Search in the Abstract Database

Search Abstracts 2013

* = Presenting author

P513. Efficacy of vedolizumab in ulcerative colitis by prior treatment failure in GEMINI I, a randomised, placebo-controlled, double-blind, multi-centre trial

B. Feagan1, B. Sands2, S. Sankoh3, C. Milch3, I. Fox3, 1Robarts Research Institute, University of Western Ontario, London, Canada, 2Mount Sinai School of Medicine, Division of Gastroenterology, New York, United States, 3Millennium Pharmaceuticals, Inc., Cambridge, United States

Background

In an exploratory analysis, we evaluated the efficacy of vedolizumab (VDZ) in subgroups of patients with ulcerative colitis and failure of specific classes of drug therapy (GEMINI I; NCT00783718).

Methods

Eligible patients had Mayo score (MS) ≥6 and endoscopic subscore ≥2 despite prior therapy. After VDZ induction doses (week 0, week 2), patients with clinical response (MS reduced ≥3 points and ≥30% from baseline, plus decrease in rectal bleeding subscore of ≥1 point or absolute rectal bleeding subscore ≤1 point) at week 6 were randomised to intravenous (IV) VDZ 300 mg every 4 weeks (Q4W), VDZ 300 mg IV Q8W, or placebo (PBO) for 46 weeks. Induction phase primary outcome was clinical response at week 6. Maintenance phase outcomes were remission (primary endpoint; complete MS of ≤2 points and no individual subscore >1 point) and mucosal healing (Mayo endoscopic subscore of ≤1 point) at week 52. Patients were categorised by failure of prior treatment with anti-tumour necrosis factor (TNF), an immunomodulator (IM) but not anti-TNF, or corticosteroids (CS) only.

Results

Overall, patients receiving VDZ had improved clinical response, remission, and mucosal healing vs PBO.

Although CIs for treatment differences included 0, VDZ efficacy was similar across prior treatment failure subgroups.

Table 1. Clinical response at Week 6 (ITT).
EndpointOverallPrior Treatment Failures
   Anti-TNFIM but not Anti-TNFCS only
 PBO
N = 149
VDZ
N = 225
PBO
N = 63
VDZ
N = 82
PBO
N = 55
VDZ
N = 96
PBO
N = 25
VDZ
N = 42
Clinical response, %25.547.120.639.034.549.020.059.5
 Difference (95% CI)21.7 (11.6, 31.7)
P<0.0001
14.4 (−1.6, 30.5)39.5 (15.3, 60.6)
*P < 0.0001.
Table 2. Efficacy endpoints at Week 52 (ITT).
 OverallPrior Treatment Failure
    Anti-TNFIM but not Anti-TNFCS only
 PBO
N = 126
VDZ Q8W
N = 122
VDZ Q4W
N = 125
PBO
n = 38
VDZ Q8W
n = 43
VDZ Q4W
n = 40
PBO
n = 61
VDZ Q8W
n = 56
VDZ Q4W n = 60PBO
n = 26
VDZ Q8W
n = 19
VDZ Q4W
n = 25
Clinical remission, %15.941.844.85.337.335.018.044.650.026.942.148.0
 Difference (95% CI) vs PBO 26.1* (14.9, 37.2)29.1* (17.9, 40.4) 31.9 (10.3, 51.4)29.7 (7.4, 49.4) 26.6 (10.4, 42.8)32.0 (16.1, 47.9) 15.2 (−12.8, 43.2)21.1 (−4.9, 47.0)
Mucosal healing, %19.851.656.07.941.947.524.653.660.026.968.460.0
 Difference (95% CI) vs PBO 32.0* (20.3, 43.8)36.3* (24.4, 48.3) 34.0 (12.6, 53.2)39.6 (18.1, 58.5) 29.0 (12.0, 45.9)35.4 (19.0, 51.9) 41.5 (14.5, 68.5)33.1 (7.4, 58.8)
*P < 0.0001.

Conclusion

VDZ showed improved efficacy vs PBO in inducing clinical response at week 6. In patients with a clinical response at week 6, clinical remission was achieved over 52 weeks in more patients with VDZ vs PBO and VDZ showed improved efficacy compared with PBO in mucosal healing at week 52, regardless of the type of prior treatment failure.