Factors Associated with Discontinuation of Anti-TNF Inhibitors Among Persons with IBD: A Population-Based Analysis

Background: Antitumor necrosis factor (anti-TNF) medications are known to be highly efficacious in persons with moderate-to-severe inflammatory bowel disease (IBD). There is a paucity of data from population-based sources to elucidate persistence with these medications in the general population of IBD. Discontinuation of anti-TNF therapy is a marker of lack of effectiveness, intolerance, and patient/physician practice preferences. Methods: We identified all persons with IBD in Manitoba who were dispensed infliximab (IFX) and adalimumab (ADA) between 2001 and 2014. Subjects were followed longitudinally to assess rates of completion of anti-TNF induction, duration of continued use, intraclass substitution, and dose adjustments. Cox proportional hazards models were used to test demographic and clinical factors associated with anti-TNF therapy discontinuation. Results: Overall, 925 of 8651 persons (10.7%) with IBD were prescribed an anti-TNF drug (705 Crohn's disease: 523 IFX and 182 ADA; 220 ulcerative colitis: 214 IFX and 6 ADA). Approximately four-fifths of persons starting on anti-TNF therapy completed induction. At 1 and 5 years, persistence rates with the original anti-TNF were approximately 60% and 40%, respectively. Immunomodulator (IM) use at the time of anti-TNF dispensation was associated with a decreased likelihood of anti-TNF discontinuation in both Crohn's disease and ulcerative colitis. ADA users with Crohn's disease who reached maintenance phase had a higher risk of discontinuation than IFX users (hazard ratio 1.64, 95% confidence interval 1.15-2.37). Conclusions: Approximately two-fifths of anti-TNF users discontinue use within 1 year of initiation, and three-fifths will have discontinued at 5 years. Concomitant IM therapy has a modest effect on discontinuation rates.