Corticosteroids (CS) are commonly used to reduce inflammatory symptoms in Crohn's disease (CD) patients. The risk of long-term side effects constrains their recommended use to shorter treatment, such as bridging therapy while switching treatments. The aim of this study was to assess the use of concomitant CS in patients beginning vedolizumab in the real-world setting.Methods:
Adult patients with CD (ICD-9 code: 555.x) initiating vedolizumab between May 1, 2014 and May 18, 2016 were extracted from the Explorys Universe electronic medical record database and insurance claims for 50 million US patients (approximately 15% of the US population). Patients with ≥180 days of follow-up after vedolizumab initiation and ≥365 days of medical history were included. The main CS of interest were prednisone, budesonide, or hydrocortisone. Concomitant CS use was defined as having ≥2 oral CS prescriptions between 30 days before and 98 days after vedolizumab initiation, including ≥1 prescription within 30 days of initiation. Successful completion of vedolizumab induction was defined as having ≥3 infusions within the first 98 days of vedolizumab therapy, and a fourth within 90 days after the third. Demographic and clinical characteristics were described at baseline as well as the use of CS (in oral, intravenous, or rectal form) and immunomodulators (azathioprine, 6-mercaptopurine, cyclosporine, tacrolimus, methotrexate) at maintenance; results were stratified by concomitant CS use.Results:
Two hundred forty CD patients initiated vedolizumab treatment; 14.2% (N = 34) used concomitant oral CS. At vedolizumab initiation, the mean age of concomitant CS patients was 40 years (standard deviation [SD]: 15.3) versus 44 years (SD: 15.3) for patients not on CS. For concomitant CS and non-CS patients, respectively, both groups had more female patients (58.8% versus 65.5%), the mean disease duration was 6.5 versus 6.0 years since diagnosis, and 79.4% versus 70.9% had been previously treated with biologics. In the 90 days before vedolizumab initiation, all-cause and CD-related hospitalizations occurred in 50.0% and 32.4% of concomitant CS patients, respectively, versus 31.6% and 18.0% of non-CS patients, respectively. CD-related surgery was performed in 5.9% of concomitant CS versus 5.8% of non-CS patients. Successful completion of vedolizumab induction in the concomitant CS group was 67.6% (95% Wilson confidence interval [CI]: 50.8–80.9) versus 65.5% (58.8–71.7) in the non-CS group. Among concomitant CS patients, 17.6% continued using CS between day 150 and 180 (95% CI, 8.3–33.5); and 14.6% of patients not initially on CS were treated with CS during the same period (95% CI, 10.4–20.0). The use of immunomodulators after day 98 was 14.7% (95% CI, 6.4–30.1) among concomitant CS patients versus 8.7% (95% CI, 5.6–13.4) among non-CS patients.Conclusions:
One out of 7 patients initiating vedolizumab in this real-world clinical practice study initiated therapy while on oral CS. Concomitant CS use occurred more frequently among patients who experienced recent hospitalizations, and had prior use of biologics. Although this was a descriptive analysis, induction completion rates for vedolizumab appeared comparable regardless of concomitant CS use at vedolizumab initiation. Of patients who received concomitant CS, more than 3-quarters did not receive any CS within 6 months of vedolizumab initiation.