A propensity score‐matched, real‐world comparison of ustekinumab vs vedolizumab as a second‐line treatment for Crohn’s disease. The Cross Pennine study II

Abstract

Background:The optimal choice of biological agents after failure of anti-tumournecrosis-factor-(TNF)α agent in Crohn's disease (CD) is yet to be defined. Aims:To assess the effectiveness and safety of ustekinumab compared to vedolizumab as second-line treatment in CD patients who failed anti-TNFα therapy.Methods: Retrospective analysis of clinical response and remission at 14 and 52 weeks to ustekinumab by physician global assessment (PGA). A propensity scorematched analysis with a cohort treated with vedolizumab was performed.Results: Of 282 patients (mean age 40 ± 15, F:M ratio 1.7:1) treated with ustekinumab, clinical response or remission was reached by 200/282 patients (70.9%) at 14 weeks, and 162/259 patients (62.5%) at 52 weeks. Overall, 74 adverse events occurred, of which 26 were labelled as serious (8.3 per 100 person-year). After exclusion of patients without prior anti-TNFα exposure and patients previously exposed to vedolizumab or ustekinumab, we analysed 275/282 patients (97.5%) on ustekinumab and 118/135 patients (87.4%) on vedolizumab. Propensity score analysis revealed that at 14 weeks, patients treated with ustekinumab were 38% (95% CI 25%-50%; P < 0.001) more likely to achieve clinical remission, while at 52 weeks, the difference of 9% (95% CI −15% to 33%; P = 0.462) was not significant. Conclusions:Ustekinumab was effective and well tolerated in this real-world cohort.While ustekinumab proved more effective at 14-weeks, we found no statistically significant differences at 52 weeks compared to vedolizumab.

Letter: potential selection bias in the real‐world comparison of ustekinumab versus vedolizumab as a second‐line treatment for Crohn’s disease

“…A recently published article authored by Lenti et al puts forward the findings from a propensity score-matched comparison of ustekinumab versus vedolizumab as a second-line treatment for Crohn's disease based on real-world data. 1 Ustekinumab was found to have more clinical effectiveness at 14 weeks and no statistically significant difference at 52 weeks compared with vedolizumab. We think that addressing some of the following issues would enhance the scientific contribution of the study.…”

Letter: potential selection bias in real‐world comparison of ustekinumab versus vedolizumab as a second‐line treatment for Crohn’s disease—authors’ reply

“…5,6 We acknowledge Drs Adomi and Ray's concern that some of the Vedolizumab patients will have not had a chance of being exposed to Ustekinumab due to the later licensing of Ustekinumab. 1,2 We do not believe that this will have affected the results as calendar year certainly has no impact on clinical outcome. The inclusion of correct variables is paramount to make propensity score analysis meaningful.…”

“…We thank Drs Adomi and Ray for the interest in our study. 1,2 Therapeutic choices for Crohn's disease have significantly increased in the last decade. While this is great news for patients, questions over appropriate sequencing of first and subsequent lines of therapy remain.…”

“…1 We explicitly excluded all ustekinumab patients with prior vedolizumab exposure and vice versa from the propensity score matched comparison. 2 We read with interest the suggestion to use the active-comparator, new-user study design. 7 Sturmer et al recommend to include new-users (as done in our study), active-comparators (as done in our study), and consideration of induction and latent periods (as done in our study), and the use of propensity scores to balance measured covariates between treatment regimens.…”

“…We corrected for age at diagnosis, disease duration and number of previous biological drugs that are associated with outcome. 2 Drs Adomi and Ray are mistaken in claiming that we did not account for some individuals in the ustekinumab cohort having used vedolizumab before in our analysis. 1 We explicitly excluded all ustekinumab patients with prior vedolizumab exposure and vice versa from the propensity score matched comparison.…”