Towards Blood-Based Prediction of Knee Osteoarthritis Progression : Evaluation of Selected Best-in-Class Biomarkers in Plasma Using ELISA

Sammanfattning

Osteoarthritis (OA) is a common and heterogeneous joint disease often associated with pain, reduced function, and decreased quality of life. Despite the high prevalence of OA in the society, there are today still no disease-modifying treatment options available. Due to the limited understanding of the underlying mechanisms in OA progression, biomarkers have gained increasing interest. This since they are believed to improve both disease understanding and enable prediction of progression. A recent study by Zhou et al. identified a so called “best-in-class” serum proteomic biomarker panel consisting of proteins associated with clinically relevant knee OA progression, which they defined as worsening in both structure and pain. Although these findings are promising, further studies are needed to investigate whether these biomarkers can be measured using more accessible methods. The aim of this thesis was to investigate “best-in-class” OA biomarker candidates in plasma using commercially available Enzyme-Linked Immunosorbent Assay (ELISA). This was made to evaluate their potential relevance in knee osteoarthritis (KOA) in the BIOFUNC cohort. Plasma concentrations of C1R, CD44, CRTAC1, DBH, RBP4, TSP1, VTDB, and KNG1 were analysed in KOA individuals at baseline and 1-year follow-up, as well as in controls. Biomarker levels were first compared between KOA individuals and controls at baseline, both individually and as a combined biomarker profile. Longitudinal changes were then evaluated within the KOA group and analysed in relation to clinical progression group, pain progression, functional performance, and combined biomarker patterns. None of the investigated biomarkers significantly distinguished KOA individuals from controls at baseline, nor did the combined biomarker profile separate the two groups. However, baseline comparisons between clinical progression groups showed differences in TSP1 and VTDB levels. Longitudinal analyses showed considerable inter-individual variability, and no clear group-specific patterns were observed for most biomarkers. Similarly, the combined longitudinal biomarker profile did not clearly separate the progression groups. Longitudinal changes in CRTAC1 were significantly associated with changes in KOOS Pain in the Progressor group. In contrast, C1R, CD44, DBH, RBP4, TSP1, VTDB, and KNG1 showed weak and non-significant associations with pain progression. Further analyses of biomarker changes in relation to functional performance showed limited associations. Altogether, the findings of this project suggest that the investigated biomarker panel had limited ability to distinguish KOA individuals from controls or consistently classify progression groups in plasma within this cohort. However, longitudinal changes in CRTAC1 may be more informative than baseline concentrations alone. These results highlight both the potential and the challenges of translating proteomic biomarker discoveries into ELISA-based plasma measurements. Further work could include larger cohorts, longer follow-up periods and integration with structural imaging together with clinical outcomes to better evaluate the clinical usefulness of these biomarkers for predicting OA progression in this cohort.