Background:Bilirubin is an antioxidant with anti-inflammatory properties. In previous reports, serum bilirubin levels were correlated with disease activity of autoimmune diseases including rheumatoid arthritis (RA). Various molecular-targeted agents have been developed for RA, and targets, such as IL-6 and TNFα, are associated with liver function. However, the association between serum bilirubin and treatment response in RA patients treated with molecular-targeted agents is still unknown.Objectives:We aimed to evaluate the role of serum bilirubin in the prediction of the early treatment response in RA patients who initiated molecular-targeted agents.Methods:We retrospectively recruited biologic naïve RA patients (n=292) with moderate-to-high disease activity from a tertiary hospital between Jan 2013 and Dec 2019. Patients with viral hepatitis, drug-induced hepatitis, or alcoholic liver disease were excluded. Molecular-targeted agents included tocilizumab (TCZ, n=40), adalimumab (ADA, n=59), etanercept (ETN, n=66), golimumab (GOL, n=60), abatacept (ABA, n=31), and tofacitinib (TOF, n=36). Clinical and laboratory data were collected from electronic medical records. Patients were categorised into an increased bilirubin group (higher serum bilirubin at 3 months than at baseline) and decreased bilirubin group (equal or lower serum bilirubin at 3 months than at baseline). At 6 months of treatment, good response (defined as a DAS28 score ≤3.2) was evaluated. Multivariate logistic regression analysis and multiple linear regression analysis were used to evaluate the association between serum bilirubin and treatment response. The variables included in the multiple logistic and linear regression analyses were age, female sex, rheumatoid factor, prednisolone, DMARDs, baseline liver enzymes, baseline DAS28 score, and components.Results:The mean serum bilirubin level at baseline was 4.7±1.8 mg/L. After 6 months of treatment, 180 (61.6%) patients achieved good responses. The mean serum bilirubin levels at 3 and 6 months were 5.3±2.3 and 5.5±2.2 mg/L, respectively. At 6 months, a good response was more frequent in the increased bilirubin group than in the decreased bilirubin group (71.2% [99/139] vs. 52.9% [81/153], p=0.001). In multivariate logistic regression analysis, the ORs among good responders at 6 months were 1.221 (95% CI 1.014–1.471, p=0.036) for baseline serum bilirubin and 1.377 (95% CI 1.146–1.654, p=0.001) for the change in serum bilirubin at 3 months. According to target agents, the mean changes in serum bilirubin from baseline to 6 months were 1.9±2.5 for TCZ, 1.0±1.5 for ADA, 0.7±1.9 for ETN, 0.6±2.2 for GOL, 0.3±1.2 for ABA, and 0.4±2.2 for TOF (Figure 1). Among the target agents, TCZ showed a significant increase in the mean serum bilirubin level at 3 and 6 months from baseline. In multiple linear regression analysis performed on TCZ, the change in bilirubin at 3 months was associated with the DAS28 score at 6 months (β=−0.349, p=0.020).Figure 1.Change in serum bilirubin during treatment with molecular-targeted agents in rheumatoid arthritis patientsConclusion:High baseline serum bilirubin and an increase in serum bilirubin during treatment are helpful to predict a good response to molecular-targeted agents, especially TCZ.Disclosure of Interests:None declared
Synovial cellular and molecular signatures stratify clinical response to csDMARD therapy and predict radiographic progression in early rheumatoid arthritis patients