Search in the Abstract Database

Abstracts Search 2019

P313 Impact of upadacitinib on the general clinical condition of patients with Crohn’s disease (CD): data from the randomised CELEST study

E. V. Loftus Jr*1, D. T. Rubin2, J. Panes3, D. Pugatch4, W. Zhou4, S. Goteti4, A. Lacerda4, S. Travis5

1Mayo Clinic, Rochester, Minnesota, USA, 2The University of Chicago Medicine, Chicago, Illinois, USA, 3Hospital Clínic Barcelona, Barcelona, Spain, 4AbbVie Inc., North Chicago, Illinois, USA, 5Oxford University Hospitals, Oxford, UK


This analysis assessed the impact of upadacitinib (UPA), an oral selective JAK1 inhibitor, on the general clinical condition of patients with CD. We assessed body weight and serum albumin levels, common measures used in clinical practice, as well as improvement in diarrhoea. Stool consistency was assessed via the Bristol Stool Chart (BSC), a patient-reported outcome measure used widely in patients with functional bowel disorders and favoured by regulatory agencies.


The placebo-controlled Phase 2 CELEST study (NCT02365649) enrolled adults with moderate to severe CD refractory or intolerant to immunosuppressants/biologics. During the induction period, patients were randomised to placebo (PBO) or UPA 3 mg, 6 mg, 12 mg, or 24 mg twice daily (BID) or 24 mg once daily (QD) for 16 weeks, followed by a 36-week double-blind extension phase. Changes over time from baseline (BL) in body weight and serum albumin levels were assessed (ANOVA). Stool consistency was assessed by change from BL to Weeks 4 and 16 in BSC score (proportion of days over the last week prior to the visit with BSC Type 6 [very soft] or 7 [liquid] stool; observed data) and proportion of patients who achieved BSC response at Weeks 4 or 16 (≥50% reduction in number of days over the last week with ≥1 BSC Type 6 or 7 stool vs. BL; non-responder imputation). UPA vs. PBO were evaluated at p = 0.1 level.


Among 220 randomised patients, mean ± SD weight was 75.3 ± 20.1 kg, mean albumin levels were 38.6–39.7 g/l, and mean ± SD BSC score was 0.9 ± 0.3 at BL. Weight significantly improved from BL to Week 12 with UPA doses ≥6 mg (range: 1.2–1.6 kg) vs. PBO (−0.6 kg; p < 0.05); at Week 16, changes remained significantly improved from PBO (0.0 kg) with UPA 24 mg BID (2.1 kg; p = 0.031). Albumin level changes from BL were significant as early as Week 2 with all UPA doses (range: 0.2–1.9 g/l) vs. PBO (−1.0 g/l; p < 0.05) and were maintained through Week 16 with UPA doses ≥6 mg (range: 2.6–4.0 g/l) vs. PBO (0.4 g/l; p < 0.05). BSC scores significantly improved from BL by Week 4 with UPA 6 mg, 12 mg, and 24 mg BID vs. PBO and were maintained to Week 16 with 6 mg and 24 mg BID (p ≤ 0.05; Figure A). A significantly greater proportion of patients receiving UPA 12 mg and 24 mg BID at Week 4 and 6 mg and 24 mg BID at Week 16 achieved BSC response vs. PBO (p < .05; Figure B).

Figure. Change from baseline to Weeks 4 and 16 in Bristol Stool Chart score (A) and proportion of patients with Bristol Stool chart response at Weeks 4 and 16 (B).


UPA induction treatment resulted in significant improvements in body weight, serum albumin levels, and stool consistency in patients with CD compared with PBO. Improvements in these parameters paralleled conventional outcomes such as CD activity index and mucosal healing.