P236 Placebo response and remission rates in Ulcerative Colitis clinical trials: Systematic review and meta-analysis
V. Jairath*1, C. Parker2, G. Zou2, 3, J.K. MacDonald2, T. Alameel4, M. Albeshir4, M. Almadi5, T. Altaweel6, N. Atkinson1, S. Biswas1, T. Chapman1, P.S. Dulai7, M. Glaire8, D. Hoekman9, A. Koutsoumpas1, E. Minas10, M.H. Mosli2, 11, M.A. Samaan2, 9, M.K. Vandervoort2, S.P. Travis12, G. D'Haens2, 9, B.G. Levesque2, 7, W.J. Sandborn2, 7, B.G. Feagan2, 3, 13
1Oxford University Hospital, Translational Gastroenterology Unit, Oxford, United Kingdom, 2Robarts Clinical Trials Inc., Robarts Research Institute, University of Western Ontario, London, Ontario, Canada, 3University of Western Ontario, Department of Epidemiology and Biostatistics, London Ontario, Canada, 4King Fahad Specialist Hospital-Dammam, Department of Medicine, Dammam, Saudi Arabia, 5King Khalid University Hospital, Division of Gastroenterology, Riyadh, Saudi Arabia, 6Mubarak Al-Kabeer Hospital, Division of Gastroenterology, Jabriyah, Kuwait, 7University of California San Diego, Division of Gastroenterology, La Jolla, United States, 8John Radcliffe Hospital, Department of Gastroenterology, Oxford, United Kingdom, 9University of Amsterdam, Academic Medical Center, Amsterdam, Netherlands, 10Wexham Park Hospital, Department of Gastroenterology, Slough, United Kingdom, 11King Abdulaziz University, Department of Medicine, Jeddah, Saudi Arabia, 12John Radcliffe Hospital, Translational Gastroenterology Unit, Oxford, United Kingdom, 13University of Western Ontario, Department of Medicine, London , Canada
Defining the magnitude and modifiers of placebo (PBO) rates in randomized controlled trials (RCTs) of ulcerative colitis (UC) is essential for the design and conduct of efficient trials and to optimize the detection of true drug-PBO differences. We conducted a contemporary meta-analysis of PBO response and remission rates in induction and maintenance phases of RCTs for active UC and assessed factors influencing these rates. We report the Results of our analysis of induction trials here.
MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials and the Cochrane IBG group specialized trials register were searched from inception to April 2014. Conference proceedings were searched between 2002-2014. Eligible studies were PBO-controlled trials of UC in adult patients which: (1) contained an induction and/or maintenance phase; (2) used the UCDAI (Mayo/Sutherland/partial equivalent) as criterion for enrolment and assessment of response/remission; and (3) evaluated the efficacy of 4 drug classes (steroids, aminosalicylates, immunosuppressives, biologics). Data were extracted independently in pairs and disagreements resolved with a third reviewer. PBO rates for each outcome were pooled using a binomial-normal model for meta-analysis of proportions.
We identified 7,587 citations and 55 studies eligible for inclusion (46 induction, 9 maintenance). Pooled PBO remission (n=40) and response (n=43) rates were 10% (95% CI 7%-13%; range 1%-49%) and 33% (95% CI 28%-38%; range 6%-92%) respectively, both with significant heterogeneity (P < 0.001). Features associated with a lower PBO response were longer disease duration prior to enrolment (33% for >5 yrs vs 47% for ≤ 5 yrs), endoscopy subscore ≥ 2 at study entry (34% for score ≥ 2 vs 46% for score ≥ 1) and requirement for improvements in endoscopy and bleeding subscores as outcome measures. Features associated with lower PBO remission rates were longer disease duration prior to enrolment (10% for >5 yrs vs 19% for ≤ 5 yrs), endoscopy subscore ≥ 2 at study entry (11% for score ≥ 2 vs 24% for score ≥ 1), the requirement for improvement in the endoscopy subscore from baseline as an outcome measure and publication date after 2005 (12% for ≤ 2005 vs 9% > 2005). No difference in PBO rates was observed when disease was classified as mild-moderate vs moderate-severe at study entry, for a UCDAI cut-point ≥ 6 vs <6, or follow-up duration.
Lower PBO response and remission rates were observed in UC induction trials enrolling patients with more active disease defined by endoscopic subscore, rather than a higher composite UCDAI. This reinforces the importance of enrolling patients and assessing outcomes using objective markers of active disease.