Background/Purpose: To compare the clinical efficacy of dose-tapering strategy of tumor-necrosis factor inhibitor (TNFi) with that of standard-dose TNFi treatment in patients with spondyloarthritis (SpA)

Methods: This observational cohort study was based on prospectively registered data of nationwide KOBIO registry. A total of 776 SpA patients with available data and more than one year of TNFi adherence were included. Quantification of TNFi dosage were expressed using dose-quotient (DQ), which was calculated as ¡®(actual prescribed dose/standard dose) x (standard dosing interval/actual dosing interval)¡¯ annually. We used linear mixed model to estimate the change in disease activity (ASDAS-CRP, BASDAI and BASFI) over 3 years between the two strategies. Classifying as dose-tapering was based on both mean DQ of patients (individual level) and DQs of all 1-year observations from all patients (time level). Incidence of discontinuation of TNFi due to inefficacy and adverse event were also compared.

Results: At the individual level, patients with mean DQ<1.0 (tapering group, n=322) were more likely to have obesity (33.5% vs. 29.4%, p<0.001) and be first TNFi users (82.9% vs. 75.1%, p=0.009) and showed higher CRP level (2.6¡¾3.1 vs. 2.1¡¾2.6 mg/dL, p=0.039) than those with standard-dose TNFi (standard-dose group, n=454). Other baseline clinical factors including age, gender, smoking status, HLA-B27 positivity, and disease activity were comparable. Change in ASDAS-CRP over time in the tapering group was not significantly different from that in the standard-dose group (p value for time*group=0.872). This result was consistent after adjustment for baseline factors (age, obesity, HLS-B27 and definite sacroiliitis) which showed relevant interaction with ASDAS-CRP changes (figure 1). At the time level, 186 (24.0%), 204 (35.6%) and 95 (41.9%) patients received reduced dose of TNFi during 0~1, 1~2 and 2~3-year period, respectively. There was no significant interaction between yearly DQ and relevant ASDAS-CRP change (p value for time*DQ=0.435) (figure 2). In the multivariable model, difference in ASDAS-CRP between the standard-dose TNFi and tapering TNFi over the entire follow-up was 0.10 (95% CI=-0.01 to 0.21). Two treatment strategies showed comparable efficacy regarding BASDAI and BASFI. Incidences of drug discontinuation due to inefficacy and adverse events between the two groups were also comparable (figure 2).

Conclusion: In SpA patients with at least one year of TNFi adherence, dose-tapering of TNFi showed a comparable clinical efficacy to standard-dose TNFi treatment.