticenterOsteoarthritis Study (MOST), and the Osteoarthritis Initiative (OAI). All three cohorts have fixed flexion, weightbearing knee radiographs that were acquired with identical protocols over multiple visits, and comparable data on demographics, KOA risk factors, and WOMAC pain and function. Our specific aims are: 1). To quantitatively assess structural and clinical progression of KOA, separately and combined; 2). To identify the structural determinants of rapid KOA progression; and 3). To determine the knee- specific probability and time to knee replacement (KR) based on structural and clinical determinants.