AB0172 METABOLIC AND LIPOPROTEIN PROFILING IN RHEUMATOID ARTHRITIS WITH THE USE OF NMR-BASED METABOLOMICS

Background Rheumatoid arthritis (RA) remains a disease with high morbidity, due to the greater prevalence of cardiovascular disease. In contrast to the general population, systemic inflammation in RA lowers the circulating levels of lipids, a phenomenon called “the lipid paradox”. In addition to that, therapy is withdrawn in 50% of patients within 5 years, due to loss of efficacy or side effects. Momentarily predictive biomarkers for drug efficacy or side effects are missing, while a personalized approach in RA therapy is imperative. Several reports support the notion that specific metabolomic profiles are good predictors for response to MTX and anti-TNF therapy. Objectives The aim of this work is to depict in detail the metabolic profile of RA patients at different time-points of therapy DMARDs/bDMARDs in order to retrieve biomarkers related to their response to a given therapy, to monitor the disease progression and to predict the optimal disease management approaches. Methods Plasma was collected from fasted RA patients according to their therapy timepoint and organized in the following groups: a) newly diagnosed, without therapy (Naïve, n=15); b) patients having received therapies previously, with unstable disease, who were evaluated before receiving a new therapy (RAb, n=23); c) patients after having received a new therapy (RAa, n=14); and d) patients receiving any standard therapy (RAs, n=54), either DMARD or bDMARD, being in a stable condition. Finally, healthy subjects were enrolled as controls (n=33). Metabolomic profiling was carried out firstly with untargeted 1 H NMR spectroscopy, and secondly with in vitro diagnostic (IVDr) NMR spectroscopy with the lipoprotein subclass analysis (B.I.LISA), to quantify absolute concentrations of metabolites and lipoproteins. The acquired data were subjected to univariate and multivariate statistical analysis to investigate clustering of the groups and define the responsible molecules. Clinical parameters, including inflammation markers, DAS28, and comorbidities, were also included in the analysis, and Spearman correlation coefficient was calculated. Results Untargeted NMR data were analyzed with multivariate supervised approach (PLS-DA) revealing distinct metabolic signatures for the 6 groups under investigation. The most defined groups being RAb and RAs, compared to controls, which indicated changes in alanine, tyrosine, lactate and acetone. Besides small molecule, significant changes were also observed in various plasma lipoproteins. For the thorough investigation of these findings, a targeted lipoprotein subclass analysis was conducted and highlighted significantly higher lipoprotein subclass concentrations, including free cholesterol (FC), cholesterol (CH), phospholipids (PL) and apolipoprotein A1 subfractions in RAs compared to controls and Naïve. Concerning metabolite differentiations, RAs patients exhibited reduced ketone bodies and organic acids compared to RAb and control individuals, respectively. All RA groups had lower concentrations of sarcosine. Correlation analysis highlighted the association of DAS28, ESR and CRP with ketone body acetoacetate (p<10 -4 ) and sarcosine (p<10 -2 ). VAS correlated with HDL triglyceride subfractions (H1TG and H2TG, p<10 -5 ) and sarcosine (p=1.8x10 -4 ). All therapies were found to correlate with lipoproteins; MTX with LDL-2 subfractions (p=5x10 -4 ), intermediate-density lipoprotein (p=2.7x10 -5 ) and acetate (p=5.9x10 -6 ), Anti-IL-6R with VLDL cholesterol subfraction V1CH, (p=1.9x10 -3 ) and V2CH (p=1.5x10 -3 ), and Anti-CD20 with triglyceride fractions, IDTG and TPTG (p<10 -3 ). Conclusion Overall, these data reveal that RA patients have a distinct metabolic signature depending on the time-point of therapy. Clinical parameters correlated with changes in ketone bodies, amino and organic acids, while therapies correlated with lipoproteins. The above analysis indicates that biomarkers revealed by metabolomic profiling can be useful in RA therapy monitoring. Disclosure of Interests None declared