Reconciliation of Advanced Joint Biomechanics and Biological Assessment of the Joint to Develop a Personalised Approach to Osteoarthritis of the Knee

Abstract

Knee osteoarthritis (kOA) is a prevalent, progressive, and multifactorial joint disease characterised by cartilage degradation, inflammation and biomechanical dysfunction. This thesis applied a multi-omics approach to investigate synovial fluid proteomics, lipidomics, and gait biomechanics in a cross-sectional cohort. Paired synovial fluid and gait data were collected from individuals with early OA (eOA), advanced OA (aOA) and near normal (nN) controls, enabling group comparisons. The aim was to characterise molecular and mechanical features of OA and explore the relationship of these datasets through an integrative analysis. Synovial fluid (SF) was acquired intraoperatively from all donors following baseline gait assessment. Gait data were collected using 3D motion capture to derive spatiotemporal and kinematic parameters, with both discrete and continuous analyses used to assess group differences. SF proteins were extracted and digested using an S-Trap protocol prior to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. Protein data were processed using Proteome Discoverer, and pathway analysis was performed using PANTHER. Lipids were extracted using a methyl-tert-butyl ether (MTBE)/methanol method and non-polar fractions were analysed by LC-MS/MS. Pathway analysis was performed using LION/web. All data layers were integrated using Multi-Omics Factor Analysis (MOFA) to identify latent factors capturing shared variance across molecular and biomechanical domains. Proteomic analysis revealed group-specific changes in synovial fluid composition as well as several highlighting protein biomarker candidates including PRDX2, DBH ACAN, and HTRA1. Lipidomic profiling demonstrated a shift in lipid profiles in aOA, including increased abundance of membrane-associated and signalling lipids (e.g., phospholipids and sphingolipids) and reduced levels of energy storage lipids (e.g., triacylglercols). Neither ‘omic analysis showed clear group level separation. Gait analysis identified progressive spatiotemporal alterations and changes in sagittal plane kinematics. MOFA revealed, for the first time, latent factors reflecting shared variation across molecular and gait domains. One factor linked specific lipid species with variations in spatiotemporal and sagittal plane changes. To our knowledge, this is the first study to integrate multi-omics data with matched biomechanical gait analysis in knee osteoarthritis.

Date of Award	24 Nov 2025
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Leela Biant (Main Supervisor), Gwenllian Tawy (Co Supervisor) & Caroline Milner (Co Supervisor)