Impact of approval of biologic disease modifying anti-rheumatic drugs on juvenile idiopathic arthritis outcomes

Introduction: Juvenile idiopathic arthritis (JIA) affects nearly 300,000 US children, causing pain and physical disability. In clinical trials, biologic disease modifying antirheumatic drugs (bDMARDs) have been shown to decrease disease activity. We examined the effects of increased use of bDMARDs in realworld settings.
Methods: We conducted a retrospective cohort study to assess outcomes in two cohorts of children diagnosed with JIA at Seattle Children’s Hospital (SCH). The exposure of interest was time period of JIA diagnosis representing pre- and post-bDMARD availability, with the earlier cohort consisting of all children diagnosed 1998-2000 (N=194) and the later cohort consisting of all children diagnosed 2015-2017 (N=240). Electronic medical records and chart review were used to assess dates of diagnosis and outcomes of achieving inactive disease, disease flare occurrence, and hospitalization for infection. Poisson regression was used to estimate relative risks (RRs) and 95% confidence intervals (CIs) for these outcomes among children in the more recent cohort relative to the early cohort, adjusted for age at diagnosis, JIA category, and receipt of any intraarticular glucocorticoid injection. Children were followed up for up to three years. The mean days to inactive disease and time in inactive disease were also assessed and compared using t-tests.
Results: Among children with three full years of follow up, the mean numbers of days to inactive disease were 237 days (sd 233) in the later cohort vs. 306 days (sd 239) in early cohort (p=0.02). The mean numbers of days in inactive disease for those in the later and early cohorts were 652 days (sd 306) and 621 days (sd 281, p=0.39), respectively. RR for achievement of inactive disease status in the late vs. early cohort was RR 1.01 (95%CI: 0.94-1.07). Among children who had achieved inactive disease status, the RR for flare occurrence was 1.21 (95% CI: 0.99-1.50). These results did not vary greatly by sex or race/ethnicity, but the risk of flare occurrence was slightly greater among children with oligoarticularpersistent JIA (RR 1.37, 95% CI: 0.86-1.97). Hospitalization for infection was not assessed due to low numbers.
Conclusion: With the exception of fewer days until inactive disease and a suggestion of increased time in inactive disease in the later cohort when bDMARDS were more in use, we did not observe strong evidence of differences in these disease outcomes between the cohorts in which these drugs were more and less available. Further studies to examine time to inactive disease, time to disease flare after inactive disease is achieved, and time in inactive disease is warranted to better understand the impact of the approval of bDMARDs on JIA outcomes outside of clinical trials.