Using metabolomics to identify novel biomarkers for knee osteoarthritis risk
Investigators: Carrie Anne Karvonen-Gutierrez, Sioban D. Harlow, Charles Burant, Bin Nan, Alla Karnovsky
Funding: National Institute on Aging, 2017-2022 (1 K01 AG 054615 01)
Osteoarthritis (OA) is a debilitating age-related disease associated with pain, stiffness and poor functioning and it is a major risk factor for mobility disability. While early osteoarthritic changes within the joint commence during mid-life (40-65 years of age), early detection of disease is limited given lack of robust or reliable OA biomarkers thereby resulting in reliance upon costly imaging modalities. Thus, late detection of OA compromises the opportunity for intervention and prevention and instead leaves only symptom management or, ultimately, joint replacement as strategies for treatment. Appreciation that underlying metabolic dysfunction is a risk factor for osteoarthritis incidence and progression suggests that biomarkers which identify individuals with disordered metabolism may be relevant for OA. Metabolomics, a newly evolving field, analyzes small molecules (metabolites) in biological specimens. Metabolomics analysis has successfully identified novel biomarkers for diagnosis, monitoring and treatment for several age-related diseases. Further, a small but growing number of studies in animal and human populations have reported that metabolomics yields potential biomarkers with good discrimination between OA patients and normal controls including metabolites associated with collagen, branched chain amino acid, energy, and tryptophan metabolism. However, no studies to date have used metabolomics to identify biomarkers for OA incidence and among age- and body size-matched individuals. To address these analytical limitations, we propose to conduct a metabolomics analysis of osteoarthritis risk within the longitudinal Michigan Study of Women?s Health Across the Nation (MI-SWAN). Specifically, 63 MI-SWAN women who developed radiographic knee OA during follow-up will be age- and BMI-matched with 63 MI-SWAN women who remained OA-free during follow-up. Banked plasma specimens from baseline (when all subjects were OA-free) will be used to conduct metabolomics analyses using the targeted lipids eicosanoids platform (Aim 1) which includes profiles from 28 eicosanoids, the lipidomics platform (Aim 2) which profiles lipids from over 10 classes including 431 unique lipid species, and an untargeted platform (Aim 3) which profiles at least 250 known compounds to identify candidate biomarkers for knee osteoarthritis risk. Relative quantitation of these metabolites will be compared within the matched pairs of women who did and did not develop incident knee OA during follow-up using paired t-tests, Wilcoxon signed rank tests, and linear regression. This K01 award will provide training and skill development in metabolomics, the associated bioinformatics considerations, and translation to clinical care, thus enabling the candidate to apply this approach to other age-related diseases and conditions.