DOP073. Efficacy of induction treatment with vedolizumab for patients with Crohn's disease who have experienced tumour necrosis factor antagonist failure or are tumour necrosis factor antagonist naive
W.J. Sandborn1, P. Rutgeerts2, J. Xu3, B. Abhyankar4, I. Fox5, 1University of California San Diego, Division of Gastroenterology, La Jolla, United States, 2Katholieke Universiteit and University Hospital Gasthuisberg, Division of Gastroenterology, Leuven, Belgium, 3Takeda Pharmaceuticals International Company, Biostatistics, Cambridge, United States, 4Takeda Development Centre (Europe) Ltd, Clinical Science, London, United Kingdom, 5Takeda Pharmaceuticals International Company, Clinical Development, Cambridge, United States
Tumour necrosis factor (TNF) antagonists have improved the care of patients with treatment-refractory Crohn's disease (CD) but may lead to primary/secondary therapy failure and increased risk of serious infection. In this post hoc analysis of data from the phase 3 GEMINI 2 and 3 studies, the efficacy of induction therapy with vedolizumab (VDZ), an investigational, gut-selective, humanized, anti-alpha4beta7 integrin monoclonal antibody, was evaluated in patients with CD and previous TNF antagonist failure and in those who were TNF antagonist naive.
Data were pooled from the randomised GEMINI 2 (NCT00783692) and GEMINI 3 (NCT01224171) trials of patients with CD who received intravenous infusions of induction therapy with VDZ 300 mg or placebo (PBO) at weeks (wks) 0, 2, and 6. In GEMINI 2, a 6-wk induction trial was followed by rerandomisation of wk 6 responders into a 46-wk maintenance trial; GEMINI 3 was a 10-wk induction trial. For the TNF antagonist failure and TNF antagonist-naive subgroups, proportions of patients in clinical remission (Crohn's Disease Activity Index [CDAI] score ≤150 points) and those with a CDAI-100 response (≥100-point decrease in CDAI score) were evaluated at wks 6 and 10. Wk 6 responders in GEMINI 2 who were rerandomised to PBO were included in the wk 10 analyses.
At baseline, pooled mean CD duration (VDZ, 9.4 y; PBO, 9.2 y) and CDAI score (VDZ, 321.7; PBO, 311.0) were similar between the VDZ and PBO groups, as was the rate of concomitant immunosuppressive use in each study (approximately one-third in each group); TNF antagonist failure was more common in GEMINI 3 (VDZ, 76%; PBO, 76%) than in GEMINI 2 (VDZ, 48%; PBO, 47%). In post analyses, the proportion of patients in remission was significantly higher with VDZ than with PBO at wk 10, but not wk 6, for the TNF antagonist failure subgroup (Table). At wks 6 and 10, proportions of patients from the TNF antagonist-naive subgroup in remission and of those from both subgroups with a CDAI-100 response were significantly higher with VDZ (Table). There were no clinically important between-group differences in adverse event profiles.
|End point||Number (%) of patients [P-value a]|
|TNF antagonist failure||TNF antagonist naive|
|VDZ (n = 263)||PBO (n = 227)||VDZ (n = 154)||PBO (n = 123)|
|Wk 6 clinical remission||35 (13.3) [0.157]||22 (9.7) [−]||35 (22.7) [0.005]||13 (10.6) [−]|
|Wk 10 clinical remission||57 (21.7) [0.0008]||25 (11.0) [−]||38 (24.7) [0.044]||19 (15.4) [−]|
|Wk 6 CDAI-100 response||87 (33.1) [0.005]||51 (22.5) [−]||62 (40.3) [0.032]||34 (27.6) [−]|
|Wk 10 CDAI-100 response||103 (39.2) [<0.0001]||51 (22.5) [−]||72 (46.8) [0.006]||37 (30.1) [−]|
|Data are from post hoc analysis. PBO, placebo; VDZ, vedolizumab.|
a P-value versus PBO and based on the Cochran–Mantel–Haenszel chi-square test.
In patients with CD for whom TNF antagonists had failed and in those who were TNF antagonist naive, remission and CDAI-100 response rates were higher with VDZ induction therapy than with PBO. In patients with previous TNF antagonist failure, the wk 10 (but not wk 6) remission rate was significantly higher with VDZ than with PBO.