Background/Purpose: The current treatment paradigm in RA is to attempt to decrease concomitant use of oral glucocorticoids (OGC). This study examined the OGC-sparing effects of Rituximab in U.S. RA patients who had switched from an anti-TNF therapy.

Methods: Retrospective cohort study using a large U.S. administrative claims database. Patients selected for study were adults aged ≥18 years who had had been diagnosed with RA between 1/1/2004-3/31/2011 had initiated Rituximab treatment after treatment with anti-TNF therapy between 3/1/2006-3/31/2011, and had OGC use within 30 days prior to Rituximab initiation. Rituximab ‘exposure periods’ were constructed using Rituximab infusion dates and the recommended administration frequency schedule, equaling the duration of time from initiating Rituximab until the first occurrence of 1) switch to a different biologic disease modifying antirheumatic drug (BDMARD), 2) 90-day gap in treatment with Rituximab, 3) disenrollment from health insurance, 4) reaching 3/31/2011, or 5) reaching a maximum study follow-up of 36 months. Patients were excluded if within the 12-month period prior to Rituximab initiation they had been diagnosed with a non-RA indication for any BDMARD.

During the Rituximab exposure periods the study outcomes, proportion of patients using OGCs and average OGC dose expressed in prednisone equivalent, were measured in sequential 90-day intervals, the first of which commenced as of Rituximab initiation. A multilevel random coefficient model was used to test for statistically significant OGC-sparing effects over time.

Results: A total of 952 Rituximab exposure periods from 938 unique patients were included for study; average age 54.7 years, 79.2% female. Table displays the study results. The proportion of patients using OGCs decreased over time, with slight non-monotonic fluctuations, from 80.6% of patients in months 1-3 to 54.1% of patients remaining on therapy in months 34-36. The mean OGC dose also decreased over time, again with slight non-monotonic fluctuations, from 5.7 mg among patients in months 1-3 to 3.2 mg among patients remaining on therapy in months 34-36. The decreases in the proportion of patients using OGCs (P<0.0001) and the average OGC dose (P<0.0001) were both statistically significant. In sensitivity analysis subset to only patients remaining on therapy through months 34-36, the OGC use patterns were similar to those from among overall sample.

Conclusion: In this study, statistically significant OGC-sparing effects were observed in RA patients who had switched to Rituximab after treatment with anti-TNF therapy. The clinical implications of such OGC-sparing effects warrant further investigation.

Table. OGC use over time since Rituximab initiation

*Average daily dose, expressed in prednisone equivalent (mg), calculated among all patients