P621 Real-world effectiveness and safety of vedolizumab and anti-TNF in biologic-naive Crohn’s disease patients: results from the EVOLVE study

B. Bressler*¹, G. Mantzaris², M. Silverberg³, P. Zezos³, D. Stein⁴, C. Colby⁵, T. Lissoos⁶, C. Lopez⁶, A. Natsios⁷, G. Radulescu⁸, H. Patel⁹, D. Demuth¹⁰, A. Yarur¹¹

¹St. Paul's Hospital, Vancouver, Canada, ²Evangelismos Hospital, Athens, Greece, ³IBD Center, Mount Sinai Hospital, Toronto, Canada, ⁴Evidera, London, UK, ⁵Evidera, California, USA, ⁶Takeda USA Inc., Chicago, USA, ⁷Takeda SA Inc., Athens, Greece, ⁸Takeda Canada Inc., Toronto, Canada, ⁹Takeda Pharmaceuticals International, Deerfield, USA, ¹⁰Takeda International - UK Branch, London, UK, ¹¹Medical College of Wisconsin, Milwaukee, USA

Background

This multi-country, retrospective chart review study assessed the effectiveness and safety of vedolizumab (VDZ) compared with anti-tumour necrosis factors (anti-TNF) agents in a real-world cohort of biologic (bio)-naïve Crohn’s disease (CD) patients.

Methods

Bio-naïve CD patients (≥18 years old) treated with VDZ or anti-TNF (May 2014 to March 2018), were included from sites in Canada, Greece and the USA. Data were collected from treatment (Tx) initiation to earliest of death, chart abstraction date or 6 months post-Tx discontinuation. Using the Kaplan–Meier method, cumulative rates of clinical response, remission, mucosal healing, dose escalation and Tx persistence were estimated over 24 months (mo). Incidence rates (per 100 person-years [PYs]) of CD exacerbations, CD-related surgeries (bowel resection, strictureplasty, colectomy), serious adverse events (SAEs) and serious infections (SIs) were assessed. A Cox proportional hazards model adjusted for baseline confounders (age, sex, albumin, C-reactive protein, disease location and duration, CD-related hospitalisations in (prior 12 mo), and disease severity) was used to compare Tx cohorts; adjusted hazard ratios (HR) with 95% confidence intervals are reported.

Results

Overall, 419 CD patients (VDZ: 177; anti-TNF: 242) [adalimumab: 125, infliximab: 111, infliximab-dyyb: 3, certolizumab pegol: (3) from 37 sites were included (median [min–max] follow-up [mo]: VDZ, 15.3 (5.0–45.9); anti-TNF, 18.1 [6.0–49.8]). Baseline characteristics are shown in Table 1.

Table 1. Baseline characteristics of real-world biologic-naive Crohn’s disease patients treated with vedolizumab and anti-TNF agents.

At 24 months, cumulative rates of clinical response, clinical remission, mucosal healing and dose escalation were similar in both Tx cohorts (Table 2). Tx persistence rates were significantly (p < 0.05) greater at 12 months (86% vs. 76%) for VDZ vs. anti-TNF patients, respectively, but did not differ at 18 (79% vs. 71%) and 24 months (71% vs. 71%). Although observed incidence rates (per 100 PYs) of CD exacerbations (17.2 vs. 25.9; p = 0.09), CD-related surgery (1.7 vs. 6.2; p = 0.079), SAEs (6.4 vs. 11.7; p = 0.16) and SIs (1.6 vs. 3.6; p = 0.13) were lower in VDZ vs. anti-TNF patients, respectively, these differences did not reach statistical significance. Adjusted HR for outcomes are shown in Table 2.

Table 2. Clinical effectiveness and safety of vedolizumab and anti-TNF agents in real-world biologic-naïve Crohn’s disease patients

Conclusion

Clinical and endoscopic outcomes in this real-world cohort suggest that VDZ and anti-TNF in CD are equally effective in a first-line biologic setting over 24 months.

Posted in: Poster presentations: Clinical: Therapy and observation (2019)