Impact and correlates of early knee osteoarthritis

Osteoarthritis (OA) is the most common form of arthritis involving chronic inflammation and structural alterations of the joint. Knee OA is the most prevalent type of OA that causes knee pain and reduced mobility. Globally over 500 million people suffer from OA, of whom around 350 million have knee OA. Knee OA and associated pain impart a substantial clinical, humanistic, and health economic burden on individuals and societies, and with an ageing population and rising prevalence of obesity globally, the burden of knee OA is increasing further. Despite this enormous unmet need, no disease-modifying OA drugs (DMOADs) are available that can arrest, slow, or reverse the progression of knee OA. The foremost reasons for the translational failure of DMOADs are likely manifolds, such as phenotypic heterogeneity and lack of sensitive biomarkers that can characterise OA, define its progression, and ascertain the clinical benefit of improving disease progression.
Furthermore, the current approach to managing knee OA is ‘reactive’ and focuses on treating established knee OA, characterised by chronic pain and radiographic evidence. However, by the time clinical or radiographic evidence of knee OA is established, significant and irreversible disease progression has already occurred, impairing disease prognosis. Consequently, to change the current ‘reactive’ approach to a ‘proactive’ one, there is a need to understand the correlates of early knee OA to enable early diagnosis, prevention, and improvement in disease progression. Hence with this background, this thesis investigated the impact and correlates of early knee OA.
We first ascertained the impact of knee pain on health health-related quality-of-life (HRQoL) in middle-aged adults utilising data from the Childhood Determinants of Adult Health (CDAH) study (CDAH, n=1567). Participants were selected from the third follow-up of the CDAH study (CDAH-3, year: 2014-19). HRQoL assessed as SF-6D health state utility values (HSUs) were generated from the participant-reported SF-12, and knee symptoms were assessed using Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) instrument. We found that knee symptoms were negatively associated with HRQoL, and an increase in knee pain over 6–9 years was associated with a reduction of HRQoL.
In the second study, we assessed the association of metabolic syndrome (MetS) with knee symptoms in middle-aged adults, utilising data from the first and third follow-ups of the CDAH study (CDAH-1, n=2447; CDAH-3, n=1549). The Fasting blood biochemistry, waist circumference and blood pressure measures were collected during CDAH-1 (year:2004-06, young-adulthood) and at 10–13-year follow-up at CDAH-3 (mid-adulthood). Participants were defined as having MetS as per International Diabetes Federation (IDF) definition, and knee symptoms were assessed using the WOMAC scale at CDAH-3. We found an independent positive association between MetS and knee symptoms. Furthermore, relative to those without MetS at either life stage, the elevation in mean knee pain scores was more pronounced for those who developed MetS after young-adulthood than those who had MetS in young-adulthood.
In the third study, we performed a comprehensive literature review to understand the current landscape of magnetic resonance imaging (MRI) and biochemical markers of knee OA to study the utility of these biomarkers in a cohort of middle-aged adults (CDAH study cohort) with knee symptoms considered as a proxy for early knee OA. Through literature review, pivotal studies in this domain, discussion with experts, and considering the budget and feasibility of analysing available serum samples and MRIs, we identified three serum biomarkers [cartilage oligomeric matrix protein (COMP), matrix metalloproteinase-3 (MMP-3), and hyaluronan (HA)] and five MRI markers [Morphological markers: cartilage thickness, cartilage volume, and subchondral bone area; Structural abnormality markers: cartilage defect and bone marrow lesions (BMLs)] to be studied further in study fourth, fifth, and sixth.
In our fourth study, we aimed to describe the associations between OA-related MRI biomarkers and knee symptoms in middle-aged adults followed over 6–9-years from CDAH-knee (n=329) study to CDAH-3 study (n=188). Knee MRI scans obtained at CDAH-knee (year: 2008–10) were assessed for morphological markers and markers of structural abnormalities. We found BMLs and cartilage defects to be positively associated with knee symptoms, whereas cartilage volume and thickness, and total bone area were weakly and negatively associated with knee symptoms. In our fifth study, we described the association between OA-related biochemical markers and knee symptoms in middle-aged adults followed for 10–13-years from CDAH-1 (n=156) to CDAH-3 (n=167). Serum samples collected during the CDAH-1 and -3 were analysed for three OA-related biomarkers (COMP, MMP-3, and HA) using non-isotopic enzyme-linked immunosorbent assay (ELISA) and knee symptoms were assessed at CDAH-3 using WOMAC scale. We found COMP, MMP-3, and HA positively associated with knee pain and symptoms in the population-based sample of middle-aged adults. Although the strength of these associations was modest, these biochemical markers measured in middle-aged adults appeared promising for further exploration as potential markers of joint pain.
In continuation to the above study, in our sixth study, we explored the associations between OA-related biochemical markers and MRI-markers in middle-aged adults followed over 10– 13-years from CDAH-1 (n=156) to CDAH-3 (n=167). The OA-biochemical markers were analysed at CDAH-1 and -3, and MRI-based imaging markers were available at CDAH-knee, which falls between CDAH-1 and -3 follow-up. We found an interdependent negative association between OA-biochemical markers and MRI-markers. The COMP and MMP-3 were negatively associated with knee cartilage thickness and volume assessed four years later, whereas cartilage volume and bone area were negatively associated with COMP and MMP-3 levels assessed 6–9-years later.
In conclusion, in middle-aged adults, the worsening of knee pain significantly negatively impacted their quality of life. MetS was associated with knee symptoms, and MetS in mid?adulthood had a more pronounced effect on the worsening of knee symptoms than MetS in young-adulthood. BMLs and cartilage defects were positively associated with knee symptoms, whereas cartilage volume, cartilage thickness, and total bone area were negatively associated with knee symptoms. OA-biochemical markers (COMP, MMP 3 and HA) were positively associated with knee pain, and there was an interdependent negative association between OA?biochemical markers and MRI-markers. These results indicate that the biochemical markers and MRI-markers can be used as markers of early knee OA.