Search in the Abstract Database

Abstracts Search 2017

* = Presenting author

P442 Incidence of pneumonia and other respiratory tract infections with vedolizumab treatment for inflammatory bowel disease: clinical trial experience

Feagan B.1, Bhayat F.*2, Khalid J.M.3, Palo W.4, Blake A.5, Shetzline M.6, Travis S.7

1Robarts Research Institute, University of Western Ontario, Robarts Clinical Trials, London, Canada 2Takeda Development Centre Europe Ltd, Global Pharmacovigilance, London, United Kingdom 3Takeda Development Centre Europe Ltd, Evidence and Value Generation, London, United Kingdom 4Takeda Development Center Americas, Inc., Global Statistics and Statistical Programming, Deerfield, United States 5Takeda Development Center Europe Ltd, Global Pharmacovigilance, London, United Kingdom 6Takeda Pharmaceuticals International Co., Clinical Science, GI, Cambridge, United States 7University of Oxford, Translational Gastroenterology Unit, Nuffield Department of Medicine, Oxford, United Kingdom


Vedolizumab (VDZ) is a humanised monoclonal antibody that binds to α4β7 integrin and selectively blocks gut-specific lymphocyte trafficking. Although upper respiratory tract infections (URTIs) have been reported with VDZ [1] its gut selectivity may reduce the risk of respiratory tract infections (RTIs), including pneumonia, compared with therapies producing systemic immunosuppression (e.g. anti-tumour necrosis factor-alpha [TNFα] agents). We aimed to determine the incidence of RTIs associated with VDZ treatment in the clinical trial setting.


Safety data from VDZ phase 3 randomised controlled trials (RCTs) in ulcerative colitis (UC; GEMINI 1) and Crohn's disease (CD; GEMINI 2) and a phase 3 open-label extension (OLE; C13008 ongoing study in both UC and CD) were used to calculate the incidence of lower RTIs (LRTIs), including pneumonia, and URTIs using the MedDRA High Level Terms “lower respiratory tract and lung infections” and “upper respiratory tract infections”. A Cox proportional hazards model was used to identify potential predictors of these AEs for the pooled GEMINI 1 and 2 results.


Of 1731 patients in the pooled RCT population and 2243 patients in the OLE, mean (SD) ages were 37.9 (12.7) years and 39.1 (13.2) years, respectively; 43% and 50%, respectively, had disease duration ≥7 years; and 57% and 59%, respectively, had prior anti-TNFα therapy. In the RCTs, the rate of LRTIs, including pneumonia, was similar in the VDZ and placebo groups (Table 1). Rates of LRTIs were higher for VDZ-treated patients with CD than with UC. Most LRTI and URTI AEs were not serious and did not result in treatment discontinuation. Predictors of increased incidence of LRTIs were identified as being a current or former smoker, female sex and prior anti-TNFα therapy (Table 2). Predictors of URTIs were: being a current smoker, concomitant narcotic use and prior anti-TNFα therapy.

Table 1. Incidence of LRTIs and URTIs – number of patients with AEs in the maintenance phase (time-adjusted incidence rate per 100 patient-years)

Table 2. Predictors of LRTI AEs in GEMINI 1 and GEMINI 2


In this post hoc analysis, VDZ treatment of IBD was not associated with an increased incidence of LRTIs, including pneumonia, compared with placebo. Continuing pharmacovigilance will supplement these data and ongoing observational studies will further characterise these findings.


[1] Takeda Pharmaceuticals America, Inc. (2014), ENTYVIO® (vedolizumab) Prescribing Information