Search in the Abstract Database

Abstracts Search 2017

* = Presenting author

P531 Discontinuation of corticosteroids among Crohn's disease patients treated with vedolizumab in the United States

Patel H.*1, Chastek B.2, Null K.D.3, Demuth D.4

1Immensity Consulting, Inc., Chicago, United States 2Optum, Eden Prairie, United States 3Takeda Pharmaceuticals U.S.A., Inc., Health Economics and Outcomes Research, Deerfield, United States 4Takeda Development Center Europe Ltd, Evidence and Value Generation, London, United Kingdom

Background

Corticosteroids (CS) are effective in the short-term induction of patients with moderate to severe Crohn's disease (CD) but not for maintenance of remission, due to the associated risks. Vedolizumab (VDZ), a humanized monoclonal anti-α4β7 integrin antibody, is approved for the treatment of adults with moderately-to-severely active CD. This study assessed VDZ treatment persistence and CS discontinuation among CD patients co-induced with CS.

Methods

Adult (≥18 years) CD patients initiating VDZ between 1 May 2014 and 30 September 2016 were identified in the US Optum Research Database. Patients with ≥12 months history (baseline) before their first VDZ claim (index date) and who completed induction (defined as ≥3 infusions in ≥98 days post-index) were included. CS-related measures included: dependence (≥80% CS use during the 6 months immediately prior to index date), co-induction with CS (CS fill for ≥28 days during the induction phase), CS discontinuation (treatment gap ≥60 days between CS fills) while on VDZ therapy. VDZ persistence was defined as no treatment gap ≥90 days between consecutive infusions. CS discontinuation and VDZ persistence were measured using the Kaplan-Meier method.

Results

A total of 242 VDZ patients were included with a mean (SD) age of 43.8 (14.6) years; 60% female, median follow-up period of 280.5 days. During baseline, 33%, 41% and 81% of patients were treated with aminosalicylates, immunomodulators, and CS, respectively; 71% of patients had received a biologic before initiating VDZ. Of CD VDZ patients, 45% (n=109) were co-induced with CS during the induction phase, of whom 17% (18/109) were CS-dependent. Overall, 43% (47/109) of CS co-induced patients discontinued their CS and among CS-dependent patients, 28% (5/18) discontinued their CS. CS discontinuation and VDZ persistence are shown in Figures 1a & 1b.

Figure 1. a. Discontinuation of corticosteroids among patients with Crohn's disease treated with VDZ. b. Persistence with VDZ among patients co-induced with and without corticosteroids.

Conclusion

This real-world study, using a nationally representative US database, showed that over half of CD patients receiving VDZ were not co-induced with CS. Among VDZ patients co-induced with CS, 43% discontinued over the follow-up period. Despite the treatment-refractory patients included in this study, the CS discontinuation rate at 26 weeks among VDZ patients was higher than what was reported from the GEMINI clinical trials. VDZ persistence was similar between CS co-induced patients versus those without CS co-induction. Future studies should examine CS-related outcomes over a longer follow-up period.

The study was funded by Takeda Development Centre Europe Ltd. No medical writing assistance was provided.