P648 Anti-TNF-α therapy, use of corticosteroids, and colectomy among paediatric and adolescent patients with ulcerative colitis: a nationwide study

Abstract Background The long-term beneficial effects of anti-TNF-α therapy are debatable referring to the need for corticosteroids and changes in colectomy rates among paediatric and adolescent patients with ulcerative colitis (UC). We aimed to investigate whether anti-TNF-α treatment reduced the use of corticosteroids and to examine colectomy rates in the era of anti-TNF-α therapy compared with a historical cohort. Methods The study population included an unselected nation-wide cohort of children and adolescents (0–20 years of age) diagnosed with UC through 1977–2016. The data were retrieved from the Danish National Patient Registry. We identified anti-TNF-α use as patients who had at least four anti-TNF-α treatments within a period of 4 months to examine a subsequent need of corticosteroid prescriptions (Figure 1). To examine the change of colectomies we calculated the cumulative risk 5 years following the diagnosis and used an adjusted Cox regression model in the comparison of colectomies between a historical cohort (1977–2003) and a cohort representing the era of anti-TNFα (2004–2016). Figure 1. Anti-TNF-α use: individual patient timeline illustration. Results We identified totally 4449 patients with UC in the study period. We identified 334 children and adolescents treated with anti-TNF-α. A total of 20.7% (69 patients) was prescribed corticosteroids in a 3-month period calculated from the fourth anti-TNF-α treatment. The proportion of steroid dependent users declined after 6 and 12 months to 6.6% and 0.6%, respectively (Figure 2). The 5-year cumulative proportion of colectomy in the historical cohort was 9.12% (95% confidence interval [CI]: 8.01–10.37) and 7.76% (95% CI: 6.70–8.98) in the era of anti-TNF-α treatment (Figure 3). The adjusted hazard ratio was 0.84 (95% CI: 0.68–1.03) for colectomy within a 5-year period in the era of anti-TNF-α compared with the historical cohort. Figure 2. Corticosteroid prescriptions among anti-TNF-α users with 3-, 6-, 9- and 12-month follow-up Figure 3. Cumulative percentage for colectomy, a 5-year follow-up Conclusions The concomitant use of corticosteroid prescriptions was virtually terminated after 12 months among patients treated with anti-TNF-α. Within a period of 5 years from the time of diagnosis, the adjusted hazard ratio for colectomy in the era of anti-TNF-α treatment was reduced, but not significantly compared with a historical cohort.