Search in the Abstract Database

Abstracts Search 2019

P649 Anti-TNF agent drug survival in patients with IBD: real-world comparisons of individual anti-TNF agents based on the Swedish National Quality Registry for IBD (SWIBREG)

I. Visuri*1, C. Eriksson1, E. Mårdberg1, O. Grip2, A. Gustavsson3, H. Hjortswang4,5, P. Karling6, S. Montgomery7,8,9, P. Myrelid4,10, O. Olén9,11,12, The SWIBREG Study Group1, J. F. Ludvigsson13,14, J. Halfvarson1

1Örebro University, Department of Gastroenterology, Faculty of Medicine and Health, Örebro, Sweden, 2Skåne University Hospital, Department of Gastroenterology, Malmö, Sweden, 3Central Hospital, Department of Internal Medicine, Karlstad, Sweden, 4Linköping University, Department of Clinical and Experimental Medicine, Linköping, Sweden, 5Linköping University, Department of Gastroenterology, Linköping, Sweden, 6Umeå University, Department of Public Health and Clinical Medicine, Umeå, Sweden, 7Örebro University, Clinical Epidemiology and Biostatistics, School of Medical Sciences, Örebro, Sweden, 8University College London, Department of Epidemiology and Public Health, London, UK, 9Karolinska Institutet, Clinical Epidemiology Unit, Department of Medicine Solna, Stockholm, Sweden, 10Linköping University Hospital, Department of Surgery, Linköping, Sweden, 11Stockholm South General Hospital, Sachs’ Children and Youth Hospital, Stockholm, Sweden, 12Karolinska Institutet, Department of Clinical Science and Education Södersjukhuset, Stockholm, Sweden, 13Karolinska Institutet, Department of Medical Epidemiology and Biostatistics, Stockholm, Sweden, 14Örebro University Hospital, Department of Pediatrics, Örebro, Sweden


Studies comparing drug survival in different anti-tumour necrosis factor (TNF) agents in IBD patients are scarce, especially for second-line anti-TNF agents. We aimed to (A) assess drug survival and predictors of response and adverse drug reactions to first-line anti-TNF treatment and (B) examine drug survival for individual anti-TNF agents when used as second-line anti-TNF.


Well-characterised patients with IBD (n = 955) starting their first anti-TNF treatment between 2006 and 2016 (Table 1), were identified from the Swedish National Quality Registry for IBD (SWIBREG). Drug survival was examined, stratified by reason for discontinuation, that is, lack/loss of clinical effectiveness or adverse drug reactions. Multi-variable Cox regression models were used to identify predictors of drug survival. Drug survival for the second anti-TNF was assessed by type of first anti-TNF agent.


Risk factors at baseline for shorter drug survival, in patients with Crohn’s disease, were use of infliximab as first-line anti-TNF (compared with adalimumab, adjusted HR = 1.95, 95% CI: 1.19‒3.18) (Figure 1A) and colonic disease (L2) (compared with ileal disease (L1) and ileocolonic disease (L3), adjusted HR = 2.16, 95% CI: 1.25‒3.74). Consistently, Crohn’s disease patients who switched from adalimumab to infliximab had shorter drug survival, compared with those who switched from infliximab to adalimumab (Figure 1B). A normalisation of CRP level at 3 months was associated with decreased risk of short drug survival in both Crohn’s disease (adjusted HR = 0.40, 95% CI: 0.19‒0.81) and ulcerative colitis (adjusted HR = 0.40, 95% CI: 0.19‒0.86). In Crohn’s disease, but not in ulcerative colitis, immunomodulators were associated with a lower risk of short drug survival due to adverse drug reactions (adjusted HR = 0.50, 95% CI: 0.31‒0.82).

Crohn's disease, n = 570Ulcerative colitis, n = 385
Male sex, no (%)298 (52)222 (58)
Median age at baseline (IQR)35 (24–48)33 (24–46)
Median disease duration in years (IQR)6 (1–16)4 (0–10)
L1 Ileal (± L4) CD, n (%)103 (18)
L2 Colonic (± L4) CD, n (%)169 (30)
L3 Ileocolonic (± L4) CD, n (%)289 (51)
L4 Upper gastrointestinal tract, n (%)6 (1)
Concomitant immunomodulators, n (%)278 (49)176 (46)
Baseline data on age, year, CD behaviour, UC extent, previous IBD surgery, CRP concentration, and smoking habits are not shown

Demographics and clinical characteristics at baseline for the 955 IBD patients treated with anti-TNF.

Figure 1. (A) Cumulative probability of remaining on the first anti-TNF treatment (Crohn’s disease). (B) Cumulative probability of remaining on the second anti-TNF treatment (Crohn’s disease). Reason for discontinuation was lack/loss of response.


Drug survival duration was longer for adalimumab compared with infliximab both when used as first anti-TNF agent and when used as second-line treatment. The consistent pattern indicates that these differences are not only explained by channelling bias (differential prescribing behaviour).