Association of Lipid Mediators With Development of Future Incident Inflammatory Arthritis in an Anti-Citrullinated Protein Antibody-Positive Population

Abstract

Objective: To determine the association of polyunsaturated fatty acid (PUFA)-derived lipid mediators with progression from rheumatoid arthritis (RA)-related autoimmunity to inflammatory arthritis (IA).

Methods: We conducted a prospective cohort study using data from the Studies of the Etiology of Rheumatoid Arthritis (SERA). SERA enrolled first-degree relatives (FDRs) of individuals with RA (FDR cohort) and individuals who screened positive for RA-related autoantibodies at health fairs (screened cohort). We followed up 133 anti-cyclic citrullinated peptide 3.1 (anti-CCP3.1)-positive participants, 29 of whom developed IA. Lipid mediators selected a priori were quantified from stored plasma samples using liquid chromatography tandem mass spectrometry. We fit multivariable Cox proportional hazards models for each lipid mediator as a time-varying variable. For lipid mediators found to be significantly associated with IA, we then examined interleukin-1β (IL-1β), IL-6, IL-8, and tumor necrosis factor (TNF) as potential statistical mediators.

Results: For every 1 natural log pg/ml increase in the circulating plasma levels of proinflammatory 5-HETE, the risk of developing IA increased by 241% (hazard ratio 2.41 [95% confidence interval 1.43-4.07]) after adjusting for age at baseline, cohort (FDR or screened), and shared epitope status. The models examining 15-HETE and 17-HDHA had the same trend but did not reach significance. We did not find evidence that the association between 5-HETE and IA risk was influenced by the proinflammatory cytokines tested.

Conclusion: In a prospective cohort of anti-CCP-positive individuals, higher levels of 5-HETE, an important precursor to proinflammatory leukotrienes, is associated with subsequent IA. Our findings highlight the potential significance of these PUFA metabolites in pre-RA populations.

Figure 1.

Lipid mediator synthesis pathways by precursor fatty acid. Shaded ovals indicate the precursor fatty acid: ALA, EPA and DHA (top row) are omega-3 PUFA and LA, DGLA and ARA (bottom row) are omega-6 PUFA. Shaded rectangles represent lipid mediators available on the targeted panel. Non-shaded rectangles indicate relevant lipid mediators in the synthesis pathways that were not quantified on the targeted panel. An asterisk (*) denotes lipid mediators with more than 30% of values

Figure 2.

Illustration of a traditional mediation model to investigate whether cytokines mediated the relationship between lipid mediators and incident IA. First, we examined the total effect by regressing the outcome (incident IA) on lipid mediator levels. Next, we tested the relationship between lipid mediators and cytokines (measured at the same study visit), α, adjusting for covariates. Then, we tested the conditional relationship between cytokines and inflammatory arthritis, β, adjusting for the lipid mediator and other covariates.