Imaging markers for osteoarthritis progression

Abstract

Joint damage in osteoarthritis (OA) involves characteristic changes in multiple joint tissue types, with alterations in the shape of subchondral bone and composition of synovial fluid occurring concomitantly with cartilage damage. Systemic risk factors such as obesity and local biomechanical processes that determine joint loading have long been shown to predispose to the disease, and more recent evidence suggests that specific genetic loci and inflammatory processes may make contributions, but the exact mechanisms driving the initiation and progression of the disease are yet to be elucidated. Advances in imaging that allow visualisation of minute changes in structures such as cartilage composition and subchondral bone shape have enabled longitudinal tracking of early disease development. It is hoped that the identification of early imaging biomarkers for later disease will not only assist in clarifying disease pathogenesis, but facilitate the development of disease-modifying interventions by shortening the lead time for assessing their effectiveness. This thesis aimed to clarify the current information on the utility of imaging biomarkers in research into the pathogenesis and treatment of osteoarthritis, and investigate potential imaging markers in the two regions that contribute the most to the disease burden of osteoarthritis; the knee and the hand. It utilises data from three clinical studies; a cross-sectional survey on quality of life indices in patients with arthritis, and the ongoing KANON (Knee Anterior Cruciate Ligament Non-operative vs. Operative Treatment) and COMBO (Combined Conservative Therapies on Clinical Outcomes in Thumb Base Osteoarthritis) trials. The chapters are presented in mixed manuscript form and journal format owing to differing stages of publication, and are designed to be read independently. Chapter one presents an introduction to the current understanding of the pathogenesis of OA and the role of imaging in clinical research. Chapter two presents published data from a cross-sectional survey that included 1039 participants via an online platform and utilised the validated ICOAP (Measure of Intermittent and Constant Osteoarthritis Pain) and EQ5D scores. The data identified the knees and hands as the regions in which pain was the most common, and in which arthropathy was the most detrimental to activities of daily living. This regional distribution differed slightly from studies in which participants were drawn from clinician referral, rather than from a convenience sample, raising points on the potential advantages of online platforms, and framing the subsequent focus on research into imaging biomarkers for OA of the knees and hands. Chapter three provides a further introduction to magnetic resonance imaging (MRI) to provide a background for the subsequent two chapters, and consists of a published book chapter that focuses on MRI as an imaging modality that enables direct visualisation of changes in both structure and composition in different joint tissues through the modification of contrast and sequences. In chapter four, post-traumatic osteoarthritis is discussed in two published manuscripts in terms of its value in investigating disease pathogenesis through imaging. Traumatic joint injury, and anterior cruciate ligament (ACL) rupture in particular, is strongly linked to the subsequent development of osteoarthritis, so provides an opportunity to track pathological changes from a defined starting point, and therefore potentially identify early imaging markers for subsequent disease. Chapter five presents data from serial MRIs in the KANON trial, which included 121 individuals who had sustained an acute anterior cruciate ligament (ACL) rupture to a previously uninjured knee. In the trial, 62 participants were randomised to undergo early ACL reconstruction and structured rehabilitation, and 59 were allocated to undergo structured rehabilitation alone, with an optional delayed ACL reconstruction. MRIs of the knee were obtained for all participants at baseline, two years, and five years, and for a subgroup of 63 participants, additional MRIs were performed at 3, 6, and 12 months. The serial MR imaging has enabled the effects of concomitant injuries such as osteochondral fractures and meniscal tears to be tracked, as well as providing longitudinal measurements of changes in cartilage and subchondral bone that can then be correlated with clinical outcomes. Data on the regional changes in the area of bone covered by cartilage (cAB) as a potential biomarker for disease, and the effects of baseline injuries to other joint structures, are presented. In chapter six, data on radiographic markers in trapeziometacarpal (base of thumb) osteoarthritis are presented from the COMBO trial. It includes two manuscripts – one published and one in the process of review – that present data from the first 100 participants included in the COMBO trial. Radiographic markers have traditionally had a poor correlation with symptomatic and functional outcomes, so generalised estimating equations are used for bilateral data to minimize the influence of interpersonal confounding factors and account for the fact that within-person measurements are not independent. The radial subluxation ratio is investigated in the manuscripts as a marker for structural and functional osteoarthritis progression, and radiographic markers of disease severity are analysed in relation to pain and functional outcomes in symptomatic disease. In summary, magnetic resonance imaging and radiographic markers are investigated in relation to their utility as indicators of osteoarthritis progression, and in acting as outcomes for the assessment of potential interventions in clinical trials.