Characterization of metabolomic profiles in late and early Osteoarthritis patients
Osteoarthritis (OA) is a prevalent disease in our society. However, a deep clinical understanding of this metabolic disease remains elusive due to the stratified nature of OA patients. Osteoarthritis presents differently in acute (no prior history of joint disease) and chronic (history of joint disease) patients. Recent studies suggest that chronic versus acute biofluids differ in the Glycine-Serine (Gly-Ser) and collagen breakdown pathways. Further investigation found that patients with a previous OA diagnosis show metabolic profiles more similar to healthy profiles as compared to acute profiels. While many studies have analyzed groupings of chronic/healthy and acute/chronic patients, few studies have compared all three groups. In order for these comparisons to be made, synovial fluid samples from chronic, acute, and healthy patients will be acquired and analyzed using Liquid Chromatography-Mass Spectrometry (LC-MS). Raw data collected from LC-MS will then be analyzed with Principal Component Analysis and Hierarchical Cluster Analysis to determine if groups naturally cluster together. Subsequently, Orthogonal Partial Least Squares-Discriminant Analysis will be used to identify metabolites that discriminate between cohorts. Metabolites with the highest discriminatory power (as determined by variable importance in projection scores) will be analyzed by pathway enrichment algorithms to identify perturbed metabolic pathways in acute versus chronic OA. Different metabolite profiles are expected for the three different groups with the acute patients as an outlier and the chronic patients relating closer to healthy patients. Acute patients should show distinct differences in the Gly-Ser pathway and collagen degradation pathway, while chronic patients should more closely resemble healthy patients. These discoveries could be used to address the heterogeneity of OA and would likely prove beneficial in the development of personalized treatment methods.