Search in the Abstract Database

Abstracts Search 2019

P239 Real-life clinical and quality of life outcomes collected remotely from patients with moderate to severe active ulcerative colitis during induction treatment with golimumab in GO OBSERVE

F. Cornillie*1, M. Flamant2, T. Haas3, E. Jörgensen4, A. Schirbel5, A. Khalifa1, M. Ferrante6, M. Govoni7, GO OBSERVE Investigators1

1MSD, Luzern, Switzerland, 2Clinique Jules Verne, Paris, France, 3Paracelsus Private Medical University, Salzburg, Austria, 4Gastroenterologie, Remscheid, Germany, 5Charité Universitätsmedizin, Berlin, Germany, 6KU Leuven, Leuven, Belgium, 7MSD Italy, Rome, Italy


Limited data are available concerning real-life experience with remotely collected patient-reported outcomes (PROs) in ulcerative colitis (UC).


GO OBSERVE is an ongoing international multi-centre observational trial with golimumab (GLM) in moderate to severe active UC patients naïve to or previously exposed to one other biological therapy. Patients receive standard subcutaneous GLM induction followed by maintenance with 100 mg or 50 mg every 4 weeks (q4wk). Mayo or partial Mayo score is collected at baseline and end of induction visit at either wk6, wk10, or wk14. Patients are asked to self-report their stool frequency score (SFS; 0–3) and rectal bleeding score (RBS; 0–3) q4wk into an electronic data capture system (EDC). Quality of life (QoL) scores are spontaneously reported by Short Health Scale (SHS) at baseline and end of induction. Partial Mayo response is defined as a decrease from baseline with ≥30% and ≥3 points and either a decrease from baseline in the rectal bleeding sub-score ≥1 or a rectal bleeding sub-score of 0 or 1. The use of concomitant UC medications is allowed per investigator’s decision. This pre-specified interim analysis reports the results at the end of induction.


In total, 102 patients were included; 88 patients have end-of-induction data for this interim analysis, including 18 patients who discontinued before wk14 due to lack of effect (n = 12), adverse event (n = 3) or withdrawal of consent (n = 3). Clinical response was achieved at either wk6, 10 or 14 in 32/88 (36.4%) patients; in 27/69 (39.1%) and 5/19 (26.3%) bio-naïve and anti-TNF exposed patients, respectively. Baseline and end of induction CRP (mg/l) was 5.20 (n = 67) and 2.20 (n = 36), respectively (p = 0.038). Baseline and end of induction median PRO2 was 4 (n = 101) and 2 (n = 68), respectively (p < 0.001) with a median change from baseline of -1 for both SFS and RBS. SHS scores were self-reported by 39 patients, with only 17 reporting SHS at both baseline and end of induction. Per cent improvement of SHS domains was: symptom burden (13%; p = 0.008), social function (20%; p = 0.015), disease-related worry (10%; p = 0.030), and sense of general well-being (10%; p = 0.167). Adverse events were reported in 20/102 patients (19.6%), including infections (n = 4), lack of efficacy (n = 9), and UC (n = 3). Serious adverse events were reported in 7 patients (6.9%) including 2 cases of severe UC.


These results from real-life practice confirm the effectiveness of GLM in active UC and show low compliance with self-reporting of PROs in UC, particularly for QoL. There is a gap between current consensus on the role of PROs in IBD and their true adoption for UC monitoring in real-life practice.