Background/Purpose: Methotrexate (MTX) is cornerstone of therapy for rheumatoid arthritis. However, response to MTX is variable and unpredictable among patients. Therefore, there is a critical need to identify potential biomarkers to improve drug response to MTX in arthritis. This study aims to evaluate MTX and circulating folates in the response to MTX treatment, and to explore metabolomic data to identify potential biomarkers associated with disease activity in collagen-induced arthritis (CIA) mouse.

Methods: The study included untreated healthy control group (n=10), MTX treated control group (n=10), untreated CIA group (n=8), and MTX treated CIA group (n=9). As endpoint analysis, disease activity score using an established scoring system were determined and paw volume using limb water volume displacement were measured. Plasma or erythrocyte samples were collected and submitted to the National Institutes of Health West Coase Metabolomics Center to measure intermediates of primary metabolism, biogenic amines, and lipids. Raw peak intensity data was utilized to identify and rank metabolites using MataboAnalyst 4.0. The metabolites with significant fold change and false discovery rate adjusted R values in the induction of disease or in the treatment of MTX were subjected to enrichment analysis using Chemical Similarity Enrichment Analysis for Metabolomics (ChemRich) and mapping data were visualized using Cytoscape 3.7.2.

Results: MTX efficacy in CIA and control mice with or without MTX treatment was evaluated. Compared to control group, CIA mice appeared significant induction in both disease activity scores (Mean±S.E.M., 9 ± 1 vs 0 ± 0, P < 0.001) and paw volume (0.48 ± 0.02 ml vs 0.35 ± 0.01 ml, P < 0.001).  CIA group treated with MTX had reduced disease activity scores (1 ± 0.4 vs 9 ± 1, P < 0.001) and paw volume (0.37 ± 0.01 ml vs 0.48 ± 0.02 ml, P < 0.001) compared to the CIA untreated group. A concentration of 5mTHF in RBCs was significantly influenced by the induction of disease (P < 0.05) and the exposure of MTX (P < 0.001). Similarly, a significant diffrence in a concentration of 5mTHF in plasma was observed between untreated CIA group and MTX treated CIA group (9.34 ± 3.18 nM vs 45.03 ± 10.84 nM, P < 0.01). RBC 5mTHF was postiviely correlated with disease activity score (⍴ = 0.740, P = 0.001) and paw volume (⍴ = 0.709, P = 0.001) in CIA mice.  Metabolomic analysis using ChemRich revealed most significant changes in the unsaturated triglycerides in the induction of disease and unsaturated phosphatidylcholines in the treatment of MTX.

Conclusion: Our study demonstrated that MTX treatment recovered a reduction in arthritis disease activity in the CIA mouse model. Associations of pharmacologic folates level with MTX efficacy suggest that erythrocyte 5mTHF levels as a potential predictor of MTX response. Metabolomic analysis of plasma generated a list of metabolites that significantly correlated with disease activity. The results suggest that metabolomic approach may lead to identify potential biomarkers in MTX response in autoimmune arthritis.

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/evaluation-of-methotrexate-efficacy-and-exploration-of-metabolites-associated-with-disease-activity-in-collagen-induced-arthritis-mouse-model/