ABSTRACT
BackgroundUnderstanding the dynamics of humoral immunity after COVID-19 vaccination is crucial in developing vaccination strategies. Antibody response patterns are more complex in patients with rheumatoid arthritis (RA) because of their underlying autoimmunity and immunosuppressive medications. The kinetics of vaccine response in RA patients are not well understood.ObjectivesTo construct a model of antibody response to COVID-19 vaccination in patients with RA.MethodsTwo patient groups were included for the study. The first group was composed of RA patients who were enrolled for influenza vaccination study between Oct 6, 2021 and November 3, 2021, in whom serial serum samples were obtained 0, 4, 16 weeks after vaccination. The second group was consecutively enrolled from outpatient clinic between October 6, 2021 and June 3, 2022, in whom serum sample was obtained once. After collecting data on demographics, vaccination and infection history of COVID-19 were obtained by self-report via questionnaire and data from Korean center for disease control. We then measured antibody titers against receptor binding domain of spike protein (anti-RBD) and nucleocapsid (anti-N), using Chemiluminescence microparticle immunosaasy (Abbott, USA) and Electrochemiluminescence immunoassay (Roche, Germany) respectively. The anti-RBD titer was log-transformed to improve normality. Time from vaccination and log of anti-RBD titer was modeled using fractional polynomial. Covariates including age, sex, BMI, underlying disease and immunosuppressive drugs were analyzed using Generalized Estimating Equations to account for repeated measured from a subject.ResultsA total of 736 patients (1042 samples) were enrolled. After excluding patients who experienced COVID-19 infection before sampling (n=84), those unvaccinated (n=44) and uncertain COVID-19 infection history (n=59), the data on 778 samples from 549 patients were analyzed (Group 1: 125, Group 2: 424). Antibody titer reached peak at 12 days after vaccination and decreased exponentially (Figure 1) which fell to 36.5% from peak after 2 months. Compared to the first vaccination, the 3rd and 4th vaccination significantly shifted anti-RBD antibody response curve (28 times, 95% CI 4~195;32 times 95% CI 4~234, respectively). However, there was no significant shift after the 4th vaccination from the 3rd vaccination (p=0.6405). Multivariable analysis showed that number of vaccinations and sulfasalazine (coefficient: 0.40, 95% CI 0.12~0.68) increased vaccine response but age (coefficient: -0.03, 95% CI -0.04~-0.02), abatacept (coefficient: -2.07, 95% CI -3.30~-0.84) and, JAK inhibitor (coefficient: -0.82, 95% CI -1.34~-0.31) decreased vaccine response.ConclusionAnti-RBD response to COVID-19 vaccination showed a peak at 12 days after vaccination and then exponentially decreased in patient with RA. The antibody response is affected by age and medications used for the treatment of RA.Table 1.ln[RBD (U/ml)]coefficient (univariable)95% CIp-valuecoefficient (multivariable)95% CIp-valuesex (female)0.17-0.22, 0.550.393---age-0.02-0.03, -0.01<.001**-0.03-0.04, -0.02<.001**DM0.11-0.27, 0.500.568---HTN-0.38-0.69, -0.070.018*---CKD0.680.07, 1.290.030*---RA duration (yr)-0.04-0.06, -0.010.001**---Pd (mg/d)-0.06-0.11, 0.000.035*---MTX use-0.23-0.52, 0.050.105---HCQ use0.01-0.28, 0.290.965---SSZ use0.450.07, 0.840.022*0.400.12,0.680.005**LEF use0.00-0.37, 0.370.988---TNF inhibitors use0.29-0.16, 0.730.208---Abatacept use-2.07-3.14, -0.99<.001**-2.07-3.30, -0.840.001**JAK inhibitors use-0.88-1.52, -0.240.007**-0.82-1.34, -0.310.002**Time (months)log(t)-1.96-2.37, -1.54<.001**-1.90-2.29, -1.50<.001**t